It's sometimes fun to wonder whether, knowing all we know today, we'd fall for another version of the Iraq/WMD bamboozle if another came down the pike. I'm afraid the answer has to be: absolutely
Look no further than the present debate about the success of the 'surge'. I think Karen DeYoung's piece
in today's Post
-- regrettably on A16 -- settles once and for all that the numbers we're hearing are basically a scam.
It's worth beginning by noting what appears to be the universal consensus that the strategic aim of the surge -- political reconciliation -- has been a complete flop. No progress and things have gotten much worse. That leaves a debate about tactical successes, which for better or worse, we're judging by various body counts. As I've struggled to get my head around this discussion I've looked -- mainly in vain -- for numbers going back some period of time with a consistent methodology since an apples to apples comparison over some period of time is the only way to make any sort of reliable judgments about change, improvement or decline.
What comes up again and again though is one basic disconnect -- the military command in Baghdad says civilian casualties have dropped dramatically. Independent press tabulations say the numbers are high and getting higher.
DeYoung's article gives us a couple bits of information that help us start to unravel the mystery. First, the military command in Baghdad is in a spat with the GAO, which the generals accuse of using a flawed methodology. (GAO's analysis basically disagreed with them on all particulars.) DeYoung's piece includes the very telling detail that the GAO is using the same methodology that the CIA and the DIA favor. So it would seem that it's not only a question of the government versus outside observers. The military command in Baghdad sounds like it's completely isolated even within the US government
on how to compute the numbers.
That doesn't inspire a lot of confidence.
We also learn from DeYoung's article that as a basic matter of categorization, the Petraeus/White House numbers don't include the deaths of people killed by our friends (new Sunni allies in al Anbar). They don't include deaths of people killed by members of their own sect (Sunni-on-Sunni, Shia-on-Shia, etc.). They count or don't count based on things like where a person has been shot in the head.
One intelligence analyst told DeYoung, "If a bullet went through the back of the head, it's sectarian. If it went through the front, it's criminal."
It's a little difficult to tell from the immediate context of the quote whether there's a little embellishment or whether that's literally true in every case about the methodology being used. But taken together what we can glean about the methodology -- which I take it is itself classified -- is that it is a classic case of presupposing the result in the methodology itself. DeYoung actually has a good quote in her piece from the Iraq Study Group that concisely explains the problem: "Good policy is difficult to make when information is systematically collected in a way that minimizes its discrepancy with policy goals."
You decide that the 'problem' is 'sectarian violence' -- not an improbable analysis in itself. Then you systematically devise a manner of counting that cuts away all sorts of carnage and mayhem that doesn't fit within an extremely narrow interpretation of the 'problem.' And voila, things are improving.
Frankly, it reminds me, painfully, of the rush of polling analyses prior to the 2004 election which showed, quite convincingly, that the numbers actually showed Kerry winning. You just had to properly weight them for Party ID!
The truth is that once you take a screwdriver and saw to the numbers smart people can come up with a lot of ways to fool others and
themselves, in many ways all the more so if the people doing the counting have a highly articulated theory behind their reasoning.
Whether this is deception or self-deception is a minor subtheme to the story. The headline is that these numbers appear to be a joke.