THE O’HANLON/POLLACK REPORT….I mentioned yesterday that I had been reading the report prepared by Michael O’Hanlon and Kenneth Pollack after their recent trip to Iraq. What I was looking for was concrete data related to improved security, as opposed to subjective impressions about morale and improved COIN efficiency.
This is not because concrete metrics are the only way to measure progress. Far from it. But it’s easy to kid yourself about this stuff, especially on a trip in which you travel in a cocoon of official Pentagon representatives. O&P, for example, found a “significant improvement in the morale of American forces.” But Tina Susman of the LA Times reports today that “signs of frustration and of flagging morale are unmistakable.” Who’s right? To be honest, I found Susman’s piece fairly weak. But at least she presented some evidence for her case, whereas O&P merely made an assertion that our troops were happier and then moved on. In the end, though, it hardly matters. Even with the best of intentions this kind of reporting is both inherently unreliable and notoriously susceptible to both conscious and unconscious confirmation bias.
The value of metrics, then, is to keep yourself honest. It’s easy to get carried away with subjective judgments, and what’s more, even if morale really is improving and U.S. troops really are embracing a smarter approach to counterinsurgency, that doesn’t matter unless it’s making an actual, tangible difference. At some point this stuff has to translate into fewer deaths, fewer car bombs, an insurgency on the run, etc. So what measurable improvements do O&P offer up? I could find only two:
“Car and truck bombs are…often less powerful than before.”
“Roughly a one-third decline in the monthly rate [of civilian casualties] since just before the surge began.”
The first point is almost certainly wrong. The data is here, and bombings don’t appear to be getting less lethal at all. Quite the opposite, in fact.
As for the one-third reduction in civilian casualties, who knows? There’s no footnote to explain how they’ve come to this conclusion. The Brookings Iraq Index (author: Michael O’Hanlon) reports a drop in civilian casualties but unhelpfully explains the data only as “estimates provided by the authors.” In his Washington Post op-ed today, O’Hanlon says merely that “the Pentagon showed us data” indicating a drop in casualties. Conversely, Leila Fadel reported last week that “statistics gathered by McClatchy Newspapers don’t support the claim.” The Christian Science Monitor reported earlier this month that Iraqi government death tolls are unreliable and AP reported that July fatalities were 23% higher than in June. What’s more, even if casualties have dropped, without regional breakdowns there’s no way of knowing how much of the decrease is due to places like Anbar and Ninawah — where everyone agrees violence has declined — and how much is due to actual surge-related progress in Baghdad and elsewhere.
Bottom line: O’Hanlon and Pollack cite only two concrete security metrics, and of those, one appears to be flatly wrong and the other is unsubstantiated and highly doubtful. Instead we get lots of phrases like “signs of progress,” “appear to be reducing,” and “our observations suggest.” To say the least, this is an unpromising track record.
It’s possible that the Iraqi Army really is getting better (though see here and here for contrary evidence from the ground); that U.S. troops really have embraced proper counterinsurgency techniques at lightning speed; and that provincial reconstruction teams are doing some good work. But it’s also true that Iraq’s infrastructure continues to decline; the police are “still a disaster”; Iraq’s economy “remains largely moribund”; unemployment is “sky high”; and among the political leadership the situation is “awful,” “worse than stalemate,” and “paranoia and backstabbing predominate.” And this is without even mentioning the upcoming election in Kirkuk or the brewing intra-Shiite smackdown in southern Iraq.
Given all that, O’Hanlon’s entreaty in the Post today that we should believe him even in the face of spotty and unreliable evidence because “Our assessments are based on our observations as well as on years of study” — well, that’s pretty weak tea, isn’t it? Considering how disastrous the political situation is, how poorly the infrastructure and the economy are doing, how often we’ve been assured of progress in the past, and the fact that most security metrics indicate that Iraq is doing worse this summer than last, I think it’s fair to ask O’Hanlon and Pollack for more evidence of progress than just regurgitation of talking points from the military brass they traveled with. Whining about how unfairly they’re being treated is a poor substitute for the healthy skepticism they should have displayed in the first place.