A friend of mine recently made some humorous remarks about the lack of WMD found in Iraq that reflect the usual view that the war was unjustified, or at least that it was unjustified on the basis of the threat of WMD. And this started me thinking about how I might try to clarify the understanding of a genuinely critical issue: How do we deal rigorously and appropriately with uncertainty about a significant threat? So I'm going to try to delve into the alpha and beta of the problem again, without surrendering the reality that there were a host of other reasons for embarking on the war, from gamma to omega.
The expectation seems to be that the Bush Administration made the wrong choice because WMD haven't been found. But this fails to recognize that there are no decision methods that rule out the possibility of error completely. It's all a matter of what kind of error you're willing to live with. And we can apply two different kinds of decision methods, neither of which has an absolute guarantee of being error-free. I'll call these methods, for the sake of brevity and consistency, alpha and beta. I'm going to forego the technical language and simply assert the following: An alpha decision method is equivalent to an assumption of innocence, while a beta method is equivalent to an assumption of guilt. Let's say we have a "glass" or container that holds signals and reasons for believing a plea of innocence. If you're willing to call the glass "partly full" until the level drops below a certain threshold you're an alpha sort of person, whereas if you insist on calling it "partly empty" until it rises above a certain threshold you're a beta sort of person. Obviously you're "playing it safe" if you're a beta guy.
I attempted in an earlier post to deal with the concept of uncertainty by reference to the Florida Recount. The thing about the Florida situation is that both sides had an asymmetric view of the threat, and tended to discount the threat the other side saw. The Bush people, for instance, tended to discount the loss of social legitimacy that would result from the appearance of having "won" the election on a legal technicality. And the Gore people tended to discount the threat associated with dragging the process out to a national distraction. Granted this is a simplification, but the point is that both sides were seen as mean and combative by the other side's partisans, because both were utilizing a beta decision method.
For most decisions the consequences of convicting an innocent person (or regime) tend to outweigh the consequences of failing to convict a guilty party, Willy Horton notwithstanding. But when the consequences of an erroneous acquittal can lead to huge numbers of casualties, the only responsible method is to view the glass as "partly empty" and to demand a fairly high threshold before you treat it as "full." And if you're concerned about a regime that Freedom House regards as the most repressive on earth, and that has a history of cultivating flowers of evil and practicing deception and bad faith for twelve years, you'd have a right--nay, an obligation--to insist that this regime be assumed guilty until proved otherwise, bad press or not.
Most people, after all, are familiar with criminal trials where the defendant is presumed innocent and must be proved guilty "beyond a reasonable doubt." This situation is the norm in many other kinds of circumstances that call for judgment and decision as well. When most employers conduct hiring interviews they don't usually presume that the interviewee is a criminal. To do so would have the cost of potentially excluding otherwise good workers. That's why most companies don't administer drug tests to employees. But the risk is that sometimes you'll mistakenly hire a sociopath or a drug abuser. And indeed over the period of many hirings you almost certainly *will* hire someone like this, if your company is large, even though the frequency of such people in the society is low. The primary objective of personnel departments is to hire as many good and qualified people as possible, and to keep them happy, in order to be competitive. Certain companies and agencies, however, have missions that demand security, so they do things that demonstrate a more cautious set of assumptions. They may conduct routine background checks or random drug tests. The fact is that if you toss out all the bathwater without being sufficiently careful you're bound to end up with a few missing babies. And as a general rule, we eschew such strategies.
This means that an alpha strategy is the "typical" decision method in our enlightened society, and beta strategies are considered rather anti-social and mean. We presume innocence because we seek to optimize performance, justice, or happiness. Those women looking for a mate who presume that all men are pigs are likely to end up single and alone, etc. And that is precisely why, even after Hans Blix expressly stated on Charlie Rose and elsewhere that the decision method he intended to follow in Iraq would be dictated by a "presumption of guilt," he and his bosses in the UNSC, and most of the non-US public, imperceptibly drifted toward the presumption of innocence regarding the Hussein regime. It's human nature.
So we began to look for reasons not to trust Saddam, as though they were hard to find. As hard as, say, WMD. And everyone focused on "the big hunt."
Well, lets examine what ought to have been done instead. The obligation of someone who applies a beta decision methodology is to diligently search for reasons to reject their assumption. And the more gravely you regard the consequences of falsely rejecting the assumption of guilt, the higher threshold you place on rejecting your assumption of guilt. And that's not human nature. After all, you're a mean person who stubbornly insists on seeing the glass as half empty, which in most cases (and especially in diplomacy) is considered an act of bad faith. Well, it is true that it's an act of low faith. But in this case low is good.
At this point in the essay I had intended so say that Bush's mistake was that he, apparently, expended little effort in fulfilling his obligation as a beta guy. He ought to have actively sought to falsify his assumption of guilt, because that's the proper way to implement the strategy. From a political perspective, however, it's risky to gather evidence for the opposition's case, which is essentially what you have to do to be methodologically rigorous. A couple of instructive examples leap to mind: The Pentagon Papers, and Rumsfeld's recent internal memo on the War on Terrorism. However, it occurs to me that you don't falsify the assumption that Saddam is guilty by looking for WMD. That may seem surprising, but it's true. The difficulty of the search, even after the country itself has been secured and Saddam ousted, suggests that simply failing to find WMD wouldn't be enough. And Saddam knew that if we tried that strategy we'd eventually get tired of it. But the fact is that we've assumed the weapons are there, so there's really no direct reason to look for them. What you look for, instead, are cooperative behaviors on the part of the Hussein regime that are inconsistent with the guilty intentions that you assume are there. And you do this by looking for WMD, attempting to provoke behaviors that reveal underlying intent. So the search for WMD is simply a means for gathering evidence about behavior, and your interest in finding WMD is secondary.
And it's simply the case that every time we pushed, Saddam pushed back or held out for as long as was possible, up to and including threats of violence and retaliation. For instance, Blix had negotiated an aerial reconnaissance regime where we gave prior warning of spy flights and agreed to send up only one plane at a time. And we had an incident where a second U-2 plane was launched "by mistake" from Saudi Arabia. And Saddam didn't simply say "Well, you said you weren't going to do that... but it's no big deal. No problem if you want two planes up." He said, instead: "Get that second plane out of the sky, or I'll shoot it down." At which point Blix ought to have said: "Go ahead and down it. We have every right according to your agreement to abide by 1441 to put up as many planes as we deem necessary." And if he had done that there'd have been little ambiguity left about Saddam's intentions. One way or the other we'd have opened a path. So Bush was doing everything he could to implement the beta guy strategy as rigorously as possible. It's just that everyone else slowly edged over to a more comfortable alpha guy behavior. Bush kept his eye on the ball, while the attention of the UNSC and everyone else wandered.
But some might argue that by pushing Saddam in such a manner we would have been taking enormous risks, or would have been insufficiently cognizant of the realities of Saddam's position. We'd either be subjecting him to a loss of esteem within his hierarchy, or flirting with the danger that he'd strike first, with some strategic WMD. As regards the first concern, I admit that confronting him in this way would have placed Saddam in something of a bind. But better him than us. And it also had the potential of creating conditions for a coup, or at least dissension in the command/control of the regime. And how is that bad, exactly? As regards the second concern, I think we could safely assume that the risk was minimal. Ironically we had pretty reliable behavioral evidence that there was no "imminent threat" of the surprise use of WMD, because Saddam was, himself, clearly facing an imminent threat on his borders, and if he'd had a strategic weapon in his arsenal he'd have at least shown it as a deterrent. He had a conspicuous example, in N. Korea, as a model for such deterrence. So he allowed his headache to increase, because he had no medication. That's not a difficult inference to make.
And there's a good deal of evidence to suggest that Bush came to precisely this conclusion. As the recent debates between Holsclaw and Schwartz covered by Daniel Drezner as well as this excellent timeline make clear, Bush was either saying we oughtn't wait until the threat became imminent (which is understandable prudence) or he was making a case for lowering the standards of what constitutes an imminent threat, to include a "grave and growing" danger. Bush was consistently and appropriately employing a "beta guy" decision method. If he made a mistake it was in conforming to the appearance of an alpha method. But, on the other hand, if he had tipped his hand the behaviors observed would have been more self conscious and difficult to interpret, because they'd have included an element of dramaturgical strategy to confuse and obfuscate even more pronounced than Saddam's usual over-the-top behavior. He had to play the game, to keep the observations as pristine and clear as possible under the circumstances.
Posted by Demosophist at October 25, 2003 01:56 PM | TrackBack