Reviews | The reason intelligence agencies keep getting it wrong

(Anthony Gerace for the Washington Post)
(Anthony Gerace for the Washington Post)

Intelligence agencies have a poor track record of crucial predictions lately, at least those that have become public.

Until the last minute, the most gloomy forecasts from American intelligence agencies last summer indicated that the Afghan government could stand up to the Taliban for three to six months after the departure of American forces. Nobody expected an almost instantaneous collapse.

Then, in mid-February this year, “US officials” – in the usual anonymous style – expected Russia to take kyiv in a matter of days and all of Ukraine in about a week. Russia’s own strategy assumed no resistance, believing the Ukrainians would greet the invaders with flowers as liberators.

Ukrainian President Volodymyr Zelensky, meanwhile, figured in US expectations as someone to evacuate from kyiv. Zelensky himself called the threat of invasion mere “psychological pressure”, which matched Ukrainian intelligence’s assessment that Russia was bluffing and not attacking.

History does not repeat itself, but it does echo. Echoes of intelligence history teach that the fundamental error of the field is too much confidence in predicting how people – individuals and entire nations – will act.

To be fair, the American assessment that Russia was preparing to invade turned out to be entirely correct. That, it seemed, depended on what intelligence could accomplish, particularly through technological means – uncovering orders, supplies, troop movements and more.

A century ago, armies began to transmit large amounts of encrypted information by radio. This created the opportunity to intercept and decipher an enemy’s communications – as Britain did with its success against Germany’s Enigma cipher during World War II. After that came satellites, fixing the movement of forces. Add in a reported network of human sources, and US agencies had a high-resolution picture of what Russia was doing.

Ukrainian agencies probably received a lot more of this image than the public – but still misinterpreted the intentions of Russian dictator Vladimir Putin. This is partly because they knew how internally divided Ukraine was. Zelensky’s approval rating was 21%. Putin could have destabilized Ukraine without invading it, a senior intelligence official there told Israeli correspondent Anshel Pfeffer. Logic said Putin didn’t need the war.

It’s dangerously easy to assume the enemy shares your logic. In early April 1940, British military intelligence had multiple signs that Germany would invade Norway – and ignored them because, logically, Germany would target Norway and Sweden together, and too few German divisions were available to conquer the two countries.

On the morning of December 7, 1941, with deciphered messages that Japan was about to attack the United States, Army Chief of Staff George Marshall prioritized Manila’s warning because, logically , which is where Japan was likely to strike. His choice contributed to a fateful delay in warning commanders at Pearl Harbor.

In early October 1973, Israeli intelligence ignored any evidence that Egypt was preparing to attack because, logically, it would not do so until it had the weapons to retake all of Sinai. Egyptian President Anwar Sadat followed a different logic.

As for Zelensky, no one foresaw that he would mobilize the Ukrainian language and send it into battle, to quote what journalist Edward R. Murrow said of Winston Churchill. But then, American assessments barely predicted that Churchill would become the wartime leader we remember. An American diplomat reported in early 1940 that when he met Churchill, the latter was “drinking whiskey and soda”. The diplomat added that “it was quite obvious that [Churchill] had consumed quite a few whiskeys before I arrived. Other American diplomats described Churchill as “old and tired” and a “has-been”.

It is remarkably difficult to predict how a person will react to a crisis, who will escape responsibility and who will rise powerfully in it. It is even more difficult to judge the mood of an entire nation, as well as to predict how it will behave in a cataclysm.

The British National Archives hold thick files of diplomatic messages from dozens of countries during World War II, all intercepted by British intelligence. Some of the cables contain reports of a country’s state of mind, based on diplomats’ conversations with officials, businessmen or anonymous sources. While perusing them, I sometimes found one that lit up with apparent prescience, given what happened in the months that followed. However, many were wrong. Perhaps the correct ones were no more prescient than a bet on the winning number at the roulette table.

Zelensky’s superb political theater surely strengthened the resolve of Ukrainians. But more was at work. The decision of a random 40-year-old citizen to volunteer for the Home Defense Force and potentially face combat with minimal training could also be the result of a conversation with a neighbor, who to his tower was influenced by his brother-in-law. Afterwards, historians will explain why such an emotion crossed a population. But this is not obvious a priori.

“The heart is deceitful above all else,” the prophet Jeremiah warned all who would try to prophesy after him. Satellites can photograph supply lines; code breakers can crack secret commands. Neither can read what a person will do under extreme duress; a person does not know himself until the time comes.

The biggest error in intelligence assessments in recent months has not been misinterpretation by leaders or nations. It was the pride, once again, of placing too much trust in these assessments.


Comments are closed.