Essential Calculations of Rationality and Irrationality in US Nuclear Deterrence

Karl Jaspers, one of the previous century’s most insightful philosophers, identified the broadest possible meanings of “rationality” and “irrationality” in human affairs. To be certain, back in the 1930s, Jaspers was not thinking specifically about military deterrence, and still more assuredly not about a then-fictional nuclear deterrence. Nonetheless, however unwitting, his conceptual observation about indissolubly reciprocal relationships existing between two utterly core aspects of human nature is well suited to a proper understanding of irrationality in world politics. Presently, this useful observation may even help to explain manifestly urgent challenges to America’s basic national security.

In traditional military studies, strategies of deterrence automatically assume enemy rationality. The reason is unambiguous. After all, from the standpoint of ordinary theorizing, this simplifying assumption is indispensable. Still, from the complementary perspective of discovering strategic “truth” – ultimately, the most critical analytic perspective – it could also be entirely false.

In the absence of rationality – that is, in those more-or-less residual circumstances where an enemy state might rank order certain preferences more highly than “staying alive” as a nation – deterrence is expected to fail. Moreover, regarding those inherently more serious and complex circumstances involving nuclear deterrence, the plausible consequences of such failure could prove catastrophic. In the worst case, incontestably, they could prove unprecedented.

There is more. Prospectively, dealing with sub-state or terrorist adversaries could present a somewhat different and potentially even more hazardous set of nuclear deterrence risks. By definition, these increasingly hard-line adversaries (e.g., ISIS, al-Qaeda, Hezbollah) generally don’t have sovereign national territories to protect. Furthermore, their core objectives could sometime include “martyrdom,” a faith-driven preference that is clearly inauspicious for any adversaries striving to maintain more orthodox deterrence strategies.

For the United States, the basic problem is easy to recognize. Going forward, America may sometime have to deal with adversaries that do not habitually or routinely conform to any ordinary definitions of decisional rationality in world politics. This particular quality is far more portentous than a merely inconvenient truth. At times, it could offer a grave or authentically existential peril.

For the most part, at least for the moment, US nuclear deterrence should continue to be examined and assessed vis-à-vis national or state adversaries, not assorted sub-state enemies. In addition, irrationality, it must be fully understood, is never the same as “crazy,” or “mad,” and must therefore be differentiated from all such imprecise or “common-sense” terminology. Accordingly, American strategic planners should expressly understand that even an irrational enemy leadership could still maintain a distinct and identifiable hierarchy of relevant preferences, albeit one in which national survival does not predictably rank at the very top.

Language matters. Using correct strategic terminology, professional military analysts would likely report that any such irrational state actors exhibit an ordering of preferences that is “consistent,” “instrumental,” and “transitive.” In principle, even certain “irrational” states could thus be rendered subject to various alternative forms of deterrence. And for any state that must more-or-less rely on credible threats of retaliatory destruction, correctly recognizing such “forms” could prove indispensable to its national security.

By usual definition, a genuinely “crazy” or “mad” leadership would have no discernible order of preferences. Rather, its strategic actions and interactions would expectedly be random and unpredictable. It follows that facing a crazy or mad adversary in world politics would be substantially “worse” than confronting “just” an irrational adversary. Although it might still be possible and determinably reasonable to attempt deterrence of an irrational enemy, there would be little or no point to seeking such protections against a seemingly “mad” one.

“Do you know what it means to find yourselves face to face with a madman,” inquires playwright Luigi Pirandello’s Henry IV. “Madmen, lucky folk, construct without logic, or rather with a logic that flies like a feather.”

What is true for individuals is sometimes also true for states. In the bewildering theatre of modern world politics, a drama that often bristles with evident or unexpected absurdities, strategic decisions that rest upon logic can quickly crumble before madness. In principle, at least, corresponding dangers could occasionally reach the most singularly threatening or existential levels. This is most notably the case whenever madness and a nuclear weapons capability would overlap.

Always, pertinent strategic questions of rationality and irrationality are not narrowly theoretical. On the contrary, they are profoundly real and current, especially in the still-adversarial dyads involving the United States with North Korea and Iran. Here, of course, a fundamental distinction must be maintained between an adversary that is already nuclear, and one that could sometime become nuclear.

Regarding Iranian intentions and capabilities, it is plausible to conclude that both have been accelerated (not constrained) by US President Trump’s unilateral decision to exit the multilateral JCPOA nuclear agreement of July 2015.

Regarding North Korean intentions and capabilities, it is overwhelmingly plausible that while the former is generally indecipherable, the latter is fixed and irreversible. For US President Trump to assume that Kim Jung-un would at any time agree to bargain away Pyongyang’s extant nuclear weapons and delivery vehicles would be nothing less than delusional.

For the foreseeable future, America’s ultimate source of national security must continue to lie in some pattern or other of sustained and flexible nuclear deterrence. Still, whatever particular patterns are codified, US nuclear posture could sometime “crumble before madness.” This suggests that in certain easily-imaginable instances involving aberrant enemy behavior, the outcome of failed retaliatory threats could on some occasions include genuinely irremediable harms.

All things considered, while the logic of deterrence has traditionally required a simplifying assumption of rationality, history also reveals the persistent fragility of any such basic expectation. To be sure, we already know all too well that nations can behave in ways that are consciously and conspicuously self-destructive. Observed Sigmund Freud: “Fools, visionaries, sufferers from delusions, neurotics, and lunatics have played great roles at all times in the history of mankind….”

There is more. Mirroring the generally unpredictable behavior of individual human beings, national leaders could sometimes assign the very highest value to preferences other than collective self-preservation, thereby producing a sort of Gotterdammerung or “Twilight of the Gods” scenario. Until now, of course, we haven’t ever witnessed such a scenario involving nuclear weapons. Nonetheless, our tormented and tormenting species’ specifically nuclear history is new and very conspicuously untested.

Shall we be reassured? For the moment, at least, no single Iranian or North Korean adversary would appear to be recognizably irrational or mad.  Harsh enemy rhetoric notwithstanding, no such adversary actually appears ready or willing to launch a first-strike against American homeland assets using nuclear weapons of mass destruction. For now, moreover, the plausible expectation that any such aggression would elicit a devastating reprisal remains seemingly sufficient to prevent any such attack.

Still, by their very nature, strategic nuclear calculations are enormously complicated. In this connection, certain faulty calculations or errors in information could sometime lead a perfectly rational enemy state to strike first; and this particular attack decision need not be the outcome of outright irrationality or madness. Technically, as corollary, all strategic judgments of rationality and irrationality must ultimately be rooted in ascertainably prior intent.

In world politics, there can be no greater power than power over death. Within this endlessly conflictual realm, still a Hobbesian bellum omnium contra omnes (a “war of all against all”), things must move perpetually in the midst of death and closely related hopes for immortality. Accordingly, certain enemy states, perhaps most likely Iran in the future, could one day decide that excising the “enemies of Allah” would be worth incurring even the most staggering costs.

From a purely military standpoint, this genocidal prospect might be reduced or avoided if the United States should become willing to undertake certain eleventh-hour “hard target” preemptions. All things considered, however, any such once-reasonable expressions of anticipatory self-defense would now be very difficult to imagine or to find cost-effective. Nuclear deterrence, not preemption, must offer the best available long-term “remedy.”

A meaningfully successful preemption is now plausibly beyond this country’s cumulative military capabilities. This does not mean that the United States would necessarily be operationally incapable of destroying a required number of Iranian or North Korean nuclear assets and infrastructures, but only that the net costs of any such venture would likely exceed the net gains. Already, such a conclusion applies to the nuclear fait accompli in Pyongyang.

Regarding Iran, which is not yet nuclear, virtually all critical Iranian nuclear assets have nonetheless been deeply hardened, widely dispersed, and substantially multiplied. For America (and for Israel), there would also be considerable political costs attached to any residual preemption. Unquestionably, a preemptive attack, even one that could become an operational failure, would elicit absolutely overwhelming shouts of both public and diplomatic condemnation. Such expectedly deafening howls of execration could (and should) factor into Washington’s (and Jerusalem’s) overall strategic decisional process.

It is perhaps plausible that certain alternative forms of preemption, including targeted assassination of nuclear scientists and/or cyber defense/cyber-warfare, could still be useful, or even necessary, but it is also unlikely that any such options could permanently obviate more traditionally expedient resorts to massive military force. It is also worth noting that such extraordinary measures would likely involve certain more-or-less conspicuous illegalities.

A “bolt-from-the-blue” CBN (chemical, biological or even nuclear) attack that is launched with an expectation of city-busting reprisals might not exhibit irrationality or madness. Within such an attacking state’s particular ordering of preferences, any presumed religious obligation to annihilate the enemy could itself represent the overriding value or preference. From the standpoint of the prospective attacker’s decisional calculus, the expected benefits of producing any such “blessedly” apocalyptic annihilation would plainly exceed the expected costs of any expected American reprisal.

Judged from this critical analytic standpoint – the perspective of the would-be attacker – a seemingly “mad” attack decision could still “make sense.”

There is more. An enemy state that exhibits explicitly-exterminatory orientations could effectively represent the individual suicide bomber in macrocosm. Whether we like it or not, it is a realistic and heuristic image. Just as certain individual Jihadists (both Shi’ite and Sunni) are now expressly willing to achieve personal “martyrdom,” so might particular Jihadist states sometime become willing to collectively “sacrifice themselves.”

Israel’s Moshe Dayan once declared: “Israel must be seen as a mad dog; too dangerous to bother.” If America’s (and Israel’s) enemies could all still be presumed rational, that is, in the ordinary sense of valuing their physical survival more highly than any other preference or combination of preferences, Washington (or also Jerusalem) could soon begin (among other things) to exploit certain latent strategic benefits of “pretended irrationality.” Recognizing that in assorted strategic situations it could be rational to feign irrationality, these two asymmetrical allies could then work systematically to create more appropriately cautionary behavior among relevant adversaries.

One overriding challenge here will be the expected expansion of asymmetrical conflicts that involve conflicts with terrorist organizations as well as with enemy states, sometimes simultaneously. A pertinent sub-set of this challenge would be developing suitable laser-based defenses against various “hybrid” adversaries, and also calculating whether nuclear weapons used elsewhere in the world could correspondingly erode still-inhibiting nuclear taboos in the Middle East or Northeast Asia. At present, it looks very much like the first actual military use of nuclear weapons in the world would involve North Korea and/or Pakistan/India.

All this should bring American planners back to certain core questions of enemy rationality. What about Moshe Dayan’s earlier advice? If America’s identifiable national adversaries were presumptively irrational in the ordinary sense, there would likely be no real benefit to any assumed US postures of pretended irrationality. This is the case because the more probable threat of any massive nuclear counterstrike linked in enemy calculations with irrationality would be no more compelling to any enemy state than if it were confronted by an expectedly rational United States policy.

Significantly, especially for US President Donald Trump, pretended irrationality can “work” only vis-à-vis fully rational adversaries. Washington could conceivably benefit from a greater understanding of the “rationality of pretended irrationality,” but only in particular reference to expectedly rational enemy states. In those more-or-less expected circumstances where such enemy states were presumed to be irrational, something else could be needed, something other than nuclear deterrence, preemption and/or active defense.

Although many commentators and scholars still believe the answer to this quandary lies in diplomatic or political settlements, this generally time-dishonored belief is born largely of frustration with incessant war. Sigmund Freud, who would have best understood the continually great attraction of power over death in world politics, would call such belief an “illusion,” or a textbook example of “wish fulfillment.”

Going forward, Washington must understand, inter alia, that irrationality need not imply madness. Even an irrational enemy state leadership could maintain an instrumental, consistent, and transitive hierarchy of wants. The first deterrent task must therefore be to identify this hierarchy among its several state enemies. Although these states might not be deterred from aggression by even the plausibly persuasive threat of massive American retaliations, they might still be dissuaded by certain other threats aimed at what they do hold to be most important.

What might be most important to America’s prospectively irrational state enemies, potentially even more important than their own physical survival as a state? One possible answer (although now limited to the Middle East area) is the avoidance of certain forms of presumed apostasy, shame, and humiliation. This would include avoiding the potentially unendurable charge that they had somehow defiled their most sacred religious obligations.

Another would be leaders’ strongly-preferred avoidance of their own violent deaths at the hand of the United States, deaths that could be attributable to certain American strategies of “targeted killing” and/or “regime-targeting.” In these cases, the particular Islamic leaders would not themselves have been persuaded by the more usually compelling benefits of “martyrdom.” This last suggestion could be problematic to the extent that, theologically, being killed by adversaries for the presumed sake of Allah ought to be regarded as a doctrinally distinct positive.

Dying for the sake of Allah, as US strategic planners should recall, could be regarded in these particular leadership contexts as a clerically-blessed passport to immortality. Again, in these adversarial contexts, there can be no earthly power that is greater than a promised power over death.

In the future, facing increasingly high levels of possible destruction, America will inevitably need to deal with both rational and irrational adversaries. These enemies, in turn, will be both state and sub-state actors. On occasion, US leaders will also have to deal with various complex and subtle combinations of rational and irrational enemies, sometimes simultaneously.

Ultimately, America must prepare to deal with “nuclear madmen,” both as terrorists and as national leaders. But first, it must fashion a suitable plan for dealing with nuclear adversaries who are neither mad, nor irrational. With such a challenging imperative, Washington should now do everything possible to meaningfully enhance its deterrence, preemption, defense, and war-fighting capabilities. This means, among other things, enhanced and explicit preparations for certain “last resort” operations.

Any recognizable last-resort preparations could enhance US preemption options by displaying a clear and verifiable willingness to accept existential risks. In this scenario, however, US leaders must always bear in mind that pretended irrationality could quickly become a double-edged sword. Brandished too flagrantly, and without sufficient nuance, any US preparations for a last resort option could sometime impair rather than reinforce America’s nuclear war-fighting options.

Undoubtedly, nuclear warfighting, wherever possible, and in any conceivable form, should be strenuously avoided by the United States. The true purpose of this country’s nuclear forces and doctrine must always be deterrence ex ante, not revenge ex post. To suggest otherwise would be to fundamentally misunderstand the nature of American national security.

Naturally, there remain some readily identifiable circumstances in which nuclear exchanges could be unavoidable, whatever Washington might actually have done to prevent them. Here, logically, some forms of nuclear warfighting could ensue, so long as: (a) enemy state first-strikes launched against American assets would not destroy America’s second-strike nuclear capability; (b) enemy state retaliations for an American conventional preemption would not destroy US nuclear counter-retaliatory capability; (c) conventional US preemptive strikes would not destroy enemy state second-strike nuclear capability; and (d) American retaliations for enemy state conventional first strikes would not destroy an enemy state’s nuclear counter-retaliatory capability.

From the standpoint of protecting its overall existential security, this means that Washington must immediately take appropriate steps to ensure the plausibility of (a) and (b), above, and the implausibility of (c) and (d).

“Do you know what it means to find yourself face to face with a madman?” Repeating this pertinent question from Luigi Pirandello’s Henry IV does have immediate relevance to America’s ultimate national security dilemma. At the same time, the steadily mounting strategic challenge to the United States and its allies will come primarily from enemy decision-makers who are not-at-all mad, and who are still more-or-less rational.

Promptly, therefore, Washington will need to fashion a comprehensive and finely-calibrated strategic doctrine, one from which various specific policies and operations could readily be extrapolated. This focused framework would identify and correlate all available strategic options (deterrence, preemption, active defense, strategic targeting, nuclear warfighting) with basic survival goals. It would also take close account of the possible interactions between these strategic options, including all determinable “synergies” between conceivable enemy actions directed against US assets. For an American president and his pertinent counselors, actually calculating these vital interactions and synergies will present a computational task at the very highest order of intellectual difficulty.

Nuclear deterrence is a “game” that America’s national security leaders must play, but to compete effectively, any would-be “winner” must first assess (1) the expected rationality of each critical opponent; and (2) the probable costs and benefits of pretending irrationality oneself. These are undoubtedly complex, interactive, synergistic, and glaringly imprecise forms of assessment, but, just the same, they constitute a much-needed foundation for America’s long-term security. Doctrinally, therefore, it is time for them to become a more integral part of Washington’s aptly relevant “order of battle.”

In the beginning, the United States had only one nuclear adversary, and was effectively able to discount decisional irrationality in Moscow. Hence, a posture of “mutual assured destruction” or MAD was successfully sustained during the Cold War. Now, however, to apply a suitable metaphor first introduced by political scientist Albert Wohlstetter and physicist J. Robert Oppenheimer, there are more than two “scorpions in the bottle,” and still further nuclear proliferation is all but certain. It follows that upcoming assessments of enemy rationality and irrationality will become increasingly bewildering, and that safeguarding US national security will require protracted competition at a uniquely daunting analytic level.

Recalling Karl Jaspers, the underlying intellectual question for President Donald Trump and his advisors will soon concern how the reciprocally rational and non-rational in world politics can best be “grasped.” To capably able to answer this multi-faceted question will require much more than anyad hoc or seat-of-the-pants threats of bravado or national belligerence. It will require a disciplined and altogether determined American victory of mind-over-mind.