Lost in Translation

Truman, Stalin and the 'thumbs up' emoji problem

Harry Truman was pleased with how the conversation with Joseph Stalin had gone. In his memoirs, Year of Decisions, the president gave the following account:

“I casually mentioned to Stalin that we had a new weapon of unusual destructive force. The Russian Premier showed no special interest. All he said was that he was glad to hear it and hoped we would make ‘good use of it against the Japanese.’”

What Truman, Secretary of State James Byrnes, Winston Churchill and others took away from the interaction at Potsdam was that Stalin did not grasp the significance of what the president had told him. They were wrong. Thanks to his network of spies at Los Alamos, Stalin had long known about the Manhattan Project. Indeed, he had learned of it years before Truman.

In his brilliant new book, The Most Awful Responsibility: Truman and the Secret Struggle for Control of the Atomic Age, the author Alex Wellerstein observes:

“Stalin’s apparent blankness was a deliberate act of non-reactiveness in the face of what was perceived to be political blackmail… What had been urged upon Truman as an act of openness was instead understood by the Soviets as a threat.”

Potsdam calling the kettle black

The Truman-Stalin problem is universal. The gap between what we say and what others hear is often so wide as to represent a threat to national security. That is because messages don’t arrive untouched or in isolation. They pass through layers — prior knowledge, assumptions about intent, insecurities, power dynamics, cultural norms and personal history. Just see below…

Humans have a seemingly endless array of cognitive biases. In this case, it is the illusion of transparency, broadly defined as the tendency to overestimate the degree to which our internal state is visible to others. One of the first examples of this phenomenon was a 1990 study conducted by Elizabeth Newton, then a Stanford University doctoral student.

In the experiment, Newton divided her fellow college students into ‘tappers’ and ‘listeners’. The former were asked to tap out the beat of a well-known song, while the latter had to guess what it was. The tappers predicted the listeners would correctly guess the song around half the time. In their heads, it was so obvious! The reality was that the listeners correctly guessed the song all of 3% of the time.

The text message is perhaps ground zero for modern-day communication problems.