April 21, 2005

Well, Somebody's Weak Grasp of the Issue, Anyway (Rapidly Updated)

by Demosophist

Larry Johnson, the fellow who broke the original story that the State Department was tucking its tail between its legs and dropping stats on the number of terrorist attacks, because the number increased from 2003 to 2004, has posted an "update," on the Counterterrorism Blog that seems to counter the contention that the data method changed before the State Department cancelled the report. (He contends that they tried to change it themselves, while they contend that they tried to revert to the old method for the sake of continuity.) He seems to be saying that the only change is that the State Department has decided to ignore the data because it's disagreeable, but others have suggested that they've decided to ignore it because a new method used to compile the numbers loses continuity with the old statistics, and that therefore the apparent increase isn't real. And if you read what he says carefully, he doesn't quite make the assertion that their version isn't true. He just blusters about it alot. What's going on? Enquiring minds want, and need, to know?

Just to make myself perfectly clear, the fundamental issue is whether or not the State Department attempted to establish an entirely new accounting method at the NCTC, just to give the impression of a smaller the number of apparent attacks, OR they tried to reinstate an old accounting method for the sake of continuity.

Johnson seems to be saying the former, but provides no proof, and he acknowledges that there are "problems with the statistics." What problems? Problems caused by a new method that simply counts more events as "terrorist attacks?" To me that would justify the State Department keeping arm's length. He characterizes the move as "cowardice" but one might also call it justifiable prudence, depending on just how bad the "data problems" really are. It would be rather foolish to legitimate an increase in the enemy's activity in an official report, if that's not actually what's happening. I also get the impression that Johnson is arguing that there's a coverup, but from what I can tell the statistics are still available, through the National Counter-Terrorism Center. No one will be hiding them. So what is being suppressed, exactly?

The bottom line, I think, is that even though Johnson calls this an "update" it adds nothing new to the debate. He's just saying the same stuff he said before, in a different way, and neither providing any new information, nor even clarifying his earlier statement. Were the 2004 statistics based on a new data gathering method, or a new definition of a "terrorist attack," or not? How hard can it be to make that clear?

Update: Though it may not be apparent to everyone, and Johnson certainly hasn't made it very clear, there are two issues at stake in this sort of data problem: continuity and accuracy. The shift to a new method of accounting that tends to count more events as terrorist attacks may be entirely justified on accuracy grounds, but may make it very difficult to track trends. The conventional way to handle this, for any data instrument that improves method in order to improve accuracy, is to publish figures based on both methods for at least a few years, long enough to build up a trend line based on the new method. That way you don't sacrifice your ability to make sense of trends, for the sake of the admitted benefits of greater accuracy.

Update 2: has another Johnson "update" that's no more helpful that the others. Well, it's helpful, perhaps, in the sense that it demonstrates conclusively that Johnson may be a card or two short of a full deck. He post the following inexplicable statement:

This is not, as some might argue, a battle over turf. Although the National Counter Terrorism Center has received a legislative mandate to coordinate terrorist issues, managing policy within the bowels of State Department requires someone who is in the building and has the ear of the Secretary of State.

A State Department official commenting on the decision to not publish the statistics on international terrorism said they would let NCTC worry about the statistics and State would do the "policy". Really? How can you formulate a coherent policy if you don't know what the numbers are?

So, just why would the State Department be in the dark about the numbers, just because they no longer publish a glossy book that contains them? Has this guy never heard of the internet? Google? What's more, using compiled numbers is a lot easier if the data are in electronic form, so the utility of posting them inside a think binding just escapse me completely. Honestly, why are people this dumb getting publishing space? [That was uncalled for, wasn't it? -ed Well, now that you mention it, yeah.]

Update 3: I was unaware that Larry was in the neighborhood, so I'm willing to concede that perhaps it's I that's stupid. Maybe we can get to the bottom of this in the comments section. It wouldn't be the first time the thread got away from me.

By Demosophist at 10:30 AM | Comments |