A casual reading of the history of national security suggests not only that the rules of evidence will be ignored in practice, but also that the notion of catastrophe encourages, even insists on, these rules being flouted. “In normal affairs,” Cardinal Richelieu declared at the dawn of the modern state system, “the administration of Justice requires authentic proofs; but it is not the same in affairs of state … . There, urgent conjecture must sometimes take the place of proof; the loss of the particular is not comparable with the salvation of the state.” As we ascend the ladder of threats, in other words, from petty crime to the destruction or loss of the state, we require less and less proof that each threat is real. The consequences of underestimating serious threats are so great, Richelieu suggests, that we may have no choice but to overestimate them. Three centuries later, Learned Hand invoked a version of this rule, claiming that “the gravity of the ‘evil’” should be “discounted by its improbability.” The graver the evil, the higher degree of improbability we demand in order not to worry about it. Or, to put the matter another way, if an evil is truly terrible but not very likely to occur, we may still take preemptive action against it.

Neither statement was meant to justify great crimes of state, but both suggest an inverse relationship between the magnitude of a danger and the requirements of facticity. Once a leader starts pondering the nation’s moral and physical extinction, he enters a world where the fantastic need not give way to the factual, where present benignity can seem like the merest prelude to future malignancy. So intertwined at this point are fear and reason of state that early modern theorists, less shy than we about such matters, happily admitted the first as a proxy for the second: a nation’s fear, they argued, could serve as a legitimate rationale for war, even a preventive one. “As long as reason is reason,” Francis Bacon wrote, “a just fear will be a just cause of a preventive war.” That’s a fairly good description of the logic animating the Cold War: fight them there—in Vietnam, Nicaragua, Angola—lest we must stop them here, at the Rio Grande, the Canadian border, on Main Street. It’s also a fairly good description of the logic animating the Nazi invasion of the Soviet Union.

We are fighting on such distant fronts to protect our own homeland, to keep the war as far away as possible, and to forestall what would otherwise be the fate of the nation as a whole and what up to now only a few German cities have experienced or will have to experience. It is therefore better to hold a front 1,000 or if necessary 2,000 kilometers away from home than to have to hold a front on the borders of the Reich.

These are by no means ancient or academic formulations. While liberal critics claim that the Bush administration lied about or deliberately exaggerated the threat posed by Iraq in order to justify going to war, the fact is that the administration and its allies were often disarmingly honest in their assessment of the threat, or at least honest about how they were going about assessing it. Trafficking in the future, they conjured the worst—“we don’t want the smoking gun to be a mushroom cloud”—and left it to their audience to draw the most frightful conclusions. [++]