NATIV Online  

Vol. 1  /  2003                                            A JOURNAL OF POLITICS AND THE ARTS  


The Bomb in the Basement:
Reconsidering a Vital Element of Israeli Nuclear Deterrence

Louis René Beres

During the middle-1980s, some students of Middle Eastern security issues began to speak plainly of Israel’s nuclear strategy with particular reference to the question of disclosure.1 Specifically, these scholars asked whether this strategy should continue to be implicit, deliberately ambiguous, and in the “basement”, or whether it should be explicit, clearly articulated, and out in the open.2 I entered the “debate” personally with a series of lectures at Israeli strategic studies centers in 1984 and 1985 and with the first edited book on the subject in 1986.3 Today, this debate no longer seems vital to Israeli strategists. As it is perfectly obvious, they reason, that Israel has some significant numbers of nuclear weapons, there really is nothing further to disclose. Israel is patently a member of the Nuclear Club. Everyone already knows this. Case closed!4

But there is a serious problem with such reasoning. The rationale of disclosure, of taking the bomb out of the “basement”,5 would not lie simply in expressing the obvious. Rather, it would lie in the informed understanding that nuclear weapons can serve Israel’s security in a number of different ways, and that all of these ways could benefit the Jewish state, more or less, to the extent that certain aspects of these weapons and associated strategies were disclosed. Indeed, as we shall now see, the pertinent form and extent of disclosure6 could soon be more critical than ever before because of the sorely misnamed “Peace Process”.

For the foreseeable future, Israel’s state enemies – especially Iran and Syria (but not excluding Egypt and Jordan) – continue to enlarge and improve their conventional and unconventional war-fighting capacities. Although no one can say for certain that such improvements are underway with especially Israel in mind, it would be prudent for Jerusalem to assume the worst. Moreover, even if enemy-state intentions do not yet parallel capabilities, this could change very quickly. Here, for example, enemy capabilities could determine intentions, occasioning chemical, biological7 or nuclear first-strikes against Israel because of expected tactical advantages.

To protect itself against enemy strikes, especially those strikes that could carry existential costs, Israel must exploit every component function of its nuclear arsenal.8 In this connection, the success of Israel’s efforts will depend in large part upon not only its particular configuration of “counterforce” and “countervalue” operations, but also upon the extent to which this configuration is made known in advance to enemy-states. Thus, before such an enemy is appropriately deterred from launching first-strike attacks against Israel, or before it is deterred from launching retaliatory attacks following an Israeli preemption, it may not be enough that it “knows” that Israel has nuclear weapons. It may also need to recognize that these Israeli weapons are sufficiently invulnerable to such attacks and/or that they are targeted on their own pertinent weapons and associated command/control systems.

To fully understand the ambiguity/disclosure question,9 we must first recall the theoretical foundations of nuclear deterrence and nuclear war-fighting as they pertain to Israel. These foundations concern prospective attackers’ perceptions of both Israel’s nuclear capability and Israel’s willingness to use such capability. Removing the bomb from Israel’s “basement” could therefore enhance Israel’s nuclear deterrence and/or nuclear war-fighting postures to the extent that it would heighten enemy-state perceptions of Jerusalem’s capable nuclear forces and/or Jerusalem’s willingness to use these forces in reprisal for certain first-strike and retaliatory attacks. Conversely, failing to remove the bomb from the “basement” could undermine Israel’s postures to the extent that it would lower enemy-state perceptions of Jerusalem’s nuclear capabilities and/or its willingness to exploit these capabilities.10

Let us look at these requirements more closely. To deter enemy attack or post-preemption retaliation, Israel must be able to prevent that enemy, by threat of an unacceptably damaging reprisal or counterretaliation, from deciding to strike. Here, security is sought by convincing the would-be rational11 attacker (irrational enemies are an altogether different problem) that the costs of a considered attack will exceed the expected benefits. Assuming that Israel’s state enemies: (1) always value self-preservation more highly than any other preference or combination of preferences; and (2) always choose rationally between alternative options, they will always refrain from attacking an Israel that is believed willing and able to deliver an appropriately destructive response.12

Two factors must communicate such belief. First, in terms of ability, there are two essential components: payload and delivery system. It must be successfully communicated to the prospective attacker by Israel that the Jewish state’s firepower and its means of delivering that firepower are capable of wreaking unacceptable levels of destruction after a first-strike or retaliatory attack. This means that Israel’s retaliatory/counterretaliatory forces must appear sufficiently invulnerable and sufficiently elusive to penetrate the prospective attacker’s active and civil defenses. It need not be communicated to the potential attacker that such firepower and/or the means of delivery are superior. The capacity to deter need not be as great as the capacity to “win”.

With the bomb kept silently in the basement, Israel’s imperative communications could be compromised perilously. Unable to know for certain whether Israel’s retaliatory/ counterretaliatory abilities were aptly formidable, enemy-states could conclude, rightly or wrongly, that a first-strike attack or post-preemption reprisal would be cost-effective. Of course, it is conceivable that continued ambiguity would be adequate for Israeli deterrence, but – then again – it might not be adequate. Were it made more plainly obvious to enemy-states contemplating attack that Israel’s basement bombs meet both payload and delivery system objectives, Israel’s nuclear forces would likely better serve their overriding security functions.

The second factor of nuclear communication for Israel concerns willingness. How may Israel convince potential attackers that it possesses the resolve to deliver an unacceptably destructive retaliation and counterretaliation? The answer to this question lies, in part, in the demonstrated strength of the commitment to carry out the threat and in the precise nuclear weapons that would be available. Here, too, continued nuclear ambiguity could create the impression of an “unwilling” Israel. Conversely, movement toward some as-yet-undetermined level of disclosure could heighten the impression of an Israel that is willing to follow through on its threats.

What, then, are the plausible connections between a more openly declared nuclear capability and enemy-state perceptions of Israel’s nuclear deterrent? One such connection concerns the relation between disclosure and perceived vulnerability of Israeli nuclear forces from preemptive destruction. Another such connection concerns the relation between disclosure and the perceived capacity of Israel’s nuclear forces to penetrate the attacking state’s active defenses. To the extent that removing the bomb from the Israeli basement, or disclosure, would encourage enemy-state views of an Israeli nuclear force that is sufficiently invulnerable to first-strike attacks and/or is capable of piercing enemy active defense systems, disclosure could represent a rational and prudent option for Israel. The operational benefits of disclosure would accrue from deliberate flows of doctrinal information about such matters as dispersion, multiplication and hardening of nuclear systems and about some other technical features of certain nuclear weapons systems. Above all else, such carefully-controlled flows would serve to remove enemy doubts about Israel’s nuclear force capabilities, doubts which – if unchallenged – could undermine Israeli nuclear deterrence. Removing the bomb from Israel’s basement might also heighten enemy-state perceptions of Jerusalem’s willingness to make good on its nuclear retaliatory threats. For example, by releasing information about its nuclear weapons that identified distinctly “usable” forces, Israel could remove enemy doubts about Jerusalem’s nuclear resolve. Here, a prospective attacker, newly aware that Israel could retaliate without generating intolerably high levels of civilian harms (possibly because of enhanced radiation13 and/or sub-kiloton weapons), would be more likely, because of Israel’s disclosure, to believe Jerusalem’s nuclear threats.

This brings us directly to the doctrinal question of “counterforce” vs. “countervalue”. Counterforce strategies are those which target an enemy’s strategic military facilities and supporting infrastructures. Such strategies may be dangerous not only because of the “collateral damage” they could produce, but also because they could heighten the likelihood of enemy first-strikes. Should Israel be “going for counterforce” with its nuclear weapons? If so, enemy knowledge of such movement could encourage preemption planning by certain enemy-states, but it could also enhance the power of Israel’s nuclear deterrent (because counterforce-targeted nuclear weapons are more likely to be judged usable). Depending upon Jerusalem’s rank-ordering of nuclear strategy values and its expectations concerning enemy-state reactions, disclosure, or taking the bomb out of the basement, could be more or less purposeful for Israel.

Countervalue strategies refer to the targeting of an enemy’s cities and industries; in effect, the targeting of civilian populations. Should Israel be content with developing the relatively inaccurate apparatus of such an “assured destruction” posture, it could probably limit the prospect of preemptive enemy first-strikes. This prospect could even be limited further if the assured destruction posture were accompanied by open and fairly precise disclosure of Israel’s nonthreatening nuclear stance. At the same time, should this posture fail to deter concerted enemy first-strikes, or enemy retaliations for Israeli preemptions, its intrinsic damage-limiting inferiority to a counterforce capability could produce much larger casualty figures. As we will soon see, excessive countervalue targeting could impair Israel’s nuclear war-fighting needs.

If, on the other hand, Israel were to start off with a declared nuclear war-fighting or counterforce posture, enemy-state perceptions of inevitable war with Israel could be enlarged. With such perceptions, belligerent leaders would have to decide whether or not it would be more gainful to await an Israeli preemption or whether to strike first themselves. Aware of this, Israel’s leaders must determine not only the optimum configuration of countervalue and counterforce, but also the most favorable means and levels of disclosure.

How should Israel choose following its incomprehensible surrender of essential territories in exchange for an illusory “peace”? If Jerusalem should opt for nuclear deterrence based on assured destruction, it would run the risk of “losing” any nuclear war that might arise. If it should choose counterforce, certain enemy-states could feel especially threatened, a condition that would likely heighten the actual prospects of nuclear weapons use.

In making its nuclear choices, Israel will have to confront a paradox. Credible nuclear deterrence, essential to security and survival – especially in a world made even more dangerous by the end of strategic depth and the creation of Palestine – would require recognizably usable nuclear weapons. If, after all, these weapons were inappropriate for any reasonable objective, they would not deter. Yet, the more usable the weapons would become in order to enhance nuclear deterrence – a usability communicated more or less effectively by a shift away from deliberate ambiguity – the more likely it is that they will actually be fired. Although this paradox would appear to recommend, inter alia, the deployment of the least-harmful forms of usable nuclear weapons, the likely absence of coordinated agreements with enemy-states on deployable nuclear weapons points toward a different conclusion: Unless Israel were to calculate that the more harmful weapons would produce greater hazards for its own population as well as for target states, there would be no tactical benefit for Israel to opt for the least injurious usable nuclear weapons.

Regarding issues of nuclear usability, an excellent study has been offered by two target planners and theater-force analysts at Los Alamos National Laboratory. While interested exclusively in the improvement of the United States' nuclear strategy, the arguments presented by Thomas W. Dowler and Joseph S. Howard II pertain instructively to Israel. In their analysis, Dowler and Howard evaluate nuclear weapons with very low yields ranging from 10 to 1,000 tons. Seeking nuclear weapons whose power is “effective but not abhorrent”, the authors detail the particular benefits of “micronukes” (weapons with a yield on the order of 10 tons or 20,000 pounds of high explosive), “mininukes” (weapons with a yield of about 100 tons), and “tinynukes” (weapons with a yield of about 1,000 tons or one kiloton).

For Israel, a micronuke employed as an earth-penetrating warhead (EPW) could destroy all but the hardest command bunkers. Deliverable by a gravity bomb, tactical cruise missile or tactical surface-to-surface missile, a micronuke EPW could also be used effectively to neutralize airfields. As a single micronuke strike could put an airfield out of commission for an extended time, use of these particular subkiloton weapons could reduce exposure of Israeli pilots to enemy defenses. This is because it would not be necessary to expect these pilots to execute follow-on strikes.

Should deterrence fail to prevent a launch of enemy missiles carrying nuclear or other mass-destruction warheads at Israeli forces, either as a first-strike attack or as an enemy retaliation for Israeli preemption, Israel would require an adequate defensive capability. To acquire such a capability, Israel could benefit from an anti-tactical ballistic missile (ATBM)14carrying a mininuke warhead with a yield of approximately 100 tons. Seeking to destroy incoming warheads in flight (simply knocking the missile off-target might not neutralize its capacity to inflict great harm on Israeli forces), a mininuke fired by Israel could provide the needed power. Such power could prove vital because an incoming nuclear, chemical or biological warhead must be destroyed as far from its target as possible.

Once an armed conflict had actually broken out between Israel and certain enemy-states, tinynuke warheads with yields of about 1,000 tons could prove effective against tank and troop units. True battlefield weapons rather than agents of indiscriminate mass destruction, these tinynukes – deliverable by tactical air-to-air surface missile, tactical surface-to-surface cruise or ballistic missile, or artillery round – could eliminate any company-sized unit. Intended for very precise operations against known troop formations, these weapons would display lethal radii on the order of 500 meters against tank crews and 600 meters against infantry, artillery and support troops. As for the collateral damage and safe standoff radii of these tinynukes, they would range only to about 1,500 meters. Moreover, utilized as airbursts, they should produce no significant local fallout.15

Returning to the original doctrinal question of countervalue vs. counterforce, Israeli planners must commence prior investigations of enemy-state inclinations to strike first and in retaliation, and of associated inclinations to strike all-at-once or in stages. Should these planners assume, for example, that certain enemy-states in the process of “going nuclear” are apt to strike in an unlimited mode – i.e. to fire all nuclear warheads immediately – Israeli counterforce-targeted nuclear warheads, used in retaliation or in counterretaliation, would likely hit only empty silos/launchers. In such circumstances, Israel’s only rational application of counterforce doctrine would be to strike first itself.

If, for whatever reason, Israel were to reject the preemption option, given the above assumptions, there would be no good reason to opt for counterforce. From the standpoint of compelling intra-war nuclear deterrence, a countervalue strategy could prove substantially more purposeful under such assumptions. Should Israeli planners assume that enemy-states “going nuclear” are apt to strike first and to strike in a limited mode, holding some significant measure of nuclear firepower in reserve for follow-on strikes, Israeli counterforce-targeted nuclear warheads, used in retaliation, could have meaningful damage-limitation effects. Here, counterforce operations could serve both an Israeli preemption or, should Israel decide, for whatever reason, not to preempt, an Israeli retaliatory strike. Moreover, should an Israeli first-strike be intentionally limited, perhaps because it would be coupled with a guarantee of no further destruction in exchange for an end to hostilities, such operations could serve an Israeli counterretaliatory strike. This is the case because Israel’s attempt at intra-war deterrence could fail, occasioning the need for follow-on strikes to produce essential damage-limitation.

In order to examine fully whether Israel would be better served by continued ambiguity or by disclosure (and if the latter, by what degrees of disclosure), as students of Israeli nuclear strategy, we must first identify and understand the reasons behind Israel’s nuclear forces. What, then, are these particular reasons? Why, exactly, does Israel need nuclear weapons? Once we can answer these antecedent questions we will be able to determine if Israel’s bomb should remain in the basement or if it should be brought, more or less, into the country’s “upper floors”. In essence, therefore, future scholarship in this area should be directed by the following hypothesis: If Israel moves beyond “deliberate ambiguity” to certain apt forms of “disclosure”, its nuclear forces will be better able to fulfil their seven essential security functions.

This does not imply, however, that if these seven essential functions were fulfilled, Israel would necessarily be secure. Although I have hypothesized that various and incremental levels of disclosure could enhance Israel’s nuclear deterrent and associated nuclear functions, Israel should never give up its territory on the assumption that it could rely entirely upon its nuclear threat. Nuclear weapons, appropriately configured and disclosed, are necessary to Third Temple Commonwealth survival, but they are not sufficient. Even a nuclear-armed state needs a broad range of weapons that are purposeful to the entire foreseeable spectrum of possible harms. It follows that those Israelis who currently argue for a withdrawal to the 1949 borders because Israel has nuclear forces are altogether mistaken. Israel must not become the first and only case in history of a state totally dependent upon nuclear threats.

Optimally, Israel will take steps to maintain its all-important conventional deterrence while, at the same time, ensuring that its nuclear deterrence is informed by doctrine. Should this doctrine be left implicit, as has been the Israeli case of “deliberate ambiguity” for decades, enemy-states would need to reconstruct expectations about Israeli capabilities and intentions. Should this doctrine be made explicit, as would be the case if the bomb were removed from the “basement”, these enemy-states could extrapolate expectations from this doctrine directly. Of course, it is conceivable that more explicit articulations of Israeli nuclear strategy would be distrusted or even discounted, but disclosure would at least provide Israel with an opportunity for some input into enemy-state calculations.

For Israel, the advantages of disclosure would likely be greater with respect to the deterrence of unconventional attacks than with respect to deterrence of large conventional attacks. This is because the presumed plausibility of Israeli nuclear reprisal is apt to be greater when unconventional weapons are used for aggression. A different assumption about disclosure’s advantages vis-à-vis large conventional attacks could be reasonable if Israel were to couple its nuclear retaliatory threats with far-reaching conventional disarmament and/ or with further territorial concessions, but such coupling, it now goes without saying, would represent a tragic and potentially irretrievable error for the Jewish state.

No less tragic for Israel would be a decision to accept some form of internationally imposed limitations on its nuclear arsenal.16 With such a decision, the question of disclosure would become moot. After all, “volitional” denuclearization consistent with expected treaty commitments would leave Israel with nothing to disclose. In consequence, Israel’s deterrence requirements would all have to be met with conventional threats and/or US “extended deterrence”. This would not be possible.

Israel requires both conventional and nuclear weapons, complementary forces and doctrines to preserve the Third Temple Commonwealth into the next millennium. Significantly, the “Peace Process” has endangered both interrelated requirements. Already, this Process, spawning shrinking strategic depth, has severely curtailed the capacities of Israel’s conventional arms. For the very immediate future, it also threatens the capacities of Israel’s nuclear weapons, a situation that would not only leave the bomb in the basement, but also bury it there.

One last word about essential Israeli nuclear deterrence of enemy unconventional attack, a need that could be served more or less effectively by some apt measure of disclosure. Normally, strategic planners, examining the requirements of nuclear deterrence, distinguish carefully between conventional and unconventional attacks. For Israel today, however, such a sharp distinction could be misleading and dangerous.

Why? From now on, it is unlikely that enemy-states would launch large conventional attacks against Israel unless these states had backup unconventional (possibly but not necessarily nuclear) forces. This means that the capacities of Israeli nuclear deterrence will now always have to be assessed vis-à-vis enemy-state unconventional weapons. Hence, the question of disclosure will now always have to be asked ultimately in reference to nuclear deterrence of unconventional weapons.

It is conceivable, especially after Israel’s unforgivable surrender of territories, that some combination of enemy-states, still effectively nonnuclear, could conclude that a combined conventional attack against Israel would be gainful. To prevent such a conclusion, thereby maintaining successful nuclear deterrence, Jerusalem would need to convince these enemy-states that their prospective combined conventional assault could elicit a fully nuclear reprisal. This task could be made easier by appropriate communications to enemy-states concerning disclosure, including purposeful communications of Israel’s awareness that the conventional/unconventional threshold might still be breached first by the “conventional” enemy-state attackers. Although it is likely that this task could also be made easier because of Israel’s already-truncated strategic depth, the net effect of such truncation for Israel would surely be negative. Halting the “Peace Process”, if this is indeed still possible, is a clear and overriding strategic imperative.



At that time, and even today, disclosure was treated generally as a dichotomous variable, i.e. only two options presented themselves: disclosure or nondisclosure. This article is premised on the assumption, inter alia, that disclosure should be treated by Israeli defense planners as a continuous variable. This would allow these planners to identify multiple disclosure options along a nuanced continuum of possibilities, and to choose that particular level of disclosure that is presumed maximally gainful to Israeli nuclear requirements.


The principal scholarly publication in this genre was Shai Feldman’s Israeli Nuclear Deterrence: A Strategy for the 1980s, New York: Columbia University Press, 1982. My argument here for disclosure is substantially different from Feldman’s, rejecting any implicit comparisons with superpower nuclear deterrence and expanding the pertinent essential functions of Israeli nuclear forces.


See ../authors/, ed., Security or Armageddon: Israel’s Nuclear Strategy, Lexington MA: Lexington Books, 1986. This anthology, which was comprised of chapters by the following scholars, was reviewed very favorably on page 1 of the August 24, 1986 New York Times Book Review: ../authors/, Alan Dowty, Gerald M. Steinberg, Avner Yaniv, Efraim Inbar, Zeev Eytan, Robert Harkavy, Avi Beker, Stephen J. Cimbala, Robert A. Friedlander, Burns H. Weston, and Avner Cohen.


An associated argument suggests that disclosure could exacerbate the regional nuclear arms race, to Israel’s distinct detriment. For example, Roger Molander and Peter Wilson maintain the following:

Overt acceptance of Israel as a nuclear-armed state would undoubtedly further stimulate nuclear weapons programs among the Arab states and Iran so that a predictable sequel would be the eventual demand to grant legitimacy to, say, nuclear arsenals in Iraq and Iran.

(See Molander and Wilson, “On Dealing With the Prospect of Nuclear Chaos”, The Washington Quarterly, Vol. 17, No. 3, pp. 23-24) There is, however, no evidence whatsoever to support this particular argument which could, by extrapolation, be used very erroneously and possibly with very grave consequences against disclosure.


An alternative image for nuclear ambiguity has been “opacity”. Shlomo Aronson refers to “opaque nuclear cases”. As he uses this particular image, undeclared bombs take on a greater analytic subtlety. According to Aronson:

The adjective ‘opaque’ is derived from physics. In this context, it can be used to describe what happens when one looks at an object through a certain type of crystal. Depending upon how you hold the crystal, you might not see the object clearly – it will be distorted. But if you hold the crystal ‘properly’, you will see the object very clearly indeed.

See the author’s The Politics and Strategy of Nuclear Weapons in the Middle East: Opacity, Theory and Reality 1960-1991, An Israeli Perspective, Albany: State University of New York Press, 1992, p. ix. Aronson credits Ben Frankel with renaming “semicovert” nuclear proliferation “opaque”.


Although not developed herein, the “pertinent form and extent of disclosure” represent critical questions for future inquiry. What, precisely, should be the optimal means of disclosure in different circumstances and in relation to different objectives? What, exactly, should be the optimal levels of disclosure in these different circumstances and in relation to the seven different stated security functions of nuclear weapons? These questions should now give rise to explicitly stated hypotheses and to further informed inquiry. Israeli military planners can have no more important or timely research obligation.


Biological weapons may be of somewhat less immediate strategic concern for Israel than chemical weapons. Although a growing number of states have or are now developing capabilities to employ living organisms (such as anthrax, lassa fever, or typhus, as opposed to inert toxins), such capabilities would have limited military value. This is because their dispersal mechanisms are difficult to manage: a change of wind could make them as lethal to the attacker as to the intended victim, and because it is difficult to sustain the living organism in biological weapons in hot climates. At the same time, precisely because biological weapons are better suited for mass-destruction than for use as dedicated military instruments, they could hold greater appeal to Israel’s irrational state enemies.


The seven essential functions that need to be addressed are as follows: (1) Deterrence of large conventional attacks by enemy-states; (2) Deterrence of all levels of unconventional attacks by enemy-states; (3) Preemption of enemy nuclear attacks; (4) Support of conventional preemptions against enemy-state nuclear assets; (5) Support of conventional preemptions against enemy-state nonnuclear assets; (6) Nuclear war-fighting; and (7) The “Samson Option”. I have investigated these functions in: ../authors/, "Security Threats and Effective Remedies: Israel’s Strategic, Tactical and Legal Options", Ariel Center for Policy Research, Policy Paper No. 102, April 2000.


It is assumed throughout here that such understanding requires what I call an appropriate “strategic dialectic”. In contrast to the standard literature of this field, which is generally satisfied with fundamentally reportorial forms of investigation, my method will be to ask and answer questions, again and again, until core problems are confronted frontally. Here we will approach our problem as an interrelated series of thoughts, where each thought presents a complication that moves inquiry onward to the next thought. Contained in this strategic dialectic is an obligation to continue thinking, an obligation that, logically, can never be fulfilled completely (because of what is called the “infinite regress problem”), but that must still be attempted as fully and as competently as possible. With such a dialectical form of investigation, we may focus upon appropriately dynamic and generic interactions/synergies.


Contrary to the prevailing conventional wisdom, nuclear deterrence and associated forms of Israel’s nuclear posture, including preemption, can fully support the authoritative expectations of international law. The adequacy of international law in preventing nuclear war in the Middle East will depend upon far more than certain formal treaties, customs, and general principles. It will depend, especially, upon the success or failure of particular country strategies in the region. If Israel’s nuclear strategy should reduce the threat of nuclear war, either because of successful forms of nuclear deterrence or because of essential preemptive strikes – possibly with the aid of some apt measure of disclosure – this strategy should be considered as an authentic component of international law enforcement. I write these words as a professor of international law, one who lectures and publishes extensively on jurisprudential matters.


Even if it could be assumed that enemy-state leaders will always meet the expectations of rational decision-making, this would say nothing about the accuracy of information used in making rational calculations. Rationality refers only to the intention of maximizing specified values or preferences. It says nothing at all about whether the information used is correct or incorrect. Hence, rational leaders of enemy-states may make errors in calculation that lead their states to war against Israel.


In keeping with standard definitions of rationality in world politics, I assume a unitary, value-maximizing decision-maker with one set of specified goals, one set of perceived options and a single estimate of the consequences that ensue from each alternative. Thus, the enemy-state decision-maker is assumed to evaluate alternatives in his strategic environment on the basis of his preferences among them; to operate according to a preference-ordering that is consistent and transitive; and to always choose the preferred alternative. An oft-ignored problem with rationality assumptions is that they concern only preference-maximizing intentions. An enemy-state may meet all of the requirements of rationality, but still commit errors in calculation that undermine deterrence.


The “neutron bomb” is often regarded as an especially unwholesome form of nuclear weapon. Yet, because its lethality is projected largely via radiation rather than through concussion and heat, the neutron bomb could prove to be especially usable, and with all attendant deterrent benefits. These issues were raised years ago in the context of US and NATO nuclear doctrine. The classic defense of nuclear radiation weapons was offered in June 1981 by Samuel T. Cohen who developed the technical-military concept of the neutron bomb in 1958. See his “Whither the Neutron Bomb? A Moral Defense of Nuclear Radiation Weapons”, Parameters, Journal of the US Army War College, Vol. XI, No. 2, June 1981, pp. 19-27.


Presently, of course, the Arrow interceptor is the centerpiece of Israel’s planned ballistic missile defense system, part of a cooperative program with the United States Department of Defense, Ballistic Missile Defense Organization. The joint project began in May 1986, when Israel and the US signed a memorandum of understanding. However “successful” Arrow might ultimately be (it is already partially deployed), it is altogether possible that offensive weapons technology exploited by Israel’s enemies would enjoy an overwhelming advantage, especially if their incoming warheads were ultimately nuclear. Considered together with prospects of enemy irrationality, this suggests a compelling argument for ongoing Israeli preparation for pertinent preemption options.


See Thomas W. Dowler and Joseph H. Howard II, “Countering the Threat of the Well-Armed Tyrant: A Modest Proposal for Small Nuclear Weapons”, Strategic Review, Vol. XIX, No. 4, Fall 1991, pp. 34-40.


Regarding recurrent US pressures upon Israel to join the NPT, this brings to mind the special foolishness of “extended nuclear deterrence” for Israel. Israeli planners should understand that, for the United States, the central role of nuclear deterrence has declined, and that extended deterrence is increasingly defined as conventional. This American view flows largely from former President Bush’s 1991 decision to eliminate ground- and sea-based tactical nuclear weapons. For a comprehensive and insightful examination of extended conventional deterrence, see: Charles T. Allan, “Extended Conventional Deterrence: In From the Cold and Out of the Nuclear Fire?”, The Washington Quarterly, Vol. 17, No. 3, Summer 1994, pp. 203-233.