To build a nuclear bomb you need uranium. That’s the easy part. Uranium ore is plentiful, but it isn’t very useful in its naturally occurring state. For bomb-making it must be enriched to increase the concentration of the isotope uranium-235. Weapons-grade uranium is around 90 per cent enriched, but at a pinch you can make do with a lesser grade, enriched to 50 per cent or so. Uranium can also be run through a reactor and reprocessed to obtain plutonium-239, the other main candidate for a nuclear bomb. Enrichment is not straightforward. At one point the Manhattan Project’s gaseous diffusion plants accounted for 5 to 10 per cent of all electricity use in the United States. But if you can acquire the fissile material you can make a bomb. Most people probably know the outline: a nuclear reaction involves a split atom emitting a number of neutrons, which in turn split the nuclei of other nearby atoms, and so on. A nuclear bomb produces a fission reaction in a supercritical mass of nuclear material contained in a tamper to make the reaction more efficient. The initial explosion heats the fissile core to billions of degrees Celsius. The bomb expands to create either an enormous fireball or a tiny sun, depending on how you look at it. A shockwave radiates outwards. As the fireball rises, air and dust is sucked into a pillar below, forming a mushroom cloud. Witnesses to nuclear tests say it feels like the air is tearing.
The nuclear weapon used at Hiroshima was primitive: sixty kilograms of enriched uranium slammed against itself. The Nagasaki bomb was of the more advanced implosion type. It contained just six kilos of plutonium, about the size of a shot put. But all fission bombs are primitive when compared to a thermonuclear (or hydrogen) bomb with a yield measured in megatons rather than kilotons. It comprises a fission bomb separated with polystyrene from a secondary fusion device. The secondary is a core of fissile material surrounded, these days, by the hydrogen isotope lithium deuteride. When the first stage is detonated, the radiation emitted by the fission explosion sets off a fusion reaction in the hydrogen isotopes of the secondary. The two stages of the bomb feed off each other, like a pair of autotrophic matryoshkas.
Large thermonuclear bombs are a thousand times more powerful than the first nuclear weapons. One will destroy a few square miles of city, the other hundreds. The first victims are killed by the sheer force and heat of the blast wave. Its diffraction causes the entire human body to be compressed, resulting in embolisms in the arteries and crushing the lungs and heart. Then there is the thermal pulse, which causes flash burns on exposed skin. A large proportion of the victims in Hiroshima and Nagasaki were killed by the heat. A smaller number were trapped in rubble and burned to death. Along with the blast and heat there is ionising radiation, which creates defects in individual atoms, morphs the blood and bone marrow, breaks chromosomes and irreparably damages cells. Victims vomit, and suffer ataxia and delirium. Less is known about longer lasting nuclear fallout. Marshall Islanders subjected to fallout in 1954 suffered ‘beta-burns’ within 24 hours and nuclear testing rendered their atolls uninhabitable.
The usual fate of revolutionary weapons is for their startling effects to be quickly nullified, or at least blunted, either by the invention of countermeasures or by everyone acquiring them. But effective defences against thermonuclear weapons have been hard to come by. Instead, their very power has constrained their use. War has always been destructive for the losers. It’s sometimes destructive for the victors too: the Soviet Union lost 13 per cent of its population in the Second World War. But until the invention of nuclear weapons, victory didn’t also mean annihilation. There is an irony in the fact that city-killer weapons arrived at the precise moment that humanity became a predominantly urban species.
Nuclear weapons haven’t been used in war since 1945, but there have been many close calls. In 1956, a B-47 bomber disappeared over the Mediterranean with two nuclear weapons on board. It was never found. In 1960, US nuclear early warning systems were accidentally triggered by the moon. The same happened with flocks of migrating geese. In 1966, a B-52 crashed mid-air and dropped three thermonuclear bombs on a Spanish village (the cores didn’t detonate). These are the accidents. The times when intentional nuclear war seemed imminent are better known. Countries with nuclear weapons often claim that only their head of state or government can order their use, but in practice states recognise that this would make them vulnerable: what if the national leader is dead? Most have allowed some delegation, so that in extremis subordinates or military commanders can order nuclear strikes.
Fred Kaplan’s 1983 book The Wizards of Armageddon was an invaluable account of early American nuclear strategy. In the immediate postwar period, US military theorists were impressed by the power of what Kaplan called the ‘absolute weapon’, but they saw it principally as an advance on the proven techniques of incendiary bombing. Bernard Brodie, an academic at Yale and then the RAND corporation, described the implications of nuclear weapons in 1946: there can be no winners in the conventional sense, and the advantage is in the threat rather than the execution. In order to deter attack, all you have to do is show that if you’re hit you will hit back. When a state lets it be known that it has the capacity to carry out a second-strike retaliation it’s almost unthinkable that it would be subjected to nuclear attack in the first place.
The US lost its nuclear monopoly in 1949, but remained by far the most powerful state in the world. The American homeland was by any measure very secure. But development of nuclear technology, and of ICBMs, was a potential challenge to its position. If security of the population, or even the state, was the main concern, it would have been rational for the US to push for arms controls on nuclear weapons technology. This was never seriously considered. The first thermonuclear device was detonated by the US on the former island of Elugelab in 1952. The Soviet Union built its first thermonuclear weapon three years later. By the 1960s the superpowers could launch them using missiles travelling at 16,000 mph.
The question for the US was this. In a world with nuclear weapons, how should it continue to exercise its global power, from Guam to Congo? It was essential to retain the imperial protectorates, which included Western Europe. During the Eisenhower administration, US policy was to threaten the Soviet Union with ‘massive retaliation’ over disagreements at the periphery of the Soviet bloc. The political leadership expressed what was already military policy. Admiral Arthur Radford summed up the thinking in American military circles when he described nuclear weapons as ‘the primary munition of war’. The Joint Chiefs of Staff had already started on target lists of Soviet cities, power plants and oil installations.
Kaplan was responsible for raising the reputation of William Kaufmann, a civilian nuclear strategist and critic of massive retaliation. Another RAND corporation man and an influential adviser to the Department of Defence on nuclear matters, Kaufmann argued that threatening to respond to minor aggression with total nuclear war was an ineffective bluff. The US was saddling itself with a choice between ‘the immeasurable horrors of atomic war’ and loss of prestige if its bluff were called. A nuclear response to non-nuclear aggression far from the homeland was like ‘a sparrow hunt with a cannon’. Practical maintenance of America’s imperial and quasi-imperial positions required conventional military forces and alliances in Germany, Taiwan and South Korea.
There were other ways of waging ‘limited war’. In 1957, Henry Kissinger argued for smaller, tactical nuclear weapons. Kaufmann and others pulled his argument apart by showing how easily the use of small nuclear weapons on the battlefield would escalate to full thermonuclear exchange. But this wasn’t enough to stop the generals from deploying tactical nuclear weapons in Europe. There are still around a hundred American nuclear bombs in bases in Germany, Italy, Belgium and the Netherlands. Trump’s deputy assistant secretary of defence, Elbridge Colby, architect of the 2018 National Defence Strategy, was an advocate of using tactical nuclear weapons in ‘possible armed conflicts with both smaller hostile rogue states and with larger near peers’.
Fear of a general thermonuclear war was at its peak in the late 1950s and early 1960s, but we now know that this was the period of US total nuclear dominance. At the time, American intelligence estimates exaggerated the number of Soviet ICBMs by a factor of ten (the mythical ‘missile gap’), and greatly exaggerated the number of Soviet warheads and Soviet bombers – which in any case couldn’t have reached the US without refuelling in vulnerable Arctic bases. In fact, until the mid 1960s, the Soviet Union wouldn’t have been able to survive an American nuclear attack, and couldn’t be confident in its capacity to launch a large second-strike retaliation.
In this context it’s worth considering the lectures of Herman Kahn, collected in 1960 as On Thermonuclear War. Kahn believed that a cataclysmic nuclear war between the US and the Soviet Union was imminent. He differed from others, though, in arguing that nuclear war need not mean mutual destruction: with careful planning and the proper civil defence measures, America could win such a war. Would twenty million or a hundred million die? The answer would affect postwar recuperation, but it wasn’t necessarily true that ‘the survivors will envy the dead.’ The most discomfiting part of Kahn’s argument was that subsequent records have shown he was probably right, though not for the reasons he gave. In 1960, mutual destruction was not assured. The US could have destroyed urban civilisation in the Soviet Union. The same was not true in reverse. The Soviet leadership could count on its weapons for retaliation against Western Europe but not the continental United States.
All nuclear strategy contains an element of madness, which Kahn seemed to personify. Much of the science of nuclear deterrence was, and still is, a matter of bluster. Kahn’s analysis wasn’t of the highest quality – there he sat, considering World War Eight (in 1973) – but he did make one useful contribution. He conceived the thought experiment 0f the ‘doomsday machine’, a device that would destroy the entire world population if any nuclear weapon were used. Although it would mean perfect deterrence, he argued that – because of the risk of accident – to build such a machine would be a great mistake.
Kahn didn’t know it at the time, but as Daniel Ellsberg later revealed, the ‘doomsday machine’ was only a slight extension of US nuclear designs. While the RAND intellectuals were theorising, the military continued to work on actual nuclear war plans, the details of which were kept secret even from US presidents. Strategic Air Command’s Emergency War Plan 1-49 included a list of seventy cities on which thermonuclear bombs would be dropped, from Moscow and St Petersburg to Berlin, Potsdam, Warsaw and most of what is now Ukraine and Belarus. In 1960, the generals completed a comprehensive plan for a first-strike attack, the Single Integrated Operational Plan, or SIOP-62. In the case of non-nuclear conflict with the Soviet bloc, the US would drop 3423 nuclear bombs on Soviet territory, Eastern Europe and China (the RAF was supposed to participate). Every city in the Soviet Union and China was to be destroyed. The power of the nuclear weapons to be used on Moscow alone was four thousand times that of the bomb used on Hiroshima. Military analysts predicted that around 600 million people would be killed, including 100 million in Western Europe and 100 million in neutral countries adjacent to the Sino-Soviet bloc such as Afghanistan, India and Japan. It would be hard to argue that any document in history contains greater evil; there is nothing in the Nazi archives that approaches it.
In 1961, some of the defence intellectuals who had spent the previous decade working on nuclear strategy were brought into the Kennedy administration, where their ideas would be tested. Thanks to Ellsberg, SIOP-62 was unearthed by the new administration, which sought to improve on it. US nuclear strategy had developed towards the theory of ‘counterforce’: missiles and bombers should target enemy nuclear forces rather than cities. Counterforce may appear sane when compared with the lunatic destruction favoured by previous plans, but the new proposals were potentially no less deadly: should the negotiations that were supposed to accompany strikes on missile installations and airfields fail, the destruction of cities would still be threatened. Besides, military targets are often near cities. And once the Soviet Union had acquired nuclear forces capable of a second strike, the risk of uncontrollable escalation came into play. Counterforce provided the logic for the arms race, and nuclear stockpiles grew far beyond any practical consideration. When there are large numbers of nuclear weapons, it no longer matters what the nominal targets are.
The revised SIOP-63 didn’t insist on the automatic destruction of China in response to a kerfuffle in Europe. But it too was a plan for a first strike. Nuclear historians speak of distinct eras in the nuclear period, but the shifts are either minor or illusory. SIOP-63 remained official doctrine until the early 1970s. Later administrations tinkered with nuclear guidance documents – the Carter administration added the summer dachas of Soviet leaders as targets – but actual planning was still the domain of Strategic Air Command. Politicians gave instructions on how nuclear war should be fought, and the generals in Omaha ignored them. In the event of a crisis, the US president would be given a binder, the ‘Black Book’, containing four or five options for nuclear war. Even the most restrained option would involve hundreds of strikes. The full details – which weapons would be used on which targets – were contained in the ‘Blue Book’, which no president or civilian leader was allowed to see until 1989. The last known equivalent to the SIOP is called OPLAN 8010-12. We know nothing about what it contains, but it’s a fair guess there is new material on China.
Kaplan’s new book, The Bomb, is both a sequel and an update to his Cold War history. After 1991 it was no longer possible to pretend that Moscow was a serious competitor to American global power. But US missiles were still pointed at Russian cities, obscure factories and fields that might serve as improvised airstrips. Over the course of the next decade, however, both the US and Russia halved their nuclear stockpiles. The Non-Proliferation Treaty had been in force since 1970, to discourage non-nuclear states from acquiring nuclear weapons, but the existing powers had only paid lip service to disarmament. Now the US could focus on non-proliferation as a tool of power politics. It often failed. The Clinton administration sought to prevent North Korea getting the bomb, and with the 1994 Agreed Framework succeeded in temporarily hindering its weapons programme. But after the US refused to finance civilian reactors for North Korea under international safeguards the agreement came under strain, and the George W. Bush administration pulled out of the deal entirely. Pyongyang acquired its first nuclear weapon in 2006.
Advocating non-proliferation is a common hobby for retired American officials with time on their hands and a less than clean conscience. Were the US actually committed to limiting nuclear weapons, it would at the very least have to declare a ‘no first use’ policy for its own nuclear arsenal. The Soviet Union, China and India have all made such a pledge in the past (Britain and France have not). Kaplan takes seriously Obama’s professed desire for ‘a world without nuclear weapons’, but the Obama administration refused to declare no first use. Its successes on nuclear matters – the Iran nuclear deal and the new START arms reduction treaty with Russia, signed in 2010 – were overshadowed by its commitments to build the next generation of US nuclear weapons systems. New ‘Ground Based Strategic Deterrent’ missiles will soon start replacing the Minuteman III. The US navy is getting new W93 nuclear warheads. The US air force will have B-21 stealth bombers ‘designed to overcome even an advanced adversary’s air defences’. In many respects Obama was a continuity president in matters of imperial management.
Ellsberg argued that every US president has used nuclear weapons ‘in the precise way that a gun is used when it is pointed at someone in a confrontation, whether or not the trigger is pulled’. Eisenhower threatened to use them in the Korean War and against China. Kennedy came close during the Cuban crisis. In 1969, Nixon threatened to use them in Vietnam. George H.W. Bush threatened Iraq with ‘nuclear retaliation’. The fact that none of these threats was carried out doesn’t mean they weren’t significant. Trump’s threat in 2017 to ‘totally destroy North Korea’ with fire and fury was a rhetorically extreme example in a long record. Eisenhower, Ford and Clinton all made similar threats. In response to North Korean missile tests, the US twice fired conventional missiles from South Korea into the Sea of Japan. In 2019, it withdrew from the Intermediate-Range Nuclear Forces Treaty. The Trump administration also decided to acquire more tactical nuclear weapons, with yields quite similar to the Hiroshima bomb. As Kaplan says, discussion of US nuclear weapons describes them as a deterrent, but ‘American policy has always been to strike first pre-emptively.’
Analysis of nuclear strategy is often approached from the perspective of one of the states involved, but it’s possible to take the planetary view. Nuclear weapons have existed for three-quarters of a century. There are few international controls and the weapons remain at the discretion of the nine states that possess them. Outside America, only four – Russia, China, India and Pakistan – have sizeable arsenals and are in a position to use them without consultation with the US. The overall number of nuclear warheads has decreased since the 1980s, but the support systems and delivery mechanisms on which nuclear war would depend have become more sophisticated.
The balance of power among nuclear states has fluctuated. The US has sought nuclear superiority over other states at all times, and has threatened to use nuclear weapons with dull regularity. It has no sustained appetite for arms control treaties and its war plans have included genocidal first strikes. Russia’s nuclear forces decayed through the 1990s and the early 2000s. Its nuclear-armed submarines weren’t even on patrol for much of that period. But over the last decade it has reversed some of the decay. It has maintained many smaller tactical nuclear weapons, ostensibly out of fear of a ground invasion from the West (and perhaps the south-east).
Since the 1960s, Britain’s nuclear forces have been based on submarine-launched ballistic missiles. The missiles are leased from the US and stored at a naval base in coastal Georgia. In March, the UK committed to increasing its nuclear stockpile by 40 per cent, reversing four decades of reduction. In keeping with local custom, this important strategic decision was slipped into page 76 of the government’s Integrated Review of Security, Defence, Development and Foreign Policy. Britain is replacing its Vanguard submarines with four bigger Dreadnought-class boats, which are due to arrive in the early 2030s. The government has also ordered new warheads, which must closely adhere to American designs so that they remain compatible with the Trident missile and its aeroshell. British politicians like to talk of Britain’s ‘independent nuclear deterrent’ but in practice its nuclear weapons are an appurtenance to US power. There is no chance they would ever be used without approval from Washington. Nor would those of Israel or France, despite the unwillingness of their leaders to look like American lackeys.
China is the only thermonuclear power committed to a policy of no first use. Its stockpile is much smaller than that of the US or Russia. It has around a hundred ICBMs, which can be kept concealed and on the move. For decades there were only a small number of ICBM silos in China, but satellite images show it is now building more than two hundred in Gansu and Xinjiang provinces. The discovery added to the scaremongering about its nuclear capacities. The Anglophone media is full of stories presenting minor developments as revolutionary advances. Fanciful ‘hypersonic’ technology and outmoded ‘fractional orbital’ systems of dubious utility are described as shocking new weapons. General Mark Milley, current chairman of the Joint Chiefs of Staff, has said Chinese tests of these technologies constitute a ‘Sputnik moment’. They are more likely to be a sign that China is trying to keep up with American advances in ballistic missile defence. The imperative is to ensure that its nuclear forces can survive a first strike. China’s Type 094 nuclear-armed submarines, of which there are now six, are more capable than previous generations. There is some debate over whether they are quiet enough to avoid tracking, but along with more secure ICBMs they may soon constitute a nuclear force which can be relied on to survive a first strike and retaliate. This would mean that the US would no longer have the option of launching an all-out attack on Chinese territory, as it has threatened to do so many times in the past. The most dangerous moment of the Cold War was in the early 1960s, when an aggressive and overwhelmingly dominant nuclear power saw itself in competition with an adversary that didn’t have equivalent nuclear forces. The US and China may be approaching a similar point.
There is a widespread belief that nuclear weapons can be thanked for the fact that there has been no total war between major powers since 1945. The most prominent exponent of this view, Robert Jervis of Columbia University’s Institute of War and Peace Studies, argued that such wars can no longer occur if statesmen are rational. In The Myth of the Nuclear Revolution, Keir Lieber and Daryl Press suggest that these grand claims are mistaken. There have indeed been no large-scale wars, but in all other respects international politics resemble the pre-nuclear age. States with nuclear weapons don’t act as if they are immune from external attack, and they still engage in reckless expeditionary wars and sometimes scuffles with one another. The main argument Lieber and Press put forward is that this is not just a legacy of old behaviour but a consequence of the nature of nuclear weapons.
Nuclear weapons are too powerful for most conceivable scenarios: in Edward Luttwak’s unimproved-on phrase, they ‘exceed the culminating point of military utility’. The only benefit they have had for the states that possess them is protection against all-out conquest of the homeland. But conventional armies are still required. Lieber and Press show that stalemate between nuclear powers doesn’t happen automatically. It takes effort to build and maintain a nuclear force that can survive a first strike and retaliate. Once states have reached stalemate with second-strike forces – the Soviet Union in the 1960s, China now – they must monitor their adversaries for signs of technological breakthrough that might allow for an attack that would disarm them.
Developments in weapons technology have made it harder to secure nuclear forces. ICBMs have become easier to find and silos easier to destroy. Ballistic missiles – especially submarine-launched ballistic missiles – are much more accurate than they were at the end of the Cold War, and new ‘super-fuses’ allow for greater control over detonation of their warheads. Lieber and Press think a US submarine-launched strike could have a greater than 96 per cent destruction rate against ICBM silos. They hold the conventional American view of the US as a relatively benign power, but the implications of their argument are clear. How confident can we be that the seventy-year gap isn’t just a lull?
The Myth of the Nuclear Revolution contains one glaring omission. Since the detonation of the first thermonuclear bomb there has been much discussion of the climatic effects of nuclear war. The main fear is of ‘nuclear winter’, the notion that firestorms would emit enough black carbon into the stratosphere to cause global temperature drops, mass crop failure and famine. The hypothesis has been dismissed in the past, but American researchers – Alan Robock, Brian Toon and others at the National Centre for Atmospheric Research in Colorado – have persisted with it. The hypothesis is contested, but the projected devastation is so severe that it would be unwise to dismiss it. The risk of climate catastrophe, as well as perhaps hundreds of millions of immediate deaths, must be accounted for in any totting up of the dangers of nuclear weapons. Yet Lieber and Press don’t mention it at all.
In March, the National Intelligence Council delivered its latest ‘global trends’ briefing to President Biden. It included the judgment that the use of nuclear weapons is ‘more likely in this competitive geopolitical environment’. Efforts to find a role for nuclear weapons in conflict have so far fallen at the feet of Luttwak’s maxim. But intentional use is not the only danger. Nuclear strategists systematically underestimate the chances of nuclear accident: it has no place in the logic of strategy. But there have been too many close calls for accidental use to be discounted. The stakes may be anthropogenic extirpation.
Lieber and Press argue that nuclear weapons ‘have made the world a better place’ and that abolishing them would lead to more conventional wars. But the assumption that nuclear weapons will indefinitely prevent large wars rests on unjustified optimism. The stronger argument against abolition is practical. Nuclear weapons can be renounced but nuclear capability can’t: our energy needs won’t allow it. And once you have that capability, the silos can always be refilled. When the only rule is the rule of force, agreements between states are always provisional. Solutions to these problems have been proposed. New treaties, such as the Treaty on the Prohibition of Nuclear Weapons, which many non-nuclear states have signed, are one approach. A reworking of the IAEA or the placement of fuel cycle facilities into international control are another. But such proposals rarely get far. Instead, the stockpiles are growing.