Commentary Magazine


The NIE and Neorealism

Norman Podhoretz has already pointed out that the new National Intelligence Estimate on Iran “represents a 180-degree turn from the conclusions of the last NIE on Iran’s nuclear program,” which were asserted with equal certitude. But the NIE has another, much more serious, problem.

The NIE asserts that “some combination of threats of intensified scrutiny and pressure”—no problem there—”along with opportunities for Iran to achieve its security, prestige, and goals for regional influence in other ways, might . . . prompt Tehran to extend the current halt to its nuclear weapons program.” My first reaction to this was to wonder why, if the U.S. has succeeded in stopping Iran’s program without any bribes, we now need to include them to prevent Tehran from starting it up again, but no matter: the NIE generously admits that it is “difficult” to specify what such a “combination” of threats and opportunities might be. The more fundamental question is whether it is in the interest of the United States—and the world—to purchase (if obtainable) a continued halt at such a price. Since Tehran’s declared goals include eradicating Israel from the face of the earth and spreading the Iranian Revolution across the entire Middle East, the answer must be that it is not.

Deciding that, of course, is not the job of the intelligence community. But the NIE’s description of its rationale for reaching its conclusion about Iran’s malleability is revealing: Tehran, it argues, halted the program in 2003 in response to unspecified “international pressure”—apparently the same kind that bore so heavily on Qaddafi—which indicates that the regime’s decisions are guided by “a cost-benefit approach.”

I winced when I read that phrase. Does anyone make decisions on that basis? States certainly do not. The phrase belongs to neorealism, to the unitary rational actor approach to the study of decision-making. The broad realist tradition—and a respectable one it is—extends back to Thucydides. Its modern and more limited variant, neorealism, exemplified today by Kenneth Waltz, Stephen Walt, and John Mearsheimer, refuses to try to understand policy-making in all its complexity. Instead, it treats states as billiard balls, ignores their leaders, politics, beliefs, and cultures, and considers only their size, their place on the table, and the position of the other billiard balls.

Neorealism has the advantage of parsimony: because it is based on simple but powerful assumptions, including the belief that nations act rationally, it generates testable predictions. But the simplifying assumptions that make it useful for scholars make it useless as a guide to how and why states actually make decisions. The only practical contribution the discipline of international relations has made in the last forty years is democratic peace theory, commonly summarized as “democracies do not fight each other.” Even if that is only mostly true, it is inexplicable to neorealists, who cannot understand why two democracies—the United States and Israel, for instance—might be allies even though the larger power has nothing much to gain from it materially.

The failure of neorealism stems directly from its assumptions. For states are not unitary. States have bureaucracies with their own agendas, factions and internal politics, and the most radical of them—like Iran—have a party machine that runs parallel to, and acts as a minder for, the official government. And states are not rational, at least not in a “cost-benefit” kind of way. Nor are they simply irrational. Rather, they have a hierarchy of preferences and seek to order them with some consistency. This kind of bounded rationality says nothing about what these preferences are, or whether they are moral, amoral, or immoral. Hitler, for instance, had two preferences: killing Jews and winning the war. And, in a bounded way, he was rational: he wanted to kill Jews more than he wanted to win, so he ran trains to Auschwitz, not the front.

The neorealist approach does have its uses. If you do not know anything about what is going on inside a country—for example, because it is a totalitarian dictatorship—a useful first cut is to ask what you would do if you were in charge. But to elevate neorealism, as the NIE has done, into a basis for offering high confidence assessments about such a state is an error. Walt and Mearsheimer’s embarrassingly amateur fantasies about the Israel Lobby demonstrate this all too clearly. For them, it is axiomatic that the United States has much more to gain from allying with the Arab oil dictators than with resource-poor Israel: the fact that the U.S. has failed to act in this way, has refused to carry out the proper “cost-benefit analysis,” can only be explained by a Jewish conspiracy. That is what passes for sophisticated thinking if you are a neorealist.

No, I do not believe that the U.S. intelligence community has stumbled into the Walt and Mearsheimer fever swamp. But the NIE’s resort to neorealist analysis is characteristic of ignorance: there is no reason to use this approach if you know what is going on. And that is the real problem. The U.S.—amazingly—publishes its National Intelligence Estimates. We make our policy in view of the entire world, and thereby impose serious constraints on our own government. We will not be able to be comfortable with Iran until we know as much about them as they know about us, and until they are as constrained by public debate as we are. And when we get that kind of Iran, we will not need high-profile but analytically shallow NIE’s.