Culture and Deterrence

The debates among American strategists before both the first (1991) and the second (2003) Gulf Wars naturally differed, because strategic circumstances differed. But at least one common element—aside from the obvious fact that Iraq was the target of both campaigns—stood forth: disagreement over whether the United States could deter a nuclear-armed Iraq. As a rule, those who believed we could opposed war in both cases, while those who were skeptical that deterrence would work inclined to favor military action to prevent a nuclear-armed Iraq from coming into being (though not necessarily to favor the second war we actually got, when and how we got it).

Clearly enough, this disagreement reflects deeper views about what deterrence is and how it works—and it is an ongoing debate, now transposed to disagreements about policy toward Iran. Those who believe in the robustness of deterrence weigh the costs and risks of U.S. military action against Iran and decide firmly against it; those who are skeptical that deterrence will work see before them a different calculus. What distinguishes these two views? In essence, different views on how the human mind navigates social realities.

Those who believe in the robustness of deterrence see a universal logic inherent in WMD that flows ineluctably from the nature of the weapons themselves. It is not as though cultural differences do not exist, or that different societies do not have different styles of reasoning—granted, they do, such advocates will admit. But all these differences are trumped by the manifest consequences of what an exchange of nuclear weapons would mean. Different reasoning styles is not the same as different rationalities or irrationality, and any rationale person will avoid personal and national suicide. That is why Richard Rhodes, Pulitzer-prize winning author of The Making of the Atomic Bomb, could write in the middle of the Kuwait crisis in 1990 that it would not matter if Iraq gained nuclear weapons capabilities because “45 years of postwar history has demonstrated that acquiring such weapons in a nuclear-armed world is inescapably self-deterring.” It is how Stanley Hoffmann, invoking the authority of Zbigniew Brzezinski, could write that “deterrence, which has worked against far greater powers, remains an effective substitute for preventative war. Israel could deter Iraq, and the US as well as other nuclear states could provide a nuclear guarantee to countries threatened by Iraq’s nuclear capacity.”

These views are supposedly buttressed by inarguable historical facts. From the time in 1947 that more than one country possessed nuclear weapons, no such weapon has ever been fired in anger, notwithstanding a series of extremely tense political crises and very deep-seated ideological conflict. Not only did the United States and the Soviet Union never use them, neither has China, France, Britain or other lesser and newer nuclear powers such as Israel, India, and Pakistan. Everyone has been either deterred or self-deterred, to use Rhodes’ language, and so, the argument goes, would Iraq, Iran, North Korea, and other states foolish enough to waste their resources on nuclear weapons programs. Facts are facts; case is closed; who could argue with that?

Me. I argued against this view more than 15 years ago, and nothing that has happened since has changed my mind: nuclear weapons are not by their very nature universally self-deterring. Deterrence is a collective psychological act, and the human mind is shaped not only by its capacity for rationality but also by emotion, moral logic and, yes, culture. Indeed, the impact of culture on the strategic realm after 9/11 is, or should be, more obvious than ever, and that impact is highly variable. I will reprise my narrower argument in a moment, but it is worth reflecting first on the logical fallacies of the robust-deterrence school of thought at their core. I say fallacies—plural—because the mistake being made here is a compound one. It is a mistake, first of all, to simply superimpose the U.S.-Soviet experience onto other geopolitical domains, but robust-deterrence advocates have also misinterpreted the U.S.-Soviet deterrence experience itself. Both mistakes involve the underestimation of the significance of culture on strategic assessment.

Mistake Number 1

A funny thing happened on the way to the Cold War. We American children of the Enlightenment are persuaded of the universal validity of certain political propositions (luckily enough, those propositions we ourselves happen to hold dear). This persuasion of ours collided in the late 1940s and 1950s with a rising social science establishment hell-bent on “hardening” its image, if not its actual work habits, as being scientific. The result is that Americans came to believe—perhaps more accurately, to assume as if by second nature—in a theory of politico-military behavior that was both universal in application and scientific in nature. Like a child who thinks that the names for objects inhere in the objects themselves, we believed we were discovering objective truths about life in the early years of the nuclear age rather than imagining (and thus inventing) such truths in culturally idiosyncratic ways.

Because these truths were supposedly objective and universal, we assumed that the Soviet leadership must have thought about these weapons and their uses just as we did. Therefore, we could plan the impact of our policies because we knew how the Soviet Union and others would react to them: namely, just as we ourselves would react.

The trouble was, they didn’t. Robert McNamara, the U.S. Secretary of Defense who first tried seriously to fuse the formal theory of the day with national policy in the nuclear age, postulated that once a strategic plateau was reached between U.S. and Soviet strategic forces, it would be possible to cap that plateau and then to negotiate sizable and stabilizing reductions. McNamara assumed that the Soviet leadership shared the view that nuclear weapons had no rational battlefield use and that any notion of practicable strategic superiority was meaningless, not just because the weapons were unusable but because neither side would allow the other to attain lopsided advantages. When Soviet behavior did not reflect McNamara’s anticipations, some people began asking why.

As it happened, Nathan Leites of the RAND Corporation had been thinking about such matters for some years. He postulated that every national leadership cadre would have what he called an “operational code” for dealing with strategic issues, a general orientation to strategy shaped by culture: language, religious beliefs, historical memory and imagination, and the whole gamut of metaphors that enable societies to make sense of the world and act purposely in it. Leites understood, along with every skeptic of positivism from Immanuel Kant to I. W. Thomas, that, in the latter’s words, people do not first see and than define, they define and then they see. The Soviets would not in all or perhaps even most cases see the world as we did, and they would not react as we would react. They would not necessarily share our assumptions or properly assess our motives, nor we theirs (unless we deciphered their operational code).

Most American strategic analysts either ignored Leites or did not understand what he was saying. B. F. Skinner was a lot more popular in those days than Kant, after all. Not until the late 1970s did the dominant view of the U.S. strategic analytical community accept the existence of distinctive strategic cultures. This happened for a variety of reasons: It had by then become impossible to explain Soviet behavior in ways that harmonized with our presumed universal understanding of strategic logic; Chinese strategic behavior and language looked odd compared to ours; and the case for taking cultural factors seriously was finally being made in language that arms and arms-control experts could understand.

Mistake Number 2

Those who still argued the positivist case for robust deterrence in 1990 were, perhaps, holdouts from the early years of the Cold War, their brains stuck in some kind of intellectual amber, still believing in objective and universal truths about deterrence. But they compounded their error with the assumption that what worked for the U.S.-Soviet relationship (for reasons they at least partly misunderstood) would also work both for lesser-included cases (e.g., a superpower deterring smaller nuclear powers) and for different cases (e.g., a smaller nuclear power deterring one or more other small nuclear powers). As I argued in 1991, there are at least six fairly obvious reasons why this assumption is mistaken.

First, U.S.-Soviet deterrence was a straightforward bilateral proposition; no other nuclear powers really played in the same weight-class. Deterrence in a nuclear-armed Middle East, which may come to include a nuclear-armed Israel, Iran, Egypt, Saudi Arabia, Pakistan, and so forth—not a far-fetched proposition, regrettably—would have to be as omni-directional as Middle Eastern antagonisms and fears, and this is not to speak of how nuclear powers outside the region would figure in all this. Calculations of sufficiency would be far less certain; states might well seek arsenals equivalent with not only one potential antagonist but several. Such a situation would conduce to arms races, crisis-instability, miscalculations, and accidents. Then there is the heightened risk of fissile materials theft and diversion along with constant fears of further proliferation convoluting still further the strategic environment. To whom does this sound like a formula for stable deterrence?

Second, first-generation nuclear weapons and delivery systems are relatively unreliable and vulnerable to preemption. Stable deterrence depends on mutually survivable forces, not mutually vulnerable ones.

Third, deterrence failure in the U.S.-Soviet case was presumed to signify massive if not total societal destruction because of the mega-tonnage involved and the unreliability of any built-in escalation brakes. Not so, necessarily, in the Middle East, with much smaller arsenals and a reluctance to shoot off all of one’s assets lest a third party take advantage after the fact. And less than ultimate stakes produce less than perfect caution.

Fourth, U.S. and Soviet caution in strategic relations stemmed from a fact we still tend to take for granted: Both leaderships actually cared about the well-being of those they ruled, even if in the Soviet case the population’s production capability rather than human value was uppermost. But we saw repeated demonstrations of mass murder inside Iraq by the Sunni ruling elite against Kurds and the majority Shiite population during Baath rule, without regard for the injury done the state, and it is not unreasonable to wonder whether the fragility of the civil bond between rulers and ruled in multiethnic and highly stratified Middle Eastern societies weakens significantly the fundamental social basis of deterrence.

Fifth, another possible discontinuity can affect the stability of deterrence: crazy states or crazy leaders in charge of highly authoritarian political cultures. Saddam Hussein was a malignant narcissist who could not have cared less about the slaughter of millions. Mao and Stalin were mass murderers with exotic personalities, too. It is not that democracies never produce scary leaders: a close study of Woodrow Wilson, for example, evokes gratitude that WMD did not yet exist during his lifetime. But on the whole, top-heavy political systems are far more prone to recruit madmen or fanatics to the pinnacle of power, and neither madmen nor fanatics are reliable stewards of nuclear deterrence because neither can be presumed to care two figs about ordinary people.

A special case in point concerns religious fanatics. During the Cold War, when nuclear deterrence theory was invented and debated, this problem simply never arose. Yes, it’s true, lots of people described Marxism-Leninism as a secular religion, and not without both reason and utility. But religion and ideology are not the same: they are not similarly organic to society and they don’t motivate and mobilize masses of people in the same ways. When President Ahmedinejad of Iran speaks in apocalyptical, millenarian terms, many Western secular sophisticates force themselves to believe he can’t be serious. Most likely, he is quite serious. When radical Muslim clerics describe why it is alright for Muslims to incidentally kill other Muslims in mass terrorist attacks—because “Allah will know his own” so that the innocent will become instant martyrs in paradise—they give every appearance of actually meaning what they say. An eleventh-grade Iranian textbook teaches that in the coming era-ending war against the infidels, Muslims cannot lose: “Either we all become free, or we will go to the greater freedom which is martyrdom. Either we shall shake one another’s hand at the victory of Islam in the world, or all of us will turn to eternal life and martyrdom. In both cases, success and victory are ours.”

How does one deter people who believe that, who are willing and even eager—from the sound of it—to turn their entire country and their entire religious sect into a suicide bomb?

Sixth, the idea that nuclear powers could extend their influence to protect non-nuclear allies was a standard-issue plank of Cold War strategic platforms. And it worked, despite the fact that protected allies were never entirely confident about it. But the idea that U.S. power protected Italy and the Netherlands and the Federal Republic of Germany from being attacked—and presumably, according to the theory, that U.S. nuclear power protected them from nuclear attack—was never based on firepower alone. It was based on willpower, and ultimately, therefore, the credibility of extended deterrence depended on persuading the target of the deterrence posture that the protecting power cared just as much about the well-being of an ally’s population as it cared about its own—or close enough, anyway, for practical purposes. Extended deterrence worked because Western democracies shared not only interests but also core principles in common.

Now, U.S. nuclear protection of Israel is credible enough, though Israel, as a nuclear power itself, doesn’t need a U.S. umbrella. But a pledge of U.S. protection for Kuwait? Egypt? Jordan? Saudi Arabia? Would those governments even want such public pledges, and would others really believe them? Could Israel, as Stanley Hoffmann seemed to suggest in 1991, really provide a credible deterrent for Kuwait to protect it from a nuclear Iraq? Who on earth would give credence to that?

Finally on this point, it is worth thinking about the possibility that a nuclear-armed Iraq, or Iran, not only would not be deterred by the United States, but could in fact deter the United States. This is the matter of inverted deterrence, a tricky but entirely realistic possibility.

The United States might have been able to deter a nuclear Iraq, and it may be able in the future to deter a nuclear Iran, from undertaking a direct attack against the United States or Israel. But could U.S. or Israeli nuclear weapons deter an aggressive nuclear-armed autocracy from stoking proxy warfare against neighbors, in Lebanon for example? From surreptitiously peddling nuclear know-how or materiel? From launching conventional aggression against a neighbor?

If Iraq had had nuclear weapons in 1991, would the United States have sent an expeditionary force to expel Iraqi forces from Kuwait? If, in the hypothetical absence of the second Gulf War, Iraq had acquired nuclear weapons in 2007 and reinvaded Kuwait conventionally in 2008, would the United States be sending an expeditionary force into the teeth of a nuclear shield to repeat its earlier successes? Under such circumstances, who, exactly, would be deterring whom? These are not easy questions to answer; but who doubts that U.S. calculations of risk and benefit would be dramatically different under such circumstances?

This is why the argument that U.S. policy should have focused more on North Korea (and less on Iraq) in 2002-03 because it was the greater nuclear threat was so foolish. Once past the nuclear threshold, as all assumed North Korea was, U.S. options narrow as its risks are magnified; all the more reason to prevent rogue regimes from getting past that threshold in the first place. But best to do it, of course, in a broader diplomatic context that does not encourage other countries to sneak under the nuclear wire before the Americans bestir themselves to act—in other words, in such a way that an act of counterproliferation policy does not elicit the need for further acts of counterproliferation policy.

A Lebanese Coda

It follows from this little thought experiment that culture isn’t the only element one needs to consider when thinking about deterrence. There are many aspects of strategic logic that, while not universal, are common enough to be appreciated across many cultures. But it is a dangerous mistake to dismiss the relevance of culture on strategic decision-making. Take the recent flare-up in Lebanon as an example.

Why did Palestinian nationalists a generation ago and Shiite fanatics today insist on believing that Israel is weak, fragile, a mere “spider’s web”? In part because in traditional Islamic lore, Jews are caricatured as weak, inferior, craven, and cowardly. Why, apparently, did Sheikh Nasrallah not expect the Israeli reaction he got from Hezbollah’s July 14 attack across Israel’s northern border? In part because he didn’t appreciate the impact of Israelis’ hearing, still in the psychological shadow of the Holocaust, a fresh barrage of anti-Semitic and eliminationist rhetoric coming from Hezbollah’s sponsors in Tehran (as well as from Nasrallah himself). Why did Israeli leaders underestimate the tenacity of Hezbollah fighters? In part because they believed mistakenly that religious fanatics cannot be trained in modern military technique, and because they could not take seriously the possibility than an entire society in southern Lebanon could really assume the mentality of a death cult.

Examples don’t end here, of course. The war on terror is suffused with cultural predicates, many of which American leaders misread. The President’s Freedom Agenda, for example, no less than the early American approach to strategic deterrence, is based on Enlightenment notions of ideal human social and political organization that are assumed to be universal but that, ironically enough, are idiosyncratically Western. The projection of these assumptions onto Muslim countries with their own idiosyncratic characteristics is worse than futile; it is counterproductive. But that is a subject for a different essay, one called, perhaps, “Culture and Inadvertent Provocation.”


Rhodes, “Bush’s Atomic Red Herring,” New York Times, Nov. 27, 1990.

Hoffmann, “The Price of War,” New York Review of Books, Jan. 17, 1991.

For an application of culture not to strategy but to law, see Lawrence Rosen, Law as Culture: An Invitation (Princeton University Press, 2006). For a good summary of cultural approaches to politics, see chapter 2 of Samuel Huntington’s The Clash of Civilizations and the Remaking of World Order (Simon & Schuster, 1996), especially pp. 41-5.

Case in point: Robert Legvold’s brilliant essay, “Strategic Doctrine and SALT: Soviet and American Views,” Survival, Jan.-Feb. 1979.

Argued in “Will Saddam Get the Bomb?” National Review, May 13, 1991.

See Anna Simons, “Making Enemies, Part Two,” The American Interest, Autumn 2006, pp. 40-1.

Quoted in Bernard Lewis, “August 22,” Wall Street Journal, Aug. 8, 2006.