The writer is a senior fellow at the Brookings Institution. He writes a monthly foreign affairs column for The Post. He worked on Nicaragua policy-making while serving at the State Department from 1986 to 1988.

What qualifies as “success” in foreign policy? The question has arisen lately because of a CIA report apparently prepared for President Obama looking into whether the United States has ever had success supporting insurgents against unfriendly governments. The report was ordered up to inform the president’s deliberations on whether and how much support to give to the Syrian opposition. According to the New York Times, the president wanted to know, “Did this ever work?” The answer CIA analysts gave was, essentially, no. And the president was persuaded. As he commented in an interview with the New Yorker, “Very early in this process, I actually asked the CIA to analyze examples of America financing and supplying arms to an insurgency in a country that actually worked out well. And they couldn’t come up with much.”

This seems an odd conclusion, given the CIA’s own track record, which ought to be regarded as at least mixed. As my Brookings colleague Bruce Riedel notes, there was at least one “spectacular success” — the arming of the mujahideen against the Soviet forces occupying Afghanistan between 1979 and 1989. As Riedel points out, not only did the U.S.-backed rebels succeed in driving the Soviets out of Afghanistan, but also the episode played a key part in bringing about the end of the Cold War.

This was true of the Reagan Doctrine more broadly. Whatever individual successes the support for anticommunist insurgencies in Afghanistan, Angola, Cambodia and Nicaragua may have achieved, collectively they played a role in Mikhail Gorbachev’s recalculation of Soviet strategy in the mid-1980s.

Sometimes such efforts succeeded in accomplishing an objective but not in the way originally conceived. Consider the Reagan administration’s support for the Nicaraguan rebel army fighting the Sandinista regime, widely known as the “contras.” The CIA report seems to regard this effort as having been an unmitigated failure. In fact, it was substantially successful. Over the course of the 1980s, what began as a small, ineffective force mostly made up of military officers of the former Somoza dictatorship gradually grew into a large and passably effective campesino army. By 1989, the Sandinistas were frightened by this growing force and what it represented in terms of discontent across the country. As a result, in an effort to end U.S. funding for the insurgency, the Sandinistas agreed to hold a free, internationally monitored democratic election — which they promptly lost. So, yes, the U.S.-backed insurgency failed to overthrow the Sandinista regime. Instead, it forced the Sandinistas to overthrow themselves. Along the way the rebels also succeeded in accomplishing what had been the original U.S. objective: to force the Sandinistas to stop providing military assistance to the Salvadoran guerrilla insurgency next door, thus allowing a somewhat democratic government to take root there.

If the CIA report failed to note these facts, it should have, for they would have provided an instructive lesson for the president on Syria. For while it may have been true that the Nicaraguan rebels probably could never have succeeded in defeating the Sandinistas, that turned out to be beside the point. So long as they were fighting reasonably effectively, there was always the possibility that they might win. War is full of surprises. Fear of this possibility led the Sandinistas to take a risk that they never would have taken otherwise. They understood a basic truth about conflicts, and apparently better than CIA analysts. As Prussian military theorist Carl von Clausewitz put it, “So long as I have not overthrown my opponent I am bound to fear he may overthrow me. Thus I am not in control; he dictates to me as much as I dictate to him.” To avoid the possibility of losing militarily, and to regain control of their own fortunes, the Sandinistas looked for a political way out.

Might Syrian dictator Bashar al-Assad make a similar calculation if he were confronted with an insurgency well-funded and consistently supported by the United States? There is no way of knowing in advance, just as there was no way of knowing in advance that the Sandinistas would risk an election to rid themselves of the contras.

In the messy world of foreign policy, as in human affairs generally, there are no certainties, just as there are no complete successes, only partial successes and partial failures. Every problem solved creates new problems. The Allies in World War II defeated Nazi Germany but then left half of Europe under the thumb of Stalin’s Soviet Union, producing an armed-to-the-teeth Cold War stalemate that would last another four decades. So would we have been better off not fighting in Europe in World War II? Support for the Afghan mujahideen against the Soviets helped lay the groundwork for al-Qaeda a decade later. Does that mean we shouldn’t have supported the mujahideen in the 1980s? Maybe it means we shouldn’t have ignored Afghanistan throughout the 1990s.

To demand guaranteed and completely successful outcomes in foreign policy is to demand the impossible, and to refuse to act without such guarantees is never to act at all. Perhaps that is what Obama wants, a justification for inaction, which the CIA dutifully delivered. But if he is genuinely trying to weigh his options, he ought to take another look at those past “failures.” Such a “failure” in Syria would look pretty good right now.