14 September 2013
Nuclear Weapons: an accident waiting to happen
By Eric Schlosser
The Guardian


In July 1956, a plane crashed in Suffolk, nearly detonating an atomic bomb. In January 1987, an RAF truck carrying hydrogen bombs skidded off a road in Wiltshire. Other near-misses remain top secret. Who is really at risk from Britain's nuclear weapons?

'Secrecy surrounding British nuclear weapon accidents has exceeded even that of the US.' Illustration: Noma Bar

In October of 1983, Mick Jones had just left the Clash, Roger Moore was still playing James Bond, Ronald Reagan was in his third year as president of the US, Margaret Thatcher had recently gained a second term as prime minister – and the cold war had entered its most dangerous phase since the Cuban missile crisis. An all-out nuclear exchange between the US, the Soviet Union and their allies seemed possible. That spring, Reagan had called the Soviets "the focus of evil in the modern world… an evil empire", and his administration was conducting an unprecedented military buildup during peacetime. Yuri Andropov, the terminally ill, deeply paranoid leader of the Soviet Union, thought the US might be planning a surprise attack. In September, Korean Airlines Flight 007, a Boeing 747 carrying almost 300 passengers, had been shot down after straying into Soviet airspace. The US was about to deploy two new nuclear weapon systems: ground-launched cruise missiles in the UK and Pershing II missiles in West Germany. And on 22 October, almost a quarter of a million people marched in Hyde Park to protest against those deployments, the largest anti-nuclear demonstration in British history.

Amid that ominous, apocalyptic mood, the Reagan administration decided to build a new submarine-launched ballistic missile. The Trident D5 would be the most accurate missile ever carried by a sub, capable of sending eight nuclear warheads halfway across the world to destroy "hard targets": Soviet missile silos and leadership bunkers. Unlike previous submarine-launched missiles, the Trident D5 wasn't designed solely as a retaliatory weapon, to be used after the US had been hit by a nuclear attack. The new missile could be launched against the Soviet Union as part of an American first strike.

Thirty years later, the cold war is a distant memory, the Soviet Union is gone, Reagan and Thatcher are gone – and a British Trident submarine is still continuously at sea, day and night, waiting for the order to fire its D5 missiles and dozens of nuclear warheads. Over the next few years, Britain will have to decide whether to replace its four ageing Trident submarines. David Cameron wants to build four new subs, at a cost of about £25bn, so that one will always be at sea, safe from attack and prepared to launch. The Labour party seems to endorse that policy: its shadow defence secretary, Jim Murphy, recently expressed support for "a continuous at sea deterrent". Though the Liberal Democrats have criticised Cameron's position, their preference hardly seems radical: building one, or perhaps two, fewer submarines in order to save money. The three main parties are broadly agreed on the need for nuclear weapons. The strongest opposition to Trident comes from politicians in Scotland, where the submarines are based. Alex Salmond, head of the SNP, has promised that if Scotland gains independence next year, its new constitution will include a ban on all nuclear weapons. This could be disastrous for the UK's nuclear deterrent: building a new submarine base and weapon storage facilities in England would take many years and cost tens of billions.

The public debate about Britain's Trident submarines and their missiles has focused mainly on the long-term costs and economic benefits of replacing them, the number of jobs that might be gained or lost, the necessity of round-the-clock patrols. Some fundamental questions have been largely absent from the discussion. How would this cold war missile, due to remain in service for another 30 years, actually be used in a 21st-century conflict? What targets would it destroy and in what circumstances? Whom is it supposed to kill? Finally, and perhaps most importantly, do the British people face a greater chance of being harmed by their own nuclear weapons, through an accident or a mistake, than by a surprise attack?

In Grand Forks, North Dakota, in September 1980, a B-52 bomber loaded with 12 hydrogen bombs and nuclear warheads caught fire. It burned for two hours, fed by a fuel pump. Only gale force winds blowing the flames away, and a fireman climbing on to the burning plane to shut off the power, averted disaster.

"The credibility of the UK's deterrent is crucial," the government's Trident Alternatives Review declared in July this year. On a single page of the report, the words "credible" or "credibility" are used seven times to describe the British nuclear strategy of "minimum deterrence". But the report never says who is to be deterred, and what the UK would actually do with its nuclear weapons should that deterrence fail. The 2010 Strategic Defence And Security Review mentions the risk of "nuclear terrorism" and "the possibility that a major direct nuclear threat to the UK might re-emerge". The implication is that North Korea, China, Pakistan, Iran or Russia might one day seek to wipe out London, despite the sizable investments some of those countries have made in the city. But the security review doesn't explain in what circumstances Britain's Trident missiles would ever be launched, citing the need to remain "deliberately ambiguous".

Since the dawn of the atomic era, the sort of ambiguity deemed essential to confuse potential enemies has enabled government officials to avoid domestic oversight and accountability. A few weeks ago, parliament voted to prevent Cameron from authorising a relatively small British attack on Syria, yet a British prime minister can authorise a nuclear attack that might kill millions without public approval or parliamentary debate. The intense secrecy that surrounds nuclear war planning has hidden not only the devastation that would be inflicted upon targets, but also the grave dangers posed by a nation's own weapons, the risk of serious accidents. And whenever those nuclear secrets are revealed, however incompletely, plans that were thought "credible" by those in power usually seem incredible, almost unbelievable, to everyone else.

Nuclear weapons are the most lethal machines ever invented, but the deterrence they provide is something intangible. "The central objective of a deterrent weapon system… is psychological," a classified Pentagon report once explained. "The mission is persuasion." The destruction of Hiroshima and Nagasaki showed what a single atomic bomb could do to a city. But widespread revulsion at the civilian casualties prompted the US to explore how nuclear weapons might be used against traditional military targets.

In 1946, the US conducted its first postwar tests of the atomic bomb. One of these tests sought to discover the effect of a nuclear blast on a fleet of warships. The results were discouraging. Of the 88 ships moored near the point of detonation, in the Bikini atoll, only five sank. The Evaluation Of The Atomic Bomb As A Military Weapon, a top-secret report sent to Harry Truman, concluded that "ships at sea" and "bodies of troops" were poor targets. "The bomb is pre-eminently a weapon for use against human life and activities in large urban and industrial areas," the report argued. Such weapons were useful, most of all, for killing and terrorising civilians. According to the report, some of the best targets were "cities of especial sentimental significance".

America's first nuclear war plan, adopted in 1948 and codenamed Halfmoon, called for 50 atomic bombs to be dropped on the Soviet Union. The number was subsequently increased to 133, aimed at 70 cities. Leningrad was to be hit by seven bombs, Moscow by eight. There seemed no alternative to the threat of mass slaughter. This US strategy was called "the nation-killing concept".

At congressional hearings in October 1949, the US had its most publicised, high-level debate about the ethics of such nuclear targeting. A group of admirals strongly condemned the war plan the US air force intended to use against the Soviets. "I don't believe in mass killings of noncombatants," Admiral Arthur W Radford testified, while Rear Admiral Ralph A Ofstie, who had toured the burnt-out cities of Japan, described the atomic blitz as "random mass slaughter of men, women and children", and said the whole idea was "ruthless and barbaric", contrary to American values.

By the mid-1950s, the American war plan had shifted from hitting "countervalue" targets (cities) to destroying "counterforce" targets (military facilities). The invention of the hydrogen bomb had created nuclear weapons hundreds of times more powerful than the bomb that destroyed Hiroshima. The Soviet Union now had its own nuclear weapons, and destroying them became the air force's principal goal.

In December 1960, the US approved its first Single Integrated Operational Plan (Siop), which specified the timing and targeting of attacks to be conducted by US forces and UK Bomber Command. This plan, in one form or another, would remain in effect for more than three decades. Most of Siop's details remain classified, but memos written during John F Kennedy's administration, at the height of the Berlin crisis in 1961, give a sense of how much destruction the Siop would unleash. It featured 3,729 targets that would be struck by 3,423 nuclear weapons. The targets were located in the Soviet Union, China, North Korea and eastern Europe. About 80% were military targets, the rest civilian. Of the "urban-industrial complexes scheduled for destruction", 295 were in the Soviet Union and 78 in China.

The Siop's damage and casualty estimates were conservative. They were based solely on blast effects, and excluded the harm that might be caused by thermal radiation, fires or radioactive fallout, which was difficult to calculate. Within three days of an initial attack by the US and UK, the full force of the Siop would kill about 54% of the Soviet Union's population and about 16% of China's population: roughly 220 million people. Millions more would subsequently die from burns, radiation poisoning and exposure.

Britain's nuclear war plans were never as conflicted as those of the US. Within weeks of Hiroshima's destruction, Clement Attlee succinctly expressed what would become the British philosophy: "The answer to an atomic bomb on London is an atomic bomb on another great city." Britain lacked the means to build enough nuclear weapons to threaten thousands of military targets in the Soviet Union. And the British population wasn't widely dispersed across a large continent. In 1955, a secret report by William Strath, an official at the Central War Plans Secretariat, concluded that if 10 hydrogen bombs were detonated along the west coast of the UK, the ensuing firestorms and radioactive fallout would immediately kill or wound about one-third of the British population. Most of the nation's farmland would be unusable for two months, and drinking water would be contaminated. Even if Britain somehow managed to destroy most of the Soviet Union's nuclear arsenal, a handful of Soviet weapons could cause much of British society to collapse.

In 1958, Bomber Command's emergency war plan called for the destruction of 44 Soviet cities. Such an attack would kill about 38 million people. One hydrogen bomb would be dropped on the centre of each city, but Moscow would be hit by four and Leningrad by two. Had Britain gone to war alongside the US in the early 1960s, Bomber Command would have been asked to destroy an additional 25 Soviet cities. As air defences improved in the Soviet Union, the number of urban areas that Britain planned to destroy unilaterally was reduced. By the late 1960s, the missiles carried by Polaris submarines served as the British strategic deterrent, and they were aimed at fewer than a dozen Soviet cities. Until the end of the cold war, the complete destruction of the Soviet Union's capital – known as the "Moscow criterion" – was the UK's main objective.

The Joint Intelligence Committee assumed that Soviet war plans would be even more brutal. According to its report on "Probable Nuclear Targets In The United Kingdom", London would be hit by eight hydrogen bombs and two atomic bombs. Edinburgh would be hit by two of each, Glasgow by four hydrogen bombs and one atomic bomb, while the British submarine bases in Scotland would be hit by four hydrogen bombs and four atomic bombs. All in all, the JIC expected that during a war with the Soviet Union, the UK would be struck by about 300 nuclear weapons.

Faslane naval base on the Clyde. Scotland is home to both Britain’s Trident submarines and the strongest opposition to them, with SNP leader Alex Salmond vowing to ban them if Scotland gains independence. This could be disastrous for the UK’s nuclear deterrent: building a new base would take years and cost billions. Photograph: Murdo Macleod

Strict official secrecy has enabled British and American war planners to choose their targets without public scrutiny. It has also facilitated the management of public opinion about nuclear weapons. The Strath report was suppressed, and Winston Churchill ordered the BBC not to broadcast news about the hydrogen bomb that might scare people. From Attlee's decision to build a British atomic bomb to James Callaghan's negotiations to obtain Trident missiles from the US, British nuclear policy has been carried out without much parliamentary oversight. "All of the key nuclear decisions were taken by a small number of very senior government ministers, meeting in informal ad hoc committees," the historians John Baylis and Kristan Stoddart have noted. "These decisions had been kept secret from the main cabinets of the day."

In the US, the design specifications of nuclear weapons don't even have to be classified. According to the Atomic Energy Act of 1946, they are "born secret": classified as soon as they exist. Maintaining secrecy about nuclear weapon designs may seem like common sense, but the secrecy justified by a need to prevent foreign espionage has routinely been used instead to hide safety problems, cover up nuclear weapon accidents and shield defence bureaucracies from embarrassment.

The Pentagon's official list of "broken arrows" – mishaps with nuclear weapons that might threaten the public – mentions 32 accidents. Yet a 1970 study by one of America's nuclear weapon laboratories, obtained through the Freedom of Information Act, stated that at least 1,200 weapons were involved in accidents between 1950 and 1968. Most of these accidents were trivial, but a number of serious ones were somehow omitted from the Pentagon's list. Moreover, the risk of accidental nuclear detonations was not fully understood by American weapon designers until the late 1960s, and it proved far greater than expected. A plane crash, a fire, a missile explosion, lightning, human error, even dropping a weapon from an aircraft parked on a runway were found to be potential causes of a nuclear explosion.

Two of the more dangerous accidents occurred in one month. On 15 September 1980, one of the engines on a B-52 bomber caught fire at Grand Forks air force base in North Dakota. The plane was carrying four hydrogen bombs and eight short-range missiles with nuclear warheads. A strong wind kept the flames away from the weapons, and a fireman climbed into the burning plane, put out the fire, and averted a disaster. Three days later a technician dropped a tool in the silo of a Titan II intercontinental ballistic missile near Damascus, Arkansas. The tool hit the bottom of the silo, bounced, struck the side of the missile, pierced the skin and caused a fuel leak. The Titan II was carrying the most powerful nuclear warhead ever built by the US. Despite a heroic effort to save the missile, it exploded – but the warhead didn't detonate. Both states could have been destroyed.

The safety problems with American nuclear weapons were kept secret until the conclusion of the cold war. A study later sponsored by the US Congress gave a "safety grade" to each type of nuclear weapon in the nation's arsenal. The grades were based on the potential risk of accidental detonation or plutonium scattering. Three weapons received an A. Seven received a B. Two received a C-plus. Four a C. Two a C-minus. And 12 received a D, the lowest grade.

The safety issues with American nuclear weapons had implications far beyond US borders. Nato forces relied on many of them. For years, the number of American nuclear weapons deployed in Britain exceeded the number of British ones. According to the historian John Simpson, in 1959 the RAF had 71 British atomic bombs and 168 American ones. In the years that followed, the nuclear weapons manufactured in Britain became remarkably similar to those made in the US, thanks to the US-UK Mutual Defence Agreement. The design of "Red Snow", the nuclear component at the heart of Britain's first widely deployed hydrogen bomb, was based on that of the American Mark 28 bomb. In 1961, Harold Macmillan was told British weapon development was "confined almost entirely to copying US designs".

The secrecy surrounding British nuclear weapon accidents has exceeded even that employed to hide US ones. In July 1992, a report by Sir Ronald Oxburgh, the chief scientific adviser at the Ministry of Defence, claimed that 19 accidents had occurred with British weapons between 1960 and 1991. The Oxburgh report suggested that none of the accidents was particularly worrisome, and according to the Ministry of Defence, the UK has no record of any accidents involving US weapons on British soil. However, while researching safety problems with American nuclear weapons, I came across information about serious British accidents that weren't mentioned by Oxburgh.

Two of the accidents occurred at RAF Lakenheath. On 27 July 1956, an American B-47 bomber was practising touch-and-go landings. The plane veered off the runway and slammed into a storage igloo containing Mark 6 atomic bombs. An American officer who witnessed the accident described what happened next in a classified telegram: "The B-47 tore apart the igloo and knocked about 3 Mark Sixes. A/C [aircraft] then exploded showering burning fuel overall. Crew perished. Most of A/C wreckage pivoted on igloo and came to rest with A/C nose just beyond igloo bank which kept main fuel fire outside smashed igloo. Preliminary exam by bomb disposal officers says a miracle that one Mark Six with exposed detonators sheared didn't go. Firefighters extinguished fire around Mark Sixes fast."

The nuclear cores of the weapons were stored in a different igloo. If the B-47 had hit that igloo instead, a large cloud of plutonium could have floated across the Suffolk countryside. Plutonium dust can be lethal when inhaled. Once dispersed, it is extremely difficult to clean up, and it remains dangerous for about 24,000 years.

Another bad accident occurred at Lakenheath on 16 January 1961. The under-wing fuel tanks of a US F-100D fighter were mistakenly jettisoned when the pilot started the engines. The fuel tanks hit the runway and ruptured, some fuel ignited, and a Mark 28 hydrogen bomb mounted beneath the plane was engulfed in flames. Firefighters managed to extinguish the fire before the bomb was badly damaged. A flaw in the wiring of Mark 28 hydrogen bombs, it was later discovered, could allow excessive heat to circumvent the weapon's safety mechanisms and cause a nuclear detonation.

The safety of the Trident D5 missile has long been a source of concern. In December 1990, the Panel on Nuclear Weapons Safety, a group of eminent physicists appointed by the US Congress, warned that its unusual design posed significant risks. To save space, the multiple warheads weren't mounted on top of the missile; they surrounded its third-stage rocket engine. And the "high-energy" class 1.1 propellant used in that rocket engine was far more likely than other propellants to explode in an accident. "The safety issue of concern here," the panel found, "is whether an accident during handling of an operational missile – viz, transporting and loading – might detonate the propellant which in turn could cause the HE [high explosives] in the warhead to detonate, leading to dispersal of plutonium, or even the initiation of a nuclear yield."

The decision to use the more energetic rocket fuel and the more unstable high explosive was made during the early 1980s, to increase the range of the Trident D5 missile and decrease the weight of its warheads. The first British Trident submarine went on patrol four years after these safety risks were discovered, and the Trident D5 missile is supposed to remain in service until 2042. The risk of explosion and plutonium dispersal is greatest when the warheads are being loaded on to the sub, unloaded from the sub, or transported by road between Scotland and the Atomic Weapons Establishment in Aldermaston, Berkshire.

The worst accident to occur during the handling of nuclear weapons in the UK, according to the Oxburgh report, came on 7 January 1987, when an RAF truck swerved to avoid another vehicle on an icy Wiltshire road. The truck, which was carrying two hydrogen bombs, went off the road and skidded on to its side. An RAF truck behind it, carrying another two bombs, went off the road, too. None of the weapons was damaged. But a recently declassified US document contains details about another serious incident in the UK, and others like it have no doubt occurred. On 17 August 1962, at an undisclosed RAF base somewhere in England, two retrorockets on a Thor missile suddenly fired while it was undergoing a routine check. The launch pad was evacuated, and when workers returned, they found that the missile's nose cone, containing the warhead, had not been dislodged. The warhead was about 60 times more powerful than the bomb that destroyed Hiroshima. "The cause of the incident," the report noted, "was failure to follow prescribed safety rules."

'A US study gave a "safety grade" to every nuclear weapon in the nation's arsenal. Twelve received a D, the lowest grade.'

The US has vastly reduced the number of its strategic nuclear weapons, by almost 90% since the Reagan era. The Siop has been replaced by another set of targets, known as the Operations Plan (OPlan) 8010, most likely designed to use nuclear weapons against Russia, China, North Korea, Syria and Iran. "Adaptive planning" allows targets in other countries to be chosen at the last minute. In June, the Obama administration publicly released a new Nuclear Weapons Employment Strategy. It claims that the US would use nuclear weapons only against military targets and "will not intentionally target civilian populations or civilian objects". One of the faults with such a counterforce strategy is that it can make nuclear weapons seem like legitimate weapons for use in a military campaign. Even the most accurate nuclear strike will cause collateral damage and lethal radioactive fallout.

As the late Michael Quinlan, a former permanent secretary at the Ministry of Defence, observed after the cold war, British war plans seemed to acquire a "very general 'to-whom-it-may-concern' character". The 2010 Strategic Defence and Security Review confirmed his opinion: "No state currently has both the intent and the capability to threaten the independence or integrity of the UK." The review called Britain's policy of minimum nuclear deterrence an "ultimate insurance policy" in "an age of uncertainty", without specifying how or against whom. And there was no mention that such a policy has traditionally emphasised the destruction of cities, not military forces.

It may indeed be necessary to threaten millions with annihilation in order to deter a nuclear attack on the UK. That argument, however, should be based on more than the trustworthiness of elected officials. Under the 1962 Nassau Agreement, which first granted the UK the use of US submarine-based missiles, those weapons were to be employed solely on behalf of Nato, unless the British government felt that its "supreme national interests are at stake". The meaning of that phrase has never been explained, and no prime minister has plausibly described a situation in which the UK would have to use its nuclear weapons unilaterally, without any support from the US or the rest of Nato. An attack that wiped out London, for example, would also kill one or two hundred thousand Americans. Over the past decade, the US has waged two wars, at a cost of almost $2tn, to avenge the death of many fewer.

The only word used more often than "credible" in official British statements on nuclear deterrence is "independent". But the Trident D5 missiles on British submarines are not, specifically, owned by the UK. They are supplied from a common Anglo-American pool, returned to the US for refurbishment and replaced with other missiles. Almost half a century ago, Harold Wilson raised questions about the "so-called independent, so-called British, so-called deterrent". Those questions have never been adequately answered.

Maintaining deliberate ambiguity about nuclear war plans is a good way for government officials to stifle meaningful debate. And while other pressing national issues may dominate the news, this one could not be more important. Britain has never had a full, vigorous debate about its nuclear weapons, based upon the facts. As other countries seek weapons of mass destruction, the stakes couldn't be higher. Will new Trident submarines protect the UK or jeopardise its future? I hope the British people get to decide.

Command And Contol, by Eric Schlosser, is published by Allen Lane, priced £25. To order a copy for £20, with free UK p&p, go to theguardian.com/bookshop.

Global Network