Institute for the Study of Asymmetric Conflict


Rational Choice Deterrence and Israeli Counter-Terrorism


Nations targeted by terrorist campaigns find themselves in a difficult situation: they feel tremendous pressure to respond in some way to terror attacks, but there is no obvious “correct” response. Passively absorbing continued attacks on civilian targets is not an attractive policy option, particularly in democracies where public dissatisfaction is a major input into government decision-making; but the alternatives to passive victimhood are often controversial, expensive, and inconvenient – and sometimes ineffective as well. It is only natural that societies on the receiving end of terrorism seek relatively painless and inexpensive ways of dealing with the phenomenon – ways in which the incidence and severity of terrorism can be reduced, and also ways in which the psycho-social impact of terrorism can be minimized.

In this light, the “holy grail” of counter-terrorism is deterrence. If enough potential terrorists can be convinced that carrying out attacks is a bad idea, terror attacks will cease; all the massive investment in target-hardening, security checks, and other means of reducing terrorism will no longer be necessary. Any strategy that provides hope of deterring terrorism is thus highly attractive. The risk, however, is that retributive policies justified as deterrents may be ineffective or even counter-productive.

Israel (and the Jewish community living there before Israeli independence) has been confronted by one of the world’s longest terrorist campaigns. Israeli/Jewish counter-terrorism efforts began in times of relative weakness before Israeli statehood, and today they continue as Israel enjoys unquestioned local military superiority. Israeli counter-terrorist policy is influenced by citizens’ reactions to terrorism itself, by Israelis’ reactions to the Holocaust and various pogroms that preceded and followed it, by moral sensitivities, and by the impact of international public opinion – which in turn has wavered between viewing Israel as the underdog and viewing Israel as an oppressive occupier. Obviously, Israel is strongly motivated to attempt to deter terror attacks; and over the years its leaders have tried in many ways to create such deterrence. Israel thus provides an excellent case-study for examining the potentials and pitfalls of deterrence as a counter-terrorism strategy.

Justifications for Punishment

Governments can act against terrorism in two very general ways: They can fight terrorism directly – by such means as reducing access to targets, intercepting terrorists before they can reach their targets, and other tactical approaches; and they can inflict punishment as a response to terror attacks. The line between punishment and tactical counter-terrorism is sometimes very difficult to draw: For example, Israel sets up roadblocks that impede Palestinian traffic, claiming that they are necessary to impede and intercept terrorists on their way to carry out attacks; but most Palestinians (and, apparently, a preponderance of foreign opinion) view these roadblocks as a form of collective punishment.

Lacking any way completely to disentangle tactical counter-terrorism from punishment, the best we can do is to analyze each counter-terrorist policy as a tactical measure (where appropriate); and, separately, as a punishment. This paper will deal only with the latter.

Punishment – the infliction of retributive harm – can be justified in several ways:

§  Just Deserts – A punishment may be inflicted in the service of “natural justice” – that is, on the basis that certain behaviors “ought” to be punished. Justice, of course, is an extremely slippery concept; however, while it may be impossible adequately to define justice (and it is certainly beyond the scope of this paper to do so!), it is quite easy to find or imagine examples of unjust punishment.

§  Deterrence – Punishment may be inflicted to prevent future misdeeds. Specific deterrence affects the future actions of the individual whose misbehavior is being punished; general deterrence reduces the probability that others will commit the same undesirable actions.

§  Revenge – Punishment is emotionally satisfying to those who feel they have been wronged by the misdeeds of the person (or group) being punished. This satisfaction may be quite independent of any abstract concept of the justice of the punishment meted out, and generally has little or no relation to the future consequences of the punishment.


These categories of justification are obviously not mutually exclusive. A just punishment may well have a deterrent effect, and is likely to be emotionally satisfying to members of the punishing society as well. On the other hand, it is quite possible for an unjust punishment to have a deterrent effect; it is even possible that unjust punishments may deter crimes of such severity that deterrence “justifies the injustice”. As a general rule, though, a punishment must provide some balance of justice and deterrence; revenge for its own sake is not generally considered an acceptable justification for punishment.

Rational Choice – a theory of deterrence

In modern criminology, the most widely accepted model for understanding deterrence is the Rational Choice Theory. According to Rational Choice, potential criminals are best understood as rational actors who weigh the costs and risks involved in committing a crime against the potential benefits of the crime. If punishment is swift, severe, and certain, few people will choose to disobey the law even if the potential illicit gain is substantial; but if punishment is uncertain, slow, or minimal, breaking the law can be seen as worthwhile even if the potential rewards are modest.

Rational Choice is an essentially economic theory of criminal behavior; and unsurprisingly, it is most successful in dealing with economically-motivated crime. Truly irrational crimes, such as intra-family murders or other crimes of passion, are beyond the effective scope of Rational Choice Theory; they are also notoriously difficult to deter.

At its most simplistic, Rational Choice Theory would seem to hold out the potential for creating precise mathematical formulae which would yield the correct punishment to deter any given crime. Reality, of course, is not so straightforward. Potential criminals do indeed weigh the risks, costs, and benefits before deciding on a course of action; but the factors they include in their intuitive calculations go far beyond the value of stolen goods and the income that would be lost due to imprisonment. The real-world factors are complicated, impossible fully to quantify, and are different for different individuals and at different times. (As a classic example of the variability of rational-choice deterrence, the social stigma of serving even a short prison sentence is likely to weigh much more heavily in the balance for a successful businessman than for an inner-city drug dealer.) So while it seems reasonable to expect people’s decision-making to be rational given their goals, beliefs, circumstances, social milieu, and so on, it can be very difficult to describe these factors accurately enough to create precisely “correct” deterrence.

Even though Rational Choice Theory cannot tell us precisely what will constitute an adequate minimum penalty to deter a particular crime, it can certainly explain the general effectiveness of punishment in deterring premeditated crime. In fact, without Rational Choice Theory or something much like it, the concept of deterrence would make no sense at all.

Is Terrorism a Rational Choice?

Given the nature of today’s terrorism – particularly suicide bombings and “suicidal shootings” – it is easy to label terrorists as irrational. Were terrorists truly irrational, it would be pointless to seek to deter terrorism – since deterrence requires a rational actor who can change his mind in order to avoid punishment. In fact, though, there is no convincing reason to consider even suicidal terrorists to be irrational; extensive studies (particularly those conducted by Ariel Merari) have failed to show any psychopathology associated with suicide terrorism. Further, terror attacks – particularly the “high-quality” attacks that result in many casualties – are not carried out impulsively, but are the result of long, meticulous planning, preparation, and indoctrination. The fact that potential suicide terrorists have changed their minds about carrying out an attack at all stages of the process, from early in their course of indoctrination until the moment before they were about to press the button to blow themselves up, shows that even the “craziest” terrorists have the ability and opportunity to make decisions based upon rational considerations.

At least in theory, then, it should be possible to deter terrorism. The question is not whether Rational Choice Theory (and deterrence in general) applies to terrorism; but rather, what strategies for creating deterrence against terrorism might be effective, and at what cost.

In evaluating the deterrent potential of various counter-terrorism strategies, the most obvious criterion of effectiveness is whether a particular strategy has shown itself to work in the real world. In some cases it is indeed possible to evaluate strategies on this basis; but more often than not, reality is too complex to yield unambiguous answers – particularly when particular strategies have both deterrent and tactical effects. Accordingly, we must look at deterrence from a theoretical as well as historical standpoint.

Individual Deterrence

Terrorism – and certainly the terror campaign against Israel in recent years – is seldom a purely personal affair. While attacks are ultimately carried out by individuals, these perpetrators have been recruited, trained, indoctrinated, equipped, and delivered to their target by organizations that have their own agendas and concerns. Further, terror attacks take place in a national and community context. So to analyze the potential for deterring terrorism, we need to examine how deterrence works – and fails – at all three levels.

Deterring the Suicide Terrorist

The most feared terrorist in Israel is the suicide bomber. The terrorist who has no thought of escape, who seeks only to find the most damaging place and time to detonate his explosives, is very difficult to intercept once he has begun the final approach to his target; and in addition to the objective impact of the casualties he causes, the very alien-ness of the suicide terrorist’s motivations creates tremendous insecurity. Clearly, suicide terrorists (and we can include in this category “suicidal shooters”, who do not kill themselves in the course of perpetrating an attack but fully expect to be killed by responding forces) are a prime target for deterrence. At the same time, they present one of the most obvious possible challenges to a retaliatory strategy: by the time any retaliation can be carried out, they are already dead.

Since punishing the guilty is clearly impossible in the case of a suicide terrorist, the only remaining option for creating deterrence is to punish the innocent. Such punishments can range from the mild to the draconian; the challenge for policy-makers is to balance the obvious injustice of punishing people who have done nothing (or very little) wrong against the benefits that effective deterrence would provide. This challenge is made especially difficult by the fact that the suicide terrorist has already demonstrated an extraordinarily high resistance to being deterred – most people, after all, would not blow themselves up for political purposes, even if they approve of others’ doing so.

Israel’s response to the challenge of deterring suicide terrorists was the demolition of the terrorist’s family home. The policy of home demolitions attempted to capitalize on the terrorist’s presumed reluctance to see his parents, siblings, and (sometimes) other relatives made homeless; however, a military investigating commission determined in early 2005 that these demolitions had never had any significant deterrent effect on suicide terrorism. (Those who favored these home demolitions frequently cited examples of parents who, to save their house, had alerted Israeli authorities to their child’s plan to carry out a suicide bombing; the IDF investigating commission discovered that there had in fact been very few of these cases – approximately twelve over several years of the “al-Aqsa Intifada”.) Home demolitions are unlikely to work as a general deterrent because they oppose transcendent benefits (furthering the terrorist’s national cause, enjoying the eternal pleasures of Paradise, and even automatic entry to Paradise for one’s relatives) with a mere economic punishment. Since his family house is likely to be replaced by the Palestinian Authority or foreign donors, and families of “shahids” receive special pensions and the like, the potential suicide bomber is likely to feel that “martyrdom” is an acceptably good deal even if his family’s house is demolished.

At the same time as it failed to deter suicide terrorists, Israel’s home-demolition policy damaged its international standing, caused internal dissention, and if anything increased Palestinian motivation to carry out terror attacks.

The fact that the policy of home demolitions failed to deter suicide terrorists does not mean that such deterrence is impossible to achieve. It is easy to imagine retaliatory policies that would create effective deterrence against suicide terrorism – for example, killing the terrorist’s parents rather than destroying their house. However, such policies would be so flagrantly unjust that they would never be acceptable to the Israeli public, much less the international community. We can generalize these observations as follows: Any policy of retaliation intended to deter suicide terrorism is likely to be either too weak to have any appreciable deterrent effect, or too harsh to be acceptable to the citizens of a democracy and the international community.

“Targeted Killing” as an Individual Deterrent

Terror-organization leaders, recruiters, bomb-makers, and other functionaries present another attractive target for deterrence; if they could be dissuaded from carrying out their tasks, terror organizations would not be able to function effectively. Actions taken against these terrorists, unlike retaliation against relatives of suicide terrorists, are reasonably defensible from a moral and political standpoint. Further, unlike retaliation against the innocent for suicide attacks, actions against living terrorists serve a tactical as well as a deterrent purpose.

Two general approaches have been used in dealing with non-suicidal terrorists: Arrest and imprisonment, and “targeted killing”. In effect, the two approaches sometimes overlap: many Israeli attempts to arrest terrorists end with the death of the target in a shootout rather than with his arrest. While this is not the stated goal of such arrest operations, it is clearly not viewed as an unacceptable outcome.

“Targeted killings” ought to be an effective deterrent. Death is a severe enough punishment to give most people pause, and members of terror organizations other than those being groomed to carry out suicide attacks are not known for being suicidal themselves. (It has often been noted that terror-organization leaders and functionaries never carry out suicide attacks; neither, in general, do their close relatives.) However, three factors reduce the deterrent effect of “targeted killings”:


§  Since “targeted killings” of terror-group leaders and functionaries are not, in general, a response to specific terror attacks, their value as an individual deterrent is somewhat limited. The possible victim of a “targeted killing” cannot avoid his death by refraining from carrying out a particular attack; he must renounce terrorism entirely and resign from his organization, or else try to turn his organization’s policy against terrorism. Either of these courses of action could well carry its own serious risks.

§  Even though they are not suicidal, terrorists – and particularly those who work with explosives – become accustomed to a high degree of risk. For a policy of “targeted killings” to be effective as an individual deterrent, the probability of being killed in this manner must be quite high.

§  For various political reasons, the use of “targeted killings” tends to be rather sporadic. Since deterrence is best created by swift, severe, and certain punishment, “targeted killings” that are undertaken only occasionally do not have the impact that they otherwise might – especially given terrorists’ acceptance of high risk.


When undertaken sporadically, “targeted killings” are likely to have some individual deterrent effect, but not one that is easy to measure or predict. Given the political costs of “targeted killings” (which are also referred to disparagingly as “extra-judicial executions”), it would appear that this tactic is best used either very selectively against especially high-value targets, or else as part of a concerted and intense campaign as described below.

Problems with Imprisonment

Arrest and imprisonment are the most common punishments for ordinary criminal behavior – but even for “normal” criminals, the deterrent effect of prison is somewhat uneven. For Palestinian terrorists, it is highly questionable whether imprisonment functions as a significant deterrent, although it may in some cases have a tactical benefit in preventing senior terrorist figures from operating freely.

To the extent that prison works as a deterrent to crime (as opposed to being simply a means of keeping the criminal off the streets for a while), it does so because of a number of ways in which it punishes the criminal:


§  He cannot earn a livelihood while in prison. This is particularly significant for criminals who are supporting a family.

§  He is removed from his family and social environment, and from his usual comforts and freedoms. This aspect of punishment is most effective when the criminal has family and friends, and lives in some comfort.

§  As a convict (and, later, as an ex-convict), the criminal acquires a social stigma. As noted above, having served time in prison can be anything from a major disgrace to a badge of honor, depending on one’s social milieu.


Palestinian terrorists imprisoned in Israel are considered by their own society to be prisoners of war, and receive benefits accordingly. Their families receive stipends to make up for their missing income; they are treated as heroes when they return home; and their release en masse is a standard demand of the Palestinian leadership in all negotiations with Israel. These factors do not completely negate the deterrent effect of imprisonment, but they certainly do reduce it.

Deterrence at the National/Community Level

As noted above, terrorism takes place in a social context. In societies that do not approve of terrorism, terrorist groups are either small or nonexistent; such terrorism as does exist is carried out by marginal extremist groups or by “rogue” individuals. Among Palestinians, however, support for terrorism is significant: Polling has shown approval of terror attacks against Israel among a segment of the Palestinian population ranging from a large minority to a substantial majority at various times.

Since anti-Israel terror organizations flourish in such an environment, it would of course be desirable to “dry up the swamp” – that is, to eliminate Palestinian tolerance of, and support for, terrorism. This could theoretically be accomplished in various ways: for example, by reducing Palestinian poverty, unemployment, and other discontents; by eliminating pro-terrorism propaganda or drowning it out with effective anti-terrorism propaganda; or by punishing Palestinian society for its support of terrorism, thus creating deterrence at the community level.

Collective punishment is manifestly unjust; and may well constitute a violation of “international law” when directed at a population under occupation. Officially, Israel does not engage in collective punishment, and never uses deterrence as a public justification for security measures that interfere with the daily lives of ordinary Palestinians. However, these security measures – roadblocks, closures, and so on – are unquestionably viewed as collective punishment by Palestinians, by much of the international community, and indeed by many Israelis. It would seem appropriate, then, to evaluate the effectiveness of these measures as a deterrent even if that is not their official purpose.

In order for deterrence to be effective, several things must be true:


§  There must be a clear causal connection in the minds of those to be deterred between the forbidden action and punishment.

§  There must be a clear and feasible course of action that will avoid punishment.

§  Punishment must be swift, severe, and certain.


Palestinian popular support for terrorism is largely passive, or active only in a very minor and diffuse way. For the Palestinian population as a whole, then – and even more so for any individual Palestinian noncombatant – there is no obvious connection between anything done or said and the punishments inflicted by Israel. The first condition for effective deterrence is thus unmet.

Although the Palestinian Autonomy conducts elections and is nominally democratic, it is not – or at least not yet – a fully functioning democracy. Ordinary citizens do not yet have effective and safe channels for bringing their complaints to those in power; accountability among leaders is only rudimentary (although, perhaps, increasing); official institutions do not have a monopoly of force; the news media are not yet fully free, independent, and skeptical; and political discourse is constrained in narrow channels, at least in public. (It is perhaps worth noting that in the recent Palestinian legislative election, no major party ran on a clear anti-terrorism platform.) Because there is no open, efficient mechanism for discussing a broad range of policy options and translating ordinary citizens’ concerns into effective government action, ordinary Palestinian citizens have no practical way to fight terrorism – and thus the second condition for effective deterrence is also unmet.

Israeli security measures are certainly severe and pervasive enough to create a great deal of unhappiness among ordinary Palestinians. However, given the very diffuse (at best) connection between the actions of ordinary Palestinians and these Israeli counter-terror measures, as well as the lack of any effective and convenient method of translating their discontents into new and effective government policies, it is hard to see any way in which this unhappiness can bring about the kind of changes that would significantly reduce Palestinian terrorism. While there have been scattered instances of resistance to terror organizations by ordinary Palestinians (largely as a result of terrorists’ blunder and bluster rather than by Israeli actions), it appears that Israel’s intrusive counter-terror measures have done more to increase Palestinian support for terrorism than to decrease it.

Organizational Deterrence

While organizations are composed of individuals, they act in many ways as independent entities, with collective interests that do not necessarily correspond to the personal interest of any particular member of the organization. For example, even terror organizations that glorify “martyrdom” and make extensive use of suicide attacks are not themselves suicidal; terror organizations, like any other type of organization, seek to perpetuate themselves – to the extent that some terror groups continue to survive and carry out attacks long after their ostensible “cause” has become obsolete.

In several ways, the terror organization is an attractive target for deterrence:


§  Unlike individual terrorists who may be indoctrinated to accept or even welcome “martyrdom”, terror organizations resist their own dissolution. There is no afterlife for organizations, and thus even religiously-motivated terror groups consistently pursue their earthly interests.

§  Terror organizations, especially those with a traditional hierarchical structure, tend to react in rational, predictable ways. The leaders of such organizations tend to be rather “professional” in their outlook; and while this makes them formidable adversaries, it also means that they will act to preserve their organization’s interests even when this means making ideological compromises.

§  Given sufficient motivation, the leadership of a traditional terror organization is capable of changing the organization’s tactics and even calling a halt to attacks.


At the same time, there are challenges facing a terror organization that decides to change its ways in response to the threat of punishment:

§  Terror groups derive much of their “street” legitimacy from the attacks they carry out. To cease to carry out attacks is to risk losing popular support. This can be countered, at least to some extent; for example, Hamas has generally adhered to a unilateral truce for almost a year, while maintaining its tough rhetorical line, continuing to threaten Israel with resumed attacks, and continuing its social-outreach activities (known as “dawa”) by providing educational, medical, and other services.

§  If the leadership of a terror organization decides to abandon terrorism, there is a possibility that part of the organization’s membership will split off and form a new, violent splinter group. (This was how the “Real I.R.A.” was created.) For deterrence at the organizational level to be fully effective, the organization’s leadership must be sufficiently strong to persuade the rank and file to follow its new policies.

§  If the organization’s leadership decides to abandon terrorism permanently, they have to find a new role for the organization or face its dissolution. Organizations that have been in the “terror business” for years find it difficult to reinvent themselves as purely political or social organizations.

Organizational vs. Individual Deterrence

When we discuss organizational deterrence, we need to understand that just as terror organizations have goals that are different from the goals of any particular individual, they also have different vulnerabilities. Organizations do not “feel pain” or other emotions; they generally react much more cold-bloodedly than individuals do. On the other hand, organizations need a constant supply of money and other material resources, as well as recruits, in order to survive. In order to function efficiently and maintain their ideological cohesiveness, organizations need to manage the development of their personnel, from new recruits to senior leadership. If any of these processes are sufficiently disrupted, an organization can lose its effectiveness or even disintegrate. 

The most obvious strategy for deterring terror organizations is to deprive them of key personnel. Imprisonment may be effective in this regard, but it is often very difficult to capture senior leaders of terror organizations; Israel has generally found it easier to kill terror-group leaders (especially given new technologies like armed unmanned aerial vehicles) than to arrest them.

A number of countries confronting terror campaigns have attempted to use a “decapitation” strategy, on the assumption that taking out the top leader of a terror organization will cause it to collapse. This has seldom worked. Most terror groups do not rely on one leader’s personality to survive, and can promote second-rank leaders to fill positions left open by “decapitation” attempts or other sporadic attacks on their leadership. A more effective strategy appears to be a focused campaign of “targeted killings” designed to eliminate enough senior operatives of a terror organization in a short enough time that the orderly process of promotion can no longer function effectively; if enough holes in the hierarchy open up at the same time, only relatively junior personnel lacking experience and reputation will be available to take over the leadership of the organization. Israel adopted this strategy in early 2004, killing Sheikh Ahmed Yassin and then Abdel Aziz Rantisi less than a month later. While Hamas has not disappeared or even renounced terrorism, the organization has behaved much more cautiously since these killings than it did previously.

While a strategy of deterrence based on the threat of aggressive attacks on a terror organization’s leadership can be effective against a traditionally-structured group like Hamas, it is less likely to work well against more chaotic, decentralized groups or “leaderless resistance” movements. Further, there is a risk that effective deterrence of traditional terror organizations may actually encourage the growth of these non-traditional terror movements.


Deterrence has been a long-standing goal of Israeli counter-terrorism; and yet, despite many years’ experience, Israel has not had very many successes in deterring terror attacks. Many of the strategies ostensibly designed to create deterrence are ineffective; subjected to analysis, they are rather obviously doomed to failure. The fact that some of these strategies have been pursued for years despite their high political cost and the lack of any good reason to believe that they would work suggests that claimed “deterrence” may sometimes be a way of rationalizing revenge.

The only strategy that shows real promise in creating effective deterrence against terrorism is to threaten terror organizations directly – not merely by sporadically attacking “key figures”, but by mounting (or at least realistically threatening to mount) a concerted attack on their entire leadership structure. While this approach is far from a panacea, it appears to work against traditionally-structured hierarchical terror organizations, which have been responsible for most anti-Israel terrorism. Whether such an approach is effective against terrorist networks like the Global Jihad is questionable, however.


Full Version