Hm. What to write about? Scooter Libby? Nah, not dull enough. Oh, what about John Negroponte's new "National Intelligence Strategy," released last week. Yes, that's the dullness we need. Important, though, especially since the new report doesn't have much to say about one of, I think, the United States' most persistent intelligence weaknesses over the past decade. More on that in a bit. For now, it's enough to note that Negroponte's report is full of resolutions to coordinate this and integrate that and improve the other, and it's all focused, naturally, on using intelligence to combat terrorism, stopping WMD proliferation, and, uh, democracy promotion. Apart from the last, which is bizarre, this is all pretty uncontroversial and likely useful—up to a point.

The idea that we need to "improve and integrate" our intelligence is always a nice platitude, but it's worth stepping back and asking how we got to where we are. In this day and age, curiously, every major foreign policy move needs to be backed, it seems, by ultra-solid intelligence—as if to give the public a veneer of objectivity for what are often simply judgment calls, hunches, guesses. That alone puts a very high degree of natural political pressure on what is otherwise an inherently imprecise process. The Bush administration may have been particularly flagrant about "stove-piping" CIA reports during the march to war in Iraq, true, but any policymaker who gets it in his or her head to pursue a course of action will have to do the same to some degree or other, because he or she will want backing from the intelligence agencies, and they can often provide no such thing.

Now Congress can always re-jigger the ways in which the various agencies are set up, as it recently did, and maybe that will alleviate some of the pressure from here, but the implicit "politicization" of intelligence can never go away completely. In a better world, policymakers would just acknowledge that intelligence is highly imperfect even in the best of times, and tell voters that ultimately, their national security decisions are primarily judgment calls, rather than obvious conclusions borne of intelligence. So many charades could be dispensed with. (Including the idea that ordinary citizens aren't informed enough to form an opinion on foreign policy decisions—a fiction that should have been buried by the Iraq war.) This cultural change will never happen, of course. So the real problem is that foreign policies that put an impossibly high burden on intelligence—the Bush doctrine and preventive war come to mind—will likely fail more often than not.

(As a side note, if only because I don't know where else to stick this, Chaim Kaufmann's "Threat Inflation and the Failure of the Marketplace of Ideas" is one of the better essays on intelligence failure during the run-up to the Iraq war; he notes, among other things, that the tendency of experts to avoid treating what they think they see as a scientific hypothesis—which would entail making predictions—was an especially ingrained failure. No amount of shuffling or re-jiggering will fix this.)

At any rate, these are badly disjointed thoughts, sorry, but I promised to say a bit about one of the most glaring and overlooked types of intelligence failure throughout U.S. history: namely, our poor ability to predict how other countries—or other people, period—will react to our actions abroad. Robert Jervis has written a few papers on this, but to put it another way, U.S. policymakers rarely seem to be able to figure out how other countries see the world, a blind spot which, during the Cold War, was more serious than the various mistaken analyses about missile gaps or mineshaft gaps or the like. The problem is that this sort of "empathy" is much very difficult to improve—trying to figure out the near-infinite set of calculations and beliefs other actors might have will always be close to impossible.

A few examples from history: In the 1950s, the U.S. failed to understand that Stalin would invade Korea—at the time, his strategy was remarkably opaque. More recently, in 1996, the Cedras junta in Haiti for some reason didn't take the Clinton administration's warnings to abdicate seriously until an invasion force was actually in the air. (Did they not think Clinton meant it? Why?) Ditto a few years later, when the Clinton administration couldn't understand why Milosevic wouldn't back down from Kosovo in the face of NATO threats. Nor did the Bush administration make any apparent attempt to understand why, in 2002, Saddam might have been acting the way he did—for instance, keeping the status of his WMDs ambiguous to fool Iran. But so long as the U.S. has a poor handle on the beliefs and calculations of other world leaders, especially its adversaries, coercive diplomacy will tend to fail. (And we haven't even touched on terrorist groups....)