Does MAD really work?

One of B.F. Skinner's pigeons in a box

David P. Barash is professor of psychology emeritus at the University of Washington; he is writing a book about nuclear deterrence.

For all its mathematical folderol – not to mention being entrusted with the fate of the Earth – deterrence is likely a superstition.

In 1948, psychologist B. F. Skinner conducted an experiment in which hungry pigeons were fed on a random schedule. Soon three quarters of them were behaving in unusual ways, depending on what each had been doing just before getting food: one rotated her body (always counter-clockwise), another swung his head like a pendulum, a third wiggled her feet, and so on. The resulting research report was titled “Superstition in the pigeon.”

We’ll never know what if anything Skinner’s pigeons were thinking. But there’s no doubt that it’s time for us to think – or rather, rethink – our reliance on deterrence.

Although conventional deterrence has existed for a long time – think of China’s Great Wall (deterrence by denial), Rome’s use of its legions (deterrence by punishment) or even the roars of a lion, and thorns of a rose bush – nuclear deterrence is of course quite new, only existing since 1945. Initially, the weapons and their presumed deterrent effect were a U.S. monopoly, thought to prevent the Red Army from rolling into Western Europe. Then, when the USSR became nuclear armed, we and they entered the MAD era (Mutually Assured Destruction), from which neither country has yet emerged even as others have joined, doctrines have been refined, and new weapons deployed.

Throughout, people have been remarkably pigeon-like, rarely questioning the underlying assumptions of nuclear deterrence, of which Winston Churchill proclaimed “Safety will be the sturdy child of terror, and survival the twin brother of annihilation.” Despite the terror, maybe deterrence really has been sturdy; after all, we have so far survived the nuclear age and avoided annihilation. But such confidence is, at best, premature.

Correlations, after all, can be spurious, as with ice cream consumption and drowning: although the two are correlated, it’s not because eating ice cream makes people drown, but because both events tend to happen in hot weather.

If a pigeon spun around and didn’t get fed, it would presumably have been disappointed, but no great harm would have been done. But if deterrence had failed (a frequent and terrifying trope among strategic planners), we likely wouldn’t be around to bemoan that particular inadequacy. And it only needs to fail once. Moreover, if you play Russian roulette indefinitely – whether with six chambers or 600 – it is mathematically certain that eventually you will take a bullet.

Following the Cuban Missile Crisis of 1962, when we came horrifyingly close to World War III, former Secretary of State Dean Acheson observed that we had avoided Armageddon by “sheer dumb luck.” And one thing we all know about luck is that eventually it runs out.

Maybe we’re like the person who has fallen from a skyscraper and who reassures herself, as she plummets down, “So far, so good.” 

The iconic argument for the success of deterrence is that the Cold War between the U.S. and the USSR never went nuclear. But aside from luck, perhaps this cheery outcome arose simply because the two countries never had any motivation sufficient for any war, conventional or nuclear. Unlike say, India and Pakistan - both of which have nuclear weapons and have also had conventional wars – the two Cold War opponents didn’t share a common border or have conflicting territorial claims. And it’s worth recalling that the Cuban Missile Crisis, rather than being a triumph of nuclear deterrence, was caused by deterrence itself, after Khrushchev sought to shore up the Soviet Union’s posture vis-à-vis the U.S., after we had deployed nuclear armed Thor missiles in the UK and Jupiter missiles in Turkey – which was ordered by President Eisenhower in hopes of, well, furthering our deterrence of the Soviets! It is reasonable to conclude that nuclear war wasn’t avoided because of deterrence but in spite of it.

The same applies to the numerous cases in which false alarms have brought deterrence to the brink of failure, as for example in 1983 when Stanislav Petrov, a mid-ranking Soviet air defense officer, received a report that five missiles, fired from the U.S., were heading toward the Russian homeland. This occurred at an especially fraught time in U.S.-Soviet relations, when the Reagan Administration was cavalierly maintaining the feasibility of surviving a nuclear war with the “evil empire,” and had recently shot down a Korean Air passenger plane, mistaking it for an American spy mission. Petrov concluded on his own that since his country’s early warning system was newly installed and liable to have some bugs, the report was probably a false alarm, so – risking serious punishment for insubordination – he didn’t pass along the alert, which would have necessitated that the ill and elderly then-President Andropov decide within minutes whether to “retaliate”  … to an attack that never happened.

As for that seemingly long nuclear peace since 1945, the historical reality is that the time-span from the beginning of the nuclear age until now isn’t really all that impressive. Not only has the U.S. been involved in many conventional wars (Korea, Vietnam, Iraq, Afghanistan), but even war-prone Europe experienced lengthy periods of peace during the 19th century alone: between the end of the Napoleonic Wars and the Franco-Prussian War, and thence until World War I, and so forth into the 20th century. Each time, peace was followed by war, and when that happened, it was fought with the weapons then available. Considering this, the decades-long absence of nuclear war – so far – may be something to savor, but is less than dispositive.

All of which makes one doubt the dogma that nuclear deterrence has worked, and that we should feel confident that it will continue to do so. Moreover, there is no evidence that nuclear threats – whether overt, via a proclaimed policy of deterrence, or implied, simply by possessing a nuclear arsenal – have conveyed increased international clout. On many occasions, non-nuclear countries have even attacked nuclear armed ones. China sent its army against U.S. forces during the Korean War, in 1951, even though the U.S. had hundreds of nuclear bombs, and Mao would not have any until 13 years later. Non-nuclear Argentina was similarly undeterred when it invaded the Falkland Islands, a territory of nuclear armed Britain. By the same token, during the first Gulf War in 1991, non-nuclear Saddam didn’t hesitate to fire Scud missiles at nuclear Israel; the government of Yitzhak Shamir didn’t play its supposed deterrence card and vaporize Baghdad in return.

There are, moreover, a number of other reasons why the Emperor Deterrence has no clothes, of which one in particular brings us to the current crisis on the Korean Peninsula. Deterrence readily elides into provocation, since doctrine, weapons, military exercises and verbal taunts lend themselves to interpretation as signaling intent to mount a first strike. Conventional military postures on both sides of the 38th parallel have long provided more than enough deterrence, with the thousands of North Korean artillery tubes roughly matched by the well-equipped South Korean military, along with about 28,000 American troops serving as a “tripwire.” But the North’s excessive anxiety about inhibiting an invasion aimed at regime change has driven the Kim government to pursue a hyperactive nuclear weapons and missile program, an example of deterrence run amok that has evoked a comparably overblown and potentially lethal response from the Trump Administration.

This action-reaction sequence italicizes one of the many deep weaknesses of deterrence: not only does it rely on the perception by each side that the other is being self-protective rather than aggressively threatening (easier said than done), it assumes that all participants will behave with cool, well-informed, thoughtful, and rational judgment – even though everything known about human behavior (perhaps especially that of Messrs. Kim and Trump) is that they can be violent, impulsive, thin-skinned, vindictive, ill-informed and downright sociopathic.

There are other problems, not least that the U.S. in particular has been moving toward smaller and more accurate nuclear weapons, especially suitable for tactical, battlefield use. This transition has been motivated by efforts to overcome one of the most troublesome aspects of deterrence, the fact that all-out nuclear war would be so horrible, and its effects so globally destructive (no matter who initiates it) that the weapons themselves aren’t really feasible; hence, they – and the deterrence they ostensibly underpin – lack credibility. The potent paradox is that the only way to imbue nuclear weapons with credibility (and thus, to bolster deterrence) is to make them relatively small and accurate enough that they are credibly usable – but the more usable they are, the more liable they are to actually be used. Add to this the fact that every war game scenario shows that such use inevitably escalates to all-out nuclear war.

The good news – and there is some – is that there are ways out of the deterrence trap. For starters, effective deterrence can be achieved, at least in the short term, with a tiny fraction of the overkill arsenals currently deployed. Despite the ignorant insistence of President Trump, there is certainly no need for more, and a crying need for fewer, eventually down to zero. Threats can be toned down, not just verbally but in terms of the weapons being deployed. The destabilizing targeting of any other country’s nuclear forces can also be ended. A fissile materials cutoff can be implemented, along with no-first-use doctrines. We can ratify the Comprehensive Test Ban Treaty, and sign on to the recently passed Nuclear Ban Treaty, already endored by more than 120 countries, and offering moral and legal authority to the delegitimation of these genocidal weapons.

We cannot mandate changes in Pyongyang’s nuclear procedures, or imbue Donald Trump with insight, thoughtful decency and a sense of international responsibility. but we can pass legislation mandating that no U.S. president – not Trump, not Pence, no one – can initiate first use of nuclear weapons, ever. Better yet, we can insure that this never happens by getting rid of these indefensible weapons altogether, along with the deeply flawed superstitious ideology of deterrence that has justified their existence.

In short, we can prove ourselves wiser than pigeons.