Nuclear Near Misses
A couple of years ago I spotted an interesting tweet from Tim Harford, pointing me to an article in the Bulletin of the Atomic Scientists that told the extraordinary story of a group of missilemen on Okinawa who were ordered to launch their Minuteman nuclear missiles during the Cuban missile crisis. The commanding officer decided not to follow the order, but it was close, and there were dissenting voices: at one stage the officer ordered two of his men to shoot a lieutenant if he went ahead with his intention to launch his missiles.
I wondered, what would it feel like to be the officer who decided to ignore the order, and who therefore saved the world from nuclear destruction? And, just as interestingly, what would it feel like to be an officer who tried to launch the missiles and was stopped?
So I did some more research. It turned out that there were, and still are, elaborate procedures set up in all nuclear nations’ armed forces to prevent an accidental nuclear war, and by and large these have worked. But the last line of defence when the machines and the procedures screw up is human common sense. I came across nine near accidental launches during the Cold War relating to humans overriding messages from the system to launch a nuclear war. All of these were covered up initially. There must have been more that have remained secret.
So what were these?
Unsurprisingly, there were a few during the Cuban missile crisis. At the height of the crisis, in October 1962, operators at the Moorestown, New Jersey radar station reported a nuclear attack from Cuba under way. Calculations showed the missile would land at Tampa in Florida. The system was running a test cassette, at the same time as a satellite was appearing over the horizon in the direction of Cuba. The system became confused and reported a missile had been fired. Strategic Air Command and the National Military Command Center at the Pentagon were informed. A few minutes later, the mistake was spotted when other radar stations confirmed that the ‘missile’ was a friendly satellite.
The US had announced a blockade of Cuba, which was enforced by the US Navy. There were Russian submarines in the area, armed with nuclear torpedoes which could take out a fleet of US ships and produce a radioactive cloud if fired. One of these submarines, the B-59, was being hunted by American destroyers. The captain, Savitsky, believing war had already started, wanted to launch a nuclear torpedo. The political officer, Maslennikov, agreed. Fortunately the submarine also happened to carry the commander of the submarine flotilla, Captain Arkhipov. All three officers had to agree before the nuclear torpedo could be launched. Arkhipov disagreed and nothing was fired.
A sentry was patrolling a fence at an air force base in Duluth, when he saw something breaking in. He raised the alarm, fearing that saboteurs were infiltrating the airfield. The wrong alarm was sounded, announcing that nuclear war had started. Warplanes armed with nuclear weapons lined up on the runway. The base commander realized what had happened, and drove his car on to the runway to stop the aircraft taking off. It turned out a bear had tried to climb over the fence.
In November 1979, Zbigniew Brzezinski, the National Security Adviser, was woken at 3 a.m. and told that 220 missiles had been launched against the US. Brzezinski decided to wait for confirmation before waking the President and advising him to launch a nuclear attack. Thirty seconds later he got it: 2,200 missiles were on their way. President Carter only had between three and seven minutes to respond. Brzezinski picked up the phone to the President when he received a third phone call. Some radar stations had not picked up any missiles at all. It turned out to be a false alarm: someone was running a test on NORAD’s early warning computer system at Cheyenne Mountain in Wyoming.
The same thing happened in 1980.
In 1983 tensions rose to a level not seen since 1962. The Americans were jumpy and the Russians even more so: they were sure that President Reagan intended to launch a surprise decapitation attack against them. So, if the commander of Serphukov-15 early warning centre near Moscow had alerted the Kremlin as soon as he saw signs of an American attack on his radar on 27 September 1983, it is likely that Andropov would have pushed the Soviet nuclear button. But the duty officer, Colonel Petrov, had been directly involved in the recent upgrade of the computer system, and he didn’t trust it. So he turned it off and turned it on again. The missiles were still coming. So he turned the computer off again. By this time, the missiles should have shown up on other radar systems and they hadn’t. So Petrov informed his commanders of the whole situation and nothing was done. The cause of the false radar reading turned out to be rays reflecting off high-altitude clouds.
Two months later, in November 1983, NATO launched a military exercise known as Able Archer 83. In this exercise NATO simulated a nuclear signalling response to a Warsaw Pact invasion of West Germany. ‘Nuclear signalling’ meant limited nuclear missile launches. For the first time, NATO made actual signals from SACEUR, the European headquarters, to its various nuclear commands, and they used new codes. The Soviets, run by ex-KGB officer Andropov, believed that if NATO launched a pre-emptive strike, they would do it under cover of an exercise like Able Archer. So when Able Archer began the Soviets prepared their missiles and bombers. General Peroots was deputy chief of staff for intelligence at the US Air Force HQ at Ramstein in West Germany and he saw the Soviet activity. He chose to do nothing. Lucky for us.
These incidents are what we know. We don’t know what we don’t know.
Which is a little scary.
Nuclear Near Misses