Deterring Rational Fanatics 9780812292046

Citation preview

Deterring Rational Fanatics

Deterring Rational Fanatics Alex S. Wilner

U N I V E R S I T Y O F P E N N S Y LVA N I A P R E S S PHIL ADELPHIA

Copyright ©  University of Pennsylvania Press All rights reserved. Except for brief quotations used for purposes of review or scholarly citation, none of this book may be reproduced in any form by any means without written permission from the publisher. Published by University of Pennsylvania Press Philadelphia, Pennsylvania - Printed in the United States of America on acid-free paper 1

3

5

7

9

10

8

6

4

2

Library of Congress Cataloging-in-Publication Data Wilner, Alex S., – Deterring rational fanatics / Alex S. Wilner. — st ed. p. cm. Includes bibliographical references and index. ISBN ---- (hardcover : alk. paper) . Terrorism—Prevention. . Deterrence (Strategy) I. Title. HV.W  .'—dc 

For Kate and Noa Tili

PB

Dorothy L

. Hodgson

CONTENTS

Chapter 1. Introduction

1

Chapter 2. Deterrence Theory: Exploring Core Concepts

16

Chapter 3. Deterring Terrorism: Contemporary Debates

37

Chapter 4. Targeted Killings: Theory, Practice, and Consequence

75

Chapter 5. Targeting the Taliban: Coercive Lessons from Afghanistan

123

Chapter 6. Moving Ahead with Deterrence Theory

172

Appendix: Research Design and Methodology

189

Notes

201

Index

245

CHAPTER 1

Introduction

For decades, deterrence theory was the veritable bedrock on which American, British, Soviet, Chinese, Indian, and Israeli foreign and military policies were based. During much of the Cold War, these and other states tailored their relations and interactions with foes and friends alike on the basis of coercion, compellence, and deterrence. Classical deterrence theory is a product of that period: The development and proliferation of nuclear weapons in par ticular necessitated that wars between great powers be avoided or contained. Academics did their part by exploring the intricacies behind the logic and theory of deterrence and by outlining the promises and pitfalls of applying the theory to the practice of warfare. Few other theories of international relations have received such an extended period of attention and assessment. With the collapse of the Soviet Union and its Communist bloc, however, deterrence theory was shunted off center stage. Major war between the great powers seemed exceptionally unlikely and rampant nuclear proliferation never occurred. Moreover, the conflicts that did emerge during the 1990s were often brushfire affairs or involved fighting within (rather than between) states, conflicts that deterrence theory was generally ill-prepared to properly address. Deterrence theory lost even more ground with the eventual refinement and ascension of competing doctrines, like counter-proliferation, preemption, and prevention, which offered states alternative ways to secure their national interests. But it was al Qaeda that sealed the theory’s decline. Its 2001 terrorist attack on the United States seemed to defy the very premise on which deterrence theory was based. Terrorists were deemed irrational, stateless, and often fanatically (and religiously) dedicated to an immutable ideological cause. That al Qaeda had resurrected the kamikaze in its suicidal operatives

2

Chapter 1

was deemed particularly problematic: “The religious orientation of the Islamists,” American political commentator Fareed Zakaria explained in 2003, “breaks down deterrence. How do you deter someone who is willing, indeed eager, to die?”1 That sentiment was expressed in U.S. counterterrorism and strategic doctrine. In a 2006 address to the United States Military Academy, President George W. Bush warned: “The enemies we face today are different in many ways from the enemy we faced in the Cold War. Unlike the Soviet Union, the terrorist enemies we face today hide in caves and shadows . . . have no borders to protect, or capital to defend. They cannot be deterred.”2 The unsettling consensus was that deterrence had lost its utility, that the theory could not be applied to counter contemporary terrorist organizations, and that terrorists were themselves altogether undeterrable. In counterterrorism, the best defense, U.S. secretary of defense Donald Rumsfeld concluded, was “a good offense.”3 The bottom line was that terrorists responded to force, not to the intricacies of coercion, and offensive and defensive counterterrorism trumped any and all talk of deterrence. This book challenges the conventional wisdom. By expanding the scope of traditional deterrence theory and pairing it with a more nuanced understanding of contemporary terrorism, a variety of deterrents can be constructed and levied against terrorist organizations. When tailored appropriately, states can use the logic of deterrence to influence, coerce, and deter terrorist groups, delimiting the type and ferocity of the violence those groups are willing to use, and influencing their behavior more broadly. Deterrence theory is not a panacea for terrorism, but it can serve as a strategic guide for state responses. And although a healthy dose of skepticism may be prudent, relegating deterrence to the dustbin of strategic theory is both shortsighted and premature.

Deterrence Skepticism In the years following 9/11, deterrence skepticism ran deep. Bush’s arguments were echoed by an array of political and strategic commentators. At least part of this skepticism was a result of our collective reaction to 9/11. The tragedy was met—quite justifiably—with an impulsive, an emotional, and even a reflexive response. Terrorism was not new, but the scale of the destruction visited on New York City and Washington, D.C.; the audacity of the attack itself (in which civilian airliners were used as guided missiles); the (supposed) religious justification underpinning the act; and the use of

Introduction

3

suicide operatives for tactical improvement seemed to overwhelm traditional forms of discourse. The unbridled and wanton destruction of life and property simply shocked the senses. In a 2003 remark that captures much of the immediate sentiment that surrounded al Qaeda’s attack, Elie Wiesel, the venerable Nobel laureate, Holocaust survivor, and author, argued that “unlike their distant predecessors,” contemporary terrorists “attack people, any people, all people at a certain place, simply because they happen to be there, at that moment.”4 According to Wiesel, there was little reason behind 9/11 except to wreak havoc and spread fear. Deterrence experts seemed to agree. British scholar Sir Lawrence Freedman, in his 2002 volume Superterrorism, writes that modern terrorists were “much readier to inflict massive loss of life and move to ever more horrific methods in their efforts to do so.”5 Whereas terrorist organizations had once sought maximum publicity—going so far as to hold press conferences; offer communiqués and explanations; and, at the very least, publicly claim responsibility for their acts of violence—today’s terrorist, skeptics calculated, wants maximum destruction above all else. Charles Krauthammer, a hawkish American political commentator, came to a natural conclusion: “the obsolescence of deterrence.” And if nonstate militant groups like al Qaeda were “inherently undeterrable” then a more aggressive and assertive U.S. foreign policy that altogether eschewed deterrence was surely needed.6 Broadly speaking, deterrence skepticism incorporates five core arguments. First, the fundamentalist religiosity of al Qaeda and the religious motivation of many contemporary militants interrupts rational decision making, contradicting rational deterrence theory. This particular view resonates with academics, decision makers, and the general public. “For the religious terrorist,” American terrorism expert Bruce Hoffman explained, “violence is first and foremost a sacramental act or divine duty executed in response to some theological demand or imperative. Terrorism thus assumes a transcendental dimension, and its perpetrators are consequently undeterred by political, moral, or practical constraints.”7 Unlike the nationalist, separatist, or Marxist militants of years past, religious terrorism approaches the divine; violence is rationalized as a religious obligation. To an extent, these arguments appear particularly relevant when interpreting al Qaeda and its followers: They present themselves as pious, devout, and dutiful Muslims. It is important to note, then, that whereas we speak in descriptive terms—of catastrophic suicide terrorism, for instance—al Qaeda speaks in religious terms: 9/11 and the thousands of acts of terrorism that followed are martyrdom

4

Chapter 1

operations; the foot soldier is not an indoctrinated or a suicidal drone but a shaheed, a venerated martyr romantically pursuing his or her religious responsibility in exchange for an afterlife in paradise; al Qaeda and its supporters represent the Islamist vanguard in a global and cosmic war with the unholy West; and this war, supported by fatwas and various religious decrees, is legitimate, meant to reassert Islamic doctrine and God’s rule over parts of the globe.8 For illustration, consider the case of the Chechen “suicide squad” dispatched by the Special Purpose Islamic Regiment (SPIR) to Moscow, Russia, in 2002 on orders to attack the Dubrovka Theatre. Operatives clipped bomb-belts around their waists and wired the pillars, walls, and ceiling of the theater with explosives, underscoring that they were, as their spokesperson explained, “more keen on dying than you [the hostages] are keen on living.” In a tape delivered to the Moscow bureau of Al Jazeera, one of the nearly two dozen female terrorists involved in the attack added: “It makes no difference for us where we will die. We have chosen to die here, in Moscow, and we will take the lives of hundreds of infidels with us.”9 Although Western audiences interpreted this contemptuous desire for murder-suicide as sheer lunacy, the squad’s message was a carefully crafted script laden with religious symbolism dating back a millennium. In 634 CE, Khalid ibn al-Walid, the commander of the outnumbered Arab army preparing to battle the Persians at Mada’in, sent a message to his adversaries: “In the name of God, the AllCompassionate, the Merciful . . . enter into our faith [lest we come with] a people who love death just as you love life.”10 The Moscow operation was absolved by invoking God’s rule. Al Qaeda is fond of tapping into these religious allegories as well. In its statement immediately following 9/11, it, too, warned that “the Americans should know that the storm of plane attacks will not abate. There are thousands of the Islamic nation’s youths who are eager to die just as the Americans are eager to live.”11 And as Osama bin Laden himself has put it: “He who God guides will never lose.”12 How can we possibly apply deterrence to that? Second, skeptics assume that fanaticism creates diverging rational contextualizations, “second realities,” and “deformed consciousnesses,” which negate deterrence theory and practice in various ways.13 This may be especially true of contemporary suicide bombers. As Lewis Dunn offers, the suicide operative’s “commitment to jihad has made the sacrifice of their lives simply an avenue to a new life after death.”14 In fact, as members of the Chechen squad admit, death is welcome. Without fear, the threat of punishment or

Introduction

5

retaliation—a critical element of deterrence theory—cannot be properly issued. And indeed, militants may desire military retaliation to galvanize would-be supporters into action or to lend credence to their worldviews. If fear is absent or military reprisal sought, deterrence will not work. Contemporary terrorists are thought to weigh the costs and benefits of their actions in ways that are irrational or, at the very least, markedly different from the manner in which Western governments weigh their own actions and expectations. An adversary “that prefers escalation regardless of the consequences,” Daniel Whiteneck reminds us, “cannot be deterred.”15 Other scholars argue that terrorists do not adhere to behavioral principles outlined in the strategic model of terrorist behavior, which assumes that terrorist groups are rational actors that attack civilians in order to achieve political ends. The work of Max Abrahms is particularly relevant here. Abrahms finds that terrorist behavior contradicts the principles of the rational model in a number of important ways: Terrorists usually fail to achieve their political goals by attacking civilians, and they continue to attack civilians even after further gain is unlikely; terrorists use violence as a first resort and usually reject opportunities to evolve into nonviolent political movements; terrorists often attack anonymously, negating the possibility for states to offer concessions; and terrorist groups resist disbanding even after having failed to achieve their political objectives.16 Other scholars add that some terrorist groups may need to perpetually sustain a terrorist campaign to prove their resiliency, vitality, and continued importance.17 This, too, may impair deterrence. And, finally, religiously self-assured terrorists are prone to be highly motivated, decidedly risk-accepting, and innately resolute. Their threshold for pain and their willingness to accept extraordinary risks may be particularly high as a result. Some may believe they have nothing to lose. All of these arguments complicate the practice of deterrence in counterterrorism. Third, some skeptics remark on the seemingly absurd objectives sought by some contemporary terrorist groups. As the adage goes: Modern terrorists don’t want an invitation to sit at the table; they want to destroy the table and kill those already seated. American scholar Mark Juergensmeyer goes so far as to suggest that the “new” terrorism appears “pointless since it does not lead directly to any strategic goal. . . . It is the anti-order of the new world order of the twenty-first century.”18 If an adversary considers conflict and warfare a success and political compromise a failure, then deterrence can never be applied. Likewise, militant objectives appear to have shifted from securing publicity to seeking maximum punishment and pain. They

6

Chapter 1

kill for killing’s sake, to serve some abstract purpose of their own fantastic construction. In this vein, 9/11 was “symbolic drama,” an attack that fit al Qaeda’s fantasy of a “pure Islamic David” combating the American Goliath. The 9/11 attacks themselves were less about structuring a classic Clausewitzian victory against a stronger adversary and more about fulfi lling the “collective fantasy” of radical Islam;19 winning may not have been al Qaeda’s primary objective or motivation. Historically, this has not always been the case. Terrorism expert Brian Michael Jenkins laments that the terrorists of the past “had a sense of morality, a self-image, operational codes, and practical concerns.”20 Less so today. Contemporary terrorism, deterrence skeptics maintain, is not only ruthless but also strategically rudderless. Everything and anything could and should be expected (hence, Rumsfeld’s proverbial “unknown unknowns”). This is problematic: Violence devoid of strategy is difficult if not impossible to deter. Fourth, deterrence skeptics note that terrorist organizations lack a “return address”—a target of value, usually territorially based, against which a retaliatory threat can be issued and carried out.21 In interstate (or, stateon-state) deterrence, threats are levied against an adversary’s territory and sovereignty, industrial base, population centers, civilian and military infrastructure, and so on. Deterrence requires that a punishment be levied against something, somewhere. During the Cold War, both Washington and Moscow threatened each other with obliteration, a credible threat (given the destructive power of nuclear weapons) that informed and influenced behavior. But terrorist organizations rarely control territory against which military threats can be issued. And the territory terrorists may at times rule—in, for example, the ungoverned patches of Pakistan, Afghanistan, Yemen, Iraq, Mali, Somalia, and Syria—almost always rests within the boundaries of an internationally recognized sovereign state, which complicates how military threats and operations can be issued and carried out. The U.S. government’s 2002 National Security Strategy made careful note of these par ticu lar dilemmas: “Traditional concepts of deterrence will not work against a terrorist enemy . . . whose most potent protection is statelessness.”22 With few targets to threaten, deterrence becomes moot. Finally, the immediate and reflexive interpretation of al Qaeda’s 9/11 attack suggested that the felling of the Twin Towers and the strike on the Pentagon was a deterrence failure of grandest proportion. After all, the attacks proved that al Qaeda had not been dismayed or deterred by the distinctive

Introduction

7

power and military prowess of the United States. During the 1990s, it was common to speak of the United States as the sole remaining superpower. In military terms, it had no peer. It was too big to bully, its strength palpable. And yet, bin Laden did not seem to care. Put simply, al Qaeda’s strike on U.S. soil, skeptics argue, was a failure of conventional U.S. deterrence. “Had these people had ballistic missile technology,” John Bolton, then the top arms control official at the U.S. State Department, declared, “there is not the slightest doubt in my mind that they would have used it.”23 But why stop there? Al Qaeda’s attack was also a failure of U.S. strategic nuclear deterrence. After all, on the eve of 9/11 al Qaeda knew full well about the U.S. nuclear arsenal. And still it was not deterred. “If al Qaeda had had a nuclear weapon,” James Kraska too easily concludes, “it would have used it.”24 Kraska’s assertion parallels Bolton’s: If al Qaeda had had other, more devastating tools at its disposal on September 10, it would have used them without reservation or care for U.S. military capabilities. In sum, not only are terrorists, like al Qaeda, and terrorism, more broadly, undeterrable, but deterrence itself as a concept, theory, and strategy failed spectacularly on September 11, 2001.

Deterrence Skepticism . . . Revisited Put together, these arguments suggest that deterrence, in both theory and practice, has grown increasingly irrelevant and is not applicable at all to counterterrorism. And yet a more nuanced appreciation of the logic on which deterrence theory is based and a more robust understanding of how terrorist organizations function can provide some counterintuitive findings. Coercing terrorists and influencing their behavior is not as impractical as skeptics assume. Terrorists may be fanatical, but they are “rational fanatics.” The late Ehud Sprinzak, a leading Israeli scholar of political violence, warned that we should not blindly accept the notion that terrorists are willing and ready to do anything at any cost. They are not. “The perception that terrorists are undeterrable fanatics who are willing to kill millions indiscriminately,” he argues, “belies the reality that they are cold, rational killers” who use violence to achieve specifically defined political and social objectives.25 Though individual militants may embrace and accept extremist ideologies, fanatical worldviews, and violence in the name of scripture, the organizations they join have strategic priorities and goals. Terrorism and political violence is

8

Chapter 1

purposeful; it is calculated violence.26 Even the suicidal terrorist willingly forfeiting his or her life in order to destroy others, is, to a certain degree, acting rationally. Professor Mia Bloom, in Dying to Kill, explains: “The organizations [using suicide terrorism] do cost-benefit analyses. . . . Suicide bombings should be disaggregated into two levels of analysis—the individual bombers who blow themselves up and the organizations that send them. To varying degrees, both parties . . . are acting rationally in the strictest sense of the term since they are pursuing goals consistent with picking the option they think is best suited to achieve their goals.”27 All terrorism, even suicide terrorism, may be a product of rational calculations, calculations that can be altered using the logic of deterrence theory if the right leverages are found and used. This holds at both the individual and group levels. “It is hard to think of any instance in history of a state or similar power grouping,” British defense expert Sir Michael Quinlan writes, “that had no assets which it wished to retain, no collective concern for the lives of its members and no interest in the survival of its ruling regime.”28 And if such concerns exist, then it is feasible—at least in theory—to identify ways to manipulate those concerns and alter behavior. Deterring terrorism is possible; it involves picking apart the terrorist threat and developing coercive leverages that speak to the unique qualities of each adversary. Likewise, terrorists may be stateless, but they are not asset-less. They have (and hold on to) things they value. As organizations, they have personnel, training facilities, gear and equipment, leaders, friends and allies, even technical expertise. Terrorists also have aspirations and goals that they would rather not unnecessarily jeopardize, and they have ideas and sentiments they believe are worth protecting. In form and function, all of these are assets that terrorists consider valuable. These assets are not (usually) territorially based and certainly do pose a problem for certain coercive processes and concepts, like deterrence by punishment or the Cold War’s mutual assured destruction (MAD). But statelessness is not a problem for deterrence theory writ large: Unique assets simply require unique threats. So although some, or even most, terrorists may scoff at traditional threats that target land-based resources in traditional ways, they may be sensitive to other threats that are tailored more specifically against certain possessions and values. If we can identify what it is that specific terrorist groups value, we can develop threats that influence behavior. Finally, 9/11 was not a deterrence failure; it was a failure to apply deterrence theory. Instead of looking up from the smoldering evidence of the

Introduction

9

World Trade Center and extrapolating from there that al Qaeda was undeterred, that U.S. conventional and nuclear power was muted, or that deterrence was altogether obsolete for managing contemporary conflict with nonstate militant organizations, we must ask whether the United States even tried to deter al Qaeda in the first place. Deterrence does not just happen. And it rarely springs forth, naturally and on its own, from the mere possession of certain military hardware. Instead, deterrence is something states do; it is something they purposefully put into motion. Deterrence is a communicative process in which military capabilities are purposefully packaged into coercive threats and then broadcast to a specific adversary in hopes of altering or manipulating the behavior of that adversary. So although Rumsfeld suggested that the 9/11 terrorists “were clearly not deterred by doing so from the massive U.S. nuclear arsenal,” simply having those weapons was not enough.29 Washington needed to translate its military and nuclear capabilities into specific coercive threats that al Qaeda could have understood. As will become evident in the following chapters, practicing deterrence requires fulfilling certain prerequisites. If and when threats are absent, muddled, miscommunicated, or weak, or if the perceived benefit of action outweighs the cost of punishment in an opponent’s decision calculus, deterrence in practice may fail while deterrence in theory analytically succeeds. Deterrence does fail, writes George Quester, “when statesmen and military leaders fail to notice, and be deterred by, powerful modes of retaliation.”30 But deterrence failure in practice should not be confused or equated with failure of deterrence theory. The distinction between theoretical and practical deterrence failure is important when discussing counterterrorism. A more nuanced evaluation of 9/11 would have to assess whether the United States had been actively, openly, and credibly attempting to deter al Qaeda and whether al Qaeda received and accurately interpreted U.S. deterrent threats. On both counts, the answer is no. First, prior to 2001 the United States failed to communicate a credible deterrent threat against al Qaeda. Neither the Bush nor the Clinton administrations had done much to discourage or dissuade al Qaeda.31 In his many audio and written messages, bin Laden describes the United States as a paper tiger, epitomizing the false courage shared by the Western world. Bin Laden considers these the lessons of pre-9/11 history: American and French disengagement from Lebanon after the bombings of the barracks in Beirut (1983) and the embassies in Kuwait City (1983); Soviet capitulation to the Afghan mujahideen (1988–89); Western withdrawal from Yemen following

10

Chapter 1

the Gold Mohur Hotel bombings (1992); no (apparent) American response to the World Trade Center bombing (1993); Western departure from Somalia following the Battle of Mogadishu (1993); a weak American response to the Khobar Towers bombing in Saudi Arabia (1996) and the U.S. Embassy bombings in Kenya and Tanzania (1998); no response to either the failed suicide attack on the USS The Sullivans or the successful attack on the USS Cole in Aden, Yemen (2000). For bin Laden (and other militants), these events highlighted the West’s inability to effectively combat terrorism and its disinterest in doing so. American Middle East scholar Bernard Lewis suggests that decades of counterterrorism neglect were interpreted as an “expression of fear and weakness rather than moderation.”32 In fact, al Qaeda may have been encouraged by the tepid U.S. response. And the process was cumulative. By 2001, al Qaeda’s experience suggested that it could target Westerners with near impunity. Al Qaeda was undeterred from carry ing out 9/11 not because it was irrational or insensitive to costs, but rather because the United States and its allies had given it little reason not to carry out that mission. In addition, al Qaeda neither properly understood nor fully appreciated the consequences of its actions. Indeed, it may have anticipated a limited or an anemic American response. “The success of the 9/11 operation,” American terrorism scholar Marc Sageman writes, “backfired on al Qaeda.”33 That the Bush administration (and the Obama administration, as well) acted fiercely, by freezing known al Qaeda sources of finance, mustering a grand coalition of the willing, destroying al Qaeda’s Afghan sanctuary and training facilities, and eliminating thousands of militant leaders and operatives in short order, may have been unanticipated. Consider Abu al-Walid alMasri’s reaction. An al Qaeda theoretician, he later denounced bin Laden for so recklessly misinterpreting U.S. capabilities, for entangling al Qaeda in a war it would lose, and for bringing about the destruction of the only purely Islamic emirate (i.e., Afghanistan). Bin Laden, al-Masri concludes, was a shortsighted and “catastrophic” leader.34 A similar argument can be made of al Qaeda’s 9/11 suicide hijackers: They too were undeterred because they went untested, over and over again. The operatives were cautious and “risk averse,” did not want to get caught, and were far from reckless, yet the U.S. security system repeatedly failed to challenge them at the “threshold levels necessary to deter their attack.”35 A threat unmade is a threat unheeded. Neither al Qaeda’s leaders nor the 9/11 operatives could have possibly interpreted a

Introduction

11

coercive threat that did not exist; they also could not fully appreciate the costs and consequences of their actions.

A Road Map to This Book The purpose of this book is twofold: to update theories of deterrence so that they better address contemporary terrorist threats and to explore whether and how targeted killings, in particular, manipulate militant behavior. In the first case, I take issue with the conventional thinking that terrorism cannot be deterred and that deterrence theory is outdated or obsolete. I suggest otherwise and illustrate how deterrence theory can be adapted, expanded, and applied to nonstate militant adversaries. In the second case, targeted killings have become an increasingly prominent (if not also controversial) feature of U.S. counterterrorism strategy. Although not an especially novel practice—targeted killings and assassinations have been around for centuries—developments in drone technology have certainly facilitated a rapid increase in their use in counterterrorism and counterinsurgency operations around the globe. In discussing drones and targeted killings in 2009, Leon Panetta, then director of the Central Intelligence Agency (CIA), was characteristically blunt: “[I]t’s the only game in town in terms of confronting or trying to disrupt the al Qaeda leadership.”36 Armed drones are here to stay and will continue to be used in targeted operations. But the strategy itself requires in-depth scholarly analysis. By assessing the effects that targeted killings have on militant behavior, this study taps into and adds to nascent debates that explore the promises and pitfalls of eliminating militant leaders in precision strikes and special operations. My argument unfolds accordingly. First, I evaluate how and where deterrence theory can be applied to counterterrorism. By assessing the distinct literatures on deterrence theory and counterterrorism, I propose ways in which the logic of coercion, punishment, denial, and delegitimization might be tailored to the specifics of nonstate militancy and political violence. This discussion leads to the exploration of a number of avenues for deterring terrorism. I then explore one of these avenues in particular: the deterrent and coercive effect of targeting and eliminating militant leaders. In brief, targeted killings are the intentional slaying of individual terrorist leaders and facilitators. In the language of deterrence theory, targeted killings represent

12

Chapter 1

a cost to planning and participating in political violence, a cost that should, in theory and in practice, influence or coerce group and individual militant behavior. By reviewing the literature on targeted killings, I assess whether and how they might function as an instrument of deterrence by punishment or denial in counterterrorism. Finally, I turn to an empirically driven analysis of how targeted killings in Afghanistan have had a coercive effect on the Taliban’s behavior. The primary methodological assumption is that a within and cross-case comparison of various Afghan targeted killings will reveal behavioral patterns on the part of the Taliban that inform particular aspects outlined in both the targeted killing and deterrence literatures. The eliminations themselves introduce a break (or control) on the longitudinal behavior of the Taliban, such that behavioral changes can be measured after each strike. The empiricism is designed as a structured and focused comparison of related cases. Four targeted killings from the Afghan theater are analyzed: Mullah Dadullah (killed May 12, 2007), Mullah Mahmud Baluch (killed June 9, 2007), Qari Faiz Mohammad (killed July 23, 2007), and Mullah Abdul Matin (killed February 18, 2008). My findings suggest that these eliminations did indeed coerce, manipulate, and alter Taliban behavior in ways outlined in theories of deterring terrorism. This book is presented in six chapters. This chapter is an introduction to deterring terrorism and exposes the various arguments skeptics have made. It positions the subject matter within the broader theoretical context of international relations theory, highlights how global events have shifted contemporary debates regarding deterrence theory’s continued utility, and anchors the subsequent exploration. Chapter 2 establishes the foundation on which deterrence theory is built. A detailed overview of the literature is presented, establishing the logical underpinnings and building blocks that inform the use of deterrence in practice. Included here are discussions of the various causal logics (like deterrence, compellence, coercion, and influence) and theoretical iterations (like deterrence by punishment versus deterrence by denial) that underpin traditional deterrence theory. Chapter 3 applies the traditional theory of deterrence to the practice of countering violent nonstate actors and terrorism. The discussion begins by first unpacking terrorist groups into their various organizational characteristics (that is, militant leaders, foot soldiers, financiers, societal and state facilitators, and ideological legitimizers) and disaggregating terrorism into its distinctive processes (i.e., recruitment, indoctrination, training, and action). Breaking terrorism into its parts and processes helps identify the individual

Introduction

13

actors and processes that may be sensitive and vulnerable to the logic of deterrence. Some terrorist actors (like financiers, for instance) may be especially susceptible to threats of punishment (that is, imprisonment or death). Similarly, certain terrorism processes (like target selection) might be manipulated by applying the logic of deterrence by denial (that is, hardening infrastructural defenses). By broadening the traditional scope of deterrence, this chapter also illustrates how core tenets of deterrence theory can be applied to unconventional and nonstate adversaries, like terrorists, insurgents, and militants. Coercive dilemmas are also reviewed: how states should identify terrorist values and goals; communicate threats with nonstate adversaries; recognize and manipulate terrorist self-restraints; use punitive and nonpunitive influences and positive inducements; and deal with asymmetries in power, capability, intent, and motivation. A separate discussion then addresses the role deterrence has in countering conventional terrorism and in countering chemical, biological, radiological, and nuclear (CBRN) terrorism. In sum, the chapter broadens traditional theories of deterrence so that core concepts can be properly applied to the intricacies of deterring terrorism and coercing nonstate militant adversaries. Chapter 4 focuses on the deterrent effect targeted eliminations have on militants and groups. Targeted killings are an example of how states can apply the logic of deterrence by punishment to counter political violence. By communicating an intention and a capability to target and capture or kill militant leaders, states might be able to coerce, influence, and deter terrorists. The chapter begins by defining the theoretical and practical parameters of targeted killings as they apply to contemporary counterterrorism. A twopart discussion then explores why targeted killings do and do not work in diminishing militant capabilities and deterring violence. For instance, although it is argued that eliminating a group’s leader may lower his followers’ morale, a leader’s death may have the opposite effect, too: creating a martyr with which to attract attention and galvanize supporters and new recruits. Other related issues, like a group’s deprofessionalization and the normative and legal dilemmas that accompany targeted killings and the use of drones to dispatch militants, are also discussed. The chapter concludes with a detailed discussion of how and why targeted killings coerce adversaries, and it establishes the theoretical groundwork on which the empirical chapter is based. Chapter 5 presents the book’s empirical findings. It begins with a description of the Taliban’s leadership and describes the 2007 U.S./NATO

14

Chapter 1

“Most Wanted” campaign, which helps situate the data set itself. The comparative analysis of the four cases is then carried out. (A description of the methodology, research design, and data collection is presented in the Appendix.) The findings suggest that the targeted killings had a coercive and deterrent effect on the Taliban, influencing the type and nature of the violence the organization was willing to orchestrate and was capable of carrying out. When assessing deterring terrorism, it is not only the amount of violence but the nature of that violence that is important. Terrorist and militant organizations have behavioral preferences. The type of violence they engage in rests as much on the effect they are trying to have as it does on their capacity and motivation to muster efforts toward particular goals. With these preferences in mind, this chapter reveals that after the targeted killings, suicide bombings, for instance, dropped by more than 30 percent. This is in keeping with the degree of difficulty, amount of time and expertise, and level of leadership that is required to coordinate effective suicide bombings. Overall levels of violence, however, are only a minor part of the analysis. The research also reveals a decrease in Taliban professionalism following the targeted killings, with drops in the proficiency of Taliban improvised explosive devices (IEDs) and suicide attacks. The data further suggest that the targeted killings influenced the Taliban’s selection of targets; soft targets were more often selected (as a percentage of all targets) after the leadership strikes. Chapter 6, which functions as the book’s conclusion, discusses these findings and offers lessons for studying contemporary deterrence theory and applying targeted killings in counterterrorism. Success in counterterrorism requires diminishing a group’s ability to organize and orchestrate acts of violence, as well as undermining a group’s motivation to use violence.37 The former flows from the use of offensive and defensive capabilities. The latter relies on the logic of deterrence, compellence, and influence and involves manipulating an organization’s willingness to use violence and its supporting community’s enthusiasm to facilitate it. Today, counter-motivation strategies are underdeveloped and understudied. In this regard, applying deterrence theory to counterterrorism is a promising field of research. By illustrating how traditional tenets of deterrence, like punishment and denial, along with newer concepts, like delegitimization and mitigation, might be used to coerce, compel, influence, and deter terrorists, this study suggests ways in which states might counter and combat political violence strategically. Although it is true that deterring terrorism will be difficult and complex, especially when compared to deterring state rivals

Introduction

15

or the former Soviet Union, targeting what terrorists value, desire, and believe might nonetheless influence militant behavior in ways that diminish the scale, scope, and ferocity of the violence they orchestrate. Deterring all terrorism may be impossible; but deterring some terrorism is feasible and, ultimately, necessary.

CHAPTER 2

Deterrence Theory: Exploring Core Concepts

The logic of deterrence is simple. It rests on convincing an adversary that the costs of taking a par ticular action outweigh the potential benefits the action is expected to provide. Deterrence is fundamentally about manipulating another’s behavior in a way that suits your needs, interests, or goals. Although classical deterrence theory is a product of the Cold War, deterrence is neither a novel nor a contemporary creation. Even the Romans understood that if you want peace, you need to be ready for war.1 And though international relations theory has popularized deterrence in the study of international crisis and conflict, to a certain extent all human behavior is dictated at least partially by the logic of deterrence. Laws of the marketplace, for instance, spell out the costs individuals risk enduring if and when they break codes of practice. Most people understand these laws and although they may see the benefits of breaking them, they refrain from doing so because of the risks and costs of getting caught. Automobile drivers act the same way when they buckle their seatbelts and purposefully avoid driving over the speed limit. Parents, too, have been known to use threats (“I will take away your allowance, if you . . .”) and inducements (“I will give you extra allowance, if you . . .”) to tailor the behavior of their children. It was the nuclear revolution that brought the logic of deterrence under the microscopic scholarship of political science and strategic studies. The vast nuclear armaments of the competing Cold War rivals necessitated that war avoidance take on new meaning, and the theories of deterrence took shape. Nobel laureate Thomas Schelling’s often-cited definition presents deterrence as “persuading a potential enemy that he should in his own interest avoid certain courses of activity.”2 Compellence is a closely related concept. Like deterrence, it is meant to influence an adversary’s behavior by

Deterrence Theory

17

threatening punishment, denying objectives, or doing both, but it is meant to encourage (rather than dissuade) a particular behavior. These core concepts are inherently related: Deterrence involves “inducing an adversary . . . not to do something”; compellence involves inducing an adversary “to do something” it might not otherwise have done.3 Although the former deals with the don’ts of adversarial relations, compellence—as its root “compel” suggests—deals with the dos. In either case, the objective is for one actor, group, or state to “prevent undesired actions” from another actor, including those actions already taken and those contemplated but not taken.4 In the parlance of deterrence theory, the actor being deterred is the challenger and the actor doing the deterring is the defender.5 Both deterrence and compellence are coercive strategies used to manipulate behavior. Coercion, explains Sir Lawrence Freedman, is the threatened and “actual application of force to influence the action of a voluntary agent” by weighing on its “strategic choices.”6 Coercive strategies are effective if they prompt voluntary changes in behavior. At its core, coercion, like deterrence, involves convincing an adversary to concede to your demands, not to force your adversary to concede by leaving that adversary physically unable to do otherwise.7 The goal of coercion, deterrence, and compellence is not to force an adversary to act in a certain way by destroying the adversary’s ability to act any other way, but rather to manipulate his or her behavior such that he or she chooses to act in the manner you desire even though he or she retains the power not to. Coercion succeeds if and when a challenger gives up even though it does not have to. Militarily crushing an opponent to the point that it can no longer muster the ability to act as it wants is not coercion. This is better understood as overt military victory using “brute force.”8 Coercion entails an active choice on the part of the challenger to act in one way and not another. At times, however, practicing coercion may involve going beyond the mere issuance of threats to the actual use of limited military force. Coercion differs from deterrence and compellence in subtle but important ways: It can include limited warfare and limited military use for manipulative or demonstrative purposes. Military strikes may serve to underscore broader coercive threats, to convince a challenger that the defender is serious and has the means to follow through with its threats, and to illustrate that the best course of action is to acquiesce to demands. Coercive air power, for instance, is an attempt to manipulate an adversary’s behavior by strategically bombing targets, rather than bombing an adversary to utter destruction.9

18

Chapter 2

The difference here, as noted earlier, is between manipulation and defeat. Deterrence, however, is behavioral manipulation short of military force. In fact, traditional theories of deterrence suggest a deterrent strategy has failed if a threat must be carried out in practice. But successful coercion often does involve the illustrative and restricted use of military force to both accentuate threats and induce desired behavior. In this sense, coercion is more akin to forceful compellence; indeed, the terms “coercive diplomacy” and “compellence” are often used interchangeably.10 Another useful way to distinguish between these processes is to appreciate how they relate to a particular political or strategic status quo. British scholar Wyn Bowen suggests that deterrence is meant to “preserve the status quo; to prevent an . . . adversary from initiating a course of action,” and that the focus of compellence is based on “altering” an existing relationship or status quo.11 In either case, the objective is to prevent undesired behavior and to coerce an adversary to comply with one’s preferences. It should come as no surprise, however, that compellence is harder to achieve than deterrence; it is far more difficult to persuade an adversary to do something else altogether (compellence) than it is to dissuade it from doing something at all (deterrence). Schelling explains: One of the great advantages of deterrent threats is that they often do not need to be articulated. They are typically addressed to some obvious overt act. If you draw a gun on somebody approaching he may stop whether or not you tell him to; if you want him to turn around, take the keys from his pocket, and throw them to you over his shoulder, you have to say something. Not only does the deterrent threat economize communications and minimize ambiguity, it permits one to create a threat without doing anything belligerent, without acknowledging, even while denying, that one is threatening.12 Compellence requires that a defender convey and communicate the extra steps it desires of its opponent while communicating how a failure to acquiesce to these demands will be met with retaliation and punishment. In practice, deterrence and compellence also vary in how easily they can be observed and scientifically measured. Because a compellent threat requires that a target do something it is not yet doing, it is easier to distinguish the success and failure of compellence than it is of deterrence. With deterrence, nothing happens; with compellence, something does. For social scientists,

Deterrence Theory

19

this fact has caused (and will continue to cause) methodological heartburn. Richard Ned Lebow and Janice Gross Stein argue that testing deterrence theory is problematic “because of the difficulties inherent in identifying deterrence successes, which leave few if any behavioural traces.”13 Or, as Colin Gray adds colorfully, “History does not usually record the dirty deeds that were deterred.”14 In measuring deterrence, it is also difficult to ascertain with certainty that a threat had any effect at all on an adversary’s behavior. It is hard to see something not happening, but even if it were possible, it is even harder to show conclusively that the defender had something (or anything) to do with the event not taking place. To measure deterrence, then, we need to know something about the challenger’s intentions, which has proven difficult to do concretely. In assessing deterrence, we need evidence that illustrates both a challenger’s intent to conduct an attack and a defender’s attempt to deter the attack.15 We would be wrong to interpret every day that an attack does not occur as a deterrent success; countless idiosyncratic reasons that have nothing to do with the issuance of threats can help explain nonevents. In studying deterring terrorism, these methodological hurdles pose unique challenges. Around these core building blocks, an expansive and a cumulative research program on the theory and practice of deterrence eventually developed.16 That maturation was a slow and deliberate process. It took several decades of research for core theoretical principles to be developed, tested, and refined. Three waves of deterrence research mark that progression.17 The process began in earnest in a preliminary, first wave of research that followed World War II. Early theorists were responding specifically to the development of nuclear weapons, illustrating their effect on the study and practice of warfare more broadly.18 Although these early deterrence theorists established the conceptual groundwork for the approaches that were to be developed decades later, their immediate theoretical and political effect on international relations was more limited. It was not until the Soviet Union and its allies eventually emerged as a capable nuclear bloc that Western policy makers and academics were united in one common goal: avoiding any and all nuclear conflicts. In the second wave of research that emerged as a reaction to these shifting international priorities, deterrence theory provided strategists with an abstract framework with which to manage the nuclear rivalry. Th is wave of deterrence research was marked by scientific modeling and the application of game-theory methodology to deterrent relations. Many core concepts that remain in use today, along with the prerequisites

20

Chapter 2

of practicing deterrence, are products of this second wave of research. In a relatively short burst of theoretical creativity, deterrence theory developed into a nuanced area of study.19 And yet, given the characteristic similarities shared by the United States and USSR and their alliance partnerships, the deterrence concepts developed during this period were respectively narrow. As a result, traditional deterrence theory referenced the bipolar setting of the era, applied most specifically to nuclear war and major interstate conflict, was oriented toward preserving the international political status quo, and sought to inform the behavior of mutually rational and unitary state actors. In time, the theoretical and deductive enterprise of the second wave gave way to the empirically driven, third wave of deterrence research. Emerging during the 1970s, the third wave was primarily geared toward evaluating and testing the concepts, models, and theories that had been previously proposed.20 Using both qualitative and quantitative methodology, a substantial research agenda emerged. Deterrence success and failure, it was illustrated, was not only based on an actor’s commitment or resolve (or both) but could also be dictated by the type and nature of the interests at stake and by the costs associated with acquiescing to a threat.21 And although in theory a deterrent was deemed effective when the expected utility of pursuing a given action was less than the expected costs of enduring a punishment, utility itself, it was found, could be measured differently by varied, though equally rational, actors.22 The third wave also tackled some of the psychological processes inherent in deterrent relations, illustrating how individual cognitive characteristics, fear, pressure, fatigue, and other human traits and organizational constraints influenced the manner in which calculations were made and decisions taken.23 An actor’s assessment of costs, benefits, and probabilities could also be influenced by misperceptions and failures to accurately or systematically interpret an adversary’s views, intentions, and positions.24 In addition, culture, values, historical development, and other social and political characteristics were also thought to influence how deterrence could be applied in practice.25 In sum, the third wave provided a more nuanced interpretation of the limitations, scopes, and boundaries associated with applying deterrence theory in practice. Distilling a generation’s worth of theory is no easy task. And yet it is clear that buried within these three waves of research are the building blocks for deterring political violence and terrorism. What follows is a discussion of the prerequisites of deterrence and an overview of the various deterrence iterations

Deterrence Theory

21

as they have been historically applied to state-based relations and as they will be further refined in the context of counterterrorism in the following chapters.

The Prerequisites of Deterrence A subset of the literature on deterrence theory focuses on the theoretical and practical prerequisites. Deterrence theory is best viewed as a game of strategic interaction where a rational actor assesses the benefits and costs of its action on expectations of the likely behavior of its opponent. Scholars Paul Huth and Bruce Russett contend that deterrence logic “must assume that decision-makers are rational expected-utility maximizers” and that actors not only rank order their preferences but choose among the different possibilities based on the “perceptions of the likelihood” of a number of outcomes.26 A deterrent is effective when it produces a cost-benefit calculus on the part of the challenger in which the expected utility of pursuing a given action is less than the expected costs of enduring a punishment. Behavior shifts as a result. Critics of rational deterrence theory note, however, that utility can be measured differently by varied—though equally rational—actors. Lebow and Stein, for instance, argue that some actors “may rationally minimize expected losses while others may maximize expected gains” and that “subjective expected utility will vary depending” on whether actors are “risk-prone or risk-averse.”27 If an adversary is willing to absorb very high costs in pursuit of an objective, the adversary becomes far more difficult to deter. These and other related criticisms rightly assume that deterrence is a psychological process in which misperception can and does happen when trying to accurately (and systematically) interpret an adversary’s intentions.28 At the extreme, decision making approaches irrationality, not because of any innate characteristics of the actor in question but rather because of the processes and strains in which the actor finds him- or herself. At times, humans are frail, such that their decisions become irregular. In terms of evaluating the applicability of deterrence to terrorism—especially suicide terrorism—the question of rationality must be addressed. If a militant is already willing to die and his organization’s modus operandi is based on a deep-seated mythology of rational suicide, are cost-benefit calculations in the manner in which Western governments understand, likely or even possible?

22

Chapter 2

In addition to rational processes, other prerequisites have been identified that address how theories of coercion are put into practice. Canadian academic Frank Harvey offers a four-point model: Potential attacks can be prevented if a defender (1) clearly “defines” the behavior considered unacceptable, (2) “communicates” to the challenger a “commitment to punish violations,” (3) retains the “capability” to either follow through with the promised punishment or deny the opponent its objectives, and (4) demonstrates “resolve” to carry out the punitive action if compliance is not met.29 Each prerequisite relates to the crucial elements of communication and credibility. In terms of communication, deterrence and compellence are very rarely effective if done in secret. This is the premise that informs the plot of Dr. Strangelove, the iconic Cold War–era film. In the movie, the Soviets have developed a doomsday device that automatically results in the detonation of fift y Soviet nuclear bombs if and when the United States attacks the Soviet Union. The device is meant to deter a U.S. first strike. But having failed to reveal its existence, the world accidentally tumbles into catastrophic and unavoidable nuclear war. “The whole point of the doomsday machine is lost if you keep it a secret!” exclaims the title character, Dr. Strangelove. A defender must communicate to an adversary that it is the target of a deterrent policy and outline what action is unwanted. If threats are vague or altogether absent, deterrence or compellence (or both) will fail. An actor must be told, often loudly and repeatedly, that it is the target of a deterrent policy; otherwise, unknowing and uncaring, there is little incentive for it to act any differently. In coercing terrorists, communication may prove especially difficult.30 Paradoxically, at times deterrence theory does appear to be working despite the fact that the defender is not purposefully intending to manipulate or coerce an adversary. For instance, if a defender has “palpable strength” it may not have to do much to convince potential challengers that attacks are not worth it.31 Here, deterrence is at work even though it is not being purposefully applied. Similarly, sometimes deterrence is the unintended outcome of other, kinetic or military operations. This may be especially true in counterterrorism, where a state pursues offensive and defensive measures that result in a dampening effect on militants even though coercion was not the expressed intention. In thinking about deterring terrorism, coercive communication and intent need to be properly parsed. Coercive credibility, however, refers to a defender’s threat and to the believability of that threat among potential challengers. It is where theory

Deterrence Theory

23

meets practice. At its core, credibility is based on a defender’s capability to do as it threatens and is related to its intent and resolve to see threats through if and when compliance is not met.32 During the Cold War and in the case of nuclear deterrence more broadly, credibility was a limiting factor. At times, threatening nuclear retaliation for some unwanted behavior was deemed absurd, even fanciful.33 That is, challengers found some nuclear threats— especially those issued to safeguard peripheral interests or those made in defense of allies—unbelievable. Why would the USSR, for instance, believe an American nuclear threat to protect its European allies if carrying out such a threat would necessarily result in American annihilation by Soviet retaliation? Nobody doubted that the United States had the capability to launch nuclear weapons. Nor did anybody doubt the effect these nuclear weapons would have. They did, however, at times doubt U.S. resolve and willpower to follow through with certain threats. It is important to note that without credibility, deterrence is weakened; coercion will fail if a challenger perceives a threat as void, incredible, or impossible to implement. In state-on-state deterrence, credibility is often communicated through declarations (like policy speeches) or by showing off and signaling military capabilities. Communicating credible resolve, Harvey continues, is “most effectively demonstrated through costly signals . . . that increase the political, economic, or military costs of the challenge.” Signals can take various forms, including action—the mobilization or deployment of military assets; statements—public (and/or backchannel) announcements of retaliation, ultimatums, and deadlines; and environmental conditions—both domestic and international support for the deterrent threat.34 For deterrence to work, then, the challenger must appreciate that a threat has been levied, understand the threat accurately, recognize what action(s) in particular is being challenged, and be sufficiently influenced by the costs and benefits of its actions. Successful deterrence further depends on the existence of a preferred and mutually shared state of affairs between adversaries. That is, behavioral preferences must overlap between challengers and defenders, even if only minimally, and common ground must be a possibility. Schelling explains: Coercion . . . requires that our interests and our opponent’s not be absolutely opposed. If his pain were our greatest delight and our satisfaction his greatest woe, we would just proceed to hurt and frustrate each other. It is when his pain gives us little or no satisfaction compared with what he can do for us, and the action of inaction

24

Chapter 2

that satisfies us costs him less than the pain we can cause, that there is room for coercion. [Deterrence] requires finding a bargain, arranging for him to be better off doing what we want—worse off not doing what we want—when he takes the threatened penalty into account. 35 If revenge, pain, and destruction are the only conditions shared by opponents, then deterrence cannot work. In this case, both sides see open conflict as the best, most desired, and only option. In thinking about deterring terrorism, these prerequisites are challenged anew. Conventional wisdom holds that terrorists are less than rational. Likewise, terrorist leaders have already accepted the utility of using violence against civilians to advance their cause and are likely to be resolute in their behavior.36 And that terrorism can be coordinated anonymously (and is otherwise difficult to attribute) leaves defenders with few targets to punish and even fewer terrorist decision makers to manipulate. Deterrence is further weakened if terrorists use violence for internal (rather than external or strategic) purposes. If militants use violence to enhance group cohesion, to attract new recruits and finances, or to compete with other militants for media and international attention, then deterrence may not work.37 And echoing Schelling’s earlier concern, some terrorists may desire destruction above all else, such that no coercive bargain is feasible because they could not care less about political compromise. They want to fight. Likewise, terrorist organizations usually function as nonstate entities, are active below the state (in terms of domestic terrorism) and above it (in terms of transnational terrorism), and usually lack a well-defined decision-making hierarchy. Communicating threats may prove difficult. And, as noted in Chapter 1, terrorists have little infrastructure to threaten, lack a home address, may be willing “to lose it all” to conduct a spectacular attack, and may actually seek retaliation.38 All of this complicates the manner in which states pursue counterterrorism deterrent threats and weakens the manner in which they will be able to apply deterrence theory to counterterrorism more generally. Finally, coercion is a dynamic process that involves at least two actors. So far, we have focused on the actions, threats, and preferences of the defender. But deterrence is not solely about what defenders can do to challengers; it is also about what challengers can do to defenders in return. In practice coercion rarely ends with a defender’s threat. Challengers, too, can react and behave in ways that affect the defender’s own cost-benefit calculation to pursue

Deterrence Theory

25

a given coercive strategy and can manipulate their willingness and ability to absorb the risks and costs involved. “Counter-coercion” refers to a challenger’s ability to interfere with and potentially diminish a defender’s preferred coercive strategy. As Frank Harvey and I explain in our contribution to Deterring Terrorism, “Counter-coercion theory is derived from assumptions about the relationship between the application of coercive tactics and an adversary’s motivation to find responses to counter those moves and/or exploit them in order to impose additional costs.” It involves coercing a coercer, responding to threats in a manner that influences a defender’s willingness to see those threats through. The idea is to nullify a coercive strategy by imposing costs on the defender. For instance, challengers may escalate a military standoff, seek to reshape the conflict environment, or try to upend a coalition aligned against it, all in hopes of convincing a defender to forgo its preferred coercive strategy. In this case, a challenger increases the costs associated with imposing costs.39 The 1999 Kosovo air campaign provides an illustration of the role countercoercion played in persuading the North Atlantic Treaty Orga ni zation (NATO) to rely on air strikes while forgoing a ground-war option. Western officials believed they could win the conflict on their terms by relying on smart bombs. NATO’s legitimacy and very cohesion depended on it minimizing civilian and allied casualties. NATO’s guiding coercive principles dictated “obvious counter-moves” by Yugoslavian president Slobodan Milošević: increasing the number of civilian and NATO military deaths.40 Milošević attempted to do so with Operation Horseshoe, the alleged plan to forcibly expel Kosovo’s Albanian population. Milošević’s preferences, in turn, informed NATO’s motivation to avoid a ground war that would have played to Milošević’s counter-coercive move. “Coercive strategies,” explain Daniel Byman and colleagues, must be “designed not only to raise the expected costs of resistance but also to anticipate and neutralize possible counter-moves that would otherwise reduce the impact of the coercive threat.”41 In Kosovo, relying on precision munitions became the best (and potentially only) way to circumvent Milošević’s counter-coercive strategy. Terrorists have enormous counter-coercion potential. They can undermine a state’s ability to turn its military capabilities into credible counterterrorism deterrent threats. For instance, a state’s nuclear arsenal is particularly ill-suited for deterring terrorism. Nuclear threats, except perhaps in the most dire of circumstances, are not going to be deemed credible by militant groups. But similar deficiencies may hold even in the case of other, more conventional,

26

Chapter 2

military capabilities, like sophisticated fighter jets and tanks. These weapon systems certainly carry a powerful and frightening punch, but they may be less than ideal for coercing lightly armed, mobile, and diff use nonstate militant groups. The usual refrain is that you cannot chase a mosquito with a sledgehammer. The crux of the problem stems from the difficulty states have in properly translating their traditional offensive military capabilities into counterterrorism deterrents or compellents, and the subsequent ability of terrorists to actively exploit these difficulties. In sum, the prerequisites of deterrence theory suggest that practicing coercion is no simple task. Several conditions, from rationality and preferences to a challenger’s capability, need to be taken into account. This is clearly evident in interstate coercive relations and will prove especially apparent in deterring terrorism. As Chapter 3 illustrates, the practice of deterring terrorism repeatedly bumps up against the prerequisites of deterrence theory. What kind of threats can states communicate with nonstate actors; how will they know if anybody is listening on the other end? Are groups like al Qaeda principally motivated by anger and a desire to make the West suffer, or is terrorism a strategic tool used for political purposes? And how are states going to package their traditional military capabilities into counterterrorism deterrents, and what capabilities in particular will resonate most with militant groups?

Deterrent Iterations: Punishment and Beyond Having outlined the basic building blocks and identified the core requisites of deterrence, the next step is to elaborate on the theory’s various delineations and processes. A deterrent is said to be effective if it produces a costbenefit calculation within an adversary’s decision making in which the expected utility of pursuing a given action is less than its expected cost. Focusing on the costs and benefits of behavioral manipulation leads to a primary fault line within deterrence theory that is of par ticular importance for deterring terrorism: the distinction between deterrence by punishment and deterrence by denial.42 Both processes address an adversary’s cost-benefit analysis but approach it from different ends; punishment adds costs while denial takes away benefits. Deterrence by punishment relies on threatening to harm something a challenger values if it decides to pursue an unwanted action. It is principally

Deterrence Theory

27

about imposing costs on an adversary’s behavior and manipulates decision making by communicating to a challenger that certain behavior will result in retaliation. The threat of punishment is perhaps the most widely understood coercive process, and in interstate relations deterrence is most actively associated with the threat of military retaliation. During the Cold War, for instance, the United States threatened to attack Moscow (a cost to the Soviets) if Moscow were to attack Washington (a cost to the Americans), and vice versa. Both sides were deterred in part because of the costs associated with initiating an attack. With the development of nuclear weapons, punishmentbased deterrence became paramount. A complicating wrinkle associated with the calculus of punishment, however, is the potential costs of inaction on the part of the challenger. That is, acquiescing to a deterrent and not carry ing out an action might also carry various costs. As Gary Schaub argues, some actors do not view “inaction as a neutral outcome” but rather as a second potential scenario that may or may not carry “some sort of loss.”43 The case of Iraq in 2002–3 is revealing. Iraq was involved in a complex, high-stakes, and multidimensional deterrent game. At one end, the United States, with international and UN support, sought to compel President Saddam Hussein to comply with weapons inspectors and to allow them full access to Iraq’s armament facilities. The United States used the threat of invasion to compel Iraq to comply. Hussein refused to cooperate fully. In the language of deterrence theory, U.S. threats of punishment failed to compel Iraq to comply with international demands. American decision makers interpreted Iraq’s evasive behavior as proof that Hussein retained WMD capabilities. Otherwise, why would he risk another round of war with the Americans?44 But Hussein’s behavior prior to Operation Iraqi Freedom was not meant solely for his American audience. At the other end of the coercive spectrum, Iraq was also signaling Iran, Baghdad’s regional and historic rival. By purposefully skirting international demands and by signaling a WMD capability, Hussein was communicating his continued strengths to the Iranians. The associated cost, then, of acquiescing to U.S. pressure and to allowing more invasive inspection of his military capability would have been to jeopardize that second, regional, and potentially more important, deterrent with Iran. The cost of being coerced by one adversary was a loss of status, strength, and security vis-à-vis another. Deterrence by punishment also assumes that inaction (or acquiescence to a threat) on the part of the challenger will be met with inaction on the

28

Chapter 2

part of the defender. This is the bargain that underscores deterrence logic. A defender must communicate to a challenger that “if you act in manner A, you will be punished in manner B; but if you don’t act in manner A, I won’t do B either.” If a challenger expects some form of punishment even after having complied with a deterrent threat, the costs of inaction and action are indistinguishable. Mutual restraint is crucial.45 Deterrence trades inaction for inaction; it only works if threats are paired with promises to refrain from unwarranted retaliation. Without credible assurances of restraint, coercive threats will go unheeded. As will become evident, mutual inaction may prove difficult (if not, at times, impossible) in deterring terrorists. Martha Crenshaw of Stanford University argues that the twin aims of destroying and deterring a single militant adversary are incompatible.46 If al Qaeda is right to assume that the United States and its allies are seeking its ultimate strategic defeat, why would its leaders believe that a deterrent relationship with the United States, in which American punishments are stayed, is ever feasible? The United States has never offered—and, indeed, may never offer— al Qaeda convincing promises of restraint in keeping with a deterrent bargain. But al Qaeda, too, would face similar shortfalls. It has vowed to strike the United States at all costs and interprets its war as a religious obligation. What guarantee do Americans have that al Qaeda will show restraint and hold its fire? None. Al Qaeda may be an especially poor bargaining partner because it may have difficulty communicating, offering, and upholding promises of both punishment and restraint.47 In either case, restraint is found wanting. Deterrence by denial functions by reducing the perceived benefits an action is expected to provide a challenger. Decision making is based on both costs and benefits, so while punishment manipulates behavior by adding to the costs, denial functions by taking away anticipated benefits. For instance, hardening national or infrastructural defenses—what Freedman calls “passive” defenses—raises the costs of attack by diminishing the probability that an adversary is likely to acquire the intended objective. While punishment deters an actor through fear, denial creates a sense of expected failure or “hopelessness” on the part of the challenger.48 The idea in this case is that challengers will back off because defenders have sufficiently shown that they are unlikely to acquire what they seek. Although defense and deterrence by denial are, for planning purposes, inextricably related, the two are conceptually distinct. Defense differs from denial, contends U.S. deterrence expert Glenn Snyder, because it seeks to reduce the “costs and risks in the event

Deterrence Theory

29

deterrence fails,” rather than attempt to influence the behavior of the actor.49 Defense mitigates while denial influences. Nonetheless, the two processes— especially in a conflict with terrorists—are distinct by only the thinnest of margins. During the Cold War, deterrence by denial was generally overshadowed by the threat of deterrence by punishment. In situations in which nuclearcapable rivals face off against each other, denial mechanisms may prove less practical because of the difficulty in credibly neutralizing the consequences of a nuclear exchange. Neither the Soviets nor the Americans could credibly threaten the other with failure when it came to nuclear weapons, long-range bombers and submarines, and ballistic missile technology. There were simply too many nuclear bombs to preemptively eliminate and no feasible way to limit the damage even one nuclear explosion would have caused. Denial, in this particular case, was simply not credible. Punishment strategies— including threats of massive nuclear retaliation, like MAD—largely took over from denial strategies as a result. However, in conventional military environments and especially in counterterrorism, denial mechanisms become more realistic. Consider NATO’s conventional strategy for defending Europe during the early and mid–Cold War, or, more broadly, of the American strategy of containing Soviet expansionism. The battlefield strategy involved denying the Soviets an easy military victory in any attempt to invade and occupy Western Europe or in any attempt to forcefully expand its global sphere of influence. In both cases, a conventional advance by Soviet forces would be met and, presumably, held (or at least stalled) by a smaller NATO force that would weigh on Soviet decision making before and during the conflict. The idea was to deny the Soviets the gains they sought by preventing them from achieving a cheap and quick military victory and to create the impression that any conventional war with NATO risked ultimately leading to tactical and strategic nuclear retaliation.50 By ensuring that any European conflict, for instance, would be a costly one, NATO denied the Soviets the value of initiating a blitzkrieg. In essence, NATO conventional forces—backed by nuclear threats— threatened the Soviet Union with failure (i.e., denial of military objectives). 51 But as Chapter 3 highlights, deterrence by denial has quietly become a cornerstone of deterring terrorism; theories of denial have been exhumed, burnished, and given new import. In a rather stunning reversal of fortune, some theorists even suggest that denial trumps punishment in deterring terrorism.52

30

Chapter 2

Deterrence by punishment and deterrence by denial represent but a fraction of what is captured by the broader logic of coercion. In general, theories of deterrence neglect to properly address the coercive value of positive inducements and rewards.53 So long as deterrence is based on manipulating an adversary’s behavior, then nonpunishment-based processes that provide coercive results are necessarily associated with the logic informing the theory. Consider that offers of reward can induce opponents to change behavior in ways that suit a defender. Paul Huth and Bruce Russett assert that a deterring actor has two primary instruments with which to influence another’s behavior: the threat of sanctions and the offer of rewards. After all, behavior is not only about avoiding a particular punishment but also about deriving a particular reward—the stick and carrot maxim so popular in diplomacy. Huth and Russett contend that “the inclusion of positive inducements as a means to deter is not standard practice in academic writings or policy debates,” despite the fact that rewards “and inducements are logically within the scope of deterrence theory.”54 Theoretically, both punishment and reward might be used to reach a common coercive goal. Nonkinetic deterrents that rely on positive inducements and reassurance, denial and mitigation, and debate and delegitimization, although not generally associated with deterrence, nonetheless function and adhere to the theory’s core tenets (changing behavior) and rely on its logical processes (manipulating costbenefit calculations). Nonkinetic deterrents do not replace traditional ones but rather accompany them into a broader strategy. In asymmetric contests between states and terrorist adversaries, these less traditional forms of deterrence gain a degree of importance.55 Miroslav Nincic has proposed two theoretical frameworks for studying how states use positive inducements to influence “renegade regimes” (in Nincic’s case, North Korea, Iran, and Libya). The “exchange model” provides “specific quid pro quos” to the challenger, supplying it with benefits that offset those it would potentially acquire with “renegade behavior.” The challenger in choosing to accept these inducements, changes its behavior in exchange. Nincic’s second, “catalytic model” is more ambitious, akin to carrot-induced regime change where positive engagement fundamentally alters the preferences of the challenger such that its behavior is influenced over time.56 In both cases, nonpunishment-based coercion is meant to produce adversarial behavior that better suits a defender. In a somewhat similar fashion, Ted Hopf, in his evaluation of Great Power credibility in Third World conflicts, finds that nonpunishment influences and inducements played an important role in

Deterrence Theory

31

determining coercive successes. He concludes that deterrence theory must “expand its scope to capture an array of deterrent instruments that the theory’s focus on military tools omits.”57 Hence, although threats are a central feature of deterrence theory, punishment and denial are but a part of what is theoretically and practically pertinent for constructing coercive relations between actors. In thinking about influencing militants, these less traditional forms of coercion gain in importance. In addition to punishment, denial, and positive inducements, deterrence has been further subdivided in other important ways. For instance, the classical literature on deterrence theory pits central deterrence against extended deterrence. The former involves deterring attacks on one’s own territory or core interests while the latter involves deterring attacks on one’s allies.58 During the Cold War, the United States sought to safeguard the security of its European, Asian, Middle Eastern, and South American allies by threatening the Soviet Union and its allies on their behalf. In this case, the United States used its nuclear and conventional threats to deter Soviet aggression against it (central deterrence) and against its allies (extended deterrence). Extended deterrence is more difficult to acquire than direct, “self-protective” deterrence because challengers are more likely to doubt that costly threats will actually be implemented in response to an attack on a third party.59 Here again, the issue is credibility. Will defenders be willing to absorb exceptionally high costs (including, in the case of the Cold War, nuclear retaliation) in exchange for the security of an allied state? The question posed during the Cold War was whether the United States, in extending its nuclear deterrent to its European allies, would willingly trade Washington for Bonn in a faceoff with the Soviet nuclear arsenal. Some doubted the trade-off altogether. Another subtheme focuses on the concept of intra-war deterrence and the various distinctions inherent to it: narrow, broad, immediate, and general deterrence. Deterrence theory does not disappear once a crisis tumbles into a war. Rather, it skips into another realm: manipulating behavior within a war, hence intra-war deterrence. So although deterrence may begin in peacetime, it also functions in wartime. At its core, intra-war deterrence suggests that even in a context of an ongoing conflict, adversaries can use threats to influence their opponent’s behavior and shape the nature of their military engagement. Its basic premise, Jack Snyder explains, is to develop “damage limitation strategies” during a war such that there exists “mutual restraint in the choice of targets and weapons” among all warring parties.60 Intra-war deterrence tackles the manner in which a conflict is conducted.

32

Chapter 2

During the Cold War, the question posed to scholars and practitioners was how the superpowers might limit their war-fighting behavior in a way that would offer both some control over escalation. Had the United States and the USSR fought a conventional war in Europe, for instance, could that war have been waged within limits (i.e., without the use of nuclear weapons against European targets)? And if nuclear weapons were indeed used in Europe, what would it take to limit that exchange to the Continent and deter the use of nuclear weapons against either the American or Russian homeland?61 Much of the theoretical work on deterrence that took place during the early Cold War was based at least partially on the notion of intra-war deterrence. For instance, the debate on the value, feasibility, and practicality of using counterforce nuclear strikes against the USSR took place against the backdrop of an imagined military conflict. Counterforce deterrent strategies, as they were developed in the 1950s and 1960s, involved targeting, striking, and (hopefully) destroying the Soviet Union’s military assets (and, notably, its long-range nuclear delivery systems) while purposefully avoiding targeting its cities and population centers. This was not based on any ethical consideration but rather on a strategic and deterrent consideration. The cities were left as hostages in hopes of deterring Moscow, which was suffering the destruction of its military, from contemplating retaliation.62 The coercive message was that further, even more painful punishment, awaited. The logic of intra-war deterrence is partially built on the distinction between narrow and broad deterrence and between immediate and general deterrence. Freedman explains that “broad deterrence involves deterring all war,” whereas “narrow deterrence involves deterring a par ticular type of military operation within a war.”63 In essence, broad deterrence stops a conflict from being initiated while narrow deterrence shapes the nature of a war once one is initiated; the latter begins in cases where the former has failed. Narrow deterrence is about forcing restraint within a confl ict; it involves communicating to an adversary the specific actions that are beyond consideration (i.e., nuclear escalation) or more specific behaviors that would result in an especially strong response (i.e., the killing of prisoners of war or a geographic expansion of a given military engagement). Consider that with the advent and use of chemical weapons during World War I, a narrow deterrent goal in most future conflicts involved deterring the use of chemical weapons on the battlefield. Chemical weapons did not disappear and states continued to go to war, but the narrow deterrent objective, in some cases, was to deter an opponent one was trying to defeat from using those weapons in battle.

Deterrence Theory

33

The same could be said of nuclear weapons after 1945. Indeed, narrow deterrence—and the logic of intra-war coercion more broadly—is often applied to fighting a war within the shadow of nuclear, biological, and chemical weapons. For instance, in a counterfactual analysis exploring U.S. and NATO options for addressing a nuclear-capable Iraq following its invasion of Kuwait in 1990, Barry Posen writes that the “coalition’s strategy must be to wage a conventional war with Iraq while deterring its resort to nuclear weapons”; that is, roll back Iraq’s military advances and defeat Iraq on a conventional battlefield while simultaneously deterring it from launching a nuclear attack during the conventional war.64 Doing so would have carried risks, of course, but Posen lays out the manner in which the United States could have communicated and signaled these narrow deterrent objectives while using Iraq’s actual conduct of the 1990–91 war and its refrain from launching its chemical weapons against Saudi Arabia, Israel, and advancing Western forces to inform his analysis. Immediate deterrence is usually pitted against general deterrence, with each sitting at one end of the spectrum. The coercive goals inherent to each concept and the parameters under which threats are carried out differ widely. In general deterrence, adversaries who have the capability to inflict harm on one another do not; instead, they use open knowledge of these capabilities along with vague threats to regulate, pacify, and shape their relationship. It is similar, then, to broad deterrence in that it offers an overarching and a strategic guideline for interstate relations. No actor is actively thinking about initiating a crisis, launching an attack, or mounting an offensive campaign, though everybody appreciates that these sorts of developments are possible, so deterrence informs their general interactions. Immediate deterrence, however, is much more pressing.65 For Pat Morgan, immediate deterrence is a coercive situation in which one side of a crisis or contest “is seriously considering an attack while the other is mounting a threat to prevent it.”66 There is a degree of imminency in this sort of situation; a challenger has taken steps toward altering the status quo and a defender takes actions to sway his decision to follow through. The coercive goal is to stop an adversary from carry ing out actions that it is actively contemplating and for which it is preparing. Some scholars present situations of immediate deterrence as “asymmetric” in nature. Here, asymmetry refers to diverging levels of offensive motivation between challengers and defenders (not to be confused with contemporary uses of asymmetry that suggest diverging levels of military power).67 On the one hand, in

34

Chapter 2

immediate deterrence, challengers have a disproportionately higher degree of motivation or intent to engage in warfare than do defenders. The deterrent is meant to add costs to seeing that motivation through to the end. On the other hand, in general deterrence, all parties are motivated to ensure that the current relationship remains intact and that adversarial motivations might therefore be considered symmetric in nature. In thinking about applying deterrence to terrorism, the concept of intra-war deterrence plays an important role because coercion take place within an ongoing security environment in which terrorist attacks and counterterrorism operations remain a distinct possibility. The topic is taken in up in subsequent chapters. Finally, the concept of cumulative deterrence has been proposed, and refined more recently for application to counterterrorism, to describe how deterrence might work over the long term and within enduring, iterated, and protracted contests between states.68 Using Israel’s experience, Uri Bar-Joseph explains: “Cumulative deterrence . . . aims at convincing the Arab side that ending the conflict by destroying the Jewish state is either impossible or involves costs and risks which exceed the expected benefits.” The coercive objective is to figuratively bank Israel’s previous military successes, along with general perceptions of its conventional and nuclear capabilities, and use that accumulated wealth of strategic might to persuade adversaries that further violence is altogether futile.69 The concept weds aspects of punishment with denial such that past Israeli military victories add credence to future coercive threats, while repeated Arab failures build momentum toward perceptions of an inevitable defeat and denial of objectives. The concept has proven both particularly relevant to the practice of deterring nonstate adversaries and exceptionally controversial among traditional deterrence scholars. So although Doron Almog, Shmuel Bar, Dima Adamsky, and Thomas Rid (who uses a criminological rather than security studies perspective) have borrowed and adapted Bar-Joseph’s concept for contemporary counterterrorism, critics have responded fiercely. Janice Gross Stein, for instance, concludes that “the concept of ‘cumulative deterrence’ so badly violates the fundamental meaning of deterrence that it loses any analytical value. That kind of strategy is better described as—war.”70 Crenshaw adds: “Almog calls for a policy of ‘cumulative deterrence’ based on a gradual wearing down of the adversary through offensive and defensive measures, . . . but it is not clear how continuous pressure and use of force can be defined as deterrence. Deterrence by retaliation involves a threat from the defender to do something new, not to

Deterrence Theory

35

continue a course of action. The defender does not say ‘we will keep hitting you over the head with this hammer so you don’t even think of attacking us’ but rather ‘we are not hitting you now but will hit you really hard later if you cross this line.’ ”71 And Tel Aviv University professor Amir Lupovici suggests further that relying on “repeated retaliation” to teach terrorists the principles of deterrence by a “cumulative learning” process might instead “lead the terrorists to the exact opposite conclusion—that deterrence fails over and over again.”72 Yet, the concept nonetheless describes Israel’s particular experience with deterrence rather well.73 And in the specific case of deterring nonstate militants, insurgents, and terrorists—where Cold War notions of deterrence success and failure no longer apply as crisply and cleanly as they once did—cumulative deterrence appears persuasive. Cold War deterrence success was absolute: Either an attack took place or it did not. And deterrence failure occurred once a defender crossed the line from issuing threats to acting on those threats. In counterterrorism, however, partial coercive success is possible. In our concluding chapter to Deterring Terrorism, Swiss scholar Andreas Wenger and I suggested that the “broadened conception of deterrence” usually applied to counterterrorism may necessitate the development of a “sliding scale of deterrence success.”74 Again, deterring all terrorism may prove impossible, but some coercive successes—at the “fringes” of terrorist behavior, against particular actors within the terrorist organization, or against particular forms of political violence— should not be discounted even in the event new terrorist attacks take place. At the same time, some scholars note that the use of military force in counterterrorism does not necessarily signify a failure of coercive intent because force is used as an instrument to underpin, establish, maintain, and strengthen de facto norms of behavior vis-à-vis violent militant adversaries. By reaching beyond the international relations literature on deterrence, Rid has been particularly effective at squaring these conceptual circles. “The mere prospect of nuclear annihilation,” he explains “contaminated the West’s conceptual, historical and operational understanding of deterrence.”75 In studies of sociology, law, criminology, and human behavior, these notions of punitive deterrence, coercion, and influence are commonplace and much less controversial. In tackling nonstate adversaries, they also seem particularly relevant. Ultimately, deterrence theory “is less brittle than sometimes thought”: Its theoretical propositions are contingent. If anticipated punishments exceed prospective gains and if other prerequisites are met, then adversaries will be coerced.76 Put more forcefully, the theory predicts cases where deterrence

36

Chapter 2

will collapse and fail. And yet, deterrence failure should not be equated with failure of deterrence theory. Indeed, deterrence theory is far more flexible than is generally assumed; its basic tenets can be molded and refined such that they then apply to less traditional adversaries and novel security concerns. In the next chapter, traditional notions of deterrence theory are broadened and applied to terrorism and counterterrorism.

CHAPTER 3

Deterring Terrorism: Contemporary Debates

Today we are in the middle of a “fourth wave” of deterrence research.1 Emerging at the end of the Cold War as a result of the collapse of the Soviet Union, the focus of this wave is directed toward mapping the contours of deterrence theory against a backdrop of novel and often asymmetric threats. In recent years, the logic of deterrence has been expanded, its application widely broadened. Research on the subject has narrowed in on asymmetric opponents, nuclear proliferants, and “rogue” states, suggesting ways in which the tenets of Cold War deterrence might be reshaped and reapplied to address contemporary security dilemmas.2 Violent nonstate actors are increasingly under the deterrent microscope, with terrorists, insurgents, militants, criminals, and even pirates serving as the primary focus.3 Deterrence theory has likewise entered the cyber world, with researchers outlining the promises and pitfalls of applying the approach to address cyber-attacks, cyber warriors, hackers, and other Web-based miscreants.4 Deterrence has also been turned inward, with research suggesting ways to deter domestic upheaval, rebellion, and violent protest.5 In addition to the growing list of contemporary targets, the practice of deterrence has also garnered some fresh, new input: Deterrence by denial is climbing out from behind the nuclear shadow;6 deterrence by delegitimization is being developed;7 extended deterrence is getting another inquisitive look;8 deterrence through religion and strategic culture has been proposed;9 and other concepts, like influence, dissuasion, and inducement, are being developed.10 The theory and practice of deterring terrorism makes up the largest subset of this fourth wave of research. And yet, as in the early phases of the Cold War, the theory and practice of deterring terrorism remains in its infancy. Core concepts are only now being developed and very few studies have taken

38

Chapter 3

empirical steps to test and refine novel theoretical propositions.11 The intention of this book is to address both of these shortfalls; the intention of this chapter is to introduce readers to the theory and practice of deterring terrorism in par ticu lar. The chapter begins by exploring two innovations developed and proposed by deterrence theorists and terrorism experts alike. The first calls for an expansion of the theoretical foundations of deterrence, a broadening of its meaning and application so that it can address novel, complex, and asymmetric threats. The second involves conceptually unpacking terrorism into its parts and processes, identifying where and how exactly deterrence might be applied to counterterrorism. Together these innovations allow us to properly pinpoint how different theories of deterrence can be properly applied to terrorism and terrorists. From there, the chapter offers an overview of the three core concepts of deterring terrorism (punishment, denial, and delegitimization) and concludes with a brief overview of the empirical work that has been carried out on the subject.

Expanding the Logic of Deterrence Applying deterrence to terrorism requires expanding the theory’s scope beyond its current focus on punishment and nuclear weaponry.12 Deterrence is a broader concept. Although (quasi-)critics like Paul Davis and Brian Jenkins, both at RAND, initially argued that “deterrence is the wrong concept” to contend with terrorism because it is “too limiting and too naïve,” they submit that “broadening the concept” and adding “an influence component” offers a way forward.13 The authors’ distinction between influence and deterrence is a little disingenuous, however. Deterrence is all about influencing opponents (broadly speaking) and is not merely based on levying threats. To set deterrence up in minimalist garb and then conclude that the approach cannot be applied to counterterrorism because it is too narrow in scope is a red herring. As David Johnson and colleagues concede, “The most useful definitions of deterrence are broad ones, which acknowledge that such actions as aggression may be deterred by many means.”14 Most scholars evaluating deterring terrorism agree, basing their investigations from an initial position that expands the meaning and applicability of deterrence. For some, like nuclear-weapons expert Lewis Dunn, influence encompasses the concept of deterrence but is intended to be applied in a way that goes beyond simply relying on punishment or denial.15 Traditional conceptions of deter-

Deterring Terrorism

39

rence matter, but Dunn’s assumption suggests that an adversary’s behavior can be manipulated in a variety of ways. To that end, Sir Michael Quinlan offers a solid conceptual starting point. His position is based on a simple understanding: Deterrence is rooted, elementally, in human affairs, applicable to all types of sociopolitical collectives, ranging from the state to the terrorist group. He writes, “Deterrence arises from basic and permanent facts about human behaviour: that in our decision-making we . . . take into account the probable consequences of our actions; that we refrain from actions whose adverse consequences seem . . . likely to outweigh the beneficial ones; and that we exploit these universal realities as one means of helping to influence others against taking action that would be damaging to ourselves.”16 For counterterrorism, the most helpful interpretations of deterrence are the broad ones that suggest a range of militant activity—from funding extremist literature to recruiting and indoctrinating new foot soldiers, and from coordinating suicide attacks against civilians to acquiring and using CBRN weapons—can be influenced by a variety of tailored deterrents. Deterring terrorism is not a one-size-fits-all scenario. And it is not that deterrence theory and practice have proven ineffective in counterterrorism, but rather that more limited, specifically defined notions of the theory—like nuclear deterrence and (for some) deterrence by punishment—may have. So as Israeli major general Doron Almog contends, although classical deterrence theory “is inapplicable to the war on terrorism,” broader concepts of deterrence—Almog’s own cumulative deterrence included—nonetheless remain relevant.17 The critical distinction is between throwing out deterrence theory in its entirety or expanding its logical boundaries in order to challenge certain aspects of the theory while developing, testing, and championing other theoretically related components. In practice, broadening deterrence in counterterrorism requires expanding the traditional values and assets associated with state-based deterrent relations, like sovereign control, territorial integrity, sociopolitical and economic survival, and so on. Although most terrorists will place little import on these values, they will cherish others, like publicity, operational and tactical success, strategic and tactical success, social cohesiveness, trust and camaraderie, popular sympathy and acceptance, religious legitimacy, prestige, personal glory and martyrdom, freedom of movement, functioning safe havens, and wealth and other material assets.18 So even though many terrorists lack territorial assets against which traditional threats can be issued, they

40

Chapter 3

do have other values that can be threatened. However, because some of these values rest outside the realm punishable by military force alone, nonkinetic deterrent instruments that target these nonphysical assets must also exist. Once uncovered, less traditional forms of deterrence that go beyond punishment-based strategies can be used. To that end, deterrence is applied to counterterrorism in all of its varied subtleties and complexities, from classical deterrent notions of coercion, threat, and punishment to more nuanced versions of tactical denial, societal influence, and religious or ideological persuasion. The logic of deterrence, however, remains the same; it is the practice that has evolved over time, in concert with changing security environments. As a result, the very notion of deterrence has been expanded, tailored, as it were, to the complexities borne of targeting disparate, diffuse, and dangerous substate entities.

Understanding Terrorism: Deconstruction and Rationality In the years since 9/11, the study of terrorism has received detailed academic treatment. Few other subjects have received as much funding and scholarly attention as terrorism; a lively debate and expansive literature exists.19 Within this literature, two conversations have had a profound effect on deterring terrorism. First, authors have conceptually unpacked terrorism, terrorist groups, and terrorist networks into their anatomically distinctive parts. From the holistically abstract—capital-T Terrorism—we get the distinctive functional parts of the terrorist system, the strategic leaders, religious leaders, financiers, recruiters, bomb makers, foot soldiers, and so on. By breaking the phenomenon of terrorism down and reducing it into a process of component functional parts, an otherwise complex puzzle becomes an altogether comprehensible, multilayered, and dynamic system. Like all social groups, terrorist organizations are human organizations with very real human problems, fears, and vulnerabilities. “Relativists do not understand the depths of their error,” warns Chris Harmon, “when they pronounce that ‘terrorism is just a word for violence we don’t like.’ . . . Terrorists are living, breathing men and women using vile but calculated means to make political gains.”20 Stripping away the nonsense offers us some relief and clarity. In one sense, terrorist groups might be conceptually approached from a business-model perspec-

Deterring Terrorism

41

tive: They are semi-transparent boxes with evident, though fuzzy, borders, delineating internal mechanisms from external environments; they are made up of internal processes held together by various ideational bonds and functional roles, purposes, and goals; they are sensitive systems that react both to inputs (the flow of tactical, monetary, and political support; community and global perceptions; and counterterrorism measures) and provide and produce outputs (terrorist plots, propaganda, fear, and violence). Systematically unpacking the phenomena provides an image of terrorism in which component parts of the terrorist system function semiautonomously from one another; it is the sum of their combined activities that results in organized political violence. And there is analytical value in disaggregating terrorism so as to better focus on specific parts or processes. Marc Sageman in Understanding Terror Networks, for instance, offers a detailed assessment of the global Salafi movement, analytically breaking the whole into its various parts. He suggests that the global jihad movement (or “social network”) is organized as follows: The system’s central staff acts as an organizing hub that has countless connections with other important hubs (“Core Arab,” “Maghreb Arab,” and “Southeast Asian” clusters) and less important nodes (smaller, less connected clusters). Connections between hubs, clusters, and nodes are afforded by lieutenants who ensure that communication with the central staff and between lesser clusters are retained. These lieutenants (usually individuals, though a group could also act as a link) are likewise connected to field commanders in charge of specific terrorist-supporting nodes in distinct geographic locales. These nodes usually look after their own affairs (financing, recruitment, reconnaissance, training, weapons procurement, and so on) but can seek advice and assistance from other clusters. Individual nodes, when led by charismatic and ambitious leaders, might eventually develop capabilities and grow in size as they attract new members and resources. The best nodes eventually transform and evolve into “small-world” networks that have a similar organizing principal as the broader network but at a local or regional level. These nodes might include al Qaeda’s various franchises in Iraq, Syria, North Africa, Yemen, Lebanon, Syria, Somalia, and elsewhere. Sageman concludes that the jihadi network organization—unlike the hierarchical organization of Leftist militant groups of the 1980s, some modern Arab groups (like Hezbollah), and other Islamist groups (like Jemaah Islamiyah)—is nearly immune to fragmentation as a result of targeted killings because its thick web of interconnectivity creates

42

Chapter 3

leadership redundancies.21 Thus, even the elimination of one hub (even a substantial one) will not eliminate the network, though it could degrade its ability to function properly. In addition to anatomizing terrorist groups, other experts decompose the very act of terrorism into its organizational phases. Here, terrorist acts are considered processes, or as “series of actions” that result in violence, rather than as single phenomena.22 Michael Powers, an early proponent of subdividing terrorist attacks into their component processes, writes that “terrorism is not just the moment at which a car bomb detonates or when salmonella bacteria sprayed on a salad bar begins to sicken restaurant patrons. Each of the steps in the . . . terrorist process—from group formation to weapons acquisition to use—is an act the terrorist can be deterred or dissuaded from taking.”23 The terrorist group is made up of leaders, indoctrinators, and foot soldiers; each act of terror is itself the result of a process of recruitment, training, and planning. Each phase of the process must take place if an act of violence is to have any success. Consider then that Mohammed Atta—the 9/11 ringleader—began laying the groundwork for that attack in Hamburg, Germany, in 1999. The idea of 9/11, however, arose many years before then within al Qaeda’s central command in Afghanistan, Sudan, and the Philippines. Sageman recounts that the 9/11 “genesis” was Abdul Hakim Murad, a little-known (and lowlevel) terrorist who had trained as a pilot in the United States in the 1980s. Murad “dreamed of using airplanes as weapons,” recounts Sageman, “fi lling them up with explosives and dive-bombing into the Pentagon or the Central Intelligence Agency Headquarters.”24 Sounds familiar. Murad became a member of an al Qaeda terrorist node in 1993 (led by his childhood friend, Abdul Basit Karim) where his ideas were eventually transmitted to Karim’s uncle, Khalid Sheikh Mohammed (KSM), renowned al Qaeda senior leader and, according to the 9/11 Commission Report, the “principal architect” of the 9/11 attacks. KSM was fascinated with the idea and, in 1996, proposed Murad’s plan to Osama bin Laden. Reportedly, bin Laden responded, “Why do you use an axe when you can use a bulldozer?” Instead of a single, explosive-laden plane, bin Laden argued, why not hijack several jets simultaneously and use them as airborne bombs against a number of targets?25 Three years later, Atta was dispatched to Europe to flesh out the plan. After the Hamburg phase, it still took Atta and his associates quite a bit of time to arrange their finances, visas, accommodations, and flight lessons; select their targets and specific dates; recruit other individuals; test their plans; and so on; until they even-

Deterring Terrorism

43

tually carried out the attack.26 At each stage an opportunity existed to detect, deter, repel, or scuttle the plan. If each unit actor involved in each phase of a terrorist plot is open to the logic of deterrence, then the possibilities for applying deterrence to counterterrorism expand exponentially. We need only to change the question from Can terrorism be deterred? to What individual elements of the terrorist system might be coerced from specific activities? or What unit actor or subphase is sensitive to what form of coercion? As Davis and Jenkins conclude, “Even the most dangerous elements in a system may be deflected from one mode of activity to another or from one set of targets to another.”27 For scholars interested in deterrence theory, decomposing terrorist groups and terrorist acts becomes much more than simply opening the black box of the phenomena. Thinking of terrorism as an “organizational phenomenon” allows us to identify precisely where deterrent leverages might be logically applied to the terrorism puzzle while also outlining how these leverages might be specifically tailored against distinct pieces of the whole.28 Understanding how the pieces all fit together gives us a better appreciation of how deterrence logic might be applied successfully. Second, terrorism experts have made headway in exploring and assessing militant and terrorist rationality. In terms of deterring terrorist organizations, especially those suff used with religious fanaticism, rational actor critiques are an important remonstrance. As the previous chapters highlight, a repeated challenge to deterrence since 9/11 has been the perceived irrationality of contemporary (super)terrorist organizations. But more prescient interpretations of terrorism suggest that it is used by “rational fanatics” seeking certain goals and objectives.29 A number of studies by American scholar Robert Pape, for instance, suggest that suicide terrorism in particular follows a pattern of strategic and coercive logic. It is used both to inflict current pain and to communicate a threat of “future pain” such that terrorist demands are met.30 The tactic and strategy is designed to acquire political goals, to compel a government to shift policies, or to wrest political control from a government (or an occupying force) altogether. What is more, Pape suggests that campaigns of suicide terrorism, when conducted against democracies, have a track record of success; democratic governments are likely to capitulate to at least some militant demands. Mia Bloom’s parallel work on suicide terrorism further supports terrorist rationality. She finds that organizations using the tactic do cost-benefit analyses, and, like our own unpacking of terrorism, Bloom differentiates between the suicide operative

44

Chapter 3

and the organization prompting and preparing their efforts.31 At both levels, rational decisions inform behavior. Terrorist organizations like al Qaeda have priorities and their leaders appreciate what it will take to achieve success. The violence used is measured to effect change in antagonistic communities and to generate support within friendly ones. Furthermore, al Qaeda has proven its capacity to learn from and reflect on tactical errors and incorporate new strategies to evolving security environments.32 Al Qaeda documents recovered by French forces in Mali in 2013, for instance, reveal disagreement between militant leaders as to how fast sharia law and punishment should be imposed on the general population under al Qaeda’s control. The fear was that by moving too quickly, al Qaeda might face public backlash, something it suffered in Iraq and elsewhere. The jihadist project in Mali, the document reads, is “a baby in its first days that is still crawling.”33 Better to be tender and see it grow. Furthermore, what militant groups may lack in military sophistication, they make up for with innovation (think of shoe bombs, liquid bombs, cartridge bombs, and underwear bombs). Terrorists also identify and take advantage of their adversary’s weaknesses. And leaders are often tech-savvy propagandists. AlShabaab’s Twitter feed, for example, went into overdrive in 2013, first during France’s lightning-quick intervention into jihadist-occupied Mali and later during al-Shabaab’s brutal attack on the Westgate Mall in Nairobi, Kenya, in which more than sixty people were shot and killed. Al Shabaab’s tweets were used to inform, coerce, and terrorize its various audiences. Finally, and perhaps most important, Islamist terrorists have constructed their goals to coincide with the perceived grievances of a transnational community. In so doing, they have all but guaranteed themselves a structural resilience that goes beyond their limited membership, along with an expansive base from which to regroup, recruit, and rebuild. This is exactly what al Qaeda elements have done in Syria as a result of the civil war. None of this negates rationality; quite the opposite. All terrorist groups are social organizations and though many may speak in religious terms, they nonetheless strive for politically defined objectives. The portrayal of Islamist terrorism by Fawaz Gerges is especially enlightening: Religion is a jihadist’s “tool for political mobilization.” Power, rather than religious conviction, feeds the motivation. Likewise, Islamist political experiments in Afghanistan, Iran, Syria, Yemen, Iraq, Mali, Sudan, Gaza, and elsewhere are distant cousins of secular authoritarian states: They are “clothed in Islamic dress” but built on traditional political constructs.34 And even in the case of suicide terrorism,

Deterring Terrorism

45

where the militant is not only prepared to die but willingly accepts his or her death as part of the operation, rationality is often in play. Both the suicide bomber and the organization planning the operation may be sensitive to costs and to countermeasures that alter their odds of success. In sum, these arguments are oddly comforting: If contemporary terrorism is a political rather than a religious phenomenon and follows rational motivations, then the United States and its allies have the right strategic toolbox to manage emerging threats.

Deterring Terrorism: Punishment, Denial, and Delegitimization When used in tandem, these conceptual insights—terrorism as a system, terrorists as rational actors, and deterrence as flexible theory—allow for the development of robust and logically coherent arguments for applying deterrence theory and practice to counterterrorism. Dozens of different actors involved in the terrorism system can be targeted with a variety of specifically tailored coercive leverages. What follows is a discussion of deterring terrorism by punishment, denial, and delegitimization.35

Punishment: Targeting What Terrorists Value

Punishment strategies offer one avenue for coercing terrorist behavior. Like state-based deterrence, the goal is to influence terrorist behavior by promising to punish something the group and its supporters value. How punishments are framed and communicated will depend on the target in question. We can distinguish between strategies that are tailored against state sponsors, societal supporters, specific militant groups, and individuals involved in terrorism. In each case, the coercive objective is the same: deterring terrorism. But the individual challengers against which coercive leverages are applied, change. Unlike the state-centricity of Cold War coercion, using punishment to deter terrorism implies influencing different actors involved in political violence in different and unique ways. To begin with, state sponsors of conventional or of CBRN terrorism can be coerced in traditional manners, with threats of military intervention and other diplomatic and economic punishment (for example, UNSC resolutions

46

Chapter 3

and unilateral and multilateral sanctions).36 This is a form of indirect deterrence, in which punishment is threatened against a state in order to compel it to refrain from assisting a terrorist group. This sort of interaction is not new; states have carried out limited retaliatory strikes against state-sponsors of terrorism since well before 9/11. Israel has historically (and repeatedly) used threats against Lebanon, Egypt, Tunisia, Jordan, Sudan, Syria, Uganda, and the territories controlled by the Palestinian Authority (PA) in hopes of influencing their respective relationships with the Palestine Liberation Organization (PLO), Fatah, Hamas, Hezbollah, and other militant groups. The United States launched precision attacks against Libya in 1986, Iraq in 1993, and Sudan and Afghanistan in 1998 in response to terrorist attacks. Turkey threatened to invade Syria in 1998 for harboring members of the Kurdistan Workers’ Party (PKK) and has in recent years launched a number of military raids into northern Iraq to hunt down Kurdish militants. In March 2012, Ethiopia sent its military into neighboring Eritrea to destroy militant bases used to train operatives for attacks in Ethiopia and Somalia. What is new, however, is the importance this method of coercion has been given in recent years. Consider that the 2011 National Military Strategy of the United States of America states explicitly that “[w]e will adapt deterrence principles to our efforts in countering extremists . . . [and] hold accountable any government or entity complicit in attacks against the United States or allies to raise the cost of their support.”37 But as in conventional deterrent relations, coercing state sponsors of terrorism is complicated. Determining whether punishment-based coercion on state sponsorship will succeed depends in part on what Shmuel Bar has dubbed the proxy– patron relationship: the extent and nature of the affi liation between the state and the terrorist organization.38 The higher the degree of affinity between a terrorist organization and a state, the more difficult coercion may be to apply in practice. In states that share a common interest, goal, or ideology with their nonstate proxies, the costs of acquiescing to compellent threats to end that support will remain high. However, deterrence by punishment will prove useful when proxy-patron affinity is low and few mutual interests are shared between hosts and organizations. These cases usually involve passive state sponsors of terrorism—defined by Georgetown University professor Daniel Byman as states that “knowingly allow” terrorists to “raise money, enjoy a sanctuary, [and] recruit” in their territory but do not directly assist them in organizing terrorism.39 Although host states may derive security dividends by tolerating an organization’s activity—through the harassment of a mutu-

Deterring Terrorism

47

ally antagonistic neighbor, for instance—they nonetheless refrain from sponsoring terrorism directly. Under such conditions, a credible punishment may compel hosts to cease their passivity and clamp down on groups active on their territory. Comparatively speaking, Iran or Pakistan, for instance, may have more to lose in ending their respective relationships with Hezbollah and various Kashmiri militants (which Tehran and Islamabad depend on for a variety of tactical and strategic purposes) than Syria did for ending its relationship with the PKK in the late 1990s. Turkey endured attacks by the PKK based in Syria, which facilitated but did not overtly sponsor their activity. Proxy–patron affinity was minimal. When Turkey mobilized its forces on Syria’s border and threatened invasion, it influenced Damascus’s utility calculation concerning its limited support for the PKK. Rather than absorb the punishment, Syria expelled the PKK’s Abdullah Ocalan, leading to his 1999 capture in Kenya by Turkish authorities. Syria may have derived some benefit from the PKK’s harassment of Turkey, but its affinity to the group and its willingness to endure Turkish punishment were relatively low. Likewise, how dependent terrorists are on a state will influence how well coercion against that state will deter terrorism. Terrorists that rely heavily on state support for access to territory, arms, logistics, and training can—if the right coercive leverages are found and applied—be influenced because the state patron itself has some control over the terrorists’ activities. In cases of truly stateless terrorism or in cases where the state itself is weak, failing, or nonexistent (as in Somalia), coercion against state sponsors will provide poor results for deterring terrorism by punishment because even if the state wanted to acquiesce to threats, it might not have the ability to do so. Boaz Atzili and Wendy Pearlman posit that what counts are the domestic strengths of the challenger—defined here as “institutional capacity, domestic political legitimacy, coherent policy-making, and territorial control”—which determine whether a challenger is actually able to acquiesce to a defender’s coercive threats to rein in terrorist groups active within their borders.40 There is another, though less common, scenario. Occasionally terrorists effectively usurp the decision-making apparatus of their host state, whereby their interests inform those of the government and their power enforces its expression in state policy. Bar calls it “reverse proxyship.” Though the state continues to exist, its decision-making power is (partially) controlled by substate actors. For example, between 2005 and 2007, the Islamic Courts Union—a precursor of today’s al Qaeda–affiliated al Shabaab—partially drove

48

Chapter 3

Somalia’s domestic (and foreign) policies. The PA, under Yasser Arafat, occasionally took cues from Hamas, and Syria increasingly looks toward Hezbollah for guidance and assistance. In these cases, state-based punishments may fail to provide the necessary incentive that compels a state to cease facilitating terrorism because the state in question cannot comply. The paradox of reverse proxyship, though, is that the more statelike a terrorist group becomes the more effective threats against it directly are. That is, the higher the degree of “stateness” (centralized leadership, political aspiration, civil involvement, territorial control) the more palpable coercive punishments become. Israel’s uncompromising blow against Hezbollah (a veritable state within a state) in 2006 helped deter the organization from again attacking Israel in solidarity with Hamas, as the Israel Defense Forces (IDF) turned its attention on Gaza militants, first in 2008–9 (Operation Cast Lead), then in March 2012 (Operation Returning Echo), and again in November 2012 (Operation Pillar of Defense). Tellingly, just months before the 2008–9 conflict, Gadi Eisenkot of Israel’s Northern Command warned Hezbollah that “what happened in the Dahiya quarter of Beirut”—a Hezbollah stronghold that Israel all but flattened in 2006—“will happen in every village from which Israel is fired on. From our standpoint, these are not civilian villages, they are military bases.” In case Nasrallah thought Israel was bluffing, Eisenkot reiterated: “This is not a recommendation. This is a plan.”41 The threat was credible given Israel’s 2006 offensive and Hezbollah understood the cost of renewed engagement: harm to its social, infrastructural, military, territorial, political assets, and overall popularity. During the ensuing Gaza conflicts, only a handful of rockets were launched from Lebanon. A vignette helps illustrate how all of this plays out in practice. The nexus between the Taliban, Pakistan, and al Qaeda is particularly informative for exploring affinity, dependence, and proxy relationships. Immediately following 9/11 Washington was clear as to its intent to punish al Qaeda and its  supporters. Bush, in an address given less than a week after the attack, warned: “All I can tell you . . . is that Osama bin Laden is a prime suspect and the people who house him, encourage him, provide food, comfort or money are on notice.”42 That notice later went out on at least two separate fronts, with threats levied against Pakistan and Afghanistan. In the former case, the United States successfully compelled Pakistan—one of the few states to have openly supported the Taliban during the 1990s—to change course. (Saudi Arabia and the United Arab Emirates [UAE] also had diplomatic relations with the Taliban; both severed those ties following 9/11.)

Deterring Terrorism

49

Former president Pervez Musharraf revealed that Richard Armitage, then U.S. deputy secretary of state, suggested that Pakistan “be prepared to be bombed . . . back to the Stone Age” if it failed to comply with U.S. demands.43 Otherwise, positive incentives and other carrots were also on the table: U.S. sanctions were to be lifted. Musharraf’s autocratic regime would be legitimized. (He had come to power during the 1999 military coup.) In addition, the United States eventually wrote off US$1 billion of Pakistan’s debt; (allegedly) pressured the International Monetary Fund (IMF) and the World Bank to expand and fast-track their loans to Islamabad; and offered billions of dollars in economic and military assistance, placing Pakistan nearly on par with Egypt and Israel as leading U.S. beneficiaries.44 These threats and incentives were convincing, and Pakistan acquiesced to U.S. pressure. In a dramatic televised speech on September 19, Musharraf went public with his decision to join U.S. efforts and to end (ostensibly) Pakistan’s support of the Taliban and al Qaeda. “In the present circumstance,” he cautioned the citizens of his country, “a wrong decision can lead to an end which will be unbearable.”45 Afghanistan was a different story. U.S. coercive threats repeatedly failed to convince the Taliban to end its support for al Qaeda. As early as September 17, Musharraf, at the behest of the United States, sent military and intelligence delegates to Kabul, including the chief of Pakistan’s Inter-Services Intelligence (ISI), General Mahmud Ahmed, armed with a “personal letter for Mullah [Mohammed] Omar” to help sway the Taliban’s position on bin Laden. They were unsuccessful. Bush then issued a clear ultimatum during his September 20 Address to the Nation, articulating U.S. threats to the Taliban: Tonight the United States of America makes the following demands on the Taliban. Deliver to United States authorities all of the leaders of al Qaeda who hide in your land. . . . Close immediately and permanently every terrorist training camp in Afghanistan. . . . Hand over every terrorist and every person and their support structure to appropriate authorities . . . [and] [g]ive the United States full access to terrorist training camps, so we can make sure they are no longer operating. These demands are not open to negotiation or discussion. The Taliban must act and act immediately. They will hand over the terrorists or they will share in their fate.46

50

Chapter 3

Within a few days, it had become clear the Taliban would not fully comply. On September 24, Rumsfeld reiterated: “We’re intent on altering behavior. . . . And that means that we want countries to stop behaving in the way that I’ve just described [harboring terrorists], and we intend to do things that will help encourage them” to comply.47 On October 7, U.S. and British forces conducted the first bombing raids over Afghanistan. Although the raids destroyed Taliban military facilities, equipment, and training camps, the precision strikes were also meant to underscore U.S. coercive demands and to communicate its intent and capability to see its threats through. Rumsfeld explained that the raids were meant to “raise the cost of doing business for foreign terrorists who have chosen Afghanistan from which to organize their activities, and for the . . . Taliban regime that continues to tolerate terrorist presence.”48 Four days later, Bush repeated his demands: “I will say it again: If you [Taliban] cough him [bin Laden] up and his people today that we’ll reconsider what we’re doing to your country. You still have a second chance. Just bring him in.” 49 This second chance went unheeded. Secretary of State Colin Powell, dispatched to Islamabad on October 15, was further unable to bend the Taliban’s position.50 Finally, on October 19 and 20, the United States launched two assaults—a helicopter-borne commando raid on one of Mullah Omar’s Kandahar compounds and a paratrooper drop on a remote Afghan airfield—that marked a shift in U.S. strategies from coercing the Taliban to targeting its leadership and militarily defeating its forces.51 The episode ended with the eventual invasion of Afghanistan and the toppling of the Taliban regime. At least part of the reason coercion failed in Afghanistan but succeeded in Pakistan has to do with affinity and dependency. Al Qaeda and the Taliban shared a common ideology, a variety of strategic goals, and local, regional, and international interests; affinity was exceptionally high. Even during the Taliban regime’s dying days, its last ambassador to the world was smugly defiant: “All the Taliban are the same. We all stand and hold our faith in Islam. On the issue of Osama [bin Laden] . . . Osama is a faith issue, and we are not going to change our faith for anyone.”52 For Musharraf, however, affinity played a much less prominent role in shaping his strategic decisions. In his September 19 speech signaling Pakistan’s break with the Taliban, he spoke of four strategic concerns: resisting Indian aggression, fighting for Kashmir, protecting Pakistan’s prestige, and safeguarding Pakistan’s nuclear and missile assets. Standing alongside the Taliban and al Qaeda at all costs and in defiance of U.S. threats did not make the cut. “When there is a

Deterring Terrorism

51

crisis,” he continued, “wisdom is a much better way than emotion.”53 Words to live by. In terms of dependency, the Taliban and al Qaeda relied on one another for existential support: Al Qaeda helped the Taliban reclaim Afghanistan from its rivals and control the country, and the Taliban gave al Qaeda territorial room to grow and develop. Freedman goes so far as to aptly suggest that the al Qaeda–Afghanistan relationship was “not so much a case of statesponsored terrorism but of a terrorist-sponsored state.”54 It is also possible, Crenshaw explains, that the Taliban may have believed that “turning over Bin Laden would cost them their tenure anyway” because acquiescing to American demands “would remove what legitimacy they had.”55 The Taliban’s existence and survival were intertwined with that of al Qaeda’s, such that the costs of meeting U.S. demands were high. Not so for Pakistan. As Musharraf revealed, Islamabad’s list of existential considerations went well beyond the Taliban regime’s continued survival. Support for the Taliban may have been important, but it was not critical. In sum, although U.S. coercion succeeded against Pakistan, it failed against Afghanistan at least partly because the Taliban risked bearing significant costs in acquiescing to U.S. demands and was unwilling (and potentially unable) to distance itself from al Qaeda as a result. CBRN, OR WMD, TERRORISM

Th reats of retaliation and punishment can also deter states from sharing chemical, biological, radiological, and nuclear weapons with terrorist organizations. In general, deterring CBRN transfers is easier to do than deterring support for conventional terrorism because the stakes are higher. The goal is to deter terrorist acquisition by associating the clandestine sponsorship of CBRN terrorism as one wholly directed by the supporting state itself. Th is is especially relevant with nuclear terrorism because terrorists are likely to require state facilitation, either directly (through sponsorship) or passively (by negligence), to gain access to or to acquire these weapons surreptitiously. 56 In one sense, the risks associated with covertly assisting terrorists in orga nizing a nuclear (or CBR) attack are the same as if the state had carried out the attack itself. If so, coercive credibility may rest on a defender’s ability to identify and trace the source of a clandestine CBRN attack: “Forensics is the linchpin.”57 Culpability can lead to the source of the weapon and traditional forms of military retaliation can follow.

52

Chapter 3

Consider then that the day North Korea detonated a nuclear device, President Bush responded by warning President Kim Jong–il that North Korea would be held “fully accountable” for transferring nuclear technology or weapons abroad.58 Washington’s goal was to deter Jong–il from doling out its weapons and spreading its technology around. (Ultimately, these threats proved weak: North Korea went on to help Syria build its own nuclear weapons program, which Israel destroyed in 2007.)59 Again, strengthening attribution capabilities that use forensic science to trace CBRN materials to their sources of origin will help signal would-be sponsors of terrorism that they cannot easily transfer weapons anonymously. Other scholars, like James Kraska, have added that criminal civil law and torts theory (along with the principles of negligence, strict liability, and agency inherent to them) might be useful for communicating threats meant to deter a state’s active and passive involvement in nuclear terrorism. Notwithstanding the potential deterrent value of these domestic, legal, and forensic-based instruments, pursuing a policy of offensive counter-proliferation might deter some states from even contemplating openly developing WMD capabilities. Writing of the Bush doctrine, Daniel Whiteneck argues that in the years since 9/11 the United States developed a credible and definitive policy to deter state sponsorship of WMD terrorism: “A nation that has launched two major military interventions—tied in large measure to supporting terrorism or WMD proliferation—and that has publicly articulated a willingness to preempt WMD acquisition . . . increases the credibility that it could be more than willing to respond in the same or greater magnitude” following a nuclear attack.60 This is fearsome stuff. Paradoxically, a WMD-capable, terrorismsponsoring state might find it in its best interest to diminish its rhetorical fervor in support of terrorism lest it suffer retaliation in response to someone else’s anonymous WMD attack. An act of WMD terrorism followed by an issuance of state sponsorship—an unlikely but possible development— offers a return address for retaliation. If, however, a WMD terrorist attack remains truly anonymous, the victim’s ensuing investigation could very well lay blame on a state’s doorstep regardless of actual culpability. Intelligence, as the 2003 Iraq War reminds us, is rarely perfect. In all of these scenarios, a state sponsor of terror suffers a crippling retaliatory attack for either directly supplying terrorists with WMDs or for talking as though it might supply terrorists with WMDs. Yet threatening retaliatory punishment in response to CBRN terrorism is problematic and far from straightforward. For starters, attacking CBRN-

Deterring Terrorism

53

capable states for supporting terrorism risks inviting CBRN retaliation, while threatening Russia or Pakistan (or, for that matter, India, France, or the United Kingdom) in response to nuclear terrorism stemming from lapses in security, is simply not credible. Some of these states are likely to retaliate, but military threats may also damage multilateral ventures meant to safeguard stockpiles and discourage proliferation. Also, threatening allies with nuclear strikes will prove counterproductive.61 Finally, under certain circumstances, weapons transfer may be more—rather than less—likely to take place. For instance, threatened states may be more likely, not less likely, to transfer CBRN weapons to nonstate proxies and allies.62 In essence, lastditch, back-against-the-wall attempts at regime survival might compel (rather than deter) a CBRN-capable state to hand over weapons to terrorist proxies in order to avoid total capitulation. Once the dust settles, Syria—in 2013–15—will make for an interesting case study. Given its ties to Hezbollah and its penchant for using its vast chemical stockpiles to attack militants and civilians alike, there are legitimate fears that Damascus might transfer its chemical weapons to militants in order to survive the civil war. Time will tell. Otherwise, forensic and attribution sciences are far from perfect. There was much uncertainty, for instance, as to the exact source of Libya’s clandestine nuclear program after it was halted in 2003, despite the fact that Tripoli supplied Washington with a uranium hexafluoride gas sample and notwithstanding the “mountain of . . . information” the International Atomic Energy Agency (IAEA) itself collected.63 And consider finally that deterring state transfers neglects the role nonstate intermediaries, criminal networks, roguegovernment elements, and individuals (for example, Pakistani A. Q. Khan) have in facilitating CBRN terrorism. Some actors involved in CBRN terrorism, like criminal networks and individuals like Khan, may be motivated by profit, suggesting that their behavior is informed by the odds of capture and the likelihood and severity of punishment. In this regard, criminals are riskaverse and may respond well to threats that increase the cost of doing business.64 And yet nuclear and radiological terrorism—which dominates the CBRN terrorism discussion—is but one form of WMD terrorism. Chemical and biological agents, for instance, have long been used in terrorist campaigns. A wide range of groups, from the Tamil Tigers, who used potassium cyanide in 1986 to poison Sri Lanka’s tea export, to Japan’s Aum Shinrikyo, who in 1995 released sarin gas on Tokyo’s subway system in a brazen act meant to kill thousands, have used them in attacks. And even before the 1995 sarin

54

Chapter 3

attack, Aum Shinrikyo reportedly carried out several other biological and chemical attacks, spraying mists of germs and microbes, including botulinum and anthrax, from trucks and high-rise buildings in busy urban centers. Thankfully, most of these attacks failed.65 The point is that unlike nuclear and radiological weapons, CB agents may be far easier to acquire, steal, or develop clandestinely. State transfer may be unnecessary, potentially diminishing the prospects for coercion. Between October 2006 and June 2007, for example, militants associated with al Qaeda in Iraq (AQI) conducted more than fifteen chlorine-enhanced and nitric-acid-enhanced bombings. No state was involved in these attacks and though Iraqi and coalition forces tried to apply threats of punishment (and denial) to stem AQI’s use of the CBRN weapons, their strategy was not altogether successful.66 The AQI case does, however, highlight one final line of inquiry. Looking ahead, some authors have explored tailoring deterrence by punishment to compel nuclear-capable terrorist groups from using the weapons they have. To be sure, this is speculative work, but the exercise has its merits. Writing of al Qaeda, Dunn suggests that the United States must influence its leaders’ assessments of whether “actual employment of a nuclear weapon would help or hurt” their cause and long-term goals.67 The idea here is to influence decisions to use, rather than acquire, nuclear (and CBR) weapons. By better understanding how terrorists perceive and internalize the utility and risk of using CBRN weapons, states might better appreciate how to deter their use.68 As Jenkins admits, “I doubt that terrorists smart enough to think about how they could acquire a nuclear weapon will be dumb enough to try. I believe that serious thinking by terrorists about how they might use a nuclear weapon . . . would cause all but a handful of nihilist fanatics to conclude that nuclear terrorism . . . would be counterproductive to their cause and their survival.”69 Following Dunn’s logic and Jenkins’s account, nuclear acquisition by al Qaeda may not necessarily and automatically lead to nuclear use. Flinch if you must, but deliberation over the costs and benefits of nuclear terrorism will inform terrorist decisions to use such weapons. Studies of al Qaeda’s strategic culture—loosely defined as the narratives, opinions, ideas, and assumptions about war and conflict that help situate and guide a group’s (or state’s) military behavior—have found that proverbial hawks and doves exist within al Qaeda’s shura (governing) council. Al Qaeda leaders held heated debates concerning the development, acquisition, and use of WMDs. Jerry Mark Long recounts that “hawks . . . pushed for authorization [to acquire nukes] . . . but others were deeply concerned. They feared pulling heaven

Deterring Terrorism

55

down upon their head in a retaliatory strike by the West.”70 Traditional coercion, in the form of threats, punishment, and retaliation, might help dissuade nuclear-capable groups from using the weapons they have managed to acquire and develop. SUPPORTIVE COMMUNITIES OF TERRORISM

Terrorist organizations do not spring forth from a void. They usually represent a specific sociopolitical, cultural, or national community and strive to correct grievances (real or imagined) that are shared by members of that community. Members of the Liberation Tigers of Tamil Eelam (LTTE, or Tamil Tigers), for instance, are predominantly recruited from Sri Lanka’s ethnic Tamil minority and the global Tamil diaspora living in India, Canada, the United States, Europe, and Australia. Historically, LTTE violence was used to achieve a sociopolitical and secessionist agenda that reflected grievances held by segments of the wider Tamil community. A variety of other groups—ranging from the Irish Republican Army (IRA) and Quebec’s FLQ (Front de Libération du Québec) to Hezbollah and Hamas, and further to Columbia’s FARC (Revolutionary Armed Forces of Colombia) and Peru’s Shining Path (Sendero Luminoso)—have been sensitive to the shifting moods of their support base. Even al Qaeda and its various branches and offshoots have taken the pulse of global Muslim attitudes and sensitivities in tailoring their behavior. This explains why al Qaeda leaders have repeatedly sought religious justification (fatwas) for indiscriminately killing women and children; for targeting apostate Sunni Muslims, Sufi and Shi’a Muslims, Hindus, Christians, and Jews; and for using suicide in their operations. It is because al Qaeda’s holy war is already considered illegitimate by hundreds of millions of Muslims that it goes out of its way to provide the theological justification for its behavior and avoid purposefully alienating its popu lar base. In the light of the supportive arc that connects terrorist organizations and their popular base, some scholars have suggested that punishment strategies targeting communities may deter terrorism. The assumption is that under pressure, these communities may revisit, alter, or altogether end their support for political violence; terrorist groups would suffer as a result. (History is replete with cases based at least partially on this form of punitive coercion.71) In this case, deterrence is again acquired indirectly, where receding popular support for terrorism translates into pressure on the militant group, which is compelled to change its behavior from the bottom up.

56

Chapter 3

Relying on the Cold War notions of massive retaliation and assured destruction, Gerald Steinberg writes that deterring terrorism requires identifying “high-value targets, including family and supporters, that will cause even the most radical leaders to weigh the costs and benefits of their actions.”72 In the specific realm of deterring nuclear terrorism, Elbridge Colby suggests expanding the target base for retaliation beyond state sponsors and terrorist groups. The idea is to threaten the armchair militant: the individuals and communities that passively support political violence by turning a blind eye to its development within their midst, and actors who give credence to terrorists by legitimizing their motives or justifying their goals.73 To a certain extent, Israel has put these notions into practice in dealing with Palestinian terrorist organizations and Hezbollah. For instance, the aforementioned threat issued during the prelude to Operation Cast Lead involved laying waste to Hezbollah’s Beirut stronghold. Today the strategy is known as the Dahiya doctrine: “Lebanon is Hezbollah and Hezbollah is Lebanon. Hezbollah shoots at us and we go to war against Lebanon.”74During the al Aqsa Intifada (2000–2005), Israel punitively demolished and sealed the homes of Palestinians who either carried out or were connected to suicide attacks. The idea was to widen the associated cost of involvement in terrorism to the suicide bomber’s social and familial network. The hope was that individual terrorists would be deterred by threats to the welfare of their families and that the families themselves would put pressure on relatives suspected of involvement in terrorism. The policy was highly controversial. But empirically, Harvard economist Efraim Benmelech and colleagues found that in the months following an Israeli punitive house demolition, fewer suicide attacks originated from the district where the homes had been destroyed.75 The punishment influenced social support for terrorism and deterred involvement in terrorism. Nonetheless, deterring terrorism by targeting peripheral supporters is problematic and has a number of critical flaws. Some punitive measures may require extreme forms of punishment that run counter to practical, humanitarian, and legal positions held by Western governments (though authoritarian regimes may face fewer similar constraints). Uri Fisher suggests, for instance, that for the United States—which he insists is “concerned with maintaining its moral authority in the world”—actually implementing certain coercive threats may be difficult.76 Others have come to the same conclusion, finding that liberal democracies may be restrained (at least initially) from doing what they must to deter a group: Japanese law, for instance, pro-

Deterring Terrorism

57

tecting religious minorities inhibited security forces from properly investigating Aum Shinrikyo as it prepared its biological and chemical attacks in the 1990s.77 The United States and other liberal democracies may be too soft to do the sorts of nasty things that are required of them to deter fanatical groups and their supporters directly. Writing of Israel, Bar concurs, explaining that Israel’s strategy of punishing peripheral and popular communities proved difficult to implement because of international and domestic constraints. So while Israel broadcast “blatant threats” to the populations of Gaza and the West Bank that described “Israel as a ‘hornet’ that if awakened could sting its enemy to death,” it simultaneously undermined this deterrent by later publicly apologizing for military operations that inadvertently killed civilians.78 The coercive message was lost in translation. The credibility of a threat designed to target a population base was weakened and nullified by a (democratic) state’s need to refrain from actually executing those threats. In this case, there is an apparent gap between coercive rhetoric and political reality. The question remains of whether threats issued against popular bases actually deter terrorism or galvanize support for it. It is a fine line. Some cultural or religious groups, for example, may be sensitive to threats of punishment that evoke feelings of dishonor and humiliation rather than fear or submission. And in the Israeli studies just discussed, although punitive demolitions (those carried out against persons related to a suicide operative) resulted in reductions in terrorism, “precautionary demolitions” (those Israel carried out to prevent or preempt the launching of attacks from a specific location but were not directly a response to suicide attacks) had the opposite effect. They led to more, not less, terrorism. From a coercive standpoint, indiscriminate retaliation can be counterproductive because it may entrench new grievances, increase support for political violence, and validate a militant group’s raison d’être.79 Deterrence by punishment needs to relate to something (a behavior, a belief, an attitude); otherwise, the coercive message is moot. And retaliation unhinged from the logic of deterrence is not coercion but vengeance. TERRORIST GROUPS

In addition to tailoring coercion against supportive states and communities, defenders can target terrorist organizations themselves with threats of punishment. Although there is some disagreement among scholars as to the strategy’s effectiveness, under certain conditions threatening organizations

58

Chapter 3

appears relevant. For instance, if and when terrorist groups gain “statelike” attributes, as Hamas and Hezbollah have, then governments will find a multitude of targets to credibly threaten. But even with truly stateless groups, finding the right leverage may simply be a matter of using one’s imagination and properly identifying the “symbols of central importance.”80 The nature of the retaliation itself might likewise influence the coercive outcome. In their work, Bryan Brophy-Baermann and John Conybeare illustrate that the rate of terrorism against a country can be disrupted by “surprise retaliations of large magnitude.” Israel’s response to the massacre at the 1972 Munich Olympic Games—in which Palestinian terrorists killed 11 Israeli athletes and coaches—is a case in point: Rates of terrorism against Israelis and Jews dropped as a result. The authors warn, however, that deviations in the terrorism rate can be fleeting, dissipating over time as terrorists raise their own “expected level of retaliation.”81 And, of course, relying on imagination, surprise, and large-scale retaliation can quickly lead to unpalatable extremes. For instance, Thomas Tancredo, a U.S. Republican who served multiple terms in Congress, suggested, as he sought the 2008 GOP presidential nomination, that “[i]f it is up to me, [the U.S. government is] going to explain that a [nuclear] attack on this homeland . . . would be followed by an attack on the holy [Islamic] sites in Mecca and Medina. That is the only thing I can think of that might deter somebody. If I am wrong fine, tell me, and I would be happy to do something else. But you had better find a deterrent or you will find an attack. There is no other way around it. There have to be negative consequences for the actions they take. That’s the most negative I can think of.”82 Setting aside the ethical issues involved in such attacks, there are strategic reasons for why Tancredo’s threat would not work. It lacks coercive credibility: Both Mecca and Medina are located in Saudi Arabia, a valuable if at times fickle American partner. Publicly placing it on a U.S. target list would jeopardize that relationship and hamper other U.S. strategic initiatives in the region (for example, Arab-Israeli peace, democratic transition, and nuclear counter-proliferation). Tancredo’s recommendation would likely backfire, too; instead of instilling fear, the United States would invite widespread scorn from the Muslim world, further alienating and angering Arab allies, elites, and moderates. Finally, targeting Islam’s holy sites provides al Qaeda and other like-minded extremists with fodder to substantiate their claims of a Western war on Islam. Violent reactions to supposedly “blasphemous” cartoons, films, and books may offer lessons for how al Qaeda and others would exploit a U.S. threat to destroy Mecca and Medina.

Deterring Terrorism

59

A more prudent strategy might involve punishing groups by threatening their tactical and strategic goals, rather than their assets. In this way, threats of punishment are combined with deterrence by denial (explored in detail below). One strategy is for defenders to offer militant organizations certain accommodations that can later be held hostage in order to influence future behavior. Robert Trager and Dessislava Zagorcheva argue that U.S. bargains with locally based militant Islamist groups helped fracture al Qaeda’s global network. The United States offered to abstain from fully involving itself in regional dynamics (partial accommodation of militant demands) in exchange for the local group’s refrain from joining al Qaeda. The coercive threat was that in uniting with al Qaeda, the regional group would force the United States to involve itself in local antiterrorism efforts, putting the terrorist group’s political, tactical, and strategic goals at risk.83 A second strategy might involve applying targeted pressure on specific terrorist leaders. Sanctions, death, and other forms of punishment might influence a leader’s behavior, and, by extension, the group’s behavior as well. Targeted killings and precision strikes carried out by drones might have a secondary coercive effect on a militant leader and group. The topic is discussed in detail in Chapter 4. But besides the threat of death or capture, leaders might also be challenged in other ways. Saudi Arabia, for instance, revoked bin Laden’s citizenship and froze his financial assets in 1994 and induced his brother, Bakr, to denounce him on behalf of the bin Laden clan. Bakr cut the family’s ties with Osama, expressing his “regret, denunciation, and condemnation” of his brother’s violent activities.84 Developments like these make it harder for leaders to operate freely and they strip away some of the leaders’ perceived legitimacy. Some militants might be susceptible to these sorts of threats. INDIVIDUAL TERRORISTS

Finally, challengers that rest below the level of the group, community, or state—such as individual militants, suicide bombers, terrorist personnel, and would-be recruits—might be deterred. In the case of the willing suicide operative, adding costs to his or her actions appears difficult. The threat and fear of punishment may have little coercive effect on an individual who has already decided to die. But other threats and other targets nonetheless exist. So although suicide bombers may largely be resistant to threats of death, they may reconsider their actions if failure meant a lifetime of imprisonment. A willingness to die may not comport with a willingness to spend the rest of one’s life in the “unglamorous, isolated, largely forgotten role of the

60

Chapter 3

prisoner.”85 If the risk of capture is high and the consequence of failure well understood, some suicide bombers might think twice before participating in an attack that risks landing them behind bars. And, as noted, threats against a suicide operative’s family or community may compel or deter some individuals indirectly. It is important to note that suicide terrorism is but a minor part of terrorist activity. Most terrorist acts are not intended to kill the perpetrator. More often than not, militants would like to go on living. So in these cases, threats of death may indeed influence behavior. Likewise, some individuals active within terrorist organizations are only superficially involved in violence. Financiers, legitimizers, logisticians, arms suppliers, and a variety of other peripheral actors may be influenced by threats of punishment, such as being found and unmasked, arrested or killed, or even shamed and ridiculed. As Chapters 4 and 5 illustrate, targeted killings represent another form of coercion that can be directed against individual terrorist leaders and facilitators. Though targeted killings are contentious, the tactic is used by a number of states in more than a dozen conflicts. For the most part, those targeted have included high-ranking terrorist leaders and commanders along with operators, facilitators, bomb-makers, and mid-level organizers. Although targeted killings are thought to degrade an organization’s capability by removing personnel, leadership, and knowledge, they may also diminish militant motivation and influence group behavior. The assumption is that eliminations represent a cost to participating in terrorism. Consider, then, that after Israel killed Hamas cofounder Ahmed Yassin in March 2004, the organization publicly appointed Abdel Aziz al Rantissi as leader. A month later, Israeli forces dropped a missile on al Rantissi’s car, killing him instantly. Dismayed, Hamas again appointed a new leader, but this time, they kept his identity hidden, lest he suffer a similar and an immediate fate. That militant leaders are keen to survive is an exploitable characteristic. Finally, individuals on the path toward violent radicalization and those contemplating joining a militant group might be swayed otherwise. Drawing people away from violence will weaken militant organizations. Violent radicalization is a complicated process, but for individuals thinking about joining militant groups, deterrence might diminish their desire to do so by diminishing the value and raising the costs of actually joining. In terms of punishment, the message is that participating in political violence is risky, costly, potentially embarrassing, and rarely legitimate. The strategy might involve using the criminal-justice system and past successes in thwart-

Deterring Terrorism

61

ing terrorist plots and incarcerating would-be terrorists as leverage against would-be terrorist recruits and radicalizing individuals. In sum, in thinking about deterring terrorism by punishment, defenders must consider many different actors and processes. Unlike Cold War punishment strategies, where focus was placed on central governments and threats relied on conventional and nuclear military retaliation, deterring terrorism by punishment is a more subtle, nuanced, and delicate enterprise. Deterring terrorism by punishment involves manipulating a variety of different actors, including those that are only superficially involved in violence, in a number of different ways.

Denial: Targeting What Terrorists Want

Deterrence by denial provides a second manner with which to coerce terrorists; it manipulates an adversary’s behavioral calculations by preventing the desired effects of its attack. Although denial was dwarfed by the threat of nuclear punishment during much of the Cold War, today, in dealing with nonnuclear adversaries and terrorists alike, theories of denial have taken center stage. In counterterrorism, a number of denial mechanisms have been proposed. First, defensive denial functions by restricting and constraining the terrorism processes. By augmenting structural defenses around potential targets, for instance, a state effectively tightens the security environment and reduces the ease with which terrorists can carry out attacks. This is intuitively understood: By impeding access, structural defenses force terrorists to reassess the costs and benefits of a particular action.86 Other scholars have gone further. They suggest that restricting terrorist financing (by uncovering, foiling, and blocking monetary transactions between terrorists and their supporters) and eliminating an organization’s domestic or foreign sanctuaries (by better policing territory or by more effectively cooperating with neighboring states) might also have a coercive effect on an organization, effectively denying it easy access to the things it needs.87 In theory, the more difficult a target is to attack, the less likely it will be. And, related to this, the higher the level of protection around a target, the more complex the attacks against it will have to be, further augmenting the risk of operational failure. In practice, defensive deterrence requires first assessing what targets terrorists most want to attack and building specific defenses that challenge

62

Chapter 3

the utility of attacking. If perpetrators come to believe that attacks are going to be difficult to mount or that they are likely to fail, they may be less willing to try. Defensive denial in counterterrorism is not particularly new. During the 1970s (and especially following the well-coordinated attack at the Munich Olympics) the United States was concerned that foreign militants might manage to steal an American nuclear bomb stored in one of NATO’s European bases. In response, U.S. scientists developed new parameter-control mechanisms that could detect and stop intruders. Eric Schlosser recounts that one such mechanism involved mounting nozzles on the walls of bomb storage facilities that could quickly fi ll rooms with sticky foam, immobilizing intruders and safeguarding the nuclear bombs.88 More recently, during the al Aqsa Intifada, Hamas, Palestinian Islamic Jihad, the al-Aqsa Martyrs Brigade, and other militant groups began by first attacking Israelis within Israel with suicide bombers dispatched against soft targets. Transportation hubs, cafés, bars, restaurants, shopping malls, and public markets were repeatedly targeted in the first half of the conflict. In response, Israel began defending public access points to bus and train depots, universities, and hospitals; restaurants and bars began placing guards outside their doors; blastproof entrances were added to buildings; checkpoints and security barriers were established in major cities and on highways; and privately owned shuttles—like Sherut, a minivan, taxi-sharing ser vice that ferries passengers between cities—began offering alternative transportation to public buses and trains. The cumulative effect was the eventual restriction of easily accessible soft targets. We like to think that a guard outside a restaurant serves the diners inside; they are protected from harm because the guard is likely to stop a would-be bomber at the door. But there is more to it: In limiting access to a target, defenses go beyond simply protecting that target to manipulating an adversary’s willingness to attack that target. The fi rst process is defense; the second is coercion. In Israel, over time, suicide bombers were forced to target military and police checkpoints over civilian targets, and many detonations occurred outside and off target. The result was a diminishment in the utility of suicide attacks in Israel and an eventual reduction in their use. Deterrence by denial need not rest on structural defenses alone. Behavioral defenses work by introducing environmental uncertainty into terrorist planning. The U.S. Department of Homeland Security suggests that “variability and unpredictability must be consciously injected into flexible prevention measures.”89 The deterrent target in question here is the terrorism

Deterring Terrorism

63

process itself; unpredictability impedes terrorist planning by introducing greater levels of uncertainty. And uncertainty translates into a denial mechanism in the planning and orchestration of terrorism. As a 2011 report published by the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland suggests, “If [terrorists] cannot understand and game security systems, they cannot have confidence that an attack will succeed.”90 In practice, applying behavioral denial involves conducting spot checks at public transportation hubs, establishing a police presence at randomly selected city intersections, and instructing security vehicles to leave their emergency lights on—as they do in Washington, D.C., for instance—to give the impression of an overwhelming security presence. Most of these tactics rely on what is often referred to as random antiterrorism measures (RAM), which “change the overall security/force protection appearance” around potential civilian and military target sites, as seen “through the eyes of a terrorists.”91 An effective RAM program presents militants with an unscheduled and even “unorthodox situation” that will complicate planning attacks and potentially deter behavior as a result. RAM defenses influence behavior because terrorists are generally risk-averse while preparing for attacks—they obey the law; are less likely to steal, speed, or do drugs; and generally try to avoid undue attention—and are usually sensitive to operational risk—which they try to diminish by casing a target in advance, uncovering defensive patterns, and exploiting security gaps.92 Behavioral denial muddles this surveillance process, negating preparation and potentially unnerving planners. R. P. Eddy, director of the Center for Policing Terrorism in New York City, suggests that defensive counterterrorism should seek to match terrorism’s unpredictability by injecting randomness into defensive tactics. To illustrate, he argues that random bag searches used in New York City’s subway system can deter would-be terrorists in a cost-effective and an efficient manner: “If you’re a terrorist you have a limited amount of resources [and] you don’t want to deploy them if you have an increased chance of . . . not being successful.”93 The point is not to catch a terrorist in the middle of the act (though that would be pretty good, too) but to give the impression to wouldbe bombers and militants that they cannot properly prepare for an attack. Even a very minor probability of getting caught (by a random bag search, in this case) may have a disproportionately large deterrent effect.94 Defensive randomness can be used in other ways, too. For instance, several dozen police officers in squad cars converge twice daily at randomly selected times

64

Chapter 3

and locations throughout New York City. The idea is to keep extremists “guessing as to when and where a large police presence may materialize at  any hour.”95 And since 2007, Los Angeles International Airport (LAX) has used a computer program, called Assistant for Randomized Monitoring Over Routes (ARMOR), to help determine when and where security personnel and canine units will converge within the terminal, and identify where temporary checkpoints will be set up on the roads leading to the airport. The number of security teams guarding LAX is finite (as is the security budget), such that not all access points and routes can be guarded permanently. Deploying security personnel in a rigid and deterministic manner will allow terrorists to identify and circumvent security planning. ARMOR allows police and other security personnel to randomize their behavior and security schedules.96 Again, the odds of catching a terrorist cell in the middle of a crime and by chance are intrinsically small, but unpredictability nonetheless impedes terrorist planning by introducing greater levels of uncertainty. In this case, establishing a perception that plans will not easily succeed provides the deterrent. No terrorist group or cell determined to strike LAX will be able to identify, learn, and apply security schedules to their own plans of  attack. That will increase their odds of failure and may influence their decision making sufficiently to deter them from attacking the airport. Finally, deterrence by mitigation functions by blunting and limiting terrorism’s social, political, and economic effect. The idea is to deny the consequences terrorists anticipate and desire. As Anthony Powers explains, mitigation sends a message that “the impact of using these weapons [or tactics] will be minimal.”97 Immediate mitigation involves limiting the proximate desires terrorists hope to acquire with their attacks: death, destruction, fear, anger, and so on. Maintaining effective first-response and public health systems, for example, that are able to manage the immediate consequences of terrorism can help save lives and calm nerves. Doing so might also help dampen the effects of terrorism, forcing those contemplating attacks to consider the utility of their plans. But mitigation extends further to impeding a group’s long-term, sociopolitical objectives. The goal is to deny terrorists their distant—rather than proximate—objectives. Mitigation rests on demonstrating that terrorism does not lead to the attainment of goals and is, in fact, counterproductive. Emphasizing the strategic futility of political violence is what counts.98 The strategy goes beyond limiting the specific tactical effect of terrorism to denying the strategic outcome it is meant to produce over the long haul. As Freedman suggests, “Over time, doubts can be in-

Deterring Terrorism

65

serted into the minds of would-be terrorists” concerning the effectiveness of particular methods and tactics.99 Still other experts stress the importance of inherent mitigating factors: A robust economy might help a target state withstand the shocks of terror, sociopolitical cohesion can give a government an effective mandate to properly respond to a terrorist threat, and societal resilience can challenge the utility of a terrorist campaign.100 Each helps deny terrorists their strategic goals. Consider that after three Palestinian suicide bombers detonated their explosives at a busy pedestrian mall on Jerusalem’s Ben Yehuda Street on September 4, 1997, the reaction most Israelis responded with was defiance. Crews spent the night “hosing down the pavement and buildings, washing away spattered blood”; shop owners immediately replaced shattered windows, reopening the following day; and crowds quickly flocked back to the mall. One shop owner explained to the New York Times: “We’re showing that life goes on . . . they can’t be allowed to disrupt our normal routine.”101 Similarly, following 9/11, critics lambasted Bush for calling on Americans to “get back to work, to go shopping, going to the theatre, to help get the country back on a sounder financial footing,” and people criticized British prime minister Tony Blair for stating that citizens “should go about their daily lives: to work, to live, to travel and to shop—to do things in the same way as they did before September 11.” But there might have been at least some validity to their pleas: When terrorism has little lasting effect, highlighting its futility may diminish its use. In practice, states can put terrorist objectives out of reach by providing allies targeted by a common adversary with financial, policing, and military assistance, or by supporting substate groups that oppose terrorist agendas and violence, or by simply refusing to acquiesce to terrorist demands. In the case of hostage operations, for instance, eliminating the tactical and strategic value of this particular tactic might require that states communicate an absolute unwillingness to negotiate under these circumstances and to refuse to trade terrorist prisoners or pay for hostages.102 A brutal but potentially important lesson comes from Russia following the 2002 and 2004 mass hostage takings at a theater in Moscow and a public school in Beslan. The Russians responded to these attacks by coordinating counterterrorism operations that resulted in the violent deaths of hundreds of hostages, sending the unequivocal message that such tactics were utterly futile. Officials appeared willing to go to nearly any length to put an end to these crises without negotiation: In Moscow, security forces pumped a toxic gas into the theater,

66

Chapter 3

subduing many of the militants but also killing dozens of civilians; in Beslan, the school was liberated following a jumbled police operation that led to the death of 331 hostages (roughly one-third the number being held). Finally, actively preparing a citizenry to withstand the shocks of terrorism—by, for instance, “engaging citizens in their own preparedness and response”—can help deny terrorists the things they seek.103 Resilience is the buzzword. If and when political violence is deemed counterproductive, less violent groups may splinter off from the core, developing and fielding more benign strategies to advance their common goals. The Palestinian case is informative. Terrorism against Israeli civilians, Max Abrahms illustrates, has repeatedly failed to advance Palestinian national aspirations over the long term. Instead, Palestinian violence strengthened Israel’s resolve to combat terrorism, helped it mend its domestic cleavages, and helped delegitimize the Palestinian cause internationally.104 These are strategic losses. It is conceivable that eventual fatigue might set in (as it has already done with members of the PLO and Fatah) and that political agendas may eventually eclipse violent ones over time.

Delegitimization: Targeting What Terrorists Believe

The logic of deterrence can be used to manipulate the political, ideological, and religious rationale that informs terrorist behavior. The objective is to reduce the challenger’s probability of achieving the desired goals by attacking the legitimacy of the beliefs that inform the behavior. Target what terrorists believe, rather than what they value or want. Although it seems self-evident, as David Lake believes, that “terrorists lack moral strictures against the use of violence,” this is only half the story.105 Terrorism is not ordinary violence but violence with a purpose. Terrorists use par ticular forms of violence that comport with specific ideological or socioreligious beliefs, values, and narratives.106 Most terrorists base their activities on a set of principles. Ideological and socioreligious beliefs not only inform terrorist behavior but also shape their goals. Targeting these narratives may—in theory—alter the behavior of future and current militants and their peripheral and active supporters. Applying the logic of deterrence to these narratives suggests that specific leverages might be developed that delegitimize the rationales and goals informing terrorist violence, de-romanticize the life of a terrorist, and de-glamorize the leaders who espouse these narratives.

Deterring Terrorism

67

Strengthening and disseminating opinions, positions, and information that contradicts the legitimization of terrorism might deter or compel individuals contemplating or taking part in violence, along with the socioreligious groups that facilitate terrorist efforts. Doing so will diminish a militant movement’s ability to recruit and indoctrinate new followers and make it harder for a terrorist group to justify its actions and goals in a way that resonates with its would-be support base. To be sure, delegitimizing terrorism is far removed from the traditional concepts of deterrence and compellence. And although punishment and denial are conceptually rooted to state-centric approaches of deterrence and might easily be applied to counterterrorism, delegitimization is something new altogether. It does not easily fit state-based models of decision making and deterrence, which usually emphasize interests over beliefs, and remains generally underdeveloped, misunderstood, and understudied. Jeff rey Knopf writes: “It is not clear that . . . [deterrence by delegitimization] represents an alternative to punishment and denial as a way to produce deterrence; it seems rather to be a par ticular way of blending the two. It would deny a terrorist organization the benefit of gaining approval and support while also imposing the punitive cost of increased criticism and loss of support.”107 Delegitimization is a product of purposefully broadening deterrence theory to fit evolving contemporary threats. Like punishment and denial, delegitimization attempts to change an adversary’s behavior, but it does so by manipulating the rationales and justifications that inform its preferences. The objective is to manipulate an adversary not by threatening punishment or denying goals but by degrading the rationales that guide its behavior. Delegitimizing terrorism involves influencing popular support, debating religious interpretations, and influencing strategic culture. Consider al Qaeda’s use of suicide terrorism. It is legitimized by relying on religious decrees that justify the taking of one’s own life. These same decrees help shape al Qaeda’s sociopolitical goals. Of importance, however, is that suicide is an otherwise blasphemous act under Islamic law and al Qaeda’s objectives are refuted by a vast majority of those that share its religious faith. But as Jerry Mark Long and I argue, al Qaeda equates the survival of its narrative as highly as—if not perhaps even more highly than—other, more traditional, material assets.108 Although al Qaeda may cherish territorial control and does indeed seek to fi ll its ranks with new recruits, what really counts is how it is perceived by its target audience and how well its narrative and behavior resonate with that audience. These nonmaterial goals and

68

Chapter 3

assets are problematic for traditional deterrence theory, which assumes that rational actors engage in forward-looking calculations of maximum utility. But for al Qaeda, symbolic actions carry weight, such that sacred and noninstrumental goals must be taken into account in any deterrent process. Deterrence by delegitimization suggests, then, that it is possible to raise the costs of participating in political violence by targeting the religious, normative, and cultural rationales that groups and individuals use to justify their participation in terrorism. In this way, delegitimization becomes one aspect of a tailored deterrent approach, in which “cultural analysis” helps us identify the ways in which to influence the “spiritual, financial, and educational foundations” of militant movements.109 Deterrence by delegitimization starts by tapping into terrorists’ “selfrestraints” and then magnifying their role in shaping terrorist behavior.110 Terrorist organizations often refrain from certain actions. This is especially evident in the case of CBRN terrorism. Some militant groups, and even those that base their narrative on religious doctrine, may be sensitive to the perceived legitimacy of acquiring and using unconventional weapons. Writing of Hamas, for example, John Parachini argues that one of its core founding leaders, Ismail Abu Shanab, explained that “the use of poison was contrary to Islamic teachings.”111 Though Hamas is a fanatically dogmatic terrorist group that has no qualms about conducting mass-casualty suicide attacks against civilians, its behavior nonetheless appears bound by some self-restriction. Consider, too, 2003 revelations by George Tenet, former director of the CIA, that al Qaeda had organized a cyanide attack in New York City. Ayman alZawahiri (who replaced bin Laden as al Qaeda’s top leader following bin Laden’s elimination in 2011) called it off because he feared that the attack “was not sufficiently inspiring to serve Al Qaeda’s ambitions” and might even humiliate the organization.112 And Khalid Sheikh Mohammed (KSM), the 9/11 mastermind, suggested that as al Qaeda was selecting targets for the 9/11 attacks, it self-imposed certain restraints on its behavior. KSM told a journalist that al Qaeda “first thought of striking at a couple of nuclear facilities [in the United States] but decided against it for fear [the attacks] would go out of control.”113 Counterintuitively, but like Hamas, al Qaeda’s violence appears to be somewhat measured. These examples suggest that under certain conditions, normative guidelines may influence terrorist activity. The compellent objective, in this case, is to expand the scope of the existing CBRN norm, reinforce adherence to it, and induce other groups not yet bound by its guidelines to accept its rationale.

Deterring Terrorism

69

Defenders can manipulate a challenger’s self-restraint by communicating how certain actions contradict religious tenets and social expectations. In al Qaeda’s case, this might be done by communicating to militants that certain types of attack—like CBRN terrorism—will galvanize global counterterrorism efforts rather than provide terrorist goals, or by illustrating that some attacks might disproportionately kill Muslims and cross certain religious boundaries as a result. Relating CBRN use to Muslim rejection of, rather than support for, al Qaeda, might sway some leaders. CBRN self-restraints can also be manipulated by associating particular egregious acts of violence with popular outrage. This raises the cost of using particular forms of terrorism (deterrence by “popular backlash”).114 Consider AQI’s barbarism. Its onetime leader, Abu Musab al-Zarqawi, was rebuked by al Qaeda’s al-Zawahiri for having alienated the broader Arab and Muslim community, whose support al Qaeda relied on for its survival. Al-Zarqawi’s 2005 attack in Amman, Jordan—in which suicide bombers killed scores of Jordanians—elicited massive protest from the Arab Street. In killing more than sixty (mostly Muslim Jordanians), al-Zarqawi was publicly and vociferously denounced; residents of his hometown even called for his death.115 The lesson here is that although al-Zarqawi orga nized tactical terrorist successes, the way he did so damaged al Qaeda’s long-term strategic goals. Turning popu lar outrage into terrorist self-restraints requires that defenders amplify and communicate the unintended damage resulting from terrorism and compel those from within the terrorist group’s sociopolitical community to condemn violence.116 Delegitimization can also be applied to the religious dialogue informing Islamist terrorism (deterrence by “counternarrative”).117 As Long illustrates, bin Laden and other al Qaeda leaders function as “lay mujtahid” (those who interpret religious texts), but they cannot simply offer any interpretation they want “or else [they] would have no legitimacy.”118 Al Qaeda is constrained by the doctrines and traditions to which it purports to adhere. The point is that Islam is multifaceted and is punctuated by religious debate concerning what is and is not legitimate behavior. There is room for debate, then, as to what actions are considered legitimate in life and marriage, in governance and industry, and, yes, even in war and conflict. Al Qaeda understands this limitation. Though it might call for the removal of kuffar (unbelievers; singular, kafir) individuals and governments in the Arab and Muslim world, it does so while very carefully gauging reactions from its broader audience. There is much skepticism regarding al Qaeda’s legitimate right to do so.119 That al

70

Chapter 3

Qaeda continues to go out of its way to provide the justification for its actions suggests that debates have yet to be won. Violent jihad is carefully regulated by sharia (Islamic religious law): The killing of women, children, and the infirm should be avoided; the mutilation of prisoners is forbidden; fair warning of hostilities is required; and agreements must be honored. At no point do religious texts “enjoin terrorism and murder.”120 It is enlightening to juxtapose these religious duties with the ferocity of the unholy terror that is being used today by al Qaeda and its supporters around the world. To a certain degree, Islamic validation for contemporary terrorism is based on a deviation, a selective (re)interpretation, and dismissal of religious texts. Again, consider suicide bombings. Although the fidayeen (one willing to die for a religious cause) has its precedents, today’s suicide bomber is a revisionist anomaly. Unlike the medieval assassin who was willing to die “at the hands of his enemies,” the modern Islamist suicide bomber dies “by his own hand.”121 This is a theological perversion relying on unjustifiable forms of violence (indiscriminate mass killing) and suicide (a sin). Ensuring that individuals contemplating suicide terrorism are aware of these differences might deter some from carrying out attacks. Manipulating debates to delegitimize certain interpretations that condone violence can help inform broader deterrent strategies. Doing so begins by ridiculing individual legitimizers. Focus here should be placed on identifying and illustrating the manner in which terrorist leaders have themselves failed to adhere to the personal, social, and religious guidelines they thrust on others. Jihadi leaders may be especially vulnerable. Evidence of moral, sexual, religious, and social indiscretion—actions that contradict expectations—should be used and widely publicized to discredit militant leaders who lecture the virtues of a life they purport (but fail) to follow. It matters, then, that Anwar al-Awlaki—the American-born militant cleric who became a leading al Qaeda in the Arabian Peninsula (AQAP) propagandist responsible for indoctrinating several Western jihadi recruits—repeatedly failed to behave in the manner he preached. Despite his puritanical defense for the sanctity of marriage, twice he was picked up by San Diego police for soliciting prostitutes.122 We all make mistakes, but in al-Awlaki’s case his mistakes, properly packaged and communicated, could have served a broader strategic objective: Ridiculing the religious and ethical bona fides of a militant cleric who purposefully used his moral character and doctrinal training to promote violence in the name of Islam and to attract followers to al Qaeda’s cause. To use a crude (if not appropriate analogy), think of the dam-

Deterring Terrorism

71

age soliciting sex has had on the reputation, careers, and status of countless American politicians and military leaders. We should expect no less for the al-Awlakis of the world. Otherwise, a militant movement’s message can likewise be challenged. The focus today is on fundamentalist interpretations of Islam, particularly its militant salafi-jihadi strain. Legitimizers operate within these traditions, commenting on how actions correspond to the dictates of religious jurisprudence. Because diverging interpretations exist, however, defenders might rely on religion itself to compel militants to forgo violence. Delegitimization strategies might embolden debates with decrees and fatwas that stigmatize al Qaeda, its leaders, and its followers. The coercive objective in this case is to manipulate terrorist violence by disseminating religious injunctions that sway the behavior and beliefs of a group’s supportive community. Consider that immediately after 9/11, bin Laden was rebuked for failing to fulfill theological requirements pertaining to jihad. He had not offered sufficient warning of the attacks, presented Americans with an opportunity to convert to Islam, or prepared authorization to kill so wantonly.123 A number of prominent jihadi scholars have also retracted their support for terrorism. Sayyid Imam al-Sharif (also known as Dr. Fadl), a onetime leader of Egypt’s Al Jihad (also known as Egyptian Islamic Jihad, or EIJ) and an early al Qaeda ideologue, backtracked in 2007, arguing that “we are prohibited from committing aggression.” In a series of articles, he explicitly forbids the practice of takfir (in which apostate Muslims are identified and targeted) along with the killing of non-Muslims in Muslim countries and members of non-Sunni Muslim sects.124 In 2007, Saudi Sheikh Khalid Bin Su’ud al-Rushud and Sheikh Abd al-Aziz al-Askar condemned al Qaeda’s use of “religious edicts permitting suicide attacks,” confirming that the “act of ‘killing the soul’ ” brings upon the “individual committing suicide suffering from Allah.”125 Another, Sheikh Salman al-Awdah (al-Ouda), attacked bin Laden publicly on television for his role in the death of hundreds of thousands of civilians: “I don’t expect a positive effect on bin Laden,” he explained, “it’s really a message to his followers.”126 And Nasir Abbas, a former leader of Jemaah Islamiyah, later wrote that “not one verse in the [Koran] contains an order for Muslims to make war on people of another religion.”127 These condemnations piled up. By 2008, al Qaeda was forced to expend a huge amount of airtime defending its legitimacy. Bin Laden himself found it necessary to reiterate that “the Muslim victims who fall during the operations against the infidel Crusaders . . . are not the intended targets.” They were unintentional, collateral

72

Chapter 3

damage. And al-Zawahiri countered Dr. Fadl with a treatise of his own.128 The risk for al Qaeda is that condemnations of this caliber influence oncesupportive communities and pious individuals fearful of religious misconduct. If terrorists perceive their war as just, moral, and defensive, promoting and communicating views that contend otherwise might influence the behavior of would-be and active supporters. When violent Islamist organizations lose religious justification, their activities resemble mere thuggery. A loss of popular support usually follows. Finally, in addition to manipulating self-restraints and narratives, other scholars have focused on influencing the social enablers of terrorist organizations. In general, the strategy is about winning the middle, the masses caught along the continuum between absolute support for and absolute rejection of al Qaeda’s mantra. As Long explained to me, what is critical is “enfeebling” al Qaeda and its franchises by delegitimizing them and reducing their pool of would-be recruits.129 Here, focus is placed on the societal bases of support that strengthen militant groups. Over the long run, widespread cognitive and attitudinal leverages might be used to modify the behavior of the militants’ enabling audience. Kim Cragin and Scott Gerwehr, for instance, contend that “strategic influence campaigns” might disrupt “the confluence of anti-Americanism, radical Islam, and general support for political violence” in ways that diminish support for al Qaeda among certain communities.130 In terms of deterrence theory, two processes are at work; recall the distinction between immediate and general deterrence. What Cragin, Gerwehr, Long, and others contemplate is more in keeping with the latter—a “general,” ideologically informed strategy that seeks to deter terrorism in the long run by inducing change in certain segments of particular societies. Thus, although “immediate” deterrence by delegitimization might attempt to influence terrorist groups from contemplating certain forms of violence (CBRN attacks, for instance), “general” deterrence by delegitimization is a forward-looking strategy that attempts to influence a community’s decision to use and support terrorism altogether. It would push terrorism toward “the margins” of acceptable sociopolitical behavior while strengthening antiterrorism norms more generally.131 Indeed, the process is already at work: That al Qaeda has found it prudent and necessary to provide retroactive justification for some of its attacks suggests that new norms, boundaries, and thresholds are being defi ned and defended. Again, consider AQI’s 2005 hotel bombings in Jordan. Initially, al-Zarqawi explained that the attack had put the United States on notice that its “backyard [military] camp” was in range

Deterring Terrorism

73

of Iraqi militants. But after AQI’s public denunciation, al-Zarqawi backpeddled, fast. He found it necessary to again explain his rationale for targeting the hotels, this time blaming Israel: “Let all know that we have struck only after becoming confident that [the hotels] are centers for launching war on Islam and supporting crusaders’ presence in Iraq and the Arab Peninsula and the presence of the Jews on the land of Palestine. . . . We did not and will not think for one moment to target” fellow Muslims.132 Too little, too late. AQI’s popularity dropped like a stone, Jordanian intelligence was galvanized to help U.S. forces track and kill al-Zarqawi (which they did the following year), and Washington turned the Iraq War back from the abyss.

Deterring Terrorism: What Comes Next If one thing is clear it is that the academic literature on deterring terrorism is heavy on theory but short on rigorous empiricism. There are a lot of competing theoretical claims for how deterrence and coercion, more broadly, can work against terrorists and terrorism, but very few studies actually test or evaluate these propositions. Like the “second wave” of deterrence theory, a large subset of the current literature on deterring terrorism is descriptive rather than illustrative; theory far outpaces empirical work. In 2011, the Influencing Violent Extremist Organization research team at the National Consortium for the Study of Terrorism and Responses to Terrorism (START), University of Maryland, identified 190 separate hypotheses for coercing terrorist behavior.133 Only a fraction had been empirically tested. And of the existing literature covered in this chapter, very little suggests either how deterrence theory might be applied to counterterrorism or how deterrence actually functioned in practice. Furthermore, most of the empirical work on deterring terrorism pays undue attention to deterrence by punishment in the Israeli–Arab/Palestinian theater.134 A handful of other studies empirically evaluate denial and disruption, while still fewer explore positive inducements and delegitimization.135 To date, the most expansive empirical evaluation of “influence operations” in counterterrorism—defined as operations that go well beyond traditional notions of deterrence and range from “pure coercion to co-optation” of militant grievances and demands—was conducted by START. Its 2011 monograph includes dozens of miniature qualitative case studies, from Somalia to Peru, spanning the spectrum of influence operations carried out against violent nonstate actors (from incorporating

74

Chapter 3

militants into the political process to utterly crushing them on the battlefield). Smaller, multiauthor, collaborative projects have also been published by the National Institute for Public Policy (2008) and Stanford University Press (2012); I coedited the latter volume. Both projects offer coercive lessons on punishment and denial derived from various qualitative case studies. The gap between description and evaluation is being bridged, if only slowly. A cumulative research program demands that a literature move forward through substantive evaluation. For the deterring terrorism literature, this is the obvious next step.

CHAPTER 4

Targeted Killings: Theory, Practice, and Consequence

Targeted killings offer one way to empirically examine the theory and practice of deterring contemporary terrorists. Targeted killings are the “intentional slaying” of individual terrorist leaders and facilitators “undertaken with explicit governmental approval.”1 In the years since 9/11, they have become a cornerstone of U.S. counterterrorism policy and strategy. Conceptually, targeted killings represent a cost to planning and participating in terrorism. Accordingly, they offer a practical case study for assessing whether and how deterrence (by punishment) might be applied to counterterrorism. The coercive effect of a sustained campaign of targeted killings on a terrorist group’s behavior can be delineated; it is not difficult to imagine how their use will influence both individual militant behavior and group behavior. Indeed, the literature on the subject suggests that their use diminishes a militant group’s capability and influences a group’s motivation and behavior. In sum, the selective targeting of terrorist leaders and facilitators represents an empirical and a quantifiable case that allows us to assess whether and how a par ticu lar subset of deterrence theory can be applied to counterterrorism. This chapter is divided into five sections. In the fi rst section, targeted killings are defined and various legal and normative considerations are assessed. A brief historical overview of contemporary targeted killings follows. The third and fourth sections assess the potential promises and pitfalls of relying on targeted killings in counterterrorism. And the final section addresses the coercive and deterrent effect targeted killings have on terrorist behavior.

76

Chapter 4

Targeted Killings Defined Targeted killings go by various names: signature strikes, long-range hot pursuit, decapitation strategies, selective targeting, surgical eliminations, targeted thwarting, man hunting, active self-defense, pinpoint prevention, focused disruption, to name but a few.2 They involve pursuing and killing individual militants with deliberate and pinpoint force. In his authoritative work evaluating their international legality, Nils Melzer suggests five definitional elements of targeted killings: They involve the use of lethal force, are designed to target specifically identified individuals (as opposed to targeting a group or using random punishment), are carried out with the deliberate intent to kill the individual in question, are used against individuals not in the physical custody of the state (distinguishing it from judicial execution), and are carried out by states and their central governments.3 Depending on the area of operation, targeted killings in counterterrorism have been carried out by special operations forces, remote-controlled and concealed bombs, and precision missile strikes involving manned and unmanned aerial vehicles (UAVs, commonly referred to as drones). An important distinction needs to be made between drone strikes and targeted killings. Drone strikes are not targeted killings; suggesting as much inaccurately muddies important conceptual definitions that distinguish between a tactic in warfare (targeted killings) and a tool of warfare (drones). Drones are certainly used to conduct targeted killings and may, in fact, greatly facilitate the tracking and targeting of individuals, but they represent one of several ways to target and kill identified militants. It is a mistake, then, to equate all U.S. drone strikes, in Pakistan or Yemen for instance, with targeted killings. There are numerous other ways in which states conduct targeted killings: They can dispatch a special operations team, as the United States did in Somalia in 2009, killing al Qaeda leader Saleh Ali Saleh Nabhan in a helicopter-borne commando raid; set up manned ambushes, as the British Special Boat Ser vice (SBS) did in killing Taliban commander Mullah Abdul Matin in 2008; or plant and detonate hidden explosives near a wanted militant, as Israel did in 1996 when it killed Hamas’s Yahya Ayyash (aka “the Engineer”) by detonating explosives hidden in the cell phone he was using. For their part, armed drones can be used for much more than simply killing wanted militant leaders in precision strikes: They can be used to attack entire groups of suspected militants based on “patterns of life” intel-

Targeted Killings

77

ligence, to help allied forces with air support during a more traditional ground offensive, to strike militant training camps and safe houses (regardless of whether key figures are present), or to destroy weapons caches and trucks shipping militant gear.4 All of these drone strikes may be pinpoint affairs, but they are not targeted killings. Indeed, targeted killings represent only a tiny fraction of what drones are actually used for. P. W. Singer, in Wired for War, illustrates that once the Predator drone was introduced in Iraq, its first missions were to destroy the Iraqi government’s television transmissions. And between June 2005 and June 2006, U.S. Predators in Iraq conducted more than 2,000 separate missions and surveyed nearly 18,500 targets, but they conducted only about 240 raids.5 Targeted killings were only a minor subset of these missions. The same appears to be true for Pakistan. According to media reports and the New America Foundation (which keeps tabs on U.S. drone strikes in Pakistan), in 2010 the United States carried out a record number of Pakistani drone strikes: 122. These attacks were estimated to have killed between 555 and 960 militants. Of those deaths, the foundation found that only eight notable militant leaders were killed. Even then, only one of those leaders, al Qaeda’s Mustafa al-Yazid— better known as Saeed al-Masri—appeared on America’s most-wanted lists.6 Clearly, drone strikes should not be automatically equated with targeted killings. The use of unmanned aircraft and other robots in counterterrorism operations is itself an increasingly contentious issue. But too often “drones” is imprecisely used as a synonym for “targeted killings.” For instance, in discussing President Barack Obama’s legal foundation for using drones and targeted killings in U.S. counterterrorism operations, Harvard University’s Jack Goldsmith writes that the “debate about targeted killing” in the United States is robust and that “the American public broadly approves of what it sees.” He follows next with polling data that show that 83 percent of respondents approve of Obama’s “use of unmanned, ‘drone’ aircraft against terrorist suspects overseas.”7 To be precise, the poll Goldsmith cites did not ask respondents about their views on targeted killings and it did not ask them to give their opinions on the use of drones to target and kill individual terrorists. It simply asked about the use of drones against terrorists living abroad, a blanket question that could be widely interpreted and was likely aimed at gauging public attitudes toward the use of robots in warfare rather than about a  specific tactic in war. Consider further the Washington Post’s extensive, three-piece special report titled “The Permanent War” (2012) that discusses

78

Chapter 4

the manner in which Obama (and Bush) transformed U.S. counterterrorism efforts to more fully incorporate targeted killings. The lengthy articles discuss the development of new blueprints—dubbed the “disposition matrix”— with which U.S. officials identify, screen, and add individuals to U.S. target lists. The articles accompany an interactive, continuously updated, feature on the Washington Post’s webpage called “Tracking America’s Drone War” that includes a database of all (suspected) U.S. drone strikes in Pakistan, Yemen, and Somalia since 2002. Links to media reports for each strike are included.8 Here again, the distinction between targeted killings (a tactic), as outlined in the articles, and drone strikes (a tool), as highlighted by the interactive component, are purposefully (though mistakenly) intertwined. Conceptually, this is a problem. Finally, consider the 2011 Human Rights Watch (HRW) call on Obama to “end CIA drone attacks” and its 2012 call on all world leaders to “ban ‘killer robots’ before it’s too late.” In explaining these initiatives, James Ross, HRW’s legal and policy director, argued: “CIA drone strikes have become an almost daily occurrence around the world, but little is known about who is killed and under what circumstances. So long as the United States resists public accountability for CIA drone strikes, the agency should not be conducting targeted killings.”9 Notwithstanding HRW’s very real concerns, once again the distinction between drone strikes and targeted killings are (in this case, purposefully) glossed over. In evaluating the advantages and disadvantages of using targeted killings in counterterrorism, we need to avoid inadvertently assessing the effects of drone strikes more generally. These are two different—though perhaps equally important—elements of contemporary counterterrorism, but they are not the same. Armed drones are a relatively new development, their deployment in counterterrorism a product of recent advances in technology.10 Targeted killings, however, have been around much longer.

Assassination and Extrajudicial Killing

Given the nature of targeted killings, there is a tendency to associate it with assassination or with extrajudicial killing (or with both). The two are not equivalent, however: Assassination has a “pejorative connotation” of murder and treachery, which should not be assumed a priori when discussing targeted killings, and involves the death of a political, rather than militant,

Targeted Killings

79

leader.11 Some scholars add that extrajudicial killings involve the purposeful murder of an individual who state authorities could otherwise have arrested, which is not usually the case with targeted killings.12 Harvard’s Alan Dershowitz puts it more bluntly: “This view is absurd: All military deaths are extrajudicial.”13 And yet these distinctions are not always clear. Consider comments offered by former U.S. secretary of defense Casper Weinberger: “It is considered lawful in warfare for a skilled and daring soldier . . . to steal into the enemy’s camp and enter the general’s tent and kill him. But it would be a forbidden assassination if someone disguised as the general’s doctor was admitted to his tent, and then killed him.”14 In both of Weinberger’s scenarios the general is killed, though only in the former case is his death presumed lawful and just. The distinction between targeted killings and assassination is further clouded because the latter is usually associated with politically motivated murders by “secret and treacherous means” of government officials, dissidents, and political or social activists. The history of the Cold War is full of cloak-and-dagger assassinations, involving everything from exploding cigars to ricin-tipped umbrellas.15 More recently, a number of Iranian scientists purportedly working on Iran’s clandestine nuclear weapons program have been killed in precision bombings and sniper attacks. Their deaths, more often than not, are deemed assassinations: former U.S. secretary of state Hillary Clinton “categorically” denied U.S. involvement in these attacks, while U.S. State Department spokesperson Victoria Nuland went further, condemning “any assassination or attack on an innocent person.”16 Contemporary campaigns of targeted killings against militant organizations are distinct from those carried out in eras past: Targets are usually mid- to high-level facilitators of terrorism with few existing political inclinations; the eliminations are often carried out in the open with the use of conventional (that is, explosive) weaponry; individual strikes constitute a much larger, iterated, and prolonged campaign and are not, as had been the case, singular events; and selected individuals usually know in advance (or at least suspect) that they are marked fugitives. These last two characteristics are pertinent, especially when deterrence and coercion are to be specifically measured. Recent targeted killings are more akin to ongoing military campaigns than the clandestine assassinations carried out in the 1960s and 1970s. It is the repeated nature of their use that provides a cumulative effect on the motivation of violent nonstate organizations. Furthermore, states often go out of their way to ensure that their

80

Chapter 4

targets are clearly informed that they are marked for death (or capture, when possible). Individual militants who know that they are wanted—by a capable and resolute state, no less—for participating in terrorism are far more likely to alter their behavior than are individuals who remain ignorant of the fact. This taps into the prerequisites of deterrence theory: Threats need to be directly communicated to challengers. Job Henning, writing of U.S. drone use in the New York Times, suggests that for targeted killings to be “adapted” to deterrence doctrine and used to “dissuade or prevent injurious acts,” requires that U.S. adversaries “know that it [the United States] can and will impose a cost if they take a particular course of action.”17 As illustration, consider that during the al Aqsa Intifada Israel compiled regular lists of suspects known to be facilitating terrorism against Israelis. Security officials passed these lists to the Palestinian Authority (PA)—which at the time controlled Gaza and the West Bank—so that arrests could be made. If the PA did not act, which was usually the case, it was understood that Israel would. For its part, the PA regularly leaked the Israeli lists to suspected militants in order to warn them of Israel’s intent. Very often, fugitives voluntarily placed themselves in PA custody to avoid being slain or captured by the Israelis.18 A militant who does not know he is wanted will have few incentives to alter his behavior, as he cannot fully appreciate all that is at stake. However, the prerequisites of deterrence theory also demonstrate that coercive threats may change behavior if (and only if) challengers are also offered a way out. As I have argued, it is not enough to simply threaten a  target; you also need to illustrate that a punishment will be stayed if behavior changes. So in the Israeli illustration, the militants that turned themselves in presumably believed that in doing so they would not be unduly targeted and killed by Israel. David Ignatius, the associate editor of the Washington Post, expands this logic when discussing the threat of U.S. drone strikes against training camps run by AQAP in Yemen and al Shabaab in Somalia. Unlike training camps in Afghan istan, Iraq, and Pakistan, these militant camps have not (yet) been targeted by drones, Ignatius suggests, because the United States is holding them hostage in a coercive bargain with al Qaeda leaders. “There is a deterrence formula implicit in this policy,” he explains.19 So long as al Shabaab and AQAP focus on the local insurgency and refrain from attacking the United States, the drones will be held at bay. Once this red line is crossed, however, specific

Targeted Killings

81

militant leaders and militant infrastructure will be targeted, killed, and destroyed.

International Legal Considerations

Notwithstanding the increasing prevalence in counterterrorism and counterinsurgency of targeted killings, the legality of targeted killings remains uncertain. On the surface, international law prohibits their use. The 1973 UN Convention on the Prevention and Punishment of Crimes Against Internationally Protected Persons, Including Diplomatic Agents bans attacks on state leaders, representatives, and officials, arguing that such acts “create a serious threat to the maintenance of normal international relations.” The idea is to ensure that even states at war retain the diplomatic ability to negotiate with one another.20 In July 2001, in response to Israel’s unprecedented campaign of targeted strikes in Gaza and the West Bank, UN secretarygeneral Kofi Annan called Israel’s action an affront to “international law, in particular to human rights law but also to general principles of law.”21 Even after 9/11, a UN report, published in the wake of a 2002 U.S. Predator drone strike in Yemen that killed Abu Ali al-Harithi—al Qaeda’s mastermind behind the 2000 USS Cole bombing and the 2002 suicide attack on the Limburg, a French oil tanker—called the American strike “truly disturbing,” labeling it “a clear case of extrajudicial killing” and an “alarming precedent.”22 Targeted killings, especially when carried out by aerial strikes, are also often considered an infringement on a state’s sovereignty because they involve the “imposition of extrajudicial punishment” on noncitizens.23 The sovereignty concern is often repeated by Pakistan, for instance, in response to U.S. drone strikes carried out in the west of the country. In August 2012, Pakistan’s high commissioner to London, England, insisted that U.S. drone strikes were “a violation of the UN charter . . . and a clear violation of [Pakistan’s] territorial sovereignty and national integrity.”24 And yet, international law also forbids the use of a state’s territory as a safe haven for terrorist training and planning. Further, it is unlawful for governments to willfully tolerate, even through negligence, terrorist activity on their territory. The 2002 strike in Yemen was justified, advocates might suggest, because the Yemeni government was either unwilling or unable to stop al-Harithi from planning acts of international terrorism. The same

82

Chapter 4

might be said today of Pakistan, Somalia, and other weak and failing states. Furthermore, the Charter of the United Nations, in Article 51, recognizes the inherent right of all states to self-defense: “Nothing . . . shall impair the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations.”25 Self-defense is central to debates over targeted killings, too. The 2010 UN Report of the Special Rapporteur on Extrajudicial, Summary, or Arbitrary Execution notes that states can “invoke the right to self-defence as justification for the extraterritorial use of force involving targeted killings.” Legal issues have arisen, however, as to whether the right to self-defense covers the use of force against nonstate actors (as opposed to states), the extent to which self-defense alone justifies targeted killings, and concerning a state’s right to “anticipatory” or “preemptive” self-defense vis-à-vis suspected terrorists.26 Accordingly, when weighing the legality of targeted killings we must take into account other important considerations, like peace and war. Counterterrorism rests somewhere between them. In Canada, the legal debate over targeted killings pivots on the peacewar axis, though a clear and careful demarcation between the assassinations of noncombatants versus the targeting of combatants is also included. The Canadian Law of Armed Conflict at the Operational and Tactical Level reads: “Assassination is prohibited. It is not forbidden, however, to send a detachment or individual members of the armed forces to kill, by sudden attack, a person who is a combatant.”27 It is in this definition where the specter of legally eliminating individuals, even by secret and surprise attack, becomes evident. The Law of Armed Conflict touches on a broader debate concerning the legal characterization of terrorists. If they are considered enemy combatants actively engaged in preparation for conflict and war against particular states, then selectively targeting and killing them is permissible under traditional conceptions of warfare. At issue, however, is the fact that, unlike soldiers, militants rarely wear discernible military uniforms, they use violence devoid of government sanction, they are usually active in a civilian setting, they prepare and conduct acts of warfare in unconventional manners, and they generally reject normative and historical conventions of modern conflict (by purposefully neglecting to differentiate between civilian and military targets, by using human shields, and by purposefully seeking urban cover to invite humanitarian devastation). As Professor Frank Harvey of Canada suggests, modern terrorists “are not soldiers in the traditional sense. . . . The war on terror is not

Targeted Killings

83

a conventional battle between factions representing two states.”28 In all but name, terrorists are combatants. In the current state of affairs, where terrorist organizations have the intent and capability to carry out mass-casualty attacks, governments are required to take measures, including targeted killings, deemed necessary to protect national interests and the safety and wellbeing of their citizens. In defending its pre-9/11 policy of targeted killings, Israel weighed in on the gray-zone straddling peace and war. Colonel Daniel Reisner, head of the international law branch of the Israeli Army’s legal division, noted that international law “only recognizes two situations: peace or war. But life isn’t as simple.”29 To be precise, Israel was not at war with Palestinian terrorist groups, and the Palestinians had neither a state apparatus nor a functioning military. And yet, Israel was certainly involved in an armed and deadly conflict and was, as a result, legally permitted to target and kill enemy combatants. Today, similar assessments are being made in the United States: John Norton Moore, a professor at the University of Virginia School of Law, concludes that if “one is lawfully engaged in armed hostility, it is not ‘assassination’ to target individuals who are combatants.”30 Indeed, in any state of war, American—and, for that matter, Canadian, British, Israeli, and French— decision makers not only have the right but the obligation to take decisive action to defend their citizens.

American Legal Considerations

In the United States, assassination is not illegal. Though the U.S. government, by executive order, has clearly and vocally renounced assassination, no American law explicitly bans its use.31 Instead, initiatives simply restrict the use of assassination to particularly extreme cases decided on by the highest authorities of government. Most of these restrictions came as a reaction to the Phoenix Program—a joint CIA–South Vietnamese attempt to pressure Viet Cong fighters with coercion and death during the Vietnam War. Estimates suggest that the program resulted in the incarceration of more than 30,000 individuals, the death of more than 25,000, and the “turning” of another 20,000. Tran Do, a Viet Cong Communist commander at the time, admitted that the program was “extremely destructive” to the command-andcontrol structure of the Viet Cong, though the program could not ultimately secure American military victory in the long run.32 Notwithstanding its

84

Chapter 4

supposed successes and failures, the Phoenix Program was extremely unpopular in the United States. In part, because of widespread public backlash, a Congressional investigation into the program began in the early 1970s. The Church Committee (chaired by Senator Frank Church) was tasked with leading the inquiry into the CIA’s efforts in Vietnam. The committee eventually ruled against the use of assassination by U.S. military forces and individuals, condemning it as a tool and tactic of U.S. foreign policy. Assassination, the committee’s final report reads, “violates moral precepts fundamental to our way of life,” contradicts traditional American notions of “fair play,” and is “incompatible with American principles, international order, and morality.” As a result, U.S. president Gerald Ford issued Executive Order 11905 in 1976. It included a “Prohibition of Assassination,” stating, “No employee of the United States Government shall engage in, or conspire to engage in, political assassination.”33 U.S. presidents Jimmy Carter and Ronald Reagan followed suit with their own initiatives. And yet, the great advantage of executive orders is that they can be repealed. In essence, any sitting U.S. president can reinterpret, modify, or rescind the ban on assassination in order to better reflect the evolving international security environment. More important, perhaps, assassination was never formally defined, giving decision makers a wider berth than one might expect when contemplating the elimination of an adversary’s leadership in times of war and peace. For instance, Reagan adopted an interpretation of the assassination ban that exempted death “incidental to a military action.” Thus, Reagan’s strike against Colonel Muammar Qaddafi’s compound (and personal tent) in April 1986, in retaliation for Libya’s involvement in the terrorist bombing of La Belle discotheque in Berlin, Germany, was not an attempt to target and kill the Libyan ruler. But Reagan was not particularly coy about his intentions either, adding that he did not “think any of us would have shed any tears” had Qaddafi been killed.34 Wink . . . nod. President George H. W. Bush offered much the same sentiment after he ordered the bombing of Saddam Hussein’s compound in 1991 during the first Gulf War: “We’re not in the position of targeting Saddam Hussein,” he professed, “but no one will weep for him when he is gone.”35 Not surprising, a sea change of sorts hit Washington in the months and years following al Qaeda’s attack on 9/11. For instance, only months before the 9/11 attacks, U.S. ambassador to Israel Martin Indyk, in response to Israel’s policy of targeting Palestinian militant leaders, commented that “the United States government is very clearly on the record as against targeted assassinations . . . they are extraju-

Targeted Killings

85

dicial killings, and we do not support that.”36 Fast-forward to U.S. secretary of defense Leon Panetta’s November 2012 speech at the Center for a New American Security: “Over the last few years, al-Qaeda’s leadership, their ranks, have been decimated. Through what has probably been the most precise campaign in the history of warfare . . . numerous other experienced operational terrorists and commanders . . . have been killed, or captured. This pressure has significantly demoralized and weakened al-Qaeda . . . [a]nd it seriously disrupted their active plotting against our homeland.”37 Panetta’s comments suggest that targeted killings may be here to stay. Indeed, since 9/11 both George W. Bush and Barack Obama have used their freedom to reinterpret the ban on assassination to help shape U.S. strategies for confronting al Qaeda and other terrorist organizations.38 Bush, for example, claimed the right to kill terrorist leaders under war powers given to the president by Congress in the aftermath of the 9/11 attacks. The Joint Resolution of Congress (otherwise known as the Authorization for Use of Military Force) passed on September 14, 2001, and was signed into law on September 18. It gave Bush authority “to use all necessary and appropriate force against those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons, in order to prevent any future acts of international terrorism against the United States.”39 The day before signing the resolution, Bush issued his famous declaration that Osama bin Laden was “Wanted, dead or alive.” Obama, who has quietly championed the use of targeted killings in counterterrorism and counterinsurgency operations and has put drone technology to unprecedented use in Pakistan, Yemen, Somalia, North and West Africa, Iraq, and elsewhere around the globe, also relies on the 2001 resolution and other legal and court provisions to do what he must. And, of course, Obama followed through with Bush’s declaration to hunt bin Laden down by dispatching Navy SEALs to Abbottabad, Pakistan, on May 2, 2011, where the al Qaeda leader was ultimately killed. Indeed, Obama, to the surprise (and chagrin) of some of his admirers, has taken a particularly hawkish position on targeting killings.40 In clarifying the Obama administration’s evolving policy to target not only foreignborn militants but American citizens as well, U.S. attorney general Eric Holder in a March 2012 speech at Northwestern University School of Law explained: “Let me be clear: An operation using lethal force in a foreign country, targeted against a U.S. citizen who is a senior operational leader of

86

Chapter 4

al Qaeda or associated forces, and who is actively engaged in planning to kill Americans, would be lawful at least in the following circumstances: First, the U.S. government has determined . . . that the individual poses an imminent threat of violent attack against the United States; second, capture is not feasible; and third, the operation would be conducted in a manner consistent with applicable law of war principles.”41 The targeted elimination of two U.S.-born al Qaeda militants—al-Awlaki (the charismatic, militant propagandist who Obama described as “the leader of external operations” for AQAP) and Samir Khan (an American citizen who was the editor of AQAP’s English-language propaganda magazine, Inspire)—in a CIA-operated drone strike in Yemen on September 30, 2011, illustrates Holder’s broader point. These Americans, the Obama administration has argued, were actively engaged in jihadi violence against the United States and were deemed legitimate targets for death as a result. But the death a few weeks later of al-Awlaki’s 16-year-old son Abdulrahman—a Denver-born American citizen living in Yemen—in a U.S. military strike raises other ethical questions. Is it just or moral for states to use coercive punishment against individuals? In which of the numerous countries where al Qaeda’s presence is suspected or known should the United States and its allies carry out targeted killings? How are targets for death determined and vetted and who has oversight over the process? What boundaries should inform targeted killing policy? What precautions exist—and which ones have yet to be developed—that limit the risk of collateral damage in aerial strikes? The case of Abdulrahman al-Awlaki is particularly troubling. Unlike his father, who denounced his American citizenship, preached a vile and dangerous militant ideology, and is blamed for indoctrinating dozens of western jihadis, including U.S. Army major Nidal Hasan (the 2009 Fort Hood shooter), Pakistani American Faisal Shahzad (the 2010 Times Square bomber), and Nigerian Umar Farouk Abdulmutallab (the 2009 “underwear bomber”), Abdulrahman was not likely a prominent or even peripheral al Qaeda leader. Writing in the New Yorker, Amy Davidson asks: “Where does the Obama administration see the limits of its right to kill an American citizen without a trial? . . . And what are the protections for an American child? You can . . . blame Abdulrahman’s death on his father—for not staying in Colorado, for introducing his son to the wrong people, for being who he was. That would be a fair part of an assessment of Anwar al-Awlaki’s character. But it’s not sufficient. He may have put his child in a bad situation, but we were the ones with the drone. One fault does not preclude another.”42 For

Targeted Killings

87

liberal democracies these questions are not purely academic. Harvard’s Dershowitz, for instance, warns that targeted killings are “essentially lawless”: “Those who authorize the hit are prosecutor, judge, and jury—and there is no appeal.”43 As a result, targeted killings seem to rest outside the realm of acceptable democratic behavior, where civil liberties, due process, and constitutional limitation are enshrined. Even in cases where legal and civilian oversight is given over the planning and use of targeted strikes, normative dilemmas endure. Some scholars insist they are “morally indefensible” regardless of whether they work: The danger is that in sanctioning their use states sacrifice their own “moral compass” in the process.44 These queries and the normative critique more broadly, represent pertinent and difficult philosophical questions that need to be more fully explored and addressed by political theorists and decision makers alike. Finally, other questions persist. How wide should the United States draw the circle around al-Zawahiri and other members of al Qaeda’s leadership? Should it target only the ticking bombs—those actively planning acts of violence—or are individuals who help prepare terrorist attacks—the “ticking infrastructure”—legitimate targets?45 What about mid-level commanders? Financiers? State supporters? Religious legitimizers? Community leaders? Properly exploring these questions may be central to determining whether targeted killings are moral and legal. For Professor Amos Guiora, who worked as an Israel Defense Forces legal adviser on the use of targeted killings and is a self-described advocate, purposefully developing and implementing a “criteria-based decision-making model” that determines who is added to the kill list (and why), is crucial: “It is the characterization of who we target and when—and how that determination is made—that raises serious questions of law and morality.” How do we know we are targeting the right people? Only when states have constructed internal review processes that properly identify targets and minimize civilian casualties might targeted killings be considered legal and moral. The reason rests on the nature of contemporary terrorist conflicts: Fighting terrorism, I suggested earlier, fits into the gray zone that straddles peace and war, that area that rests between traditional and less traditional conflict. Within this framework, specifically identifying individuals who pose an imminent threat to security helps determine the legality and legitimacy of targeted killings. By illustration Guiora suggests that Israel, having relied on targeted killings for decades, has gone to some length to ensure its target lists meet the highest standards of a criteria-based process. Not so in the United States. Under Obama, Guiora

88

Chapter 4

and others argue, those deemed “likely” to be members of a terrorist group are defined as legitimate targets for death. But guilt by association and a flexible interpretation of what constitutes an imminent threat may not be enough.46 To be fair, change is coming. In a May 2013 speech, Obama signaled a policy shift that would, among other things, codify and institutionalize the use of U.S. drones in counterterrorism.47 The announcement was matched with a dramatic decline in the number of U.S. drone strikes in Yemen, Pakistan, and Somalia. Time will tell whether these shifts satisfy the various ethical, normative, and legal dilemmas critics have rightly underscored.

Targeted Killings in Practice Notwithstanding these legal and moral concerns, targeted killings have become a tactic of choice for countering nonstate militant organizations. This is especially evident in the post-9/11 era. To highlight how prolific and widespread the tactic has become, a selection of the more pertinent examples of targeted killings, from a variety of recent and ongoing conflicts, is offered. Though the list is not exhaustive, it does show just how far-reaching the practice has become.

Israel

Perhaps more than any other state, Israel has refined the practice of selectively targeting individual terrorists. Even before the al Aqsa Intifada, Israel aggressively pursued individual terrorist facilitators all over the world. In the 1950s, Israel killed senior Egyptian military officials responsible for planning operations against Israelis by sending them mail and package bombs.48 In the 1960s, package bombs were used against German scientists developing missile technology on behalf of the Egyptian government, prompting some German scientists to return to Europe. In response to the massacre at the 1972 Munich Olympics, Israel eliminated Black September as a functioning terrorist orga nization by selectively killing most of its leaders and members over a period of a decade. In 1973, Israeli commandoes, some apparently dressed as women, strolled around Beirut, Lebanon, hunting down Fatah members, killing Arafat’s deputy, Mohammed Yusuf al-Najjar and Fatah spokesperson, Kamal Nasir. In 1979, a suspected Israeli

Targeted Killings

89

bomb planted in Beirut killed Ali Hassan Salameh, Fatah’s top commander. In the 1980s and following Israel’s intervention in southern Lebanon in operations meant to halt PLO activities, Israel tried on a number of occasions to kill Yasser Arafat. In 1988, commandos killed Arafat’s top aide, Khalil al-Wazi (Abu Jihad), in his home in Tunisia—Israel officially acknowledged the attack in 2012.49 Israel targeted Hezbollah leaders Ragheb Harb in 1984 and Abbas al-Musawi in 1992, gunned down Palestinian Islamic Jihad’s (PIJ) founding leader, Fathi Shaqaqi, in Malta in 1995, and killed notorious Hamas bomb maker Ayyash in Gaza with an exploding cell phone in 1996. The following year, two Israeli Mossad agents were caught in Jordan after poisoning Hamas leader Khaled Mashal. In an embarrassing turnabout, Israel was forced to provide Mashal with the antidote and release Sheikh Ahmed Yassin (the “spiritual leader” of Hamas) in exchange for the captured agents. (Israel later killed Yassin in a 2004 aerial strike in Gaza City.) During the al Aqsa Intifada—a conflict in which more than 130 suicide bombings took place within Israel (and many more were foiled or disrupted)— Israel greatly increased the tempo of its targeted killings. In the first year alone, reports suggest that Israel conducted at least forty targeted killings. Since then, it is believed that well over 200 Palestinian terrorist facilitators have been targeted in precision strikes and operations.50 Unlike Israel’s previous targeted attacks—many of which were conducted clandestinely by small groups of Special Forces active overseas—those that took place during the 2000–5 conflict were more often conducted in the open with sophisticated aerial weaponry. For instance, one of the first targeted attacks, which took place in November 2000 against Hussein Abayat, a Tanzim leader, saw a missile fired from an Israeli helicopter in broad daylight. In the years following Abayat’s death, Israel continued its policy of tracking and targeting known terrorist facilitators. As Steven David notes, most of those targeted were “mid-level fighters, important enough to disrupt a terrorist cell but not so important as to provoke murderous retaliation,” or, one might add, international condemnation.51 Nonetheless, Israel also carried out several very high-level attacks, killing Salah Shehadeh, the leader of Hamas’s Izz ad-Din al-Qassam Brigades in 2003, Yassin in 2004, and Dr. Abdel Aziz al-Rantissi, Yassin’s successor, also in 2004. In most cases, those targeted by Israel knew in advance that they had been marked for capture or death. After the al Aqsa Intifada ended, Israel continued its policy of targeting suspected Palestinian and Islamist terrorists. During its 2006 war with

90

Chapter 4

Hezbollah, for instance, the Israel Defense Forces (IDF) twice bombed bunkers where Hassan Nasrallah, Hezbollah’s secretary general, was thought to be hiding. He escaped. “The third time, we won’t miss” warned General Dan Halutz.52 In 2008, Hezbollah’s top commander Imad Mughniyeh, was blown up in a car bomb in Damascus, Syria (with suspicion falling on Israel). In January 2009, during Israel’s Operation Cast Lead, Hamas’s interior minister, Said Seyam, was killed in an IDF strike on his brother’s home. In 2010, Mahmoud al-Mabhouh, a senior Hamas leader, was killed in his Dubai hotel room; though Israel denied involvement, the attack has been attributed to it. Also in 2010, Israeli air strikes killed Islam Yassin and Mohammed Nimnim, purportedly senior members of a Gaza-based al Qaeda ally, Army of Islam. And in April 2011, Sudan blamed Israel for an air strike that killed two passengers traveling in a car near the Red Sea city of Port Sudan. Though neither Sudan nor Israel has confirmed the victims’ identities, Israeli analysts have suggested the target was a Hamas weapon’s supplier.53 In March 2012, Israeli missiles killed Zohair al-Qaisi, head of the Popular Resistance Committee, as he was driving his car near Gaza City. His death sparked a week-long military skirmish between the IDF and Gaza militants. A few months later, in the (literal) opening shot of Operation Pillar of Defense (November 2012), the IDF bombed the car Hamas chief Ahmed Jabari was traveling in. He was killed instantly. Jabari had replaced Mohammed Deif as Hamas’s top military leader after he was critically wounded in a 2006 strike. Jabari purportedly survived numerous previous attempts on his life, including a 2004 air strike on his home that killed several relatives. He also escaped Israeli attempts to nab him, including a 2008 operation that went sour only after his car turned down the wrong street.54 Jabari’s eventual elimination sparked an eight-day military conflict between Hamas and Israel.

Some of the Others

Besides Israel, a number of other states have used targeted killings. In 1988, a British Special Air Ser vice (SAS) team attacked and killed a number of (reportedly unarmed) IRA militants in Gibraltar. In the decades before 9/11, the United States also carried out precision strikes and operations. In 1985, for example, Americans were likely involved in the car bombing of Mohammed Hussein Fadlallah, a Hezbollah leader implicated in the U.S. marine barracks attack in 1983. In 1989, Abdullah Azzam, bin Laden’s mentor, was

Targeted Killings

91

killed in a bomb blast many suspect was at least coordinated by the United States.55 In response to the 1998 U.S. Embassy bombings in Kenya and Tanzania, American forces targeted bin Laden in a cruise missile attack on his Afghan compounds and training facilities. In 1991, 1993, and 2003, U.S. forces bombed Saddam Hussein’s official residences, including his command bunkers. The United States did much the same against Serbian President Slobodan Milošević, striking his personal villa during the 1999 Kosovo air campaign. And in 1993, Colombian police, reportedly with American assistance, cornered and killed Pablo Escobar, a notorious drug baron. Colombia has gone on to successfully target and kill several FARC leaders in recent years. For instance, Tomas Medina Caracas, who headed FARC’s smuggling operations, was killed in 2007. Raúl Reyes (Luis Edgar Devia), a FARC Secretariat member, was killed in 2008. In a massive 2010 military assault on a FARC jungle base, Colombian soldiers killed Mono Jojoy (Jorge Briceno), the group’s military commander. And finally, Alfonso Cano, FARC’s No. 1, was killed by Colombian forces in 2011. Peru, too, with apparent help from the CIA, successfully tracked (and captured) Shining Path leader Dr. Abimael Guzman in 1992. Turkey captured Abdullah Ocalan, the Kurdistan Workers Party (PKK) leader, in a global operation in 1999, and, more recently, in 2008, used Special Forces to kill a number of PKK leaders active in northern Iraq, including Saban Aktas, Mustafa Iscan, and Abdulkerim Ertas. Iran also has targeted Kurdish militant leaders. Most notably, in 1989, after arranging to meet and negotiate with Dr. Abdul Rahman Ghassemlou, leader of the Kurdistan Democratic Party of Iran (KDPI), in Vienna, Austria, Iranian agents killed him instead. His successor, Dr. Sadegh Sharafkandi, was later killed in Berlin, Germany. Russia, too, is known to have targeted militant leaders, killing Zelimkhan Yandarbiyev (placed on the UN Security Council’s terrorist blacklist for his role in the Second Chechen War) in 2004 by strapping a bomb to his SUV in Qatar. Moscow also targeted and killed Aslan Maskhadov, a former president of Chechnya, who led insurgents during the Second Chechen War in 2005; Shamil Basayev, mastermind behind the 2004 Beslan school attack, in 2006; Samir Saleh Al-Suwailem (Emir Khattab), a Chechnyan terrorist leader, in 2002; Khattab’s successor, Abu Walid, with a guided missile in 2004; AbdulKhalim Sadulayev, a top Chechnyan commander, in a 2006 Special Forces operation; and Said Buryatsky (Alexander Tikhomirov), locally dubbed the “Russian Bin Laden,” in 2010. And France, just weeks after its military intervened in Mali in January 2013, reportedly killed Abou Zeid, a top Algerian

92

Chapter 4

leader of al Qaeda in the Islamic Maghreb (AQIM). The Philippines, too, have carried out a string of operations against the Abu Sayyaf Group (ASG), an al Qaeda–allied terrorist organization. In 1998, government forces killed Abdurajak Janjalani, the organization’s founding leader, in a raid. Abu Sabaya and Ghalib Andang, both one-time ASG leaders, were killed and captured in 2002 and 2003, respectively. And between 2006 and 2007, in an unrelenting campaign to hunt down members of the group, government forces killed Binang Sali, an ASG bomb expert; Jundam Jamalul, one of the group’s regional commanders; and Khaddafy Janjalani, brother of Abdurajak and ASG’s then leader. Still later, Philippine forces killed Abu Sulaiman, the mastermind behind the 2004 Manila Bay ferry bombing, in January 2007. Finally, in Saudi Arabia, in a 2004 raid, government forces killed Abdul Aziz al-Muqrin, leader of al Qaeda’s local branch and mastermind behind a string of attacks in Riyadh.

The United States, Post-9/11

It was the 2001 felling of the Twin Towers, however, that introduced a global and more widespread use of targeted killings in counterterrorism and counterinsurgency. Besides the most recent campaign of targeted killings in Afghanistan (which is largely discussed in the following chapters), al Qaeda’s global leadership has been decimated since 2001. The United States has led the charge and the list of those targeted is, to be blunt, a long one.56 Following are some of the more prominent and high-profi le targets. Mullah Mohammed Omar, the Taliban’s chief and al Qaeda’s top Afghan ally, escaped a strike on his compound in 2001. Muhammad Atef (Abu Hafs al-Masri) was killed in a Kabul air strike in 2001. Al-Harithi, introduced earlier, was killed by a CIA drone in Yemen in 2002. KSM, “architect of 9/11,” was caught in Pakistan in 2003, while Ali Abd al-Aziz (Ammar al-Baluchi), 9/11’s primary bagman, and al Qaeda lieutenant, Tawfiq Bin Attash, were both captured in Pakistan in 2003. Taliban commander Nek Mohammed was killed in a U.S. drone strike in South Waziristan, Pakistan, in 2004. Mohammed’s death marked the first known U.S. drone strike to occur in Pakistan. Hamza Rabia and Haitham Yemeni were both killed by guided Predator missiles in Pakistan in 2005. Abu Farraj al-Libbi, al Qaeda’s No. 3, was captured in a 2005 sting operation in Pakistan. Al Qaeda in Iraq’s (AQI) leader, al-Zarqawi, was killed in a 2006 air strike in Iraq. Mullah

Targeted Killings

93

Akhtar Mohammad Osmani, a top aide to Mullah Omar and a leading Taliban commander, was killed in an air strike in Afghanistan in 2006. Al Qaeda’s al-Zawahiri survived a 2006 U.S. drone strike in Pakistan’s remote Bajaur region. Abu Laith al-Libi, al Qaeda’s Taliban liaison, was killed in a 2008 Predator drone strike in North Waziristan. Abu Khabab al-Masri (Midhat Mursi al-Sayid Umar), al Qaeda’s “chemical weapons expert” who had a $5 million bounty on his head, was killed in a 2008 drone strike in Pakistan. Taliban commanders Eida Khan and Wahweed Ullah were killed in Predator drone strikes in 2008 in Pakistan. Osama al-Kini (Fahid Msalam), a Kenyan al Qaeda leader who helped plan the 1998 U.S. Embassy bombings in East Africa, was killed in 2009. Tehrik-e-Taliban (TTP) leader Baitullah Mehsud was killed in a U.S. drone strike in South Waziristan in 2009. His successor, Hakimullah Mehsud (Baitullah’s cousin) was repeatedly targeted by U.S. forces and was eventually killed in November 2013.57 Mustafa Abu al-Yazid (Saeed al-Masri), al Qaeda’s operational leader for Afghanistan and a top-echelon al Qaeda figure (some experts suggest he was al Qaeda’s No. 3 at the time of his death), was killed in a missile strike in May 2010. He was replaced by Egyptian Sheikh Fateh al Masri, who lasted only months in his new role before he, too, was killed in a U.S. drone strike in September 2010. Mohammed Amin, a leader of the Islamic Movement of Uzbekistan who served as a Taliban deputy shadow governor for Takhar province, was killed in a September 2010 strike. Abdallah Umar al-Qurayshi, a senior al Qaeda commander from Saudi Arabia who led Arab militants in eastern Afghanistan, was killed in an International Security Assistance Force (ISAF) operation in 2010 in eastern Afghanistan. The strike also killed al Qaeda explosives expert Abu Atta al Kuwaiti. Abu Miqdad al-Masri, an al Qaeda veteran and associate to bin Laden, was killed in October 2011 in a drone strike in Pakistan. Abu Hafs al-Najdi (Abdul Ghani), who directed al Qaeda’s operations in Kunar Province, Afghanistan, was killed in a NATO air strike in April 2011. Ilyas Kashmiri, the operational commander of Harkat-ul-Jihadal-Islami (HuJI), a blacklisted al Qaeda affiliate seeking to incorporate Jammu and Kashmir into Pakistan, was killed in a June 2011 drone strike in Pakistan. However, although U.S. officials confirmed his death in July 2011, he was purportedly sighted in 2012, meeting with TTP leaders in Pakistan.58 As noted earlier, AQAP propagandists al-Awlaki and Khan were killed in a Yemen strike in September 2011. Atiyah Abd al-Rahman, a seasoned militant who rose to al Qaeda’s No. 2 position after bin Laden’s 2011 killing, was himself killed in Pakistan in August 2011. His successor, Abu Yahya

94

Chapter 4

al-Libi, was also killed in a CIA drone strike in Pakistan in June 2012. Aslam Awan, a Pakistani national who spent several years living in Britain and was considered an al Qaeda “external operations planner” was killed in a drone strike in January 2012 in Pakistan. Fahd Mohammed Ahmed al-Quso (or al-Qusaa), a militant listed on the FBI’s most-wanted list who was indicted for his role in the 2000 bombing of the USS Cole and appeared in an al Qaeda video taking credit for the 2009 “underwear plot” against the United States, was killed in a drone strike in Yemen in May 2012. The strike ended al-Quso’s illustrious career of repeatedly evading death and capture: He escaped from a Yemeni prison in 2003 and apparently survived targeted air strikes in 2009, 2010, and 2011. Badruddin Haqqani, who organized the Haqqani network’s daily operations, was killed in August 2010 in a strike in Pakistan’s North Waziristan. The Haqqani network was founded by Badruddin’s father, Jalaluddin Haqqani, and is primarily run by Badruddin’s brothers, Sirajuddin and Nasiruddin. Blacklisted in 2012 by the U.S., the Haqqani network is responsible for a number of spectacular attacks, including complex suicide bombings against the American and Indian embassies and NATO headquarters in Kabul. Adnan al-Qadhi, an al Qaeda operative who was involved in planning the 2008 attack on the U.S. Embassy in Sana’a, Yemen, was killed in 2012 in a drone strike outside Yemen’s capital. And in January and February 2013, U.S. strikes killed Taliban leaders Mullah Nazir, Faisal Khan, and Wali Mohamed, and al Qaeda leader Yasi al Kuwaiti. In Somalia, Abu Taha al Sudani, Fazul Mohammed, and others were killed in a joint Ethiopian-American strike in 2007; Hassan Turki was killed in 2008; and Aden Hashi Ayro (head of al Qaeda ally al-Shabaab) was killed in May 2008. In April 2011, Ahmed Abdulkadir Warsame, a conduit between al Qaeda and al-Shabaab, was captured as he traveled by boat from Yemen to Somalia. He was interrogated abroad by the United States for two months before being sent to New York City in July 2011 to be tried in a civilian court. Fazul Abdullah Mohamed, al Qaeda’s leader in East Africa and a top commander for al-Shabaab responsible for the U.S. Embassy bombings in Kenya and Tanzania and the 2002 attacks on Israeli tourists in Kenya, was killed at a security checkpoint in Somalia in 2011.59 Mohamed’s close associate, Bilal al-Barjawi, was killed in a 2012 U.S. drone strike near the Somali capital. Al-Barjawi grew up near London, England. Apparently, the deadly strike took place soon after al-Barjawi “was stupidly brazen enough” to call his wife on his cell phone to congratulate her on giving birth in a London hospital.60 And in the weeks following al-Shabaab’s brutal September 2013

Targeted Killings

95

attack on Westgate shopping mall in Nairobi, Kenya, U.S. Special Forces failed to capture Abdulkadir Mohamed Abdulkadir (aka Ikrima), thought to have directed a number of al-Shabaab’s attacks in Kenya, but did kill Ibrahim Ali, the group’s most prominent explosives expert, in an air strike. British Pakistani Rashid Rauf, a purported ringleader of the 2006 UK-based plot to destroy trans-Atlantic airliners with liquid explosives, was targeted and thought killed in November 2008 in Pakistan. Though some analysts suspected he survived the attack, his family suggested otherwise in 2012.61 Two other prominent British terrorism suspects, Mohammed Khan and Ibrahim Adam, were killed by drone strikes in 2010 and 2011, respectively. And Briton Abdul Jabbar, apparently tapped in 2010 by al Qaeda leaders in Pakistan to direct a new Qaeda offshoot, the Islamic Army of Great Britain, for “Mumbai-style” attacks in Europe, was killed in a September 2010 drone strike. As intelligence of the European plot was uncovered, a flurry of U.S. drone strikes pounded Pakistan (security alerts, threat-level assessments, and travel advisories were issued and raised in the West). In September 2010 alone, over two dozen drone strikes were carried out in Pakistan, a record, and in October, drones killed several militants with German citizenship. The plot unraveled soon afterward.62 Arguably, selectively targeting and killing suspected terrorist and militant leaders and facilitators has become a cornerstone of U.S. counterterrorism policy and strategy. What follows is a discussion of the practical and coercive consequences targeted killings have on the capability and behavior of terrorist organizations.

Why Targeted Killings Work Targeted killings affect a militant group’s operational capability and degrade motivation in at least five distinct ways.

Elimination of Threat

First, for smaller terrorist groups and individual cells—where leadership, knowledge, and power is likely more centralized than it would be in a larger organization—targeted eliminations can effectively destroy the group. Without quickly replacing fallen leaders, small organizations may well disband

96

Chapter 4

and cease to exist. Likewise, militant organizations founded or led by particularly charismatic leaders may also be especially susceptible to decapitating strikes—assaults remove an essential strategic, ideological, or practical linchpin from the organization.63 This is particularly relevant if and when leaders play a critical or an oversized role in keeping the group together and on track. This assumption partly rests on the notion that terrorist organizations depend on the work of a few vital individuals. Isaac Ben-Israel and colleagues note, for instance, that the number of key personnel in Hamas—those members that are actively “engaged in preparing an act of terror”—number in the low hundreds. A state only needs to “neutralize” a quarter of them, they explain, for the organization’s campaign of violence to suffer dramatically.64 The argument holds that terrorism is a process that requires a production line of activity—from scouting targets to preparing bombers—if a campaign of violence is to take place. This line of reasoning taps into the organizational view of terrorism introduced earlier: Terrorism is the product of individual parts and processes fulfilling particular jobs and functions. Destroy or eliminate certain parts, and the whole is likely to suffer as a result. Consider the regression of the Shining Path following the 1992 capture of its leader Guzman in Peru; the PKK’s stumble following Ocalan’s arrest in Kenya; the setback for Palestinian Islamic Jihad following Israel’s 1995 strike against Shikaki; and the complete collapse of a-Saika, a Syrian controlled faction of the PLO, after the mysterious death of its leader Zuheir Mohsein in France in 1979.65 Al Qaeda, too, is susceptible here. The 2006 removal of AQI’s al-Zarqawi helped initiate the organization’s precipitous decline. Though other factors, including the 2007–8 U.S. troop surge, internecine rivalries between al Qaeda and other insurgent groups, popular disdain among both Sunni and Shi’a communities for AQI’s murderous rampages, and the deployment of more robust indigenous Iraqi security forces certainly played a role in AQI’s decay, the replacement of charismatic al-Zarqawi with ho-hum Abu Hamza al-Muhajir (aka Abu Ayyub al-Masri) had anything but a positive effect on the group’s immediate future. Less than two years after alMasri’s promotion, the United States dropped its bounty on al-Masri from $5 million to a measly $100,000, a clear sign that U.S. and Iraqi security officials did not think much of his capabilities.66 Only after al-Masri’s death in 2010 (during a joint Iraqi-American raid on his safe house) and AQI’s eventual renewal between 2011and 2014 under the stewardship of Ibrahim Awwad Ibrahim Ali Badri (aka Abu Du’a) has AQI slowly clawed its way back from near extinction. Spurred on, in part, by Islamist and jihadi successes in neigh-

Targeted Killings

97

boring Syria, AQI (renamed the Islamic State of Iraq and the Levant) bounced back considerably in Spring 2014. An indication of AQI’s recovery came with the U.S. State Department’s bounty for Abu Du’a: $10 million (twice that offered for al-Zarqawi). That was at par with other terrorist heavyweights, like the Taliban’s Mullah Omar and Lashkar-e-Taiba’s (LeT) Hafiz Mohammad Saeed (the suspected mastermind behind LeT’s 2008 attack on Mumbai, India), and just below Public Enemy No. 1, al Qaeda’s al-Zawahiri.67 These Iraqi trends were reiterated in the leaked (and later declassified) U.S. National Intelligence Estimate (NIE), Trends in Global Terrorism: Implications for the United States. Though at the time most media concentrated on the NIE’s finding that the U.S. invasion of Iraq and the subsequent Iraqi jihad had become a cause célèbre for Islamists and a breeding ground for tomorrow’s terrorists, the report offered other important suggestions: “The loss of key leaders in rapid succession, probably would cause the group to fracture into smaller groups.”68 An organization’s immediate effectiveness is certainly affected by the loss of its leadership, though the long-term consequences are less certain. The evolving case of AQI seems to bear these competing lessons out. As for individual terrorist cells, the loss of a leader can be especially crippling. In the weeks following the Madrid train bombings of March 11, 2004, Spain’s security ser vices tracked part of the terrorist cell—including its suspected ringleader, Sarhane Ben Abdelmajid Faked—to an apartment building in Leganes, a suburb of Madrid. After a short gun battle with security officers, the terrorists blew themselves up, destroying their apartment and killing an officer. In the investigation that followed, Spanish authorities found more than 200 detonators of the kind used in the March 11 train attacks and the foiled April 2 train attack, several kilograms of explosives, suicide vests, and a car, packed with explosives, parked outside. As Spain’s interior minister Ángel Acebes expressed at the time, the Madrid bombers “were going to keep on attacking.”69 That Spain had both the investigative and policing ability to track, engage, and trap the group expeditiously neutralized a specific threat in one fell swoop. With a single raid and the removal of a group’s mastermind, a highly capable terrorist threat was altogether removed.

Staying Home and Out of Sight

Second, the threat of assassination forces leaders to look after their personal safety rather than attend to the operational success of their organization.

98

Chapter 4

Targeted individuals are forced to go deeper underground, to distance themselves from other leaders and facilitators of violence, and to restrict their communication with followers lest they inadvertently tip off intelligence officers. In a 2004 letter to bin Laden, AQI’s al-Zarqawi stresses this persistent dilemma: “What is preventing us from making a general call to arms,” he protests, “is the fact that the country of Iraq has no mountains in which to seek refuge, or forest in which to hide. Our presence is apparent and our movement is out in the open. Eyes are everywhere.”70 All of this has a consequence for groups planning terrorism. Diminishing levels of topdown leadership will leave a mark. It takes considerable effort and time to locate safe houses where leaders can hide, plan their affairs, or simply collect their thoughts, while securing safe passage to and from safe houses can be daunting. Ariel Sharon, Israel’s prime minister at the height of the al Aqsa Intifada, explained his country’s use of targeted killing: “The goal . . . is to  place the terrorists in varying situations every day and to ‘unbalance’ them.”71 The threat of twenty-four-hour surveillance—especially high-tech surveillance involving drones and satellite technology—can hinder leaders’ freedom of movement, can throw them off balance, and can divert their energy toward taking precautionary measures rather than planning terrorist operations. In 2009, a Taliban commander gave his fighters three (rather simplistic) instructions for protecting themselves against drones: “mobility, secrecy, and anonymity.” They were to travel in small groups, conduct all communications orally, and avoid satellite and SMS (text messaging) technology. Meetings were to be called at the last minute.72 The threat of precision strikes forced these militants to mask their identities and change their behavior. By targeting and eliminating par ticu lar individuals that fi ll critical positions and forcing others to run and hide, a group’s command and control mechanisms are substantially disrupted. In the meantime, communication between leaders and operators breaks down, complicating both short-term tactical planning and long-term strategic planning. For example, following Israel’s 1992 elimination of Hezbollah’s al-Musawi, the orga nization enforced strict radio silence.73 Although this may have prevented Israel from monitoring the group’s actions and hindered its attempt to track other leaders, it also unquestionably limited Hezbollah’s ability to plan and coordinate acts of terrorism against Israelis. Flipping the argument around, consider what happened when the United States suspended all

Targeted Killings

99

drone strikes in Pakistan—because of deteriorating U.S.-Pakistan diplomatic relations—for more than two months (December 2011 through January 2012), the longest such lull in three years: Al Qaeda and Taliban militants took advantage of the “greater freedom of movement” to develop new strategies, accelerate their attacks, and patch things up between feuding factions.74 There are also psychological consequences to being targeted. Individuals in line to lead militant groups know that they are purposefully placing themselves in the crosshairs. Fear can be paralyzing: “We now often sleep in the river beds or under the eucalyptus trees,” one Pakistani militant explained.75 Living as a fugitive can result in the severance of ties with friends and loved ones. Indeed, in an age of sophisticated signals intelligence (SIGINT), contact with family can prove lethal. E-mail correspondence and phone conversations can be intercepted, revealing a target’s location. Revelations in 2013 concerning the apparent role of the National Security Agency (NSA) in identifying militants abroad and in assisting CIA drone operations seem to bear the argument out. For example, the October 2012 drone strike in Pakistan that killed Hassan Ghul, a close associate of bin Laden’s, was made possible based on information the NSA gleaned from an e-mail exchange between Ghul and his wife.76 Similar intelligence is critical to combating homegrown and domestic terrorism, too. In Canada, for instance, intercepted e-mail correspondence facilitated the arrest of members of the Toronto 18, a group planning atrocities in Ontario, and in foiling Mohammad Khawaja, a would-be Canadian jihadist with ties to a British terrorist cell.77 And life underground, while appealing for some, can get tiring. “Oncepowerful motives to join” a terrorist organization, Christopher Harmon reminds us, “do not always translate into certainty about staying.”78 Fatigue for a given cause can set in; and, with that, a change in behavior and ultimately a diminishment of terrorism. In Italy’s Cold War struggle against radical leftwing terrorist organizations, one prominent counterstrategy was to force leaders and militants deeper underground, denying them the freedom of movement they needed to operate effectively. The effects, according to one member quoted in a study conducted by the National Institute for Public Policy, were grueling: “What I’d really like to do is go away. Just leave, take a long, long trip somewhere, get away, body and mind, somewhere different. I am so tired, and when you enter this long tunnel that life has become, you just need to forget the idea of a future. There are no roads out of here. One

100

Chapter 4

way out, of course, would be the Revolution. But let’s not kid ourselves. More likely, it will be prison. Or worse. You don’t think about it, of course, but then you can hardly imagine going on like this for the rest of your life either. . . . The life we lead does not encourage solidarity, but rather tension, resentment, and constant conflict.”79 By forcing militant leaders to worry about their welfare and well-being, targeted killings help reduce a group’s capability to coordinate acts of violence while diminishing both individual and group confidence and morale.

Deprofessionalization

Third, a prolonged campaign of targeted killings may leave terrorist organizations in general disarray. Finding individuals that are able and willing to replace eliminated leaders takes time and energy, regardless of the fact that not just any substitute will do. Few individuals have the skills needed to effectively manage a militant organization, let alone design and build effective bombs that do not prematurely detonate. “Of 10,000 would-be bin Ladens,” academic Patrick Porter asserts, “most lack ability.”80 The same can be said for mid-level terrorist operators and facilitators—those who forge passports, arrange safe houses, train militants, acquire financial support and weapons, and so on. Although many terrorist leaders are highly educated individuals, holding graduate, legal, and medical degrees—al Qaeda’s Zawahiri is an eye surgeon; Hamas leaders al-Rantissi and Mahmoud al-Zahar both received medical degrees; the Shining Path’s Guzman was a professor of philosophy; PIJ’s Shaqaqi was a medical doctor; the masterminds of the bungled 2007 Glasgow and London car bombings met at the University of Cambridge; and Theodore “Ted” Kaczynski (the “Unabomber”) held a PhD in mathematics from the University of Michigan—higher education does not necessarily translate into solid military and terrorist knowledge. Furthermore, attracting and recruiting the right people to a life of violent hardship that invariably comes with joining a terrorist organization can be difficult. Individuals who have spent years (as well as a small fortune) training as engineers, for instance, may be more interested in joining the private sector, where income and occupational longevity are all but guaranteed, than a terrorist organization, where the pay and lifestyle is, perhaps, less attractive. Even in the case of suicide bombers, it is not enough to simply equip and send out a great

Targeted Killings

101

many operatives on self-destruct missions—the individuals need to have the intellect and training to know where to go, who to target, how and when to detonate their bombs, and what to do in case of mishap. Otherwise, missions will fail. This is precisely what happened to the Taliban. By relying on “mentally unsound” and physically disabled individuals, drug addicts, and children to conduct its suicide attacks, its tactical success rate suffered.81 Likewise, suicide bombers must also be trained not to renege on their decision to die and need to know, too, never to get caught. In fact, there is very real danger in sending out poorly trained or unstable suicide operatives. Dud bombers—those who fail in their attempt to detonate their explosives— get caught and can crack under interrogation, causing irreparable damage to their group, cell, or operator. Everybody remembers British shoe bomber Richard Reid—who, in 2001, nearly detonated an explosive hidden in his boot while on a flight from Paris to Miami—but most have forgotten about his compatriot Saajid Badat. Equipped with a shoe bomb of his own, Badat wavered at the last moment, balked when it came time to carry out the attack. He hid his device at his parents’ house. He was later caught and imprisoned; he then promptly decided to cooperate with British and U.S. authorities. The damage he caused al Qaeda was substantial: he gave evidence in the United States relating to a foiled al Qaeda suicide operation in New York City and testified against Sulaiman Abu Ghaith, bin Laden’s son-in-law, during Abu Ghaith’s 2014 trial in the United States.82 Leaders in hiding face the related problem of motivating their followers. Calling for members to brazenly carry out acts of terror from behind a black ski mask—or worse, from the comfort of a posh villa in a neighboring country far from the frontlines—while operatives die on the battlefield, is less effective than doing so openly, in sight, and alongside members of the organization. Leading from behind is rarely a winning strategy; if militants lack motivation, they may be less likely to show up for work. The cumulative effect of eliminating the top echelons of a terrorist group—the engineers, trainers, recruiters, and so on—might deprofessionalize an orga nization in the long run. By forcing the hasty replacement of important leaders with less well trained or ill-suited individuals, targeted killings make terrorist organizations less effective. The number of skilled operators may be limited, after all, so that while a group might remain able to attract young recruits willing to act as foot soldiers, bodyguards, and suicide operatives, until these young recruits are trained as new leaders, a group’s overall ability to inflict harm may be

102

Chapter 4

diminished. If the number of key personnel involved in the production line of terrorism centers on a small cadre of professionals, then thinning the ranks of those professionals can have a negative effect on a group’s capability.83 A 2007 report published by the United Nations just after a concerted U.S./NATO targeted killing campaign in Afghanistan began, reads: “Following counter-insurgency operations in the south and east [of Afghan istan], the Taliban have lost a significant number of senior and mid-level commanders. In Hilmand, Kunar, Paktya and Uruzgan Provinces, insurgent leaders have been forced to put foreigners in command positions, further undermining the limited local bases of support.”84 It would seem, based on this UN report, that in some conflicts it is not enough to have well-trained and charismatic leaders; those leaders must also be of the right sociopolitical, ethnic, and cultural fit. The long-term survival of a terrorist or militant movement hinges on broader support afforded to it by the community. If foreigners—even Muslim foreigners in this par ticu lar case—are considered ill-suited by the community at large to lead their members into battle, then the pool of potential commanders and facilitators shrinks even further. At least for the Taliban, indigenous characteristics matter a lot. Evidence of a terrorist group’s deprofessionalization is highlighted in the case of Hamas during Israel’s campaign of elimination between 2000 and 2005. Although the terror organization carried out many more attacks on Israelis in 2005 than it did in 2002, its general effectiveness, measured by the number of victim deaths, decreased precipitously. That is, even though more attacks were planned, only a miniscule proportion of them were ultimately successful. And, of the attacks that were carried out, their outcome was limited. Other variables (like the West Bank security barrier, Israeli mobilization, military checkpoints, domestic vigilance, and Palestinian infighting) affected Hamas’s efficiency to launch attacks on Israel; it appears, as Byman suggests, that the “drop-off occurred partly because [of] Israel’s targeted killings,” which “shattered” Hamas’s ability to train, equip, and send out the best suicide bombers.85 Michael Eisenstadt, too, takes note of Hamas’s increasing carelessness. He remarks that after several months of Israeli targeted killings, a greater number of terrorist attacks had been foiled, while even successful attacks resulted in fewer than expected casualties. Although these failures could be attributed to “luck, good intelligence, and an alert citizenry,” there were signs, too, of “poor planning, inadequate prepa-

Targeted Killings

103

ration, and sloppy execution.” For example, in one attempted bus bombing, wires protruding from the bag the suicide operative was carry ing, alerted the bus driver, who with the help of other passengers subdued the bomber. In another attempt against the Tel Aviv Central Bus station, the female suicide bomber panicked and fled on foot when approached by security agents. And in a successful restaurant bombing, the operative lifted his shirt, showing off his explosive vest to diners before setting it off, giving people enough time to shout out warnings and take cover. Eisenstadt concludes that the campaign of targeted killing had “depleted the ranks of the most experienced Palestinian planners,” forcing terrorist organizations active in Gaza and the West Bank to expedite the process of preparing suicide operatives in order to prove that they were still in the fight.86 The bombers went out, but they were less than ready.

Infighting and Purges

Fourth, the elimination of leaders can lead to power struggles and internecine conflicts within militant groups and movements. This may be especially so if and when surviving members suspect a traitor or spy was involved in directing or assisting the targeted attack. Infighting should not be taken lightly. In February and March 2011, four senior Taliban commanders, three of whom led several hundred fighters in Helmand province, Afghanistan, were mysteriously killed or wounded. The events were reported to have spread a “climate of paranoia” and “spy mania” among insurgents. Surviving leaders were forced deeper underground; others refused to answer their phones out of fear of tipping off unseen surveillance drones. Divisions then emerged between the Taliban’s top leadership in Pakistan and midlevel commanders fighting in Afghanistan, with some fighters reportedly refusing to obey orders.87 This case is not unique to the Taliban. Historically, a number of terrorist groups have succumbed to paranoia, triggering purges among and between members. Out of fear and suspicion, for instance, Sabri alBanna (aka Abu Nidal) ordered and orchestrated a string of bloody internal purges in the late 1980s within his Abu Nidal Organization (ANO). Nidal’s downfall was driven in part by a diligent CIA plan to covertly undermine his perceptions of control and security. The CIA “fueled [Nidal’s] hysteria over plots against him,” a U.S. spy explained, and he panicked as a result.88 In

104

Chapter 4

Lebanon and Libya, ANO went on to execute its own; hundreds were killed. The once formidable organization was “wrecked internally by this selfdestruction.”89 In short order, ANO became near inoperable, and today it no longer exists. Likewise, the Tamil Tigers in Sri Lanka eliminated dozens of its personnel following the defection of Colonel Karuna Amman, along with a large number of his followers, in 2004. By 2009, after more than three decades of warfare, the Sri Lankan government destroyed what remained of the orga nization, killing its leader, Velupillai Prabhakaran. The defections and purges could not have helped the Tamil Tigers’ cause. And the Japanese Red Army, in 1971, went through a bout of brutal infighting that began after a prolonged and heated debate over ideological and tactical practices. Members were beaten and killed, and by 1972 the group had all but ceased to exist. Other recent examples of feuds being instigated by targeted killings or captures comes from the Afghan-Pakistan theater. For example, in 2007 a gulf emerged between various Islamist factions active in Pakistan’s tribal zone of South Waziristan: Foreign fighters associated with al Qaeda ally the Islamic Movement of Uzbekistan fought local Pakistani Taliban. Reports suggest that as many as several hundred local and foreign militants were killed and injured. Syed Saleem Shahzad (the renowned Pakistani journalist and bureau chief for Asia Times Online who was assassinated by Pakistan’s Inter-Services Intelligence in 2011) recounts that ideological differences between local and foreign fighters was primarily to blame. “Many of the foreign volunteers,” he writes, were “Takfirists, who regard ‘bad Muslims’ as the real enemy.” The result was that trained fighters of Arab, Chechen, and Uzbek origin persuaded Pakistani militants to carry out attacks against apostate Pakistan rather than coalition forces in Afghanistan. Yet, indigenous Pakistani Islamists did not always accept takfiri practices. Shahzad suggests that many local organizations were uncomfortable with al Qaeda’s relentless targeting of other Muslims. Eventually, a number of Afghan and Pakistani Taliban revolted; their primary objective was to target NATO personnel and Westerners.90 Although on the surface the spat seems to have had little to do with a specific targeted killing, closer examination of the case does reveal that it had everything to do with the absence of strong leadership. Mullah Omar, the Taliban’s fugitive leader, reportedly sent Mullah Dadullah from Afghanistan to Pakistan to help mend the divisions that emerged between the local and foreign fighters. As Omar’s esteemed envoy, Dadullah was able to do so for a short period of time in 2006. Yet he was

Targeted Killings

105

killed in a targeted attack in May 2007, cutting short his mediating efforts. Dadullah’s removal might not have been all that damaging had Mullah Omar or another leading Taliban figure been able to carry on with his intervention, but no one was able or willing to do so. The feud, left unchecked, eventually erupted. It could not have helped that Mullah Omar chose to remain in hiding rather than risk getting caught or killed himself. Thus, although Dadullah’s killing and Omar’s trepidation were not the cause for the Islamist bloodbath, their absence was certainly felt. Again, in 2010, the Taliban’s unity and operational capability suffered following the capture of Mullah Barader—the day-to-day leader of the Taliban’s Afghan insurgency and perhaps Mullah Omar’s closest aide. According to Anand Gopal, Baradar’s arrest led directly to a bifurcation of the Taliban’s surviving leadership. Two networks emerged. Confusion, Gopal explains, “over the chain of command, absent during Baradar’s tenure” featured prominently both among Taliban leaders and rank-and-fi le foot soldiers. These divisions created friction between competing Taliban personalities, further diminishing the unity of the insurgency.91 And, in 2012, the death of the TTP’s Mehsud in a CIA drone strike in western Pakistan apparently again “exposed centuries-old rivalries” among surviving members. Mullah Fazlullah was quickly chosen as the new leader, despite the fact that he was not a member of the Mehsud tribe (which has dominated the Pakistan Taliban since its inception in 2007) and notwithstanding the fact that he was apparently hiding in Afghanistan (in order to avoid U.S. drones buzzing the skies of Pakistan). But instead of having Fazlullah return to Pakistan to take charge and mend fences, the TTP appointed Khalid Haqqani as its functional leader, handing him the reins of day-today operations.92 Time will tell how well the TTP manages with two leaders; though both logic and history tell us that it may well face daunting challenges.

Lowering Morale

Finally, a fift h consequence of targeted killings is the impact it has on the morale of a terrorist orga nization’s members and, especially, on those individuals known to have been placed on a state’s wanted list. A string of successful targeted killings—especially when they involve inventive measures like booby-trapped cars, exploding cell phones, and drone-launched

106

Chapter 4

guided missiles—can lead to a feeling of despair among members of a terrorist group. Each successful targeted killing is a reminder of the long arm of the state’s abilities and reiterates that death and capture are often sudden and unanticipated. As noted earlier, after Israel proved its willingness and capacity to target and kill individuals, some wanted Palestinian militants willingly turned themselves in to authorities to avoid being killed. Over time, widespread feelings of anguish can influence not only individuals active in the group but, more generally, how well the group itself attracts new members and followers. In an enlightening 2007 story for Israel’s national newspaper, Haaretz, journalist Avi Issacharoff follows a group of Al Aqsa Martyrs Brigade militants known to be on Israel’s wanted list. The men each have “social problems” because of their status as wanted terrorists, Issacharoff explains. They are unwelcome even in their own communities: When the men visit coffee shops, patrons are quick to leave, lest they get inadvertently caught up in an Israeli raid; taxis refuse to give them lifts; and barbers and shop owners shut their doors when the men approach. The most difficult problem the men face, relates Issacharoff, is that they had “become ineligible for marriage.” Fathers refuse to allow their daughters to marry the blacklisted men to protect their daughters from the dangers that association with wanted militants might entail. In the West Bank, Israel capitalized on the despondency by offering amnesty to wanted men if they handed over their weapons to the Palestinian Authority, swore off violence, and agreed to some form of surveillance on a probationary period. The deal was that so long as they refrained from terrorism, Israel would refrain from targeting them—a classic deterrent bargain. Hundreds of men reportedly joined the program.93 The message repeated among once-active terrorists in the West Bank was that terrorism did not pay. The constant threat of elimination can heighten suspicion not only within organizations but between a group and its broader social community as well. Driving a wedge between the two is critical because distance between society and militants likely stymies the flow of recruits and structural support. Every successful targeted killing is the result of careful intelligence gathering conducted by state security forces. In some cases, the most valuable intelligence stems from well-placed informants and spies working in close proximity to the target in question. These sources transmit updated and relevant human intelligence (HUMINT) to government officials on

Targeted Killings

107

the specific location of wanted militants. As a result, after successful targeted killings, a period of heightened suspicion within the group and between the group and the community often sets in. Members of the community and new recruits might be questioned as to their loyalties. In some cases, suspected collaborators—many of whom likely are innocent civilians— are killed, their bodies left hanging in trees or alongside roads for others to see. Other times, messages accompany the corpses: “Anyone who dares spy for the Americans will meet the same fate,” read one note left beside the bullet-riddled bodies of three Pakistanis accused of aiding the U.S. war effort.94 In another particularly gruesome case, Ghulam Nabi was beheaded by an adolescent boy on suspicion of having betrayed the whereabouts of Taliban leader Mullah Osmani. Osmani was a top-tier leader, part of a Taliban triumvirate that included Mullah Dadullah and Mullah Omar. Osmani was killed in December 2006 when American forces destroyed his vehicle in a precision missile strike. Apparently, a Royal Air Force surveillance aircraft tracked Osmani after he used a satellite phone, facilitating his targeted elimination. But the Taliban thought otherwise and blamed Nabi. Like many other jihadi snuff videos, his beheading was recorded with praise for the Taliban, overlain with a jihadi song. That Nabi’s killer was a child as young as twelve (and that other boys appear pinning Nabi to the ground) only further entrenches the macabre scene. Canadian journalist Graeme Smith, in his award-winning 2013 book The Dogs Are Eating Them Now, recalls that the original video was hand delivered to Kandahar governor Asadullah Khalid. Apparently, Asadullah had personally known Nabi, who was in fact a spy. The video was widely distributed in Pakistan and Afghanistan as a warning to other would-be spies and traitors. It now lives permanently online.95 With every witch hunt and lynching of a suspected collaborator, suspicion around and within the group is heightened. Trust may go wanting. As Gal Luft writes, “The paranoid environment in which terrorists operate reduces their effectiveness drastically. Without [trust], the organization becomes disjointed; information cannot be disseminated; people do not feel a part of a team; lessons are not learned properly.”96 With greater suspicion, information concerning leaders is more carefully protected, operational planning is kept to a minimum of insiders, and recruitment is done more carefully, if only to weed out potential double agents. In each case, the terrorist

108

Chapter 4

organization’s overall ability to plan and organize acts of terrorism becomes more difficult.

Why Targeted Killings Do Not Work There are, predictably, a number of good reasons for why targeted killings might induce, rather than reduce, the scope and prevalence of terrorism. Four concerns are explored below.

Civilian Casualties and Missed Targets

First, the manner in which a target is pursued and killed can have unanticipated consequences. Aerial strikes, whether by drone, jet, or helicopter, can kill innocent bystanders and destroy property, such that even so-called “precision attacks” in densely packed urban areas can backfi re, often spectacularly. When individuals not involved in political violence pay the price of counterterrorism, ordinary men and women may be radicalized, at risk, even, of siding with and joining militant movements. Faisal Shahzad, the Pakistani American who nearly set off a car bomb in Times Square, New York City, in May 2010 justified his radicalization and recruitment to terrorism accordingly: “When the drones hit, they don’t see children.”97 He was out for vengeance. Causing widespread destruction and death in an attempt to kill a single militant, even when unintentional, is the very antithesis of a smart counterinsurgency strategy.98 Thus, while Israel got its man when it dropped a oneton bomb on Shihada’s residence in 2002, the attack unintentionally destroyed a neighboring home, killing many civilians (including nine children) and injuring scores more. The bombing was deemed a tactical success because Shihada was removed, but it was also a strategic failure because militants were able to portray Shihada and the others killed as heroes of the resistance, using their deaths to attract attention and support for their cause. The Israeli strike was also widely condemned internationally, forcing Israel to once again defend its choice of tactics and targets. The United States has similarly been at fault. Its 2006 strike against al-Zawahiri in Pakistan, for instance, killed eighteen Pakistanis; it appears al-Zawahiri was not even present at the site of the attack.99 Drone strikes in Somalia, Iraq, Yemen, and elsewhere

Targeted Killings

109

have had similar ends with some nongovernmental organization (NGO) reports suggesting that a disproportionate number of civilians have died in these strikes, raising anti-American hostility among residents and fraying diplomatic relations.100 (However, a number of studies suggest that civilian casualties have largely been exaggerated.)101 Thus, although states may have the legal and military justification to carry out targeted strikes, the cost, in potentially spurring radicalization and promoting terrorist recruitment, can be high. The scholars and surveys that have begun to address the impact of targeted killings (and drone strikes) on civilians and society mostly indicate hostile public reactions. Using anecdotal evidence, terrorism expert Mia Bloom asserts that Israel’s use of targeted killings “infuriate[s] the wider Palestinian audience,” increases militants’ ability to mobilize popular support, and even “encourage[s] terrorism.”102 Some authors studying U.S.-led strikes in Pakistan seem to agree, though much of this research focuses on public attitudes concerning drone strikes writ large, rather than on targeted killings more specifically. For instance, in their survey analysis, Clay Ramsay and colleagues found that more than 80 percent of Pakistanis interviewed believe that U.S. drone strikes were unjustified.103 And Aliya Robin Deri finds further that UAV technology itself (which can be used to dispatch individuals in a cold, mechanical, and surgical way) may interact with “Pakistan’s cultural codes of honor” in ways that encourage “tribal Pakistanis to defend their violated honor.”104 Neither do drone strikes appear to diminish militants’ abilities to propagate their message among would-be and actual supporters.105 Yet, the role targeted killings, drone strikes, and other counterterrorism measures have in shaping the relationship between members of society and militant organizations is not well understood. So although some reports argue that drone strikes in par ticu lar galvanize support for militants, other studies fully disagree. For instance, Brian Williams finds that civilians living in areas of Pakistan that are governed by the Taliban are surprisingly supportive of U.S. drone strikes. Why? Because popu lar revulsion with the Taliban’s austere religious governance seems to trump public anger over U.S. drone warfare. The “simplistic paradigm that Predator drones drive Pakistanis into the arms of the militants,” Williams concludes, “needs to be reevaluated.”106 Once again, however, other Pakistani public surveys conducted in Taliban strongholds challenge Williams’s claims: In 2010, the New America Foundation found that civilians living in Pakistan’s Federally Administered Tribal Areas (FATA) do indeed strongly

110

Chapter 4

oppose the presence of al Qaeda and Afghan and Pakistani Taliban in the area, but they also vehemently oppose U.S. drone strikes and other military initiatives.107 Arguably, additional research is needed to parse these topics more effectively. Otherwise, unsuccessful targeted killings—those that fail to kill their targets or that miss their objectives altogether—can make the state organizing the attack look weak, disorganized, or, worse, inept. Routinely failing to kill or capture intended targets can also embolden the individual being sought. For instance, before his death in 2012, al Qaeda’s al-Quso evaded death on at least three separate occasions and managed to escape prison, too. “Every time we do a hit on him and he survives,” FBI agent Ali Soufan noted, “his reputation becomes more significant.”108 Al-Quso’s legend grew over time, which al Qaeda likely used to attract recruits. Likewise, killing the wrong person can have damaging results as well. In 1973, Israeli Mossad agents killed Ahmed Bushiki, an Algerian working in Norway, believing that he was Ali Hassan Salameh, a senior member of the group responsible for the massacre at the Munich Olympics. (Israel did eventually find and kill Salameh in Beirut several years later.) Norwegian-Israeli diplomatic relations soured badly as a result of the failed operation. Getting caught conducting a targeted killing can also lead to diplomatic demarches and strained relations. Israeli-Jordanian and Israeli-Canadian ties were severely tested when Mossad agents, who had entered Jordan on falsified Canadian passports, were caught trying to poison Hamas’s Mashal in 1997. In sum, the potential gain offered by eliminating terrorist facilitators might well be offset if and when plans go awry.

International Condemnation

Beside collateral damage and failed attacks, there is the very real cost of potentially losing international and regional legitimacy by pursuing a policy and strategy of targeting individuals. Although the United States, Canada, Britain, and a great number of their allies accept (if cautiously and behind closed doors) the need to eliminate terrorist facilitators, many more states equate targeted killing with assassination and would prefer to see a redoubling in efforts to capture rather than eliminate suspected militants. One potential international outcome of the continued moral debate is a fracturing of political consensus among allies. The risk is that targeted killings

Targeted Killings

111

(when supported by one state but rejected by another) might splinter political and military alliances, damaging fragile international partnerships. The U.S.-Pakistan relationship is today especially sensitive to American efforts to target and kill militants (usually in drone strikes) in western Pakistan.109 But as the United States expands its counterterrorism footprint in Yemen, Syria, Iraq, North and West Africa, Somalia, and elsewhere, by potentially further intensifying its use of drones, a number of other U.S. partnerships may become increasingly untenable. Furthermore, as the UN continues to develop its own role in reviewing and investigating U.S. drone strikes and targeted killings, traditional U.S. allies may themselves be placed in a sticky situation. For example, in June 2012, Christof Heyns, the UN special rapporteur on extrajudicial killings, offered his most stinging critique of U.S. drone strikes, suggesting that some may even constitute “war crimes.”110 That U.S. drones have also killed German and British citizens suspected of joining jihadi groups overseas might help sharpen domestic barbs, as well. In the United Kingdom, questions were raised as to the role British intelligence played in directing and feeding U.S. drone strikes abroad.111 If civilians are killed in these strikes, are Britons culpable? What about Google? If its ser vices provide the NSA with intelligence on militants that lead to the death of civilians overseas, what are shareholders going to do? It is possible that international unease over targeted killings might bubble over into open rebukes of U.S. policy, both by Western allies and corporate entities alike, just as Bush’s policies concerning Guantanamo Bay, rendition, and enhanced interrogations were widely and loudly censured. Given that combating global terrorism effectively requires as wide a coalition as possible—and might include the need to partner with less traditional allies—divisions regarding the legitimacy of targeted killings can have a lasting and detrimental political effect. However, if targeted killings are declared by a majority of states as a just and effective tactic, the danger is that new norms and standards of behavior concerning assassination will be introduced into global affairs. As David cautions, “Norms may not be determinative [to action], but neither are they irrelevant.”112 As the norm against assassination is weakened, there is concern that their use and popularity as a solution to foreign policy and security dilemmas will rise as a result. That is, other states and even nonstate actors may use targeted killings—and, if  possible, drone technology—against their enemies and adversaries. In October 2012, for instance, Hezbollah sent an Iranian-manufactured drone

112

Chapter 4

into Israel. The drone was unarmed—and rather crude—and Israel easily shot it down, but Nasrallah played up the event anyway. “Possession of such an aerial capacity,” he crowed, “is a first in the history of any resistance movement. The [drone mission] is part of an overall deterrence strategy.”113 Furthermore, as champions of the tactic, Western leaders might well invite reciprocity. Philip Alston, a UN special rapporteur on extrajudicial executions, warned in 2010 of a future dystopia in which dozens of countries use drone attacks beyond their borders against those they label terrorists: “I’m particularly concerned” he added, that the United States seems “oblivious to this . . . when it asserts an ever-expanding entitlement for itself to target individuals across the globe.”114

Militant Retaliation

Another downside to eliminating specific terrorist leaders is the likelihood that their organization will try to retaliate. This makes sense. An aggrieved organization will want to prove its capability despite the loss of its leader while exacting revenge for the strike itself.115 After Hezbollah’s Musawi was killed by Israel in 1992, for example, Hezbollah retaliated by attacking the Israeli Embassy and the Asociación Mutual Israelita Argentina (AMIA) Jewish community center, both in Buenos Aires, Argentina, in 1992 and 1994. Hundreds of people were killed and injured. And after Israel killed Popular Front for the Liberation of Palestine’s (PFLP) secretary general Abu Ali Mustafa in 2001, the group assassinated Israeli Minister of Tourism, Rehavam Ze’evi. Likewise, four suicide attacks followed Israel’s killing of Ayyash in 1996. Even more damning, Naomi Chazan suggests that in Israel’s case there is “a clear pattern of assassinations yielding increased rocket attacks . . . to even more audacious suicide bombs.”116 Tit for tat. Pakistan’s TTP also retaliated against the United States after it killed its leader, Baitullah Mehsud, in 2009. In an audacious plan pulled straight out of a Hollywood blockbuster, TTP used a double agent, Jordanian doctor Humam Khalil Abu Mulal al-Balawi, to gain access to a CIA base (Camp Chapman) near Khost, Afghanistan, where Balawi detonated a suicide bomb, killing seven CIA operatives and a Jordanian spy. In a posthumous video released to Al Jazeera, Balawi is shown sitting next to Baitullah’s replacement, Hakimullah Mehsud, praising the TTP. (The attack was later dramatized by Kathryn Bigelow in Zero Dark Thirty, a 2012 film that portrays the decade-long hunt for bin

Targeted Killings

113

Laden.) Indeed, noting the spike in deaths caused by suicide attacks in Afghanistan and Pakistan between 2004 and 2009, Leila Hudson and colleagues suggest that “it is probable” that there is a “substantive relationship” between the number of drone strikes and levels of militant retaliation.117 Although militant reprisal might be avoided by carry ing out targeted killings in a clandestine manner, covert attacks may be more difficult to do and will negate much of the deterrent effect the elimination might have had; remember, secret attacks only poorly communicate coercive intent. Besides, some observers question the tit-for-tat retaliatory argument—that targeted killings invite further acts of terrorism—altogether. After all, the stated intent of al Qaeda, Hezbollah, Palestinian Islamic Jihad, and so on, is to attack Americans, Israelis, and Britons. None of these groups need an excuse to do so.118 Violence, in this case, may not beget violence. The very real downside is that of unintentionally creating martyrs and popular heroes from targeted leaders that ultimately stimulates recruitment. From Afghanistan to Yemen, fallen leaders are often hailed as champions of their cause. Palestinian organizations put on “lavish funeral processions” to drive home the point.119 Streets are named after targeted leaders, candies and fruits are handed out in public squares, relatives are congratulated. Not surprisingly, being the victim of a targeted killing can actually enhance a group’s standing among the community. It is proof that the organization has done something to merit the state’s attention and resources. As one jailed militant recounts in an interview with Jerrold Post and colleagues, “Major [terrorist] actions became the subject of sermons in the mosque.” Public ostracism was later levied against those who failed to enlist in the militant cause.120

Unknown Successors

In removing a militant leader, there is the related risk that states will be left with something worse, a replacement that proves a more vicious, charismatic, or fanatical opponent. Consider that after Israel killed Hezbollah’s Musawi in 1992, he was replaced by Nasrallah, who went on to become one of the most effective militant leaders in contemporary history. He has built Hezbollah into a formidable substate organization capable of both launching sophisticated attacks against Israel and controlling—with popular sympathy—large swaths of southern Lebanon. Today, Nasrallah’s Hezbollah is a prominent player in Lebanese and Syrian politics (and war) and is a close ally of Iran.

114

Chapter 4

One way states might avoid creating martyrs is to target facilitators that are less well known to the public. Mid-level commanders will have a lower profi le than those at the top, though their role in the overall functioning of the terrorist orga nization will be nonetheless crucial. Targeting these individuals might have the intended effect of impeding and deterring terrorism while not breeding the next generation of young recruits and seasoned leaders. On the flip side, eliminating a group’s leadership might have the unintended result of removing charismatic leaders that might have otherwise played a critical role in negotiating a future settlement to a protracted conflict. Had the United States managed to kill Libya’s Qaddafi in 1986 during Operation El Dorado Canyon, he would not have been around to lead his country in from the cold in 2003, renouncing everything from international terrorism to WMD proliferation.121 The same might be said of some militant leaders: More than a few have traded violence for politics. Ezer Weizman, an Israeli cabinet member during Israel’s 1988 killing of al-Wazir in Tunisia, argued as such: Israel is “trying to fi nd Palestinians to talk to us. I don’t think the assassination contributes to this. Liquidating individuals will not advance the peace process.”122 The same might be said today of the Taliban in Afghanistan. Although skeptics, including this one, are suspicious of calls for finding a negotiated settlement between Kabul, the West, and the Taliban, there might be some merit to Weizman’s general critique. If the most popu lar leaders—those able to rally supporters around a cause—are killed today, the list of tomorrow’s political partners is cut short. Here is where targeted captures, rather than killings, may prove especially attractive. Indeed, many states are quick to point out that they would rather capture than eliminate individual terrorists. For starters, dead men don’t talk.123 The apprehension of militants and leaders provides the state with invaluable operational, tactical, and strategic intelligence that can lead to further detentions and other tactical successes. Bin Laden himself, for instance, was eventually found and killed because of the intelligence that was acquired from a senior al Qaeda leader, Abu Faraj al-Libbi, captured in Pakistan in 2005. Al-Libbi was considered al Qaeda’s No. 3 at the time of his arrest and was able to give interrogators the pseudonym of one of bin Laden’s designated personal couriers. The runner was eventually traced to the Abbottabad compound where bin Laden was hiding. In another case, intel-

Targeted Killings

115

ligence gleaned from Ahmed Sidiqi, an Afghan German militant with ties to al Qaeda who was arrested in Kabul in July 2010, led to the discovery of al  Qaeda’s 2010 plot to use western militants in “Mumbai-style” shooting attacks in various European cities. Sidiqi’s information was corroborated by other captured militants in France and Germany and the plot was eventually thwarted. Capturing a leader also avoids inadvertently creating a martyr of a dead one. Pictures and videos of incarcerated leaders retain propaganda value for counterterrorism efforts. Think of the unflattering pictures associated with KSM’s 2003 capture.124 Wearing a stretched-out and dirty white T-shirt, KSM looks disheveled, grubby, and (only slightly) drugged. The picture now accompanies almost all Internet searches of KSM and is not helping al Qaeda’s cause. Or consider the embarrassing tirade Guzman shouted from his Peruvian prison cell. Clad in a striped prison jumper, yelling at journalists from his outdoor cage, he looked deranged, nothing like a heroic Maoist crusader. Guzman’s diatribe is widely available on YouTube. Captured leaders can also be put on trial, providing their victims with some form of justice, and, at times, they can renounce their militant views and convince others to join them in laying down their arms. The obvious drawback is that targeted captures are difficult to carry out, especially within the context of a foreign war. As slain Tamil Tiger leader Velupillai Prabhakaran illustrated in his desperate bid to escape encroaching Sri Lankan forces in 2009, militant leaders may be especially unwilling to surrender. That they may fight to the death suggests that targeted captures place security personnel in positions of unwarranted danger. And there is the related risk of operational failure; a wanted individual may be more likely to elude his or her captors than to survive a pinpoint precision strike. Other scholars have suggested, too, that some militant leaders can be influenced without the use of threats of death or capture. In their counterintuitive research, Elena Mastors and Jeff rey Norwitz explain that “leadership frameworks” that take into account a leader’s “personal characteristics, operating environment, advisory process, and information environment” can help security officials build a detailed and highly personalized roadmap for better understanding what makes specific leaders tick. With that knowledge, a covert and tailored psychological operation might be devised that influences the leader. (The psychological operation against Abu Nidal, discussed earlier, is a case in point.) Using al Qaeda’s al-Zawahiri as an

116

Chapter 4

example, Mastors and Norwitz suggest that counterterrorism officials write and then leak fictitious letters on his behalf that play to his characteristics and personality flaws. We could have him make disparaging, derogatory, even racist remarks as to the quality of al Qaeda’s new leadership in Somalia, Yemen, North Africa, Syria, Iraq, and so on. Or we might surreptitiously nettle the ideological, tactical, and strategic divisions he has previously had to contend with. The fractious relationship between local Syrian jihadist groups, like Jabhat al-Nusra, and foreign jihadist organizations active in Syria, like the al Qaeda–linked Islamic State of Iraq and the Levant (ISIL, or ISIS), may be especially sensitive to Western interference. The goal would be to embarrass and discredit al-Zawahiri, to force him to defend himself publicly, and to oblige him to continually mend fences. And that might give security officials a potential opportunity to track him down while sowing division among the militant movement more broadly.125

But . . . Do Targeted Killings Work?

Finally, the long-term effect, deterrent or otherwise, of a sustained campaign of targeted killing on al Qaeda and other militant groups is uncertain. It is worth reviewing this literature in some detail.126 Although military and political leaders in the United States are quick to point out the number of dead terrorists as a metric for gauging successes in their struggle against militant groups, there remains the simple fact that targeted killings have yet to, and may never, have the desired impact of eliminating terrorism altogether. American officials have likened targeted killings to “clipping toenails” (or “mowing the lawn,” as some Israelis put it) because al Qaeda and others seem capable of eventually cultivating a new cohort of leaders to replace fallen comrades.127 Those groups seem to have a deep bench of willing trailblazers. Byman argues further that the policy is potentially much less useful against highly “decentralized groups,” where “true decapitation” is not possible.128 Although Israel managed to severely undercut PIJ activity by eliminating one man, it would be difficult to repeat that success with Hezbollah or Hamas. In a similar vein, Austin Long of Columbia University suggests that a critical factor that helps determine whether targeted killings are effective

Targeted Killings

117

in diminishing terrorism (or of destroying militant groups) is the level of “institutionalization” within the group. Organizations that have effective hierarchical decision-making structures and specialization within ranks can bestow on (new) leaders a degree of authority, inoculating the group as a whole from the effects of leadership decapitation. Turning the tables, this is exactly what happens when Western military leaders are killed or otherwise removed from their positions; organizational bureaucracy kicks in. New leaders are quickly selected and are promptly granted legitimate authority. This is precisely what happened when CIA director (and retired four-star general and Iraq War hero) David Petraeus suddenly resigned from his position in a sex scandal that rocked Washington in November 2012. The transition, first to acting director Michael Morell and later to John Brennan, was seamless. Americans would not have expected otherwise. Long hypothesizes, then, that militant groups with low levels of institutionalization will suffer more readily from targeted killings than groups that can provide smooth and structured leadership transition. He offers a preliminary test using the rise and fall of various Iraqi militant groups in  the mid-2000s. Although his thesis is not particularly novel—the idea that charismatic militant leaders are more difficult to replace has long been proposed—Long’s argument that institutionalization plays an important role does suggest that the more bureaucratic a militant group becomes, the more likely it is to survive leadership strikes.129 Michael Freeman suggests otherwise. Militant leaders perform “two large aggregated functions”: They inspire followers and militants, and they help manage and guide an orga nization. The leader’s role in providing either or both of these functions waxes and wanes over time and as the orga nization evolves. The “greatest expected utility for leadership targeting,” Freeman theorizes, will occur when leaders provide both functions. So while organizational bureaucracy or institutionalization may soften the effects targeted killings have on groups, the precise function and role of the leader being removed is potentially far more important.130 A handful of other studies (some qualitative, others quantitative) have sought to evaluate whether targeted killings correlate with a militant group’s mortality rate—organizational dissolution, collapse, and decline— and reduction in group capability. Findings are generally mixed. Audrey Kurth Cronin, in her landmark 2009 book How Terrorism Ends, fi nds that the effects of leadership removal (by death or capture) on a militant group’s

118

Chapter 4

rate of survival will vary, depending on the structure of the organization, the degree to which the group “fosters a cult of personality” in a specific leader, the availability of replacement leaders, the group’s ideological motivation, and the nature of the removal itself (death versus capture). After comparing a number of qualitative cases, she concludes that capturing militant leaders (and publicly placing them on trial and jailing them) is more effective for terminating a group than is killing leaders outright.131 A number of other scholars have focused on the Israeli case study. In their quantitative analysis of Israel’s campaign of targeted killings of Palestinian militant leaders between 2000 and 2004, Mohammed Hafez and Joseph Hatfield tackle various assumptions outlined in the theoretical literature: deterrence, backlash, disruption, and incapacitation. They conclude that Israeli targeted killings neither diminished nor increased the rates of Palestinian political violence (including suicide bombings and other forms of terrorist activity). Furthermore, a decline in terrorist success rate was likely a result of defensive rather than offensive Israeli countermeasures. They write that although targeted killings “may signal a determination to fight back,” may show an Israeli “commitment not to succumb” to militant demands, and may placate public anger and calls for government action, they “should not be presented as a proven solution to patterns of political violence.”132 Other Israel-based case studies corroborate these fi ndings. Edward Kaplan and colleagues, for instance, fi nd that in the specific case of suicide bombings against Israeli citizens, preventative arrests rather than targeted killings influenced rates of violence. Targeted killings serve only to “knock out individual” terrorists while captures and arrests facilitate interrogation, which can lead to intelligence on the terrorist network and to other more effective counterterrorism measures.133 Of the large-N quantitative studies that go beyond the limited Israeli case to offer a more global assessment of targeted killings, three studies in particular stand out. Again, however, each is primarily focused on assessing whether targeted killings influence the mortality rate of militant groups; counter-capability rather than counter-motivation or coercion is the primary focus. Overall findings are mixed. Jenna Jordan offers one of the first systematic and quantitative studies of the effects that leadership targeting has on militant groups. Using a data set of 298 cases of elimination—she excludes cases in which leaders are removed by internal purges and infighting— encompassing ninety-six militant organizations, she finds very little evidence

Targeted Killings

119

supporting theoretical claims that targeted killings hasten a group’s collapse. In fact, targeted killings may be counterproductive, extending rather than shortening the longevity of an organization. Otherwise, her findings also illustrate that a militant group’s “age, size, and type”—that is, religious, ideological, or separatist motivation—can influence how effective targeted killings are on rates of decline and degradation: Older and larger organizations are more resistant to targeted killings and collapse than younger and smaller groups. And, based on three additional qualitative studies of Spain’s Euskadi Ta Askatasuna (ETA), Colombia’s FARC, and Hamas, she finds only mixed results for the effects leadership removals have on a group’s capacity to conduct lethal attacks. Jordan concludes that “leadership decapitation seems to be a misguided strategy.”134 The other quantitative studies suggest otherwise, however. Patrick Johnston’s research uses 118 decapitation attempts (forty-six of which successfully removed a leader) from ninety counterinsurgency settings, and fi nds using statistical analysis that eliminations do have positive net effects: They increase the odds of conflict “termination” and “government victory” and reduce the intensity and frequency of militant attacks. Digging deeper, Johnston also fi nds that targeted killings are more effective than captures: “There is a statistically significant relationship between operations that resulted in the deaths of militant leaders and quicker war termination and government victory.”135 The second study, by Bryan Price, corroborates some of Johnston’s conclusions. Price begins by illustrating that terrorist groups should be especially susceptible to decapitation because they have certain innate characteristics (“they are violent, clandestine, and valuesbased”) that make leadership succession particularly difficult. Theoretically, leadership removal should affect a group’s longevity and durability. Price, using a data set that includes 207 militant groups active between 1970 and 2008 and 299 cases (one more than Jordan) in which militant leaders are killed, captured, or otherwise replaced—because of natural death, expulsion, or resignation—finds that “leadership decapitation significantly increases the mortality rate of terrorist groups.” This was true regardless of how leaders were removed, suggesting that targeted killings and captures (and natural deaths, too) provide similar results, and regardless of the group’s size, suggesting that large, robust groups are just as likely to decline as small ones after a leader is removed.136 In sum, targeted killing should be considered one among many different counterterrorism and

120

Chapter 4

counterinsurgency tactics, to be applied in concert with other mea sures that seek to tackle the long-term determinants of terrorism, militancy, and political violence.

Targeted Killings as Coercion From the earlier discussion, the presumed deterrent effect of targeted killings on the short- and long-term behavior of terrorist organizations can be surmised. Turning targeted killings into a deterrent requires that a state’s threat to use them against a militant organization is both communicated clearly and perceived as credible. These are the prerequisites of deterrence theory. Credibility is perhaps most easily achieved by demonstrating capability: locating, pursuing, and attacking wanted individuals. When will a militant group appreciate that a threat to kill its leaders is credible? When its leaders are targeted and killed.137 The use of targeted killings bolsters the credibility of a state’s threat to use them again in the future. And in terms of communicating the threat to militant groups and leaders, states can easily publish wanted lists and disseminate them within the community in which a target is suspected of hiding. Israel, as the earlier examples illustrate, did so often by passing lists of wanted militants on to the Palestinian political leadership. When Israel proved able to target individuals with precision force (capability), illustrated its resolve to do so repeatedly (credibility), and ensured that individuals appreciated that they were wanted (communication), it managed to change the behavior of some militant groups and their individual members and leaders. In 2012, Israel updated the tactic by taking its coercive message to the Internet, relying on recent advances in online social media to more effectively get its message across. At the very onset of Operation Pillar of Defense—the November 2012 conflict with Gaza—Israel killed Hamas chief Jabari in a precision attack. The Israel Defense Forces via their Twitter account (@IDFSpokesperson), immediately offered followers a link to a short, black-and-white YouTube video of the Jabari strike, presumably taken from an Israeli surveillance drone. “In case you missed it,” the tweet read, “IDF Pinpoint Strike on Ahmed Jabari.”138 (The ten-second video got several million views in its first week.) That IDF message was then followed with an explicit warning to other militant leaders (see Figure 4.1). It is interesting to note that the tweet itself used language that IDF brigadier general Yoav (Poly) Mordechai used earlier in a televised press

Targeted Killings

121

Figure 4.1. Screenshot of IDF tweet, November 14, 2012.

conference, in which he stated: “I encourage all members of Hamas not to walk around above ground.” But his quote barely registered with the international media.139 It went viral only after the IDF Twitter handle repeated it. In this particular case, the Jabari strike communicated and illustrated the IDF’s capability, while the tweet—picked up and widely reposted online— carried the promise to other Hamas leaders and Jabari’s replacements in particular that more of the same remained a possibility. This was the intent of the IDF’s tweets of war.140 In an interview with BuzzFeed—a popular, New York–based online platform that collates viral web content—Lieutenant Colonel Avital Leibovich, an IDF spokesperson who heads the IDF’s Interactive Media branch and manages its social  media outreach operation, suggests that “when . . . you want to convey a message of deterrence to an audience, then [Twitter is] a good tool to do it.”141 A well-articulated and clearly communicated campaign of targeted killings that is carried out systematically and punctuated by some successes can have a number of coercive, deterrent, or compellent effects: 1. Leaders and high-ranking terrorist facilitators are kept hidden from the public: Their names, faces, and identities are kept secret and their statements are issued by way of audio or written mediums rather than at public gatherings. 2. Lower-ranking individuals become less willing to replace fallen leaders or are altogether harder to find: Power vacuums emerge as a result, and splinter groups surface. 3. Organizations become plagued by power struggles: Purges are carried out, and unions and alliances are broken. 4. Leaders change the manner in which they communicate with one another: To avoid inadvertently revealing their location, they begin relying on less sophisticated means of communicating with other militants. 5. Organizations begin attracting poorer-quality recruits: Children and the mentally ill are increasingly sent on suicide missions, for instance.

122

Chapter 4

6. Group morale suffers: There is an increasing rate of defection and abandonment. 7. Suspicion within the organization and between it and its supportive popular community is heightened: Purported traitors and spies are killed, angering members of the community. 8. The organization’s overall professionalism decreases: Attacks become sloppy, and failure rates and arrests increase. 9. Terrorist attacks become less sophisticated, in keeping with a less professional leadership cadre: Shifts occur in the selection of victims (toward easier or softer targets) and in the type of attacks carried out (toward less sophisticated assaults). Many of these coercive effects are evident and measurable. A careful historical analysis of a terrorist organization’s behavior following a targeted killing can help identify when and where these developments took place. Doing so provides an analysis of deterrence (by punishment) theory as it applies to terrorism and political violence. Of these deterrent influences, the last two—decreases in professionalism and changes in the type, nature, and sophistication of terrorist attacks—are the easiest to elaborate on, identify in practice, and measure systematically in a quantifiable way. The other coercive effects can be assessed from a careful reading of the historical (that is, behavioral) record. Chapter 5 evaluates the coercive effect targeted killings had on Taliban behavior following the removal of four militant leaders.

CHAPTER 5

Targeting the Taliban: Coercive Lessons from Afghanistan

On March 5, 2007, Daniele Mastrogiacomo, an Italian journalist, along with his Afghan driver and an Afghan journalist, Ajmal Naqshbandi, were abducted by Taliban forces while reporting from southern Afghanistan. The kidnapping was conducted by forces associated with Mullah Dadullah Akhund, the most high-ranking and prominent Taliban leader in Afghanistan. Besides being a top-ranked member of the Taliban’s shura (or leadership) council, Dadullah was also one of three military strategists dispatched by the Taliban’s supreme leader, Mullah Mohammed Omar, to organize the Taliban’s insurgent activities. Mastrogiacomo’s kidnapping made headlines. After promptly executing the Afghan driver, the Taliban broadcast a plea by Mastrogiacomo for the removal of Italian troops from the country. In an unprecedented move, the Afghan government cut a bargain with Dadullah after only two weeks, agreeing to a controversial prisoner swap. Mastrogiacomo was released on March 19 in exchange for five captured Taliban leaders, including Mullah Shah Mansoor Ahmad (aka Mansour Dadullah), Dadullah’s younger brother.1 The prisoner swap was met with ferocious disapproval, both in Afghanistan and overseas. Denunciations even flowed in from the United Nations, where an Afghan spokesman reiterated that “the UN does not negotiate with terrorists.” The assertion was repeated by the U.S. Embassy in Kabul, stating that “it is U.S. policy not to make concessions to terrorists’ demands.” Some Italian newspapers even called the swap “straight-out repugnant.” In Kabul, a beleaguered spokesperson for Afghan president Hamid Karzai

124

Chapter 5

called the prisoner exchange “an exceptional case” reflective of Afghanistan’s strong relations with Italy and assured the international community that “it would not happen again.”2 Indeed, the release of such a high-ranking Taliban leader as Dadullah’s brother along with four other commanders was a contradiction of NATO, the International Security Assistance Force, and Afghan policy. How could any of them—let alone Washington—have ever agreed to the swap? The answer came a few weeks later. Most observers were unaware that Mullah Mansoor was covertly monitored the minute he stepped out of prison and led coalition forces directly to his brother’s hideout. Using sophisticated signals and drone technology, American forces working with Task Force Orange (TF-Orange), a clandestine unit dispatched by the Pentagon to collect signals intelligence overseas after 9/11, tracked Mansoor to a Taliban training base in Quetta, Pakistan. Once there, Mansoor was happily reunited with his brother, Dadullah, and ISAF had finally tracked down the elusive Taliban commander. Dadullah’s satellite phone was placed under surveillance, his every move diligently tracked. In time, Dadullah and Mansoor led a group of fighters over the Pakistani border into Afghanistan’s southern province of Helmand. Once in Afghanistan, a precision attack was planned for Dadullah’s removal. Using the SIGINT, his exact location was pinpointed to a compound in Bahram Chah, Helmand province, a mere fifteen miles from the Pakistani border. A precision air strike, a commonly used tactic for carry ing out targeted killings in Afghanistan at that time, was ruled out because it could not guarantee that Dadullah would be killed. Instead, special operations forces with the British Special Boat Ser vice launched a coordinated strike on the compound using both a helicopter-borne and a ground-based assault. On May 12, 2007, SBS soldiers, along with a number of Afghan Special Forces, stormed the compound and, in the ensuing battle, killed Dadullah along with two of the five men released in the Mastrogiacomo exchange (Mullah Hamdullad and Mullah Ghaffar).3 Over the following days, Dadullah’s body was recovered and its image broadcast to the world. At the time, ISAF commented that although Dadullah would “most certainly be replaced” the insurgence had nonetheless “received a serious blow.” The governor of Kandahar Province, Asadullah Khalid, concurred, arguing that Dadullah’s elimination was “a huge loss for the Taliban; it will certainly weaken their activities.” Rahimullah Yusufzai, an editor for the Pakistani

Targeting the Taliban

125

newspaper The News and a Taliban expert argued further: “I think [Dadullah’s killing] is the biggest loss for the Taliban in the last six years. I don’t think they can find someone as daring and important as Dadullah.” 4 Complex operations like this one have become increasingly prevalent in the Afghan conflict. ISAF as well as American and Afghan government and military officials have pinned their hopes on the tactical and deterrent effects of targeted killings. Today, even more than during Dadullah’s reign, precision strikes, including drone strikes, are common in Afghanistan and neighboring Pakistan. Although the exact number of militant leaders and facilitators that have been tracked and killed has never been disclosed, by most accounts, between 2006 and 2008—around the time of Dadullah’s elimination—50 to 200 targeted killings had been conducted against Taliban, al Qaeda, and other insurgent leaders in Afghanistan by international, American, and Afghan forces.5 Since then, American drone strikes in Pakistan have expanded exponentially, from 53 separate sorties in 2008 to nearly 120 in 2010. According to statistics compiled by the New America Foundation, of the 277 drone strikes carried out in Pakistan between 2004 and 2011, between 1,387 and 2,163 militants had died. That is an average of roughly 6 militants killed in every strike.6 Although drone strikes are accurately described as precision attacks, it would be a mistake to also consider each strike a targeted killing. As I argued earlier, by definition, targeted killings are carried out with the intention of eliminating a single (and known) individual, not a group of purported militants. Many of the 277 drone strikes have, for instance, targeted groups of insurgents, training facilities, safe houses, and militant infrastructure. The rationale, often repeated, of carry ing out targeted killings is rather simple: “You disrupt the network when you take out the leadership.” 7 Over time, gaps appear within the command-and-control structure, weakening a group’s ability to function effectively in the long term. This chapter assesses the deterrent and coercive effects of targeted killings on the behavior of the Taliban. It begins with a description of Afghanistan’s most-wanted terrorists and ISAF’s “Most Wanted campaign”—officially announced in October 2007—and offers a summary of the more prominent targeted killings that have taken place in Afghanistan and Pakistan since 2006. From there, a comparative analysis of three targeted killings—Mullah Dadullah, Qari Faiz Mohammad, and Mullah Abdul Matin—is made. A second, cross-case analysis is then offered, using aggregate data that includes a fourth elimination,

126

Chapter 5

Mullah Mahmud Baluch. Both sets of comparisons help identify the coercive effect these targeted eliminations had on Taliban behavior.

Afghanistan’s Most-Wanted Terrorists Deconstructing the violence in Afghanistan is not easy. A number of militant and terrorist organizations are active in the country and bordering regions.8 In addition to the Taliban (and “neo-Taliban,” as some observers put it) and al Qaeda, a number of less renowned groups also plan attacks, bombings, and ambushes. The Haqqani network, for instance, established by Jalaluddin Haqqani, trains foot soldiers in the lawless regions of North Waziristan, Pakistan, and sends them over the border into eastern Afghanistan to target multinational and Afghan soldiers. The Islamic Movement of Uzbekistan, led by Tahir Yuldash until he was killed by a Predator drone strike in August 2009, is also active in Afghanistan. Headquartered in Waziristan, the group is closely allied to al Qaeda and regularly equips Taliban forces, in both Afghanistan and Pakistan, with men and equipment. The Anwar ul-Haq Mujahed, otherwise known as the Tora Bora Military Front, established by Moulvi Younus Khalis (a mujahedeen leader from the Soviet era) and led by his son Anwar Ul-Haq, organizes operations in Afghanistan’s Nangarhar province in and around the fabled mountains of Tora Bora. Khalis announced his decision to wage jihad against NATO in 2005, and established an alliance with the Taliban in order to spread the insurgency to the provincial capital of Jalalabad. The Hezb-i-Islami Afghanistan (Islamic Party of Afghanistan, HIA), though sidelined during much of the Taliban’s reign, has reemerged as a capable militant organization. Led by Gulbuddin Hekmatyar, the HIA has carried out a number of large-scale attacks, including an April 2008 attack on a Kabul military parade attended by Karzai that killed a number of Afghans (including a parliamentarian) and sent shock waves around the world. Finally, the Tehrik-e-Taliban Pakistan (TTP), a consolidation of various Taliban-affi liated groups active in western Pakistan’s Tribal Region, formed in 2007 under the leadership of Baitullah Mehsud. Following Mehsud’s death, TTP leadership was handed over to his deputy, Hakimullah Mehsud, who the United States killed in 2013. The TTP poses both a serious regional and a serious international challenge. It not only carries out devastating attacks in Afghanistan (like the 2009 suicide attack against

Targeting the Taliban

127

the CIA facility at Camp Chapman) and Pakistan (including the May 2011 attack on a military academy in northwest Pakistan that killed more than eighty people), but it also claimed responsibility for the failed 2010 car bombing at Times Square in New York City. Though each group may share common aspirations (the removal of foreign troops and aid workers from Afghanistan, the establishment of an Islamic government in Kabul—and, for some, in Islamabad as well—the imposition of strict sharia, and so on) and though each group may work with one another toward these goals, they are nonetheless distinct organizations. Of all the groups, the Taliban is the largest, most organized, and by far the most active in Afghanistan. It draws recruits from various indigenous populations in Afghanistan and Pakistan along with foreign fighters from Uzbekistan, Chechnya, Saudi Arabia, Libya, Egypt, and elsewhere.9 The Taliban also claims to have broad popular support in a number of Afghanistan’s provinces and, ostensibly, acts on behalf of Afghans. Unlike al Qaeda, for instance, which is not only led primarily by non-Afghans but fights in Afghanistan in order to reestablish the territorial base it lost in 2001, the Taliban’s war is first and foremost an attempt to “liberate” Afghans from “Western influences” and to reestablish Islamic rule in the region. That some (perhaps even most) Afghans shudder at the prospects of a second Taliban regime is of no consequence; the Taliban sees itself as the religious vanguard with a duty to protect Afghans and the integrity of Islam in the region. The Taliban is organized hierarchically, with a supreme council dictating the group’s long-term goals, strategy, and general behavior. In this regard, the Taliban’s control and command structures are more akin to Hezbollah’s or Hamas’s—both of which have centralized hierarchical leadership—rather than al Qaeda’s, which has a more loosely defined network structure. The Taliban’s centralization carries down through all levels of the organization. Thus, the shura appoints various regional commanders who themselves sit on top of another organizational pyramid. Below them are various districtlevel commanders looking after even smaller groups active in specific geographic locations.10 In this way, operational reports and requests from the battlefield work their way up through the organization until, if they merit more serious attention, they reach the shura council itself. Nothing, it would seem, is left to chance. Consider then the Taliban’s layeha (book of rules), a small nine-page pamphlet reportedly distributed in 2006. (A longer Taliban “pocket book,” which is discussed later, appeared in 2009.)

128

Chapter 5

A copy of the pamphlet was acquired and published by the Swiss weekly journal Die Weltwoche and was republished in German and later in English by Newsweek. The booklet offers the individual Taliban fighter thirty simple rules to follow, offering input on everything from attracting the right kind of individual to the Taliban’s cause, to the virtues of refraining from smoking cigarettes. For our immediate discussion on hierarchy, consider rules six, ten, and twelve: If a Taliban fighter wants to move to another district, he is allowed to do so, but he must ask and receive permission from his group leader; each foot soldier is “responsible to his commander for the money he spends and the equipment he uses”; a group of fighters cannot approach another group’s fighters to join it unless “written permission” is granted by higher-level commanders.11 Even the movement of foot soldiers and the firing of weapons is partially directed from above. Not surprising, membership in the Taliban’s leadership council is closely guarded. Updated and authoritative lists are difficult, if not impossible, to obtain, though a rough sketch of the Taliban’s leadership can be generally surmised. Between 2006 and 200812—the period under investigation here—the Taliban’s top echelon was made up of these leaders: • At the top sat Mullah Mohammed Omar, the one-eyed Taliban leader who ruled Afghanistan between 1996 and 2001. A recluse since the American invasion, Mullah Omar is currently in hiding and has turned over the Taliban’s daily war-fighting decisions to his senior commanders in the shura. Between 2006 and 2008, those leaders included Dadullah, Akhtar Usmani (Osmani), and Mullah Obaidullah, in particular and (according to some) Haqqani as well. On occasion, Omar releases audio recordings urging his fighters on, though his role within the Taliban has increasingly become ideological rather than tactical in nature. Nonetheless, as Taliban spokesperson Zabibullah confirms, “His word is law . . . No one can or dares to challenge Mullah Omar.”13 • Mullah Dadullah, the notorious commander introduced earlier, was at one time described as the Taliban’s top military commander in Afghanistan and a strong proponent of extreme and exceptionally violent tactics that included indiscriminate suicide bombings, mass kidnappings, and public beheadings. • Akhtar Usmani was a top military commander and was reportedly considered a successor to Omar in the event of his death. He was

Targeting the Taliban

129

also considered a “moderate voice” of the Taliban’s shura, at times apparently questioning the necessity of Dadullah’s brutality. • Mullah Obaidullah was defense minister under the Taliban regime. A top commander, he was instrumental in equipping insurgents with weapons and in organizing attacks. In addition to Mullah Omar and his top three commanders, the other shura members between 2006 and 2008 included the following: Amir Khan Muttaqi (Motaki), the Taliban representative during the 1990s; Mullah Mohammed Hassan Rehmani, Kandahar’s Taliban-era governor; Mullah Abdul Ghani Berader, a regional commander; Hafiz Abdul Majid (Majeed), Omar’s personal messenger; Abdul Razaq , the Taliban’s former interior minister; Qudratulah Jamal (aka Hamid Agha), information minister under the Taliban and the insurgency’s “chief propagandist”; and Mullah Abdul Kabir , the Taliban’s Nangarhar province governor and popular insurgent leader. All of these leading Taliban figures were, at one time, wanted men. NATO and Afghan forces repeatedly tried to capture or kill them, and have had, over the past few years, considerable success in doing so. The shura council members listed here have been heavily attrited: Usmani was killed in a NATO air strike in Helmand in December 2006; Obaidullah was captured in Pakistan in 2007 and later died in prison (confi rmed in February 2012);14 Dadullah was killed in May 2007; Berader was nearly killed in a precision air strike in August 2007 and was later arrested in Pakistan in 2010 in a joint CIA-ISI (Inter-Services Intelligence) operation;15 and Kabir was arrested in Pakistan in February 2010.16 Some of the others have also survived targeted operations. Besides these leading Taliban figures, other militant commanders and terrorist leaders have been openly sought and hunted down by U.S., NATO, and Afghan forces. In October 2007, American and Afghan forces began a formal “Most Wanted” campaign in Afghanistan, offering cash rewards for information leading to the capture or death of a dozen insurgent leaders suspected of conducting and organizing suicide attacks, planting roadside bombs, and ambushing security personnel. All over the country, more than 300,000 posters and several hundred billboards went up with the names and pictures of the wanted individuals (see Figure 5.1). Rewards ranged from US$20,000 to US$200,000. As Lieutenant Colonel Rob Pollock, a U.S. military spokesperson explained at the time, the campaign was “not necessarily a new program” but rather an attempt to get “more visibility” on par ticular individuals.17 NATO was communicating its intentions to capture or kill

130

Chapter 5

Figure 5.1. Afghanistan’s “most wanted” poster. Bill Roggio, “US Places Bounty on Senior Taliban and al Qaeda Leaders,” Long War Journal, January 26, 2008.

these men. Though the individuals making up the Taliban shura were also wanted men, those specified in the 2007 campaign were not the top-rung Taliban leadership but rather the mid- and high-level regional operators active in Afghanistan and Pakistan. Likewise, a number of different militant groups active in the region were represented. The list included Yuldash, the Islamic Movement of Uzbekistan’s leader introduced earlier; Saraj (Siraj) Haqqani, a senior leader of the Haqqani network; Abu Laith al-Libi, considered al Qaeda’s top commander for Afghanistan and Pakistan; Mohammad Rahim, the Taliban’s liaison with al Qaeda and Helmand province’s senior Taliban representative; ul-Haq, the Tora Bora Military Front’s leader, introduced earlier; Dr. Amin al-Haq, bin Laden’s chief of security; Mullah Sadiq, a Taliban liaison between Afghan and Pakistani factions; Baitullah Mehsud, Pakistan’s TTP leader; Darim Sedgai, a Haqqani network explosives expert; Dost Mohammed, the Taliban commander for Nuristan province; Mullah Nassar, the Taliban’s shadow governor for Ghazni province; and Qari Baryal, a Taliban chief active in Kapisa province’s Tagab Valley. Like the Taliban’s leadership, these figures have been repeatedly targeted. Between 2007 and 2008, Yuldash survived a number of attempts on his life and was eventually killed in a U.S. drone attack in August 2009. Saraj Haqqani survived numerous attempts on his life, including two separate attacks in February 2010 in which scores of other militants, including Saraj’s brother, were killed. Mehsud also survived a number of attacks before he was eventually killed by a U.S. hellfire missile dropped on his Pakistani

Targeting the Taliban

131

safe house in 2009. Baryal was killed in January 2011 in a precision air strike carried out by ISAF in Kunar Province, Afghanistan. Rahim was captured in Pakistan in August 2007 and handed over to the CIA, which held him secretly for several months before transferring him to Guantanamo Bay. AlHaq was captured in Pakistan in January 2008 but was released in September 2011. Al-Libi was killed in a January 2008 U.S. Predator drone strike in Pakistan. Sedgai was confirmed killed by “unknown” gunmen in Pakistan in January 2008. And in March 2013, ISAF and media reports confirmed that Dost Mohammed had been killed in a coalition drone strike in Kunar Province, Afghanistan.18 In addition to those listed here, open-source reporting suggests that several dozen other regional-level and district-level commanders, field lieutenants, bomb makers, and facilitators have been captured or killed in targeted operations and strikes since 2006. The list is exceptionally impressive.19 Lower-level facilitators, perhaps even dozens, also have been specifically targeted by ISAF as well as American and Afghan forces, but their names are rarely reported. These are the minor, local-level leaders and cell operatives. Their death or capture is usually accompanied by a short report that simply states that a “known Taliban leader,” “local group commander,” or “facilitator” had been targeted with precision force or killed in an operation. These are Afghanistan’s most-wanted militants. Their capture or death in carefully calculated precision strikes has become a central plank on which NATO and U.S. military strategy rests. What follows is an evaluation of four targeted strikes and their tactical and deterrent effects on militant behavior.

Coercion and Targeted Killings Measuring the coercive effects of targeted killings on militant organizations can be done by comparing several targeted killings and identifying the patterns of behavioral change by the group. Following is a comparative analysis of three targeted killings in Afghanistan; a fourth case is added to the aggregate data. As discussed in the Appendix, each case is compared along two overarching themes: militant professionalism (foiled, failure, and success rates) and the nature of militant behavior (type of violence and target selection). Each case study is further subdivided into two time frames: before and

132

Chapter 5

after the targeted strike. In each case, a comparison between the two periods reveals changes in Taliban behavior that is related to the particular elimination. The three cases under review include two senior Taliban leaders (Dadullah and Faiz Mohammad) and one regional-level Taliban subcommander (Matin). Each case begins with a brief description of the leader and is followed by an analysis.

Mullah Dadullah

Mullah Dadullah is the most notorious Taliban leader to have been targeted and killed in Afghanistan since the beginning of hostilities in 2001. As noted, Dadullah’s elimination was given a great deal of time, energy, and attention. The fact that ISAF decided to pursue him in a complex and dangerous operation—rather than bombard his known location from the relative safety of the air—and thereby risk the lives of coalition and Afghan soldiers in the process, reiterates the fact that in Dadullah’s case, NATO and Kabul wanted to absolutely get it right. Even more important, however, Dadullah’s death— much like Abu Musab al-Zarqawi’s 2006 elimination in Iraq and Osama bin Laden’s 2011 killing in Abbottabad—needed to be unequivocally established and widely publicized. It did not help that coalition and Afghan forces had already been embarrassed by Dadullah on a number of occasions: In February 2006, Afghan officials reported that he had been killed in a Helmand fi refight; in May 2006, Afghan officials again mistakenly claimed that he had been captured in Kandahar. After both claims, Dadullah contacted the media to report that he was feeling just fine.20 Gruesome as it sounds, Dadullah’s body needed to be recovered and broadcast. (Al-Zarqawi’s body was also retrieved and later shown to the public; bin Laden’s body was retrieved by U.S. forces and then buried at sea.) The idea was to use images of Dadullah’s corpse to galvanize Afghan (and Western) support for NATO’s mission. But it was also assumed that publicizing Dadullah’s demise would send a decisive message to his militant comrades regarding the intelligence and military capability of coalition and Afghan forces—a message that carried coercive and deterrent qualities. Dropping precision munitions would have likely killed Dadullah just the same, but Kabul was after much more. It wanted to use Dadullah’s elimination to dissuade others from taking his place, to create disorder within surviving ranks, to send other Taliban leaders into hiding, to keep Mullah Omar underground,

Targeting the Taliban

133

and to deal the Taliban a psychological blow. Is it any wonder, then, that former Kandahar Governor Asadullah Khalid spent the days following Dadullah’s death parading his body before television cameras? Dadullah was the Taliban’s top military strategist, a member of the Taliban’s inner circle, one of Mullah Omar’s closest confidants, the “moral authority” and “soul” that defined the Taliban, and a brutal, “blood-thirsty sadist” on the battlefield. It was Dadullah who insisted on adding suicide tactics to the Taliban arsenal, a tactic that was uncommon in Afghanistan before 2005. In a September 2006 interview with ABC, Dadullah claimed to have personally trained 500 suicide bombers ready for deployment and to having more than 12,000 foot soldiers under his command. “We have no shortage of fighters” he boasted. “In fact,” he continued, “we have so many [volunteers] that it is difficult to accommodate and arm and equip them. Some of them have been waiting for a year or more for their turn to be sent to the battlefield.” This was all bravado, of course, and proved a little hollow by mid2007, but it fed the cult-like status Dadullah had carefully cultivated among his followers. Perhaps Dadullah’s most important role was as brazen propagandist and charismatic champion of the Taliban’s cause and strategy. He often paraded in front of the media and starred in a number of violent jihadi snuff videos, all in an attempt to attract recruits. Germany’s Der Spiegel went so far as to argue that Dadullah was “worshipped like a saint” by his young followers. The late Asia Times reporter and Taliban expert Syed Saleem Shahzad wrote that Dadullah “derived loyalty because he fought alongside his men and suffered the same harsh conditions” as they did. He was an adept leader, both on and off the battlefield, a savvy propagandist, and a prized Taliban commander.21 In the days and weeks following his elimination, analysts suggested that his removal would have an effect on the Taliban’s 2007 spring offensive. “Replacing Dadullah,” a Western intelligence officer told Time magazine, “won’t be easy for the Taliban, in part because his own brutality may have eliminated a number of potential successors.”22 The likeliest of candidates, Usmani and Obaidullah, had previously been incapacitated. Shahzad adds that Dadullah’s demise marked “a shift of the Taliban’s military command into the hands of ‘non-Taliban’ and non-Kandahari commanders of southeastern Afghanistan.” Over time, Shahzad suggested, changes in leadership would weaken Mullah Omar’s grip over the direction of the Taliban’s insurgency, as younger, less staunchly supportive allies fi lled vacated positions.23 The loss of the charismatic commander was a huge blow to the Taliban

134

Chapter 5

because he was a symbol of Taliban success and no replacement of equal stature was evident. Elsewhere, Shahzad suggests further that Dadullah’s killing “set in motion a major change within the Taliban’s command structure.” In fact, one month following Dadullah’s elimination, Mullah Omar established new guidelines for his field commanders, suggesting that Dadullah’s targeted killing had a near immediate and identifiable effect on Taliban motivation and behavior. Consider Mullah Omar’s new rules of engagement: (1) No member of the Taliban’s central command will be allowed to work in southern Afghanistan because of the high risk of death; (2) control over military strategy would be decentralized and passed down to lower-level district commanders; (3) individual district strategies would be relayed to the Taliban’s appointed “shadow governor” who would then communicate the strategies to central command, and from there the central leadership would patch together a broader strategy for the country as a whole; (4) personality cults like Dadullah’s would be discouraged because the death of a “hero . . . demoralizes followers”; and (5) the Taliban’s media wing would be centralized, with four spokesmen appointed to specific conflict zones so that in the event of an arrest, the individual could divulge only a limited amount of intelligence. It is interesting to note that all four spokesmen were to use the same name when contacting media—Qari Yousuf Ahmadi (or Ahmedi). The name soon appears in media reports from Afghanistan.24 Each of these developments represents a shift in Taliban behavior that is in keeping with theories of deterring political violence and, more broadly, with the psychological consequences of targeted killings (demoralization, confusion, fear, and diminishment of morale). But empirically speaking, did Dadullah’s killing have the coercive effect it was expected to? Predictably, yes, and no. The following assessment involves a comparison of the three weeks before and after Dadullah’s death and incorporates data from all of Afghanistan. OVERALL VIOLENCE

There was a sharp increase in the number of insurgent and terrorist attacks in the weeks following Dadullah’s targeted killing. Attacks by IEDs, for instance, jumped from twenty-nine attempts before Dadullah’s elimination to forty-six attempts afterward. A similar surge is evident in small arms and rocket (SA/R) fire, with forty-one attacks before and seventy-five after his death. Only suicide bombings decreased, from seventeen to twelve (Figure 5.2).

Targeting the Taliban

135

75

80 70 60 46

50

41

40 30

29 17

20

12

10

5

5

0 IED

SA/R Before TK

Suicide Bombings

Assassinations

After TK

Figure 5.2. Acts of terrorism (total)—Dadullah. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “SA/R” stands for “small arms and rocket fire.”

On the surface, it would seem that Dadullah’s elimination had very little coercive influence on the overall level of insurgent violence in Afghanistan. His death may have actually encouraged his followers to increase their immediate activities, if only to highlight that their leader’s death would have little impact on their ability and will to continue fighting. General Mohammad Zahir Azimi, a senior Afghan defense ministry spokesperson, argued that following Dadullah’s elimination, Kabul witnessed “a kind of revengeful reaction from Taliban.”25 A similar sentiment was expressed by Mullah Omar, through fictitious spokesperson Yousuf Ahmadi: “There are hundreds and thousands of mujahedin who have fought under Dadullah and there are hundreds and thousands of mujahedin who are able to replace Mullah Dadullah very well. . . . [His death] is not going to slow down the Taleban Jihad.”26 Omar’s sentiment is reflected in the data: All acts of terrorism except for suicide attacks increased after Dadullah’s targeted killing. Taken at face value, it would seem that Dadullah’s elimination had almost no positive effect (deterrent or otherwise) on his organization’s ability, determination, and motivation to organize and take part in terrorist violence. TALIBAN PROFESSIONALISM

Yet, the surge in violence was actually much less impressive than at first glance. A closer look at the data reveal that although the overall number of attacks may have increased following Dadullah’s killing, overall professionalism

136

Chapter 5 17

18

15

Number of IEDs

16

14

14 12

11

12

10 8

6

6 4 2 0 Controlled

Simple Before TK

Failed After TK

Figure 5.3. IED professionalism—Dadullah. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.”

decreased significantly. So even though more IED attacks were carried out after his death, there were also higher levels of IED failure. After the targeted killing, one in every three attempted IEDs failed to detonate (or was uncovered and neutralized), a figure that had been closer to one in five before Dadullah’s elimination (Figure 5.3). Even then, however, IED attacks remained just as deadly, killing, on average, two individuals for every successful detonation (the same ratio as before Dadullah’s death). Furthermore, of all IED victims, hard targets made up the majority of those killed, 72 percent before and 90 percent after Dadullah’s killing (Table 5.1). Data on small arms and rocket fire reveal an opposite trend. Besides the near doubling of SA/R attacks after Dadullah’s killing, on average the attacks got deadlier. Before Dadullah’s killing, almost seven militants were killed for every one ISAF soldier, whereas after the targeted attack, just over four insurgents died for every coalition soldier. However, SA/R fire against the Afghan National Army (ANA) and Afghan National Police (ANP) reveals a different trend altogether: Before the targeted raid, just over two insurgents died for every three Afghan security officers in SA/R attacks, while after the Dadullah killing, that ratio reversed itself to just under four dead ANA/ANP for every five insurgents killed (Table 5.2).27 Finally, at first glance, the professionalism of Taliban suicide bombings seems to have dramatically improved following Dadullah’s death. For instance, before the targeted strike, more than half of all suicide bombing missions failed; after the Dadullah strike, that figure dropped to one in four.

Table 5.1. IED Success and Kill Ratios—Dadullah Period IED success

Before TK

After TK

Detonations

23

31

Impact ISAF/PRT ANA/ANP Private security Afghan government Civilians

Deaths/injuries 3/3 31/28 0/0 7/3 7/9

3/9 46/24 9/11a 4/0 2/11

Note: “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.” The remaining acronyms can be found in the body of this chapter. a Th is includes a double bombing against a PSG team, with the fi rst bomb detonated against the convoy and a second detonated against first responders securing the site a few moments later. In total, four PSG and three ANP were killed.

Table 5.2. SA/R Success and Kill Ratios—Dadullah Period Small arms and rocket fire Total

Before TK

After TK

41

75

Success rates ISAF/PRT Insurgents ANA/ANP Insurgents

SA/R deaths a

3 20 30 23

11b 49 39 50

Note: “TK” stands for “targeted killing”: “SA/R” stands for “small arms and rocket fire.” The remaining acronyms can be found in the body of this chapter. a The May 7 SA/R attack outside a high-security prison that killed two U.S. troops and wounded two more was carried out by an individual wearing an ANA uniform. Initial reports suggested the soldier suffered from mental illness, though often Taliban fighters have used ANA and ANP uniforms to ambush members of the ISAF. Th is became evident in 2012 and 2013 when so-called “insider” attacks became a very serious security problem. Furthermore, in this par ticu lar case, the Taliban expressly issued a claim of responsibility for the attack. “Afghan Who Shot GIs Have Been Mentally Ill,” Associated Press, May 7, 2007. b These figures include two SA/R attacks (May 14: five ISAF, twelve ANA, and ten insurgents dead; May 26: one ISAF dead and four injured) that followed IED ambushes.

138

Chapter 5

Likewise, although five suicide bombers killed only themselves before Dadullah’s elimination, not one suicide bomber failed to kill at least one other individual after his targeted elimination. Yet of these post-Dadullah bombings, only one operative succeeded in killing his noncivilian target, an ISAF soldier on foot patrol in Konduz. The latter kill ratio therefore reflects the fact that civilians suffered the brunt of these suicide attacks. Similarly, bomberto-nonbomber kill ratios rose after Dadullah’s killing, from 1:2 before to nearly 1:5 afterward; the bombers appear to have gotten deadlier (see Figure 5.4 and Table 5.3). TYPE AND NATURE OF ATTACK

Number of Suicide-Bombing Attempts

An improvement in the quality of suicide attacks following Dadullah’s targeted killing is an unanticipated (and a theoretically contradictory) fi nding. However, a deeper and more nuanced reading of the data illustrate that post-Dadullah suicide-bombing successes were in great part a result of a Taliban shift toward selecting softer targets. Taliban suicide blasts became deadlier, overall, because of the type of target they were used against. Consider that eleven of the thirteen known targets of suicide bombers before Dadullah’s elimination (85 percent) were hard targets (ISAF, provincial reconstruction team [PRT], ANA, ANP, and private security guards [PSG]). After Dadullah’s killing, known hard-target selection dropped substantially to 40 percent. In fact, no suicide bomber targeted ANA or ANP forces specifically following Dadullah’s elimination (though 18 16

2

14

2

12 10

5

2 1

8

9

Pre-TK

Post-TK

8 6 4 2 0 Successful

Stopped

Arrested

Premature

Figure 5.4. Suicide-bombing professionalism—Dadullah. “TK” stands for “targeted killing.”

Targeting the Taliban

139

Table 5.3. Suicide Bombings: Success—Dadullah Period Suicide bombings Total attempts Body-borne Vehicle-borne

Before TK 17 4 4

12 4 5

Success rates Bombers Nonbombers Number of detonations in which only bomber dies

After TK

Deaths 16 35 5

a

9 44 0

Note: “TK” stands for “targeted killing.” a Another sixteen Taliban were killed following a suicide attack when ISAF and ANA forces returned fire following a coordinated Taliban ambush.

they were the favored target before his death). Instead, after Dadullah’s killing, well over half of all known targets of suicide bombers were soft targets: international officials, government of Afghanistan (GoA), NGO, and civilians (see Table 5.4). This shift toward soft-target selection helps explain why post-Dadullah bombing successes increased precipitously. Soft targets, especially civilians and GoA officials, cannot easily defend themselves against suicide bombers and are less able to mitigate the effects of a bomb’s blast. They are simply easier to attack and more readily suffer the consequences. Consider further that of the nine failed suicide bombings recorded before Dadullah’s death, five (and possibly a sixth) were preempted by vigilant security personnel. On April 23, for instance, the ANA received a tip concerning a potential bomber and, on locating the man, gave chase. In the pursuit, the bomber’s vest exploded, having been either struck by a bullet or purposefully detonated. His was the only death recorded. And again, two days later, the ANP suspected a suicide attack after smoke appeared from under the clothes of a 15-year-old bomber. He was shot and arrested.28 It is difficult, if not impossible, however, for civilian and government targets to do the same. Soft targets may lack the updated intelligence and the means to defend themselves as effectively as security forces. The fact that soft targets made up 60 percent of post-Dadullah suicide bombings helps explain why bomber success rates increased over that same period.29 It was not that the quality of the bombers

140

Chapter 5

Table 5.4. Suicide-Bombing Target Selection—Dadullah Period Suicide bombing Total Target selection ISAF ANA PSG ANP International government Government of Afghanistan NGO Civilian Unknown

Before TK

After TK

17

12

3 4 0 4 0 1 0 1 4

3 0 1 0 0 3 0 3 2

Note: “TK” stands for “targeted killing.” The remaining acronyms can be found in the body of this chapter.

improved, but rather that the quality of the targets diminished. This is critical. Alone, the data on suicide-bombing professionalism is misleading. A more refined assessment that takes into account the target selection of suicide bombers after Dadullah’s elimination reveals a shift in target selection that clarifies the sudden increase in suicide-bombing professionalism. The data on the use of small arms and rocket fire also exhibit a Taliban shift in target selection, with attacks against ANP forces jumping from 24 percent of total attacks before Dadullah’s elimination, to more than 40 percent afterward (Figure 5.5). Of all armed groups working in Afghanistan, ANP forces may be the easiest to attack successfully. Though armed and numbering in the tens of thousands, ANP forces often lack the sophistication, equipment, and training to properly contend with heavily armed and battle-trained insurgents. This was especially so in 2007, before NATO and other UN partners put serious emphasis (and funding) on training Afghan police. What is more, under-protected and far-flung ANP checkpoints dot the Afghan landscape and represent a fi xed position for Taliban fire. In addition to defending themselves against insurgent attacks, ANP personnel must also try to impose law and order and conduct counter-narcotic operations, all for a salary of less than $100 a month. An official with Afghanistan’s Ministry of Interior (MoI) put it best: “Our enemies attack with

Targeting the Taliban SA/R Targets (Pre-TK)

SA/R Targets (Post-TK) Civilian, 2

Civilian, 6 NGO, 1

141

GoA, 8 ISAF, 13

ISAF, 19

GoA, 4 Int’l Gov, 1 ANP, 31 ANP, 10

ANA, 6

ANA, 13 PSG, 2

Figure 5.5. SA/R target selection—Dadullah. “TK” stands for “targeted killing”; “SA/R” stands for “small arms and rocket fire.”

RPGs, missiles and other sophisticated weaponry, while our police defend themselves only with old Kalashnikovs.”30 It would seem that following Dadullah’s targeted killing, insurgent groups were desperate to prove their mettle and did so by targeting easy prey. Doubling SA/R attacks against ANP positions in the weeks following Dadullah’s death was the easiest (and least-sophisticated) way to do that. And targeting the weakest of all armed security groups ensured the highest odds of success at the lowest cost to the Taliban. One final shift in insurgent strategy that is clearly evident following Dadullah’s killing is the nature of Taliban assassinations. Although the number of monthly Taliban assassinations in 2007 (where evident) usually numbered in the low, single digits, a pattern is immediately evident in the Dadullah case. In the three weeks before the raid, five assassination attempts were recorded: GoA senator Abdul Sabur Farid Kuhestani was assassinated by gunfire outside his residence; an ANP criminal investigator was shot and killed outside his home; Abdul Hamid Khan, an irrigation project’s commanding security officer, was assassinated; the governor of Sari Pul province was uninjured when a bomb detonated near his home; and Filip Velach, the Czech Republic’s highest ranking Afghan diplomat, escaped an attack on his convoy in Paktia Province that resulted in the death of several insurgents and Afghan soldiers. After Dadullah’s targeted killing, another five assassinations took place, but, unlike those before his death, most were part of a protracted Taliban attempt to kill Kandahar governor Khalid in par ticu lar. Th ree of the five attempts were on Khalid’s life. Furthermore, each attempt used suicide bombers, whereas previous assassinations were usually conducted using less sophisticated methods. That two suicide bombers were dispatched the very

142

Chapter 5

week of Dadullah’s elimination (one on May 17 that hit the governor’s convoy, injuring cultural minister Karim Khurram, and a second two days later that again attacked Khalid’s SUV) and a third the week after (on May 26 against Khalid’s residence) is a likely indication that the Taliban desperately wanted to get even with Governor Khalid for so blatantly savoring Dadullah’s death.31 The Taliban was out for revenge. In sum, Dadullah’s case is insightful. First, his elimination did not deter his comrades from carry ing out acts of political violence. Instead, his death led to an increase in the tempo with which some forms of attack (IED and SA/R)—but not others (suicide bombings)—were carried out. Second, even though a greater number of attacks took place after Dadullah’s death, there was a related decrease in Taliban professionalism. Third, target selection following Dadullah’s elimination shifted toward softer targets. Fourth, Dadullah’s death provoked a visceral Taliban reaction to kill high-profi le members of the Afghan government, notably Governor Khalid, the very man who presented Dadullah’s corpse to the world. Finally, apart from the empirical analysis offered here, Dadullah’s killing also appears to have led to a substantial degree of spy mania within Taliban ranks. According to Rahimullah Yusufzai, a Pakistani journalist and expert on the Taliban, after Dadullah’s elimination, “suspicion [was] falling even on trusted men and [was] creating tension in Taliban ranks.”32 Suspected spies were rounded up, tortured, and beheaded, though most of those spies were reportedly little more than local Afghan chiefs and individuals who had had some contact with NATO and Afghan forces or the GoA. Others have reported that because of the supposed threat of spies within the organization, some Taliban leaders had become “extra careful in selecting fighters to serve as their bodyguards.”33 As previous chapters highlight, when the fear of future eliminations grips a militant organization to the point that bloody purges, defections, and betrayals become common, the group’s overall capability and motivation diminishes. To a certain extent, this is a coercive success born of targeted killings.

Qari Faiz Mohammad

On July 23, 2007, coalition and Afghan soldiers conducted a targeted operation in Helmand province, killing Qari Faiz Mohammad. Lieutenant Colonel Claudia Foss, ISAF spokesperson, noted at the time that the precision

Targeting the Taliban

143

operation against Mohammad would ensure “insurgent disruption,” adding that “the Taliban leadership has suffered another severe setback.”34 Details of the raid have remained fuzzy, though the strike—targeting a “known Taliban commander”—was conducted as part of a multipronged and extensive two-day ISAF operation in Helmand that resulted in the death of more than fift y insurgents, a number of Afghan forces, and one ISAF soldier.35 Though Mohammad was not as well-known in the West as other senior-level Taliban leaders, he nonetheless retained a critically important position within the orga ni zation and had been the target of an earlier U.S. operation (in 2004). Afghan analysts posit that Mohammad was the chairman of the Taliban’s military council, a leading strategist for the insurgency, a chief financier for Taliban forces, and a close and personal associate of Mullah Omar’s.36 Following the incapacitation of the shura’s big three—Usmani, Obaidullah, and Dadullah—it is possible, even likely, that Mohammad had risen to a position of greater prominence. Even then, unlike Dadullah, Mohammad sought neither public nor media attention and generally remained within the shadows of the Afghan insurgency, as instructed by Mullah Omar. Yet, Mohammad’s elimination in July 2007 did have a notable coercive effect on the Taliban’s behavior. Mohammad’s case study, like Dadullah’s, rests on events and developments spanning a period of a month and a half—three weeks before and three weeks after the targeted killing—and uses data from the entire country. OVERALL VIOLENCE

Unlike the Dadullah case, Mohammad’s targeted elimination corresponds with a sharp decline in all major acts of violence and terrorism. Attacks using IEDs and SA/R both dropped by more than 10 percent, while suicide bombings declined by half. Only assassination attempts rose during the Mohammad case, by a factor of almost two. Taliban abductions had become a prevalent tactic by July 2007 and were evident in the data collected during the Mohammad time period. Though kidnappings had always been a threat— particularly to international civilians working with the government of Afghanistan, coalition government employees, media personnel, NGO staff, and Afghan civil-society leaders—the period under review illustrates consistently higher levels of abductions than previously (and subsequently) recorded (Figure 5.6). Finally, unlike Dadullah’s elimination, the Mohammad killing did not elicit the same visceral, revenge-driven reaction from Taliban forces.

144

Chapter 5 111

120

95

Total Attempts

100 80 60 40

44 39 17

20

8

12

7

16 16

s Ab

du

ct

tio in a ss sa As

m Bo e Su

ici d

io n

n

s bi ng

/R SA

IE

D

0

Before TK

After TK

Figure 5.6. Acts of terrorism (total)—Mohammad. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “SA/R” stands for “small arms and rocket fire.”

These preliminary fi ndings corroborate theoretical expectations. Mohammad’s targeted killing not only resulted in a drop in the use of complex acts of violence, like suicide bombings, but also led to an increase in less sophisticated forms of violence, like the assassination and abduction of soft targets. Joanna Nathan, a senior researcher at the International Crisis Group, explains the Taliban’s periodic use of less formidable forms of violence and terrorism as an attempt to “create a general air of instability” rather than an attempt to gain impressive tactical victories. In Afghanistan “there’s a saying,” she posits, “that foreigners have the watches and insurgents have the time.”37 Given that the Taliban will never be able to outgun ISAF, NATO, or the United States, its strategy is one of attrition: to simply wear down the other side. That is not to argue that the Taliban would not take over Kabul, Jalalabad, or Kandahar if it had the means. Indeed, during the summer of 2007, the Taliban tried—and failed badly—to muster platoon-like forces for traditional or conventional battle with Afghan and coalition forces. At every point during that campaign, the Taliban was turned back, having suffered significant losses. Since then, Afghan militants have relied heavily on hit-and-run tactics, like IED attacks, suicide bombings, and SA/R ambushes, and generally have avoided engaging NATO and Afghan forces in full-frontal attacks. That change in behavior was born of necessity; the Taliban insurgents were denied the fruits of their labor. Their increasing reliance on intimidation,

Targeting the Taliban

145

hostage-taking, public executions, beheadings, and assassinations can likewise be attributed to their relative weakness. So although it might be true that the Taliban desires an “air of instability” in Afghanistan, the organization nonetheless also has a rank order of tactical preferences where certain acts of political violence are accorded greater value than others. Above all else, the Taliban desires control over territory and victory over ANA/ISAF forces in traditional military engagements. The group tried that in 2006 and 2007, and failed. The second-most-preferred outcome is high kill ratios of Western forces through large bombing attacks against hard targets (ISAF and ANA). The idea, in this case, is to shift public support in the West against NATO’s mission by successfully (and repeatedly) killing Western soldiers. When these sorts of attacks become too difficult to carry out—or when success rates continue to drop—the Taliban would likely consider hitting secondary targets (ANP and the GoA) instead. Small arms and rocket fire against lightly armed groups (ANP, GoA, and civilian transport trucks) that do not risk high Taliban casualty rates might come next on the Taliban’s list of preferences, with intimidation tactics against Afghan civilians coming last. The point is that speculation on Taliban preferences allows us to demonstrate coercive successes by highlighting how certain military tactics result in shifting Taliban behavior. If the Taliban was prevented from successfully carrying out its most-preferred tactic, as was indeed the case between 2006 and 2008, then this represents an important finding for studying coercion in counterterrorism and counterinsurgency. A closer inspection of the Mohammad data reveal a number of expected trends associated with deterring political violence and targeted killings, including deprofessionalization and changes in target selection. TALIBAN PROFESSIONALISM

Professionalism decreased (in some cases, precipitously) following Mohammad’s targeted killing. For instance, IED attacks failed twice as often following Mohammad’s death (Figure 5.7). Of these failures, most of the IEDs were uncovered by security personnel and dismantled. There was, however, one notable premature IED detonation in the weeks preceding Mohammad’s killing. On July 12, an explosion occurred at a suspected IED factory in Kandahar. Upon searching the premises, ANA found two dead bomb makers, along with a large cache of weaponry, including four suicide vests, an IED manual, fuses, explosives, and remotely activated IED switches.38 It is possible that this event impeded IED attacks in the three weeks following

146

Chapter 5

Number of IEDs

25

23

20 15

12

16

14 11

10

7

5 0 Controlled

Simple Before TK

Failed After TK

Figure 5.7. IED professionalism—Mohammad. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.”

Mohammad’s targeted killing, though there were likely many such factories in Afghanistan. Furthermore, in the eleven days between the Kandahar blast and the targeted raid, the Taliban could have dispatched another IED cell to the city to replace the one that self-destructed. Besides a drop in overall IED attacks and a diminishment in IED professionalism, there was also a recorded decrease in IED effectiveness after Mohammad’s targeted killing. Fewer people were killed, comparatively speaking, for every IED attack. Consider that before the Mohammad raid, a total of sixty-eight individuals were killed by thirty-seven successfully detonated IEDs—almost a 2:1 ratio. After the targeted raid, only nineteen deaths were recorded in twenty-three IED detonations—less than a 1:1 ratio. This is a noteworthy development; not only were IEDs failing more often, but the ones that did explode caused far less damage. And yet, of IED victims, only security personnel were killed in IED blasts following Mohammad’s elimination. Before then, twenty-three civilians (roughly one-third of the total) were killed by insurgent bombs (Table 5.5). It is possible that the Taliban made a conscious and concerted effort around the time of Mohammad’s elimination to limit civilian casualties in order to avoid alienating their popular bases of support, something the Taliban seems to do often enough. After all, the hearts-and-minds dictum applies equally to the Taliban and other militant groups, as it does to ISAF, the United States, and the Afghan government. But if trying to win over the Afghan population while killing them indiscriminately sounds counterproductive, that’s because it is. The Taliban stands firm, however: “The target of the bombings that the mujahideen carry out on the sides of roads,” reads one Taliban statement from April 2007, “is always the foreign occupation troops

Targeting the Taliban

147

Table 5.5. IED Kill Ratios—Mohammad Period IED success

Before TK

After TK

Detonations

37

23

Impact ISAF/PRT ANA/ANP PSG Civilians

Deaths/injuries 10/1 29/34 6/4 23/25

6/4 11/24 2/0 0/5

Note: “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.”

and their agents. The mujahideen denounce attacks that result in civilian casualties.”39 Later, in 2009, the Taliban published a lengthy book of rules, or a “military code of conduct,” for its foot soldiers. Al Jazeera provides a partial translation of the document online. Of importance, the document reads: “[Taliban] Governors, district chiefs and line commanders and every member of the Mujahideen must do their best to avoid civilian deaths, civilian injuries and damage to civilian property. Great care must be taken. . . . Suicide attacks should only be used on high and important targets. A brave son of Islam should not be used for lower and useless targets. The utmost effort should be made to avoid civilian casualties.” 40 For every Afghan civilian killed by Taliban violence, others are alienated from the Taliban’s cause. But even then, more Afghans die at the hands of the Taliban and other terrorist groups than by ISAF, ANA, or ANP activity.41 Data on small arms and rocket fire reveal an opposite trend. Though SA/R attacks dropped in the period following Mohammad’s killing, the attacks carried out nonetheless got deadlier. Before the Mohammad raid, twenty-two insurgents were killed for every one ISAF soldier targeted with SA/R fire. That figure dropped to thirteen insurgents for every one ISAF after the raid (Table 5.6). Likewise, against ANA and ANP forces, almost three insurgents died for every one security personnel before the targeted killing, a figure that declined to two insurgents for every one Afghan security officer afterward. Finally, during the time frame incorporated in the Mohammad case study (July 2 through August 14, 2007) a number of rockets were fired against coalition and Afghan forces; a total of twenty-five as opposed to only five

148

Chapter 5

Table 5.6. SA/R Kill Ratios—Mohammad Period Small arms and rocket fire Total

Before TK 111

Success ratesa ISAF/PRT Insurgents ANA/ANP Insurgents PSG Insurgents GoA Insurgents

After TK 95 SA/R deaths

7b 155 50 126 6 10 2 11

6 80 67 129 3 7 0 3

Note: “TK” stands for “targeted killing”; “SA/R” stands for “small arms and rocket fire.” The remaining acronyms can be found in the body of this chapter. a The figures included here relate to the cumulative results (e.g., number of deaths) of specific SA/R engagements between Taliban insurgents and their respective targets (e.g., ISAF/PRT, ANA/ANP, PSG, GoA). b Another thirty ISAF were reported injured, some seriously.

recorded during the Dadullah period (April 21 through June 2, 2007). Though details on rocket fire is sketchy, those recorded here are neither rocket propelled grenades (RPGs)—which are usually reported as either “RPG” or “gunfire” in security reports and are, for the purpose of this study, considered small arms fire—nor the more sophisticated rockets launched by other militant groups—like Hezbollah’s Fajr, Zelzal-2, and Katyusha rockets or Hamas’s Qassam and Grad rockets. Taliban rockets rest somewhere in the middle and, between 2006 and 2008, most were crude mortars (the most sophisticated being the 107 mm multiple-rocket launch system) and various antitank weapons. When in power, the Taliban regime relied heavily on vehicle-mounted AT-1 Snappers—a Soviet-era weapons system that was largely destroyed in the first weeks of Operation Enduring Freedom in October 2001. Some evidence points to more sophisticated Taliban rocket technology, including Chinese-made HN-5 shoulder-fired missiles, which are effective against low-flying helicopters, and a number of U.S.-made Stinger missiles, leftover from the anti-Soviet conflict of the 1980s. In 2002, for instance, U.S. forces destroyed an abandoned cache of thirty such missiles, and

Targeting the Taliban

149

ISAF helicopters have on occasion been fired on by sophisticated surfaceto-air missiles.42 One notable example came on July 22, 2007, when the Taliban fired its first heat-seeking surface-to-air missile (SAM) at a U.S. Hercules in Nimruz province. The attack was widely reported in the international press and was used to underscore the severity of the Afghan security environment.43 (The pilots used evasive maneuvers to avoid being hit.) But in general, and regardless of the specific type of projectiles, most of the Taliban’s rocket and mortar fire misses its mark. Of the twenty-five rockets launched during the time period covered in this case study, twenty-two failed to hit their target, many of them exploding harmlessly in open fields (Table 5.7). Suicide bombings during the Mohammad case-study time period also decreased in both total attempts and professionalism. Before the Mohammad elimination, just over one in every three bombings failed. (In one successful attack, militants forced a young child to push a cart toward an ANP position. Once in position, they remotely detonated the bomb hidden in the cart, killing the child and two ANP.)44 But after the Mohammad raid, well over half of all suicide attacks failed (see Figure 5.8 and Table 5.8). The ratio of successful bombers who nonetheless failed to kill other individuals in their attacks decreased slightly in the weeks following the targeted killing, from 67 percent before to 57 percent afterward.45 Nonetheless, overall bomberto-nonbomber kill ratios dropped significantly. Before Mohammad’s death, on average, one nonbomber died for every attempted suicide attack.46 After the targeted killing, it took roughly two bombings to kill one nonbomber. More specifically, bomber success rates against hard targets decreased even more precipitously following Mohammad’s elimination. Before the targeted Table 5.7. Rocket-Fire Target Selection—Mohammad Period Rocket fire

Before TK

After TK

Launches Misses

10 8

15 14

Impact Soft targets Hard targets Note: “TK” stands for “targeted killing.”

Deaths/injuries 1/2 0/4

0/4 0/0

Chapter 5

Number of Suicide-Bombing Attempts

150 18 16 14 12 10 8 6 4 2 0

2 2 1

3

12

1 1 3

Before TK Successful

After TK Stopped

Arrested

Premature

Figure 5.8. Suicide-bombing professionalism—Mohammad. “TK” stands for “targeted killing.”

Table 5.8. Suicide-Bombing Kill Ratios—Mohammad Period Suicide bombing Total attempts

Before TK

After TK

17

8

Success rates

Deaths/injuries

Bombers ISAF ANA/ANP International government Civilians

16 1/17 16/19 0/2 0/0

7 0/3 1/4 0 3/14

Note: “TK” stands for “targeted killing.” The remaining acronyms can be found in the body of this chapter.

raid, roughly one security personnel was killed for every suicide bomber. After the raid, only one security officer died in total. TYPE AND NATURE OF ATTACK

Notwithstanding their failure rates, suicide bombers consistently targeted hard targets during the period under review. Of the twelve suicide detonations preceding Mohammad’s targeted killing, only one took place against a soft target. (A Turkish vehicle convoy, with diplomatic plates, was struck by a body-borne suicide attacker in Kabul on July 18, 2007.)47 And during both halves of the Mohammad case study, ISAF suffered the brunt of all attacks

Targeting the Taliban

151

while ANP forces were targeted on every third occasion. Unlike the suicidebombing barrage targeting Governor Khalid following Dadullah’s targeted killing, no similar trend is evident following the Mohammad raid. In similar fashion, hard targets were the primary target of choice for SA/R fire. If anything, there seems to have been a slight shift toward targeting military actors following the raid. ISAF and ANA forces, for instance, made up roughly 20 percent of SA/R targets before the raid, and 30 percent after. Unlike the Dadullah case, which recorded a substantial swing in SA/R attacks against ANP forces after Dadullah’s death, no similar change is recorded here, with ANP targeted consistently at just under 50 percent. Finally, the postMohammad period does reveal a decrease in the number of SA/R attacks against soft targets, which accounted for a quarter of all SA/R engagements before and one-sixth of engagements after the raid (Figure 5.9). Target selection for IED attacks again seems to indicate a shift away from soft targets. Of suspected IED target selection, attacks against ISAF or SA/R Targets (Before TK)

SA/R Targets (After TK) Civilian, 10

Civilian, 10

ISAF, 23

NGO, 5

ISAF, 22 GoA, 7

ANA, 1

GoA, 13

PSG, 3 ANA, 7 PSG, 3 ANP, 39

ANP, 48

Suicide-Bombing Target (Before TK)

Suicide-Bombing Target (After TK)

Int’l Gov, 1 ANP, 2 ISAF, 5 ANP, 4 ISAF, 4 PSG, 1

ANA, 1

Figure 5.9. SA/R and suicide-bombing target selection—Mohammad. “TK” stands for “targeted killing”; “SA/R” stands for “small arms and rocket fire”; “Int’l Gov” stands for “international government.” The remaining acronyms can be found in the body of this chapter.

152

Chapter 5

ANA personnel jumped from barely one in twenty before the raid to nearly half afterward. Likewise, IED attacks against ANP dropped from half of all IEDs to a quarter (Figure 5.10). Strikes against soft targets dropped in similar value. As noted in the introduction to this section, assassinations and abductions increased markedly following Dadullah’s death in May 2007. Although abductions remained constant over the Mohammad reporting period, a slight increase in assassinations was evident. However, while all seven assassination attempts prior to Mohammad’s killing were against civilians, half of those conducted afterward were against (off-duty) security personnel. Of the abductions, Afghans made up the majority of those kidnapped, except, most notably, for twenty-three South Korean Christian missionaries abducted on July 19, 2007 (Table 5.9). (The Taliban killed two of the male South Korean hostages and released the others following negotiations with Korean officials.)48 In sum, findings from the Mohammad case study suggest that, unlike the Dadullah case, overall acts of terrorism dropped significantly in the weeks following the targeted elimination. The IED attacks, SA/R engagements, and suicide bombings all dropped—suicide bombings by 50 percent. Over the same period, Taliban forces conducted many more assassinations and continued abducting civilians at a much higher rate than was evident during the Dadullah case study. Both findings help corroborate the study’s theoretical assumption that as leaders are eliminated their groups find it more difficult to properly organize terrorist acts and thus purposefully decide to rely on less sophisticated forms of violence. Although some commentators IED Targets (After TK)

IED Targets (Before TK)

NGO, 1 ISAF, 1 Civilian, 5

GoA, 1

ISAF, 4

PSG, 2

GoA, 1 Int’l Gov, 1 ANP, 7 ANP, 13

ANA, 3 PSG, 1

Figure 5.10. IED target selection—Mohammad. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “Int’l Gov” stands for “international government.” The remaining acronyms can be found in the body of this chapter.

Targeting the Taliban

153

Table 5.9. Abduction and Assassination Targets—Mohammad Period Before TK

After TK

Abductions Total

16

16

Targets Civilian (Afghan and international) ANA/ANP personnel Beheadings

46a 5 8

25b 4 0

Total

7c

12d

Targets Government of Afghanistan ANA/ANP personnel International government Civilian Success rate (killed)

6 0 0 1 4

4 6 0 2 12

Assassinations

Note: “TK” stands for “targeted killing.” a Including interpreters, GoA workers and supporters, engineers, the South Korean missionaries, and two German engineers. b Including a number of GoA workers, supporters, and judges; medical professionals; businessmen; one Danish journalist; and one Pakistani engineer. c Targets included a chief of Showak district, Paktia province; a district manager of Mata Khan district, Paktika province (killed); a pro-government cleric Mawlana Ahmadzada (killed); Khost’s provincial education department, deputy head, Sayed Usman Hussaini (killed); MP Mula Pirumqul, targeted twice, in Takhar province; a former ANP officer; and a GoA elder in Khost province (killed). d Targets included a Balkh district governor; two government district administrators (both in Paktika); a former ANA commander, Haji Fazal Mohammad; four judges (each killed); a man who owned a shop on a Paktya PRT base (killed); a criminal investigator from Wardak (killed); a civilian translator (killed); and four (presumably off-duty) ANP officers killed in their homes.

have interpreted the Taliban’s reliance on abductions as a successful shift toward “emotional warfare” that is as powerful a tool as any bomb, a more appropriate interpretation would argue that the shift was born of desperate necessity and is rather a sign of weakness.49 Indeed, prior to 2005, some Taliban fighters would have thought it beneath them to trade individuals for money. Only unruly thugs trade hostages; the Taliban behave with pious merit. In

154

Chapter 5

fact, the 2006 and 2009 Taliban rule books strictly forbid the trading of hostages and prisoners for money, except in extreme cases. Accordingly, the Taliban’s persistent use of foreign captives to coerce coalition governments— or, worse, to help pay for weapons—is a sign of Taliban regression, not progression. Overall, professionalism also dropped during the Mohammad case, as it did during the Dadullah case. The IED, rocket, and suicide attacks not only failed more often following the targeted killing, but their success rates fell as well. Fewer people were killed for every IED planted, rocket launched, and bomber dispatched. The data on small arms fire, however, show an opposite trend, with these attacks dropping in number but increasing in professionalism. Unlike the Dadullah case, however, there was no evident shift in Taliban target selection after Mohammad’s killing. Although a shift toward softer targets is expected, in no category did such a trend develop: After the Mohammad killing, all suicide detonations targeted security personnel; SA/R fire against security forces rose from 68 percent to 75 percent; and IED detonations, where evident, targeted ISAF and ANA personnel much more often following the Mohammad raid. It is important to note that in the six-week period covered by the Mohammad case, a number of regional developments took place that could have influenced some of these findings and, especially, this last trend regarding target selection. As in any complicated social environment, uncontrollable external variables always exist. The trick is to identify which ones matter and explain why. First, the Mohammad elimination was not the only one to have taken place in Afghanistan during the period under review. In the weeks leading to Mohammad’s death, two low-level commanders were arrested and a number of other commanders were killed. The loss of these commanders along with the death of Mohammad slightly muddies the analytic waters. The removal of these leaders could have superficially amplified this case study’s findings. Likewise, Mohammad’s elimination came only ten weeks after Dadullah’s targeted killing, which was an exceptionally traumatic event for the Taliban. It is possible that even by July, Dadullah’s loss was still being felt. For instance, Shahzad wrote that “amid the demoralization” that followed Dadullah’s targeted killing, “the entire Taliban leadership left Helmand, Uruzgan, Zabul and Kandahar [provinces] and sat idle in . . . Quetta, Pakistan, for several weeks.”50 Saleh Mohammad Saleh, a Radio Free Afghanistan correspondent, agreed. In mid-July, Saleh obtained

Targeting the Taliban

155

a pamphlet—often referred to as Taliban Night Letters, used to warn or inform local Afghan populations of Taliban demands—reportedly signed by midlevel Taliban commanders from southern Afghanistan. In it, the commanders oppose the Taliban shura’s decision to place non-Afghan al Qaeda leaders in charge of Taliban operations. The move followed the Taliban’s shura council meeting in Quetta, in late June or early July, just after Dadullah’s elimination. “We criticize the decision of Mullah Mohammad Omar” the letter reads, and “don’t accept any other commander.” It continues: “If [the shura] continue on this path, we will leave the movement. We will fight until the end against foreign troops . . . but the decision of the leadership council in Quetta was a wrong decision. They want to appoint Uzbek or Chechens instead of a Taliban commander. And Mullah Mohammad Omar, you should know that Pashtuns never want to be slaves. We will not accept a Chechen or Uzbek commander. It is still unclear whether Uzbeks and Chechens are good Muslims. Death is better than accepting their commands.”51 The letters were, of course, condemned as American “propaganda” and “disinformation” by Taliban representatives. And yet in August 2007, Mullah Omar, in a rare message, called on all Afghans to put aside their differences and join his movement’s military campaign against Western troops in Afghanistan. He argues that “the enemies of the religion of Islam . . . have launched satanic propagandas under the slogans of democracy and freedom,” and pleads, “We have to put aside all our internal, regional, and linguistic differences and get united against the enemy.”52 All of this is to say that Mohammad’s elimination occurred within a months-long Taliban hangover that originated with Dadullah’s death. Other regional events might also have influenced the Mohammad casestudy findings. On July 10, for instance, Pakistani forces stormed the Red Mosque (or Lal Masjid) in Islamabad, crushing a standoff with Islamist militants (and students). The raid reportedly killed hundreds of civilians and militants, effectively putting an end to the Taliban-Pakistani truce that had been established in the spring. Al Qaeda’s then No. 2, Ayman al-Zawahiri, was quick to denounce the raid with a message, titled The Aggression Against Lal Masjid, calling for a Pakistani jihad. Catastrophic suicide attacks against Pakistanis occurred soon after. Indeed, the raid on the Red Mosque unleashed a wave of terrorism in the country, some of which was orga nized by elements allied and associated with Afghanistan’s Taliban movement. It is possible that Pakistani militants allied with the Afghan Taliban shifted their efforts following the raid in Islamabad, draining supplies and personnel that

156

Chapter 5

would have otherwise reached Afghanistan. If so, new efforts to attack Pakistanis might have reduced the amount of violence recorded in Afghanistan over the latter half of the Mohammad case study. And yet, in the weeks following Mohammad’s elimination, more than a thousand insurgents were reported to have crossed into Afghanistan’s Kunar, Laghman, Nuristan, Kandahar, and Wardak provinces from Pakistan. Though more than a dozen ISAF/ANA raids were reported to have incapacitated roughly four hundred insurgents in that period, several hundred fresh fighters would remain available to plan and carry out acts of violence in Afghanistan in the weeks under review. That Mohammad’s elimination took place while hundreds of fresh Taliban and foreign fighters poured into Afghanistan might help explain other findings revealed in the case. Recall that on nearly all counts, professionalism and terrorist success decreased while target selection seemed to favor NATO and Afghan security forces. It is possible that the well-rested Taliban fighters from Pakistan were eager to attack ISAF and Afghan forces despite the loss of some key Taliban leaders.

Mullah Abdul Matin

In a dramatic precision attack, Special Forces attached to Britain’s SBS intercepted regional Taliban subcommander Mullah Abdul Matin, along with his lieutenant, Mullah Karim Agha, as they drove their motorcycles across the desert near Gereshk, Helmand, on the night of February 18, 2008. Using surveillance intelligence and rapid transport, an SBS team stealthily deployed ahead of Matin’s convoy. Once in position, the SBS ambushed Matin with small arms fire as he drove by, killing him and Agha (along with a bodyguard). Matin had been sought by Helmand-based coalition forces for more than a year, his death or capture considered a top priority. He was thought to have been the mastermind of a number of suicide bombings targeting British soldiers and was directly implicated in the deaths of two coalition servicemen (Royal Marine Gary Wright in October 2006 and Sargeant Dave Wilkinson in July 2007), the injury of more than a dozen other NATO soldiers, and the death and injury of a number of Afghan security personnel. He was also responsible for other acts of terrorism and had organized a number of kidnappings. British Forces spokesperson, Lieutenant Colonel Simon Millar reported that “Matin’s been a priority target for some time. He commanded fighters and reported directly to the highest Taliban levels of

Targeting the Taliban

157

command.” The raid in February had been the second attempt on Matin’s life; he was almost captured by British forces in December 2007 in Musa Qala, Helmand.53 Matin was not a high-ranking Taliban commander but rather a regionallevel field commander. Under the Taliban regime he had been appointed the Musa Qala district governor and it is assumed that during the Taliban’s insurgency he played a leading role in southern Afghanistan and, in particular, in Helmand province. His death, ISAF noted, was a “severe setback” for the Taliban.54 Because Matin was a regional rather than a national leader, his death was not as significant to the organization as that of Dadullah or Mohammad; his removal, we can assume, would have had an effect on Taliban behavior but to a lesser extent and degree than Dadullah’s or Mohammad’s. The Matin case study corresponds to Matin’s leadership role in southern Afghanistan (in particular, in Nimruz, Helmand, Kandahar, Zabul, and Uruzgan provinces) over a period of one month, two weeks before and two weeks after the strike. OVERALL VIOLENCE

Matin’s elimination did not have the same effect on the type and nature of the Taliban violence in southern Afghanistan as Dadullah’s and Mohammad’s eliminations had for all of Afghanistan. In general, Matin’s targeted killing both corroborates and contradicts theoretical expectations. For instance, though attempted suicide bombings dropped by half following his elimination, SA/R fire doubled while IED attacks remained unchanged. Likewise, assassination attempts decreased following the Matin raid, while abductions increased (Figure 5.11). Furthermore, even the precipitous drop in suicide bombings following Matin’s targeted killing can be partially explained by other factors. In the weeks before his elimination, for example, two Taliban suicide bombers killed more than 160 Afghan civilians. Both attacks were successful assassination attempts carried out in heavily populated civilian gatherings. The public condemnation of these attacks was ferocious. Civilian bystanders suffered the brunt of these attacks and voiced their disgust with the Taliban. It is conceivable that Taliban forces decided to stem the number and intensity of their suicide bombings in the weeks following these assassinations in order to not further inflame public sentiment. Although this period of relative calm coincided with Matin’s elimination, it may not have been a result of it. Then again, as I suggest in the coming section, other interpretations are possible.

158

Chapter 5

Total Attacks

30 25

24 21 21

20 15

12 8

10

5

4

5

3

2

5

ns Ab

du

ct io

at io As sa ss in

Bo m bi de Su

ici

n

s ng

/R SA

IE

D

0

Before TK

After TK

Figure 5.11. Acts of terrorism (total)—Matin. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “SA/R” stands for “small arms and rocket fire.”

Taliban Professionalism

As in the Dadullah case, increases in overall activity following Matin’s elimination were met with decreases in professionalism. It is important to note that the trend is especially evident when measuring suicide attacks. Of the eight suicide bombings carried out prior to Matin’s death, half failed (Figure 5.12).55 But after the Matin raid, all suicide attacks attempted in southern Afghanistan failed: One vehicle-borne suicide bomber prematurely detonated his payload in Kandahar (February 19); a suicide bombing cell was captured (with seven arrests and three car bombs destroyed) in Zabul province (March 2); a potential bomber was arrested by the ANA after his family reported his activities to authorities (March 3); and another fifteen suicide vests (along with IED-making materials) were found and destroyed in Helmand (March 4). Following Matin’s elimination, no suicide attack took place in all of southern Afghanistan (except for the premature suicide detonation in Kandahar, which occurred the morning after the targeted killing and had presumably been organized before the Matin raid). It is possible that Matin’s elimination scrambled Taliban ranks and diminished the Taliban’s ability to organize suicide attacks in the region. After all, Matin was a known suicidebomb expert with a history of a long list of bloody successes. He also was politically well-connected and was considered a powerful regional leader. He successfully oversaw an effective suicide-bombing network (and cam-

Targeting the Taliban

159

9 8

Attempts

7 6 5 4

1 2 1 1

3 2

4

2

Before TK

After TK

1

1

0

Successful

Stopped

Arrested

Premature

Figure 5.12. Suicide-bombing professionalism—Matin. “TK” stands for “targeted killing.”

paign) in southern Afghanistan, such that in his absence a leadership void was created. That a number of suicide vests and car bombs were located and destroyed and a number of bombers arrested following the Matin raid might indicate that his replacements were less able than he was.56 Furthermore, like suicide attacks, data on small arms and rocket fire during the Matin case reveal a decline in professionalism (Table 5.10). Though SA/R attacks doubled following Matin’s killing, overall each attack became far less deadly. Before the Matin raid, roughly four Taliban died in SA/R fire for every one security personnel; after the targeted killing, the ratio dropped to roughly sixteen Taliban killed for every one coalition or Afghan officer. During the Matin case study, however, IED professionalism remained unchanged (see Table 5.11 and Figure 5.13). During both halves of the study, a total of twenty-one IED attacks were reported. Of these, seven failed before the targeted raid and six after the raid. Most of these failures were a result of IEDs being uncovered by vigilant coalition and Afghan forces (on occasion, following tips offered by local Afghan nationals). There were two notable exceptions. On February 5, before Matin’s elimination, an eight-member IED cell was preemptively attacked by an ISAF air strike as its members were preparing to conceal a roadside bomb in Kandahar province. Three days later, also in Kandahar, another four insurgents, including a known “IED expert,” were reportedly killed when their IED exploded prematurely. The destruction of these IED cells may have affected IED success rates in the period after Matin’s death. Indeed, IED success rates dipped (if only

Table 5.10. SA/R Kill Ratios—Matin Period Small arms and rocket fire

Before TK

After TK

12

24

Total Success ratesa

SA/R deaths

ISAF/PRT Insurgents ANA/ANP Insurgents GoA Insurgents

0 25 6 1 0 1

0 12 3 7 0 30b

Note: “TK” stands for “targeted killing.” The remaining acronyms can be found in the body of this chapter. a The figures included here relate to the cumulative results (e.g., number of deaths) of specific SA/R engagements between Taliban insurgents and their respective targets (e.g., ISAF/PRT, ANA/ANP, GoA). b Taliban forces attacked a GoA poppy eradication team in Helmand on February 28, 2008. During the battle, a number of ANA and ANP reinforcements arrived, and in the ensuing three-hour engagement, thirty Taliban were reported killed, including a regional subcommander, Mullah Naqeebullah. “Taliban Die in Fighting in Southern Afghanistan,” Asian Political News, March 2, 2008.

Table 5.11. IED Failure—Matin Period IED attempts Total Failed Uncovered Premature Preempted

Before TK

After TK

21

21a

7 5 1 1

6 5 1b 0

Note: “TK” stands for “targeted killing.” a Th is includes one large car bomb that was initially reported as a suicide attack but, on further investigation, was later classified as a remote-controlled IED (Kandahar, February 19, 2008). b Four Taliban were killed in a premature detonation of an IED in Helmand province.

Targeting the Taliban 18

17

16

Number of Events

14

161

14

15

14

13

12 10 8

8 6

7

6

7

5

4 2 0 Before TK IED Detonations Civilian Deaths

After TK Military Deaths Civilian Injuries

Military Injuries

Figure 5.13. IED professionalism—Matin. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.”

slightly) following Matin’s killing. Before the raid, fourteen IED detonations were recorded, killing nineteen individuals—a little over one death for every IED. After the Matin raid, fi fteen IED detonations were recorded, killing twelve. A similar decrease is evident concerning the specific number of security personnel killed by IEDs: Before Matin’s targeted killing, roughly 70 percent of those killed in IED attacks were ISAF, ANA, or ANP officers; afterward, only 40 percent of casualties were security personnel. TYPE AND NATURE OF ATTACK

The Matin case does not shed much light on suicide-bomber target selection; only one bomber detonated his explosive vest post-Matin and he did so prematurely. His intended target is unknown. Nonetheless, of the four detonations before the targeted elimination, two were directed against ISAF and two were assassination attempts against Afghan government officials and their allies. Although the attacks on ISAF had little effect, the attacks on the GoA killed and injured hundreds of Afghans. In the first attack, a bomber detonated his vest at the entrance of a mosque in Lashkar Gah, Helmand, on January 31, killing the province’s deputy governor, Haji Pir Mohammed, as he approached the building for evening prayers. Several other Afghans were killed and injured. That the attack took place at a mosque enraged many

162

Chapter 5

Afghans. In the second case, a devastating suicide attack took place at a dogfight, a popular spectator sport, in Kandahar on February 17. Reports suggest the attack was an assassination attempt against Abdul Hakim Jan, a former Taliban commander turned GoA ally and recently appointed ANP chief. Eyewitnesses report that the bomber weaved his way through the crowd, getting to within meters of Jan before detonating his device. More than 180 individuals were killed or injured, including Jan (whose body was never found). It was, for a time, considered the deadliest suicide attack in Afghanistan’s history. The bombing received widespread and vociferous condemnation, both within Afghanistan and abroad. Though the Taliban denied responsibility for the attack—something the orga nization does often enough when too many Afghan civilians die in its attacks—it was nonetheless widely suspected of having orga nized Jan’s assassination. After abandoning its cause, the Taliban considered Jan a traitor and Karzai puppet. By removing him, they not only got even but also sent an undeniable coercive message to other would-be defectors: Switching sides was a costly and an unforgivable offense. No significant shift in target selection was recorded with regard to SA/R attacks. Both ISAF and ANA forces were targeted evenly, at roughly 25 percent, before and after Matin’s death. Likewise, the ANP was the target of more than half of all SA/R engagements during both time frames. Again, soft-target selection remained constant at roughly 17 percent (Figure 5.14).57 The only notable shift in target selection was the targeting of three transport trucks driven by local Afghans. Of suspected IED targets, a shift toward targeting ANP and away from targeting ISAF and ANA forces is evident. ISAF and ANA were targeted by every third IED before Matin’s death but only by every eighth IED afterward. For their part, Afghan police were targeted by one-quarter of all IEDs before Matin’s elimination and by half of all IEDs afterward. The assassinations and abductions recorded in the Matin case offer inconclusive findings; assassinations dropped while abductions rose in the reporting period (see Tables 5.12 and 5.13). As mentioned, two assassination attempts using suicide bombers killed two GoA allies. The other three targets were also GoA-affi liated individuals, including Khalid, the governor of Kandahar, who was targeted on two separate occasions, and Abdul Salaam, a district chief for Helmand province. Of these three attempts, two were conducted with remote-controlled IEDs. That four of the five assassination attempts were carried out with complex weapons, as opposed to SA/R fire, is

Targeting the Taliban

163 IED Targets (After TK)

IED Targets (Before TK) ISAF, 1 GoA, 2

ISAF, 1 GoA, 2

ANA, 2

ANP, 4

ANP, 2

SA/R Targets (Before TK) GoA, 2

SA/R Targets (After TK) Civilian, 3

ISAF, 3

ISAF, 4

GoA, 1 ANA, 2

ANP, 12

ANP, 7

Figure 5.14. IED and SA/R target selection—Matin. “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “SA/R” stands for “small arms and rocket fire.” The remaining acronyms can be found in the body of this chapter. Table 5.12. Assassinations—Matin Period Assassinations

Before TK

After TK

Total

5

2

Targets Government of Afghanistan Success rate (killed)

5 2

2 0

Note: “TK” stands for “targeted killing.”

a change in tactic from the Dadullah case. In like fashion, both assassination attempts after Matin’s killing—one against Governor Khalid and another against a GoA official—were carried out with IED detonations. As for abductions, the majority of those targeted were civilians, with a few ANP abducted following SA/R exchanges. Before the targeted strike on Matin, an Afghan pharmacist working with a local NGO and a pro-Karzai mullah

164

Chapter 5

Table 5.13. Abductions—Matin Period Abductions

Before TK

After TK

Total

3

5

Targets Civilian (Afghan and international) ANA/ANP personnel

2 2

6 1

Note: “TK” stands for “targeted killing.” The remaining acronyms can be found in the body of this chapter.

were abducted. After the targeted raid, one international civilian, American Cyd Mizell, and her Afghan escort Abdul Hadi were abducted. Both worked with the Asian Rural Life Development Foundation, an international NGO, and were reportedly killed in February 2008. The other abductees were truck drivers. The Matin case study offers a few insights not necessarily evident in either the Dadullah or Mohammad cases. First, Matin’s elimination did not result in a significant reduction in the overall level of terrorism in southern Afghanistan. Like the Dadullah case, the number of SA/R attacks actually increased following the elimination, while the number of IED attacks remained unchanged. Only suicide bombings dropped in number after Matin’s elimination, a finding shared by both the Dadullah and Mohammad cases. Second, professionalism, in most categories, did drop. Most important, suicide-bombing successes fell to zero. Likewise, after Matin’s killing, almost two weeks went by before suicide missions were again carried out. That Matin was a suicide-bombing expert may have put a damper on the Taliban’s ability to carry out such attacks in southern Afghanistan following his death. The IED professionalism also dropped and SA/R attacks became far less deadly after Matin’s death, even though the frequency of those attacks doubled. Shifts in target selection, however, were inconclusive. Little data were available on suicide-bombing targets, though suspected IED target selection did shift toward softer security forces. The Taliban’s behavior in the period after Matin’s killing changed in other ways not evident when relying on the data alone. For instance, in the weeks following the Matin raid, Taliban forces spent a substantial amount

Targeting the Taliban

165

of time and energy sabotaging cell-phone towers in southern Afghanistan. On four separate occasions, Taliban operatives destroyed towers and related infrastructure, believing that the technology helped NATO, the United States, and the ANA identify, track, and target their leaders in precision strikes. Communications experts have argued that the demolitions had little effect on the ability of the United States to locate insurgent leaders because it does so by relying on satellite and other high-tech sources of communications intelligence that have little to do with intact telecommunication towers. What the sabotages do illustrate, however, is a clear attempt on the part of the Taliban to better protect its leadership. Around the time of the Dadullah and Mohammad killings, the Taliban tried to protect its leadership by capturing Afghan “spies” and executing them in public (and on video), in an effort to deter other Afghans from working with Kabul or NATO. However, in early February 2008, Mullah Omar ordered his fighters to stop the beheadings because after sixteen months of doing so the Taliban’s brutality had cost it some support among Afghans. Instead, Omar instructed his followers to simply shoot suspected spies. The Taliban charm offensive continued: Suicide bombers were to avoid targeting civilians; schools were not to be burned to the ground; and militants were banned from cutting off the ears, lips, or tongues of their captives. It was reported that Mullah Omar even removed some of his more brutal commanders from their posts.58 Of note, during the weeks immediately before and after the Matin killing—which covers a period of time beginning only weeks after Mullah Omar’s injunction—no beheadings were reported at all. In addition, the Taliban shifted its strategy from deterring Afghans from joining U.S. efforts to destroying the imagined technological sources that they believed were threatening the survival of its leadership. The Taliban began destroying the telecommunication infrastructure after it failed to stem the wave of targeted killings by way of public intimidation and coercion. Destroying communications towers was a second (albeit even less effective) option and was a behavior shift directly resulting from the campaign of targeted killings. Furthermore, as in the Mohammad case, a number of potentially important sociopolitical developments took place and overlapped the Matin study period. Before Matin’s targeted elimination, for instance, ISAF and the ANA conducted a number of raids in the south of the country, which resulted in the incapacitation of roughly forty insurgents (including two “low-level”

166

Chapter 5

leaders). And regionally, secret negotiations between the Pakistani government and antigovernment groups—namely, with Mehsud’s Tehrik-e-Taliban, which was blamed for a string of deadly attacks, including the 2007 assassination of former Pakistani prime minister Benazir Bhutto—produced a ceasefire. A truce in Pakistan risked influencing levels of violence in Afghanistan by freeing Taliban resources for the Afghan campaign. Finally, in the two weeks prior to Matin’s killing, two other high-ranking Taliban and al Qaeda leaders were incapacitated. Mullah Mansoor Dadullah, Mullah Dadullah’s brother, was reportedly captured by Pakistani troops on February 11. Though Mansoor had nominally replaced his brother as Taliban commander, his importance at the time of his capture was unclear. A number of analysts suggest that Mullah Omar was unhappy with Mansoor after he took over his brother’s duties and subsequently fired him from his post.59 The Taliban statement surrounding Mansoor’s sacking suggested that he had “not obey[ed] the rules of the Islamic emirate [Afghanistan] and violate[d] it. Therefore it was decided not to appoint any post in the emirate to him.”60 Mansoor’s demotion is an important development for studying the effects of targeted killings in counterterrorism. It signaled, first and foremost, that the Taliban was having difficulty finding a suitable commander to replace Dadullah, as theories of targeted killings suggest. It also revealed a potential lack of morale (Mansoor was not willing to lead as his brother had); a lack of depth in the Taliban’s leadership pool (elite positions were offered on the basis of familial ties and tribal clanship); and the Taliban’s difficulty in attracting and retaining the best quality leaders (Mansoor, a weak leader, was nonetheless the best option). Mansoor’s sacking was also an embarrassment for Mullah Omar and illustrates his own weaknesses in running the Taliban’s Afghan insurgency at arm’s length from the safety of his hideout. Alongside Mansoor’s capture, another important elimination took place prior to the Matin time frame: Al Qaeda’s Abu Laith al-Libi—a top al Qaeda leader in the region—was killed in a precision, CIA-operated drone strike in Pakistan on (or around) January 29, 2008. Al-Libi was active in orga nizing attacks in Afghanistan and Pakistan, and his death was a significant loss for al Qaeda’s regional ambitions. (It was also reported, though ISAF could not corroborate Afghan statements, that Mullah Abdul Barri, an important leader in Helmand, was also killed.)61 As in the Mohammad case study, the loss of important senior Taliban and al Qaeda leaders around

Targeting the Taliban

167

the time of Matin’s targeted killing might have affected the case study’s overall findings.

Aggregate Observations In a 1989 New York Times editorial, American political scientists David Newman and Bruce Bueno de Mesquita suggested that targeted killings can have a deterrent and cumulative effect on an adversary’s behavior. “A reputation for preparedness to retaliate against aggression,” they write, “can insure peaceful relations even among the most hawkish adversaries. Paradoxically, the failure to build such a tough reputation can lead to the very acts of hostility [actors] wish to avoid.” 62 In the realm of targeted killings, the authors attest, a similar conclusion is borne out. Signaling a willingness and capability to track and kill an adversary’s leadership, and then proving the ability to do so with often-repeated strikes, can have a deterrent effect on the adversary’s behavior. In Afghanistan, ISAF as well as American and Afghan security forces have shown both a willingness to target their adversary’s leadership and an ability to do so effectively. For this reason, the prerequisites for deterrence— identified earlier as resolutely communicating a clear, costly, and credible retaliatory threat—are not only evident but also satisfied. Constructing and publicizing “Most Wanted” campaigns and offering reward money for useful information and intelligence clearly communicates the costs associated with facilitating terrorism. Tracking and capturing or killing identified leaders with precision attacks further communicates capability and credibility and shows resolve to carry out threats. In Afghanistan, the extensive list of captured and killed terrorist and insurgent leaders illustrates that coalition and Afghan security forces have the ability to do as they threaten. Militant leaders in Afghanistan know they are marked men. Indeed, the current crop of Taliban leaders are far less willing to reveal their faces to journalists— let alone conduct in-depth and personal interviews, like Dadullah did—for fear of being easily identified and targeted by security forces. The days of the Taliban cheerfully mugging for the camera are over. The aggregate data from all four cases (the three targeted killings presented in the body of this chapter along with the fourth case of Mullah Baluch), are informative. First, they reveal that overall violence generally increased

168

Chapter 5 250

Number of Attacks

207 200

180

150 109

116

100 43

50

29

0 IED

SA/R

Suicide Bombings

Type of Attack Before TK

After TK

Figure 5.15. Total attacks (aggregate). “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device”; “SA/R” stands for “small arms and rocket fire.”

following the targeted eliminations (Figure 5.15). This is an unanticipated development.63 The theoretical literature on targeted killings suggests that eliminations should result in a general diminishment of violence, but empirical studies repeatedly find otherwise. For instance, in their analysis of Israeli targeted killings (2000–4), Hafez and Hatfield conclude that “targeted assassinations have no significant impact on rates of Palestinian violence.”64 Jordan corroborates their finds concerning Hamas. And she writes of Colombia’s FARC: “[T]he casualty rate increased by over 100 percent after having experienced seven years of leadership decapitation.” 65 On the surface, this project, along with the Hafez and Hatfield, Jordan, and other related studies, contradicts theoretical expectations: Targeted killings do not appear to diminish overall levels of terrorism and political violence. Yet, a closer examination of the Afghan data does corroborate the literature’s most basic theoretical principle: Targeted killings influence the type of violence terrorists are capable of planning and force them to conduct less preferred forms of activity. That detail is lost if we focus only on overall levels of violence. We need to dig deeper. Violent, nonstate organizations have behavioral preferences. The Taliban is no exception. The type of violence it orchestrates rests as much on the effect it is trying to have on NATO and Afghans alike as it does on its capacity and capability to plan these acts. As discussed, suicide attacks are the Taliban’s preferred tactic—they are an effective form of violence; provide

Targeting the Taliban

169

the greatest consequence (both in kill ratios and psychological effect); can be directed against hard targets; are difficult to detect, thwart, and mitigate; and have a proven track record of killing coalition and Afghan soldiers. But suicide bombings are also the most sophisticated type of violence to plan and the most difficult to organize effectively; in addition, they take a considerable amount of time, energy, and expertise to mount successfully. Improvised explosive devices are the Taliban’s second-most-preferred tactic—IEDs have proven deadly against the ANP and other lightly armored NATO and ANA personnel carriers, they are cheaply constructed, and they provide a deadly concentrated explosive blast. They are also less sophisticated than suicide bombs and may be easier to organize effectively. Another benefit is that those planting IEDs can do so again, whereas suicide operatives usually only get one chance. The downside of IEDs, however, is that they offer the Taliban less control. They cannot be consistently directed against particular targets, they can be detected and diff used more easily than suicide bombers, they can be mitigated with proper armor, and they are often unintentionally triggered by civilians. Small arms and rocket fire is perhaps the Taliban’s least preferred tactic—it is most effective when used against soft targets, against Afghan and international officials, against lightly armed ANP forces, and in complex ambushes that involve other forms of explosive devices. However, SA/R attacks against security forces can be easily mitigated and usually result in a disproportionately high rate of Taliban casualties. No matter how well armed, Taliban forces will always be outgunned by NATO. Likewise, Taliban SA/R attacks are usually successfully repelled (and often fail to achieve their tactical objective) and the heavy concentration of militants in one location can be easily attacked with aerial support. Furthermore, Taliban rocket fire is crude, uncontrolled, and ineffective. In sum, SA/R attacks are the least sophisticated type of violence and the easiest to organize, but they provide the worst results. With these Taliban preferences in mind, the aggregate data on overall violence are illustrative. After the targeted killings, suicide attacks dropped by almost 35 percent—this is in keeping with the degree of difficulty, amount of time and expertise, and level of leadership that is required to coordinate effective suicide bombings. It is possible that the decrease in suicide attacks spurred a concomitant rise in less sophisticated forms of violence, with IED attacks increasing by 6 percent and SA/R attacks by 15 percent following the four targeted eliminations. As leaders and terrorist facilitators were killed, the Taliban shifted its behavior toward less sophisticated forms of violence

170

Chapter 5

Number of Attacks

that took less energy and time to organize. This is a shift in behavior specifically outlined in the theory: As terrorist organizations succumb to the effects of a protracted campaign of targeted killings, their overall ability to operate at a high level of sophistication, day in and day out, decreases and the selection and use of less formidable forms of violence increases. Overall violence, then, is only part of the empirical story and much more nuance is needed to properly assess the coercive effects of targeted killings. The aggregate data highlight the way targeted killings relate to Taliban professionalism. For the two most sophisticated forms of terrorism (suicide bombings and IED attacks), there was a significant decrease in professionalism following the targeted killings. Before the eliminations, there were seven 100 90 80 70 60 50 40 30 20 10 0

88 77

39 25 18

15

Suicide Bombings Before TK

21

14

Suicide Bombings After TK Success

IED Before TK

IED After TK

Failure

Figure 5.16. Total professionalism (aggregate). “TK” stands for “targeted killing”; “IED” stands for “improvised explosive device.” Known Suicide-Bombing Targets (Before TK)

Known Suicide-Bombing Targets (After TK)

Soft, 6

Soft, 6

Hard, 24

Hard, 13

Figure 5.17. Suicide-bombing target selection (aggregate). “TK” stands for “targeted killing.”

Targeting the Taliban

171

more successful suicide attacks than failures; after the elimination, one suicide attack failed for every two attempted. Changes in IED professionalism are even more pronounced. Before the targeted killings, only one-fift h of all IED attacks failed; after the targeted killings, one-third of all attempted IED attacks failed (Figure 5.16). The aggregate data also inform the effects that targeted killings had on terrorist target selection. Suicide bombers, for instance, were directed against softer targets far more often in the weeks following the targeted strikes (Figure 5.17). In sum, as militant leaders are eliminated, the expectation is that their surviving forces attack easier targets in order to exact a potentially greater payoff for less risk. In Afghanistan, targeted killings became a prominent counterterrorism and counterinsurgency tactic. Pinpoint eliminations were considered the coercive half of a two-pronged strategy. As former ISAF spokesperson for Kandahar David Marsh revealed, “There’s a mixture going on of bringing out the less committed [Taliban]” with various amnesty, development, and make-work programs, “and getting rid of . . . tier one [leaders].”66 The strategy is generally supported by the government of Afghanistan. General Mohammad Zahir Azimi, GoA defense spokesperson, reiterated that “we can’t kill all the Taliban,” but that Afghan and international forces could “get rid of the Taliban leadership and reconcile the rest of them who are just ordinary people that have joined the Taliban under certain situations.”67 My empirical findings offer a comparative analysis of the coercive half of the general strategy. In the following chapter, a number of patterns, evident from the data analysis, are discussed, as they relate to and inform the theory and practice of deterring terrorism and political violence.

CHAPTER 6

Moving Ahead with Deterrence Theory

Developing and testing theories of deterring terrorism is a promising academic endeavor with very real practical applications. For decades, deterrence theory informed foreign and military policies around the world. Its tenets offered political and military leaders a guide for how to fight current conflicts within limits, while planning ahead to avoid the next war altogether. That changed with the fall of the Iron Curtain; deterrence soon became a theory in search of meaningful occupation. In the post–Cold War era, a few scholars applied deterrence to novel contemporary threats, like rogue states and WMD proliferaters, domestic and civil wars, and humanitarian crises and genocide, but it seemed that the theory had peaked within the study of international relations. And on September 11, 2001, deterrence theory absorbed another near-fatal blow. Although states had encountered terrorism before, al Qaeda’s attack on the United States was something new, more brutal and shocking. The old terrorism was a menace; the new terrorism appeared existentially frightening, irrational, and unbelievable. Walter Laqueur, a steady hand within the annals of terrorism research, explains: “Terrorism has been on the international agenda for a long time, but . . . it was relegated to a lowly place. From time to time, following some spectacular attack, terrorism would figure prominently in the media for a few days. . . . But when calm returned the issue would be forgotten, for there seemed to be no particular urgency to deal with it. There were always some very important domestic and foreign issues that would take precedence. . . . This has now changed.”1 Al Qaeda’s 2001 attack on the United States spurred renewed interest in and placed greater importance on the study of terrorism and counterterrorism. There was understandable urgency, too. That suicidal terrorists had rather easily turned planes into guided missiles suggested that

Moving Ahead with Deterrence Theory

173

other cruelly ingenious, mass-casualty attacks were on the horizon. Recall that the anthrax (Bacillus anthracis) attacks began on September 18, just a week after the Twin Towers were destroyed. By mid-October, when the first anthrax deaths were confirmed, there was open (though ultimately erroneous) speculation that al Qaeda had fired its second shot at the United States: biological terrorism.2 Against this backdrop, the contours of the emerging struggle against terrorism were taking shape and deterrence was not part of the equation. “Evil has escaped the prison house of deterrence,” Canadian politician and Harvard scholar Michael Ignatieff wrote. “The logic of deterrence that once kept state violence in some kind of check has no traction with loners and the cult leaders of global terrorism.”3 That skepticism was widespread . . . but it was also premature. Now, more than a decade since 9/11, we are in a better position to assess whether and how deterrence theory might be applied against nonstate militants, and actually work to keep ourselves safe. There are many and obvious differences between applying deterrence against the former Soviet Union or another state adversary and applying it against terrorist and militant organizations like al Qaeda; some of these differences are large.4 But they appear increasingly manageable, too. Theories of deterrence can be expanded and adapted in ways that incorporate unique aspects inherent to violent nonstate adversaries. And terrorists themselves are part of an entire constellation of different actors, each as important to the next in orchestrating acts of political violence. Deterring, compelling, influencing, or coercing some of these actors will help us manage, contain, and counter the terrorist threat. This book suggests that it is a mistake to assume that terrorism or terrorists cannot be deterred. Post-9/11 deterrence skepticism may have been a predictable reaction to the nature and scope of al Qaeda’s challenge, but the skepticism was also largely misplaced. Targeting what terrorists value, desire, and believe can influence the type and ferocity of the violence they organize. Doing so will require developing innovative strategies that combine punishment with nonpunishment tactics and applying them to the spectrum of state, community, and individual facilitators of political violence. Success in countering terrorism involves undermining the motivation that informs an organization’s willingness to use violence and its supporting community’s enthusiasm to facilitate it. This requires better understanding the religious, cultural, and ideological factors and local conditions that foster political violence and tailoring our coercive tactics, strategies, and policies to influence and shape behavior. In Chapters 2 and 3, I offer a wide range of options

174

Chapter 6

for applying the logic of deterrence theory to terrorism: Punishment, denial, and delegitimization each hold promise. Drilling down on theories of punishment, I then select one coercive process—targeted killings—to evaluate empirically. Targeted killings represent but one of a variety of different costs that can be used to threaten and manipulate militant behavior. With this twofold exploration of deterrence and targeted killings, this book offers lessons for both the theory of deterring terrorism and the practice of targeted killings in counterterrorism and counterinsurgency. In this concluding chapter, a variety of theoretical, practical, and empirical lessons from both sides of the equation are discussed in greater detail.

Targeted Killings as Coercion: Reinterpreting the Data The first lesson stems from the empirical analysis of Taliban targeted killings: It is clear that overall levels of terrorism and political violence are not, on their own, robust measurements of deterrence success or failure. Nor do levels of political violence tell us much about whether targeted killings influence militant behavior more broadly. In each case, some form of political violence increased in the period after each elimination. This was most evident in the Dadullah case, in which overall acts of terrorism rose precipitously for all types of violence (except suicide bombings) after Dadullah’s death. Arguably, moving beyond an assessment of only the most easily accessible measurement of a group’s behavior (that is, overall levels of violence) is critical. Measuring deterrence and coercion—or, for that matter, measuring counterterrorism successes more broadly—solely on the amount of violence an organization is able to orchestrate, is to badly misinterpret the available data. Flawed policy is a concomitant risk. It is not enough to base an assessment of security policy and counterterrorism strategy by simply pointing to increasing levels of militant violence, as some academics, journalists, and politicians are apt to do when it comes to the Afghan conflict and to counterterrorism more broadly. Any increase in terrorism needs to be properly parsed and contextualized. A more nuanced way to measure the effect of counterterrorism and of targeted killings in Afghanistan (and elsewhere) is to assess the type and nature of the violence itself and to track how and why different types of violence change over time. A second lesson for properly exploring deterrence by punishment in counterterrorism is offered by gauging changes to Taliban professionalism.

Moving Ahead with Deterrence Theory

175

Doing so also augments our analysis of overall terrorism trends. For instance and as previously noted, in Dadullah’s case, IED attempts increased by almost 50 percent following Dadullah’s removal, but IED success rates also dropped by more than 10 percentage points. In Matin’s case, too, suicidebombing rates fell overall, and the effectiveness of the suicide bombings decreased to zero. In both cases, superficial measurements that rely only on overall levels of violence were cast aside by going beyond the easiest of observations. So although the overall number of attacks may have gone up following Dadullah’s elimination, more bombs failed to go off, too. That matters. It is critical to note here that changes in levels of militant success and professionalism can help reveal a deteriorating organization that is having trouble attracting, retaining, and replacing fallen leaders and facilitators. Deprofessionalization speaks to the literature on targeted killings—as a group’s leaders are killed, less capable individuals move up the ranks. Difficulties in recruitment and retention relate to deterrence by punishment—as a group faces a focused counterterrorism campaign, would-be recruits begin to have second thoughts about joining, moving up the food chain, or sticking around. Even then, though, measuring professionalism by itself (or in relation to overall levels of violence alone) says nothing as to why militant success rates rise or fall over time. To do that effectively, we need to drill down even deeper; we need to have a closer look at militant target selection. Thus, a third lesson born of the analysis is the necessity and importance of evaluating what type of victim is targeted with what particular form of violence. In general, the easier a target can be attacked, the more success a militant group is likely to have: Choosing to target unprotected civilians over heavily armed military personnel will likely result in (superficial) improvements to a group’s observable success rate. Kill rates will increase and the uptick will be evident. For example, during the Mohammad case study the Taliban conducted many more assassinations after Mohammad’s elimination than before, while after Dadullah’s death, suicide attacks targeted soft targets more often than before his elimination. Both trends show a dedicated shift in militant behavior: Soft targets replaced hard ones. These trends necessarily affect the data on terrorism success rates and professionalism but should not be taken on their own as indicators of a healthy militant orga nization scoring multiple victories. In fact, the opposite may be true. Militant target selection that favors softer targets not only reveals a potentially demoralized organization but may unintentionally lead to the group’s eventual demise if and when civilians and supportive constituents face the brunt of their violent actions and decide to

176

Chapter 6

push back (as communities have in Iraq, Mali, Pakistan, and elsewhere, for instance). This is especially relevant in the case of Afghanistan, where the Taliban must retain popular sympathies if it hopes to ever again govern the country. In sum, the targeted killings explored in this book resulted in a short-term change in the Taliban’s effectiveness, professionalism, and motivation. These findings inform both the literature on targeted killings (and counter-capability) and on deterrence theory (and counter-motivation). That targeted killings are today being used by several countries involved in ongoing conflicts with a variety of nonstate adversaries suggests that the next step is to broaden the empirical base in order to provide international and cross-case comparisons of the deterrent effects targeted killings have in distinct conflict environments (as Cronin, Jordan, Price, Johnston, and others have done in studying targeted killings and militant group “mortality” rates).5 Doing so will generalize this study’s findings, further refine existing research on the coercive effects of targeted killings, and offer an important addition to the nascent literature on deterring terrorism by punishment. A related caveat is this: Although the elimination of Taliban leaders seems to have affected the behavior of the organization (within the time frame presented here), other factors much less closely related to specific U.S. operations also influence Taliban behavior. At times, threats of punishment and military coercion will be overshadowed by other influencing factors and processes. For example, as noted, Afghan support for the Taliban’s cause—or, for that matter, local support for American military and political objectives—may have an important influence on the Taliban’s decision to use certain forms of violence. Hearts and minds really do matter, and they cut both ways. Mullah Omar is just as fearful of alienating Afghan civilians as are NATO, the UN, and the United States. When innocent Afghans die in coalition strikes and operations, anger flares against the West. We take that as a given. But the very same logic holds in the case of insurgents: When Taliban suicide or IED attacks disproportionately kill civilians (purposefully or accidentally), popular resentment and anger is directed against it. The horrendous suicide bombings that devastated Afghan civilians during the period covered in the Matin case study are a working example. All of this suggests that other factors in addition to the ongoing campaign of targeted killings influence Taliban decisions to rely on particular forms of violence and targets over the long haul. Likewise, although this empirical study suggests that targeted killings have an immediate influence on militant behavior, the longer-term coercive

Moving Ahead with Deterrence Theory

177

effects of targeted killings are less evident. Clearly, the Taliban leaders studied in this book—and the dozens of others that have been killed in other targeted operations—have been, and will continue to be, replaced. New fighters repeatedly settle into vacated leadership positions and the Taliban continues its insurgent campaign in hopes of wresting control of Afghanistan from Western forces and their Afghan allies. Along the way, they have also learned a thing or two about resilience. They have developed ways to circumvent, evade, or limit the immediate effect of each targeted killing: Power has been decentralized and leadership positions compartmentalized so that the death of one leader has a local, rather than a national or regional, consequence. And charismatic leaders like Dadullah no longer regale journalists with personal anecdotes over tea and biscuits, lest these meetings inadvertently help direct the drones. Indeed, in March 2008, a young Taliban subcommander noted that following Dadullah’s killing, a number of changes to the Taliban’s command-and-control structure had been implemented: “After Dadullah’s death, we have a motto that everyone has become a Dadullah. . . . There are a lot of small commanders now.”6 This decentralization of power is one immediate consequence of targeted eliminations that might potentially have a lasting coercive effect. If everybody is fearful of becoming the next Dadullah, then the behavior of each person will have been influenced over the long term. More research will be needed to explore these issues in greater detail.

Deterrence and Asymmetry: Interests, Resolve, and Power I turn next to the lessons for deterring terrorism derived from the various theoretical explorations. An important though generally understudied concern involves the role asymmetry plays in determining how states calculate, measure, and issue coercive threats against much weaker adversaries, like nonstate militant groups. A number of asymmetries differentiate states and militant groups, and each influences how deterrence theory might be successfully applied. First, there is an asymmetry in the very interests states and militants bring to the table. These interests shape both parties’ willingness to contemplate certain actions and to pursue those actions to the end once they are started. At issue is how important each actor interprets the stakes at hand, which informs how far and how long that party might be willing to carry on with a

178

Chapter 6

given conflict or deterrent relationship. Arguably, existential threats—those that threaten the very livelihood of the actor in question—have greater importance than other, lesser threats. Actors at risk of being snuffed out will show greater resolve to continue pursuing a given action than those facing nonexistential threats. They may be especially skillful at practicing coercion, too, because their strategic position adds credence and credibility to their deterrent threats. Before 9/11, as Laqueur reminds us earlier in this chapter, most states did not consider terrorism an especially grave threat; rarely was it an existential challenge to core national interests. Of course, Israelis months into the al Aqsa Intifada, or Sri Lankans at the high point of the LTTE’s insurgency, or Malians during al Qaeda’s near conquest in 2013, or Iraqis during ISIS’s 2014 expansion probably thought otherwise: Terrorism and militant groups certainly did pose a significant challenge to these countries. But for the United States and much of the Western world, terrorism has long been a nuisance, an irritant that sits well down the totem pole of national security threats. Before 2001, U.S. political will to actively and purposefully dismantle al Qaeda was much lower than al Qaeda’s interest in striking the United States, and diverging levels of attention and interest marked each actor’s behavior toward the other. After 9/11, that asymmetry changed markedly. Arguments contending that al Qaeda’s interest in achieving its goals and its resolve to do that are higher than those held by the West to protect itself from terrorism, are increasingly misplaced. The scale of the 9/11 attacks, along with the subsequent fear of WMD terrorism, realigned the interest and resolve asymmetries that once defined relations between al Qaeda and the West. That the years following 9/11 was marked by a multitude of small-scale, domestic, and often al Qaeda–inspired terrorist attacks in North America, Europe, and Australia—to say nothing of the daily carnage visited on civilians living in places like Iraq, Syria, Pakistan, Yemen, Nigeria, and Somalia—further strengthened U.S. resolve to treat terrorism as an appropriately important security challenge. As a result, a new symmetry of resolve now exists between al Qaeda (and its partners) and the West. And that should help improve our deterrent and coercive bargain with militants.7 We simply have more at stake, so our coercive threats appear appropriate and credible. As a vignette, consider French military intervention into Mali in January 2013. To the surprise of many observers, Paris took it upon itself to unilaterally fast-track a UN-mandated intervention into the besieged African country and to rid it of al Qaeda–linked militants threatening to overrun the capital of Bamako.

Moving Ahead with Deterrence Theory

179

French foreign minister Laurent Fabius called the mission a national and global necessity: “We—not just the French, but all nations—have to combat terrorism.”8 This from “perfidious France,” a country many Americans were fond of referring to as “cheese-eating surrender monkeys” only half a decade ago.9 After 9/11, terrorism (at least in appearance) has become everybody’s problem, narrowing the gap between terrorist and counterterrorist goals, interests, and resolve. This improves our coercive position because we are—and, more important, appear to be—committed to thwarting al Qaeda and deterring terrorism. A second asymmetry is the obvious technological fissure in military capability that exists between states and nonstate militant actors. No terrorist organization has ever come close to matching the military muscle of the United States; no group is likely to, ever. Only on very rare occasions, in which a state and its military are especially weak and a militant adversary is especially strong, do capabilities reach a level of parity. The case of Mali is once again informative. Having suffered an earlier coup, by 2013 the government had lost control of its northern territory to a patchwork of Islamist (and secular Tuareg) militant groups allied with al Qaeda. Malian power and military capability was at an absolute nadir. At the same time, the militant Islamists, flush with sophisticated weapons pillaged from the retreating Malian troops and purchased or stolen from Libyan stockpiles (which had become vulnerable following NATO’s 2011 intervention against Muammar Qaddafi’s Libyan regime) were at their zenith. The asymmetry in power soon became obvious. In early January 2013, the Islamists pressed their advantage and moved south with thousands of heavily armed and battle-tested fighters. They made quick work of their Malian adversaries, who were undergunned, disorganized, and totally incapable of defending Malian sovereignty.10 The capital seemed likely to fall. In this particular case, the militant groups’ strength was equal to, if not greater than, the state’s. Only French intervention righted the scales. For nonstate militant organizations, the Mali case is the rare exception. Terrorist groups by their very nature are weak actors. Understanding how power asymmetries shape a terrorist group’s attempt to circumvent disparities in capability in order to coerce a stronger adversary (while avoiding traditional military confrontation) will affect how deterrence in counterterrorism will work. Here, Bar’s notion of the power of weakness—the relative power that the weaker side in a contest has over the stronger party—is particularly

180

Chapter 6

important.11 Militants neutralize features of a state’s technological advantage by either finding defenses against counterterrorism tactics or creating circumstances that prevent the state from using its technological and tactical advantages. For example, bin Laden went off-grid following the U.S. invasion of Afghanistan, limiting his personal contact with his network to a few trusted couriers who carried his correspondence around the region. That was a behavioral and defensive adaptation on his part that circumvented U.S. prowess in collecting electronic and signals intelligence. Bin Laden survived another decade by laying low. Or consider the case of Hamas, which fires crude rockets into Israel from densely populated neighborhoods, public grounds, mosques, and schoolyards. Doing so puts limitations on how the Israeli Air Force can respond by purposefully placing Israelis in a position to inflict unwanted (and unwarranted) collateral damage on Palestinian noncombatants.12 The flip side of the power of weakness is the weakness of power. Strong states retain their power on the basis of a variety of factors— like maintaining an open, accessible, and integrated economic system; a large critical infrastructure; and an adaptable and educated populace—that also represent vulnerabilities exploitable by weaker nonstate adversaries.13 The bottom line is that the asymmetry in capabilities that marks state–militant relations skews traditional conceptions of power that are critical to practicing deterrence. Traditional military capabilities—like tanks, fighter jets, or nuclear weapons—that are useful for deterring state adversaries lose much of their coercive value and credibility when used against (relatively) weak adversaries like nonstate actors. By manipulating both the power of weakness and the weakness of power, terrorists are able to circumvent a state’s strategic preferences, altogether diluting the feasibility of practicing coercion, deterrence, or compellence the way the state may want to. Moving forward, we need to better assess how states can repackage their inherent strengths to the service of deterring terrorism.

Deterrence and Coercive Communication: Tailoring Lost Messages A third lesson drawn from this study involves coercive communication. Scholars applying deterrence to counterterrorism usually combine various aspects of the theory in order to illustrate how the logic of deterrence can be

Moving Ahead with Deterrence Theory

181

used against distinctive elements of a terrorist group. Deterrence by denial and deterrence by punishment are, for example, woven together with the logic informing general deterrence and immediate deterrence, and the entire lot is applied simultaneously to counter and to deter terrorism. The result is a thick web of coercion—or, as Davis and Jenkins posit, a “portfolio of influences”14—each tailored against one particular actor within the terrorism constellation. By expanding deterrence theory and practice and by applying it across a spectrum of disparate actors, we appear to have the ability to fine-tune deterrence to a remarkable degree. The problem with tailoring deterrence so finely, however, is that we risk muddying and weakening our coercive messages and intentions along the way. (Recall that communication is a core requisite of practicing deterrence.) What we think should deter or coerce behavior does not because we unintentionally miscommunicate our goals, capabilities, or intentions. For example, consider a state threatening to use military force to retaliate against a militant group’s leadership, safe havens, and sponsors, while also threatening to deny the group access to domestic targets and further promising to mitigate the social, economic, and political effects an attack might have. The first coercive message is that terrorism will be costly and painful to the perpetrator; the second and third messages are that terrorism will be difficult to accomplish and will not amount to much even if successful. These threats are all equally and logically consistent with theories of deterrence, appear sensible and credible for deterring terrorism, and may well sway militant behavior. More problematic, however, are the hidden implications and assumptions that inform each threat. Threats of punishment are based on an expectation (or, at least, a hope) that terrorist attacks will not take place at all. Threats of denial and mitigation, however, are based on an expectation that attacks are likely (or unavoidable) but manageable. The difference is subtle but significant: By practicing each together, a state risks sending contradictory coercive messages. A challenger might interpret threats of punishment as evidence that the state does not believe that it is capable of defending against or mitigating the effects of political violence; alternatively, it might interpret threats of denial and mitigation as evidence that the state is concerned about its ability or willingness (or both) to see retaliatory threats through.15 Either way, the risk is that independent coercive messages inadvertently interfere with or negate others, altogether jeopardizing the practice of deterrence.

182

Chapter 6

Evidently, strengthening coercive communications is critical to deterring terrorism. How do we do that? To begin with, states might consider limiting what they hope to achieve with deterrence and tailoring their coercive communication to these specific goals. Deterring all terrorism may not be feasible, but deterring some terrorism is.16 Picking and choosing our deterrent battles will help streamline the communication process. As Frank Harvey and I explain, “The more numerous the coercive threats and the broader their scope, the more complex the interactions between adversaries become, and the higher the odds that intentions and aims will be poorly communicated and improperly interpreted.”17 In thinking about deterring terrorism, simplicity is probably best; complexity invites miscommunication and coercive failure. One way to approach simplicity is to reexamine broad and narrow deterrence as they relate to counterterrorism. In broad deterrence, all acts of aggression are on the docket: States try to limit all occurrences of war or aggression with another actor and seek to pacify relations more generally. But narrow deterrence focuses on a particular unwanted behavior and often occurs within an ongoing conflict: States try to limit the occurrence of a specific behavior while other acts of aggression are ongoing. Put another way, narrow deterrence may involve deterring the especially nasty things actors contemplate while fighting wars. In deterring terrorism and in communicating threats to nonstate actors, narrow deterrence may be especially relevant. The U.S. conflict with al Qaeda and its allies and franchises, for example, is ongoing; it is a hot, rather than cold, war. This does not mean, however, that deterrence is irrelevant, but it does suggest that narrow deterrence will likely trump general deterrence for the time being. Although the conflict with al Qaeda will endure and sporadic terrorist attacks on American and allied interests are likely, defenders might nonetheless be able to use coercion and deterrence to influence where and how future terrorist attacks take place. The goal of deterrence is not (necessarily) an end to all terrorism (though that would be nice, too) but rather limiting the occurrence of certain forms of violence. The coercive objective here is to communicate to militants the sorts of acts that are especially heinous (and especially unwanted), those acts that would result in a particularly firm and unprecedented response. For most states, deterring WMD terrorism fits into this category. Officials in the United States can concentrate on developing threats of punishment and denial as they relate to WMD terrorism in particular and can communicate those threats clearly, individually, and credibly. Those threats would func-

Moving Ahead with Deterrence Theory

183

tion separately from other U.S. counterterrorism measures and deterrent strategies and would not necessarily relate to conventional terrorist attacks. To a certain degree, this is already happening. In the case of deterring WMD terrorism by punishment, the United States has made it clear to North Korea, Iran, and others that the transfer of nuclear weapons and related material to nonstate militants will not be tolerated. Consider remarks given by Stephen Hadley, national security adviser under President Bush: “The United States will hold any state, terrorist group, or other non-state actor fully accountable for supporting or enabling terrorist efforts to obtain or use weapons of mass destruction, whether by facilitating, financing, or providing expertise or safe haven for such efforts.”18 The coercive threat is limited specifically to WMD terrorism, was clearly (and repeatedly) issued, and appears credible. At the same time, deterring WMD terrorism by denial has likewise been put into practice. Notably, in April 2010, President Obama convened the Nuclear Security Summit in Washington, D.C., which brought together leaders from dozens of countries to discuss international measures for safeguarding stocks of weapons-grade plutonium and uranium. Obama’s goal was to find ways to deny terrorists easy access to nuclear material by getting countries to more effectively protect their domestic stockpiles and by improving multilateral mechanisms that control the flow of nuclear materials globally.19 Tightening controls would hamper the theft and illicit trade of nuclear materials, denying terrorists easy access to the materials they would need to conduct nuclear or radiological attacks. The coercive message in this case was that such materials were exceptionally difficult to obtain and that militants would likely fail if they tried. Both cases highlight the manner in which the United States and its partners have stripped deterring WMD terrorism away from deterring or combating conventional terrorism more broadly. The threat has been given special import. By focusing on one subset of political violence, the coercive message can be fine-tuned and tailored to the specific event and associated actors. This improves the likelihood that would-be challengers—in this case, those contemplating or facilitating CBRN terrorism—will receive and accurately interpret threats of punishment and denial. Elsewhere and under different conditions, other forms of conventional terrorism—like masscasualty bombing attacks against civilians or assaults on nuclear power stations—might also become the specific focus of a tailored coercive communication strategy. Either way, in minimizing our deterrent expectations and in

184

Chapter 6

sharpening our deterrence communications, we better ensure that the theory of deterrence is properly applied.

The Defeat-Deter Paradox: Restraint and Intra-War Dynamics A final theoretical and practical dilemma that needs to be further explored in thinking about deterring terrorism is the issue of restraint.20 Deterrence theory stipulates that if a challenger acquiesces to a threat (that is, if the action is successfully deterred or compelled), a defender will not carry out a punishment. Challenger inaction (or action in the case of compellence) will be met with defender inaction. Carrying out a coercive threat even in the event an adversary bends to the demands of the defender is counterproductive. This is equally true whether dealing with state or nonstate adversaries. In their study of deterring rebellion and domestic dissent, for instance, T. Clifton Morgan and Patrick Regan write that “increasing the probability that punishment will be inflicted on the innocent”—that is, on citizens undeserving of punishment—“serves to undermine deterrence and increase the probability that an individual will join the rebellion.”21 Restraint is key: People need to believe that the state will not punish them if they decide against joining a rebellious or militant movement. If challengers believe otherwise, that they will indeed suffer regardless of whether they change their behavior, then deterrence will fail because the costs and benefits of inaction and action are virtually indistinguishable. Punishment and retaliation devoid of restraint is not coercion but revenge. What matters is the defender’s ability to practice restraint—refraining from unwarranted and unnecessary punishment—and assurance— guaranteeing passivity in response to challenger inaction. In countering terrorism, both restraint and assurance prove problematic because deterrence is pursued alongside other offensive counterterrorist actions and activities that go on regardless of coercive success or failure. That is, states try to thwart attacks and dismantle militant organizations while pursuing coercive goals that seek to deter terrorism and manipulate militant behavior. Although perhaps necessary, kinetic counterterrorism potentially negates assurances of coercive restraint. Critics of deterring terrorism have argued that this is especially relevant with deterrence by punishment, where the line between deterrence goals and counterterrorism intentions is muddied or altogether

Moving Ahead with Deterrence Theory

185

nonexistent.22 A state can try to defeat and destroy its terrorist adversary or it can try to deter and coerce that organization with threats of punishment, critics explain, but it cannot do both simultaneously. I call this coercive puzzle the “defeat-deter paradox.”23 The United States, for instance, has vowed to crush al Qaeda and its associates. Al Qaeda accurately interprets American goals as an attempt to fully dismantle it (however difficult doing so may be). For both adversaries, restraint and reassurance are found wanting and mutual inaction may prove problematic if not impossible. Proponents of the defeat-deter paradox argue that punishment strategies are ineffective in deterring terrorism because they contradict stated (and expected) objectives of defeating terrorism. In his interviews with Israeli officials, for example, Bar found that the conventional wisdom in Israel was that terrorists could not be deterred and that they needed to be “fought until extirpated.”24 The same is heard in Washington, D.C.: Having “put al Qaeda on the path to defeat,” Obama argued in 2011, “we will not relent until the job is done.”25 In this case, coercion lacks credibility because reassurances that our counterterrorism and security goals are indeed limited are suspect or are missing altogether. Working around the defeat-deter paradox can be accomplished in two ways. First, as Chapter 3 illustrates, threats of punishment can be issued against a whole assortment of actors involved in and associated with terrorism. As noted, there are numerous types of challengers against which a variety of different and specifically tailored threats can be applied (notwithstanding dilemmas associated with actually communicating these threats). Distinguishing among punishment strategies that are tailored to state sponsors of terrorism, societal supporters of terrorism, specific terrorist groups, or individuals involved in terrorism can help circumvent the defeat-deter paradox. In each of these cases, the coercive objective is the same: deterring terrorism. But the individual challengers against which coercive leverages are applied come in and out of focus. It is important to note that as one actor is being coerced, another may be destroyed without falling into the trap of invalidating coercive threats at the behest of waging a war on terrorism. For example, a state sponsor of terrorism might be compelled to stop supporting militancy by threats to its interests while associated militants are simultaneously killed in counterterrorism operations. By illustration, Israel, in the mid1980s, compelled Tunisia to stop hosting the PLO by threatening (and carry ing out) limited strikes on its territory, and it simultaneously attacked the PLO directly by killing its key leader (Khalil al-Wazi). The former involved coercion, the latter defeat.

186

Chapter 6

A second way to address the defeat-deter paradox is to apply the logic of intra-war deterrence to counterterrorism. As I suggested earlier, deterrence does not end once hostilities begin. Rather, deterrence theory is used to coerce behavior within warfare. In general, the dilemma of deterring an adversary that one is also trying to defeat has been addressed by theories of deterrence involving states.26 It is possible to deter certain behavior (or a specific form of warfare) while simultaneously engaging in open conflict with that same adversary. The logic of intra-war deterrence places emphasis on the threats defenders communicate to challengers while concurrently carrying out military operations toward their ultimate defeat. Deterring terrorism by punishment is inherently indebted to the logic of intra-war deterrence because coercion will be practiced within an ongoing conflict with militants in which the prospects for terrorism and counterterrorism remain high. States and their militant adversaries are in a state of war. Episodic terrorist violence or counterterrorism operations, however, do not negate the state’s ability to influence or manipulate the location, nature, and intensity of future terrorist attacks. The coercive objective is not the cessation of all political violence but rather the changing of a specific militant behavior. A state might try to deter terrorist escalation, for example. The intra-war coercive goal is to delimit violence. Or a state might try to manipulate a militant group’s choice of targets. The intra-war objective here is to influence a group’s choice of victims (from civilian to hard targets, for example) or to deter attacks on specific targets or on an entire class of targets. Or militants might be compelled to focus on regional rather than international grievances and political dynamics, binding the geographic reach of their violence. In sum, the defeat-deter paradox can be addressed in a variety of ways.

Final Thoughts This book evaluates the immediate coercive effect of a particular form of deterrence by punishment (targeted killings) on militant (Taliban) behavior within the context of a specific and an ongoing conflict (the war in Afghanistan, circa 2007–8). But its larger purpose is to shed light on how classical and contemporary deterrence theory can be recast and applied to counter terrorism, insurgency, and political violence more generally. By thinking beyond conventional theories, the logic of deterrence can be applied to political violence in novel and unique ways. Traditional threats of punishment can

Moving Ahead with Deterrence Theory

187

be broadened and joined by less conventional threats of denial, mitigation, and delegitimization. The theoretical fit is not always perfect, of course, and dilemmas, paradoxes, and practical limitations do present themselves. But these challenges are nothing new: Deterrence has faced shifting international landscapes and evolving security environments before. The theory has repeatedly adapted in response, expanding its core concepts and causal relations to address emerging security threats. That same process is at work today with terrorism and nonstate militancy, forcing deterrence theory into new, uncharted, and untested territory. We are in the early stages of that process and much work still needs to be done.27 Yet the benefits of applying deterrence theory and practice to terrorism outweigh the current theoretical and empirical challenges. Some terrorist groups can be defeated outright, and militant movements seem to come and go.28 The revolutionary European Marxists of the 1960s, 1970s, and 1980s, for example, were largely beaten in Germany, Italy, France, and elsewhere. The same goes more recently for the LTTE in Sri Lanka, Aum Shinrikyo in Japan, and the Shining Path in Peru. But for other groups and militant movements, defeat may prove especially daunting. Al Qaeda and the larger militant Islamist movement is a case in point. Although it appears that al Qaeda’s central leadership—those individuals responsible for the group’s activities during the 1990s and early 2000s—has largely been dismantled, al Qaeda has nonetheless established offshoots and franchises that thrive in new geographic locales. It also inspires individuals who are unaffiliated with the organization to fight in its name. “Clear cut victory,” against al Qaeda and Islamist militancy, General Sir David Richards, head of Britain’s armed forces, concluded in 2010, can “never be achieved.” We must instead strive to “contain” it sufficiently such that the threat of terrorism is properly managed.29 Deterrence theory has a role to play in any such strategy. Forcing terrorist organizations to change their behavior, to limit the scope of their attacks, to reorganize their internal structures, or to shift tactics by relying on punitive measures (like targeted killings) is part and parcel of that overall containment and management process. Our next step is to better understand how other, nonpunishment-based coercive processes like denial and delegitimization further deter terrorism. Since 9/11, offensive and defensive measures have stood as the two central pillars of counterterrorism. In the coming decade, deterrence will rise as the third.

APPENDIX

Research Design and Methodology

The empirical research described in this book is designed as a structured and focused comparison of related cases. It follows closely—though also goes beyond—a number of innovations in comparative and qualitative analysis developed by Alexander George and Andrew Bennett1 Th is Appendix describes the book’s methodology, research design, and data collection.

Methodology Comparative methodology is based on certain attributes associated with the experimental method. At its most basic level, the experimental method compares “two equivalent groups” in which one (the experimental group) is exposed to a particular stimulus and the second (the control group) is not. Any divergence between the two groups can be attributed to the stimulus.2 Social scientists, however, face restrictions in establishing reliable controls for experimental analysis; we cannot easily conduct laboratory work in the humanities. 3 Comparative analysis alone, then, provides uncertain fi ndings. More often than not, the cases we analyze diverge in a number of ways other than that dictated by the control. Th is may affect our research and fi ndings. In the study of deterrence, these faults are common and wellappreciated.4 George and Bennett provide a number of alternatives to purely controlled comparative analysis, arguing more generally that their qualitative social science approach offers one of many ways to increase our “confidence” in a theory.5 This study’s research design includes a number of innovations in comparative analysis that are meant to address certain

190

Appendix

shortfalls associated with conducting a qualitative comparative analysis of deterrence theory.

Before-After Research Design

Identifying cases that are as similar as possible to one another can involve locating a “control” along a “single longitudinal case.”6 We can build two cases out of a larger one by focusing on a particular, time-bound event that breaks the single larger case into two smaller subcases. Each subcase is similar to the other but for the occurrence of the given event. This leaves investigators with two environments that are as similar as they might be, with one taking place before and the other after a particular event. Arnold Lijphart argues that these before-after comparisons (or “diachronic” investigations) maximize comparability by introducing a more suitable control mechanism: Comparing “the same unit at different times” allows for greater ability to control how divergent the various cases are.7 The before-after approach is particularly wellsuited to this book. Each targeted killing acts as a break within the larger, longitudinal behavior of the Taliban, creating two distinct environments for comparison that are intimately and inherently similar. Divergences in militant behavior before and after the eliminations allows for a comparative analysis of the deterrent, coercive, and influencing outcome of targeted killings.

Intra-National Investigation

Intra-national investigation, as opposed to inter-national investigation, further refines comparative analysis by taking advantage of the par ticu lar national characteristics of each case. These national characteristics are unique and serve to further standardize the study.8 Evaluating Afghan targeted killings, as opposed to comparing Afghan cases with cases from Somalia, Yemen, Pakistan, Iraq, Chechnya, Gaza, or elsewhere, ensures that par ticu lar commonalities inherent to the Afghan conflict and to the Taliban’s experience are used to inform the research. This is not to suggest, however, that the lessons offered by country-specific investigations do not lend themselves to international events or research. Quite the opposite. Findings from intra-national case studies can lead to the development of

PB

Sect

tion

Appendix

191

theory that can then be (re)applied and tested against other international cases. The Afghan cases, for instance, offer findings that inform international comparisons of targeted killings across geographic, social, and historical spectrums.

Within-Case Analysis

PB

Within-case analysis goes beyond traditional comparative methods. Instead of attempting to analyze the “causal power of a particular variable” as it is applied from case to case, within-case approaches stress the “causal path in a single case.” One way of doing within-case analysis is to use congruence methods. This particular approach begins with a theory and, relying on its tenets and assumptions, uses it to explain the outcome of a particular event or relationship. The approach evaluates how congruent a particular empirical outcome is to the theory under evaluation; it traces the causal chain between variables and illustrates how well findings comport with theoretical expectations.9 In this book, the goal is to illustrate a substantive pattern of congruency between theory and a multitude of Afghan targeted killings. And yet, George and Bennett caution against interpreting consistency between a theory’s predictions and case outcomes as theoretical validation.10 At issue is measuring the significance of comparative findings. Simply uncovering a relationship between two variables says nothing as to the strength of that relationship. Nor does it fully account for spurious relationships. Addressing these methodological hazards requires that we use various safeguards.

Process-Tracing

Process-tracing is one way to assess the causal significance of congruent findings; it outlines the way in which independent and dependent variables are related.11 George and Bennett explain that significance can be surmised by providing a “plausible . . . argument that the deductive theory or empirical generalization being employed is powerful and well validated, that it fits the case at hand extremely well.”12 Process-tracing helps locate the causal pathway between the independent and dependent variables, offers an alternative to purely controlled comparisons by assessing whether (and

192

Appendix

to what degree) differences between the cases influence the causal relationship, and uncovers alternative manners in which the observed outcome might have occurred. Furthermore, it addresses the issue of significance by determining the strength of the relationship uncovered, identifying possible intervening or antecedent variables, and explaining “deviant cases” when and if they arise.13 Finally, process-tracing can also effectively deal with the dilemma of “equifinality,” in which different causal patterns can lead to similar observable outcomes.14 (In plain English: What we see could have come about by different means.) Equifi nality weakens comparative analysis because it undermines the assumption that various cases with similar outcomes share a common explanation. This is especially pertinent to the study of deterrence theory because the outcome (being deterred) is not only difficult to observe but might also result from variables or causal pathways not fully considered. As Achen and Snidal remind us, “pacific intentions are easily confused with successful deterrence, and vice versa.”15 We cannot simply assume that an attack or a war that does not take place is a deterrence success: It is possible and perhaps even likely that the challenger never intended to attack in the first place. Accordingly, tracing the coercion process can strengthen a study’s findings by noting the phenomenon’s exact causal pathways.

Research Design Between 2006 and 2008—the historical context from which the comparative cases were selected—two assumptions informed the U.S.-led campaign to eliminate Afghan Taliban leaders. First, a sustained attack on the operational leadership of the Taliban and associated groups would eliminate the functional echelons of the Afghan insurgency, disrupt day-to-day militant planning, and diminish the capabilities of the group, thereby reducing levels of violence against Afghans and coalition forces and allowing for U.S. counterinsurgency efforts to take root and flourish. Second, targeted killings would represent a tangible cost to participating in violence, challenge the Taliban’s cost-benefit calculations, and influence individual and group behavior accordingly. The first assumption speaks to the counter-capability notions of targeted killings: eliminations diminish what a group can accomplish in practice. The second relates to the logic inherent in deterrence theory: Targeted killings influence motivation and shape behavior.

PB

Sect

tion

Appendix

193

The relationship under investigation in this book is the coercive effects that targeted killings have on the behavior of militant organizations. Two specific aspects of Taliban behavior were measured in particular: the group’s overall professionalism and the type or nature (or both) of the group’s attacks. Changes in Taliban professionalism in the period after the eliminations would suggest that the strikes diminished the capability of the Taliban. Deprofessionalization can be attributed to a militant group’s inability to continue attracting the best facilitators and planners to its cause. Shifts in the nature, sophistication, and type of Taliban attack, however, would suggest that the targeted killings influenced the motivation and behavior of surviving leaders and foot soldiers. Following targeted killings, changes toward easier-to-organize forms of violence can be attributed to a shift in the Taliban’s cost-benefit calculation. Taliban professionalism was measured in a number of ways:

PB

1. Success rate of attacks: determined by the number of dead and injured “targets” in each attack and the ratio between dead and injured “targets” and dead and injured militants for every attack 2. Failure rate of attacks: determined by the number of unsuccessful attacks, including premature explosions, and attacks that fail to meet their primary objective16 3. Foiled rate of attacks: determined by the number of foiled, uncovered, or thwarted attacks The success, failure, and foiled rates can be measured both before and after a particular targeted killing. In this way, a comparative analysis of how these rates change as a result of a specific elimination can be made. Data that inform our assessment of the Taliban’s professionalism but that fall beyond quantifiable measurements—like information on internal divisions, drops in morale, defections, and recruitment issues—can also be added to the overall assessment. In sum, if, following a targeted attack, the Taliban fails to achieve its intended tactical objectives, kills fewer of its targets with each attack, or launches poorly organized attacks more often than it did prior to the targeted killing, then we can argue that its capability and general behavior was influenced as a result of the elimination. If a similar outcome is evident in each of the individual cases, then an even stronger argument can be made. The second behavioral variable—the type and nature of the Taliban’s activity—was measured using the following criteria:

194

Appendix

1. Type of attack a. Suicide bombings (body-borne vs. vehicle-borne attacks)17 b. Improvised explosive devices (IEDs) and roadside bombs18 c. Small arms fire19 d. Rocket or mortar attack e. Frontal assault20 f. Kidnappings and assassination g. Beheading, torture, and other acts of intimidation 2. Type of target a. Soft targets (civilian infrastructure, schools, hospitals, markets, journalists) b. Soft government targets (Afghan, provincial, or local government personnel or facilities) c. Soft international targets (foreign embassies; NGO, UN, or diplomatic facilities) d. Private security guards (either Afghan or international personnel) e. Afghan National Police (ANP) f. Afghan National Army (ANA) g. Coalition forces (NATO/ISAF or U.S./international military personnel associated with Operation Enduring Freedom [OEF]) Measuring how Taliban attacks and target selection shift before and after a targeted killing can offer insight into how the removal of individual facilitators and leaders influences the group’s behavior. The assumption here is that certain types of political violence are easier to plan than others, and that some types of targets are easier to attack than others. For instance, attacking a NATO base with a complex, daytime, frontal assault involving dozens of Taliban foot soldiers will take greater organizational ability than planting a victim-operated IED (that is, a repurposed land mine) alongside a desert or mountain road. In like fashion, carry ing out a vehicle-borne suicide attack against the Afghan National Army will take more time, energy, and sophistication than kidnapping an individual working for an under-defended, international NGO. In deciding what, when, and how to attack, militant organizations must weigh how best to use their limited resources, consider where to place their energy, and assess what type of violent behavior will have the best odds of achieving a desired tactical or strategic effect. Targeted killings can influence those decisions and shape militant behavior.

PB

Sect

tion

Appendix

PB

195

A number of issues need to be further discussed. First, some attacks cannot easily be associated with a particular target. Victim-operated IEDs hidden alongside roads, for instance, rarely distinguish between civilian or military vehicles before they explode. Accordingly, the intended targets of IED attacks are not considered in the data set unless the bombs were remotely activated by militants (that is, “complex” IEDs). Second, most terrorist activity in Afghanistan, especially suicide bombings, kill and injure Afghan civilians. In general, however, civilians are not the intended targets, even though they may suffer the brunt of these attacks. In cases where civilians made up all or most of those killed or injured in a given terrorist attack, careful attention was paid before linking the event with intent to purposefully target civilians. Accordingly, a suicide attack on a NATO convoy that killed civilians but no soldiers would not be attributed as an attack on a soft target, although a suicide attack outside a mosque might be, depending on whether other possible targets were present (government or police officials, for instance). Third, some of the violence in Afghanistan is not attributable to any militant organization. Personal, tribal, ethnic, and religious grievances often lead to death and injury, and theft, narcotic manufacturing, weapons smuggling, and other forms of unlawful activity can all be associated with violent acts. Accordingly, where attribution for an act of violence was uncertain or in cases in which media, police, or government investigations reported the violence as distinct and separate from militant activity (which occurred more often than expected), the event itself was not included in the data set.

A Note on Definitions

Critics might argue that American and coalition efforts—including targeted killings—in Afghanistan are more in line with counterinsurgency (COIN) doctrine than with counterterrorism. The Taliban, after all, is largely a grassroots and popular movement that uses violence—against civilians and military personnel alike—to consolidate its political power and position in the country. International efforts to stabilize Afghanistan post-9/11 have applied COIN principles (including protection of civilian populations, development of economic and political structures, and public diplomacy) alongside security and military operations in hopes of fracturing the public and popular support Afghans lend to the Taliban. The elimination of insurgent leaders is not purely about countering or deterring Taliban violence but about separating

196

Appendix

or isolating militant leaders from their followers and the broader community (a core COIN principle). In a similar vein, labeling militant groups as terrorists and calling their violent actions terrorism, runs into certain definitional dilemmas. Is the Taliban an insurgent or a terrorist group? The issue is more contentious than many assume. Concise definitions are muddied because of the considerable overlap that exists between terrorism and other forms of political violence, like insurgency, militancy, guerrilla warfare, and civil war.21 And the labels themselves lead to serious legal, political, and even moral repercussions. In this study, “terrorism” is the use of indiscriminate violence against noncombatants by nonstate actors with the purpose of generating fear in order to “signal,” communicate, or advance particular sociopolitical objectives. This definition implies that terrorism, as Bruce Hoffman notes, creates “psychological repercussions beyond the immediate victim.”22 It is political violence. Furthermore, violent nonstate militant organizations act independently from states and often lack sovereign territorial control. Even then, definitional distinctions are far from clear-cut and academics have had a field day splitting hairs. Daniel Byman, for instance, finds parallels between insurgency and terrorism, noting that nearly all insurgent groups rely on terrorism at some point.23 The Taliban is a prime example: As Chapter 5 highlights, it does indeed target international and Afghan military forces, but it also, more often than not, targets unarmed Afghan and international civilians. Other scholars, like Austin Long, separate “regional armed groups” (militias, guerrillas, and civil war belligerents) from “transnational” militant groups and “spontaneous terror cells,” like the various homegrown, al Qaeda–inspired groups that have sprouted in the West.24 And still other experts, like RAND’s John Arquilla, use the concept of “network” to describe some, but not all, violent nonstate groups.25 Recent research even equates global counterterrorism as a “global counterinsurgency” against an international assemblage of like-minded, jihadi-inspired groups and individuals.26 Here, too, the Taliban seems to fit the various molds. That it has a deepseated ideological affinity to, is operationally associated with, and shares a strategic vision with al Qaeda, the Islamic movement of Uzbekistan, the Haqqani network, Tehrik-e-Taliban, and other groups that are indeed considered and labeled terrorist groups by a variety of states and international organizations, truly complicates the matter. It does not help to call the Taliban an insurgent group when it conducts local attacks against ISAF but also to label it a terrorist group when it purposefully targets civilians, religious

PB

Sect

tion

Appendix

197

scholars, UN and aid workers, Afghan and foreign officials, civil society actors (teachers, NGO representatives), members of the judiciary, and other so-called “soft targets,” or when it builds lasting operational and strategic ties with widely accepted designated terrorist organizations. The bottom line is that these labels are critically important but also exceptionally complicated to apply. A recognized solution is not at all apparent. Remaining cognizant of these dilemmas, this study conceptually separates targeted attacks against Afghan Taliban leaders as distinct from other COINoriented international efforts in Afghanistan and equates these attacks as counterterrorism, akin to similar operations conducted in Somalia, Iraq, Pakistan, Yemen, and elsewhere. The Taliban is deemed a violent nonstate actor that has a strong penchant for terrorism, though other labels, such as insurgent and militant, are periodically used.

Data Collection

PB

The study required two types of data: reliable information on individual targeted killings and detailed data on militant activity both before and after selected eliminations. In both cases, systematic data collection (organized by day and location) was carried out by relying on a number of open-sourced and (semi)closed-sourced references.

Open Sources

All open-sourced references were located online or accessed in print. From these various references, historical information on the Afghan confl ict (between 2006 and 2008) along with specific details concerning individual militant leaders and specific acts of terrorism and counterterrorism was collected. Many of these sources, especially the media resources, offer daily and archived analysis on the Afghan conflict along with compilations of documents and other official reports. From these references, I tabulated every act of violence (pertinent to the study and in keeping with the various guidelines presented in this Appendix) in the weeks before and after each of the four targeted killings. For example, if an IED was reported by Reuters in a story listed by the Pajhwok Afghan News agency in one of its briefings, corroborating information was sought from other sources and then the data

198

Appendix

point was added to the set. Major events involving death and injury were easy to verify using triangulation. More difficult to substantiate were cases of foiled or failed attacks, arrests, and (sometimes) the exact number of dead or injured. If data points could not be confirmed, they were either excluded from the data set or treated as suspect information in the final analysis. In this manner, I created a detailed data set of daily security incidents for all of Afghanistan centered on each targeted killing. Data were mined from the research and reports published by a number of academic and private institutions, including the following: Long War Journal (www.longwarjournal.org/); Jamestown Foundation (www.jamestown.org /programs/gta/); Project for Defense Alternatives (www.comw.org/pda/); Institute for War & Peace Reporting (www.iwpr.net); Reporters Without Borders (http://en.rsf.org/#); International Crisis Group (www.crisisgroup.org); Center for Defence Information (Afghanistan Update); and Human Security Report Project (www.hsrgroup.org). Data on daily events were also mined from a number of regional media outlets, including the following: Embassy of Afghanistan in Ottawa, Daily News Bulletin/Archives (www.afghanemb-canada.net); Pakistan News Service, Pak Tribune (www.paktribune.com/news); Afgha.com, English News (www.afgha.com); Ariana Afghanistan Worldwide Broadcasting (www.e-ariana .com/ariana/eariana.nsf); Pajhwok Afghan News (www.pajhwok.com); Eurasianet, Afghanistan Daily Digest (www.eurasianet.org/resource/afghanistan); Afghan News Center (www.afghanistannewscenter.com/news/archive). Other Western media outlets were also used (New York Times, Washington Post, Canadian Broadcasting Corporation, British Broadcasting Corporation, Globe and Mail, and National Post).

(Semi)Closed Sources

In Afghanistan, many NGOs, private citizens, security firms, and international organizations are active in rebuilding and stabilizing the country. Many of these groups rely on third-party security assessments to safely conduct their affairs in a volatile security environment. Daily, weekly, and monthly security assessments can be purchased or acquired from a number of private security firms. For instance, individuals working with the International Republican Institute (IRI) in Afghanistan (a nonprofit and nonpartisan organization that promotes democratic and political reform) relied on the reporting

PB

Sect

tion

Appendix

PB

199

expertise of Global Security Ser vices (a full-service private security agency) for information on immediate security incidents, follow-up reporting of incidents, and weekly or bi-monthly Afghan-wide security reports.27 Staffers working with the UN Development Program (UNDP), however, got their daily Afghan security reports from the UN Department of Safety and Security (UNDSS) which acquires its data from both private and public sources.28 In both cases, daily incident reports allow employees and volunteers to conduct their affairs with precise knowledge as to the evolving security environment within which they work. Incident reports also include information on potential future attacks and offer security warnings to that effect. For example, the Strategic Security Solutions International (SSSI) Daily Situation Report for Afghanistan on July 7, 2008 notes the suicide attack on the Indian Embassy in Kabul and reports “of possible further attacks with up to three VBIEDs [vehicle-borne IEDs] active in Kabul.” It cautions that movement in and around the city be limited to essential activity only. The same report also notes that a group of 300 insurgents were known to have moved into the Dangam and Sirkanay districts of Kunar Province.29 This pertinent material retains important analytic value for studying and analyzing militant behavior on a daily basis. Security reports like these are compiled using highly classified and sensitive intelligence sources and are not generally made available to the public.30 In compiling the data sets, I was able to acquire a number of (semi)closed sources from several private security firms. The reports are “semi” closed in that they are not usually made available to the public, although I was allowed access to them. These daily situation reports were invaluable in corroborating open-sourced information. They also led to a substantial cache of underreported events, namely—and most important for the purpose of this study—on foiled and failed attacks and arrests. Sources of daily and weekly (semi)closed reports used in this study included SSSI (a division of Universal Guardian Ser vices Group), Daily Situation Report (Afghanistan); Afghanistan NGO Safety Office (ANSO), Weekly Security Report; GardaWorld, Afghanistan Intelligence Summary (bi-weekly reports); and GardaWorld, Afghanistan Intelligence Overview (monthly reports).31

PB

Dorothy L

. Hodgson

NOTES

Chapter 1 1. Fareed Zakaria, “America’s New World Order,” Newsweek (September 15, 2003). 2. U.S. president George W. Bush, Commencement Address at the United States Military Academy, May 27, 2006. 3. U.S. secretary of defense Donald Rumsfeld, “21st Century Transformation of U.S. Armed Forces,” Address to the National Defense University, January 31, 2002. 4. Elie Wiesel, “Why Terror Must Be Unmasked and Defeated,” International Peace Academy: Fighting Terrorism for Humanity (Conference Proceedings) (September 2003), 13. 5. Lawrence Freedman, “Superterrorism: Policy Responses,” in Superterrorism: Policy Responses, ed. Lawrence Freedman (Malden, Mass.: Blackwell Publishing, 2002), 4. See also Keith B. Payne, Testimony, House Armed Ser vices Committee, Subcommittee on Strategic Forces, July 18, 2007. 6. Charles Krauthammer, “The Obsolescence of Deterrence,” Weekly Standard 8:13 (December 2002). 7. Bruce Hoff man, “Old Madness, New Methods: Revival of Religious Terrorism Begs for Broader U.S. Policy,” Rand Review 22:2 (1998–99), 15. 8. See, for instance, Jerry M. Long, “Strategic Culture, Al-Qaida, and Weapons of Mass Destruction,” Defense Threat Reduction Agency (Advanced Systems and Concepts Office), 2006; Daniel Byman, “Al-Qaeda as an Adversary,” World Politics 56 (2003); Jerry M. Long, “Confronting an Army Whose Men Love Death,” Joint Force Quarterly 44:1 (2007); Peter Bergen, The Osama bin Laden I Know: An Oral History of al Qaeda’s Leader (Toronto: Free Press, 2006). 9. Michael Wines, “Hostage Drama in Moscow: The Moscow Front,” New York Times, October 25, 2002. 10. Quoted in Long, “Confronting an Army,” 79. Original quote from Ehsan YarShater (ed.), The History of al-Tabari, vol. 11 (Albany: State University of New York Press, 1993), 44. 11. “Al-Qaeda Threatens Fresh Terror Attacks,” BBC News, October 10, 2001. 12. “Bin Laden, America ‘Filled with Fear,’ ” CNN Online, October 7, 2001.

202

Notes to Pages 4–8

13. Barry Cooper, “Unholy Terror: The Origin and Significance of Contemporary, Religion-Based Terrorism,” Studies in Defence and Foreign Policy, no. 1 (Fraser Institute, 2002). 14. Lewis Dunn, “Can al Qaeda Be Deterred from Using Nuclear Weapons?,” Occasional Paper, No. 3, Center for the Study of Weapons of Mass Destruction (National Defense University), 2005, 7. 15. Daniel Whiteneck, “Deterring Terrorists: Thoughts on a Framework,” Washington Quarterly 28:3 (2005) 187. 16. Max Abrahms, “What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy,” International Security 31:1 (2008); Max Abrahms, “Are Terrorists Really Rational? The Palestinian Example,” Orbis 48:3 (2004); Max Abrahms and Karolina Lula, “Why Terrorists Overestimate the Odds of Victory,” Perspectives on Terrorism 6:4–5 (2012). 17. Eric Larson, “Exploiting al-Qa’ida’s Vulnerabilities for Delegitimization,” in Counter Violent Extremism: Scientific Methods and Strategies (Topical Strategic MultiLayer Assessment and Air Force research laboratory multidisciplinary white paper in support of counterterrorism and counter-WMD), 2011. 18. Mark Juergensmeyer, “Understanding the New Terrorism,” Current History 99 (2000), 158. 19. Lee Harris, “Al Qaeda’s Fantasy Ideology: War Without Clausewitz,” Policy Review 114 (2002); John Gearson, “The Nature of Modern Terrorism,” in Freedman, Superterrorism. 20. Brian Michael Jenkins, “The New Age of Terrorism,” in McGraw-Hill Homeland Security Handbook, ed. David Kamien (New York: McGraw Hill 2006), 118. 21. Richard Betts, “The Soft Underbelly of American Primacy: Tactical Advantages of Terror,” Political Science Quarterly 117:1 (2002), 31–32; Derek Smith, Deterring America: Rogue States and the Proliferation of Weapons of Mass Destruction (Cambridge: Cambridge University Press, 2006), 39; Andrew Brown and Lorna Arnold, “The Quirks of Nuclear Deterrence,” International Relations 24:3 (2010), 306. 22. Government of the United States of America, The National Security Strategy of the United States of America (September 2002); President George W. Bush, “Graduation Speech at West Point,” United States Military Academy, West Point, N.Y., June 1, 2002. 23. William Broad, Stephen Engelberg, and James Glanz, “Assessing Risks, Chemical, Biological, Even Nuclear,” New York Times, November 1, 2001. 24. James Kraska, “Torts and Terror: Rethinking Deterrence Models and Catastrophic Terrorist Attack,” American University International Law Review 22:3 (2007), 362. 25. Ehud Sprinzak, “Rational Fanatics,” Foreign Policy (September– October 2000), 73. 26. A lively debate focuses on this very premise. For research advocating terrorist strategic rationality, see, among others, Martha Crenshaw, “An Organizational Approach to the Analysis of Political Terrorism,” Orbis 29:3 (1985); Bruce Hoff man, Inside Terrorism (New York: Columbia University Press, 1998); Martha Crenshaw, “The

Notes to Pages 8–10

203

Logic of Terrorism: Terrorist Behavior as a Product of Strategic Culture,” in Origins of Terrorism, ed. Walter Reich (New York: Cambridge University Press, 1990); Betts, “The Soft Underbelly”; Robert Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005); Mia Bloom, Dying to Kill: The Allure of Suicide Terrorism (New York: Columbia University Press, 2005); Louise Richardson, What Terrorists Want (New York: Random House, 2006). For research contending otherwise, see, for instance, Scott Atran, “Mishandling Suicide Terrorism,” Washington Quarterly 27:3 (2004); Abrahms, “Are Terrorists Really Rational?”; Scott Atran, “The Moral Logic and Growth of Suicide Terrorism,” Washington Quarterly 29:2 (2006); Abrahms, “What Terrorists Really Want”; Assaf Moghadam, “Motives for Martyrdom: Al-Qaida, Salafi Jihad, and the Spread of Suicide Attacks,” International Security 33:3 (2008–9). 27. Bloom, Dying to Kill (2007 ed.), 80, 84. 28. Michael Quinlan, “Deterrence and Deterrability,” Contemporary Security Policy 25:1 (2004), 14. 29. Rumsfeld, “21st Century Transformation.” 30. George Quester, “Some Thoughts on ‘Deterrence Failure,’ ” in Perspectives on Deterrence, ed. Paul Stern et al. (Oxford: Oxford University Press, 1989), 54–55. In her critique of deterrence theory, Canadian scholar Janice Gross Stein points to a different lesson: “The theory of deterrence . . . is almost self-confirming in large part because of the microeconomic model of rationality. Almost all failures . . . are either the result of poor implementation of the strategy, or they are beyond [the theory’s] scope.” Janice Gross Stein, “Deterring Terrorism, Not Terrorists,” in Deterring Terrorism: Theory and Practice, ed. Andreas Wenger and Alex Wilner (Stanford, Calif.: Stanford University Press, 2012), 51. 31. I explore these arguments in “Deterring the Undeterrable: Coercion, Denial, and Delegitimization in Counterterrorism,” Journal of Strategic Studies 34:1 (2011). See also Jeff rey Knopf’s discussion in “Wrestling with Deterrence: Bush Administration Strategy After 9/11,” Contemporary Security Policy 29:2 (2008), 232–33. 32. Bernard Lewis, “Free at Last? The Arab World in the Twenty-First Century,” Foreign Affairs 88 (2009). See also Robert McFarlane, “From Beirut to 9/11,” New York Times, October 22, 2008. A similar assessment is offered by Shmuel Bar and others concerning Israel’s supposed deterrence failure against Hezbollah at the onset of hostilities in 2006. Shmuel Bar, “Deterring Nonstate Terrorist Groups: The Case of Hizballah,” Comparative Strategy 26:5 (2007); Amos Malka, “Israel and Asymmetrical Deterrence,” Comparative Strategy 27:1 (2008); Barak Mendelsohn, “Israeli Self-Defeating Deterrence in the 1991 Gulf War,” Journal of Strategic Studies 26:4 (2003), 92. 33. Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), 51. 34. Quoted in Fawaz Gerges, “Are We Safe Yet?: A Foreign Affairs Roundtable,” Foreignaffairs.com (September 2006).

204

Notes to Pages 10–18

35. Robert Anthony, “Deterrence and the 9–11 Terrorists,” Institute for Defense Analyses (Doc. 2802), 2003, 2. 36. “U.S. Airstrikes in Pakistan Called ‘Very Effective,’ ” CNN, May 18, 2009. 37. See my discussion in “Counter-Capability and Counter-Motivation: Combating Terrorism in Canada,” in National Security Strategy for Canada in the Post-9/11 World, ed. D. McDonough (Toronto: Toronto University Press, 2012); Daniel Byman, The Five Front War: The Better Way to Fight Global Jihad (Hoboken, N.J.: Wiley, 2008), esp. chaps. 2 and 7; J. Michael Waller, Fighting the War of Ideas Like a Real War (Washington, D.C.: Institute of World Politics Press, 2007).

Chapter 2 1. Michael Quinlan, “Deterrence and Deterrability,” Contemporary Security Policy 25:1 (2004), 11. 2. Thomas Schelling, The Strategy of Conflict (Cambridge: Cambridge University Press, 1960), 9. 3. Thomas Schelling, “Thinking About Nuclear Terrorism,” International Security 6:4 (1982), 72. 4. Frank Harvey and David Carment, Using Force to Prevent Ethnic Violence: An Evaluation of Theory and Evidence (Westport, Conn.: Praeger, 2001), 46– 47. 5. At times, these labels have proven malleable and difficult to use. See my discussion in Alex Wilner, “Fencing in Warfare: Th reats, Punishment, and Intra-War Deterrence in Counterterrorism,” Security Studies 22:4 (2013), n. 15. 6. Lawrence Freedman, Deterrence (Cambridge: Polity Press, 2004), 26. 7. Karl Mueller, “Strategies of Coercion: Denial, Punishment, and the Future of Air Power,” Security Studies 7:3 (1998), 184; Daniel Byman and Matthew Waxman, “Kosovo and the Great Air Power Debate,” International Security 24:4 (2000), 9. 8. Thomas Schelling, Arms and Influence (New Haven, Conn.: Yale University Press, 1996), 3. 9. Robert Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996). 10. See, for example, Janice Gross Stein, “Deterrence and Compellence in the Gulf, 1990–91: A Failed or Impossible Task?,” International Security 17:2 (1992), 170– 73; Martha Crenshaw, “Coercive Diplomacy and the Response to Terrorism,” in The United States and Coercive Diplomacy, ed. Robert Art and Patrick Cronin (Washington, D.C.: U.S. Institute of Peace Press, 2003), 305– 47. 11. Wyn Bowen, “Deterrence and Asymmetry: Non–State Actors and Mass Casualty Terrorism,” Contemporary Security Policy 25:1 (2004), 58. 12. Schelling, “Thinking About Nuclear Terrorism,” 73. See also David Johnson et  al., Conventional Coercion Across the Spectrum of Conventional Operations: The

Notes to Pages 19–20

205

Utility of U.S. Military Forces in the Emerging Security Environment (Santa Monica, Calif.: RAND, 2002), 14. 13. Richard Ned Lebow and Janice Gross Stein, “Deterrence: The Elusive Dependent Variable,” World Politics 42 (1990), 336. See also Wyn Bowen, “Deterring Mass-Casualty Terrorism,” Joint Force Quarterly (2002), 26. 14. Colin Gray, “The Reformation of Deterrence: Moving On,” Comparative Strategy 22:4 (2003), 438. 15. Lebow and Stein, “Deterrence,” 343. 16. This section is partially based on Alex Wilner and Andreas Wenger, “Linking Deterrence to Terrorism: Promises and Pitfalls,” in Deterring Terrorism: Theory and Practice, ed. Andreas Wenger and Alex Wilner (Stanford, Calif.: Stanford University Press, 2012), 5–7. 17. Robert Jervis, “Deterrence Theory Revisited,” World Politics 31:2 (1979); Lawrence Freedman, Deterrence (Malden, Mass.: Polity Press, 2004), 21–25. 18. Bernard Brodie (ed.), The Absolute Weapon (New York: Harcourt, Brace, 1946); Oskar Morgenstern, The Question of National Defense (New York: Random House, 1959). 19. See, for instance, Patrick Morgan, Deterrence: A Conceptual Analysis (Beverly Hills, Calif.: Sage Publications, 1977); Patrick Morgan, Deterrence Now (Cambridge: Cambridge University Press, 2003); Herman Kahn, On Thermonuclear War (Princeton, N.J.: Princeton University Press, 1960); Bruce Russett, “The Calculus of Deterrence,” Journal of Conflict Resolution 7:2 (1963); Paul Huth and Bruce Russett, “What Makes Deterrence Work? Cases from 1900 to 1980,” World Politics 36:4 (1984). 20. See, among others, Alexander George and Richard Smoke, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press, 1974); Russett, “The Calculus of Deterrence”; Morgan, Deterrence; Glenn Snyder and Paul Diesing, Conflict Among Nations (Princeton, N.J.: Princeton University Press, 1977). 21. George and Smoke, Deterrence; Huth and Russett, “What Makes Deterrence Work?”; Trevor Salmon, “Rationality and Politics: The Case of Strategic Theory,” British Journal of International Studies 2:3 (1976); Gary Schaub, “Deterrence, Compellence, and Prospect Theory,” Political Psychology 25:3 (2004). 22. Richard Ned Lebow, Between Peace and War: The Nature of International Crisis (Baltimore: Johns Hopkins University Press, 1981); Richard Ned Lebow and Janice Gross Stein, “Rational Deterrence Theory: I Think, Therefore I Deter,” World Politics 41:2 (1989); Lebow and Stein, “Deterrence.” 23. Ole Holsti, Crisis, Escalation, War (Montreal: McGill–Queen’s University Press, 1972); John Steinbruner, “Beyond Rational Deterrence: The Struggle for New Conceptions,” World Politics 28:2 (1976); Robert Jervis, Richard Ned Lebow, and Janice Gross Stein, Psychology and Deterrence (Baltimore: Johns Hopkins University Press, 1985); Christopher Achen and Duncan Snidal, “Rational Deterrence Theory and Comparative Case Studies,” World Politics 41:2 (1989); Frank Harvey, “Rigor Mortis or Rigor,

206

Notes to Pages 20–24

More Test: Necessity, Sufficiency, and Deterrence,” International Studies Quarterly 42:4 (1998); Steve Walt, “Rigor or Rigor Mortis? Rational Choice and Security Studies,” International Security 23:4 (1999); Frank Harvey, “Practicing Coercion: Revisiting Successes and Failures Using Boolean Logic and Comparative Methods,” Journal of Conflict Resolution 43:6 (1999); Jeff rey Berejikian, “A Cognitive Theory of Deterrence,” Journal of Peace Research 39:2 (2002). 24. Robert Jervis, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976); Robert Jervis, “Deterrence and Perception,” International Security 7:3 (1982–83); Kevin Woods and Mark Stout, “Saddam’s Perceptions and Misperceptions: The Case of ‘Desert Storm,’ ” Journal of Strategic Studies 33:1 (2010). 25. Keith Payne, Deterrence in the Second Nuclear Age (Lexington: University of Kentucky, 1996); Keith Payne, “The Fallacies of Cold War Deterrence and a New Direction,” Comparative Strategy 22:5 (2003); Colin Gray, “The Reformation of Deterrence: Moving On,” Comparative Strategy 22:5 (2003). 26. Huth and Russett, “What Makes Deterrence Work?,” 499. 27. Lebow and Stein, “Rational Deterrence Theory,” 208–9. 28. See, for instance, Steinbruner, “Beyond Rational Deterrence,” 223–45; Berejikian, “A Cognitive Theory of Deterrence,” 165–83; Jervis, “Deterrence and Perception.” 29. Frank Harvey, “Practicing Coercion,” Journal of Conflict Resolution 43:6 (December 1999), 840– 43; Harvey and Carment, Using Force to Prevent Ethnic Violence, esp. chap. 4. 30. See my discussion of this in Frank Harvey and Alex Wilner, “Counter-Coercion, the Power of Failure and the Practical Limits of Deterring Terrorism” in Wenger and Wilner, Deterring Terrorism. 31. Freedman calls this “deterrence lite.” Lawrence Freedman, “Framing Strategic Deterrence: Old Certainties, New Ambiguities,” RUSI Journal 154:4 (2009), 47. 32. D. Marc Kilgour and Frank C. Zagare, “Credibility, Uncertainty, and Deterrence,” American Journal of Political Science 35:2 (1991). 33. Alexander George and Richard Smoke, “Deterrence and Foreign Policy,” World Politics 41:2 (1989), 177. For a popu lar and an entertaining exploration of this dilemma, see Eric Schlosser, Command and Control: Nuclear Weapons, the Damascus Accident, and the Illusion of Safety (New York: Penguin, 2013). 34. Harvey, “Rigor Mortis or Rigor, More Test,” 676. For the role iteration plays in interstate deterrent relations, see Jonathan Shimshoni, Israel and Conventional Deterrence: Border Warfare from 1953 to 1970 (Ithaca, N.Y.: Cornell University Press, 1988). 35. Thomas Schelling, Arms and Influence, 4. 36. National War College, Combating Terrorism in a Globalized World: Report by the NWC Student Task Force on Combating Terrorism (May 2002), 42. 37. Gary Schaub Jr., “When Is Deterrence Necessary?: Gauging Adversary Intent,” Strategic Studies Quarterly 3:4 (2009), 64– 69.

Notes to Pages 24–29

207

38. Andrew Morral and Brian Jackson, “Understanding the Role of Deterrence in Counterterrorism Security,” Occasional Paper, RAND, 2009, 7; Richard Betts, “The Soft Underbelly of American Primacy: Tactical Advantages of Terror,” Political Science Quarterly 117:1 (2002), 19–36. 39. Harvey and Wilner, “Counter-Coercion,” 97. 40. Frank Harvey, “Getting NATO’s Success in Kosovo Right: The Theory and Logic of Counter-Coercion,” Conflict Management and Peace Science 139 (2006), 150–53. 41. Daniel Byman, Matthew Waxman, and Eric Larson, Air Power as a Coercive Instrument (Washington, D.C.: RAND, 1999), 21; Daniel L. Byman, Matthew Waxman, and Jeremy Shapiro, “The Future of U.S. Coercive Airpower” in Strategic Appraisal: United States Air and Space Power in the 21st Century, ed. Zalmay Khalilzad and Jeremy Shapiro (Santa Monica, Calif.: RAND, 2002). 42. For deterrence by denial, see Glenn Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton, N.J.: Princeton University Press, 1961); John Mearsheimer, Conventional Deterrence (Ithaca, N.Y.: Cornell University Press, 1983); Jonathan Shimshoni, Israel and Conventional Deterrence: Border Warfare from 1953 to 1970 (Ithaca, N.Y.: Cornell University Press, 1988). Some scholars also differentiate between retaliation and punishment. The former threatens pain in the event an action is taken. The latter threatens repeated pain until the unwanted action is reversed. See  Jim Writz, “Deterrence as Strategy: The Strategic Importance of Deterrence by Denial,” conference paper presented to Deterrence by Denial: Theory, Practice, and Empiricism, University of Toronto, October 2013. 43. Gary Schaub Jr., “Deterrence, Compellence, and Prospect Theory,” Political Psychology, 25:3 (2004), 406. 44. Other explanations abound. See my discussion in Wilner, “Fencing in Warfare” n. 26. 45. Brad Roberts, “Deterrence and WMD Terrorism: Calibrating Its Potential Contributions to Risk Reduction,” Institute for Defense Analyses (Paper P-4231) (June 2007), 7; Robert Jervis, “Deterrence, Rogue States, and the U.S. Policy,” in Complex Deterrence, ed. T. V. Paul, Patrick Morgan, and James Wirtz (Chicago: University of Chicago Press, 2009), 136. 46. Martha Crenshaw, “Will Th reats Deter Nuclear Terrorism?” in Wenger and Wilner, Deterring Terrorism. See also Paul Davis and Brian Jenkins, “A System Approach to Deterring and Influencing Terrorists,” Conflict Management and Peace Science 21:1 (2004), 4. 47. Tom Dannenbaum, “Bombs, Ballots, and Coercion: The Madrid Bombings, Electoral Politics, and Terrorist Strategy,” Security Studies 20:3 (2011), 312. 48. David Johnson, Karl Mueller, William Taft, 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), 17. 49. Snyder, Deterrence and Defense, 3.

208

Notes to Pages 30–33

50. John Mearsheimer, “Prospects for Conventional Deterrence in Europe,” Bulletin of the Atomic Scientists 41:7 (1985), 158. 51. See Austin Long, Deterrence: From Cold War to Long War (Washington, D.C.: RAND, 2008), 46–52, 57–58; Michael Gerson, “Conventional Deterrence in the Second Nuclear Age,” Parameters 39:3 (2009) 32– 48; Samuel Huntington, “Conventional Deterrence and Conventional Retaliation in Europe,” International Security 8:3 (1983–84), 35– 40; Richard Harknett, “The Logic of Conventional Deterrence and the End of the Cold War,” Security Studies 4:1 (1994). 52. Morral and Jackson, “Understanding the Role of Deterrence”; David Auerswald, “Deterring Nonstate WMD Attacks,” Political Science Quarterly 121:4 (2006); Robert Trager and Dessislava Zagorcheva, “Deterring Terrorism: It Can Be Done,” International Security 30:3 (2005– 6); Jim Smith, “Strategic Analysis, WMD Terrorism, and Deterrence by Denial,” in Wenger and Wilner, Deterring Terrorism; John Gearson, “Deterring Conventional Terrorism: From Punishment to Denial and Resilience,” Contemporary Security Policy 33:1 (2012). 53. Paul Huth and Bruce Russett, “Testing Deterrence Theory: Rigor Makes a Difference,” World Politics 42:4 (1990); Jervis, “Deterrence Theory Revisited”; Ted Hopf, Peripheral Visions: Deterrence Theory and American Foreign Policy in the Third World, 1965–1990 (Ann Arbor: University of Michigan Press, 1994). 54. Huth and Russett, “Testing Deterrence Theory,” 471. 55. For the role “deterrence by reward” may play in counterterrorism, see Lee Dutter and Ofira Seliktar, “To Martyr or Not to Martyr: Jihad Is the Question, What Policy Is the Answer?” Studies in Conflict and Terrorism 30:5 (2007), 437–38. 56. Miroslav Nincic, “Getting What You Want: Positive Inducement in International Relations,” International Security 35:1 (2010), 138–83. 57. Hopf, Peripheral Visions, 241. 58. Huth and Russett, “What Makes Deterrence Work?,” 497–99. 59. Johnson et al., Conventional Coercion, 12. 60. Jack Snyder, The Soviet Strategic Culture: Implications for Limited Nuclear Operations (Santa Monica, Calif.: RAND, 1977), v. 61. W. Andrew Terrill, Escalation and Intrawar Deterrence During Limited Wars in the Middle East (Carlisle, Pa.: U.S. Army War College, Strategic Studies Institute, 2009), 4. 62. See Long’s discussion of the subject in Long, Deterrence, 30; Schlosser, Command and Control, 276–306. 63. Freedman, Deterrence, 32–34. Other scholars have used the narrow–broad dichotomy to describe other deterrent functions, which somewhat muddies the typology. See Troy Thomas, Stephen Kiser, and William Casebeer, Warlord Rising: Confronting Violent Non-State Actors (Oxford: Lexington, 2005), 186–88. 64. Freedman, Deterrence, 33; Barry Posen, “U.S. Security Policy in a NuclearArmed World, or What If Iraq Had Had Nuclear Weapons?” in The Coming Crisis: Nuclear Proliferation, U.S. Interests, and World Order, ed. Victor Utgoff (Cambridge,

Notes to Pages 33–37

209

Mass.: Belfer Center for Science and International Affairs, 2000), 174–79; Paul Davis and John Arquilla, Deterring or Coercing Opponents in Crisis: Lessons from the War with Saddam Hussein (Santa Monica, Calif.: RAND, 1991). 65. Jervis, “Deterrence, Rogue States, and the U.S. Policy,” 135. 66. Morgan, Deterrence, 28; Freedman, Deterrence, 40–42. See also Vesna Danilovic, “Conceptual and Selection Bias Issues in Deterrence,” Journal of Conflict Resolution 45:1 (2001), 99–101; Morgan, Deterrence Now, 2003. 67. Frank Zagare and D. Marc Kilgour, “Asymmetric Deterrence,” International Studies Quarterly 37:1 (1993), 2. 68. Parts of this section are based on my discussion in Wilner, “Fencing in Warfare,” 746– 47. 69. Uri Bar-Joseph, “Variations on a Theme: The Conceptualization of Deterrence in Israeli Strategic Thinking,” Security Studies 7:3 (1998), 156–57; Doron Almog, “Cumulative Deterrence and the War on Terrorism,” Parameters 34:4 (2004–5); Shmuel Bar, “Deterrence of Palestinian Terrorism: The Israeli Experience” in Wenger and Wilner, Deterring Terrorism; Thomas Rid, “Deterrence Beyond the State: The Israeli Experience,” Contemporary Security Policy 33:2 (2012), 128–30, 137–42. 70. Janice Gross Stein, “Deterring Terrorism, Not Terrorists,” in Wenger and Wilner, Deterring Terrorism, 64. 71. Crenshaw, “Will Threats Deter Nuclear Terrorism?,” 143. 72. Amir Lupovici, “The Emerging Fourth Wave of Deterrence Theory: Toward a New Research Agenda,” International Studies Quarterly 54 (2010), 722. 73. Dima Adamsky, “Change and Continuity in the Israeli Approach to Deterrence,” conference paper presented to Deterrence by Denial: Theory, Practice, and Empiricism, University of Toronto, October 2013; Rid, “Deterrence Beyond the State.” 74. Andreas Wenger and Alex Wilner, “Deterring Terrorism: Moving Forward,” in Wenger and Wilner, Deterring Terrorism, 318. 75. Rid, “Deterrence Beyond the State,” 125. 76. Achen and Snidal, “Rational Deterrence Theory,” 152.

Chapter 3 1. Jeff rey Knopf, “The Fourth Wave in Deterrence Research,” Contemporary Security Policy 31:1 (2010); Amir Lupovici, “The Emerging Fourth Wave of Deterrence Theory: Toward a New Research Agenda,” International Studies Quarterly 54 (2010); Jeff rey Knopf, “Terrorism and the Fourth Wave in Deterrence Research,” in Deterring Terrorism: Theory and Practice, ed. Andreas Wenger and Alex Wilner (Stanford, Calif.: Stanford University Press, 2012). 2. T. V. Paul, Patrick M. Morgan, and James J. Wirtz (eds.), Complex Deterrence: Strategy in the Global Age (Chicago: University of Chicago Press, 2009); Robert Art and Patrick Cronin (eds.), The United States and Coercive Diplomacy (Washington,

210

Notes to Page 37

D.C.: U.S. Institute of Peace 2003); Derek Smith, Deterring America: Rogue States and the Proliferation of Weapons of Mass Destruction (Cambridge: Cambridge University Press, 2006). 3. See Wenger and Wilner, Deterring Terrorism; Michael Bahar, “Attaining Optimal Deterrence at Sea,” Vanderbilt Journal of Transnational Law 40:1 (2007); David Auerswald, “Deterring Nonstate WMD Attacks,” Political Science Quarterly 121:4 (2006), 555– 67. For a popu lar account of how the United States has applied “new deterrence” to combat terrorism, see Eric Schmitt and Thom Shanker, Counterstrike: The Untold Story of America’s Secret Campaign Against Al Qaeda (New York: Times Books, 2011). 4. For a prescient exploration of cyber deterrence, see John Arquilla and David Ronfeldt, “Cyberwar Is Coming!,” Comparative Strategy 12:2 (1993). For more recent work, see Eric Sterner, “Retaliatory Deterrence in Cyberspace,” Strategic Studies Quarterly 5:1 (2011); Matthew Crosston, “World Gone Cyber MAD,” Strategic Studies Quarterly 5:1 (2011); Martin C. Libicki, Cyberdeterrence and Cyberwar (Santa Monica, Calif.: RAND, 2009). 5. Carla M. Machain, T. Clifton Morgan, and Patrick Regan, “Deterring Rebellion,” Foreign Policy Analysis 7:3 (2011), 295–316. For an earlier, Cold War, exploration of deterring revolution, see Jack Goldstone, “Deterrence in Rebellions and Revolutions,” in Perspectives on Deterrence, ed. Paul Stern et al. (Oxford: Oxford University Press, 1989), 222–50. 6. In October 2013, I convened an international conference on deterrence by denial at the Munk School of Global Affairs, University of Toronto, Canada. Conference proceedings are being packaged into a volume (2014–15). Otherwise, see Paul Kapur, “Deterring Nuclear Terrorists,” in Paul, Morgan, and Wirtz, Complex Deterrence; Jim Smith, “Strategic Analysis, WMD Terrorism, and Deterrence by Denial,” in Wenger and Wilner, Deterring Terrorism. 7. Alex Wilner, “Deterring the Undeterrable: Coercion, Denial, and Delegitimization in Counterterrorism,” Journal of Strategic Studies 34:1 (2011), esp. 26–31; Alex Wilner, “Apocalypse Soon? Deterring Nuclear Iran and Its Terrorist Proxies,” Comparative Strategy 31:1 (2012), esp. 31–34; Brian Michael Jenkins, “No Path to Glory: Deterring Homegrown Terrorism,” Testimony, U.S. House of Representatives, Homeland Security Committee, May 2011; Jerry Mark Long and Alex Wilner, “Delegitimizing al-Qaida: Defeating an ‘Army Whose Men Love Death,’ ” International Security 39:1 (2014). 8. Andreas Wenger and Stefanie von Hlatky (eds.), The Future of Extended Nuclear Deterrence in Europe (under review, 2013). 9. Shmuel Bar, “God, Nation, and Deterrence: The Impact of Religion on Deterrence,” Comparative Strategy 30:5 (2011), 428–52; Jeff rey Lantis, “Strategic Culture and Tailored Deterrence: Bridging the Gap Between Theory and Practice,” Contemporary Security Policy 30:3 (2009), 476–78; Long and Wilner, “Delegitimizing al-Qaida.”

Notes to Pages 37–39

211

10. Jonathan Hagood, “Dissuading Nuclear Adversaries: The Strategic Concept of Dissuasion and the U.S. Nuclear Arsenal,” Comparative Strategy 24:2 (2005), 173–84; Gary Ackerman and Lauren Pinson, I-VEO [Influencing Violent Extremist Organizations] Empirical Assessment: Literature Review and Knowledge Matrix, National Consortium for the Study of Terrorism and Responses to Terrorism (2011); John P. Sawyer and Amy Pate, I-VEO [Influencing Violent Extremist Organizations] Empirical Assessment: Case Studies of Historical Efforts to Influence Violent Extremist Organizations, National Consortium for the Study of Terrorism and Responses to Terrorism (2011). 11. For the few exceptions, see Robert Anthony, “Deterrence and the 9–11 Terrorists,” Institute for Defense Analyses, May 2003; Robert Trager and Dessislava Zagorcheva, “Deterring Terrorism: It Can Be Done,” International Security 30:3 (2005– 6); Gary Geipel, “Urban Terrorists in Continental Europe After 1970: Implications for Deterrence and Defeat of Violent Nonstate Actors,” Comparative Strategy 26:5 (2007); Shmuel Bar, “Deterring Nonstate Terrorist Groups: The Case of Hizballah,” Comparative Strategy 26:5 (2007); Amos Malka, “Israel and Asymmetrical Deterrence,” Comparative Strategy 27:1 (2008); Shmuel Bar, “Deterring Terrorists: What Israel Has Learned,” Policy Review 149 (2008); Alex Wilner, “Targeted Killings in Afghanistan: Measuring Coercion and Deterrence in Counterterrorism and Counterinsurgency,” Studies in Conflict and Terrorism 33:4 (2010); Sawyer and Pate, I-VEO Empirical Assessment; and the empirical chapters by Shmuel Bar, Michael Cohen, David Romano, and Fred Wehling in Wenger and Wilner, Deterring Terrorism. 12. This discussion is adapted from Wilner, “Deterring the Undeterrable,” 10–5. 13. Paul Davis and Brian Jenkins, Deterrence and Influence in Counterterrorism: A Component in the War on al Qaeda (Santa Monica, Calif.: RAND, 2002), 9–13, 61; Paul Davis and Brian Jenkins, “A System Approach to Deterring and Influencing Terrorists,” Conflict Management and Peace Science 21:1 (2004). 14. David E. Johnson, Karl P. Mueller, and William H. Taft Conventional Coercion Across the Spectrum of Operations (Santa Monica, Calif.: RAND, 2002), 11. 15. Lewis Dunn, “Influencing Terrorists’ Acquisition and Use of Weapons of Mass Destruction,” NATO Defense College Workshop, August 2008, 2. See also Lewis Dunn, “Deterrence Today: Roles, Challenges, and Responses,” Proliferation Papers, Institut Français des Relations Internationales (2007), 17–22; and Lewis Dunn, Next Generation: Weapons of Mass Destruction and Weapons of Mass Effects Terrorism, Defense Threat Reduction Agency, Advanced Systems and Concepts Office (2008), sec. 3. 16. Michael Quinlan, “Deterrence and Deterrability,” Contemporary Security Policy 25:1 (2004), 11. 17. Doron Almog, “Cumulative Deterrence and the War on Terrorism,” Parameters (Winter 2004–5), 5. 18. A Concept for Deterring and Dissuading Terrorist Networks (Unclassified Draft Report), U.S. Office of the Secretary of Defense (2005), 7–9.

212

Notes to Pages 40–43

19. See, for instance, John Horgan and Kurt Braddock, Terrorism Studies: A Reader (New York: Routledge, 2012). 20. Christopher Harmon, “The Myth of the Invincible Terrorist,” Policy Review 142 (2007), 62. 21. Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), 137–39, 140– 42, 171–73. See also John Arquilla and David Ronfeldt (eds.), Networks and Netwars: The Future of Terror, Crime, and Militancy (Santa Monica, Calif.: RAND: 2001). 22. Trager and Zagorcheva, “Deterring Terrorism,” 98. 23. Michael Powers, “Deterring Terrorism with CBRN Weapons: Developing a Conceptual Framework,” Chemical and Biological Arms Control Institute , Occasional Paper No. 2, February 2001, 4. 24. Sageman, Understanding Terror Networks, 164. 25. Ibid., 164– 65. 26. For works on Osama bin Laden, 9/11, and al Qaeda, see Peter Bergen, The Osama bin Laden I Know: An Oral History of al Qaeda’s Leader (Toronto: Free Press, 2006); U.S. Government, National Commission on Terrorist Attacks, 9/11 Commission Report (New York: W. W. Norton, 2004); Michael Scheuer, Osama Bin Laden (Oxford: Oxford University Press, 2011). For more on Atta and the Hamburg cell more specifically, see Sageman, Understanding Terror Networks, 103–7. 27. Davis and Jenkins, “Deterrence and Influence,” 22. 28. Ariel Merari, “Terrorism and Threats to U.S. Interests in the Middle East,” Testimony, U.S. House of Representatives, Special Oversight Panel on Terrorism, July 13, 2000. 29. Ehud Sprinzak, “Rational Fanatics,” Foreign Policy (2000), 66–73. See also Andrew Kydd and Barbara F. Walter, “Sabotaging the Peace: The Politics of Extremist Violence,” International Organization 56:2 (2002); Harmon, “Myth of the Invincible Terrorist”; Martha Crenshaw, “An Organizational Approach to the Analysis of Political Terrorism,” Orbis 29:3 (1985); Martha Crenshaw, “The Logic of Terrorism: Terrorist Behavior as a Product of Strategic Culture,” in Origins of Terrorism, ed. Walter Reich (New York: Cambridge University Press, 1990); Richard Betts, “The Soft Underbelly of American Primacy: Tactical Advantages of Terror,” Political Science Quarterly 117:1 (2002); Lee Dutter and Ofira Seliktar, “To Martyr or Not to Martyr: Jihad Is the Question, What Policy Is the Answer?” Studies in Conflict and Terrorism 30:5 (2007), 432–36 30. Robert Pape, “Suicide Terrorism and Democracy: What We’ve Learned Since 9/11,” Policy Analysis 582 (2006), 4; Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005). For Pape’s critiques, see Scott Atran, “The Moral Logic and Growth of Suicide Terrorism,” Washington Quarterly 29:2 (2006), 132–39; Scott Atran, “Mishandling Suicide Terrorism,” Washington Quarterly 27:3 (2004); Scott Ashworth, Joshua Clinton, Adam Meirowtiz, and Kristopher Ramsay, “Design, Inference, and the Strategic Logic of Suicide Terrorism,” American Political Science Review 102:2 (2008), 269–73.

Notes to Pages 44–49

213

31. Mia Bloom, Dying to Kill: The Allure of ST (2005; New York: Columbia University Press, 2007). 32. Assaf Moghadam, “How al Qaeda Innovates,” Security Studies 22:3 (2013). 33. Mark Doyle, “Mali Islamists Warned About Sharia in al-Qaeda ‘Manifesto,’ ” BBC, February 26, 2013. 34. Fawaz Gerges, Journey of the Jihadist: Inside Muslim Militancy (Toronto: Harcourt, 2007), 11–14, 39– 45. 35. Parts of the following discussion first appeared in Wilner, “Deterring the Undeterrable,” 14–29 and Alex Wilner, “Fencing in Warfare: Threats, Punishment, and Intra-War Deterrence in Counterterrorism,” Security Studies 22:4 (2013), 750– 62. 36. See Michael Cohen, “Mission Impossible? Influencing Iranian and Libyan Sponsorship of Terrorism,” in Wenger and Wilner, Deterring Terrorism; Stephen Collins, “Dissuading State Support of Terrorism: Strikes or Sanctions?, Studies in Conflict and Terrorism 27:1 (2004), 1–18; Bruce Jentleson and Christopher Whytock, “Who ‘Won’ Libya? The Force-Diplomacy Debate and Its Implications for Theory and Policy,” International Security 30:3 (2005– 6), 47–86. 37. U.S. Joint Chiefs of Staff, National Military Strategy of the United States of America (February 2011), 6. 38. Bar, “Deterring Terrorists,” 30– 40; Wilner, “Deterring the Undeterrable,” 14– 16. The National Institute for Public Policy, in a 2008 report, distinguishes between a “host” state—one that allows a terrorist or militant organization to “function within its territory”—and a “patron” state—one that provides leadership; direction; political, financial, and material support; or sanctuary to an organization. Keith Payne et al., “Deterrence and Coercion of Non-State Actors: Analysis of Case Studies,” National Institute for Public Policy (2008), 8. 39. Byman excludes states that sponsor terrorism, have tried but failed to hinder it, are unaware of its presence, or lack the ability to counter it. Daniel Byman, “Passive Sponsors of Terrorism,” Survival 47:4 (2005– 6), 118. 40. Boaz Atzili and Wendy Pearlman, “Triadic Deterrence: Coercing Strength, Beaten by Weakness,” Security Studies 21:2 (2012). 41. Quoted in “Israel Warns Hezbollah War Would Invite Destruction, Reuters, October 3, 2008. 42. U.S. president George W. Bush, Comments from the Pentagon, September 17, 2001. 43. Armitage denied the conversation: “It would be completely out of character for me to threaten the use of military force when I was not authorized to do so. I don’t command aircraft and could not make good on such a threat.” Melissa Block, “Armitage Denies Making ‘Stone Age’ Threat,” All Things Considered (NPR), September 22, 2006; Tim Reid, “We’ll Bomb You to Stone Age, U.S. Told Pakistan,” Times Online, September 22, 2006. 44. Martha Crenshaw, “Coercive Diplomacy and the Response to Terrorism,” in The United States and Coercive Diplomacy, ed. Robert Art and Patrick Cronin (Washington, D.C.: U.S. Institute of Peace Press, 2003), 335; Stephen Cohen, “America and

214

Notes to Pages 49–52

Pakistan: Is the Worst Case Avoidable?,” Current History 104:680 (March 2005), 132; Paul Blustein, “A Tighter Hand in Doling Out Global Aid?,” Washington Post, September 20,2001; Bessma Momani, “The IMF, the U.S. War on Terrorism, and Pakistan,” Asian Affairs 31:1 (2004), 41–50. 45. Pakistan president Pervez Musharraf, Transcript of Televised Speech, September 19, 2001. Elements within Pakistan’s intelligence community, however, appear to have continued supporting various militant groups, including al Qaeda, well after September 2001. See, for instance, Carlotta Gallmarch, “What Pakistan Knew About Bin Laden,” New York Times Magazine, March 19, 2014. 46. U.S. president George W. Bush, Address to the Nation, September 20, 2001. 47. John Burns, “Pakistanis Fail in Last-Ditch Bid to Persuade Taliban to Turn Over bin Laden,” New York Times, September 29, 2001; U.S. secretary of defense Donald Rumsfeld, Exchange with Media, September 24, 2001. 48. U.S. secretary of defense Donald Rumsfeld, News Conference, October 7, 2001. 49. U.S. president George W. Bush, “Bush on State of War,” October 11, 2001. 50. U.S. secretary of state Colin Powell, Remarks with President Pervez Musharraf, October 16, 2001. 51. Thom Shanker and Eric Schmitt, “G.I. Raid Struck Taliban Leader’s Compound,” New York Times, October 21, 2001; Michael Gordon, “First of the Stinging Jabs,” New York Times, October 21, 2001. 52. John Burns, “Don’t Doubt Steadfastness of Taliban, Envoy Insists,” New York Times, October 20, 2001. 53. Musharraf, Transcript, September 19, 2001. 54. Lawrence Freedman, “The Third World War?,” Survival 43:4 (2001), 74. 55. Crenshaw, “Coercive Diplomacy,” 346. 56. Sammy Salama and Lydia Hansell, “Does Intent Equal Capability? Al-Qaeda and Weapons of Mass Destruction,” Nonproliferation Review 12:3 (2005), 621. 57. Caitlin Talmadge, “Deterring a Nuclear 9/11,” Washington Quarterly 30:2 (2007), 25; William Dunlop and Harold Smith, “Who Did It?,” Arms Control Today 36 (2006); Siegfried Hecker, “Toward a Comprehensive Safeguards System: Keeping Fissile Material Out of Terrorists’ Hands,” Annals of the American Academy of Political and Social Science 607 (2006); Nuclear Forensics: Role, State of the Art, and Program Needs, American Physical Society and American Association for the Advancement of Science, 2005. 58. David Sanger and Thom Shanker, “U.S. Debates Deterrence for Nuclear Terrorism,” New York Times, May 8, 2007. 59. David Makovsky, “The Silent Strike,” New Yorker, September 17, 2012; Elliott Abrams, “Bombing the Syrian Reactor: The Untold Story,” Commentary magazine, February 2013. 60. Daniel Whiteneck, “Deterring Terrorists: Thoughts on a Framework,” Washington Quarterly 28:3 (2005), 190.

Notes to Pages 53–56

215

61. Michael Levi, “Deterring State Sponsorship of Nuclear Terrorism,” Council Special Report 39 (2008), 4–5. 62. Jasen J. Castillo, “Nuclear Terrorism: Why Deterrence Still Matters,” Current History 102 (2003), 426. 63. Matthew Phillips, “Uncertain Justice for Nuclear Terror: Deterrence of Anonymous Attacks Through Attribution,” Orbis 51:3 (2007), 434–35. 64. Auerswald, “Deterring Nonstate WMD Attacks,” 555–59. 65. William Broad, “Sowing Death: A Special Report; How Japan Germ Terror Alerted World,” New York Times, May 26, 1998; Tim Ballard et al., “Chronology of Aum Shinrikyo’s CBW Activities,” CNS Reports, Monterey Institute of International Studies (2001); Theodore Karasik, Toxic Warfare (Santa Monica, Calif.: RAND, 2002); Jonathan Tucker (ed.), Toxic Terror: Assessing Terrorist Use of Chemical and Biological Weapons (Cambridge, Mass.: MIT Press, 2000). 66. Fred Wehling, “A Toxic Cloud of Mystery: Lessons from Iraq for Deterring CBRN Terrorism,” in Wenger and Wilner, Deterring Terrorism, 267– 69. 67. Lewis Dunn, “Can al Qaeda Be Deterred from Using Nuclear Weapons?,” Occasional Paper No. 3, Center for the Study of Weapons of Mass Destruction, 2005, 24. For other research that explores the relationship among group motivation, WMD acquisition, and WMD use, see Nancy Kay Hayden, “Terrifying Landscapes,” Defense Threat Reduction Agency (2007); Jerrold Post, “Psychological and Motivational Factors in Terrorist Decision Making,” in Tucker, Toxic Terror. 68. Brian Michael Jenkins, “The Terrorist Perception of Nuclear Weapons and Its Implications for Deterrence,” in Wenger and Wilner, Deterring Terrorism, 119–20. See also Brian Michael Jenkins, Will Terrorists Go Nuclear? (New York: Prometheus Books, 2008), esp. chaps. 4 and 5. 69. Jenkins, “The Terrorist Perception of Nuclear Weapons,” 119–20. See also Jenkins, Will Terrorists Go Nuclear?, esp. chaps. 4 and 5. 70. Jerry Mark Long, “Strategic Culture, Al-Qaida, and Weapons of Mass Destruction,” Defense Threat Reduction Agency, Advanced Systems and Concepts Office (2006), 22. 71. Nazi Germany razed Ukrainian, Belarusian, Polish, and French villages to punish dissidents and partisans. Soviet Russia did much the same in Estonia, Lithuania, Poland, Afghanistan, and elsewhere. Syria’s Hafez al-Assad leveled the city of Hama in 1982 to smother a local uprising, and his son and heir, Bashar, again used the tactic during the Syrian civil war. Iraq used chemical weapons in Halabja, northern Iraq, in 1988 to stifle the Kurdish independence movement. Russian forces repeatedly hammered the Chechen capital of Grozny in the 1990s to quell a separatist movement. 72. Gerald Steinberg, “Rediscovering Deterrence After September 11, 2001,” Jerusalem Letter/Viewpoints, No. 467, Jerusalem Center for Public Affairs (2001). 73. Elbridge A. Colby, “Expanded Deterrence,” Policy Review 149 (2008).

216

Notes to Pages 56–61

74. Dexter Filkins, “After Syria: Hezbollah Looks to a Future Without Assad,” New Yorker, February 25, 2013, 55. 75. Edraim Benmelech, Claude Berrebi, and Esteban Klor, “Counter-SuicideTerrorism: Evidence from House Demolitions,” National Bureau of Economic Research, Working Paper 16493, October 2010. 76. Fisher recounts a 1985 incident involving the abduction of a Soviet diplomat by Hezbollah. In retaliation, the KGB responded by killing a relative of a Hezbollah leader and returning the mutilated corpse (the man’s severed genitals were apparently stuffed in his mouth) to the relatives. The implicit threat was that more of the same would be meted out. The diplomat was promptly released. Uri Fisher, “Deterrence, Terrorism, and American Values,” Homeland Security Affairs 3:1 (February 2007), 4, 13. 77. Payne et al., “Deterrence and Coercion of Non-State Actors,” 42– 43. 78. Bar, “Deterring Terrorists,” 29– 42. 79. Benmelech, Berrebi, and Klor, “Counter-Suicide-Terrorism,” 14–15; Ariel Merari, “Israel Facing Terrorism,” Israel Affairs 11:1 (2005), 228–32; Jason Lyall, “Does Indiscriminate Violence Incite Insurgent Attacks? Evidence from Chechnya,” Journal of Conflict Resolution 53:3 (2009); Thomas Rid, “Deterrence Beyond the State: The Israeli Experience,” Contemporary Security Policy 33:2 (2012), 140; James Lebovic, Deterring International Terrorism and Rogue States (London: Routledge, 2007), 126–27. 80. Wyn Bowen, “Deterrence and Asymmetry: Non-State Actors and Mass Casualty Terrorism,” Contemporary Security Policy 25:1 (2006), 62. 81. Bryan Brophy-Baermann and John Conybeare, “Retaliating Against Terrorism: Rational Expectations and the Optimality of Rules Versus Discretion,” American Journal of Political Science 38:1 (1994), 203–9. 82. Chris Dorsey, “Tancredo Says Threat of Attack on Holy Sites Would Deter Terrorism,” IowaPolitics.com, July 31, 2007. 83. Trager and Zagorcheva, “Deterring Terrorism,” 104–5, 108–10. 84. 9/11 Commission, 62– 63. 85. Combating Terrorism in a Globalized World, National War College (2002), 44. 86. M. Elaine Bunn, “Can Deterrence Be Tailored?,” Strategic Forum, no. 225, Institute for Strategic Studies (NDU) (January 2007), 3. 87. Other scholars are much less sanguine about the feasibility of turning off the financial taps, suggesting that terrorists can fi nd alternative ways to fi nance their affairs. Peter Fitzgerald, “Tightening the Screws: The Economic War Against Terrorism,” National Interest 66 (2001–2); Loren Yager, “Terrorist Financing: Better Strategic Planning Needed to Coordinate U.S. Efforts to Deliver Counter-Terrorism Financing Training and Technical Assistance Abroad,” U.S. Government Accountability Office (2005); Payne et al., “Deterrence and Coercion of Non-State Actors,” 20; Mark Basile, “Going to the Source: Why Al Qaeda’s Financial Network Is Likely to Withstand the Current War on Terrorist Financing,” Studies in Conflict and Terrorism 27 (2004); Michael Levi, “Combating the Financing of Terrorism: A History and Assessment of the

Notes to Pages 62–66

217

Control of ‘Threat Finance,’ ” British Journal of Criminology 5 (2010); Johan Bergenas et al., “Killing Lions, Buying Bombs,” New York Times, August 9, 2013. 88. Eric Schlosser, Command and Control: Nuclear Weapons, the Damascus Accident, and the Illusion of Safety (New York: Penguin, 2013), 373. 89. Report of the Future of Terrorism Task Force, U.S. Department of Homeland Security (2007), 6. 90. Ackerman and Pinson, I-VEO Empirical Assessment, 23. Morral and Jackson highlight three types of defensive measures that increase uncertainty: unknown defensive capabilities, random measures, and measures that force terrorists to “adopt . . . more-risky tactics.” Andrew Morral and Brian Jackson, “Understanding the Role of Deterrence in Counterterrorism Security,” Occasional Paper, RAND, 2009, 10–14. 91. U.S. Air Force Reserve, 911 Airlift Wing Instructions 31–210, “Installation Random Antiterrorism Measures (RAM) Program” (November 19, 2004). 92. Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 1998), esp. chap 8. 93. R. P. Eddy, “Making New York Safer,” Council on Foreign Relations Policy Symposium: The Terrorist Threat in New York, September 8, 2006. 94. See Anthony, “Deterrence and the 9–11 Terrorists.” 95. Eric Schmitt and Thom Shanker, “U.S. Adapts Cold-War Idea,” New York Times, March 18 2008. 96. James Pita et al., “Deployed ARMOR Protection: The Application of a Game Theoretic Model for Security at the Los Angeles International Airport,” Proceedings of the 7th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Portugal 2008. 97. Powers, “Deterring Terrorism with CBRN Weapons,” 15. 98. James Smith and Brent Talbot, “Terrorism and Deterrence by Denial,” in Terrorism and Homeland Security, ed. Paul Viotti, Michael Opheim, and Nicholas Bowen (New York: CRC Press 2008), 54–9. Elsewhere, Smith (writing of CBRN terrorism) relates strategic futility with denial of legitimacy: “The goal is to create the perception in the mind of the terrorist leadership that it cannot achieve its objectives or fulfi ll its ideological goals through the use of WMD terrorism.” Smith, “Strategic Analysis,” 176. 99. Lawrence Freedman, Deterrence (Cambridge: Polity Press, 2004), 123–24. 100. Almog, “Cumulative Deterrence,” 12–14. 101. Joel Breenberg, “Defiant Israelis Crowd Jerusalem Mall Where Suicide Bomb Attack Killed 4,” New York Times, September 6, 1997. 102. Nasser Weddady, “How Europe Bankrolls Terror,” New York Times, February 16, 2013. 103. Brian Jenkins, Unconquerable Nation: Knowing, Our Enemy Strengthening Ourselves (Santa Monica, Calif.: RAND, 2006), 157. 104. Max Abrahms, “Are Terrorists Really Rational? The Palestinian Example,” Orbis 48:3 (2004), 542; Abrahms, “Why Terrorism Does Not Work,” International

218

Notes to Pages 66–69

Security 31:2 (Fall 2006), 72–75; Max Abrahms and Karolina Lula, “Why Terrorists Overestimate the Odds of Victory,” Perspectives on Terrorism 6:4/5 (2012). 105. David Lake, “Rational Extremism: Understanding Terrorism in the Twenty– First Century,” Dialog–IO (Spring 2002), 17. 106. For the handful of studies that touch on the concept of delegitimization in deterrence, see Joseph Lepgold, “Hypotheses on Vulnerability: Are Terrorists and Drug Dealers Coercible?,” in Strategic Coercion: Concepts and Cases, ed. Lawrence Freedman (New York: Oxford University Press, 1998), 144–46; Long, “Strategic Culture, Al-Qaida, and Weapons of Mass Destruction”; Lantis, “Strategic Culture and Tailored Deterrence,” 476–78; Jenkins, “No Path to Glory”; Bar, “God, Nation, and Deterrence”; Wilner, “Apocalypse Soon?,” 31–34; Cheryl Benard, “The Mechanics of De-Legitimization,” in Countering Violent Extremism: Scientific Methods and Strategies (September 2011); Eric Larson, “Exploiting al-Qa’ida’s Vulnerabilities for Delegitimization,” in Countering Violent Extremism: Scientific Methods and Strategies (September 2011); Long and Wilner, “Delegitimizing al-Qaida.” 107. Knopf, “Terrorism and the Fourth Wave.” 108. Long and Wilner, “Delegitimizing al-Qaida.” 109. Lantis, “Strategic Culture and Tailored Deterrence,” 470, 477. 110. Brad Roberts, “Deterrence and WMD Terrorism: Calibrating Its Potential Contributions to Risk Reduction,” Institute for Defense Analyses (Paper P-4231) (June 2007), 8–17. 111. John Parachini, “Putting WMD Terrorism into Perspective,” Washington Quarterly 26:4 (2003), 45. 112. Quoted in Schmitt and Shanker, “U.S. Adapts Cold War Idea.” 113. KSM quoted in John Stone, “Escalation and the War on Terror,” Journal of Strategic Studies 35:5 (2012), 645. 114. Knopf coined the term “popular backlash,” channeling Lewis Dunn’s preliminary work. Dunn envisions deterring al-Qaida’s use of WMD materials by weighing on its concerns of a “backlash among the wider Muslim audience.” Lewis Dunn, “Influencing Terrorists’ WMD Acquisition and Use Calculus,” [excerpt from Next Generation WMD and WME Terrorism], Defense Threat Reduction Agency (2008), 13. 115. Shmuel Bar and Yair Minzili, “The Zawahiri Letter and the Strategy of AlQaeda,” in Current Trends in Islamist Ideology, vol. 3, ed. Hillel Fradkin, Husain Haqqani, and Eric Brown (Washington, D.C.: Hudson Institute, 2006); Frank Douglas, “Waging the Inchoate War: Defining, Fighting, and Second-Guessing the ‘Long War,’ ” Journal of Strategic Studies 30:3 (2007), 401–9; Fawaz Gerges, “Buried in Amman’s Rubble: Zarqawi’s Support,” Washington Post, December 5, 2005. 116. Patrick Porter, “Long Wars and Long Telegrams: Containing Al-Qaeda,” International Affairs 85:2 (2009), 300–305. 117. Counter-narrative is usually applied to deterring violent radicalization, recruitment, and popu lar support for militancy. Paul Kamolnick argues that by using an “Islamically rooted legal case” al Qaeda’s constituency might be deterred from

Notes to Pages 69–73

219

joining it or be compelled to leave it. Paul Kamolnick, “Delegitimizing al-Qaeda: A Jihad-Realist Approach,” Strategic Studies Institute, U.S. Army War College (2012), 6, 20. Michael Jacobson, “Learning Counter-Narrative Lessons from Cases of Terrorist Dropouts,” in Countering Violent Extremist Narratives, National Coordinator for Counterterrorism (NCTb), Government of the Netherlands (2010), 73–79. 118. Long, “Strategic Culture, Al-Qaida, and Weapons of Mass Destruction,” 9. 119. Daniel Byman, The Five Front War: The Better Way to Fight Global Jihad (Hoboken, N.J.: Wiley, 2008), 40– 42. 120. Bernard Lewis, The Crisis of Islam: Holy War and Unholy Terror (New York: Random House, 2004), 39. 121. Ibid., 153 122. For the circumstances and controversy surrounding these events, see Jeremy Scahill, Dirty Wars: The World Is a Battlefield (New York: Nation Books, 2013), 37–38. 123. Giles Kepel, The War for Muslim Minds (Cambridge, Mass.: Harvard University Press, 2005), 100–104; Larson, “Exploiting al-Qa’ida’s Vulnerabilities,” 112–14; J.  Vahid Brown, “Cracks in the Foundation: Leadership Schisms in al- Qa’ida, 1986–2006,” Combating Terrorism Center (2006). 124. Lawrence Wright, “The Rebellion Within: An al Qaeda Mastermind Questions Terrorism,” New Yorker, June 2, 2008; Ronald Sandee, “Core al-Qaida in 2008: A Review,” NEFA Foundation (2009), 12–15; Ellen Knickmeyer, “Egyptian Extremist Rewriting Rationale for Armed Struggle,” Washington Post, July 15, 2007. 125. Asaf Maliach, “Saudi Religious Scholars Come Out Against Al-Qaeda’s Use of Religious Edicts Permitting Suicide Attacks Against Muslims,” International Institute for Counter-Terrorism (2007). 126. Peter Bergen and Paul Cruickshank, “The Unraveling: The Jihadist Revolt Against bin Laden,” New Republic, June 11, 2008. 127. Atran, “Moral Logic,” 142. 128. Bergen and Cruickshank, “The Unraveling”; Sandee, “Core Al-Qaida in 2008,” 11–12; Shaun Waterman, “Al Qaeda Adopting Defensive Tone,” Washington Times, August 13, 2008. 129. In conversation with Jerry Mark Long, July 2012. See also Long and Wilner, “Delegitimizing al-Qaida.” 130. Kim Cragin and Scott Gerwehr, Dissuading Terror: Strategic Influence and the Struggle Against Terrorism (Santa Monica, Calif.: RAND, 2005), xi–xii. 131. Steve Simon and Jeff Martini, “Terrorism: Denying al Qaeda Its Popu lar Support,” Washington Quarterly 28:1 (2004), 132. 132. Gerges, Journey of the Jihadist, 257– 63. 133. Ackerman and Pinson, I-VEO Empirical Assessment, 3–7; Sawyer and Pate, I-VEO Empirical Assessment. 134. See, for instance, Victor Le Vine and Barbara Salert, “Does a Coercive Official Response Deter Terrorism? The Case of the PLO,” Terrorism and Political Violence 8:1 (1996); Brophy-Baermann and Conybeare, “Retaliating Against Terrorism”; Bar,

220

Notes to Pages 73–77

“Deterring Nonstate Terrorist Groups”; Malka, “Israel and Asymmetrical Deterrence”; Bar, “Deterring Terrorists”; Benmelech, Berrebi, and Klor, “Counter-Suicide-Terrorism”; Shmuel Bar, “Deterrence of Palestinian Terrorism: The Israeli Experience,” in Wenger and Wilner, Deterring Terrorism; Rid, “Deterrence Beyond the State,” 128–30, 137– 42. Other punishment case studies focus on Afghanistan, Chechnya, Libya, Sudan, Iraq, Iran, and elsewhere. See Wilner, “Targeted Killings in Afghanistan”; Cohen, “Mission Impossible?”; Lyall, “Does Indiscriminate Violence Incite Insurgent Attacks?,” 331– 62; Crenshaw, “Coercive Diplomacy,” 315– 42; John Nevin, “Retaliating Against Terrorists,” Behavior and Social Issues 12 (2003). A number of other authors empirically evaluate the effects of retaliating against terrorists and their state sponsors but measure the counter-capability (rather than coercive) effects of retaliation. Strictly speaking, these are not studies of deterrence but rather of offensive counterterrorism. See, for instance, Walter Enders and Todd Sandler, “The Effectiveness of Antiterrorism Policies: A Vector-Autoregression-Intervention Analysis,” American Political Science Review 87:4 1993; Peter Chalk, “The Response to Terrorism as a Threat to Liberal Democracy,” Australian Journal of Politics and History 44:3 (1998); Gary Lafree, Laura Dugan, and Raven Korte, “The Impact of British Counterterrorist Strategies on Political Violence in Northern Ireland: Comparing Deterrence and Backlash Models,” Criminology 47:1 (2009). 135. Anthony, “Deterrence and the 9–11 Terrorists”; Trager and Zagorcheva, “Deterring Terrorism”; Geipel, “Urban Terrorists”; Wehling, “A Toxic Cloud of Mystery”; Payne et al., “Deterrence and Coercion of Non-State Actors,” 15, 22–24; David Romano, “Turkish and Iranian Efforts to Deter Kurdish Insurgent Attacks,” in Wenger and Wilner, Deterring Terrorism; Long and Wilner, “Delegitimizing al-Qaida.”

Chapter 4 1. Steven David, “Fatal Choices: Israel’s Policy of Targeted Killing,” Mideast Security and Policy Studies 51 (2002), 2. 2. Parts of this chapter first appeared in Alex S. Wilner, “Targeted Killings in Afghanistan: Measuring Coercion and Deterrence in Counterterrorism and Counterinsurgency,” Studies in Conflict and Terrorism 33:4 (2010). 3. Nils Melzer, Targeted Killing in International Law (Oxford: Oxford University Press, 2008), 3– 6. 4. “Pakistan: U.S. Drones Kill 18 Suspected Militants,” CBS/Associated Press, August 24, 2012; “U.S. Drone Attacks Kill 17 Militants in Pakistan,” New York Times, October 2, 2010; Nasser Arrabyee, “Qaeda Leader Reported Dead in Yemen Attack,” New York Times, October 18, 2012; Aamir Iqbal, “Drone Strike in Pakistan Kills 4 Suspected Militants,” CNN, September 24, 2012; Mark Mazzetti and Eric Schmitt, “U.S. Changes Tactics in Pakistan,” New York Times, October 28, 2008.

Notes to Pages 77–80

221

5. P. W. Singer, Wired for War: The Robotics Revolution and Conflict in the 21st Century (New York: Penguin, 2009), 35. 6. New America Foundation, Drone Wars Pakistan: Analysis, http://counterter rorism.newamerica.net/drones (accessed March 2014); Greg Miller, “Increased U.S. Drone Strikes in Pakistan Killing Few High-Value Militants,” Washington Post, February 21, 2011; Peter Bergen and Katherine Tiedemann, “Washington’s Phantom War,” Foreign Affairs 90:4 (July–August 2011), 12–13. 7. Jack Goldsmith, “Fire When Ready,” Foreign Policy, March 19, 2012; Washington Post–ABC News Poll, February 1– 4, 2012, www.washingtonpost.com/wp-srv/poli tics/polls/postabcpoll_020412.html (accessed March 2014). 8. “Special Report: The Permanent War,” Washington Post, October 2012, www .washingtonpost.com/world/national-security/plan-for-hunting-terrorists-signals-us -intends-to-keep-adding-names-to-kill-lists/2012/10/23/4789b2ae-18b3-11e2-a55c-39 408fbe6a4b_story.html (accessed March 2014). 9. Human Rights Watch, “U.S.: End CIA Drone Attacks,” December 19, 2011; Human Rights Watch, “Ban ‘Killer Robots’ Before It’s Too Late,” November 19, 2011. 10. Singer, Wired for War, 32–37; “Flight of the Drones,” Economist, October 8, 2011. 11. David, “Fatal Choices,” 2. 12. Amos Guiora, “Targeted Killing as Active Self-Defense,” War Crimes Research Symposium, Case Western Reserve University School of Law, October 2004, 330. See also Michael Walzer’s discussion of the subject in Amos Harel, “Professor Michael Walzer Discusses If and When Terrorists Should Be Assassinated,” Haaretz, May 18, 2011. 13. Alan Dershowitz, “The Rule of Proportionality,” New York Times, November 15, 2012. 14. Casper Weinberger, quoted in David, “Fatal Choices,” 15, 16. Brian Jenkins makes a similar assertion in “Assassination: Should We Stay the Good Guys?” Los Angeles Times, November 16, 1986. 15. David Silverstein, “Reviving the Assassination Option,” American Enterprise 12:8 (December 2001), 36; Richard Edwards, “Poison-Tip Umbrella Assassination of Georgi Markov Reinvestigated,” Telegraph, June 19, 2008; Duncan Campbell, “638 Ways to Kill Castro,” Guardian, August 3, 2006. 16. Saeed Kamali Dehghan and Julian Borger, “Ira nian Nuclear Chemist Killed by Motorbike Assassins,” Guardian, January 11, 2012; “Tensions Rise as Hillary Clinton Denies Any U.S. Role in Apparent Bomb Attack While Iran Calls for Strong UN Condemnation,” Al Jazeera, January 12, 2012. 17. Declassifying “kill lists,” Henning continues, would help “put enemies on notice.” Job Henning, “Embracing the Drone,” New York Times, February 20, 2012. 18. Samantha Shapiro, “The Year in Ideas: A to Z; Announced Assassinations,” New York Times Magazine, December 9, 2001.

222

Notes to Pages 80–83

19. Ignatius does not mention the suspected U.S. missile strike on an AQAP base in Ma’jalah, Yemen, on December 17, 2009. David Ignatius, “Drone War Strategies,” Real Clear Politics, October 5, 2011; Bill Roggio, “Evidence Presented of U.S. Involvement in 2009 Airstrike in Yemen,” Long War Journal, June 7, 2009. 20. Convention on the Prevention and Punishment of Crimes Against Internationally Protracted Persons, Including Diplomatic Agents, United Nations Treaty, December 14, 1973, 168; Bruce Berkowtiz, “Is Assassination an Option?” Hoover Digest 1 (2002). 21. Tim Weiner, “The Nation: Terminator; Making Rules in the World Between War and Peace,” New York Times, August 19, 2001. 22. Extrajudicial, Summary, or Arbitrary Executions: Report of the Special Rapporteur, UN Document E/CN.4/2003/3 (January 13, 2003), 16. 23. Gal Luft, “The Logic of Israel’s Targeted Killing,” Middle East Quarterly 10:1 (2003). 24. Chris Woods, “CIA Drone Strikes Violate Pakistan’s Sovereignty, Says Senior Diplomat,” Guardian, August 3, 2012. 25. Charter of the United Nations and Statute of the International Court of Justice (June 26, 1945), Chapter 7, Article 51. 26. UN General Assembly, Human Rights Council 14th Session, Report of the Special Rapporteur on Extrajudicial, Summary, or Arbitrary Execution, May 28, 2010, 12–13. 27. Law of Armed Conflict at the Operational and Tactical Levels, Department of National Defence, Government of Canada, 2001, secs. 6-4, 6-5. 28. Frank Harvey, “Is America Being Fair to Guantanamo Detainees?,” Globe and Mail, June 17, 2008. For U.S. views, see John Bellinger III (legal adviser, U.S. Department of State), “Legal Issues in the War on Terrorism,” London School of Economics (October 2006); Harold Hongju Koh (legal adviser, U.S. Department of State), “The Obama Administration and International Law,” Annual Meeting of the American Society of International Law, Washington, D.C. (March 2010); Robert Cassidy, Counterinsurgency and the Global War on Terror: Military Culture and Irregular War (Stanford, Calif.: Stanford University Press, 2008), 9–11; David Tucker, “Fighting Barbarians,” Parameters 28 (1998). 29. Quoted in Weiner, “The Nation.” For more on Israel’s evolving targeted-killings policy, see Avery Plaw, Targeting Terrorists: A License to Kill? (Burlington, Vt.: Ashgate, 2008), esp. chaps. 2 and 3; Shmuel Bar, “Deterrence of Palestinian Terrorism: The Israeli Experience,” in Deterring Terrorism: Theory and Practice, ed. Andreas Wenger and Alex Wilner (Stanford, Calif.: Stanford University Press, 2012); Daniel Byman, A High Price: The Triumphs and Failures of Israeli Counterterrorism (Oxford: Oxford University Press, 2011), esp. sec. 5. 30. John Norton Moore, quoted in Steven David, “Israel’s Policy of Targeted Killing,” Ethics and International Affairs 17 (2003). Professor Louis René Beres places terrorists in the same category as the pirates of the eighteenth century, labeling them

Notes to Pages 83–85

223

“hostes humani generis, common enemies of human kind.” Both groups are international outlaws who fall within the scope of “universal jurisdiction,” their action punishable by any nation or state. Louis René Beres, “Assassination of Terrorists May Be Law-Enforcing,” www.freeman.org/m_online/nov97/beres2.htm (accessed March 2014); Louis René Beres, “International Law and the Killing of Imad Mughniyeh,” Israel Journal of Foreign Affairs 2:2 (2008), 82. 31. Catherine Lotrionte, “When to Target Leaders,” Washington Quarterly 26:3 (2003), 73. 32. See Silverstein, “Reviving the Assassination Option,” 36–37. 33. Interim Report: Alleged Assassination Plots Involving Foreign Leaders, United States Senate, November 1975; Executive Order 11905, United States Foreign Intelligence Activities, February 1976. 34. Indeed, the raid itself was “consciously” conducted in a manner “that made Qadhafi’s death possible.” Tim Zimmerman, “Coercive Diplomacy and Libya,” in The Limits of Coercive Diplomacy, 2nd ed., ed. Alexander George and William Simons (Boulder, Colo.: Westview Press, 1994), 203–4; Daniel Schorr, “Stop Winking at ‘The Ban,’ ” Christian Science Monitor, September 21, 2001. 35. Schorr, “Stop Winking.” See also George Stephanopoulos, “Why We Should Kill Saddam,” Newsweek, December 1, 1999. 36. Joel Greenberg, “Israel Affirms Policy of Assassinating Militants,” New York Times, July 5, 2001. 37. U.S. secretary of defense Leon Panetta, “The Fight Against al Qaeda: Today and Tomorrow,” Speech Before the Center for a New American Security, Washington, D.C., November 20, 2012. 38. A slew of popular books and articles have been written about the subject: Daniel Klaidman, Kill or Capture: The War on Terror and the Soul of the Obama Presidency (New York: Houghton Mifflin Harcourt, 2012); Kevin McGrath, Confronting Al Qaeda (Annapolis, Md.: Naval Institute Press, 2011), esp. chaps. 2 and 7; Jack Goldsmith, Power and Constraint: The Accountable Presidency After 9/11 (New York: W. W. Norton, 2012), esp. pt. 2; Eric Schmitt and Thom Shanker, Counterstrike: The Untold Story of America’s Secret Campaign Against Al Qaeda (New York: Times Books, 2011); Mark Mazzetti, The Way of the Knife (New York: Penguin, 2013); Jeremy Scahill, Dirty Wars (New York: Nation Books, 2013); Linda Robinson, One Hundred Victories (New York: PublicAffairs, 2013); Jack Goldsmith, “Fire When Ready,” Foreign Policy, March 19, 2012; Peter Bergen and Katherine Tiedemann, “The Hidden War,” Foreign Policy, December 21, 2010. 39. Joint Resolution, 107th United States Congress, September 14, 2001; Josh Meyer, “CIA Expands Use of Drones in Terror War,” Los Angeles Times, January 29, 2006; Barton Gellman, “CIA Weighs ‘Targeted Killing’ Missions: Administration Believes Restraints Do Not Bar Singling Out Individual Terrorists,” Washington Post Sunday, October 28, 2001; Dana Priest, “Covert U.S. Hit Teams Try to Kill Saddam,” Washington Post, March 29, 2003.

224

Notes to Pages 85–88

40. Greg Miller, “Under Obama, an Emerging Global Apparatus for Drone Killing,” Washington Post, December 28, 2011; Greg Miller, “Plan for Hunting Terrorists Signals U.S. Intends to Keep Adding Names to Kill Lists,” Washington Post, October 24, 2012; Greg Miller and Julie Tate, “CIA Shifts Focus to Killing Targets,” Washington Post, September 2, 2011; Jane Mayer, “The Predator War,” New Yorker, October 26, 2009; Karen DeYoung, “A CIA Veteran Transforms U.S. Counterterrorism Policy,” Washington Post, October 25, 2012. 41. U.S. Department of Justice, “Attorney General Eric Holder Speaks at Northwestern University School of Law,” Chicago, Ill., March 5, 2012 42. Another American, Ahmed Hijazi, was killed in a 2002 drone strike in Yemen—the first such strike outside Afghanistan—though his death did not spark nearly as much consternation. Amy Davidson, “An American Teenager in Yemen,” New Yorker, October 18, 2011; Mark Mazzetti, Eric Schmitt, and Robert Worth, “TwoYear Manhunt Led to Killing of Awlaki in Yemen,” New York Times, September 30, 2011; Glenn Greenwald, “The Killing of Awlaki’s 16-Year-Old Son,” Salon, October 20, 2011; Peter Finn and Greg Miller, “Anwar al-Awlaki’s Family Speaks Out Against His Son’s Death in Airstrike,” Washington Post, October 18, 2011. 43. Alan Dershowitz, Why Terrorism Works: Understanding the Threat, Responding to the Challenge (New Haven, Conn.: Yale University Press, 2002), 120. 44. Naomi Chazan, “Assassinations as Weapons of War,” Israel Journal of Foreign Affairs 2:2 (2008), 87. For other normative research, see, in particular, Claire Finkelstein, Jens David Ohlin, and Andrew Altman (eds.), Targeted Killings: Law and Morality in an Asymmetrical World (Oxford: Oxford University Press, 2012); Asa Kasher and Amos Yadlin, “Military Ethics of Fighting Terror: Principles,” Philosophia 34 (2006); Paul Robinson, “The Ethics of the Strong Against the Tactics of the Weak,” Philosophia 36 (2008); Yael Stein, “By Any Name Illegal and Immoral,” Ethics and International Affairs 17:1 (2003); Michael Gross, Moral Dilemmas of Modern War (Cambridge: Cambridge University Press, 2010), chap. 5. 45. Zeev Schiff, “Outgoing Shin Bet Deputy Chief Weighs Complex Issues in Fighting Terrorism,” Haaretz, September 10, 2003; Uri Blau, “License to Kill,” Haaretz, November 27, 2008. 46. Amos Guiora, “Drone Strikes: What the U.S. Could Learn from Israel,” Los Angeles Times, September 29, 2011; Amos Guiora and Laurie Blank, “Targeted Killing’s ‘Flexibility’ Doctrine That Enables U.S. to Flout the Law of War,” Guardian, August 20, 2012; Guiora, “Targeted Killing as Active Self-Defense,” 325–30; Amos Guiora, “License to Kill,” Foreign Policy, July 13, 2009; Amos Guiora, “The Importance of CriteriaBased Reasoning in Targeted Killing Decisions,” in Finkelstein, Ohlin, and Altman, Targeted Killings. 47. Peter Baker, “In Terror Shift, Obama Took a Long Path,” New York Times, May 27, 2013; Declan Walsh, “Drone Issue Hovers More Than Ever, Even as Strikes Ebb,” New York Times, October 24, 2013; Greg Miller et al., “Documents Reveal

Notes to Pages 88–94

225

NSA’s Extensive Involvement in Targeted Killing Program,” Washington Post, October 16, 2013. 48. Yossi Melman, “Targeted Killings: A Retro Fashion Very Much in Vogue,” Haaretz, March 24, 2004. 49. Associated Press, “Israel Confi rms Killing Arafat Deputy in 1988,” November 1, 2012. 50. In the initial wave, Israel targeted individuals from a number of different organizations, including nineteen Hamas operatives, nine from PIJ, twelve from Fatah’s Al-Aqsa Martyrs Brigade, and two from the Popu lar Front for the Liberation of Palestine (PFLP). Luft, “The Logic of Israel’s Targeted Killing”; Peter Cullen, “The Role of Targeted Killing in the Campaign Against Terror,” Joint Force Quarterly, 48 (2008). 51. David, “Fatal Choices,” 5. 52. Dexter Filkins, “After Syria: Hezbollah Looks to a Future Without Assad,” New Yorker, February 25, 2013, 54. 53. Israel clandestinely bombed Sudan on two other occasions, destroying an Iranian weapons convey near Port Sudan in 2009 and a munitions factory near Khartoum in 2012. Avi Issacharoff, “Report: Hamas Arms Trafficker Killed in Sudan Air Strike,” Haaretz, April 6, 2011. 54. Ibrahim Barzak, “Ahmed Jabari, Hamas Military Chief, Topped Israel’s Wanted List,” Associated Press, November 14, 2012; “Get Jabari: Israel’s Bid to Nab the Hamas Chief Who Oversaw Shalit’s Capture,” Times of Israel, July 26, 2013. 55. Abdel Bari Atwan, The Secret History of al Qaeda (Berkeley: University of California, 2006), 43– 45. 56. The cases listed here (and the others that are not) can be reviewed online at a number of different institutional sites, including Th e Long War Journal (www .longwarjournal.org/pakistan-strikes-hvts.php), the New America Foundation (http:// natsec.newamerica.net/), and The Bureau of Investigative Journalism (www.thebureau investigates.com/category/projects/drones/). NATO’s International Security Assistance Force (ISAF) also keeps track of militant leaders killed in Afghanistan (www.isaf.nato .int/). 57. See Stanford University’s Mapping Militant Organizations project: “Tehreeke-Taliban,” (http://www.stanford.edu/group/mappingmilitants/cgi-bin/groups/view /105) (accessed March 2014); “Hakimullah Mehsud Killed by Drone, Pakistan Taliban Say,” BBC News, November 1, 2013. 58. Mark Hosenball, “U.S. Now Believes Qaeda Leader Kashmiri Is Dead,” Reuters, July 7, 2011; Bill Roggio, “Al Qaeda Leader Ilyas Kashmiri Spotted at Taliban Meeting,” Long War Journal, March 7, 2012. 59. He and his associates apparently “got lost” driving around Mogadishu and accidentally arrived at the checkpoint, which culminated in the lethal shootout. Jeff rey Gettleman, “Somalis Kill Mastermind of Two U.S. Embassy Bombings,” New York Times, June 22, 2011.

226

Notes to Pages 94–99

60. “A Very British Execution?,” Economist, January 25, 2012; Michael Burleigh and Sarah Rainey, “The 7/7 Widow and a Boom in British Jihad,” Telegraph, March 3, 2012. 61. Bill Roggio, “Al Qaeda Leader Rashid Rauf Killed in Drone Strike, Family Says,” Long War Journal, October 29, 2012. 62. “Briton Suspected of Plotting Mumbai-Style Attacks Killed by Drone,” Telegraph, September 30, 2010; “Drone Attacks ‘Linked’ to Suspected Europe Terror Plot,” BBC, October 6, 2010; Tom Peter, “Al Qaeda Plot in Europe Possibly Revealed by German Terror Suspect,” Christian Science Monitor, September 29, 2010; Mark Mazzetti and Souad Mekhennet, “Drones Kill Westerners in Pakistan,” New York Times, October 4, 2010. 63. Christopher Harmon, “The Myth of the Invincible Terrorist”, Policy Review 142 (2007), 59. 64. Isaac Ben-Israel, Oren Setter, and Asher Tishler, “R&D and the War on Terrorism: Generalizing the Israeli Experience,” in Science and Technology Policies for the Anti-Terrorism Era, ed. A. James (Amsterdam: IOS Press, 2006), 54. For counterarguments, see Or Honig, “Explaining Israel’s Misuse of Strategic Assassinations,” Studies in Conflict and Terrorism 30 (2007), 571–72. 65. Melman, “Targeted Killings,” 4. 66. Lennox Samuels, “Al Qaeda Nostra,” Newsweek, May 20, 2008; Abdul Hameed Bakier, “A Profile of al-Qaeda’s New Leader in Iraq,” in Unmasking Terror, ed. Jonathan Hutzley (Washington, D.C.: Jamestown Foundation, 2007); Michael Schmidt and Eric Schmitt, “Leaving Iraq, U.S. Fears New Surge of Qaeda Terror,” New York Times, November 5, 2011; Qassim Abdul-Zahra and Lara Jakes “Al-Qaeda Rebounding in Iraq, Officials Say,” Associated Press, October 10, 2012. 67. U.S. Department of State, Rewards for Justice, www.rewardsforjustice.net /index.cfm?page=Wanted_Terrorist&language=english (accessed April 2014). 68. National Intelligence Estimate, Trends in Global Terrorism: Implications for the United States (September 2006), 2– 4. 69. “Madrid ‘Ringmaster’ Dies in Blast,” BBC, April 2, 2004. 70. “Zarkawi’s Cry,” National Review Online, February 12, 2004. 71. Ari Shavit, “Sharon Is Sharon Is Sharon,” Haaretz, April 12, 2001. 72. Jamestown Foundation, “Pakistani Taliban Commander Describes CounterMeasures Against UAV Attacks,” Terrorism Monitor 7:31 (2009), 1–2; Michael Evans, “Death from Above: How Predator Is Taking Its Toll on al- Qaeda,” Times, January 3, 2009. 73. Luft, “The Logic of Israel’s Targeted Killing.” See also Hillel Halkin, “Israelis Must Kill or Be Killed,” Wall Street Journal, August 28, 2001. 74. Relations suffered as a result of the bin Laden raid in May 2011, bungled American-NATO air strikes in November 2011 that killed twenty-five Pakistani soldiers, and the January 2012 arrest of CIA agent Raymond Davis in Lahore. Eric Schmitt, “Lull in Strikes by U.S. Drones Aids Militants in Pakistan,” New York Times,

Notes to Pages 99–104

227

January 7, 2012; Salman Masood and Eric Schmitt, “Tensions Flare Between U.S. and Pakistan After Strike,” New York Times, November 26, 2011. 75. Eric Schmitt and Jane Perlez, “Attacks by U.S. Drones Radicalizing Al Qaeda,” New York Times, February 25, 2009; Greg Miller and Julian Barnes, “Higher-Tech Predators Targeting Pakistan,” Los Angeles Times, September 12, 2008. 76. Greg Miller, Julie Tate, and Barton Gellman, “Documents Reveal NSA’s Extensive Involvement in Targeted Killing Program,” Washington Post, October 16, 2013. 77. Colin Freeze, “Khawaja Lived for Jihad, Crown Alleges,” Globe and Mail, June 24, 2008; Christie Blatchford, “ ‘Down with the J,’ and Out of Their Senses,” Globe and Mail, June 24, 2008. 78. Harmon, “The Myth of the Invincible Terrorist,” 60. 79. Keith Payne et al., “Deterrence and Coercion of Non-State Actors: Analysis of Case Studies,” National Institute for Public Policy (2008), 20. 80. Patrick Porter, “Long Wars and Long Telegraphs: Containing al-Qaeda,” International Affairs 85:2 (2009), 301. 81. Brian Glyn Williams, “Mullah Omar’s Missiles,” Middle East Policy 15:4 (2008), 38– 42. 82. “Would-Be Plane Bomber Has Sentence Cut,” BBC News, April 16, 2012. 83. Honig disagrees, writing that “terrorists are easily replaced and technical knowledge can be too easily disseminate.” In his view, the “barrel” of would-be Palestinian terrorists, for instance, is infinite; Israel cannot kill its way out of it. Still others, however, disagree with Honig, suggesting instead that groups do not have a “bottomless pit” of talent from which to pull. Hillel Frisch explains that targeted killings diminish a group’s capability and motivation to conduct certain acts of terrorism. Ben-Israel, Setter, and Tishler, “R&D,” 54; Honig “Explaining Israel’s Misuse of Strategic Assassinations,” 571; Hillel Frisch, “Motivation or Capabilities? Israeli Counterterrorism Against Palestinian Suicide Bombings and Violence,” Mideast Security and Policy Studies, no. 70 (Begin-Sadat Center for Strategic Studies), 1–5; Bar, “Deterrence of Palestinian Terrorism,” 214–15. 84. United Nations, Situation in Afghanistan and Its Implication for International Peace and Security, Report of the Secretary-General, September 21, 2007, 2. 85. Daniel Byman, “Do Targeted Killings Work?”, Foreign Affairs 85:2 (2006), 103. 86. Michael Eisenstadt, “Preemptive Targeted Killings as a Counterterror Tool,” PeaceWatch, no. 342, August 2001. 87. Carlotta Gall, “Signs of Strain as Taliban Gird for More Fighting,” New York Times, March 30, 2011; Carlotta Gall, “Midlevel Taliban Admit to a Rift with Top Leaders,” New York Times, February 21, 2011. 88. Duane Clarridge, quoted in Elena Mastors and Jeffrey Norwitz, “Disrupting and Influencing Leaders of Armed Groups,” in Armed Groups: Studies in National Security, Counterterrorism, and Counterinsurgency ed. Jeffrey Norwitz (Newport, R.I.: Naval War College, 2008), 328–29. 89. Harmon, “The Myth of the Invincible Terrorist,” 62

228

Notes to Pages 104–109

90. Syed Saleem Shahzad (trans. Donald Hounam), “Al-Qaida: The Unwanted Guests,” Le Monde Diplomatique, July 2007; “Pakistan Rebel Clash ‘Kills 100,’ ” BBC, March 21, 2007; Jane Perlez and Eric Schmitt, “Pakistan’s Spies Tied to Slaying of a Journalist,” New York Times, July 4, 2011. 91. Anand Gopal, “Serious Leadership Rifts Emerge in Afghan Taliban,” CTC Sentinel 5:11–12 (November 2012), 9–11. 92. Maria Golovnina and Asim Tanveer, “Leader’s Death Plunges Pakistan Taliban into Dangerous Disarray,” Reuters, November 14, 2013; Declan Walsh, “Pakistani Taliban Pick Hard-Liner as Leader, Imperiling Proposed Peace Talks,” New York Times, November 7, 2013; “With Fazlullah Away, Khalid Haqqani to Oversee TTP Operations,” Express Tribune, November 15, 2013. 93. Avi Issacharoff, “In the West Bank, Wanted Militants Are Made to Feel Unwanted,” Haaretz, November 13, 2007. 94. “Three Slain Men Discovered in Tribal Region,” International Herald Tribune, October 4, 2010. 95. Graeme Smith, The Dogs Are Eating Them Now: Our War in Afghanistan (Toronto: Alfred A. Knopf, 2013), 100; “Jihadist Video Shows Boy Killing Pakistani Militant,” Associated Press, April 21, 2007; Michael Smith, “Taliban Leader ‘Killed’ After RAF Tracks Phone,” Times, December 25, 2006. 96. Luft, “The Logic of Israel’s Targeted Killing.” 97. Jo Becker and Scott Shane, “Secret Kill List Proves a Test of Obama’s Principles and Will,” New York Times, May 29, 2012. 98. David Kilcullen and Andrew McDonald Exum, “Death from Above, Outrage Down Below,” New York Times, May 17, 2009. 99. “Pakistan Rally Against U.S. Strike,” BBC, January 15, 2006. 100. International Human Rights and Conflict Resolution Clinic (Stanford University) and Global Justice Clinic (New York University), Living Under Drones (2012); Gregory Johnson, “The Wrong Man for the CIA,” New York Times, November 19, 2012; Bureau of Investigative Journalism, Covert Drone War, www.thebureauinvestigates .com/category/projects/drones/ (accessed April 2014); Sudarsan Raghavan, “In Yemen, U.S. Airstrikes Breed Anger, and Sympathy for al-Qaeda,” Washington Post, May 30, 2012; Amnesty International, “Will I Be Next?”: U.S. Drone Strikes in Pakistan, October 2013; Human Rights Watch, Between a Drone and al-Qaeda: The Civilian Costs of U.S. Targeted Killings in Yemen, October 2013; Mark Mazzetti, “Drone Strikes May Work Against U.S. in Pakistan,” New York Times, March 23, 2009. 101. Mathew Fricker, Avery Plaw, and Brian Glyn Williams, “New Light on the Accuracy of the CIA’s Predator Drone Campaign in Pakistan,” Terrorism Monitor 8:41 (2010); New America Foundation, Drone Wars Pakistan: Analysis, http://counter terrorism.newamerica.net/drones (accessed April 2014); Bill Roggio and Alexander Mayer, “Charting the Data for U.S. Airstrikes in Pakistan, 2004–2012,” Long War Journal, www.longwarjournal.org /pakistan-strikes.php (accessed April 2014); Bill Roggio

Notes to Pages 109–111

229

and Bob Barry, “Charting the Data for US Airstrikes in Yemen, 2004–2014,” Long War Journal, http://www.longwarjournal.org /multimedia/Yemen/code/Yemen-strike.php (accessed October 2013); Meg Braun, “Counting Civilian Casualties in CIA’s Drone War,” Foreign Policy, November 2, 2012; Peter Bergen, “Washington’s Phantom War,” Foreign Affairs 90:4 (2011), 12–18; Brian Glyn Williams, “Accuracy of the U.S. Drone Campaign: The Views of a Pakistani General,” CTC Sentinel 4:3 (2011); Joshua Foust, “Targeted Killing, Pro and Con: What to Make of U.S. Drone Strikes in Pakistan,” Atlantic, September 26, 2012; Sebastian Abbot, “New Light on Drone War’s Death Toll,” Associated Press, February 25, 2012; Declan Walsh, “In a Surprise, Pakistan Says Fewer Civilians Died by Drones,” New York Times, October 30, 2013. 102. Mia Bloom, Dying to Kill: The Allure of Suicide Terrorism (New York: Columbia University Press, 2005; 2007 edition), 93. 103. Others, like Mirza Jan and colleagues, find that most Pakistanis living in Waziristan also condemn “drone attacks of NATO forces in Waziristan,” notwithstanding the fact that it is the United States and not NATO that controls the drones flying over Pakistan. Mirza Jan, “Military Operations in Waziristan: Public Perceptions in Pakistan on Terra,” Gomal University Journal of Research 27:1 (2011), 105; Clay Ramsay et al., “Pakistani Public Opinion on Swat Conflict, Afghanistan, and the U.S.,” World Public Opinion.Org (Program on International Policy Attitudes, University of Maryland), 8. 104. Aliya Robin Deri, “ ‘Costless’ War: American and Pakistani Reactions to the U.S. Drone War,” Intersect 5 (2012), 10. 105. Megan Smith and James Igoe Walsh, “Do Drone Strikes Degrade al Qaeda? Evidence from Propaganda Output,” Terrorism and Political Violence 25 (2013). 106. Brian Williams, “The CIA’s Covert Predator Drone War in Pakistan, 2004– 2010: The History of an Assassination Campaign,” Studies in Conflict and Terrorism 33 (2010), 883, 886. See also Farhat Taj, “CIA Drone Strikes in Pakistan’s FATA Region and the ‘Loss’ of Actionable Intelligence: A Pashtun Perspective,” Terrorism Monitor 8:15 (2010), 6. 107. New America Foundation, Public Opinion in Pakistan’s Tribal Regions, September 2010, http://newamerica.net/publications/policy/public_opinion_in_pakistan_s _tribal_regions (accessed March 2014). 108. Craig Whitlock and Mohammed al-Qadhi, “Al-Qaeda Fugitive in Yemen Gets Away Again,” Washington Post, July 16, 2011. 109. Leila Hudson, Colin Owens, and Matt Flannes, “Drone Warfare: Blowback from the New American Way of War,” Middle East Policy 18:3 (2011), 128–30. 110. Owen Bowcott, “Drone Strikes Threaten 50 Years of International Law, Says UN Rapporteur,” Guardian, June 21, 2012. 111. Michael Savage, “Challenge to Britain’s Spry Role in Drone War,” Times, September 18, 2012; Terri Judd, “British Intelligence Officers Accused of Murder over Drone Attacks in Pakistan,” Independent, October 24, 2012.

230

Notes to Pages 111–116

112. David, “Fatal Choices,” 14. 113. Dana Khraiche, “Nasrallah: Some Christian Parties Seek Inter-Muslim Strife,” Daily Star (Lebanon), November 12, 2012; Ashley Fantz, “Hezbollah Claims It Sent Drone over Israel, But Expert Calls It ‘Rinky-Dink,’ ” CNN, October 12, 2012. 114. Charlie Savage, “U.S. Should Limit Attacks by Drones, UN Asserts,” New York Times, June 3, 2010. See also Ward Thomas, “Norms and Security: The Case of International Assassination,” International Security 25:1 (2006). 115. In his statistical study of al Qaeda’s retaliation, however, Daniel Hepworth finds no significant link between targeted killings and revenge attacks. Hepworth, “Terrorist Retaliation? An Analysis of Terrorist Attacks Following the Targeted Killing of Top-Tier al Qaeda Leadership,” Journal of Policing, Intelligence and Counter Terrorism 9:1 (2014). 116. Chazan, “Assassinations as Weapons of War,” 86. 117. Hudson, Owens, and Flannes, “Drone Warfare,” 126 118. Eisenstadt, “Preemptive Targeted Killings,” 2. 119. Michael Oren, before becoming Israeli ambassador to the United States, suggested otherwise in a 2006 Wall Street Journal special feature: “[D]amage to Israel’s image will likely be temporary. Who today remembers Abdel Aziz Ranitisi and Sheikh Yassin?” Michael Oren, “Stop Terror at Its Source,” Wall Street Journal, June 28, 2006. See also David, “Fatal Choices,” 11. 120. Jerrold Post, Ehud Sprinzak, and Laurita Denny, “The Terrorists in Their Own Words: Interviews with 35 Incarcerated Middle Eastern Terrorists,” Terrorism and Political Violence 15:1 (2003), 178. 121. Bruce Jentleson and Christopher Whytock, “Who ‘Won’ Libya? The ForceDiplomacy Debate and Its Implications for Theory and Policy,” International Security 30:3 (2005– 6). 122. David, “Fatal Choices,” 10. See also Honig, “Explaining Israel’s Misuse of Strategic Assassinations,” 564– 65. 123. Marisa Porges, “Dead Men Share No Secrets,” New York Times, September 24, 2012; Marc Thiessen, “Dead Terrorists Tell No Tales,” Foreign Policy, February 8, 2010; Armin Krishnan, “Targeting Individuals: Overcoming the Dilemmas of Secrecy,” Contemporary Security Policy 34:2 (2013). 124. See George Tenet’s enlightening account in At the Center of the Storm (New York City: Harper, 2008), 250–53 125. Mastors and Norwitz, “Disrupting and Influencing Leaders,” 324, 328–38. 126. For an overview and a critique of the empirical literature on targeted killings, see Stephanie Carvin, “The Trouble with Targeted Killings,” Security Studies 21:3 (2012), 528–55. 127. Barton Gellman, “CIA Weighs ‘Targeted Killing’ Missions,” Washington Post, October 28, 2001; Yoni Dayan, “Mowing the Lawn,” Yediot Ahronot (Ynet), November 27, 2012; Efraim Inbar and Eitan Shamir, “Mowing the Grass: Israel’s Strategy for Protracted Intractable Conflict,” Journal of Strategic Studies, 2013.

Notes to Pages 116–120

231

128. Byman, “Do Targeted Killings Work?,” 100. 129. Parts of Jenna Jordan’s work focus on leadership charisma: “When Heads Roll: Assessing the Effectiveness of Leadership Decapitation,” Security Studies 18 (2009). Austin Long, “Assessing the Success of Leadership Targeting,” CTC Sentinel 3:11–12 (2010), 19–21. 130. Michael Freeman, “A Theory of Terrorist Leadership (and Its Consequences for Leadership Targeting),” Terrorism and Political Violence 26:3 (2014). 131. Audrey Kurth Cronin, How Terrorism Ends (Princeton, N.J.: Princeton University Press, 2009), esp. chap. 1. 132. Mohammed Hafez and Joseph Hatfield, “Do Targeted Killings Work? A Multivariate Analysis of Israel’s Controversial Tactic During Al-Aqsa Uprising,” Studies in Conflict and Terrorism 29 (2006), 361– 63, 367–79. 133. Edward Kaplan, Alex Mintz, Shaul Mishal, and Claudio Samban, “What Happened to Suicide Bombings in Israel?,” Studies in Conflict and Terrorism 28 (2005), 232. Other scholars have used Israeli stock-market data to gauge public reactions and perceptions of success and failure to Israeli targeted killings. Asaf Zussman and Noam Zussman, “Assassinations: Evaluating the Effectiveness of an Israeli Counterterrorism Policy Using Stock Market Data,” Journal of Economic Perspectives 20:2 (2006). See also David Jaeger and M. Daniele Paserman, “The Shape of Things to Come? Assessing the Effectiveness of Suicide Attacks and Targeted Killings,” Institute of the Study of Labor (IZA, Bonn, Germany), Discussion Paper 2890 (2007). 134. Jordan, “When Heads Roll,” 719–55; Robert Wright, “The Price of Assassination,” New York Times, April 13, 2010. 135. Th is study does not include targeted killings of Islamist militant leaders from the post-9/11 era. Patrick Johnston, “Does Decapitation Work?,” International Security 36:4 (2012), 75. 136. Bryan Price, “Targeting Top Terrorist: How Leadership Decapitation Contributes to Counterterrorism,” International Security 36:4 (2012), 9–46. Price’s and Johnston’s findings are (partially) substantiated by two other quantitative studies: Aaron Mannes, “Testing the Snake Head Strategy: Does Killing or Capturing Its Leaders Reduce a Terrorist Group’s Activity?,” Journal of International Policy Solutions 9 (2008); Dane Rowlands and Joshua Kilberg, “Organizational Structure and the Effects of Targeting Terrorist Leadership,” Centre for Security and Defence Studies, Carleton University (Canada), Working Paper 09 (2011). Jenna Jordan, following an earlier appearance with both Price and Johnston on an International Studies Association (ISA) annual convention panel, assesses both projects in “ISSF Article Review,” H-Diplo (2012). 137. William Casebeer and Troy Thomas, “Deterring Violent Non-State Actors in the New Millennium,” Strategic Insight 2:12 (2003), 4. Melman argues, instead, that overusing targeted killings “dissipates” the mystery and fear they produce if used sparingly. See Melman, “Targeted Killings.” 138. Robert Mackey, “Israel’s Military Begins Social Media Offensive as Bombs Fall on Gaza,” New York Times, November 14, 2012.

232

Notes to Pages 121–125

139. Annie Lubin, “IDF Warning to Hamas: Don’t Show Your Faces Above Ground,” Arutz Sheva (IsraelNationalNews.com), November 14, 2012; Neal Ungerleider, “Inside the Israeli Military’s Social Media Squad,” FastCompany.com, November 20, 2012. 140. Emily Greenhouse, “The Tweets of War,” New Yorker, November 19, 2012. 141. John Herrman, “Israel Defense Forces Social Media Head Defends Tweets,” BuzzFeed.com, November 17, 2012; Lauren Bohn, “Israel and Hamas Battle on Social Media as Well,” Associated Press, November 15, 2012.

Chapter 5 1. The others included Ustad Yasir and Muft i Latifullah Hamkimi, both onetime Taliban spokespersons, and two regional commanders, believed to have been Hafi z Hamdullad and Abdul Ghaffar. Parts of this chapter originally appeared in Alex Wilner, “Targeted Killings in Afghanistan: Measuring Coercion and Deterrence in Counterterrorism and Counterinsurgency,” Studies in Conflict and Terrorism 33:4 (2010). 2. Richard Owen and Tim Albone, “Protests over Taliban Exchange for Journalist,” Times, March 21, 2007; Declan Walsh, “Afghans Admit Doing Deal with Taliban to Free Italian Hostage,” Guardian, March 21, 2007; “Anger at Taliban Exchange for Journalist,” Australian, March 22, 2007. 3. Michael Smith, “SBS Behind Taliban Leader’s Death,” Sunday Times, May 27, 2007; “Special Boat Ser vice Operations—Hunting Mullah Dadullah,” Special Boat Ser vices, May 12, 2007, www.specialboatservice.co.uk /mullah-dadullah.htm; “Raid Kills Taliban Exchanged for Italian,” Reuters, May 16, 2007; Matthias Gebauer, “Taliban Leader Mullah Dadullah: The Star of Afghanistan’s Jihad,” Der Spiegel Online, March 1, 2007. 4. International Security Assistance Force (ISAF), Press Release No. 2007-370, “Mullah Dadullah Lang Killed in Security Operation,” May 13, 2007; Taimoor Shah and Carlotta Gall, “Key Taliban Leader Is Killed in Afghanistan in Joint Operation,” New York Times, May 14, 2007; “Taliban Military Commander Mullah Dadullah Killed,” Associated Press, May 13, 2007. 5. “U.S. Offers $200,000 to Catch ‘Most Wanted’ Taliban,” Associated Press, October 1, 2007; Alisa Tang, “Gen.: Attacks Down Sharply in E. Afghanistan,” Associated Press, February 24, 2008. 6. See New America Foundation, http://counterterrorism.newamerica.net/drones (accessed April 2014). For another set of statistics, see Bill Roggio and Alexander Mayer, “Charting the Data for U.S. Airstrikes in Pakistan, 2004–2014,” Long War Journal, www.longwarjournal.org/pakistan-strikes.php (accessed April 2014). 7. Jason Straziuso, “U.S. Offers Up to US$200,000 in New ‘Most Wanted’ Campaign,” Associated Press, October 2, 2007.

Notes to Pages 126–128

233

8. For information on the groups listed here, see Syed Saleem Shahzad, “The Taliban Talk the Talk,” Asia Times, April 11, 2008; Syed Saleem Shahzad, “Rise of the NeoTaliban (Part I),” Asia Times, November 13, 2007; Syed Saleem Shahzad, “Rise of the Neo-Taliban (Part II),” Asia Times, November 14, 2007; Antonio Giustozzi, “The Resurgence of the Neo-Taliban,” openDemocracy, December 14, 2007; Antonio Giustozzi, Koran, Kalashnikov, and Laptop: The Neo-Taliban Insurgency in Afghanistan (New York: Columbia University Press, 2008); Jason Burke, “The New Taliban,” Observer, October 14, 2007; Mark Mazzetti, Scott Shane, and Alissa Rubin, “Brutal Haqqani Crime Clan Bedev ils U.S. in Afghanistan,” New York Times, September 24, 2011; Jeff rey Dressler, “The Haqqani Network and the Threat to Afghanistan,” Foreign Affairs, November 11, 2011; Jamestown Foundation, “Leader of Uzbek Militants Calls for Jihad in Pakistan,” Terrorism Focus 5:5 (2008); “Tahir Yuldash: U.S. Fiasco Is Nearing. Look Us Up in Washington,” Ferghana News Agency, October 15, 2007; Rahimullah Yusufzai, “New Taliban Group Named After Tora Bora Setup,” ABC News, February 26, 2007; Muhammad Tahir, “Gulbuddin Hekmatyar’s Return to the Afghan Insurgency,” Terrorism Monitor 6:11 (May 28, 2008); Shaiq Hussain, “New Breed of Taliban Commanders Lined Up,” Nation, July 26, 2008; Sohail Abdul Nasir, “Baitullah Mehsud: South Waziristan’s Unofficial Amir,” Terrorism Focus 3:26 (July 5, 2006); Bill Roggio, “Pakistani Taliban Unites Under Baitullah Mehsud,” Long War Journal, December 15, 2007; “Al-Qaeda: A Hydra-Headed Monster,” Special Report, Economist, July 17, 2008. 9. See, for instance, David Rohde, “Foreign Fighters of Harsher Bent Bolster Taliban,” New York Times, October 30, 2007. 10. Mansoor Dadullah explained in a July 2007 interview that he had “120 commanders under” his control and that “each one of them ha[d] a certain number of subcommanders, who in turn each ha[d] a group of soldiers under them.” Urs Gehriger and Sami Yousafzai, “ ‘Ours Is a Global Struggle’: An Interview with Taliban Military Chief Mansoor Dadullah,” World Politics Review, July 3, 2007. 11. Christopher Dickey, “The Taliban’s Book of Rules,” Newsweek, October 15, 2007; Urs Gehriger, “Layeha (Regelbuch) fur dei Mudschaheddin,” Die Weltwoche, August 2007. 12. For details covering this period, see Graeme Smith, “The Taliban,” Globe and Mail, November 27, 2006; “Taliban Names Anti-U.S. Leadership Council,” Reuters, June 24, 2003; “Taliban Form ‘Resistance Force,’ ” CNN, June 24, 2003; Syed Saleem Shahzad, “Taliban’s New Commander Ready for a Fight,” Asia Times, May 20, 2006. Other analysts suggest that the Taliban’s shura is comprised of other individuals. Rahimullah Yusufzai, a veteran journalist based in Pakistan and resident editor of the News International, adds Abdur Razzaq Nafiz, Jalaluddin Haqqani, Saifur Rahman Mansoor, Mullah Mansoor, and Mullah Mohammad Rasul to the list. See Rahimullah Yusufzai, “Taliban Confirm Usmani’s Death in Afghan Air Strike,” News International, December 27, 2006. Asia Times’s Shahzad argues further that Haqqani, though not an original member of the Taliban, was appointed deputy chief in 2006 and is also a leading member of the shura. See Syed Saleem Shahzad, “Pakistan Gains from Taliban Split,”

234

Notes to Pages 128–131

Asia Times, May 9, 2007; Syed Saleem Shahzad, “U.S. Shooting in the Dark in Afghanistan,” Asia Times, June 28, 2003. For a historical breakdown of the Taliban leadership in southern and western Afghanistan in par ticu lar, see Scott Baldauf and Owais Tohid, “Taliban Appears to Be Regrouped and Well-Funded,” Christian Science Monitor, May 8, 2003, and Shahid Afsar, Chris Samples, and Thomas Wood, “The Taliban: An Organizational Analysis,” Military Review (May–June 2008), 58–65. For a list of the Taliban’s regional leadership, see Bill Roggio, “The Afghan Taliban’s Top Leaders,” Long War Journal (February 23, 2010); Graeme Smith, “The Canadians Try to Kill Everybody,” Globe and Mail, November 27, 2006. 13. Sami Yousafzai and Ron Moreau, “The Mysterious Mullah Omar,” Newsweek, February 25, 2007. 14. Sami Yousafzai and Ron Moreau, “While Pakistan Burns,” Newsweek, November 9, 2007; “Senior Afghan Taliban Leader, Mullah Obaidullah, Is Dead,” BBC News, February 13, 2012; “Afghan Taliban Demand Pakistan Explain Ex-Minister’s Death,” Daily Times, February 14, 2012. 15. Original reports suggest that Mullah Berader was killed in an air strike in August 2007 (“Top Taliban Commander Killed in Southern Afghanistan,” Radio Free Europe/Radio Liberty, August 30, 2007), though later reports suggested that he survived the attack (“Taliban Commander Says Reports of Death Premature,” News International, September 14, 2007). For details on Berader’s capture, see Mark Mazzetti and Dexter Filkins, “Secret Joint Raid Captures Taliban’s Top Commander,” New York Times, February 15, 2010. 16. Pir Zubair Shah and Dexter Filkins, “Pakistan Reports Capture of Taliban Leader,” New York Times, February 22, 2010. 17. “U.S. Offers Rewards for ‘Bad Guys’ in Afghanistan,” CNN, October 1, 2007; “U.S. Offers $200,000 to Catch ‘Most Wanted’ Taliban,” October 1, 2007; Matt Dupee, “Roundup: ‘Most Wanted’ Taliban Commander Killed, Others Targeted,” Afgha.com, January 27, 2008. For a critique of the program, see “Counter-Terror Bounties,” International Herald Tribune (with Oxford Analytica), February 25, 2008. 18. For details on these events, see “Tahir Yuldash,” Ferghana News, 2007; Bruce Pannier, “IMU Announces Longtime Leader Dead, Names Successor,” Radio Free Europe/Radio Liberty, August 17, 2010; Matt Dupee, “Unprecedented Coalition Strike Nails the Haqqani Network in North Waziristan, Long War Journal, March 13, 2008; Pir Zubair Shah, “Missile Kills Militant’s Brother in Pakistan,” New York Times, February 19, 2010; “Sources: Drone Strikes Kill 29 in Pakistan,” CNN, February 2, 2010; Bill Roggio, “Report: Strike Targets Baitullah Mehsud’s Hideout in Pakistan, Long War Journal, June 15, 2008; Declan Walsh, “Pakistan’s Top Taliban Leader Baitullah Mehsud Killed in U.S. Drone Attack,” Guardian, August 7, 2009; Bill Roggio, “ISAF Targets al Qaeda–Linked Taliban Commanders in the Afghan East,” Long War Journal, January 9, 2011; Mar Mazzetti, “CIA Secretly Held Qaeda Suspect, Officials Say,” New York Times, March 15, 2008; Bill Roggio, “Report: Osama bin Laden’s Security Coordinator Captured in Pakistan,” Long War Journal, January 6, 2008; Bill Roggio,

Notes to Pages 131–136

235

“Pakistan Frees al Qaeda Commander: Report,” Long War Journal, September 21, 2011; Bill Roggio, “Siraj Haqqani’s Deputy Killed in Afghanistan,” Weekly Standard, December 14, 2007; “Counter-Terror Bounties,” International Herald Tribune, December 5, 2008.; Bill Roggio, “Senior Leader of Haqqani Network Killed in Pakistan,” Long War Journal, January 26, 2008; Combined Joint Task Force—101, Press Release, “Coalition Forces Confirm Darim Sedgai Death,” January 26, 2008; “Report: Airstrike Killed Taliban Shadow Governor Sheikh Dost Mohammed,” Long War Journal, March 3, 2012. 19. For instance, see the list compiled by Bill Roggio and Alexander Mayer: “Senior al Qaeda and Taliban Leaders Killed in U.S. Airstrikes in Pakistan, 2004–2013,” Long War Journal, http://www.longwarjournal.org/pakistan-strikes-hvts.php (accessed March 2014). 20. See “Top Taliban Commander ‘Arrested,’ ” BBC News, May 19, 2006; “Key Taliban Leaders Captured?,” CBS, May 19, 2006; Omid Marzban, “Mullah Dadullah: The Military Mastermind of the Taliban Insurgency,” Terrorism Focus 3:11 (March 21, 2006). 21. For details on Dadullah, see Matt Dupee, “Mullah Dadullah Killed in Helmand Fighting,” Afgha.com, May 13, 2007; Kim Barker, “Taliban Leader’s Death a Big Blow,” Chicago Tribune, May 14, 2007; Syed Saleem Shahzad, “Dadullah’s Death Hits Taliban Hard,” Asia Times, May 15, 2007; United Nations Assistance Mission to Afghanistan (UNAMA), Suicide Attacks in Afghanistan (2001–2007), September 2007; Kamran Bokhari, “Taliban Psyops: Taking the Fight Abroad,” Global Intelligence, STRATFOR, November 2, 2006. See also Dadullah’s various interviews at Live Leak, http://item.liveleak.com/2/browse?q=mullah+dadullah& = (accessed April 2014); English transcript: “Taleban Leader Says bin-Laden Alive, to Speak on Video,” Afghan News Bulletin, no. 1677 (Embassy of Afghanistan in Canada), May 2, 2007); Graeme Smith, “Taliban Scramble After Top Commander Killed,” Globe and Mail, May 14, 2007; Matthias Gebauer, “Mullah Dadullah, the Star of Afghanistan’s Jihad,” Der Spiegel, March 1, 2007. Shahzad, “Dadullah’s Death Hits Taliban Hard.” 22. Rachel Morarjee, “After a Taliban Leader’s Death,” Time, May 14, 2007; Rachel Morarjee and Mark Sappenfield, “Death of Taliban Chief Leaves Void,” Christian Science Monitor, May 13, 2007. 23. Shahzad, “Dadullah’s Death Hits Taliban Hard.” 24. Syed Saleem Shahzad, “Taliban a Step Ahead of U.S. Assault,” Asia Times, August 11, 2007; Rod Norland, “One Voice or Many for the Taliban, But Pegged to a Single Name,” New York Times, June 14, 2011. 25. “Afghanistan in its ‘Bloodiest’ Year: Defence Official,” Agence France-Presse, December 2, 2007. 26. “Taliban Chief Says Brother to Replace Dadullah,” Agence France-Presse, May 14, 2007. 27. These figures do not include insurgents killed in air strikes during or following SA/R engagements. For instance, on May 11, 2007, an estimated forty insurgents were killed in ISAF air strikes called in during an SA/R exchange with Afghan forces.

236

Notes to Pages 139–149

28. Habib Rahman Ibrahimi, “Would-Be Suicide Bomber Held in Kabul,” Pajhwok Afghan News, April 25, 2007; Abdul Matin Sarfaraz, “Bomber Killed in Takhar Suicide Blast,” Pajhwok Afghan News, April 25, 2007. 29. Failed suicide attacks after Dadullah’s killing include the following: On May 17, a truck carry ing explosives was stopped at Kabul’s eastern gate and its occupants were arrested; on May 18, two suicide bombers (on foot and working together) were arrested in Kabul; and on May 30, a single bomber was shot and arrested. 30. UN Integrated Regional Information Network (IRIN), “Afghanistan: Police Casualties High Due to Lack of Training, Equipment,” June 13, 2008. 31. The other assassination attempts included an attack on MP Ahmad Khan and on Ghazni’s ANP chief (gunmen attacked his home, killing his wife, two sons, and two nephews). 32. Jason Burke, “Hunt for ‘Traitors’ Splits Taliban,” Observer, May 27, 2007. 33. Rahimullah Yusufzai, “Spies in Their Ranks Worry Taliban,” News International, May 18, 2007. 34. International Security Assistance Force (ISAF), Press Release No. 2007-576u, “Update: ISAF Confirms Taliban Leaders Targeted,” July 29, 2007. 35. ISAF waited almost a week to confirm Mohammad’s death, issuing a statement to that effect on July 29. By then, reports of the raid in question had been overshadowed by other, more immediate stories, like the ongoing Korean hostage saga and the death of Afghanistan’s last king, Mohammad Zahir Shah. “Six NATO Troops Killed in Afghanistan,” Agence France-Presse, July 24, 2007; United Kingdom Ministry of Defence, “British and Afghan Forces Launch New Offensive Against the Taliban,” July 26, 2007. 36. Bill Roggio, “Chairman of Afghanistan’s Taliban Military Council Killed,” Long War Journal, July 29, 2007; Alastair Leithead, “Long Haul Fight to Defeat the Taleban,” BBC News, June 25, 2007; “ISAF: Chairman of the Taliban Military Council Killed in Helmand,” Afgha.com, July 29, 2007; “Afghan Soldiers Injured by RemoteControlled Bomb,” Associated Press, May 30, 2004. 37. Soraya Sarhaddi Nelson, “Taliban’s Shift ing Tactics Define Afghanistan Conflict,” National Public Radio, March 4, 2008. 38. For visual reference, see “Anatomy of IED,” Long War Journal, March 2008, www.longwarjournal.org /multimedia/IED-Anatomy/index.html. 39. “Taliban Says Had No Role in Deadly Khost Bombings,” Reuters, April 23, 2007; “Official: al-Qaida Near Tipping Point?,” Associated Press, February 7, 2008. 40. “Taliban Issues Code of Conduct,” Al Jazeera, July 28, 2009. 41. Human Rights Watch, “The Human Cost: The Consequences of Insurgent Attacks in Afghanistan,” 19:6 (April 2007). 42. Philip Smucker, “Taliban Uses Weapons Made in China, Iran,” Washington Times, June 5, 2007; Scott Baldauf and Ashraf Khan, “New Guns, New Drive for Taliban,” Christian Science Monitor, September 26, 2005; “Taliban Want Heavier Weap-

Notes to Pages 149–158

237

ons to Attack Canadians,” Canadian Press, May 8, 2006; Declan Walsh, “U.S. Covered Up Fatal Taliban Missile Strike on Chinook,” Guardian, July 25, 2010. 43. “Taliban’s Failed First Use of SAM Still Worrisome,” Washington Times, July 29, 2007. 44. In another case, this time a failed bombing attempt, a Pakistani adolescent was arrested. He had been coerced into carry ing out a suicide attack in Afghanistan by his Taliban handlers. He was later reunited with his family. 45. The discrepancy between successful suicide bombers (three) and bombers who successfully detonated their explosives but failed to kill others (four) is accounted for by the fact that in three of the five recorded failed suicide bombings the bombers successfully detonated their payloads but altogether missed their targets. Hence, these three individuals “succeeded” in detonating their explosives but “failed” as suicide bombers. 46. In one failed vehicle-borne suicide attack, two bombers were killed. 47. One Afghan civilian was killed and two international civilians were reported injured in the attack. 48. Choe Sang-Hun, “Freed by Taliban, 19 South Korean Hostages Will Face Relief and Anger Back Home,” New York Times, September 2, 2007. 49. Iqbal Sapand, “The Taliban: Kidnapping, Inc.,” NBC News, February 13, 2008; David Montero, “New Strategy in Taliban’s Offensive,” Christian Science Monitor, April 11, 2007. 50. Shahzad, “Taliban a Step Ahead.” 51. Ron Synovitz, “Afghanistan: Skeptics Urge Caution over Purported Hekmatyar Cease-Fire,” Radio Free Europe/Radio Liberty, July 19, 2007. 52. Sayed Salahuddin, “Omar Urges Afghans to Unite Against Western Troops,” Reuters, August 18, 2007. 53. For details on Matin, see “Key Taliban Warlord Killed in Ambush,” Scotsman, February 22, 2008; International Security Assistance Force (ISAF), “Successful Operations Target Top Taleban Leader,” Press Release No. 2008- 075, February 21, 2008; Tom Dunn, “SBS Swoop to Kill Taliban Chief,” Sun, February 22, 2008; Matt Dupee, “Helmand: Coalition Operations Kill Three Top Taliban Commanders,” Afgha.com, February 22, 2008; “Mullah Abdul Matin,” Afgha.com, February 23, 2008. 54. ISAF, “Successful Operations Target Top Taleban Leader,” Press Release. 55. Failed suicide attacks included two vehicle car bombs uncovered in Kandahar (February 11) and another vehicle car bomb preemptively destroyed in an ISAF air strike in Zabul province. In the Zabul case, the car belonged to the director of the Peace and Reconciliation Program, retained by militants after they murdered him on January 20. Members of the ISAF recognized the vehicle as it was being driven around by a suicide bomber, and they destroyed it. In another failed attack, a suicide bomb prematurely detonated as it was being constructed in Helmand on February 9. Afghan officials reported that Mullah Abdul Wasay, a militant bomb maker, died in that explosion (along with another militant—possibly the suicide bomber—along with members of Wasay’s family).

238

Notes to Pages 159–172

56. Comparing suicide-bomber success rates is impossible given that no deaths (other than the bombers’ deaths) were reported in the weeks following Matin’s elimination. 57. The intended targets of the remaining two reported rocket attacks are unknown. 58. UN Integrated Regional Information Network (IRIN), “Afghanistan: Taliban Chief Orders Change in Mode of Executions,” September 2, 2008; Alissa Rubin, “Taliban Overhaul Image to Win Allies,” New York Times, January 21, 2010. 59. “Taliban Sack Military Commander,” BBC News, December 29, 2007; Bill Roggio, “Mullah Omar Confirms Firing of Mullah Mansoor Dadullah,” Long War Journal, January 2, 2008; Bill Roggio, “Taliban Dismisses Senior Afghan Commander [Update],” Long War Journal, December 29, 2007 60. “Taliban Sack Military Commander.” 61. Syed Saleem Shahzad, “Taliban Take a Hit, But the Fight Goes On,” Asia Times, February 2, 2008; Claude Salhani, “Jihad Turning Point?” Washington Times, February 5, 2008; Robert Reid, “Fatal Qaida Missile Strike a U.S. Victory,” ABC News, February 1, 2008. Combined Joint Task Force–101 confirms the raid but not Barri’s death; Combined Joint Task Force–101, “ANSF Secure Remote Village in Helmand Province,” Press Release, February 20, 2008. 62. David Newman and Bruce Bueno de Mesquita, “Repeal Order 12,333, Legalize 007,” New York Times, January 26, 1989. 63. Sudden increases in violence immediately following the eliminations might be based on the Pashtun tribal code of badal (revenge). Badal is the “means of enforcement by which an individual seeks personal justice for wrongs done against him or his kin group.” Though a complicated social institution, retaliation does play a prominent role in Afghan society. Other customary laws related to dispute resolution, like nanawatai (forgiveness), melmastia (hospitality), and jirga (consensus building), are also inherent in Pashtun society. See Thomas Barfield, Afghan Customary Law and Its Relationship to Formal Judicial Institutions, United States Institute for Peace (Washington, D.C.), June 26, 2003. 64. Mohammed Hafez and Joseph Hatfield, “Do Targeted Killings Work? A Multivariate Analysis of Israel’s Controversial Tactic During Al-Aqsa Uprising,” Studies in Conflict and Terrorism 29 (2006), 361– 63, 367–77. 65. Jenna Jordan, “When Heads Roll: Assessing the Effectiveness of Leadership Decapitation,” Security Studies 18 (2009), 749–53. 66. Sardar Ahmad, “Forces in Afghanistan Shift Focus to Taliban Leaders,” Agence France-Presse, January 2, 2007. 67. Ibid.

Chapter 6 1. Walter Laqueur, No End to War: Terrorism in the Twenty-First Century (New York: Continuum, 2003), 7.

Notes to Pages 173–181

239

2. Felicity Barringer, “New Tactic of Terrorists Is to Attack the Media,” New York Times, October 15, 2001; Dana Canedy and Alex Kuczynski, “Second Case of Anthrax Leads F.B.I. into Inquiry,” New York Times, October 9, 2001; William Broad, Stephen Engelberg, and James Glanz, “Assessing Risks, Chemical, Biological, Even Nuclear,” New York Times, November 1, 2001. 3. Michael Ignatieff, The Lesser Evil: Political Ethics in an Age of Terror (Princeton, N.J.: Princeton University Press, 2004), 151–52. 4. See my discussion with Andreas Wenger in “Deterring Terrorism: Moving Forward,” in Deterring Terrorism: Theory and Practice, ed. Andreas Wenger and Alex Wilner (Stanford, Calif.: Stanford University Press, 2012). 5. See Chapter 4 for a detailed description of the literature on targeted killings. 6. “Younger Leadership for Taliban in Afghanistan,” Telegraph, March 25, 2008. 7. See, for instance, Wyn Bowen, “Deterrence and Asymmetry: Non-State Actors and Mass Casualty Terrorism,” Contemporary Security Policy 25:1 (2006), 55–56. 8. The phrase is all too reminiscent: On September 13, 2001, Jean-Marie Colombani, editor of France’s Le Monde, titled his contribution “Nous sommes tous Américains” (We are all Americans). Jean-Marie Colombani, “Nous sommes tous Américains,” Le Monde, September 13, 2011; Steven Erlanger, Alan Cowell, and Adam Nossiter, “French Pledge More Troops for Mali as Airstrikes Continue,” New York Times, January 15, 2013. 9. Gary Younge and Jon Henley, “Wimps, Weasels and Monkeys: The U.S. Media View of ‘Perfidious France,’ ” Guardian, February 11, 2003. 10. Lydia Polgreen, “Mali Army, Riding U.S. Hopes, Is Proving No Match for Militants,” New York Times, January 24, 2013; Tiemoko Diallo, “Mali Islamists Capture Strategic Town, Residents Flee,” Reuters, January 10, 2013; “Where Have the Jihadists Gone?” Economist, February 2, 2013; C. J. Chivers, “Looted Libyan Arms in Mali May Have Shifted Conflict’s Path,” New York Times, February 7, 2013. 11. With an eye on Israel, Hamas, Fatah, and Hezbollah, Bar explores ways in which militant groups equalize power asymmetries with their state adversaries. Shmuel Bar, “Israeli Experience in Deterring Terrorist Organizations,” Institute for Policy and Strategy, Herzliya Conference, Israel, January 2007. 12. In Deterring Terrorism, Frank Harvey and I relate this to “counter-coercion”: behavior that alters a stronger party’s ability or willingness to rely on its preferred coercive tactic or strategy. See Frank Harvey and Alex Wilner, “Counter-Coercion, the Power of Failure and the Practical Limits of Deterring Terrorism,” in Wenger and Wilner, Deterring Terrorism. 13. Thomas Homer-Dixon, “The Rise of Complex Terrorism,” Foreign Policy (January–February 2002); Lawrence Freedman, “The Third World War,” Survival 43:4 (2001); Frank Harvey, Smoke and Mirrors: Globalized Terrorism and the Illusion of Multilateral Security (Toronto: University of Toronto Press, 2004), 16. 14. Paul Davis and Brian Jenkins, Deterrence and Influence in Counterterrorism: A Component in the War on al Qaeda (Santa Monica, Calif.: RAND, 2002), 11.

240

Notes to Pages 181–187

15. I originally explored these issues with Andreas Wenger in, “Deterring Terrorism: Moving Forward,” in Wenger and Wilner, Deterring Terrorism, 307–8, 310, 314. A similar problem is evident in applying “pre-event” and “post-event” deterrence by denial in the specific context of deterring radiological terrorism. See Wyn Bowen and Jasper Pandza, “Preventing Radiological Terrorism: Is There a Role for Deterrence?,” in Wenger and Wilner, Deterring Terrorism, 191–92, 197. 16. Jeff rey Knopf, “Terrorism and the Fourth Wave in Deterrence Research,” in Wenger and Wilner, Deterring Terrorism, 37. 17. Harvey and Wilner, “Counter-Coercion,” 104. 18. Stephen Hadley, “Remarks to the Center for International Security and Cooperation,” Stanford University, Stanford, Calif. (February 8, 2008). David Sanger and Helene Cooper, “Bush, in Asia, Makes Appeal on North Korea,” New York Times, November 16, 2006. 19. U.S. president Barack Obama, Opening Statement, Nuclear Security Summit, Washington, D.C., April 13, 2010. 20. I explore the issue in greater detail in “Fencing in Warfare: Threats, Punishment, and Intra-War Deterrence in Counterterrorism,” Security Studies 22:4 (2013). 21. T. Clifton Morgan and Patrick Regan, “Deterring Rebellion,” Foreign Policy Analysis 7:3 (2011), 313. 22. Martha Crenshaw, “Will Threats Deter Nuclear Terrorism?,” in Wenger and Wilner, Deterring Terrorism; Martha Crenshaw, “Coercive Diplomacy and the Response to Terrorism,” in The United States and Coercive Diplomacy, ed. Robert Art and Patrick Cronin (Washington, D.C.: U.S. Institute of Peace, 2003), 311; Brad Roberts, “Deterrence and WMD Terrorism: Calibrating Its Potential Contributions to Risk Reduction,” Institute for Defense Analyses (Paper P-4231) (June 2007), 30–32. 23. See Wilner, “Fencing in Warfare.” 24. “Deterrence is relevant to states,” one official told him, but “[t]errorist organizations have to be eradicated.” Shmuel Bar, “Deterrence of Palestinian Terrorism: The Israeli Experience,” in Wenger and Wilner, Deterring Terrorism, 209, n. 3. 25. U.S. president Barack Obama, “The Way Forward in Afghanistan,” June 22, 2011. 26. Thomas Schelling, Arms and Influence (New Haven, Conn.: Yale University Press, 1966); Lawrence Freedman, Deterrence (Malden, Mass.: Polity Press, 2004), esp. chap. 2; W. Andrew Terrill, “Escalation and Intrawar Deterrence During Limited Wars in the Middle East,” Strategic Studies Institute (2009); Barry Posen, “U.S. Security Policy in a Nuclear-Armed World, or What If Iraq Had Had Nuclear Weapons?,” in The Coming Crisis: Nuclear Proliferation, U.S. Interests, and World Order, ed. Victor Utgoff (Cambridge, Mass.: Belfer Center for Science and International Affairs, 2000), 175. 27. See Thomas Schelling’s comment in his foreword to Deterring Terrorism, ed. Wenger and Wilner. 28. Audrey Kurth Cronin, How Terrorism Ends (Princeton, N.J.: Princeton University Press, 2009); David C. Rapaport, “The Four Waves of Modern Terrorism,” reprinted

Notes to Pages 187–191

241

in Attacking Terrorism: Elements of a Grand Strategy, ed. Audrey Cronin and James Ludes (Washington, D.C.: Georgetown University Press, 2004). 29. Sean Rayment, “Britain’s Top Soldier Says al-Qaeda Cannot Be Beaten,” Telegraph, November 13, 2010.

Appendix 1. Alexander George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (Cambridge, Mass.: MIT Press, 2005), ix, 74–79; Arnold Lijphart, “Comparative Politics and the Comparative Method,” American Political Science Review 65:3 (1971), 691. 2. Lijphart, “Comparative Politics,” 683. 3. Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994), 75–82; George and Bennett, Case Studies, 25–30. 4. For research on the par ticu lar difficulty associated with case-study selection in evaluating theories of deterrence, see Vesna Danilovic, “Conceptual and Selection Bias Issues in Deterrence,” Journal of Conflict Resolution 45:1 (2001); Richard Ned Lebow and Janice Gross Stein, “Deterrence: The Elusive Dependent Variable,” World Politics 42:3 (1990); Frank Harvey, “Getting NATO’s Success in Kosovo Right: The Theory and Logic of Counter-Coercion,” Conflict Management and Peace Science 23 (2006); Frank Harvey, “Rigor Mortis or Rigor, More Tests: Necessity, Sufficiency, and Deterrence Logic,” International Studies Quarterly 42 (1998); Christopher Achen and Duncan Snidal, “Rational Deterrence Theory and Comparative Case Studies,” World Politics 41:2 (1989); Barbara Geddes, “How the Cases You Choose Affect the Answers You Get: Selection Bias in Comparative Politics,” Political Analysis 2 (1990); David Collier and James Mahoney, “Insights and Pitfalls: Selection Bias in Qualitative Research,” World Politics 49:1 (1996). 5. George and Bennett, Case Studies, 153. 6. Ibid., 166. 7. Lijphart, “Comparative Politics,” 689. 8. Ibid. 9. George and Bennett, Case Studies, 179, 181–82. 10. Ibid., 183. 11. Building counterfactuals, integrating quantitative and qualitative methodologies, or assessing cases in which the projected outcome occurs in the absence of the independent variable (or cases in which the projected outcome fails to materialize though the independent variables are present) represent other ways in which causal significance can be ascertained. Barbara Geddes, “How the Cases You Choose Affect the Answers You Get: Selection Bias in Comparative Politics,” Political Analysis 2 (1990); King, Keohane, and Verba, Designing Social Inquiry; Gary King, “On Political Methodology,” Political Analysis 2 (1990), 1–25; George and Bennett, Case Studies, 230–32.

242

Notes to Pages 191–199

12. George and Bennett, Case Studies, 184. See also Andrew Bennett and Alexander George, “Process Tracing in Case Study Research,” MacArthur Foundation Workshop on Case Study Methods (paper presentation), October 1997. 13. George and Bennett, Case Studies, 201. 14. Ibid., 161, 207. 15. Achen and Snidal, “Rational Deterrence Theory,” 163. 16. As in the case when a suicide bomber kills only him- or herself. The failure here is that, despite a successful detonation, the bomber does not kill anyone else. 17. Vehicle-borne suicide attacks include those carried out by militants riding motorcycles or bicycles. 18. A further distinction was made between “simple” IEDs (which are victimoperated explosives that detonate when individuals or objects sever tripwires or trigger pressure mats, for example) and “complex” IEDs (which are remotely activated by terrorists using radio or cellular transmissions or other electronic devices). 19. Includes ambush attacks and the use of shoulder-launched missiles and other rocket-propelled grenades. 20. Attacks involving numerous foot soldiers who coordinate a complex attack on a fi xed, defensive position (a security checkpoint, police headquarters, an ISAF base, and so on). 21. I thank an anonymous reviewer for highlighting the issue. 22. Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 2006), 40; Bruce Hoffman and Gordon McCormick, “Terrorism, Signaling, and Suicide Attack,” Studies in Conflict and Terrorism 27 (2004), 248–51. 23. Daniel Byman, “Understanding Proto-Insurgencies,” Journal of Strategic Studies 31:2 (2008), 167–70. 24. Austin Long, Deterrence: From Cold War to Long War (Santa Monica, Calif.: RAND, 2008). 25. John Arquilla, “The End of War as We Knew It?,” Third World Quarterly 28:2 (2007). 26. Frank Douglas, “Waging the Inchoate War: Defining, Fighting, and SecondGuessing the ‘Long War,’ ” Journal of Strategic Studies 30:3 (2007); David Kilcullen, “Countering Global Insurgency,” Journal of Strategic Studies 28:4 (2005). 27. Personal correspondence with Shuvaloy Majumdar, (former) resident country director and chief of mission, International Republican Institute, Afghanistan. 28. Personal correspondence with Bev Toomer, (former) chief field security adviser, UNDP, Afghanistan. 29. Strategic Security Solutions International (Afghanistan), Daily Situation Report, July 7, 2008. 30. I was nonetheless given a number of archived UNDSS reports thanks to the Afghan Research and Evaluation Unit, though only in great confidence. I have agreed not to reveal any specific sources.

Note to Page 199

243

31. The most useful (semi)closed sources were the GardaWorld Summaries and Overviews. A conversation with Neil Mackinnon, GardaWorld’s director of Crisis Response and Intelligence, revealed how GardaWorld collected its information. The Afghan intelligence reports were collected on an hourly basis from a number of sources, including ISAF and UN personnel, local intelligence sources (U.S., Afghan, and international personnel), and a variety of other human intelligence sources active in Afghanistan. Relevant information was tabulated and packaged and then sold to various NGOs, private-sector groups, and other individuals who needed immediate assessments on the level of insurgent activity in their working environments. Personal correspondence with Neil Mackinnon, GardaWorld, director of Crisis Response and Intelligence.

INDE X

abductions (kidnappings), 128, 194, 216n76; after Dadullah killing, 152; as “emotional warfare,” 153; after Matin killing, 157, 162, 163, 164; after Mohammad (Qari Faiz) killing, 143; of soft targets, 144 Abdulmutallab, Umar Farouk, 86 Abrahms, Max, 5, 66 Abu Ghaith, Sulaiman, 101 Abu Nidal (Sabri al-Banna), 103– 4, 115–16 Abu Nidal Orga nization (ANO), 103– 4 Abu Sayyaf Group (ASG), 92 Achen, Christopher, 192 Adamsky, Dima, 34 Afghanistan, 6, 10, 114, 176; data sets collected in, 197–99; national army and police , 136, 138– 41, 145–56, 158, 160– 65, 169, 194; al Qaeda in, 42, 127, 130; Soviet war in, 9, 126; Tora Bora Military Front, 126, 130; U.S. precision attacks in response to terrorism, 46, 50 Aggression Against Lal Masjid, The (al-Zawahiri), 155 Ahmad, Mullah Shah Mansoor (Mansour Dadullah), 123–24, 166, 233n10 Ahmadi (Ahmedi), Qari Yousuf, 134, 135 Almog, Doron, 34, 39 Alston, Philip, 112 Annan, Kofi, 81 al Aqsa Intifada, 56, 62, 80, 88, 98; as existential challenge to Israelis, 178; suicide bombings during, 89 Al-Aqsa Martyrs Brigade, 62, 106, 225n50 Arafat, Yasser, 48, 88, 89 Armitage, Richard, 49, 213n34 Arquilla, John, 196 assassination, 78–81, 84 Assistant for Randomized Monitoring Over Routes (ARMOR), 64

Atef, Muhammad (Abu Hafs al-Masri), 92 Atta, Mohammed, 42 Atzili, Boaz, 47 Aum Shinrikyo, 53–54, 57, 187 Awlaki, Abdulrahman al-, 86 Awlaki, Anwar al-, 70, 71, 86, 93 Ayyash, Yahya (“the Engineer”), 76, 89, 112 Azimi, Gen. Mohammad Zahir, 135, 171 backlash, popu lar, 69, 218n114 Badat, Saajid, 101 Badri, Ibrahim Awwad Ibrahim Ali (Abu Du’a), 96, 97 Balawi, Humam Khalil Abu Mulal al-, 112 Baluch, Mullah Mahmud, 12, 126, 167 Bar, Shmuel, 34, 46, 57, 203n32, 239n11; on power of weakness, 179; on reverse proxyship, 47 Barader, Mullah, 105 Bar-Joseph, Uri, 34 Barri, Mullah Abdul, 166 Baryal, Qari, 130, 131 behavioral defenses, 62– 63 beheadings, 128, 145, 165, 194 Beirut Marine barracks bombing (1983), 9 Ben-Israel, Isaac, 96 Benmelech, Efraim, 56 Bennett, Andrew, 189, 191 Berader, Mullah Abdul Ghani, 129, 234n15 Beslan, mass hostage taking in, 65, 66, 91 Bhutto, Benazir, 166 Bigelow, Kathryn, 112 bin Laden, Osama, 4, 7, 48, 71, 90, 93; Afghan compound targeted by cruise missile, 91; Bush’s “Wanted” declaration about, 85; criticized within al Qaeda, 10; denounced by brother, 59; elimination of, 68, 85, 93, 114, 132; evasion of U.S. signals

246

Index

bin Laden, Osama “continued” intelligence, 180; as “lay mujtahid,” 69; planning of 9/11 attacks and, 42; on United States as paper tiger, 9; Zero Dark Thirty fi lm about hunt for, 112–13. See also al Qaeda Black September, 88 Blair, Tony, 65 Bloom, Mia, 8, 43– 44, 109 Bolton, John, 7 bomb makers, 40, 60 Bowen, Wyn, 18 Brennan, John, 117 Brophy-Baermann, Bryan, 58 Bueno de Mesquita, Bruce, 167 Bush, George H. W., 9, 84 Bush, George W., 2, 10, 48, 52, 65, 78; counterterrorism policies of, 111; reinterpretation of ban on assassination, 85 Byman, Daniel, 25, 46, 102, 116, 196, 213n39 Canada, 82, 99, 110 Carter, Jimmy, 84 central deterrence, 31 Central Intelligence Agency (CIA), 11, 68, 91, 103, 226n74; Camp Chapman base in Afghanistan, 112, 127; drone strikes and, 78, 86, 92, 94, 105, 166; leadership transition in, 117; militants as prisoners of, 131; Pakistani intelligence cooperation with, 129; Phoenix Program and, 83, 84; planned attack on headquarters of, 42 challengers: counter-coercion by, 24–25; deterrence by denial and, 28; inaction by, 184; self-restraint of, 69; shared interests with defenders, 23–24 Chazan, Naomi, 112 Chechens/Chechnya, 4, 91, 127, 190 chemical, biological, radiological, nuclear (CBRN) weapons, 13, 32, 33, 51–55, 183; anthrax attacks (2001), 173; delegitimization and, 68, 69, 72, 217n98; tailored deterrents and, 39 Clausewitz, Carl von, 6 Clinton, Hillary, 79 coercion, 1, 11, 12, 120–22, 131–32; counter-coercion, 25; Dadullah killing and, 132– 42; defi ned, 17; dilemmas faced by states, 13; of groups, 57–59; of individual terrorists, 59– 61; induce-

ments/rewards and, 30; intricacies of, 2; less traditional forms of, 31; liberal democracies and, 56–57; Matin (Mullah Abdul) killing and, 156– 67; military force and, 17–18; Mohammed (Qari Faiz) killing and, 142–56; in ongoing security environment, 34; punitive, 55, 215n71; reinterpretation of data on targeted killings, 174–77; state-centric Cold War coercion, 45 Colby, Elbridge, 56 Cold War, 1, 2, 45, 99; assassinations and, 79; classical deterrence theory as product of, 16; credibility of nuclear deterrence, 23; end of, 37; limits of war-fighting behavior, 32; mutual assured destruction (MAD), 8; mutual destruction as threat, 6; nuclear punishment as principal threat, 61; punishment-based deterrence in, 27, 29 Colombia, 55, 91 compellence, 1, 12, 14, 67; as “coercive diplomacy,” 18; communication and, 22; defi ned, 16–17 conventional weapons, 25–26 Conybeare, John, 58 cost-benefit calculus, 21, 24, 192; defensive denial and, 61; deterrence by denial and, 28; effectiveness of deterrence and, 26; suicide terrorism and, 8, 21 counterinsurgency (COIN), 11, 81, 120, 145, 192; civilian casualties and, 108; counterterrorism distinguished from, 195–96, 197 counterterrorism, 5, 12, 24, 35, 38; broadened conception of deterrence and, 35; coercion in, 145; defeat- deter paradox and, 184; denial mechanisms, 61, 62; deterrence theory and, 7, 75; drones used in, 11, 88; influence operations in, 73; international partnerships damaged by, 111; military force used in, 35; offensive and defensive, 2, 220n134; success requirements for, 14 credibility, 23, 30, 120 Crenshaw, Martha, 28, 34, 51 criminal-justice system, 60 criminology, 34, 35 Cronin, Audrey Kurth, 117–18, 176

Index Dadullah, Mullah, 12, 104–5, 123, 124–25, 129; interviews conducted with, 167, 177; tactics supported by, 128; targeted killing of, 132– 42, 151–52, 155, 164, 175, 177, 236n29 Dahiya doctrine, 56 data collection, 197–99 David, Steven, 89, 111 Davidson, Amy, 86 Davis, Paul, 38, 43, 181 decapitation strategies, 76, 116, 117, 119, 168 defeat-deter paradox, 184–86 defenders: CBRN attacks and, 51; challengers’ self-restraint and, 69; cost-benefit calculations of, 24; cumulative deterrence and, 35; restraint by, 184; shared interests with challengers, 23–24 Deif, Mohammed, 90 delegitimization, 11, 14, 30, 37, 45, 73, 174; as core concept of deterrence, 38; of Palestinian cause, 66; punishment joined by, 186–87; strategic futility and, 217n98; as targeting of what terrorists believe, 66–73 denial, 11, 12, 45, 173, 181, 187; as core concept of deterrence, 38; defensive, 61– 62; nuclear weapons and, 37; punishment distinguished from, 26–31; as targeting of what terrorists want, 61– 66; as threat against groups, 59 deprofessionalization, 13, 100–103 Dershowitz, Alan, 79, 87 deterrence: asymmetry and, 177–80; broad and narrow, 32–33, 182, 208n63; coercive communication and, 180–84; as communicative process, 9; compellence compared with, 17–18; cumulative, 34–35; in Dr. Strangelove, 22; expanded logic of, 38– 40; extended, 37; failure of, 6–7, 8, 9, 35–36, 203n32; immediate and general, 32, 33–34, 181; inter-state, 6, 23; intra-war, 31, 34, 184–86; logic of, 16; military force and, 18; by mitigation, 64– 65; necessity of, 15; prerequisites of, 20, 21–26; punishment versus denial, 26–31; skepticism toward, 2–11, 203n30; as unintended outcome, 22 deterrence theory, 1, 31, 67; development and testing of, 172–74; as game of strategic interaction, 21; methodology of research in, 189–92; prerequisites of, 80;

247

research design in, 192–95; targeted killings and, 11–12; updating of, 11; waves of research in, 19–20, 37, 73 Deterring Terrorism (Wenger and Wilner, eds.), 25, 35, 239n12 dissuasion, 37 Dr. Strangelove (fi lm), 22 Dogs Are Eating Them Now, The (Smith), 107 drones, 11, 13, 76–77, 80, 85, 88; Iranianbuilt Hezbollah drone, 111–12; militants’ attempts to evade, 98; Predator drones, 77, 81, 93, 126, 131 Dunn, Lewis, 4, 38–39, 54, 218n114 Dying to Kill (Bloom), 8 Eddy, R. P., 63 Egypt, 46, 49, 88, 127 Eisenkot, Gadi, 48 Eisentadt, Michael, 102–3 escalation, 5, 32 Ethiopia, 46, 94 Euskadi Ta Askatasuna (ETA), 119 exchange model, 30 extrajudicial killings, 78, 81, 111 Fabius, Laurent, 179 Fadlallah, Mohammed Hussein, 90 FARC (Revolutionary Armed Forces of Colombia), 55, 91, 119, 168 Fatah, 46, 66, 88–89 fatwas, 4, 55 FBI (Federal Bureau of Investigation), 94, 110 fidayeen, 70 fi nanciers, 40, 87 Fisher, Uri, 56, 216n76 foot soldiers, 39, 40, 42, 101, 193 Ford, Gerald, 84 Fort Hood shooter (2009), 86 Foss, Lt. Col. Claudia, 142– 43 France, 53, 91–92, 96; intervention in Mali, 44, 178–79; militant captured in, 115; revolutionary Marxists in, 187 Freedman, Sir Lawrence, 3, 17, 64– 65 Freeman, Michael, 117 Front de Libération du Québec (FLQ), 55 GardaWorld, 199, 243n31 Gaza, 44, 48, 57, 80, 81, 89, 120, 190; deprofessionalization of groups in, 103; George, Alexander, 189, 191

248

Index

Gerges, Fawaz, 44 Germany: Berlin discotheque bombing, 84; Nazi regime, 215n71; al Qaeda in, 42, 115; revolutionary Marxists in, 187 Gold Mohur Hotel bombings (1992), 10 Goldsmith, Jack, 77 Gray, Colin, 19 Guantanamo Bay detention camp, 111, 131 Guiora, Amos, 87–88 Gulf War, first (1991), 84 Guzman, Abimael, 91, 96, 100, 115 Hadley, Stephen, 183 Hafez, Mohammed, 118, 168 Halutz, Gen. Dan, 90 Hamas, 46, 48, 62, 96, 119; delegitimization deterrence and, 68; deprofessionalization of, 102–3; Izz ad-Din al-Qassam Brigades, 89; leaders killed by Israel, 60, 76, 89, 90, 102, 225n50; “statelike” attributes of, 58; support base of, 55; Taliban compared to, 127; weapons of, 148 Haqqani, Badruddin, 94 Haqqani, Jalaluddin, 126 Haqqani, Nasiruddin, 94 Haqqani, Sirajuddin, 94 Haqqani network, 94, 130, 196 Harb, Ragheb, 89 Harithi, Abu Ali al-, 81, 92 Harkat-ul-Jihad-al-Islami (HuJI), 93 Harmon, Chris, 40, 99 Harvey, Frank, 22, 23, 25, 82–83, 182, 239n12 Hasan, Nidal, 86 Hatfield, Joseph, 118, 168 Hekmatyar, Gulbuddin, 126 Hezb-i-Islami Afghanistan [HIA] (Islamic Party of Afghanistan), 126 Hezbollah, 41, 46, 47, 90, 203n32; as ally of Iran, 113; Iranian-built drone of, 111–12; Jewish targets in Argentina attacked by, 112; kidnapping of Soviet diplomat by, 216n76; stated intent of, 113; “statelike” attributes of, 58; support base of, 55; Syria and, 48, 53, 113; targeted killings of leaders of, 98; U.S. operation against, 90; weapons of, 148 Hoff man, Bruce, 3, 196 Holder, Eric, 85–86 Homeland Security, U.S. Department of, 62

Hopf, Ted, 30–31 hostages, 4, 65– 66, 145, 152, 153–54 house demolitions, 56, 57 How Terrorism Ends (Cronin), 117–18 human intelligence (HUMINT), 106–7 Human Rights Watch (HRW), 78 Hussein, Saddam, 27, 84, 91 Huth, Paul, 21, 30 Ignatieff, Michael, 173 Ignatius, David, 80, 222n19 improvised explosive devices (IEDs), 14, 169–71, 197; civilian casualties from, 176; after Dadullah killing, 134, 136–37, 175; after Matin killing, 157– 64; after Mohammad (Qari Faiz) killing, 143– 47, 151–54; vehicle-borne, 199; victim-operated (“simple”) versus remotely operated (“complex”), 195, 242n18 India, 53 inducement, 37, 73 Indyk, Martin, 84 influence, 12, 13, 35, 37; compellence and, 16–17; expanded logic of deterrence and, 38 Influencing Violent Extremist Orga nization research team, 73 Inspire (AQAP propaganda magazine), 86 insurgents, 37, 125, 196 intentions, 19, 21, 192; misinterpretation of, 20, 182; muddied or weakened, 181, 184; NATO intentions in “Most Wanted” campaign, 129–30 international relations theory, 12, 172 International Security Assistance Force (ISAF), 93, 124–25, 131, 166, 167, 225n56; Dadullah killing and, 132, 136, 137; Matin killing and, 165; Mohammad (Qari Faiz) killing and, 142–51, 154, 156; Taliban attacks on, 149, 194 intra-national investigation, 190–91 Iran, 44, 91, 113, 183; Hezbollah drone built by, 111–12; Iraq’s WMD deterrence against, 27; nuclear weapons program, 79; as “renegade regime,” 30 Iraq, 6, 27, 44, 176, 190; chemical weapons used against Kurds, 215n71; civilian casualties of terrorism in, 178; drone strikes in, 85, 108; invasion of Kuwait, 33; Predator drones in, 77; al Qaeda in, 41,

Index 54, 116; U.S. precision attacks in response to terrorism, 46 Iraq War (2003), 52, 73, 97 Irish Republican Army (IRA), 55, 90 Islam, 58, 67, 70–71 Islamic Courts Union, 47– 48 Islamic Movement of Uzbekistan, 93, 104, 130, 196 Islamic State of Iraq and the Levant (ISIL or ISIS), 116 Islamist groups, 41, 44, 187 Israel, 33, 73, 80, 114; credibility of threats made by, 57; cumulative deterrence used by, 34–35; Hamas leaders killed by, 60; Hezbollah drone in, 111–12; morale of militants undermined by, 106; punitive coercion used by, 56; response to Munich Olympics massacre, 58, 110; suicide bombings in, 62; targeted killings used by, 76, 81, 83, 88–90, 102, 231n133; threats against states linked to terrorism, 46, 185; thwarted terrorist attacks in, 102–3; war with Hezbollah (2006), 48, 203n32 Israel Defense Forces (IDF), 48, 87, 90, 120–21 Issacharoff, Avi, 106 Italy, radical left in, 99, 187 Jabari, Ahmed, 90, 120–21 Jabhat al-Nusra, 116 Japan, 53, 56–57 Japa nese Red Army, 104 Jemaah Islamiyah, 41, 71 Jenkins, Brian Michael, 6, 38, 43, 54, 181 Jews, 55, 58, 73 jihad, 4, 41, 44, 71 Al Jihad (Egyptian Islamic Jihad [EIJ]), 71 Johnson, David, 38 Johnston, Patrick, 119, 176 Joint Resolution of Congress [Authorization for the Use of Military Force] (2001), 85 Jordan, 46, 69, 72, 73 Jordan, Jenna, 118–19, 168, 176 Juergensmeyer, Mark, 5 Kaczynski, Theodore (“Ted”), 100 Kamolnick, Paul, 218n117 Kaplan, Edward, 118 Karzai, Hamid, 123, 162 Kashmir, 50, 93

249

Kenya, 47, 96; U.S. Embassy bombing (1998), 10, 91, 93, 94; Westgate Mall attack, 44, 95 Khalid, Asadullah, 107, 124, 133, 151, 162, 163 Khan, A. Q., 53 Khan, Samir, 86, 93 Knopf, Jeff rey, 67, 218n114 Korea, North, 30, 52, 183 Kosovo air campaign (1999), 25, 91 Kraska, James, 7, 52 Krauthammer, Charles, 3 Kurdistan Democratic Party of Iran (KDPI), 91 Kurdistan Workers’ Party (PKK), 46, 47, 91, 96 Kuwait, 33 Lake, David, 66 Laqueur, Walter, 172, 178 Lashkar-e Taiba (LeT), 97 law, international, 81, 83 leaders, death or capture of, 115, 117–19; creation of martyrs, 113, 114; effects of, 13, 60; elimination of threat and, 95–97; leadership redundancies and, 41– 42; unknown successors, 113–16 Lebanon, 9, 41, 46, 56, 113 Lebow, Richard Ned, 19, 21 Leibovich, Lt. Col. Avital, 121 Lewis, Bernard, 10 Libbi, Abu Farraj al-, 92, 114 Liberation Tigers of Tamil Eelam [LTTE] (Tamil Tigers), 53, 55, 104, 115, 178, 187 Libi, Abu Laith al-, 130, 131, 166 Libi, Abu Yahya al-, 93–94 Libya, 30, 46, 84, 114, 127; Malian crisis (2013) and, 179; nuclear program, 53 Lijphart, Arnold, 190 Limburg (French oil tanker), suicide attack on, 81 Long, Austin, 116–17, 196 Long, Jerry Mark, 54–55, 67, 69, 72 Long War Journal, 198 Los Angeles International Airport (LAX), 64 Luft, Gal, 107 Lupovici, Amir, 35 Madrid train bombings (2004), 97 Mali, 6, 44, 91, 176, 178–79

250

Index

Manila Bay ferry bombing (2004), 92 Marsh, David, 171 martyrdom, 3– 4, 39, 113 Marxists, 3, 187 Mashal, Khaled, 89 Mastrogiacomo, Daniele, 123, 124 Matin, Mullah Abdul, 12, 76, 125, 156–67, 175 Mehsud, Baitullah, 93, 112, 126, 130–31,166 Mehsud, Hakimullah, 93, 112 Melzer, Nils, 76 militants, 37, 186; armchair militants, 56; asymmetries in confl icts with states, 177–80; retaliation for targeted killings, 112–13; soldiers contrasted with, 82–83; surrender of, 106; terrorist designation and, 196 Milošević, Slobodan, 25, 91 mitigation, 14, 30, 64– 65, 169, 181, 187 Mogadishu, Battle of (1993), 10 Mohamed, Fazul Abdullah, 94, 225n59 Mohammad, Qari Faiz, 12, 125, 132, 142–56, 164, 175, 236n35 Mohammed, Dost, 130, 131 Mohammed, Khalid Sheikh (KSM), 42, 68, 92, 115 monetary transactions, blocking of, 61 Moore, John Norton, 83 Mordechai, Yoav (Poly), 120–21 Morell, Michael, 117 Morgan, Pat, 32–34 Morgan, T. Clifton, 184 Mossad, 89, 110 Mughniyeh, Imad, 90 Muhajir, Abu Hamza al- (Abu Ayyub al-Masri), 96 Mumbai attack (2008), 97 Munich Olympic Games massacre (1972), 58, 62, 88, 110 Musawi, Abbas al-, 89, 98, 113 Musharraf, Pervez, 49, 50, 51 Muslims, 3, 55 mutual assured destruction (MAD), 8, 29 Nasrallah, Hassan, 48, 90, 112, 113 National Consortium for the Study of Terrorism and Responses to Terrorism (START), 63, 73 National Institute for Public Policy, 74, 99 National Military Strategy of the United States of America (2011), 46

National Security Agency (NSA), 99, 111 National Security Strategy (U.S. gov’t., 2002), 6 New America Foundation, 77, 109, 125 Newman, David, 167 New York City, 64, 68; random bag searches in subway, 63; Times Square bomber (2010), 86, 108, 127. See also 9/11 (September 11, 2001), attacks Nigeria, 178 Nincic, Miroslav, 30 9/11 (September 11, 2001), attacks, 1, 2–3, 40, 75; asymmetry and, 178; as backfi re on al Qaeda, 10; as blow to deterrence theory, 172–73; as deterrence failure, 6–7; as failure to apply deterrence theory, 8–9; planning of, 42– 43, 68; al Qaeda statement following, 4; resumption of normal routine after, 65; as symbolic drama, 6; targeted killings policy and, 84–85, 92–95 9/11 Commission Report, 42 nongovernmental organizations (NGOs), 109, 140, 243n31; kidnapping of staff of, 143, 163– 64; security assessments and, 198, 199; as targets of terrorist attacks, 139, 141, 151, 152, 194, 197 North Africa, 41, 85, 111, 116 North Atlantic Treaty Orga ni zation (NATO), 25, 33, 62, 165, 226n74; Afghan i stan operations, 93, 94, 102, 124, 129–31; civilian casualties and, 176; deterrence by denial of Soviet Union, 29; drone strikes in Pakistan and, 229n103; intervention against Qaddafi (2011), 179; Taliban attacks on, 156, 168– 69, 194, 195 Norwitz, Jeff rey, 115–16 Nuclear Security Summit (2010), 183 nuclear weapons, 1, 6, 29; clandestine Ira nian program, 79; counterforce nuclear strikes, 32; deterrence theory and, 19; as ill-suited to deter terrorism, 25; narrow deterrence and, 33; nuclear proliferation, 1, 37; stolen by militants, 62; war avoidance and, 16 Obaidullah, Mullah, 128, 129, 133 Obama, Barack, 10, 77–78, 85–86, 183, 185 Ocalan, Abdullah, 47, 91, 96

Index Omar, Mullah Mohammed, 49, 92, 97, 104–5, 123; civilian casualties and, 176; guidelines for field commanders, 134, 165; in hiding, 132; Taliban’s top echelon and, 128, 133, 155, 166 Operation Cast Lead, 48, 56, 90 Operation El Dorado Canyon, 114 Operation Enduring Freedom, 148, 194 Operation Horseshoe, 25 Operation Iraqi Freedom, 27 Operation Pillar of Defense, 48, 90, 120 Operation Returning Echo, 48 Osmani, Mullah Akhtar Mohammad, 93, 107 Pajhwok Afghan News agency, 197, 198 Pakistan, 6, 48, 53, 156, 176, 190; civilian casualties of terrorism in, 178; Federally Administered Tribal Areas (FATA), 109–10; Inter-Services Intelligence (ISI), 49, 104, 129; support for militant groups, 49, 214n45; U.S. relations with, 99, 111 Pakistan, drone strikes in, 76–78, 85, 92–95, 105; expansion of, 125; popu lar opinion about, 109–10, 229n103; suspension of, 98–99; Taliban leaders killed by, 131, 166 Palestine Liberation Orga nization (PLO), 46, 66, 89, 96, 185 Palestinian Authority (PA), 46, 48, 80, 106 Palestinian Islamic Jihad (PIJ), 62, 89, 96, 100, 113, 116, 225n50 Palestinian terrorists, 56, 58, 63, 83 Panetta, Leon, 11, 85 Pape, Robert, 43 Parachini, John, 68 Pearlman, Wendy, 47 Pentagon, attack on, 42 “Permanent War, The” (Washington Post special report), 77–78 Peru, 55, 73, 91, 187 Petraeus, David, 117 Philippines, 42, 92 Phoenix Program, 83–84 pirates, 37, 222n30 Popu lar Front for the Liberation of Palestine (PFLP), 112, 225n50 Popu lar Resistance Committee, 90 Porter, Patrick, 100 Posen, Barry, 33 Post, Jerrold, 113 Powell, Colin, 50

251

Powers, Anthony, 64 Powers, Michael, 42 Prabhakaran, Velupillai, 104, 115 Price, Bryan, 119, 176 prisoner exchanges, 123–24 process-tracing, 191–92 propaganda, 41, 115 proxy-patron relationship, 46 psychology, in deterrent relations, 20 publicity, 3, 5, 39 punishment, 11, 173, 174; case studies in, 74; changes in Taliban professionalism and, 174; compellence and, 18; as core concept of deterrence, 38; denial distinguished from, 26–31; deterrence by, 8; deterrence strategies beyond, 40; as goal of militants, 5; intra-war deterrence and, 186; random, 76; retaliation distinguished from, 207n42; targeted killings as, 13; targeting what terrorists value, 45–51 Qaddafi, Col. Muammar, 84, 114, 179 al Qaeda, 3, 26, 92, 166, 172; asymmetry in relations with West, 178; decline of deterrence theory and, 1–2; delegitimization used against, 67–73, 218–19n117; fracturing of global network, 59; franchises in various countries, 41, 187; internal policy disagreements, 44; 4; lack of restraint by, 28; in Mali, 178; “Mumbaistyle” attacks in Europe planned by, 95, 115; Muslims targeted by, 104; nuclear weapons and, 7, 54; quality of new leadership of, 116; stated intent of, 113; support base of, 55, 69; Taliban and, 49, 51, 93; targeted killings of leaders of, 11, 85, 87, 93–94. See also bin Laden, Osama; 9/11 (September 11, 2001), attacks al Qaeda in Iraq (AQI), 54, 72–73, 92, 96–97, 98 al Qaeda in the Arabian Peninsula (AQAP), 70, 80, 86, 93, 222n19 al Qaeda in the Islamic Maghreb (AQIM), 92 Quester, George, 9 Quinlan, Sir Michael, 8, 39 Quso (Qusaa), Fahd Mohammed Ahmed al-, 94, 110 Rahim, Mohammad, 130, 131 Ramsay, Clay, 109

252

Index

random antiterrorism measures (RAM), 63– 64, 217n90 Rantissi, Abdel Aziz al-, 60, 89, 100 Reagan, Ronald, 84 reassurance, 30, 185 recruiters, 40 Regan, Patrick, 184 Reid, Richard, 101 religion, 37, 44, 67, 69–70 restraint, 69, 184, 184–85; mutual, 28, 31; narrow deterrence and, 32; terrorist self-restraint, 13, 68, 72 retaliation, 5, 9, 34–35; CBRN/WMD attacks and, 51, 52; compellence and, 18; counter-capability effects of, 220n134; nuclear, 31; public announcements of, 23; punishment distinguished from, 207n42; as vengeance, 57, 184 reverse proxyship, 47 Rid, Thomas, 34 “rogue” states, 37, 172 Ross, James, 78 Rumsfeld, Donald, 2, 6, 9 Russett, Bruce, 21, 30 Russia, 53, 65– 66, 91, 215n71. See also Soviet Union (USSR) Saeed, Hafiz Mohammad, 97 safe havens/houses, 39, 77, 125 Sageman, Marc, 10, 41, 42 Salafi movement, 41 Salameh, Ali Hassan, 89, 110 Saleh, Saleh Mohammad, 154–55 Saudi Arabia, 33, 48, 92, 127; bin Laden clan and, 59; holy sites of Islam located in, 58; Khobar Towers bombing (1996), 10 Schaub, Gary, 27 Schelling, Thomas, 16, 18, 23–24 Schlosser, Eric, 62 Sedgai, Darim, 130, 131 al-Shabaab, 44, 47, 80, 94–95 shaheed (martyr), 4 Shahzad, Faisal, 86, 108 Shahzad, Syed Saleem, 104, 133, 134 Shaqaqi, Fathi, 89, 100 Sharif, Sayyid Imam al- (Dr. Fadl), 71, 72 Sharon, Ariel, 98 Shi’a Muslims, 55, 96 Shining Path (Sendero Luminoso), 55, 91, 96, 100, 187

shoe bomber, 101 signals intelligence (SIGINT), 99, 124, 180 signature strikes, 76 Singer, P. W., 77 small arms and rocket (SA/R) fi re: after Dadullah killing, 134–37, 141; after Matin killing, 157– 60, 162, 163– 64; after Mohammad (Qari Faiz) killing, 142, 143– 44, 147– 48, 235n27; mitigation of, 169 smart bombs, 25 Snidal, Duncan, 192 Snyder, Glenn, 28–29 Snyder, Jack, 31 Somalia, 6, 46, 73, 76, 80, 111, 190; civilian casualties of terrorism in, 178; drone strikes in, 78, 108; al Qaeda in, 41, 116; targeted killings in, 94; weak or failing state in, 47– 48; Western withdrawal from, 10 Soviet Union (USSR), 2, 15, 20, 173; Afghanistan war of, 126; Cold War deterrence and, 27, 31; collapse of, 1, 37; credibility of nuclear deterrence and, 23; limits of war with United States, 32; NATO strategy to defend against, 29; nuclear arsenal of, 19, 31; punishment of dissidents and partisans, 215n71; retaliation against Hezbollah, 216n76. See also Russia Spain, 97, 119 Special Air Ser vice (SAS), British, 90 Special Boat Ser vice (SBS), British, 76, 124, 156 Special Purpose Islamic Regiment (SPIR), 4 Sprinzak, Ehud, 7 Sri Lanka, 55, 104, 115, 178, 187 states: asymmetries with militant groups, 177–80; coercive dilemmas and, 13; credibility of military threats, 25–26; patron states and terrorists, 185, 213n38; as sponsors and hosts for terrorists, 46– 47, 52, 213n38; state sovereignty argument, 81; targeted killings and, 79–80 Stein, Janice Gross, 19, 21, 34, 203n30 Steinberg, Gerald, 56 strategic culture, 37 Sudan, 42, 44, 46, 90, 225n53 Sufi Muslims, 55 suicide bombings (suicide terrorism), 3– 4, 14, 45, 89, 118, 121; body-borne and

Index vehicle-borne, 194; civilian casualties and, 176; cost-benefit calculations and, 8, 21; deprofessionalization of, 100–101, 122, 136, 138; deterrence of individuals, 59– 60; failed attacks, 149, 237n55, 237nn44– 45, 242n16; individual bombers and organizations, 8; innovations in types of bombs, 44; in Israel, 62; mitigation and, 65; mythology of rational suicide, 21; 9/11 hijackers, 10; al Qaeda and, 67– 68; as retaliation for targeted killings, 112–13; social/family network of bombers, 56; strategic coercive logic of, 43; tailored deterrents and, 39; as Taliban tactic, 128, 144, 149–51, 158–59, 168– 69, 175; as theological perversion, 70 The Sullivans, USS, 10 Sunni Muslims, 55, 96 Superterrorism (Freedman, ed.), 3 Syria, 6, 46, 48, 215n71; civilian casualties in, 178; Hezbollah ties, 48, 53, 113; Kurdistan Workers’ Party (PKK) in, 46, 47; nuclear weapons program, 52; al Qaeda in, 41, 44, 116 takfir practice, 71, 104 Taliban, 12, 48, 98, 99, 190; bin Laden and, 50; cost-benefit calculations of, 192, 193; deprofessionalization of, 101; infighting in, 105; as insurgents and terrorists, 196–97; layeha (book of rules), 127–28; leadership shura of, 127–29, 233n12; in Pakistan, 49, 109–10, 156; popu lar sympathy in Afghanistan for, 176; prisoner exchanges with, 123–24; professionalism of, 131, 145–50, 158– 61, 164, 170–71, 193; al Qaeda and, 51, 93; U.S./NATO “Most Wanted” campaign (2007) and, 13–14, 126–31, 130 Tancredo, Thomas, 58 Tanzania, U.S. Embassy bombing (1998), 10, 91, 93, 94 Tanzim, 89 targeted killings, 11–12, 75; assassination associated with, 78–81; civilian casualties and, 87, 108–10; as coercion, 120–22, 174–77, 192–93; defined, 76–88; efficacy of, 95–108; inefficacy of, 108–20; international condemnation of, 110–13; leadership redundancies and, 41– 42;

253

legality (American), 83– 88; legality (international), 81– 83; practiced by United States, Israel, and others, 88–95 targeted killings, in Afghanistan, 12, 14, 92, 123–26, 131–32, 167–171; counterinsurgency (COIN) doctrine and, 195–96; of Dadullah, 132– 42; intra-national investigation of, 190–91; of Matin (Mullah Abdul), 156– 67; of Mohammed (Qari Faiz), 142–56; of most-wanted terrorists, 126–32, 130, 167 Tehrik-e-Taliban (TTP), 93, 105, 112, 126–27, 130; Bhutto assassination and, 166; strategic vision shared with Taliban, 196 Tenet, George, 68 terrorists/terrorism, 26, 37, 40; coercion of groups, 57–59; deprofessionalization of groups, 13, 100–3, 175, 227n83; deterrence by denial and, 29; domestic, 24; fi nancing of, 13, 39, 216n87; individual, 59– 61; infighting and purges in groups, 103–5, 118; intentions of, 19; Islamic scholars’ retracted support for, 71–72; lowered morale among, 105–8, 122; parts and processes of, 12–13, 38, 96; as rational actors, 5; religious, 3; “return address” of, 6; seen as irrational, 1–2; self-restraint of, 68, 69; stateless, 47; states’ ability to influence behavior of, 2; supportive communities of, 55–57, 67; target selection by, 13, 122, 131, 140, 170, 194; terrorism defined, 196–97; transnational, 24; as weak actors, 179 “Tracking America’s Drone War” (Washington Post webpage), 78 Trager, Brad, 59 training camps, militant, 77, 80, 125 Tunisia, 46, 89, 114, 185 Turkey, 46, 47, 91 Twitter/tweets, 44, 120–21 Uganda, 46 Ul-Haq, Anwar, 126, 130 “Unabomber,” 100 uncertainty, defensive measures and, 63– 64, 217n90 Understanding Terror Networks (Sageman), 41

254

Index

“underwear bomber” (2009), 86, 94 United Arab Emirates (UAE), 48 United Kingdom (UK), 53, 95, 110, 111 United Nations (UN), 27, 102, 140, 176, 197, 243n31; Department of Safety and Security (UNDSS), 199, 242n30; Development Program (UNDP), 199; French intervention in Mali and, 178; Security Council (UNSC), 45– 46, 91; targeted killings and, 81, 82, 111 United States, 20, 56, 57, 110, 165; civilian casualties and, 176; Cold War deterrence and, 27, 31; credibility of threats made by, 58; debate on targeted killings, 77; legality considerations about targeted killings, 83–88; limits of war with USSR, 32; nuclear arsenal of, 7, 9; Pakistan relations with, 99, 111; political will to dismantle al Qaeda, 178; precision attacks in response to terrorism, 46; resolve and willpower of, 23; as sole remaining superpower, 7; strategic doctrine of, 1; targeted killings practiced post-9/11, 92–95; threats against states linked to terrorism, 49; threat to invade Iraq, 27. Usmani (Osmani), Akhtar, 128–29, 133 Uzbekistan, 127 vengeance, 24, 57, 184 Vietnam War, 83–84 violence, 14, 20, 41, 113; actors superficially involved in, 61; calculated, 8; costs of participating in, 68; logic of deterrence and, 186; for political and social objectives, 7; self-restraint of terrorists and, 68; targeted killings and insurgent/ terrorist violence, 134– 68, 174–77, 238n63; type of, 131; without strategy, 6

war crimes, 111 warfare, 5, 82, 186; cumulative deterrence as war, 34; deterrence theory and, 1; gray zone between war and peace, 83, 87; guerilla warfare, 196; limited, 17 Wazi, Khalil al- (Abu Jihad), 89, 185 Waziristan, 92, 93, 94, 104, 126, 229n103 weapons of mass destruction (WMD), 27, 51–55, 114, 172, 182–83 Weinberger, Casper, 79 Wenger, Andreas, 35 West Bank, 57, 80, 81, 103, 106 Whiteneck, Daniel, 5, 52 Wiesel, Elie, 3 Williams, Brian, 109 World Trade Center bombing (1993), 10 Wright, Gary, 156 Yazid, Mustafa al- (Saeed al-Masri), 77, 93 Yemen, 6, 41, 44, 116, 190; attack on USS Cole, 10, 81, 94; civilian casualties of terrorism in, 178; drone strikes in, 76, 78, 85, 86, 92, 94, 108, 224n42; Western withdrawal from, 9–10 Yuldash, Tahir, 126, 130 Yusufzai, Rahimullah, 124–25, 142 Zagorcheva, Dessislava, 59 Zakaria, Fareed, 2 Zarqawi, Abu Musab al-, 69, 72–73, 97; elimination of, 92, 96, 132; on inability to hide, 98 Zawahiri, Ayman al-, 68, 69, 72, 87, 100; call for Pakistani jihad, 155; civilians killed in strike against, 108; failed drone strike against, 93; psychological operation against, 115–16; as Public Enemy No. 1, 97 Zero Dark Thirty (fi lm, 2012), 112–13