YOU ARE HERE: LAT HomeCollections

Why we're clueless

The foreign policy debate in Washington is blinkered and conventional.

February 18, 2007|Michael May | MICHAEL MAY is professor emeritus of engineering at Stanford University and a senior fellow at Stanford's Institute for International Studies.

BOTH PROPONENTS and foes of the "surge" in Iraq made firm, confident predictions about what would certainly happen if their preferred course of action were not followed.

"If the U.S. walks away now, extremist elements will take over Iraq," said one group of government officials and pundits. "No, no -- the proposed surge is too little, too late," said another. Some insisted that "the surge will only inflame anti-American sentiment."

But here's the only thing we know for sure: When the collection of politicians and pundits we call "Washington" makes predictions about countries in which the U.S. has "vital interests" (and especially about countries with which we have had bad relations), the predictions -- even when they contradict one another! -- are almost always wrong.

Consider the record. Washington didn't predict the fall of the shah in Iran, or the end of the Cold War, or the disintegration of the Soviet Union. Nor did our pundits (whether left, right or center) predict the war between Vietnam and China after the U.S. lost more than 50,000 service members in the region to prevent the spread of monolithic communism, or, for that matter, China's turn toward becoming a capitalistic and trading giant. There are innumerable other examples in Latin America, Africa and the Middle East.

Everything is fine as long as nothing much is happening. But when something happens, especially if it's sudden or revolutionary, we usually don't see it coming. Since World War II at least, Washington's success rate at predicting change in countries with which we have a hostile relationship is close to zero.

Why this failure when most of the events mentioned were years in the making? In part, it's that we don't look at the right things. Our intelligence apparatus is better at counting missiles than at following the intricate politics of a central committee or overhearing the street talk in Tehran -- or Baghdad. But even good intelligence on those matters would have had a hard time making it into the Washington debate.

That's because Washington debates tend to be narrow; they fall within a range of acceptable conventional wisdom and close out anything that seems to go out on a limb. We wonder: Is the Soviet Union trying to build a first-strike capability or not? Is Vietnam a tool of Red Chinese expansion? Does Saddam Hussein have nukes? Intelligence is asked to cast light on just those acceptable questions. The assumptions underlying those debates are seldom challenged. They are based on politically acceptable views of the other state and, career-wise, it doesn't pay to challenge them. Indeed, movers and doers in Washington often do not even have time to challenge them. There is little or no intelligence outside the axis of debate.

As a result, any analyst who forecast that, say, the Soviet Union would for its own purposes peaceably release its European satellites, or that the heirs of Mao would turn to capitalism before his body was cold in the grave, would have been quickly marginalized.

When researchers test predictions about, for instance, investments or the effectiveness of a new drug, they begin with the assumption, for testing purposes, that the theory or the advice leads to no better result than what could be attained by chance, or that the drug is no more effective than a placebo. It's up to the data to prove that the advice or the drug is effective, not the other way around. That's called starting with the "null hypothesis."

The null hypothesis means that there is no ground to believe or disbelieve any prediction. It's a way of guarding against bias.

The record of official Washington predictions gives us no reason to depart from the null hypothesis. Whatever the prediction, we should consider it just as likely that something entirely different will happen.

If the U.S. withdraws from Iraq, for instance, the civil war could grind on for years, or, conversely, Prime Minister Nouri Maliki could bring order to Baghdad within a year, probably by ethnic cleansing. The surge could lead to more needless deaths, or Gen. David Petraeus could bring a better kind of order out of chaos, at least for a while. If President Bush attacks Iran, it could help cool the situation, or we might have to hunker down for a longer, bigger war. Or perhaps the situation may be resolved not by actions from the U.S. but by initiatives from within the area, as were the earlier Soviet and Chinese situations.

Whatever we do, the situation a few years from now is likely to be different from what's currently predicted by any party to the Washington debate.

Los Angeles Times Articles