Old-fashioned intelligence problems in Syria

Barack Obama’s administration is taking a lot of heat for newfangled intelligence activities, but it is old-fashioned intelligence problems that bedevil the administration in Syria. As a recent Wall Street Journal article (available here, but behind a subscription wall) explains, the intelligence community is divided on a basic question that figures heavily in the policy debate over what to do: Is Syrian President Bashar al-Assad doomed, despite recent battlefield successes, or has he sufficiently changed his political fortunes so that he may "win" his battle with the rebels?

The reported division among intelligence analysts provides a good teaching moment to explain the limits of intelligence, insofar as it informs policy. At most, intelligence can inform policy. It rarely if ever is dispositive in the sense of making absolutely clear that one policy alternative is "right" and the others are "wrong." And all too often, the intelligence is sufficiently ambiguous that reasonable people can disagree as to the policy implications. In those cases — the majority of the cases, in my experience — the final policy decision rightly hinges on judgments and bets that only the policymaker can make.

Consider how this plays out in the Syria case. The Obama administration is making a bet about the future, a bet that hinges on multiple predictions about what will and won’t happen. One big prediction is whether Assad will fall without substantially more robust U.S. assistance to the rebels. This is not the only prediction related to Assad that matters. Other subsidiary ones, like whether there is an available alternative regime — another strongman or some sort of power-sharing agreement — that would better suit U.S. interests, also matter. 

But Assad’s fate is a biggie, and knowing the answer to that question would go some distance to determining the costs and benefits of Obama’s current hands-off policy. The Obama administration so far has been betting that Assad will fall without ramped-up U.S. prodding. This is what the administration devoutly wishes will happen, because it would neatly and cheaply resolve the contradiction between the central pillar of the Obama doctrine of "no more U.S. ground interventions in the Middle East" and the obvious U.S. national interest in seeing Iran’s strategic ambitions in the region thwarted. The problem is that no one — not Obama, not the intelligence community (IC), not Assad himself — knows the answer. In fact, in the current instance, the IC (and probably the Obama policymaking community itself) is divided on the question.

There is no alternative, therefore, but to make a bet. And while the IC is betting in its estimates, only policymakers have the assignment — what I would call the political competence — to make the bet fully, implementing it into a policy decision one way or the other.

The Obama administration might have greater confidence in its bet if the IC were united in its own estimate (though the administration might also come to doubt such a consensus on a thorny prediction like this, suspecting groupthink at work). But even then the administration would not, or should not, have total confidence that the IC estimate was correct. And in the present situation, with some analysts betting one way and others betting the other way, the only thing that Obama can be sure of is that, if his bet comes a cropper, someone in the IC could say "I told you so."

Some of the administration’s intel problems in Syria are of its own making, or at least exacerbated by choices the administration made. A more activist policy earlier would have likely improved information gathering inside Syria — a safe-haven zone, for instance, would have been an intelligence bonanza. Most of the problem, however, is inherent in the business and should not be blamed on the administration.

Bottom line: Yes, we should work to improve the quality of intelligence collection and analysis, but the buck will never stop until it lands in the Oval Office.