April 21, 2005

Is Counterterrorism Really Dying? (Updated)

by Demosophist

OK, I'm still not sure whether I'm out to lunch on this issue of the counterterrorism report, or whether it's Larry Johnson who's missing the boat. It's certainly possible that each of us it outclassing the other on some aspect of the analysis. For the sake of argument, I'm willing to concede that the accounting method used to define a terrorist act hasn't changed at all from 2003 to 2004 and that continuity isn't the issue. I'm not 100% convinced of that, for one thing because Larry characterizes the whole debate as a "mere squabble over method," and as a methodologist, I know of no such thing. Issues of method, as I learned from Thelma Z. Lavine, go right to the heart of what we call "reality," and it just doesn't get any more momentous or important than that. But again, I'll concede the point that the method hasn't changed, and also that the State Department attempted to change it, failed, and then cancelled the Annual Report.

Well, first of all if the current definition of a terrorist act involves no attempt to suss out the issue of intent, or leaves it ambiguous, then it might well be appropriate to change the method... to make it a better reflection of the kind of act we're attempting to stop. And I realize it may be extraordinarily difficult to define intent, operationally. But again for the sake of argument let's assume the worst case, that the State Department didn't care about accuracy, and they just wanted to keep the numbers low so as not to appear to be "losing" the War on Terror.

That, in itself, leaves a lot of questions open. For one thing, why is that bad? It's certainly the case that however well-meaning or underhanded the attempt, it failed spectacularly. They haven't exactly kept things under wraps. For another, as Larry mentions on the Counterterrorism Blog, one has to remain cognizant of details embedded in the data to have any prayer of formulating coherent policy. I assume that we want the State Department to have a coherent counterterrorism policy, even if we don't want them leading the charge... so the criticism is well-taken, if the decision not to publish the report implies that the State Department would not, then, be made aware of these embedded truths in the data. (These usually take the form of cross-tabulations, or summary variables categorized by another variable, sometimes called "cross-sums" or "tab-sums.") If the only, or the best, way to disseminate this form of knowledge is through the kind of published report that the State Department has just curtailed, then indeed we're stepping from the twilight deliberately into the dark. It's a bad thing.

But there's more to determining whether this is true, than simply making the assumption that it is. Perhaps white papers, or periodic reports, or even having their own in-house analysts to continually process data, is the better way for the State Department to go? Perhaps a continuously monitored website, part of which is available to the public, of the sort that's represented on NationMaster, or on the OECD's Factbook. This is certainly a better tool than a predigested report, right?

Another issue concerns how one interprets the data. If the number of attacks is escalating, but the geographical area where they take place is shrinking, is it clear that we're losing, or winning? Certainly the level of violence escalated dramatically at the Battle of the Bulge, but it was the beginning of the end for the Third Reich. And I can't help but feel that if we'd had today's data dissemination and analysis technology during the Tet Offensive, with the ability to substitute our own rather than relying on MSM's flawed interpretation, we'd have drawn very different conclusions about the meaning of the event.

Finally, we are talking about the State Department here, not the Intelligence Services themselves. To what extent does this refusal to publish really reflect deep-seated policy? I'm in over my head here, because I just don't know that much about the internal culture of the State Department... but it would seem to me that Condi Rice is probably a great deal more empirical than her predecessor, so the conclusion that they're abandoning empirical analysis may be jumping the gun a bit. They may have simply decided that they didn't want the hassle of publishing this report. After a brief attempt to alter the method to better reflect the intent behind these acts of terrorism they may have decided that the political ramifications of publishing the report were getting in the way of making sound policy decisions. Believe me, it happens. Policy and advocacy are always in a tug of war, and people invested with the power to make decisions tend to look at advocacy as unnecessarily messy and distracting, because sometimes... it is.

Finally, who ought to "take the lead" on counterterrorism? As I said in the comments to my earlier post, the State Department's historical addiction to stability suggets it's not the ideal institution to adopt that role. Granted, the State Department needs a counterterrorism policy, but should that be derived independently of the Intelligence Services or the DOD? To what extent should they lead, or follow? I know I have my preferences, but it's not really my decision... and this may be one of those issues that isn't resolved for several administrations anyway. The decision will impact careers, income, and the distribution of power and responsibility within government, and I'm not wise enough to know what ought to happen. But I do think it'd probably be a mistake so simply assume that the status quo is best.

Update: Larry Johnson mentions Kashmir as a revealing example of why "details buried in the data" are critical to formulating coherent policy:

One of the uncomfortable answers buried in the data concerns the attacks in Kashmir. The groups responsible for these attacks--e.g., the Lashkar Tayibba, the Harakat ul-Ansar--have received funding and support from elements of Pakistan's intelligence service. In the past these groups were training in Afghanistan alongside Al Qaida. In other words one of our key allies in the war on terrorism has been a sponsor of terrorism.

Well, if it's actually the case that a large proportion of these attacks are being generated by an ally (and attacks in Kashmir account for almost half of the attacks counted in 2004) then we certainly ought to factor that into how we deal with the Pakistanis. But the fact that there's a kind of schizophrenia within Pakistani Intelligence isn't exactly new information, so I'm not sure what Johnson would propose, beyond turning Pakistan from an ally into an enemy. Depending on which one of its "personalities" wins the internal struggle, that may happen anyway... but the real foreign policy issue is whether we're prepared to push this ally into the realm of greater political and civil freedom, and historically the State Department hasn't been very supportive of such a policy.

On a related note, the Kashmir problem may be on the verge of a solution, anyway. If so, the biggest problem faced by Kashmir, one of the most beautiful places on the planet, may be how to handle the honeymoon tourist trade.

By Demosophist at 06:22 PM | Comments |