Predictions About Protests Were Wrong Analysts not only failed to see revolutions coming; they failed to adjust their thinking about how far the upheavals would spread once they were under way. Forecasts can be wrong when experts rely heavily on knowledge of how things worked in the past.
NPR logo Predictions About Protests Were Wrong, Predictably

Predictions About Protests Were Wrong, Predictably

Predictions don't always turn out to be accurate.

Analysts inside and outside government sometimes get caught flat-footed by changing circumstances.

It might have been impossible for intelligence agencies to predict that the current season of protests and regime changes in North Africa and the Middle East was about to occur.

Some of the conditions that have led to demonstrations and conflict, such as high unemployment and the lack of political outlets for dissent, had existed for decades. Analysts couldn't know exactly when suppressed discontent would suddenly turn into a cascading series of protest movements, says Marina Ottaway, director of the Middle East program at the Carnegie Endowment for International Peace.

"Even in retrospect, it's very difficult to know why now," she says.

No one seemed to predict in advance that revolutionary fever would break out in the region. Even after it did, many analysts failed to recognize how far and how fast it would spread.

The protests deviated greatly from historical patterns. But even some who saw that a new day appeared to be dawning were made cautious in their pronouncements, either because of political sensitivities in dealing with longtime allies, or owing to a desire for firmer evidence that regime change could be a real possibility.


"The Tunisian example is unlikely to lead to similar upheavals."

— Stephen M. Walt, professor of international relations at Harvard University, on, Jan. 16, 2011

NPR's Melissa Block: And do you see any scenario, Ambassador Kurtzer, where President Mubarak would indeed step down, where there would be regime change?

Kurtzer: I don't see that under almost any circumstances. There have been riots and demonstrations in the past. I think this is a different sort. But so far, there's no indication that the two main pillars of power in Egypt, which is the army and the security services, have defied orders or have refused to stand up for the regime. And as long as that continues, I think Mubarak is secure in his leadership role. What he's trying to find is a comfort zone where he can be a little responsive but also show determination that he's not going to buckle to the streets.

—Daniel C. Kurtzer, former U.S. ambassador to Israel and Egypt; professor of Middle Eastern policy studies at Princeton University; on NPR, Jan. 28, 2011

"As analysts, what we can point to is the existence of conditions on the ground that make this kind of development possible," Ottaway says. "But we are never sure when the potential for an uprising turns into an uprising."

Falling Back On Past Patterns

Middle East and North Africa analysts have been, to some extent, misled during this revolutionary moment by their own deep knowledge of the region or individual countries. Decades-old certainties suddenly became shaky and it's been difficult to shift gears in a hurry.

"Sometimes the people who are the most expert are least able to see things that are outside the past norms that they're accustomed to," says Philip Zelikow, a historian at the University of Virginia and former executive director of the 9/11 Commission.

It's like having a trusty old map that has successfully guided you through foreign territory for decades. Suddenly, you reach a detour and you're put onto an uncharted, unpaved road. Your map is no longer a reliable guide, but it's difficult to break old habits and cast it aside when you find yourself lost.

"We have seen this before," says Jack Goldstone, director of the George Mason University Center for Global Policy and an expert on revolutions. "Area experts are so heavily invested, they refuse to accept that their expertise will not help them just when it seems most needed, to address crucial changes. Because Egypt had been stable since 1952, no one who has become an Egypt expert in the last half-century has studied revolutions in any depth, so they have little idea how to interpret what is happening."

That's one reason some pundits have been called out for having made inaccurate predictions, like that unrest would not spread beyond Tunisia or that Libyan leader Moammar Gadhafi would not be seriously challenged.

More important, this dynamic is one factor that has led to outside governments being slow to recalibrate their positions in response to changes on the ground.

France, for instance, at first offered assistance to Tunisian President Zine El Abidine Ben Ali in quelling protests in his country. Its foreign minister has since resigned over complaints about her interactions with the Ben Ali regime.

And the Obama administration wasn't able to pull off a completely smooth shift from supporting Egyptian President Hosni Mubarak to declaring that it was time for him to go.

"It takes a while, because your first instinct is you've got to be careful," says Lawrence Korb, a former Defense Department official who is now a senior fellow at the Center for American Progress. "Given what people in that part of the world think about the U.S., you have to be even more careful."

The Politics Of It All

It reportedly took some days for the State Department and political officials in the White House to reach the same conclusion that the time had come to part with Egyptian President Hosni Mubarak. Such conflicting headwinds within governments also help explain why intelligence agencies and government analysts are sometimes cautious about recognizing that upheaval is really happening.

"There are often political reasons why people don't want to probe too much," says Stephen M. Walt, a Harvard University professor of international relations who recently wrote a blog post conceding that his earlier prediction that revolutionary contagion would not spread had been proven wrong.

Intelligence agencies or government types may see that an opposition movement is growing, but they may be reluctant to talk openly about it; they may be wary of getting too far ahead of the official position.

Even absent WikiLeaks, Walt says, it's tricky for government analysts to state bluntly how strong they think opposition movements might be. Such assessments might contradict public statements and prove sticky diplomatically.

'Accountability Ping-Pong'

It has sometimes proved difficult for U.S. intelligence agencies to adjust their thinking to changing realities, says Philip Zelikow, the former executive director of the 9/11 Commission.

For decades, they planned for the possibility of a Soviet attack. That's one reason they failed to anticipate the Sept. 11, 2001, terrorist attacks, according to Zelikow. "They didn't turn toward the enemy most likely to attack," he says.

They also failed to think about how al-Qaida had previously turned other modes of transportation — cars, boats — into bombs. If they had done so, they might have been on guard for the possibility that they would do the same with planes and reacted to the warning signs that known terrorists were learning to fly.

"To anticipate 9/11 didn't require an out-of-the-box thinker," Zelikow says.

But Philip Tetlock, a professor of management and psychology at the University of Pennsylvania who has written about how expert predictions are shaped by knowledge of the past, says the intelligence community gets "caught in a game of accountability pingpong."

After being blamed for failing to connect the dots that presaged the 9/11 attacks, U.S. intelligence agencies soon got blamed for "overconnecting the dots," Tetlock says, and for practically guaranteeing that Iraq possessed weapons of mass destruction.

Forecasters who can't spot warning signs or adapt to changes in real time are bound to be blindsided. But intelligence professionals are bound to come under criticism regardless of whether they're too cautious or get too far ahead of the pack, Tetlock argues.

"If they overpredict the instability of autocratic regimes, they get blowback from the State Department for undermining relationships," he says. "If they underpredict, they get blamed for leaving the president unprepared."

Still, Walt says it's part of the job of both intelligence agencies and the diplomatic corps to "think about unlikely but not implausible contingencies," though he says he's not sure it was discussed in detail this time around.

"How much had we really thought about what government policy was going to be if Mubarak was going to collapse?" he asks. "I'm willing to bet there wasn't anybody who had said our policy would have to change."

And while former 9/11 Commission director Zelikow agrees that part of the job of intelligence agencies is to be thinking about various scenarios, he says it's not their job to forecast the future.

He says intelligence analysts should not be expected to make concrete predictions about future events and their exact timing.

"An intelligence agency's job is not to handicap a horse race or pick a stock," Zelikow says.

Instead, the job is more analogous to an engineer building a dam. He can't tell you when the dam is going to burst, but he should be able to analyze points of weakness and help construct designs that protect against them.

"The job of professionals who should try to deal with highly improbable events is not to guess right when it would happen," Zelikow says. "It's to figure out when there are high possibilities of black swans, and what factors might make them happen."

People who forecast unlikely but not implausible events, like revolution in long-stable countries, may be touted as prophets in retrospect. But because such predictions tend to be wrong far more often than not, they're not often taken seriously at the kind of places where analysts get paid to think about other nations — government agencies such as the Defense Department and the CIA, as well as think tanks and universities, says Rajan Menon, a political scientist at City College of New York.

"The evidence for foreseeing cataclysmic changes is simply not there most of the time," he says. "So when you're asked the basis for your offbeat prediction, you're reduced to being seen as a soothsayer, whereas you want to be seen as a hardheaded professional."