If Iran was presumed to be rational, in the usual sense of valuing its national physical survival more highly than any other preference, or combination of preferences, Washington and/or Jerusalem could begin to consider certain benefits of pretended irrationality. Years ago, Israeli Gen. Moshe Dayan, warned: "Israel must be seen as a mad dog; too dangerous to bother." In this crude but effective metaphor, Dayan had already understood that, sometimes, it can be rational to feign irrationality.
There may have been a core element of precisely such reasoning by President Kennedy, during his handling of the 1962 Cuban Missile Crisis. Many years ago, when I was cochairing a panel at the Naval Academy with Admiral Arleigh Burke, the former chief of Naval Operations repeated to me what had earlier been published by Ted Sorensen, JFK's biographer: Kennedy believed that his actions (the quarantine of Cuba) entailed potentially even odds of a nuclear war with the Soviet Union.
But what if an Iranian adversary were presumed to be irrational in the sense of not caring most about its national survival? Here, there would be no deterrence benefit to assuming any posture of pretended irrationality. The more probable threat of a massive nuclear counterstrike by Israel and/or the United States in this case would likely be no more compelling in Tehran than if Iran's enemy were considered rational.
"Do you know what it means to find yourself face to face with a madman?" inquires Luigi Pirandello's Henry IV. While this query does have residual relevance to our current security concerns with Iran, the mounting strategic challenges from that country will be more apt to come from decision-makers who are not "mad," and who are still more-or-less "rational." Soon, with this idea in mind, an idea plainly accepted by Meir Dagan, Israel, and the United States, preferably in close cooperation with each other, will need to fashion a focused strategic doctrine from which essential policies and operations can be suitably extrapolated.
This framework for decision would identify and correlate all available strategic options (deterrence; pre-emption; active defense; strategic targeting; and perhaps nuclear war fighting) with critical survival goals. It would also take very close account of possible interactions between these discrete strategic options. Calculating these interactions or "synergies" will present a computational task on the very highest order of intellectual difficulty.
Nuclear strategy is a "game" that sane and rational decision-makers must play, but to compete effectively, a would-be victor must always first assess (1) the expected rationality of each opponent; and (2) the probable costs and benefits of pretending irrationality oneself. These are undoubtedly complex and glaringly imprecise forms of assessment, ones that will require corollary refinements in intelligence and counterintelligence, but they are also indispensable.