The unleashed power of the atom,” Albert Einstein wrote in 1946, “has changed everything save our modes of thinking, and we thus drift toward unparalleled catastrophe.” Winston Churchill noted in 1955, however, that nuclear deterrence might produce stability instead and predicted that “safety will be the sturdy child of terror, and survival the twin brother of annihilation.” Einstein’s view became the touchstone of the modern peace movement. Churchill’s view evolved into mainstream Western nuclear strategy and doctrine. Both argued that the nuclear revolution had fundamentally transformed international politics. Both were wrong.

Since the 1940s, nuclear weapons have greatly affected defense budgets, political and military posturing, and academic theory. Beyond that, however, their practical significance has been vastly exaggerated by both critics and supporters. Nuclear weapons were not necessary to deter a third world war. They have proved useless militarily; in fact, their primary use has been to stoke the national ego or to posture against real or imagined threats. Few states have or want them, and they seem to be out of reach for terrorists. Their impact on international affairs has been minor compared with the sums and words expended on them.

The costs resulting from the nuclear weapons obsession have been huge. To hold its own in a snarling contest with the Soviet Union during the Cold War, the United States spent $5–$10 trillion maintaining a vast nuclear arsenal—resources that could have been used more productively on almost anything else. To head off the imagined dangers that would result from nuclear proliferation, Washington and its allies have imposed devastating economic sanctions on countries such as Iraq and North Korea, and even launched a war of aggression—sorry, “preemption”—that killed more people than did the nuclear bombs dropped on Hiroshima and Nagasaki.

The time has long since come to acknowledge that the thinkers of the early nuclear age were mistaken in believing that the world had been made anew. In retrospect, they overestimated the importance of the nuclear revolution and the delicacy of the balance of terror. This spurred generations of officials to worry more about nuclear matters than they should have and to distort foreign and security policies in unfortunate ways. Today’s policymakers don’t have to repeat the same mistakes, and everybody would be better off if they didn’t.


Over the decades, the atomic obsession has taken various forms, focusing on an endless array of worst-case scenarios: bolts from the blue, accidental wars, lost arms races, proliferation spirals, nuclear terrorism. The common feature among all these disasters is that none of them has ever materialized. Either we are the luckiest people in history or the risks have been overstated.

The cartoonist and inventor Rube Goldberg received a Pulitzer Prize for a 1947 cartoon showing a huge atomic bomb teetering on a cliff between “world control” and “world destruction.” In 1950, the historian John Lewis Gaddis has noted, no U.S. official could imagine “that there would be no World War” or that the superpowers, “soon to have tens of thousands of thermonuclear weapons pointed at one another, would agree tacitly never to use any of them.” And in 1951, the great philosopher Bertrand Russell put the matter simply: 

Before the end of the present century, unless something quite unforeseeable occurs, one of three possibilities will have been realized. These three are:—

1. The end of human life, perhaps of all life on our planet. 

2. A reversion to barbarism after a catastrophic diminution of the population of the globe. 

3. A unification of the world under a single government, possessing a monopoly of all the major weapons of war. 

The novelist and scientist C. P. Snow proclaimed it a “certainty” in 1960 that several nuclear weapons would go off within ten years, and the strategist Herman Kahn declared it “most unlikely” that the world could live with an uncontrolled arms race for decades. In 1979, the dean of realism, Hans Morgenthau, proclaimed the world to be moving “ineluctably” toward a strategic nuclear war and assured us that nothing could be done to prevent it. 

A 1982 essay by the author Jonathan Schell asserted that the stakes were nothing less than the fate of the earth and concluded that soon “we will make our choice.” Schell continued: “Either we will sink into the final coma and end it all or, as I trust and believe, we will awaken to the truth of our peril . . . and rise up to cleanse the earth of nuclear weapons.” In the spirit of the times, the following year, a chart-topping pop song traced the dangers of accidental nuclear war, and the year after, Brown University students passed a referendum demanding that the university health service stockpile suicide pills for immediate dispensation to survivors in the event of a nuclear attack. 

Disasters were certainly possible, and a healthy appreciation of the dangers nuclear weapons posed eventually led to the development and spread of best practices in strategy and safety. But prudence in controlling tail-end risks sometimes evolved into near hysteria. Nuclear exchanges were assumed to be easy to start, hard to stop, and certain to end up destroying life on earth. 

Nuclear proliferation has been a perennial source of fear. During the 1960 U.S. presidential campaign, John F. Kennedy predicted that there might be “ten, 15, or 20” countries with a nuclear capability by the next election, and similar declarations continue. And since 9/11, nuclear terrorism has been the nightmare of choice.

Ever since the dropping of the bomb, in short, Armageddon and apocalypse have been thought to be looming just over the horizon. Such fears and anxieties were understandable, especially at first. But they haven’t been borne out by the lived record of the nuclear era.


Fine, one might concede. In retrospect, perhaps the risks were exaggerated. But at least there is a retrospect—which there might not have been without nuclear weapons, since they staved off a third world war, right?

Actually, no. Nuclear strategy—a theoretical and nonexperimental enterprise—has been built on a grand counterfactual: the notion that without the prospect of nuclear devastation hanging over its head, the postwar world would have collapsed into a major conflict yet again. But this turns out to be just a story, and less history than fable.

The nuclear-deterrence-saved-the-world theory is predicated on the notion that policymakers after 1945 were so stupid, incompetent, or reckless that, but for visions of mushroom clouds, they would have plunged the great powers back into war. But the catastrophic destruction they experienced in their recent war (one they had tried to avoid) proved more than enough to teach that lesson on its own, and there is little reason to believe that nuclear weapons were needed as reinforcement.

Nuclear weapons have proved useless in conventional or guerrilla warfare, lousy at compellence, and not very good at deterrence.

Moreover, the Soviet Union never seriously considered any sort of direct military aggression against the United States or Western Europe. After examining the documentation extensively, the historian Vojtech Mastny concluded that the strategy of nuclear deterrence was “irrelevant to deterring a major war that the enemy did not wish to launch in the first place.” He added: “All Warsaw Pact scenarios presumed a war started by NATO.” In 1987, George Kennan, the architect of containment himself, had agreed, writing in these pages, “I have never believed that [Soviet leaders] have seen it as in their interests to overrun Western Europe militarily, or that they would have launched an attack on that region generally even if the so-called nuclear deterrent had not existed.” 

Moscow’s global game plan stressed revolutionary upheaval and subversion from within, not Hitlerian conquest. Given Russia’s calamitous experience with two world wars, a third was the last thing Soviet policymakers wanted, so nuclear deterrence was largely irrelevant to postwar stability. Nor has anyone ever come up with a compelling or even plausible rationale for using such weapons in conflicts short of total war—because there simply aren’t many targets that can’t be attacked as effectively with conventional weapons. 

Nuclear weapons have also proved useless in conventional or guerrilla warfare, lousy at compellence (think Saddam Hussein refusing to leave Kuwait), and not very good at deterrence (think the Yom Kippur War or Argentina’s seizure of the Falklands). There are circumstances in which such weapons would come in handy—say, in dealing with a super-aggressive, risk-acceptant fanatic leading a major country. But that has always been a remote possibility. The actual contribution of nuclear weapons to postwar stability, therefore, has been purely theoretical—extra insurance against an unlikely calamity.


Great powers are one thing, some might say, but rogue states or terrorist groups are another. If they go nuclear, it’s game over—which is why any further proliferation must be prevented by all possible measures, up to and including war. 

That logic might seem plausible at first, but it breaks down on close examination. Not only has the world already survived the acquisition of nuclear weapons by some of the craziest mass murderers in history (Stalin and Mao), but proliferation has slowed down rather than sped up over time. Dozens of technologically sophisticated countries have considered obtaining nuclear arsenals, but very few have done so. This is because nuclear weapons turn out to be difficult and expensive to acquire and strategically provocative to possess.

They have not even proved to enhance status much, as many expected they would. Pakistan and Russia may garner more attention today than they would without nukes, but would Japan’s prestige be increased if it became nuclear? Did China’s status improve when it went nuclear—or when its economy grew? And would anybody really care (or even notice) if the current British or French nuclear arsenal was doubled or halved?

Alarmists have misjudged not only the pace of proliferation but also its effects. Proliferation is incredibly dangerous and necessary to prevent, we are told, because going nuclear would supposedly empower rogue states and lead them to dominate their region. The details of how this domination would happen are rarely discussed, but the general idea seems to be that once a country has nuclear weapons, it can use them to threaten others and get its way, with nonnuclear countries deferring or paying ransom to the local bully out of fear. 

Except, of course, that in three-quarters of a century, the United States has never been able to get anything close to that obedience from anybody, even when it had a nuclear monopoly. So why should it be true for, say, Iran or North Korea? It is far more likely that a nuclear rogue’s threats would cause its rivals to join together against the provocateur—just as countries around the Persian Gulf responded to Saddam’s invasion of Kuwait by closing ranks to oppose, rather than acquiescing in, his effort at domination.

If the consequences of proliferation have so far proved largely benign, however, the same cannot be said for efforts to control it. During the 2008 U.S. presidential campaign, Senator Barack Obama of Illinois repeatedly proclaimed his commitment to “do everything in [his] power to prevent Iran from obtaining a nuclear weapon—everything,” and his opponent, the Republican senator from Arizona John McCain, insisted that Iran must be kept from obtaining a nuclear weapon “at all costs.” Neither bothered to tally up what “everything” entailed or what the eventual price tag of “all costs” would be.

If the consequences of proliferation have so far proved largely benign, however, the same cannot be said for efforts to control it.

All they needed to do was consider the fate of one country to understand the potentially disastrous consequences of such thinking. The Iraq war had been sold as an act of preventive counterproliferation, with President George W. Bush pointedly warning that “the United States of America will not permit the world’s most dangerous regimes to threaten us with the world’s most destructive weapons.” A nuclear Iraq was considered unacceptable because it would “hold [its] neighbors hostage.” Put aside for a moment the fact that Saddam had actually mothballed his covert weapons of mass destruction programs years earlier, so that the war turned out to be unnecessary by its own rationale. Imagine that Saddam, with his resentful population and unreliable army, had managed to acquire a modest nuclear capability. What would have happened then? What could and would he have done with the weapons? Something worse than launching the war to prevent Iraq from going nuclear, which, along with its aftermath, has killed hundreds of thousands of people and destabilized an entire region?

As for nuclear terrorism, ever since al Qaeda operatives used box cutters so effectively to hijack commercial airplanes, alarmists have warned that radical Islamist terrorists would soon apply equal talents in science and engineering to make and deliver nuclear weapons so as to destroy various so-called infidels. In practice, however, terrorist groups have exhibited only a limited desire to go nuclear and even less progress in doing so. Why? Probably because developing one’s own bomb from scratch requires a series of risky actions, all of which have to go right for the scheme to work. This includes trusting foreign collaborators and other criminals; acquiring and transporting highly guarded fissile material; establishing a sophisticated, professional machine shop; and moving a cumbersome, untested weapon into position for detonation. And all of this has to be done while hiding from a vast global surveillance net looking for and trying to disrupt such activities.

Terrorists are unlikely to get a bomb from a generous, like-minded nuclear patron, because no country wants to run the risk of being blamed (and punished) for a terrorist’s nuclear crimes. Nor are they likely to be able to steal one. Notes Stephen Younger, the former head of nuclear weapons research and development at Los Alamos National Laboratory: “All nuclear nations take the security of their weapons very seriously.”

The grand mistake of the Cold War was to infer desperate intent from apparent capacity. For the war on terrorism, it has been to infer desperate capacity from apparent intent. 


For nearly three-quarters of a century, the world has been told it is perched precariously on Rube Goldberg’s precipice, perennially at risk of plunging into apocalyptic devastation. But oddly enough, both we and the weapons are still here. Understanding their actual impact and putting them into the proper context would enable policymakers to view nuclear matters more sensibly. 

In practice, that would mean retaining the capabilities needed to respond to the wildly unlikely nightmare scenario of having to deter a possible future Hitler while pruning nuclear arsenals and stepping back from dangerous strategies and postures. It would mean working with North Korea to establish a normal condition in the region and worrying about reducing its nuclear capabilities later. There is nothing wrong with making nonproliferation a high priority—indeed, it would do a favor to countries dissuaded from pursuing nuclear weapons by saving them a lot of money and pointless effort. However, that priority should be topped by a somewhat higher one: avoiding policies that can lead to massive numbers of deaths under the obsessive sway of worst-case fantasies.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • JOHN MUELLER is Adjunct Professor of Political Science and Woody Hayes Senior Research Scientist at Ohio State University and a Senior Fellow at the Cato Institute. He is the author of Atomic Obsession: Nuclear Alarmism From Hiroshima to Al Qaeda. 
  • More By John Mueller