At the height of the Cold War, with alarms screaming, one Soviet officer stopped to think. It cost him his career. It may also have saved the world.

The morning of the 26th September 1983 had started quietly at Serpukhov-15, a Soviet military townlet about 70km south of Moscow. Despite its small size, the base was important, for it was home to the primary control centre for “Oko,” a satellite-based missile launch detection system.

If the USSR was subject to nuclear assault then the men and women monitoring Oko would be the first to know. If the Soviets were to have any hope of retaliating then a swift, early warning from them was critical.

It was a heavy burden to bear, one made worse by the fact that tensions between the superpowers that summer were higher than they had been since the Cuban Missile Crisis several decades before. Indeed at the very highest levels the Soviets had become convinced that a pre-emptive nuclear strike by America was not just likely, but practically inevitable.

And then suddenly that September morning Oko erupted into life. Sirens blared and red lights blazed.

Missile launch detected.

Instantly the men and women of Serpukhov-15 sprang into action. Calculations were made and systems checked for failure, all of them came back clear. More and more data was coming in now and all of it seemed to paint a terrible picture. A single, deadly, Minuteman III ICBM had been launched from Malmstrom Air Force Base in Montana, targeted on the Soviet Union.

The staff of the control centre looked to the man who sat at their centre — forty-four year old Lieutenant-Colonel Stanislav Petrov.

“Everyone jumped from their seats, looking at me.” He later recounted. “What could I do? There was an operations procedure that I had written myself. We did what we had to do. We checked the operation of all systems — on 30 levels, one after another. Reports kept coming in: All is correct.”

The rules were clear. As the man in charge it was Petrov’s responsibility to report the missile launch. The data was irrefutable.

The trouble, Petrov realised as he began to automatically run through the notification procedure, was that deep down, for reasons he couldn’t yet explain, this just didn’t feel right.

As the West slept on, oblivious to the mounting danger, Lieutenant Colonel Petrov paused.

"Call me back"

Although it may not have felt like it to those living through it, throughout the bulk of the Cold War the risk of an actual nuclear launch was relatively low. In part, this was because both sides of the conflict were critically aware that the risk of an accidental launch — and the likely terrible chain of events that would precipitate — was something that needed to be carefully managed. This meant careful consideration of how to approach control of both nuclear weapons themselves and the decision-making process behind their use.

Finding a balance between speed and effectiveness, however, was a constant balancing act. In our look at the British Space Programme we described, briefly, one of the bizarre situations this lead to in the UK, whereby nuclear missiles on Vulcan bombers stationed in Scotland were allowed to have either fuel or a warhead, but never both.

Both the USA and the USSR faced similar problems, but on a grander scale, and mistakes in both the handling of weapons and — perhaps more critically — the detection of their use by the other side were not unheard of.

On the night of the 9th November 1979 for example, Zbigniew Brzezinski, National Security Advisor to President Jimmy Carter, was awakened by the sound of the phone he kept next to his bed.

Brzezinski answered it. On the other end of the line was Lieutenant Colonel William Odom, his military assistant. Odom calmly informed Brzezinski that NORAD’s early warning systems had detected the launch of 250 missiles from the Soviet Union. In line with his responsibilities, Odom told him, Brzezinski should inform the President.

Adrenaline flowed through Brzezinski’s veins. Could it possibly be true? There’d certainly been no sign of any impending attack from the intelligence services. Pausing to think, he remembered that the time required to initiate a retaliatory launch on the orders of the President was three to seven minutes.

He ordered Odom to get the B-52s of Strategic Air Command in the air, but to do nothing else.

“Confirm it.” He told Odom. “And call me back.”

Brzezinski quietly got out of bed and headed to his office. Looking over at his wife, he decided to leave her sleeping. If Odom was correct, he reasoned, then in half an hour they were both as good as dead anyway. Let her sleep.

The news didn’t get any better when Odom called again a minute later. NORAD, he explained, was now reporting 2,200 missiles in the air. It was an all-out attack.

Resigned to destruction, Brzezinski put down the phone. He paused for a minute to gather his thoughts, and then prepared to ring the President.

As he was about to do so the phone suddenly rang one final time. It was Odom yet again.

None of the other warning systems had picked up a launch, Odom told him. There was some kind of a problem at NORAD. Everyone was standing down. The whole thing was a false alarm.

It would later emerge that a faulty chip in NORAD’s computer systems had resulted in test data simulating an attack accidentally being routed to its live monitoring displays. Unaware of the issue, the controllers had thought a real nuclear assault was underway before the mistake had been caught.

More launches

Whether Petrov was aware of NORAD’s own close call (which had eventually leaked to the press) or not is unknown. He was critically aware though that every moment he delayed in issuing a Launch Warning increased the risk not just to his country, but to his own career. The pressure on him to pick up the phone and alert his superiors was enormous.

It just didn’t feel right though. Oko was still relatively new and Petrov had seen more than a few bugs emerge during its installation. Who was to say there weren’t more? Whilst it was only one missile, there was just enough leeway in his own orders to allow him to hold off a little longer without risking punishment.

He demanded his staff check everything again, but no sooner had he begun issuing orders when more alarms started blaring.

More launches. Five now in total.

This was serious. Serious enough to trigger automatic alerts to those higher up the chain of command that something was happening.

Petrov was out of time. It was no longer a case of waiting to call his superiors. They were about to call him. And when they did, whatever his own doubts, according to the letter of his orders, he had to issue an official Launch Warning.

As if on cue, the phone on his desk rang.

A gap in the network

Under normal circumstances (or at least what passed for normal in the Cold War) Petrov issuing an official Launch Warning wouldn’t necessarily have been a disaster. Just as aware of the risks of a false launch detection as the Americans, the Soviets had originally intended to have a system of checks and balances that would produce an experience for decision-makers similar to that of Brzezinski in 1979.

To achieve this, in the early 1970s the Soviet Union had embarked on an ambitious plan to update and improve its detection systems for nuclear attack.

Originally, all Soviet early warning had been radar based, with a focus on Europe and the known likely approaches from the US. Now the Soviets envisioned a three stage warning system, which would allow long, medium and short range detection, and would be better set up to deal with the ICBM threat.

Short range detection would happen via the existing radar system, upgraded and expanded. The satellite-based Oko system would provide long range detection by picking up ICBM launches thermally from space. In between the two would sit a new network of over-horizon (OH) radar stations able to overcome the curvature of the Earth and fill the gap between them.

By 1983, the existing radar network had been upgraded and Oko had finally been brought online. Critically though, the OH radar network had not been built.

This meant that far from having a flexible, nuanced system, the Soviets only really had two possible responses when a Launch Warning was issued based on Oko’s data. They could either wait until the apparently inbound missiles were close enough to be picked up by the short range radar stations, by which point they would be mere minutes away from their targets, or fire based on Oko’s long range warning alone.

The problem was that in September 1983, unbeknownst to the West, the Soviet leadership felt so threatened that it was leaning heavily towards the latter approach.

A nation in decline

Understanding why this situation existed (and also how the West missed it) means going back to the beginning of the decade. The 1980 US Presidential election took place against a backdrop of national uncertainty. Facing economic troubles at home and still shaken by the Vietnam War and Iran hostage crisis abroad, US voters responded well to Ronald Reagan’s campaign rhetoric which focused on low taxes and small government at home, and a more robust approach to enemies such as the Soviet Union abroad.

Reagan’s anti-Soviet campaigning did not go un-noticed by the Kremlin, who grew increasingly concerned as the election drew near. Reagan seemed to be threatening to usher in a new period of antagonism and anti-Sovietism and this was something the Soviet Union could ill afford. For though the USSR still presented itself as a force to be reckoned with to the outside world, behind the scenes its leadership were fully aware that they were slipping further and further behind the West — militarily, economically and technologically.

This decline had been obvious to those in positions of power for a number of years. In time, Gorbachev would come to refer to it as the “Era of Stagnation,” which would only end with his own ascension to the Soviet Presidency in 1985.

The effects of this stagnation were felt across all areas of the economy and society, not least by the Soviet military. Since 1977, Soviet investment in its nuclear missile programmes had effectively remained static, if not declined, and the increasing Soviet commitment to detente had as much been a quiet acceptance of the fact that it could no longer compete with the US directly as anything else. Meanwhile, a disastrous attempt to reassert order in Afghanistan through conventional warfare not only demonstrated the failing power of the Soviet army, but also provoked international outrage.

By 1981 the Soviet Union was a world power in serious decline, torn between the need to present a strong image both abroad and at home, and a need to reform and cut back spending if it was to survive.

Poking a wounded bear

Into this environment the Reagan administration exploded like a hand grenade. If the Soviets had hoped they would moderate their campaign stance on election they were soon disappointed.

Under Reagan, spending on defence increased to record levels for peacetime, whilst the new President’s own rhetoric continued to harden. Meanwhile, an explicit policy of undermining the Soviet Union through covert political-psychological acts was also pursued.

Flights aimed at spooking the USSR and probing its radar, something that had fallen away in the last decade, began again. From 1981 onwards the US Navy also began exercising in areas explicitly aimed at reminding the Soviets that they did not control the ocean.

Publicly the Soviet Union reacted to these provocations with alarm, threats and manoeuvres of their own. To the White House, this was seen largely as bluster to which both they and the Pentagon were happy to respond. Both sides, they believed, were simply playing the same game that they had for decades.

What they failed to realise, however, was that behind the scenes the gerontocracy in charge of the stagnating USSR were no longer playing the same game. Increasingly aware of the growing gap between US military power and their own, and feeling increasingly surrounded and probed for weaknesses, the Soviet leadership was starting to get genuinely scared.

Taken in 1975, this photo highlights the ageing nature of Soviet leadership. From left, Mikhail Suslov, Viktor Grishin, Leonid Brezhnev, Yury Andropov, Mikhail Zimyanin, Konstantin Chernenko and Alexei Kosygin.

Operation RYaN

In May 1981, at a major KGB conference in Moscow, ailing Soviet Premier Leonid Brezhnev denounced the aggressive policies of new US President Ronald Reagan. The head of the KGB, Yuri Andropov then took the stage and dropped a political bombshell on those assembled: The Soviet leadership, he announced, was convinced that the Americans were now actively preparing to launch a nuclear war. As a result, he was launching Operation RYaN — a worldwide KGB operation aimed at providing as much warning as possible as to the timing of that attack.

Almost as soon as Andropov had finished talking, orders and instructions were secretly dispatched to KGB station officers at key locations throughout the West. These detailed, at great length, various signs and actions that could be taken to mean that a NATO attack was imminent. All stations were ordered to begin reporting these signs as soon as they occurred.

Psychologically, the phrasing behind the RYaN orders was both interesting and potentially catastrophic. In them, Andropov didn’t ask for evidence that an attack might happen — that was simply taken for granted. Instead he asked for signs as to when.

Soon, keen to demonstrate their ability and efficiency, station chiefs were passing on information they claimed identified the latter, despite the fact that (as many would later admit after the fall of the Soviet Union) the feeling on the ground was that Moscow was misreading the mood of NATO and there was no chance of such an attack actually happening.

That Andropov was so sensitive about an impending attack was not entirely surprising. A hardliner who believed in flexing military muscle from a position of strength, it made complete sense to him that the US would aim to do the same. How closely Brezhnev’s own beliefs matched Andropov’s ultimately didn’t matter, because in November 1982 the Soviet Premier died, and KGB chief Andropov was chosen to take his place.

Warning signs

By 1982, Andropov was already 67 years old and — like Konstantin Chernenko, who would succeed him just 18 months later — already unwell. Both his age and his past as head of the KGB meant that, of all Brezhnev’s potential successors, he was the man most likely to be convinced of the possibility of a US attack.

A veteran of the Second World War, like many of his political generation Andropov was determined never to make the same mistake as Stalin, who had been caught off guard by Hitler’s surprise attack on the Soviet Union in 1941. As head of the KGB, Andropov was also intimately aware of the truth behind that surprise — it wasn’t that the intelligence hadn’t been there, it was that Stalin had refused to accept it. If placed in the same position Andropov was determined to act based on data and not ignore the information in front of him.

Andropov believed he had made this attitude clear to the US both publicly and privately. Unfortunately his public pronouncements were, again, largely taken by most of the West as part of the Cold War game. Increasingly, however, one NATO power was starting to suspect that things were more serious than the US and others believed — the British.

British doubts were based on a rather fortuitous promotion. In 1982, KGB officer Oleg Gordievsky was promoted to Colonel and transferred to the Soviet embassy in London, where he was ordered to take over as KGB Resident — station chief.

Unknown to Moscow though, Gordievsky had already been turned by British intelligence agency MI6.

Gordievsky’s new role opened him up to the full workings of Operation RYaN. It immediately became obvious to both Gordievsky and his handlers that politically the Americans were playing an incredibly dangerous game. Unfortunately, MI6 was unable to persuade the CIA that the Soviets genuinely believed that the US was contemplating a pre-emptive attack. With only a few exceptions, the CIA’s analysts were unwilling to even consider it. America, they insisted, didn’t do “Pearl Harbours.” The whole world knew that!

They asked for direct access to MI6’s source but, worried about compromising such a high-profile asset, the agency refused and for now at least the warning was dismissed.

A most dangerous year

In 1983 the situation went from bad to worse.

First, on the 8th March, in a heated speech before the British Parliament, Reagan referred to the Soviet Union as the “Evil Empire” and described how nuclear war could be seen as an extension of the old battle between good and evil.

Unknown to the US, the speech sent shockwaves through Moscow. Combined with US plans to deploy Pershing II nuclear missiles to Germany, to the Soviets this seemed to be the strongest indication yet that nuclear war was imminent.

Then, later that month, Reagan announced plans for the Strategic Defense Initiative — “Star Wars” — a system which would destroying Soviet missiles before they reached their targets.

In the West, the SDI announcement was greeted with a more than a little cynicism. To the Soviet leadership though, now gripped by a technological inferiority complex, the idea that Star Wars might not work was barely considered. From their perspective in just one month the US had announced a willingness to attack, plans to move missiles so close that the USSR would have little time to respond in an emergency, and long term plans to remove the Soviets’ ability to retaliate at all.

Publicly, Andropov announced that the US were undermining the very principles of deterrence. He did exactly the same privately when ex-US Ambassador Averell Harriman visited the new Soviet Premier off-the-record in June. As Harriman would report back to the White House, not only did Andropov seem genuinely worried about American intentions, but he also insisted that the US might force the Soviets to adopt a principle of Launch on Warning, and in doing so the risk of an accidental war as much as a deliberate one.

A tired-looking Andropov (centre) at the Supreme Soviet in June 1983

Once again, the warning signs were ignored. The Soviets, the US analysts insisted, were still just playing the game. Andropov was posturing in a desperate attempt to get the Pershing II missiles withdrawn. Nothing more, nothing less.

Countdown to war

On 1st September 1983, Korean Air Lines Flight 007 (KAL 007) from New York to Seoul via Anchorage took to the air. Unbeknownst to its crew, from the moment it had departed Alaska the flight had begun to drift off course. By the time it had crossed the Pacific it was almost 300km north of where it was meant to be — right over Soviet Kamchatka. What followed was nothing less than a tragedy.

In March 1983, the US Navy had once again engaged in a series of simulated attacks on Soviet territory aimed at probing their defences. This included overflights of Soviet bases on the disputed Kuril Islands. In the aftermath of these exercises, Soviet high command had dismissed the commanders at those bases for failing to respond with adequate force. Andropov had also quietly changed the Soviet rules of engagement nationwide for unknown aircraft. Paranoid and embarrassed, both globally and locally, the Soviets were determined not let the same thing happen again.

For the crew and passengers of KAL 007, the results of this policy were to prove fatal. Their course deviation took them into the same vicinity and the local Soviet forces reacted with brutal efficiency. Whether the plane was civilian or not didn’t matter. From the General in charge of Sokol Air Base to the pilots themselves, no one stopped to question their orders. KAL 007 was shot down, and 269 people died.

In the aftermath of the disaster neither the US nor Soviet governments covered themselves in glory. The Soviets initially denied their role in the plane’s destruction, then deliberately misled rescuers as to the location of its remains and interfered with the international search and rescue effort.

Meanwhile, sensing an almost unprecedented opportunity for a major propaganda advantage, the US released a wealth of intelligence and recordings making the Soviet Union’s culpability clear. On the 5th September Reagan condemned the flight’s destruction as a “crime against humanity [that] must never be forgotten.”

On the 9th September the Soviets publicly defended their actions and refused to apologise. They protested, farcically, that the plane had been a CIA spy plane. The evidence to the contrary was so overwhelming that the suggestion was treated in the West with the scorn it deserved. This was just Soviet propaganda, the US insisted, just another lie and part of the Cold War game.

The trouble was, as documents released after the collapse of the USSR would later show, once again the West had badly misread the situation. Driven to the point of absolute paranoia about US intentions, such a nadir had been reached in US / Soviet relations that Andropov himself actually believed this was true.

As New York airports began denying Soviet flights permission to land in the city for the UN, in direct contravention of the UN Charter, Andropov and the Soviet leadership felt that there was now no question about what was coming. The Soviet press, acting as a voice for the government, declared Reagan to be a madman and, ominously, compared him to Hitler.

In the US, the comparison was dismissed as tasteless and disgraceful, once again though the meaning beneath the insult was entirely missed. It wasn’t Hitler the dictator that the Soviet government were comparing Reagan to, it was Hitler the surprise invader.

As far as Andropov and the Soviet leadership were concerned, they had been manoeuvred into a corner. The US had manufactured a casus belli, and they were about to declare war.

Petrov makes his decision

All the above may seem like a particularly lengthy diversion from our main story, but it is vital context if one is to understand the enormous consequences of what Petrov did next.

If Lieutenant Colonel Petrov followed his orders, then with five missiles apparently in the air he had to issue a Launch Warning. It didn’t matter whether he believed it or not. It didn’t matter that the Soviet leadership had already decided that the Americans were about to launch a pre-emptive nuclear attack. It didn’t matter that as a result, either literally or figuratively, the Soviet Union was operating at a state of Launch on Warning.

Petrov, like the men who had shot down flight KAL 007 had one job to do. Follow his orders. Accept the data. Issue a Launch Warning.

And, as he sat there amidst the chaos and alarms, with an intercom in one hand and the phone in the other, that’s exactly what Lieutenant Colonel Petrov didn’t do.

Because he’d finally realised what was bugging him.

“When people start a war, they don’t start it with only five missiles,” He told the Washington Post many years later. “You can do little damage with just five missiles.”

Ignore the data. Ignore the alarms. There was no nuclear attack, Lieutenant Colonel Petrov told the anxious, senior officials on the other end of the phone. It was a bug in the system.

“I didn’t want to make a mistake. I made a decision, and that was it.”

Petrov put down the phone, and waited for the world to end.

The aftermath

The world didn’t end, of course. Because Petrov was right.

It would take some time to discover what had happened that morning. Eventually though, Soviet technicians worked it out.

Unwilling to use geostationary satellites as part of their early warning system, the Oko designers had instead opted for a number of satellites in high elliptical orbits instead.

Only seven of the planned nine satellites had been in space and operational at the time of the false alarm, and because of the nature of their orbits only one, Cosmos 1382, had been in position to get line-of-sight on Malmstrom Air Force Base.

The satellites themselves relied on infrared heat detection in order to establish a launch and the system’s designers had naturally been careful to ensure that another source of heat — such as, say, the sun — couldn’t accidentally be picked up instead. The trouble was that in a very specific set of circumstances, with a single satellite at a specific angle, the sun at a specific position in the sky, and the wrong type of cloud at a very specific attitude, it turned out it was entirely possible for one of Oko’s satellites to register a false positive. Multiple false positives, in fact.

Which was exactly what had happened that morning.

Once the flaw in the warning system had been identified, it was quietly fixed. Geostationary satellites were also added, in order to ensure there was always a second position from which to take a reading.

Not that this would help Petrov of course. Initially praised by his superiors for his actions, once it became clear that there was a flaw in Oko the need to cover up their own embarrassment exceeded their desire to recognise that the officer in charge had made the correct call. Also, it was impossible to ignore the fact that whilst Petrov had in fact done the right thing, in order to do so he had actually disobeyed orders — a worrying trait in a Soviet officer.

Eventually, after a gruelling investigation process, Petrov was quietly admonished and removed from command on a technicality — he hadn’t written up his war diary as events unfolded during those critical minutes.

“I had a phone in one hand and the intercom in the other,” he explained, futilely, to the investigators, “and I don’t have a third hand.”

His protestations landed on deaf ears. Reassigned to lower grade work, his path to further promotion blocked, Petrov left the army shortly afterwards to take care of his ailing wife and work as a defence contractor. Today, he lives on a small pension in Moscow.

At the time, few people realised how close the world had come to nuclear war, indeed that obliviousness would, just a month later, almost result in disaster again during a military exercise called Able Archer. By that point, however, enough evidence had begun to mount in the West that the Soviets weren’t acting. In a final, desperate effort to convince the US of what was happening MI6 also finally gave the US full access to Gordievsky. It was a decision that would nearly prove fatal for the KGB officer, but that’s a story for another time. What mattered was that US analysts finally realised that the Soviets really did think the US was out to get them.

Neither then though, nor since, has the world ever truly come as close to nuclear war as it did that September night in 1983. A fact that would remain unknown until well after the Cold War had ended.

“Foreigners tend to exaggerate my heroism,” Petrov once told me “I was just in that place at that moment.”

He went on to insist he was just doing his duty.

That may well be true, but it is also slightly misleading. On that September morning Lieutenant Colonel Stanislav Petrov stopped and thought, rather than blindly followed orders. In doing so he did indeed do his duty, but it was not just to his country — it was to the human race.

And for that he deserves to be remembered.

John Bull

The Day the World Didn’t End