Overconfidence at the top threatens to plunge the world into war

As the world (including the self-styled peacemaker-in-chief in the White House) holds its breath for the announcement of this year’s Nobel Peace Prize, spare a moment to ponder the growing risk of war, including world war.

Even a cursory scan of today’s major military powers suggests that both their leaders and policy elites are dangerously overconfident, and could — as in 1914, say — sleepwalk into disaster out of what international-relations scholars call mutual optimism.

If you’re not worried yet, consider a study, the largest and most international of its kind, that comes to exactly this conclusion. Jeffrey Friedman at Dartmouth College just published the findings of surveys he’s been giving (between roughly 2016 and 2022) to about 2,000 national-security officials from more than 40 Western countries — men and women, North Americans and Europeans, civilians and service members.

Friedman’s questions took the form of statements to which the officials had to attach probabilities. A few samples: The United States is the only country in the world that has stealth aircraft. (The correct answer is no.) There are more active-duty military personnel in the European Union than in Russia (yes). Jihadi terrorists in the preceding years killed more people in France than in the US (yes). There are more refugees from Syria than from Venezuela (at the time, yes).

Starting in 2020, Friedman told me, he started asking every question in two versions. For example, half of the participants received this variant: “What are the chances that Boko Haram has killed more civilians than ISIS since 2010?” The other half got: “What are the chances that ISIS has killed more civilians than Boko Haram since 2010?”

As you’ve guessed, Friedman wasn’t after quizzing the officials’ knowledge, but after gauging what I think of as their “intellectual humility” (or its absence, hubris). And the data were clear: Participants were wildly overconfident.

When participants estimated that statements had a 90% chance of being true, those statements were true just 58% of the time — basically, a coin flip. Even when participants felt completely certain — assigning a 0 or 100% chance — they were wrong more than 25% of the time. There was no difference between men and women, Americans and Europeans, brass and civilians.

Moreover, the participants weren’t just wrong randomly, but prone to false positives in particular — that was the point of flipping the questions. You’d think that if you ask a large number of rational experts to assign probabilities to either ISIS or Boko Haram being more lethal than the other, the averages should sum to 100%. But they consistently (for 244 of the 280 questions in the experiment) added up to much more.

In other contexts, such bias toward false positives suggests that people are more likely, say, to send an innocent person to prison than to set a guilty person free. In international relations it helps explain, for example, why advisers in the White House in 2002 felt certain that Saddam Hussein was trying to build nuclear weapons (when he wasn’t) and were confident that they could not only topple his regime but also stabilize and democratize Iraq quickly (when they couldn’t).

In venturing hypotheses for this perilous cognitive asymmetry, Friedman points to the work of psychologists such as the late Amos Tversky and Daniel Kahneman, the world champions of exposing cognitive biases. One is the availability heuristic, our human tendency to exaggerate the probability of whatever comes readily to mind, and to brush aside other possibilities.

In 2002, for example, it was much easier to imagine that Saddam was importing aluminum tubes to build centrifuges for enriching uranium than to consider that he just wanted the metal to make conventional rockets (which turned out to be the case) or something else entirely.

Another trap is the so-called acquiescence bias, our tendency to say yes before even considering the content of a proposition. This gets worse by multiples when you add groupthink, peer pressure or outright fear. That’s why authoritarian regimes tend to err more disastrously than open societies do. Think of Vladimir Putin’s invasion of Ukraine in 2022, which his counsellors and generals assured him would take a matter of days.

The bad news that follows from Friedman’s research is twofold. First, memories of the last world war have faded, and the current generation of leaders and experts — from China and Russia to the US and elsewhere — is showing signs of waning humility and growing hubris, similar to European leaders in the summer of 1914.

Second, the mightiest military power on the planet, the United States, is moving away from a culture of open and objective analysis and toward groupthink and motivated reasoning based on loyalty tests to the leader— what one might call a war on expertise.

There’s also good news, though. Friedman discovered in his surveys that you can dramatically boost humility and improve results by giving officials just two minutes of training, in effect priming them to be aware of their biases.

The stakes in international relations are often war and peace, life and death. Consider some of the questions that the White House currently has to grapple with. Did the US in fact “obliterate” Iran’s nuclear program, or merely set it back for a while? Is Russia waging hybrid war against European NATO countries only to harass the alliance, or to test its vulnerabilities for a full-bore attack? Does North Korea have plans to attack the South, or China to seize Taiwan? If it comes to war, who would be more likely to win?

Here are my suggested lessons from Friedman’s research to leaders of all countries: First, value expertise and recognize that its job is to tell truth to power, not to flatter you. Second, don’t allow advisers to present single scenarios, but insist on alternative hypotheses — then flip them, so that positives become negatives.

Above all, don’t reward confidence (and certainly not showmanship) among your officials, but humility. And always, always, always stay humble yourself.