Deterring the Undeterrable: Coercion, Denial, and Delegitimization in Counterterrorism

Abstract

This article argues that deterrence theory can be applied to counterterrorism. Doing so requires broadening the traditional concept of deterrence by punishment, expanding deterrence by denial to include defense, mitigation, and strategic hindrance, and developing deterrence by delegitimization to influence the political, ideological, and religious rationales informing terrorist behavior. In practice, deterring terrorism requires tailoring threats against state and individual facilitators, diffusing the intended consequences of terrorism, and manipulating terrorist self–restraints. When these and other deterrent leverages are applied simultaneously against various actors and processes involved in terrorism, coercion can be achieved.

Key Words:

• Deterrence
• Coercion
• Terrorism

In 2006, President George W. Bush suggested that the cornerstone of American foreign policy – deterrence theory – was no longer a viable strategic principle. ‘Unlike the Soviet Union’, he remarked, ‘the terrorist enemies … hide in caves and shadows … have no borders to protect, or capital to defend. They cannot be deterred.’1 Deterrence skepticism reached a crescendo following 9/11. Generally, skeptics assume the following: (i) the fundamentalist religiosity of Al-Qa'eda and others negates rational decision-making; (ii) fanaticism creates diverging contextualizations and a penchant for risk-accepting, maximalist, and resolute behavior; (iii) an individual who actualizes a splendid life after death fears neither retaliation nor punishment; and (iv) non-state organizations lack a ‘return address’, a territorial target against which threats can be issued and fulfilled. The unsettling consensus borne of these arguments is that deterrence cannot be applied to combat terrorist organizations and that terrorism is altogether undeterrable.

While these findings intuitively resonate with policy-makers and the general public, they amount to opinion pieces rather than hard-nosed evaluations. Post-9/11 deterrence skepticism is based on instinct rather than rigorous theoretical argumentation. A more nuanced appreciation for the logic upon which deterrence theory is based and a more robust understanding of the terrorism phenomenon suggests a variety of deterrents can be levied against terrorist organizations.2 When tailored appropriately, states can utilize deterrence to influence the behavior of terrorist groups and delimit the type and ferocity of the violence they employ. While deterring terrorists will be more difficult than deterring state adversaries, both processes share the same inherent logic: an adversary's behavior is manipulated by applying coercive, diplomatic, or ideological leverages against his assets, goals, and beliefs.

Understanding why deterrence can be applied to counterterrorism is important for three reasons.

First, the subject represents an understudied feature of deterrence theory. Despite the prevalence terrorism now has in international affairs (to say nothing of the cumulative research program associated with deterrence) little scholarship has broached the topic effectively.

Second, identifying how deterrence theory can be applied to terrorism will lead to policy relevant findings. Success in counterterrorism requires two things: diminishing a group's capability to organize acts of violence and undermining a group's motivation to employ violence.3 While the former demands the use of military and defensive capabilities, the latter rests on deterrence, compellence, and influence. Today, counter-motivation strategies are underdeveloped.

Third, the current systemic environment is marked by the complexity of emerging threats. Deterrence today will require greater flexibility, accurate interpretations concerning the motivations, interests, and cultural norms of a variety of adversaries, tactical options that speak to a spectrum of threat contingences (from non-state to near-peer competitors), and a broadened interpretation of deterrence stability.4 This study, in evaluating one particular security dynamic – terrorism – helps refine the emerging deterrence paradigm.

This article posits that deterrence in theory and practice can be applied to counterterrorism. The argument is structured as follows. The first section introduces the tenets of deterrence theory and explains why 9/11 was not a deterrence failure but rather a failure to apply deterrence – an understated proposition that undermines deterrence skepticism. The second section suggests that a broadened conception of deterrence applied to an ‘organizational’ interpretation of terrorism reveals the mechanisms by which the logic of deterrence can be applied to counterterrorism. With that, deterrence by punishment, denial, and delegitimization are identified, in sections three, four, and five, respectively. Each case is informed by recent events. The article concludes by drawing out the theoretical implications of the analysis and suggests avenues for further research.

The Logic of Deterrence

The concept of deterrence is an ancient one; even the Romans understood that ‘if you want peace, make ready for war’.5 It was the nuclear revolution that brought the logic of deterrence under the microscopic scholarship of International Relations. Classical deterrence theory is a product of the Cold War, a conflict in which nuclear armaments necessitated that war avoidance take on new meaning. Thomas Schelling's oft-cited definition presents deterrence as ‘persuading a potential enemy that he should in his own interest avoid certain courses of activity’.6 Influencing behavior through fear involves two processes: deterrence–‘inducing an adversary … not to do something’, and compellence– inducing another ‘to do something.’7While the former deals with the ‘don'ts’ of inter-unit relations, compellence – as the root ‘compel’ suggests – deals with the ‘dos’. It requires that the defender convey what action is expected and how inaction will be met with retaliation. A useful way to distinguish the two is to relate them to a status quo. Wyn Bowen suggests that deterrence is meant to ‘preserve the status quo’– to prevent another from ‘initiating a course of action’– while compellence is based on altering an existing relationship.8 In either case, the objective is to prevent undesired behavior and coerce an adversary to comply with one's preferences.

Deterrence has been sub-divided into casual logics and processes.9 Most importantly for counterterrorism is Glen Snyder's distinction between deterrence by threat of punishment and deterrence by denial.10

Deterrence by punishment relies on threatening to harm something an adversary values. During the Cold War, Washington threatened to attack Moscow, and Moscow, Washington. Both were deterred because of the costs associated with initiating an attack. A complicating matter involving the calculus of punishment is the potential costs of inaction on the part of the challenger.11 While pursuing an unwanted action may result in punishment, acquiescing to a deterrent might also carry a cost. In order to be effective, then, a punishment must carry more weight than the cost of inaction in the challenger's calculus.

The second process, deterrence by denial, functions by reducing the perceived benefits an action is expected to provide. Hardening national or infrastructural defenses – what Sir Lawrence Freedman calls ‘passive’ defenses – raises the costs of attack by diminishing the probability that an adversary is likely to acquire his objective. While punishment deters an actor through fear, denial does so by creating a sense of ‘hopelessness’ on the part of the victim.12 Defense produces similar outcomes but differs from denial because it seeks to reduce the cost in case deterrence fails rather than attempt to influence another's behavior.

The logic upon which deterrence theory rests is on persuading a target that the costs of taking an action outweigh the possible benefits the action might provide. A deterrent is effective when it produces a cost-benefit calculus in which the expected utility of pursuing a given action is less than its expected cost. During the Cold War and in inter-state relations more generally, deterrence was most actively associated with the threat of retaliation. With the development and proliferation of nuclear weapons, punishment-based deterrents became the cornerstone of American (and Russian, Chinese, Israeli, etc.) deterrent strategies. By the 1960s, Freedman asserts, ‘the role of nuclear weapons’ became paramount in restraining state behavior; ‘so long as both sides [of a contest] were confident that they could inflict utter hell on the other, then a wider political equilibrium would be possible’.13 But deterrence by punishment, nuclear or conventional, represents but a tiny fraction of what is encapsulated by deterrence logic. So long as deterrence is based on manipulating an adversary's behavior, then non-punishment based processes that provide deterrent results are necessarily associated with the logic informing the theory.

Consider that offers of reward can induce an opponent to change his behavior in ways that suit a defender. Paul Huth and Bruce Russet contend that while ‘the inclusion of positive inducements as a means to deter is not standard practice in academic writings’, its exclusion ‘cannot be justified on grounds of strict logic’.14 Rewards, if meant to manipulate behavior, rest logically within deterrence theory. In his evaluation of Great Power credibility in Third World conflicts, Ted Hopf, finds that non-punishment influences played an important role in determining deterrence 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’.15 Hence, while threats are a central feature of deterrence, punishment is but a minor slice of what is theoretically and practically pertinent to constructing deterrent relations between actors.

Non-kinetic deterrents that rely on positive inducements and reassurance, denial and mitigation, and debate and delegitimization, while not generally associated with deterrence, nonetheless function and adhere to the theory's core tenets (changing behavior) and rely on its logical processes (manipulating cost-benefit calculations). Non-kinetic 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.

Deterrence Failures

Deterrence offers compelling arguments that inform the behavior of states in conflict scenarios. However, as Christopher Achen and Duncan Snidal argue, ‘its propositions are contingent: if the expected punishment exceeds the gain, then opponents will be deterred’, implying that ‘some conceivable punishment would deter, but not that any particular one will’.16 Put forcefully, the theory is able to predict cases where deterrence will fail.17 And yet deterrence failure in practice should not be confused with failure in deterrence theory. If and when threats are absent, muddled, miscommunicated, or weak, or the perceived benefit of action outweigh the cost of punishment, deterrence in practice may fail all the while deterrence in theory analytically succeeds.

The distinction between theoretical and practical deterrence failure is an important one when discussing terrorism. An intuitive assessment of 9/11 suggests the attack was a deterrence failure of grandest proportion, that Al-Qa'eda is undeterrable, and that deterrence theory is increasingly irrelevant. ‘Most of what we believed was true about deterrence’, Keith Payne offers, ‘is now misleading because international conditions have changed so dramatically.’18 And yet a more nuanced evaluation of 9/11 would have to ask whether or not the US had been actively, openly, and credibly attempting to deter Al-Qa'eda in the first place and whether Al-Qa'eda received and accurately interpreted these deterrents. On both counts, the answer is no.

First, prior to 2001 the US failed to issue a credible deterrent against Al-Qa'eda. Neither the Bush nor Clinton Administrations had done much to discourage Al-Qa'eda.19 In his many audio and written messages, Osama bin Laden describes the US 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 barrack and embassy bombings in Beirut and Kuwait City (1983/84); Soviet capitulation to the Afghan Mujahidin (1988/89); Western withdrawal from Yemen following the Gold Mohur bombings (1992); no 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 embassy bombings in Kenya and Tanzania (1998); no response to the failed attack on USS The Sullivans or the successful attack on USS Cole (2000).

For bin Laden, these events highlighted the West's incapacity and disinterest in effectively fighting terrorism. Bernard Lewis suggests that decades of counterterrorism neglect were interpreted as an ‘expression of fear and weakness rather than moderation’.20 Al-Qa'eda was encouraged by these tepid responses.21 By 2001, its experience suggested that it could target Westerners with impunity. Al-Qa'eda was undeterred from doing so not because it was irrational but rather because the US and its allies had given it little reason not to. Concomitantly, Al-Qa'eda neither properly understood nor fully appreciated the consequences of its actions.22 Marc Sageman suggests that it anticipated ‘a limited US response’ to 9/11.23 That the Bush administration acted otherwise, mustering a grand coalition, destroyed Al-Qa'eda's Afghan sanctuary, and eliminated thousands of operatives in a matter of weeks, was unanticipated.

Consider Abu al-Walid al-Masri's reaction. A leading Al-Qa'eda theoretician, he went on to denounce bin Laden for so recklessly misinterpreting US capabilities, for entangling Al-Qa'eda in a war it would lose, and for destroying the only purely Islamic emirate. Bin Laden, al-Masri concludes, was a shortsighted and ‘catastrophic leader’.24 In sum, Al-Qa'eda is not undeterrable, nor was 9/11 a deterrence failure. Rather, the US failed to properly apply deterrence theory to counterterrorism.

Expanding Deterrence

Applying deterrence to terrorism requires expanding the theory's scope beyond its current focus on punishment and nuclear weaponry. As suggested, deterrence is a broader concept. While critics like Paul Davis and Brian Jenkins initially argued that ‘deterrence is the wrong concept’ to contend with terrorism because it is ‘too limiting and too naïve’, they concede that ‘broadening the concept’ and adding ‘an influence component’ offers a way forward.25 The authors’ distinction between influence and deterrence is a little disingenuous, however. Deterrence is all about influencing opponents (broadly speaking) and not merely based on levying threats. To set deterrence up in minimalist garb and then concede 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’.26 Most scholars evaluating deterring terrorism agree, basing their investigations from a methodological position that expands the meaning and applicability of deterrence. For Michael Dunn, ‘influencing’ encompasses the concept of deterrence but is intended to ‘point to a broader set of actions that might be pursued than simply punishment or denial’.27 Traditional conceptions of deterrence matter, but Dunn's assumption suggests that an adversary's behavior can be manipulated in a variety of ways.

Michael Quinlan offers a solid, conceptual starting point. His position is based on a simple paradigmatic understanding: deterrence is rooted, elementally, to human affairs, applicable to all types of collectives, ranging from the state and nation 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.28

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 deterrence by punishment – may have. As Doron Almog contends, classical deterrence theory ‘is inapplicable to the war on terrorism’, though broader concepts of deterrence – Almog's own ‘cumulative deterrence’ included – remain pertinent.29 Herein rests the critical distinction, between throwing out deterrence theory in toto and expanding its logical boundaries in order to challenge certain aspects of the theory while developing, testing, and championing other theoretically related components.

For counterterrorism, the most practical interpretations of deterrence are the broad ones that suggest a range of terrorist activity – from suicide attacks against civilians to acquiring and using chemical, biological, radiological, or nuclear (CBRN) weapons – can be influenced by a variety of tailored deterrents. In practice, applying broadened deterrence to counterterrorism requires expanding the traditional ‘values’ associated with state-based deterrent relations, like sovereign control, territorial integrity, socio-political and economic survival. While most terrorists will place little import in these values, they will in others, like publicity, operational and tactical success, acquiring strategic and tactical aims, 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.30 While many terrorists lack territorial assets against which deterrents can be issued, they do have other values that can be threatened. However, because these values rest outside the realm punishable by military force, non-kinetic deterrent instruments that target these non-physical assets should also exist. Once uncovered, less traditional forms of deterrence that go beyond punishment-based strategies can be applied in order to coerce terrorists. Doing so requires that we disaggregate the associated actors and roles inherent to terrorism.

Unpacking Terrorism

Critics of rational deterrence theory note that utility can be measured differently by various actors. Richard Ned Lebow and Janice Gross Stein, for instance, argue that some actors ‘may rationally minimize expected losses while others may maximize expected gains’ and that ‘subjective expected utility’ is informed by predispositions towards risk-prone or risk-averse behavior.31 Other critics note that deterrence is a psychological process in which cognition, fear, pressure, fatigue, and other qualities inform how calculations are made and decisions taken.32 In like fashion, Robert Jervis contends that an adversary's assessment of costs, benefits, and probabilities can be misunderstood.33 In terms of deterring terrorist organizations, especially those infused with religious fanaticism, rational actor critiques are an important remonstrance.

A repeated challenge since 9/11 has been the perceived irrationality of modern super-terrorist organizations. John Gearson suggested that the ‘objectives’ of modern terrorists had ‘changed from achieving ends to simply punishment.’34 Mark Juergensmeyer argued that the ‘new terrorism’ appeared ‘pointless since it does not lead directly to any strategic goal’.35 And Jenkins lamented that while traditional terrorists ‘had a sense of morality, a self-image, operational codes, and practical concerns’, today's terrorist did not.36

More prescient interpretations of terrorism, however, suggest that it is used by ‘rational fanatics’.37 A number of studies by Robert Pape, for instance, suggest that suicide terrorism in particular follows a pattern of strategic logic. It is utilized, he explains, ‘to inflict enough pain and threaten enough future pain to overwhelm the target country's interest in resisting the terrorists’ demands’.38 The tactic is designed to acquire political goals, to compel a government to shift policies, and/or to wrest political control from a government or occupying force. What is more, Pape suggests that campaigns of suicide terrorism, when conducted against democracies, have a proven track record of success; governments are apt to capitulate to at least some terrorist demands. Though Pape's work has been criticized on theoretical and methodological grounds, his work does highlight the rationality informing terrorist activity.39 In like fashion, terrorist organizations like Al-Qa'eda have priorities and their leaders appreciate what it will take to achieve success. The violence employed is measured to affect change in antagonistic communities and generate support within friendly ones.

Furthermore, Al-Qa'eda has proven its capacity to learn from and reflect upon tactical errors and incorporate new strategies to evolving security environments. What it lacks in military sophistication, it makes up for in coercive innovation. Terrorist leaders are also technology-savvy propagandists who appreciate the ‘power of weakness’ (and relatedly, the ‘weakness of power’) and use their adversary's limitations to their benefit.40 And perhaps most importantly, Islamist terrorists have constructed their goals to coincide with the perceived grievances of a transnational community. In so doing, they have guaranteed themselves a structural resilience that goes beyond their limited membership, along with an expansive base from which to regroup, recruit, and rebuild.

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. Fawaz Gerges' portrayal of Islamist terrorism is especially enlightening. ‘Religion’, he writes, is a jihadist's ‘tool for political mobilization’. Power, rather than religious conviction, feeds the motivation. Likewise, Islamist political systems in Afghanistan, Iran, Sudan, and Gaza are distant cousins of secular authoritarian states. They are ‘clothed in Islamic dress’ but built on traditional political constructs.41 These findings are oddly comforting: if modern terrorism is a political rather than religious phenomenon, then the United States and its allies have the right strategic toolbox to manage emerging threats. What follows is an exploration of three deterrent strategies (punishment, denial, and delegitimization) that can be used to deter, compel, and influence terrorist behavior.

Deterrence by Punishment: Targeting what Terrorists Value

Coercing and Compelling States

Punishment strategies offer one avenue for influencing terrorist behavior. They will prove most practical at the state and individual levels. For starters, states that sponsor terrorist organizations can be threatened with military intervention and other punishments.42 This is a form of indirect deterrence, where punishment is threatened against a state in order to compel it to refrain from facilitating terrorism. This sort of interaction is not new; states have threatened and carried out military strikes against state-sponsors of terrorism well before 9/11.43 American strikes on Afghanistan and Sudan following the 1998 embassy bombings and Israel's retaliatory invasion of southern Lebanon in 1982, come to mind. What matters most in determining whether or not punishment-based deterrents and compellents on state sponsorship will succeed is the ‘proxy–patron relationship’: the extent and nature of the affiliation between the state and terrorist organization.44 Of critical importance is the degree of dependence terrorists have on state support and the level of affinity states share with their terrorist proxies.

In the first case, state-based punishments will work well in instances where terrorists are highly dependent on patron support. In these relationships, states can exert control over non-state proxies by withholding financial or structural assistance. They simple need to be compelled to do so. Bar, in his analysis of Israel's deterrent relationship with Hizballah, Fatah, and Hamas, illustrates that punishment strategies worked well against the former two (who were dependent on Iran, Syria, Lebanon, and the Palestinian Authority) but much less so against Hamas (which was the most independent).45 The more dependent a terrorist organization, the more likely punishment-based deterrents levied against sponsors will provide results. Pakistan represents a good case. One of the few states to have established diplomatic and military relations with the Taliban regime in Afghanistan in the 1990s, Islamabad was compelled otherwise by the United States soon after 9/11. Former President Pervez Musharraf revealed that Richard Armitage, then US Deputy Security of State, compelled Pakistan to cut ties with the Taliban by threatening military force. ‘Be prepared to be bombed’, Armitage is quoted as saying. ‘Be prepared to go back to the Stone Age.’46 The threat was credible given the severity of Al-Qa'eda's attack and the political mood in the US. Eventually, Pakistan acquiesced.

In the second case, 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 Daniel Byman as states that ‘knowingly allow’ terrorists to ‘raise money, enjoy a sanctuary, [and] recruit’ on their territory but do not directly assist them in organizing terrorism.47 While host states may derive security dividends by tolerating an organization's activity – through the harassment of a mutually antagonistic neighbor, for instance – they nonetheless refrain from sponsoring terrorism. Under such conditions, a credible punishment may compel hosts to cease their passivity and clamp down on groups active on their territory. In the 1990s, Turkey endured attacks by the Kurdistan Workers' Party (PKK) based in Syria, which facilitated but did not 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 passive support for the PKK. Rather than absorb the punishment, Syria expelled the PKK's Abdullah Ocalan, leading to his 1999 capture in Kenya by the Turkish authorities.

There is a third, though less common, scenario. Occasionally, terrorists effectively usurp the decision-making apparatus of their host, 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 sub-state actors. For example, in the late 1990s, the Taliban took many of its policy cues from Al-Qa'eda. Freedman has aptly suggested that the Al-Qa'eda–Afghanistan relationship was ‘not so much a case of state-sponsored terrorism but of a terrorist-sponsored state’.48 A similar phenomenon took place between the Islamic Courts Union and Somalia between 2005 and 2007. The Palestinian Authority, under Yasser Arafat, occasionally did the same with Hamas and Syria increasingly looks toward Hassan Nasrallah, Secretary General of Hizballah, for guidance.

In these cases, state-based punishments may fail to provide the necessary incentive that compels a state to cease facilitating terrorism. The paradox of reverse proxyship, though, is that the more state-like a terrorist group becomes the more effective threats against it directly are. That is, the higher the degree of ‘stateness’ (i.e. centralized leadership, political aspiration, civil involvement, territorial control) the more palpable coercive punishments. Israel's uncompromising blow against Hizballah (a veritable ‘state within a state’) in 2006 helped deter the organization from again attacking Israel in 2008/9 as its military bombarded Hamas in Gaza. Tellingly, just months before the Gaza conflict, Major General Gadi Eisenkot of Israel's Northern Command warned Hizballah that ‘what happened in the Dahiya quarter of Beirut’ (a Hizballah 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.’49 The threat was credible given Israel's 2006 offensive and Hizballah understood the cost of renewed engagement: harm to its social, infrastructural, military, territorial, and political assets. During the ensuing Gaza conflict, all of half a dozen rockets were launched from Lebanon.

Coercion and CBRN Terrorism

Punishments might also deter states from sharing CBRN weapons with terrorist organizations. In general, deterring CBRN transfers – the likeliest manner in which terrorists will gain access to these weapons – is easier to do than deterring support for conventional terrorism because the stakes are simply higher.50 The goal is to deter terrorist acquisition by associating the clandestine sponsorship of CBRN terrorism as one wholly directed by the supporting state itself. The day North Korea detonated a nuclear device, President George W. Bush responded by warning President Kim Jong-il that the ‘transfer’ of weapons ‘would be considered a grave threat’ and that North Korea would be held ‘fully accountable’.51 Washington's goal was to deter Pyongyang from dolling out its weapons. Strengthening attribution capabilities that use forensic science to trace CBRN materials to their sources of origin will help signal to potential state-sponsors that they cannot easily transfer weapons anonymously.52

And yet, threatening retaliatory punishment in response to CBRN terrorism is problematic. For starters, attacking CBRN-capable states for supporting terrorism will almost surely invite retaliation. Likewise, threatening Russia or Pakistan (or for that matter, India, France, or the UK) in response to CBRN terrorism stemming from lapses in security, is simply not credible.53 Not only will these states likely retaliate, but threats damage multilateral ventures meant to safeguard stockpiles and risk encouraging proliferation. Furthermore, forensic and attribution sciences are far from perfect. There is continued uncertainty, for instance, as to the source of Libya's clandestine nuclear program, despite the fact that Tripoli supplied Washington with a uranium gas sample in 2004.54

Finally, deterring state transfers neglects the role non-state intermediaries, criminal networks, rogue government elements, and individuals, like A.Q. Khan, have in facilitating CBRN terrorism. The dilemma was reiterated by Stephen Hadley, National Security Advisor under President Bush. The US ‘will hold any state, terrorist group, or other non-state actor fully accountable’, he suggested, for assisting CBRN terrorism, ‘whether by facilitating, financing, or providing expertise or safe haven’.55 Some actors involved in CBRN terrorism, like criminal networks and individuals like Khan, are usually 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 risk-averse and will respond better to deterrents that increase the cost of doing business than other state-based facilitators. If so, of all actors potentially involved in CBRN terrorism, this subset may be the most susceptible to the logic of deterrence by punishment.56

Coercion and Individuals

State-based punishment strategies will prove ineffective in the case of weak, failing, and collapsed states. Foreign Policy's 2009‘Failed State Index’ suggests that well over 30 states lack full territorial control. Somalia, Sudan, Iraq, Afghanistan, Pakistan, Bangladesh, Sri Lanka and Lebanon, are on the shortlist – a veritable who's who of international terrorism.57 Weak polities like these may be willing to support terrorists in exchange for financial or security gains or, in truly dysfunctional environments, may have little choice but to accept terrorist predation on their territory. Terrorists appreciate this and use domestic chaos borne of state weakness to their benefit. When coercing states becomes impractical, threatening other elements of the terrorist organization can provide deterrent results. To that end, individual terrorists can be threatened with death, sanctions, capture, and other punishments. Saudi Arabia, for example, revoked Osama bin Laden's citizenship and froze his assets in 1994. It further compelled his brother, Bakr, to denounce him publicly on behalf of the family. In 1996, Sudan, as a result of Saudi and Egyptian pressure, expelled him. And the United States, between 1998 and 1999, tracked his movements extensively.58

Targeted killings are another form of coercion that can be directed against individual terrorist leaders and facilitators. Targeted killings are the ‘intentional slaying’ of individuals ‘undertaken with explicit governmental approval’.59 Though targeted killings are legally and morally contentious and potentially degrade particular norms of international behavior, they are employed by a number of states in over a dozen conflicts.60 While the elimination of mega-terrorists – like the Taliban's Mullah Dadullah and Baitullah Mehsud, Al-Qa'eda's Abu Laith al–Libi, Al-Qa'eda in Iraq's Abu Musab Zarqawi, and Al-Shabaab's Aden Ayro – make headlines, targeted killings are being utilized far more often than is generally acknowledged: Israel has conducted over 250 since 2000; American and NATO forces have targeted individuals in Afghanistan, Yemen, Somalia, Pakistan, and Iraq on several hundred occasions since 2002; and Russia has carried out several dozen strikes against Chechnyan groups since 2005.61 For the most part, those targeted have included high-ranking terrorist leaders and commanders along with operators, facilitators, bomb-makers, and mid-level organizers.

While targeted killings degrade an organization's coercive capability by removing personnel, leadership assets, and military knowledge, they also diminish individual motivation and influence group behavior. The first process is generally understood. The literature on targeted killings suggests that they reduce command and control, impede communication, force facilitators underground, restrict freedom of movement, degrade operational capability, create power vacuums, instigate purges and feuds, decrease recruitment, and altogether de-professionalize terrorist organizations.62 With diminished capability, groups find it more difficult to coordinate sophisticated attacks and to obtain their tactical and strategic objectives.

Less understood is the second process. The assumption is that eliminations represent a significant cost to participating in terrorism; a cost that translates into a deterrent. Brad Roberts explains that ‘the leaders of Al-Qa'eda … are inspired by martyrdom – but not their own’.63 That they are keen to survive is an exploitable characteristic. The threat of elimination has a psychological impact: fear can be paralyzing. Eliminations also remind terrorists of the state's superior abilities and reiterate that death can come suddenly. Targeted killings also lower morale. A string of eliminations can cause despair among surviving members. Eventually, lowered morale can diminish a group's ability to recruit. Likewise, life on the run can get tiring. ‘Once-powerful motives to join’ an organization, writes Christopher Harmon, ‘do not always translate into certainty about staying.’64 Fatigue can set in, followed by a diminution in the desire to carry on with terrorism.

Of all punishment-based deterrents, targeted killings are the easiest to communicate effectively. Ensuring targeted killings retain a deterrent quality requires that the threat be perceived as credible and its use clearly communicated. Doing the former involves demonstrating an ability to locate, track, and attack wanted individuals repeatedly, establishing credibility through iterated action. Unlike interstate deterrence, in which credibility is often communicated through declarations and the signaling of capabilities, counterterrorism deterrence derives from the repeated use and demonstration of particular capabilities.65 Each targeted killing communicates the state's technological, intelligence-gathering, and military ability to do as it threatens. In Afghanistan, for example, the Taliban's Shura Council has been heavily attrited: Akhtar Usmani, Mullah Obaidullah, Mullah Dadullah, and Mullah Berader have all either been captured or killed since 2006. These successes have communicated the Coalition's willingness, capability, and ‘intelligence dominance’ to target the Taliban's leadership. The result has been a shift in Taliban structure, motivation, professionalism, and behavior.66 Effectively communicating a policy of targeted killings can be accomplished by publishing and disseminating wanted lists. Israel did so during the Al-Aqsa Intifada (2000–05) when officials passed information to the Palestinian Authority, which shared it with various sub-state organizations. On occasion, wanted men voluntarily placed themselves in custody to avoid being attacked. In Afghanistan, the Coalition began a ‘most-wanted’ campaign in 2007, offering rewards for information on various individuals known to be facilitating terrorism in the region. In both cases, the threat of elimination was communicated effectively and wanted individuals understood that they were marked for death or capture.

Deterrence by Denial: Targeting what Terrorists Want

Deterrence by denial provides a second manner with which to coerce terrorists. Denial manipulates an adversary's behavioral calculus by preventing the desired effects of an attack. Within nuclear dyads, denial mechanisms proved less practical because of the difficulty in credibly neutralizing the consequences of nuclear exchange. In conventional military environments and particularly in counterterrorism, however, denial mechanisms become realistic. In the context of NATO engagement in Europe, John Mearsheimer explains that deterring conventional Soviet aggression was a result of NATO's ability to ‘prevent the Soviets from winning a quick and decisive victory’.67 By ensuring that any European conflict would be a costly one, NATO denied the Soviets the value of initiating a Blitzkrieg. In countering terrorists, denial mechanisms work in a similar fashion: they create conditions where the achievement of goals seems unattainable. By removing tactical or strategic options, denial mechanism force terrorists to make decisions on less-preferred behavior, plans, and outcomes.68

Counterterrorism denial mechanisms exist. First, defensive denial functions by restricting and constraining the terrorism processes. By augmenting structural defenses around potential targets, 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. In theory, the more difficult a target is to attack, the less likely it will be. In practice, defensive deterrence requires properly assessing what targets terrorists most want to attack and building specific defenses that challenges the utility of doing so. During the Al-Aqsa Intifada, Hamas, Palestinian Islamic Jihad, the Al-Aqsa Martyrs Brigade, and others began by first attacking Israelis within Israel with suicide bombers dispatched against soft targets. Transportation hubs, cafés, bars, restaurants, shopping malls and markets were repeatedly targeted in the first half of the conflict. In response, Israel began defending public access points to bus and train depots, restaurants and bars began placing guards outside their doors, blast-proof entrances were added to buildings, checkpoints and security barriers were established, and privately-owned shuttles began offering alternative transportation services in major cities. The cumulative effect was the restriction of easily accessible soft targets by 2004. Eventually, suicide bombers were forced to target military and police checkpoints and to detonate outside and off target. The result was a diminishment in the utility of suicide attacks in Israel and an eventual reduction in their use.

Defensive denial need not rest on structural defenses alone. Behavioral defenses work by introducing environmental uncertainty into terrorist planning. The US Department of Homeland Security suggests that ‘variability and unpredictability must be consciously injected into flexible prevention measures’.69 The deterrent target in question is the terrorism process itself; unpredictability impedes terrorist planning by introducing greater levels of uncertainty. And uncertainty translates into a denial mechanism. In practice, this involves conducting spot checks at public transpiration hubs, establishing a police presence at randomly selected city intersections, and/or instructing security vehicles to leave their emergency lights on, giving the impression of an overwhelming security presence. These defenses influence behavior because terrorists are generally risk-averse while preparing for attacks; they obey the law, are less likely to steal, speed, or take drugs, and avoid attention. In this case, establishing a perception that plans will not easily succeed provides the deterrent.

Deterrence by mitigation functions in a similar way but involves blunting and limiting terrorism's social, political, and economic effect. The idea is to deny the immediate consequences terrorists anticipate and desire. Broad mitigating factors include a robust economy, an effective political system, and a resilient society able to withstand the shocks of terrorism. More specific factors include first response and public health systems able to manage the immediate consequences of violence. Both sets of factors challenge the utility of terrorism and make it a less attractive option. When terrorism has little lasting effect, highlighting its futility will diminish its use. As Colin Gray concludes, ‘if we are to perform competently in deterrence we need to address … how, by our policies, we can negate the political effects of tactically successful terrorism’.70 By not rewarding terrorism, we challenge its efficacy.

Denial extends further to impeding a group's long-term socio-political objectives. The premise rests on demonstrating that terrorism does not lead to the attainment of goals and is counterproductive. Doing so requires emphasizing the strategic futility of violence.71 While related to mitigation, this strategy goes beyond limiting the specific tactical effect of terrorism to denying the socio-political outcome it is meant to produce. Freedman suggests that ‘over time, doubts can be inserted into the minds of would-be terrorists that … particular method[s] of promoting their cause [are] … less effective than they might have expected’.72 The strategy is to increase the costs of acquiring political goals. States can put terrorist objectives out of reach by providing other states targeted by a common adversary with financial and military assistance, dissuading states from acquiescing to terrorist demands, supporting sub-state groups that oppose terrorist agendas, and altogether restricting terrorist activity.

‘From a policy perspective,’ suggest Robert Trager and Dessislava Zagorcheva, ‘the ability to hold political ends at risk’ is important because doing so ‘stands by far the best chance of fracturing the global terrorist network’.73 When violence becomes counterproductive, groups may develop more benign strategies to advance their goals. The Palestinian case is informative. Terrorism against Israeli civilians, Max Abrahms illustrates, ‘has not worked to advance Palestinian national aspirations or quality of life’ over the long term.74 Instead, Palestinian violence strengthened Israel's resolve to combat terrorism, helped it mend its domestic cleavages, and delegitimized the Palestinian cause. These are strategic losses. That no functioning Palestinian state yet exists speaks volumes regarding the long-term utility of terrorism. It is not inconceivable that eventual fatigue will set in (as it has already with members of the Palestinian Liberation Organization and Fatah) and that political agendas may eventually eclipse violent ones.

Identifying the factors that determine whether or not denial mechanisms will influence terrorist behavior is nonetheless troublesome. One lesson drawn from Israel's engagement with suicide terrorism in Lebanon during the 1980s suggests that denial mechanism established in theater were instrumental in forcing Hizballah to reassess its utility. In his testimony before the US House of Representatives, Ariel Merari suggested that suicide attacks against the Israel Defense Forces in Lebanon dropped ‘very, very significantly’ after 1985 because measures adopted by the Israelis ‘proved effective in preventing most of the suicide attacks’, leading those organizing them to cease their use because they were ‘not bringing any results.’75

While this case is instructive, more recent ones contradict denial expectations. Bruce Hoffman notes that the 2006 Al-Qa'eda plot to detonate liquid bombs on transatlantic flights questions the value of denial mechanisms. Airplanes are perhaps the most hardened soft target, nearly impossible to attack effectively. Theoretically, Al-Qa'eda should have been deterred from directing its energies against airlines because of the difficulty involved in doing so and the exceptionally high risk of operational failure. Though the 2006 plot was foiled, Al-Qa'eda was not deterred from trying.76 And it again targeted airlines in 2009, with Umar Farouk Abdulmutallab's near miss over the skies of Detroit, Michigan, and in 2010, with the cargo planes plot. A similar finding is offered by the 2008 Al-Qa'eda-linked attack on the US Consulate in Istanbul, Turkey. American diplomatic offices are heavily guarded, exceptionally difficult to attack, and offer very low odds of success. In this particular case, for instance, three of the four assailants were killed outside the facility's entrance. Nonetheless, Al-Qa'eda was not deterred from facilitating the attack.

Both examples suggest that the utility of attacking some targets go beyond ease of access and odds of success. The value of targeting hardened soft targets is perhaps less about the outcome of the attack than about the sensationalism of the attack itself. Al-Qa'eda repeatedly targets commercial aircraft despite the difficulty in doing so because the plots themselves generate value. Success is a bonus, not a necessity. Speaking of the consulate attack, Ihsan Bal suggests that Al-Qa'eda ‘chose one of the best protected buildings in Turkey, not because they wanted to blow it up, but because they knew it would attract world attention’.77 This speaks to the counter-coercive power of terrorism. Counter-coercion is a challenger's ability to counteract and interfere with a defender's deterrent.78 These leverages diminish the potency of a deterrent by placing costs on a defender's preferences and neutralizing threats. This is an underappreciated area of investigation that is highly pertinent to deterring terrorism because terrorists retain enormous counter-coercive and counter-deterrent strengths. They have the ability to diffuse, stem, and limit the effect of particular deterrents in ways that undermine state capabilities and resolve. Small attacks translate into disproportionate security losses, and even foiled attacks are usually perceived as deterrence failures.79 Moreover, terrorists' counter-coercion capabilities are not usually contradicted by pointing to counterterrorism successes.

At issue is that deterrence is not a static game, where a defender constructs and implements a deterrent and the challenger simply weighs the costs and benefits of absorbing particular threats. He, too, can act and behave in ways that influences the defender's utility calculus in order to manipulate his willingness to pursue particular deterrents. Gauging how best to deter terrorists through denial will require assessing how individual groups value attacking particular targets, how they measure tactical successes, and how they weigh unsuccessful/foiled attacks in their utility calculations. This will require detailed information, data, and intelligence on terrorist preferences.80 Only then will denial mechanisms be specifically and properly tailored to challenge particular terrorist values consistently.

Deterrence by 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 his goals by attacking the legitimacy of the beliefs that inform his behavior. While it seems self–evident, as David Lake believes, that ‘terrorists lack moral strictures against the use of violence’, this is only half the story.81 Most terrorists base their activities on a set of principles. Ideological and socio-religious beliefs not only inform terrorists' behavior but also their goals. Al-Qa'eda's use of suicide terrorism, for example, is legitimized by relying on religious decrees that justify and sanitize the taking of one's own life. These same decrees help shape Al-Qa'eda's socio-political goals. Of importance, however, is that suicide is an otherwise blasphemous act under Islamic law and Al-Qa'eda's objectives are refuted by a vast majority of those that share the Muslim faith. Applying the logic of deterrence to these inherent divisions suggests that specific leverages might be developed that delegitimize and ridicule the rationales and goals informing terrorist violence. Strengthening and disseminating opinions, positions, and information that contradicts the legitimization of terrorism might deter or compel individuals contemplating and/or taking part in violence along with the socio-religious groups that facilitate terrorist efforts.

To be sure, delegitimizing terrorism is far removed from the traditional concepts of deterrence and compellence, but its purpose and practical application share common theoretical and logical themes. Like deterrence, delegitimization attempts to change an adversary's behavior but it does so by manipulating the rationales and justifications that inform his preferences. The objective is to cause an adversary to alter or abandon a preferred behavior not by threatening punishment or denying goals, but by degrading the rationales that motivate and guide his behavior. Delegitimizing terrorism involves influencing popular support, debating religious interpretations, and manipulating strategic culture.

Delegitimizing mechanisms raise the costs of participating in terrorism by challenging the normative, religious, and socio-political rationales individuals rely upon when participating in violence. Deterrence by delegitimization starts by tapping into terrorists' ‘self-restraints’ and magnifying their role in utility calculations that shapes violent behavior.82 Terrorist organizations regularly refrain from certain actions, especially evident in the case of CBRN weaponry.83 John Parachini suggests that religious beliefs instruct the degree to which some Islamists are willing to go in acquiring unconventional capabilities. Citing Hamas' Abu Shannab, who explained that ‘the use of poison was contrary to Islamic teachings’, Parachini notes that CBRN is considered illegitimate by some terrorist organizations.84 In the case of Hamas, though a fanatically dogmatic terrorist group that has no qualms conducting mass-casualty attacks against civilians, its behavior is bound by self-restraint. Consider further 2003 revelations by George Tenet, former Director of the Central Intelligence Agency, that Al-Qa'eda had organized a cyanide attack in New York City. Ayman al-Zawahiri called it off because he feared that the attack ‘was not sufficiently inspiring to serve Al-Qa'eda's ambitions’ and might humiliate the organization.85 Like Hamas, Al-Qa'eda's violence is measured. Both cases suggest that under certain conditions, normative guidelines delimit terrorist activity. The compellent objective, in this case, is to expand the scope of the existing CBRN norm, reinforce adherence to it, and induce others not yet bound by its guidelines to accept its rationale.

Defenders can manipulate a challenger's self-restraint by communicating how certain actions contradict religious tenets and social expectations. This can be done by illustrating how and why CBRN terrorism would galvanize global counterterrorism efforts rather than fulfill terrorist goals, emphasizing that CBRN terrorism would kill Muslims alongside non-Muslims. This would cross religiously accepted boundaries as a result, relating CBRN use to Muslim revolt against rather than support for terrorism, and compelling religious scholars to contradict justifications for CBRN use.86 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. According to Colin Gray, ‘terrorists lose when their outrages delegitimize their political causes’.87

Consider how Zarqawi's barbarism in Iraq was rebuked by Zawahiri for having alienated the broader Arab and Muslim community, whose support Al-Qa'eda relied on for its survival.88 Zarqawi's 2005 attack in Amman solicited similar protest from the Arab street. In killing over 60 mostly Muslim Jordanians, Zarqawi was vociferously denounced; residents of his hometown even called for his death.89 The lesson here is that though Zarqawi organized tactical successes, the way he did so damaged Al-Qa'eda's long-term strategic goals. Turning popular 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 socio-political community to condemn violence.90

Delegitimization can also be applied to the religious dialogue informing Islamist terrorism. ‘Speaking for Islam,’ writes Freedman, ‘is bin Laden's objective, so this is a war about the future of Islam.’91 In an evaluation of Al-Qa'eda's ‘strategic culture’, Jerry Long illustrates further that though bin Laden ‘functions as a kind of lay mujtahid’– one who interprets religious texts – he is ‘not free to give just any interpretation, or else he would have no legitimacy’.92 The point is that Islam is multifaceted and is punctuated by religious debate concerning what is and is not legitimate behavior.93 To be sure, bin Laden's holy war is considered religiously illegitimate by hundreds of millions of Muslims. Al-Qa'eda understands this limitation and goes out of its way to provide the theological justification for its behavior. That it continues to do so suggests that debates have yet to be won.

Lewis demonstrates that violent jihad is carefully regulated by Sharia 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, Lewis reiterates, do religious texts ‘enjoin terrorism and murder’.94 It is enlightening to juxtapose these religious duties with the ferocity of the unholy terror that is being employed today. Islamic validation for modern terrorism is based on a deviation, a selective (re)interpretation, and dismissal of religious texts.

Consider suicide bombings. While the ‘fedayee(n)’ (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 Islamic suicide bomber dies ‘by his own hand’.95 This is a theological perversion relying on unjustifiable forms of violence (indiscriminate mass killing) and suicide (a sin). Ensuring individuals contemplating suicide terrorism are aware of these differences might deter some from carrying out attacks.

Manipulating religious debates in ways that delegitimize interpretations that condone terrorism can help inform broader deterrent strategies. Doing so requires debating and ridiculing individual legitimizers. As Hoffman instructs, ‘religious terrorism demands vastly revised … diplomatic and cultural strategies’.96 The focus today is on fundamentalist interpretations of Islam, particularly its Salafi strain. Legitimizers operate within these traditions, commenting on how actions correspond to the dictates of religious jurisprudence. Because diverging interpretations exist, defenders can rely on religion itself to compel terrorists to forgo violence. Delegitimization strategies might embolden debates with decrees and fatwas that stigmatize Al-Qa'eda, its leaders, and followers. Consider that immediately after 9/11, bin Laden was rebuked for failing to fulfill theological requirements pertaining to jihad.97 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.

Prominent jihadi scholars have also retracted their support for terrorism. Sayyid Imam Sharif (aka Dr Fadl), a onetime leader of Egypt's Islamic Jihad and an early Al-Qa'eda 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.98 In 2007, Saudis Sheikh Su'ud al-Rushud and Sheikh al-Askar condemned Al-Qa'eda'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’.99 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.’100 And Nasir Abas, a former leader of Jemaah Islamiyah, suggested that ‘not one verse in the [Koran] contains an order for Muslims to make war on people of another religion’.101

By 2008, Al-Qa'eda was compelled to expend half its airtime defending its legitimacy.102 Bin Laden himself reiterated that ‘the Muslim victims who fall during the operations against the infidel Crusaders … are not the intended targets’ and Zarwahiri countered Dr Fadl with a treatise.103 The risk for Al-Qa'eda is that condemnations of this caliber influence once-supportive 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 will influence the behavior of would-be supporters. When Islamist organizations lose religious justification, their activities resemble thuggery, their objectives become suspect, and their support is eliminated.

Conclusion

Applying deterrence theory to terrorism is a promising field of research. This study illustrates how the traditional tenets of deterrence (punishment and denial) and less conventional notions (delegitimization) can be utilized to coerce, compel, and influence terrorist behavior. Three principle policy recommendations follow.

First, post-9/11 deterrence skepticism is misplaced. While it is true that deterring terrorism will be more difficult to do than deterring the Soviet Union, targeting what terrorists value, desire, and believe will influence the type and ferocity of the violence they organize. Doing so will require developing innovative strategies that combine punishment with non-punishment tactics and applying them against the spectrum of state, community, and individual facilitators.

Second, success in counterterrorism involves undermining the motivation that informs an organization's willingness to employ violence and its supporting community's enthusiasm to facilitate it. Doing so requires a better understanding of the religious, cultural, and ideational factors and ‘local conditions’ that foster terrorism.

Third, deterrence theory must be better expressed in counterterrorism practice and policy. The complexities of the threat demand a more robust understand of how deterrent leverages will work in practice, how they will be applied to a given situation, and how they might interact with other counterterrorism strategies.

Further research is needed in four areas. First, theorists will have to address the practical prerequisites of deterrence and evaluate whether and how they challenge the theory's applicability to counterterrorism. Successful deterrence is conditional: a defender must (i) clearly define the action to be avoided, (ii) communicate a willingness to punish violations, (iii) retain the capability to punish and/or deny as threatened, and (iv) demonstrate resolve to retaliate if demands are not met.104 These are essential conditions of deterrence; doing otherwise invites failure. In countering terrorism, fulfilling each might prove difficult. It is not clear, for instance, what action defenders are trying to deter – is it all acts of terrorism in all political and geographic contexts? Likewise, communicating deterrents with decentralized, non-hierarchical, and non-state adversaries is a challenge – who is it states expect to communicate with and what control do these actors have over the nature, scope, and ferocity of the terrorism threat? And finally, signaling capabilities along with resolve will be taxing – what relevant actions, statements, ultimatums, and ‘costly signals’ will be effective? These practical dilemmas will need to be addressed, first in theory and then in practice.

Second, future research will have to work around another even more daunting dilemma. Deterrence theory assumes that acquiescence on the part of the challenger will be met with inaction on the part of defender. If a challenger expects some form of punishment even after having complied with a deterrent threat, the costs of inaction and action are indistinguishable and deterrence will fail. What matters is restraint– refraining from unwarranted punishment – and assurance– guaranteeing passivity. In countering terrorism, both will prove problematic because deterrent strategies are pursued alongside anti-terrorist actions. While necessary, anti–terrorism negates assurances of restraint. The United States has vowed to crush Al-Qa'eda and its associates. If terrorists were to acquiesce with Western demands, could they realistically believe that American forces would stop hunting them down? Strategies that disable terrorists but do not simultaneously weaken deterrents will have to be explored.

Third, methodological dilemmas involved in measuring deterrence success and failure will have to be addressed. This is related to longstanding dilemmas concerning the development of practical matrices for identifying deterrence in practice. Lebow and Stein argue that testing deterrence theory is problematic ‘because of the difficulties inherent in identifying deterrence successes, which leave few if any behavioural traces, and of inferring the intentions of would be challengers’.105 Intention to attack is hard to positively identify. It is even more difficult to show conclusively that a defender had something to do with a non-event. Circumventing these dilemmas in studying deterring terrorism requires precise knowledge of an adversary's intention and capability. Future studies identifying deterrence outcomes in counterterrorism will have to build on terrorist declarations, statements, and doctrine and illustrate how specific deterrents led to changes in terrorist capability (weaponry, infrastructure, recruitment, leadership), motivation (frequency of attacks, following through with threats) and tactics (sophistication and professionalism of attacks).106

Finally, deterrence says very little as to how conflict situations cease. ‘Deterrence theory, explains Patrick Morgan, ‘is about cancelling a challenger's desire to attack and not about how to end the conflict.’107 While deterrence offers a respite to violence, it does little to improve the environmental conditions that require deterrence's continual use. Accordingly, interdisciplinary research that better situates the logic of deterrence to counterinsurgency, socio-political development, and peace studies will help suggest ways in which deterrence might not only influence behavior but also inculcate and diffuse pacific norms and create the conditions that lead to peace rather than war.108

Acknowledgements

For their thoughtful comments, I am indebted to Frank Harvey, T.V. Paul, Denis Stairs, Major-General E.S. Fitch, Andreas Wenger, Victor Mauer, and the journal's anonymous reviewer. For their generous support, I thank the Security and Defence Forum at the Canadian Department of National Defence and the Center for Security Studies at the ETH, Zurich, Switzerland.

Notes

¹George W. Bush, Commencement Address, US Military Academy, 27 May 2007.

²Terrorism is the use of violence by non-state actors against noncombatants with the purpose of generating fear to communicate (and achieve) socio-political objectives. This definition implies that terrorism, as Bruce Hoffman notes, creates ‘psychological repercussions beyond the immediate victim’. Terrorism relates well to deterrence logic, suggests Sir Lawrence Freedman, because it is a ‘coercive strategy’ meant to compel states by threatening pain. Delineating it from other forms of political violence (insurgency, civil and guerrilla warfare, genocide) is nonetheless problematic because of the overlap that exists between these phenomena. See Bruce Hoffman, Inside Terrorism (New York: Columbia UP 2006), 40; Lawrence Freedman, Deterrence (Malden, Canada: Political Press 2004), 121–2; Daniel Byman, ‘Understanding Proto-Insurgencies’, Journal of Strategic Studies 31/2 (April 2008), 167–70; David Kilcullen, ‘Countering Global Insurgency’, Journal of Strategic Studies 28/4 (Aug. 2005), 597–617.

³Daniel Byman, The Five Front War: The Better Way to Fight Global Jihad (Hoboken, NJ: Wiley 2008), Chs. 2 and 7.

⁴Freedman, Deterrence, 54–9, 99–100; Keith Payne, ‘Bush Administration Strategic Policy: A Reality Check’, Journal of Strategic Studies 28/5 (Oct. 2005), 779–81; Payne, ‘Deterrence: A New Paradigm’, National Institute for Public Policy (Dec. 2003); T.V. Paul, ‘Complex Deterrence: An Introduction’ in T.V. Paul et al. (eds), Complex Deterrence: Strategy in the Global Age (Chicago UP 2009), 8–20.

⁵Michael Quinlan, ‘Deterrence and Deterrability’, Contemporary Security Policy 25/1 (April 2004), 11.

⁶Thomas Schelling, The Strategy of Conflict (Cambridge, MA: Harvard UP 1960), 9

⁷Thomas Schelling, ‘Thinking about Nuclear Terrorism’, International Security 6/4 (Spring 1982), 72.

⁸Wyn Bowen, ‘Deterrence and Asymmetry: Non-State Actors and Mass Casualty Terrorism’, Contemporary Security Policy 25/1 (Spring 2004), 58.

⁹The most pertinent include narrow and broad deterrence, extended and central deterrence, immediate and general deterrence, and nuclear and conventional deterrence. For an overview, see Freedman, Deterrence, Ch. 2.

¹⁰Glen Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton UP 1961), 13–19.

¹¹Frank Harvey, ‘Practicing Coercion: Revisiting Successes and Failures Using Boolean Logic and Comparative Methods’, Journal of Conflict Resolution 43/6 (1999), 842–3.

¹²David Johnson et al., Conventional Coercion across the Spectrum of Conventional Operations (Santa Monica, CA: RAND 2002), 17.

¹³Freedman, Deterrence, 16.

¹⁴Paul Huth and Bruce Russett, ‘Testing Deterrence Theory: Rigor Makes a Difference’, World Politics 42/4 (July 1990), 471.

¹⁵Ted Hopf, Peripheral Visions: Deterrence Theory and American Foreign Policy in the Third World, 1965–1990 (Ann Arbor: UP 1994), 241. See also, Jeffrey W. Knopf, ‘Three Items in One: Deterrence as Concept, Research Program, and Political Issue', in T.V. Paul et al. (eds), Complex Deterrence: Strategy in the Global Age (Chicago UP 2009), 33–7.

¹⁶Christopher Achen and Duncan Snidal, ‘Rational Deterrence Theory and Comparative Case Studies’, World Politics 41/2 (1989), 152.

¹⁷Frank Harvey, ‘Rigor Mortis or Rigor, More Test: Necessity, Sufficiency, and Deterrence’, International Studies Quarterly 42/4 (1998), 676–83.

¹⁸Keith Payne, Testimony before the House Armed Services Committee, Subcommittee on Strategic Forces, Washington DC, 18 July 2007.

¹⁹Jeffrey Knopf, ‘Wrestling with Deterrence: Bush Administration Strategy after 9/11,’Contemporary Security Policy 29/2 (Aug. 2008), 232–3.

²⁰Bernard Lewis, ‘Free at Last? The Arab World in the Twenty-first Century’, Foreign Affairs 88/2 (March/April 2009).

²¹A similar assessment is offered by Shmuel Bar and others concerning Israel's supposed deterrence failure against Hizballah at the onset of hostilities in 2006. Shmuel Bar, ‘Deterring Non-state Terrorist Groups: The Case of Hizballah’, Comparative Strategy 26/5 (2007), 472; Amos Malka, ‘Israel and Asymmetrical Deterrence’, Comparative Strategy 27/1 (2008), 2–5; Barak Mendelsohn, ‘Israeli Self-Defeating Deterrence in the 1991 Gulf War’, Journal of Strategic Studies 26/4 (Dec. 2003), 92.

²²It is possible that Al-Qa'eda devised 9/11 to compel the US to further its global military intervention in hopes of catalyzing greater socio-political upheaval in the Muslim world and/or in order to bait and bog down its opponent in costly asymmetric conflicts.

²³Marc Sageman, Understanding Terror Networks (Philadelphia: Univ. of Pennsylvania Press 2004), 51.

²⁴Quoted in Fawaz Gerges, ‘Are We Safe Yet? A Foreign Affairs Roundtable’, <www.foreignaffairs.com/discussions/roundtables/are-we-safe-yet> (Sept. 2006).

²⁵Paul Davis and Brian Jenkins, Deterrence and Influence in Counterterrorism: A Component in the War on al Qaeda (Santa Monica, CA: RAND 2002), 9–13, 61; and Paul Davis and Brian Jenkins, ‘A System Approach to Deterring and Influencing Terrorists’, Conflict Management and Peace Science 21/1 (2004), 3–15.

²⁶Johnson et al., Conventional Coercion, 11.

²⁷Lewis Dunn, ‘Influencing Terrorists’ Acquisition and Use of Weapons of Mass Destruction’, NATO Defence College Workshop, 5 Aug. 2008, 2. See also, Dunn, ‘Deterrence Today: Roles, Challenges, and Responses’, Proliferation Papers, IFRI (2007), 17–22; and Dunn, Next Generation: Weapons of Mass Destruction and Weapons of Mass Effects Terrorism (Defense Threat Reduction Agency, Advanced Systems and Concepts Office 2008), sect. 3.

²⁸Quinlan, ‘Deterrence and Deterrability’, 11.

²⁹Doron Almog, ‘Cumulative Deterrence and the War on Terrorism’, Parameters 34/4 (Winter 2004/5), 5.

³⁰‘A Concept for Deterring and Dissuading Terrorist Networks’ (unclass. draft report), US Office of the Secretary of Defense (2005), 7–9.

³¹Richard Ned Lebow and Janice Gross Stein, ‘Rational Deterrence Theory: I Think, Therefore I Deter’, World Politics 41/2 (1989), 208–9.

³²Jeffrey Berejikian, ‘A Cognitive Theory of Deterrence’, Journal of Peace Research 39/2 (2002), 167–73.

³³Robert Jervis, ‘Deterrence and Perception’, International Security 7/3 (Winter 1982/83), 3–30.

³⁴John Gearson, ‘The Nature of Modern Terrorism’, in Lawrence Freedman (ed.), Superterrorism: Policy Responses (Malden, MA: Blackwell 2002), 11.

³⁵Mark Juergensmeyer, ‘Understanding the New Terrorism’, Current History 99/636 (2000), 158.

³⁶Brian Jenkins, ‘The New Age of Terrorism’, in David Kamien (ed.), McGraw-Hill Homeland Security (Santa Monica, CA: RAND Reprint 2005), 118.

³⁷Ehud Sprinzak, ‘Rational Fanatics’, Foreign Policy (Sept./Oct. 2000), 66–73.

³⁸Robert Pape, ‘Suicide Terrorism and Democracy: What We've Learned Since 9/11’, Policy Analysis 582 (2006), 4; idem, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House 2005).

³⁹Scott Atran suggests that Pape's 2005 dataset of over 300 suicide attacks (1980–2003) was outdated by 2006. Using updated data he finds that suicide terrorism is not meant to compel adversaries but to increase a ‘sponsoring organization's political “market share”’ among supportive communities. Others have criticized Pape on methodological grounds. See Scott Atran, ‘The Moral Logic and Growth of Suicide Terrorism’, Washington Quarterly 29/2 (2006), 132–9; Atran, ‘Mishandling Suicide Terrorism’, Washington Quarterly 27/3 (2004), 67–90; Scott Ashworth et al., ‘Design, Inference, and the Strategic Logic of Suicide Terrorism’, American Political Science Review 102/2 (2008), 269–73.

⁴⁰Amos Malka, ‘The Power of Weakness vs. The Weakness of Power: Asymmetrical Deterrence’, Institute for Policy and Strategy, Hezliya Conference, Israel, Jan. 2008.

⁴¹Fawaz Gerges, Journey of the Jihadist: Inside Muslim Militancy (Toronto: Harcourt 2007), 11–4, 39–45.

⁴²Bruce Jentleson and Christopher Whytock, ‘Who “Won” Libya? The Force–Diplomacy Debate and its Implications for Theory and Policy’, International Security 30/3 (2005/6), 51–5.

⁴³Martha Crenshaw, ‘Coercive Diplomacy and the Response to Terrorism’, in Robert Art and Patrick Cronin (eds), The United States and Coercive Diplomacy (Washington DC: US Institute of Peace 2003), 314–35.

⁴⁴Bar offers four relationships: surrogates (groups are an extension of the state), proxies (groups do the bidding of the state); partnerships (groups form alliances with the state); and reverse proxyship (groups inform state behavior). Shmuel Bar, ‘Deterring Terrorists: What Israel Has Learned’, Policy Review 149 (June/July 2008), 29–42.

⁴⁵Shmuel Bar, ‘Israeli Experience in Deterring Terrorist Organizations’, Institute for Policy and Strategy, Herzliya Conference, Israel, Jan. 2007.

⁴⁶Tim Reid, ‘We'll bomb you to Stone Age, US told Pakistan’, The Times Online, 22 Sept. 2006.

⁴⁷Byman 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 (Winter 2005/6), 118.

⁴⁸Lawrence Freedman, ‘The Third World War?’, Survival 43/4 (Winter 2001), 74.

⁴⁹Quoted in Reuters, ‘Israel Warns Hezbollah War Would Invite Destruction’, 3 Oct. 2008.

⁵⁰Caitlin Talmadge, ‘Deterring a Nuclear 9/11’, Washington Quarterly 30/2 (2007) 21–34; Elbridge Colby, ‘Expanded Deterrence: Broadening the Threat of Retaliation’, Policy Review 149 (June/July 2008), 43–59.

⁵¹David Sanger and Thom Shanker, ‘US Debates Deterrence for Nuclear Terrorism’, New York Times, 8 May 2007.

⁵²William Dunlop and Harold Smith, ‘Who Did It? Using International Forensics to Defeat and Deter Nuclear Terrorism’, Arms Control Today 36 (Oct. 2006), 6; Siegfried Hecker, ‘Toward a Comprehensive Safeguards System: Keeping Fissile Material out of Terrorists’ Hands', The ANNALS 607 (2006), 121–32; Nuclear Forensics: Role, State of the Art, and Program Needs (Washington DC: American Physical Society & American Association for Advancement of Science 2005).

⁵³Michael Levi, ‘Deterring State Sponsorship of Nuclear Terrorism’, Council Special Report 39 (2008), 4–5.

⁵⁴Matthew Phillips, ‘Uncertain Justice for Nuclear Terror: Deterrence of Anonymous Attacks through Attribution’, Orbis 51/3 (2007), 434–5.

⁵⁵Stephen Hadley, ‘Remarks to the Center for International Security and Cooperation Center for International Security and Cooperation’, Stanford Univ., California, USA, 8 Feb. 2008.

⁵⁶David Auerswald, ‘Deterring Nonstate WMD Attacks’, Political Science Quarterly 121/4 (2006), 555–9.

⁵⁷ Foreign Policy, ‘Failed States Index 2009’ (Summer 2009).

⁵⁸Government of the USA, 9/11 Commission Report (22 July 2004), 62–3; PBS Frontline, ‘Interview with Michael Scheuer’ in The Cell Next Door, Jan. 2007.

⁵⁹Steven David, ‘Fatal Choices: Israel's Policy of Targeted Killing’, Mideast Security and Policy Studies 51 (2002), 2; and Nils Melzer, Targeted Killing in International Law (Oxford: OUP 2008), 3–8.

⁶⁰For normative critiques, see Freedman, Deterrence, 118–20; Asa Kasher and Amos Yadlin, ‘Assassination and Preventive Killing’, SAIS Review 25/1 (2005), 41–57; Yael Stein, ‘By Any Name Illegal and Immoral: Response to ‘Israel's Policy of Targeted Killing’, Ethics and International Affairs 17/1 (2003), 127–37; Naomi Chazan, ‘Assassinations as Weapons of War’, Israel Journal of Foreign Affairs 2/2 (2008), 85–90.

⁶¹Graham Turbiville, ‘Hunting Leadership Targets in Counterinsurgency and Counterterrorist Operations’, Joint Special Operations University Report 07–6 (2007), 52–73; Peter Cullen, ‘The Role of Targeted Killing in the Campaign against Terror’, Joint Force Quarterly 48 (2008), 23–9.

⁶²See, among others, Daniel Byman, ‘Do Targeted Killings Work?’Foreign Affairs 85/2 (2006), 95–111; Michael Eisenstadt, ‘“Pre-Emptive Targeted Killings” as a Counter-Terror Tool: An Assessment of Israel's Approach’, PeaceWatch 342 (28 Aug. 2001); 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 29/4 (2006), 359–82; Hillel Frisch, ‘Motivation or Capabilities? Israeli Counterterrorism against Palestinian Suicide Bombings and Violence’, Mideast Security and Policy Studies 70 (2006), 1–32; Gal Luft, ‘The Logic of Israel's Targeted Killing’, Middle East Quarterly 10/1 (2003), 6–13.

⁶³Brad Roberts, ‘Deterrence and WMD Terrorism: Calibrating its Potential Contributions to Risk Reduction’, Institute for Defense Analyses Paper P-4231 (Alexandria, VA: 2007), 13.

⁶⁴Christopher C. Harmon, ‘The Myth of the Invincible Terrorist’, Policy Review 142 (2007), 60.

⁶⁵For the role iteration plays on inter-state deterrent relations, see Jonathan Shimshoni, Israel and Conventional Deterrence: Border Warfare from 1953 to 1970 (Ithaca, NY: Cornell UP 1988).

⁶⁶Alex Wilner, ‘Targeted Killings in Afghanistan: Measuring Coercion and Deterrence in Counterterrorism and Counterinsurgency’, Studies in Conflict and Terrorism 33/4 (April 2010), 307–29.

⁶⁷John Mearsheimer, ‘Prospects for Conventional Deterrence in Europe,’Bulletin of the Atomic Scientists 41/7 (1985), 158.

⁶⁸See Gary Geipel, ‘Urban Terrorists in Continental Europe after 1970: Implications for Deterrence and Defeat of Violent Nonstate Actors’, Comparative Strategy 26 (2007), 439–67; Robert Anthony, ‘Deterrence and the 9-11 Terrorists’, Institute for Defense Analyses Document 2802, May 2003; M. Elaine Bunn, ‘Can Deterrence be Tailored?’, Strategic Forum 225 (2007), 3.

⁶⁹US Dept. of Homeland Security, Report of the Future of Terrorism Task Force (Washington DC: Jan. 2007), 6.

⁷⁰Colin Gray, ‘Thinking Asymmetrically in Times of Terror’, Parameters 32/1 (Spring 2002), 8–11.

⁷¹James Smith and Brent Talbot, ‘Terrorism and Deterrence by Denial’, in Paul Viotti, Michael Opheim and Nicholas Bowen (eds), Terrorism and Homeland Security (New York: CRC Press 2008), 54–9.

⁷²Freedman, Deterrence, 123–4.

⁷³Robert Trager and Dessislava Zagorcheva, ‘Deterring Terrorism: It can be Done’, International Security 30/3 (2005/6), 60–1, 89.

⁷⁴Max Abrahms, ‘Are Terrorists Really Rational? The Palestinian Example’, Orbis 48/3 (2004), 542; and Max Abrahms, ‘Why Terrorism does not Work’, International Security 31/2 (2006), 72–5.

⁷⁵Ariel Merari, Testimony before the Special Oversight Panel on Terrorism, US House of Representatives, 13 July 2000.

⁷⁶Personal correspondence with Bruce Hoffman, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland, 10 March 2009.

⁷⁷Associated Press, ‘Investigators Eye al-Qaeda Link in Embassy Attack’, 10 July 2008.

⁷⁸Daniel Byman and Matthew Waxman, ‘Defeating US Coercion’, Survival 41/2 (1999), 108–15.

⁷⁹I am grateful to Frank Harvey for providing this observation. For reference, see Frank Harvey, ‘Getting NATO's Success in Kosovo Right: The Theory and Logic of Counter-Coercion’, Conflict Management and Peace Science 23/2 (2006), 150–1.

⁸⁰Louise Richardson, What Terrorists Want (New York: Random House 2006).

⁸¹David Lake, ‘Rational Extremism: Understanding Terrorism in the Twenty-First Century’, Dialog–IO (Spring 2002), 17.

⁸²Roberts, ‘Deterrence and WMD Terrorism’, 8–17.

⁸³Nancy Kay Hayden, Terrifying Landscapes: A Study of Scientific Research into Understanding Motivations of Non-State Actors to Acquire and/or Use Weapons of Mass Destructions, Defense Threat Reduction Agency, Advanced Systems and Concepts Office (June 2007), 18–21.

⁸⁴John Parachini, ‘Putting WMD Terrorism into Perspective’, Washington Quarterly 26/4 (Autumn 2003), 45.

⁸⁵Quoted in Eric Schmitt and Thom Shanker, ‘US Adapts Cold-War Idea to Fight Terrorists’, New York Times, 18 March 2008.

⁸⁶Lewis Dunn, ‘Can al-Qaeda be Deterred from Using Nuclear Weapons?’, Occasional Paper 3, Center for the Study of Weapons of Mass Destruction (2005), 24; Daniel Whiteneck, ‘Deterring Terrorists: Thoughts on a Framework’, Washington Quarterly 28/3 (Autumn 2005), 188–94.

⁸⁷Gray, ‘Thinking Asymmetrically’, 7.

⁸⁸Shmuel Bar and Yair Minzili, ‘The Zawahiri Letter and the Strategy of Al-Qaeda’, in Hillel Fradkin et al. (eds) Current Trends in Islamist Ideology, Vol. 3 (Washington DC: Hudson Institute 2006); and Douglas, ‘Waging the Inchoate War’, 401–9.

⁸⁹Fawaz Gerges, ‘Buried in Amman's Rubble: Zarqawi's Support’, Washington Post, 5 Dec. 2005.

⁹⁰Patrick Porter, ‘Long Wars and Long Telegrams: Containing Al-Qaeda’, International Affairs 85/2 (2009), 300–5.

⁹¹Freedman, ‘The Third World War?’, 63.

⁹²Jerry Long, ‘Strategic Culture, Al-Qaida, and Weapons of Mass Destruction’, Defense Threat Reduction Agency, Advanced Systems and Concepts Office (2006), 9.

⁹³Dunn, Next Generation, 11–14, 25–6.

⁹⁴Bernard Lewis, The Crisis of Islam: Holy War and Unholy Terror (New York: Random House 2004), 39.

⁹⁵Ibid., 153

⁹⁶Bruce Hoffman, ‘Old Madness, New Methods: Revival of Religious Terrorism Begs for Broader US Policy’, Rand Review 22/2 (1998/99), 15.

⁹⁷Giles Kepel, The War for Muslim Minds (Cambridge, MA: Harvard UP 2005), 100–4.

⁹⁸Lawrence Wright, ‘The Rebellion Within: An al-Qaeda Mastermind Questions Terrorism’, New Yorker, 2 June 2008; Ronald Sandee, ‘Core Al-Qaida in 2008: A Review’, NEFA Foundation (2009), 12–15.

⁹⁹Asaf Maliach, ‘Saudi Religious Scholars come out against Al–Qaeda's Use of Religious Edicts Permitting Suicide Attacks against Muslims’, International Institute for Counter–Terrorism, Herzliya, Israel (July 2007).

¹⁰⁰Peter Bergen and Paul Cruickshank, ‘The Unraveling: The Jihadist Revolt against bin Laden’, New Republic, 11 June 2008.

¹⁰¹Atran, ‘The Moral Logic’, 142.

¹⁰²Shaun Waterman, ‘Al Qaeda Adopting Defensive Tone’, Washington Times, 13 Aug. 2008.

¹⁰³Bergen and Cruickshank, ‘The Unraveling’; Sandee, ‘Core Al-Qaida in 2008’, 11–12.

¹⁰⁴Frank Harvey, ‘Practicing Coercion’, Journal of Conflict Resolution 43/6 (1999), 840–3.

¹⁰⁵Richard Ned Lebow and Janice Gross Stein, ‘Deterrence: The Elusive Dependent Variable’, World Politics 42/3 (1990), 336–43.

¹⁰⁶Almog, ‘Cumulative Deterrence’; and Wilner, ‘Targeted Killings in Afghanistan’.

¹⁰⁷Patrick Morgan, ‘Taking the Long View of Deterrence’, Journal of Strategic Studies 28/5 (Oct. 2005), 754–62.

¹⁰⁸Freedman has done so with regard to inter–state deterrent relations. See Freedman, Deterrence, Ch. 4.