Today’s Must Read

Pentagon officials are angry with the Government Accountability Office, whose recent report disputes that the surge has reduced sectarian violence in Iraq. And with good reason: over the weekend, General David Petraeus boasted to The Australian that sectarian violence was down a staggering 75 percent in Baghdad, a “hugely important” figure he surely wants to trumpet before Congress next week. An anonymous DOD official insisted to the Washington Post that the GAO is “factually incorrect” on this important metric, but provided no evidence for the claim. And that fits a pattern for the Pentagon’s gripes with the GAO.

Repeatedly throughout the document (pdf), the GAO takes a note of caution on sectarian violence, which is the thirteenth benchmark. It says it’s “not clear” whether the violence is in fact down, since “measuring such violence requires understanding the perpetrator’s intent, which may not be known.” Recall that in its July report, the White House plainly stated that “trends supplied over time by [the U.S. military command in Iraq] demonstrate a decrease in sectarian violence, particularly in Baghdad, since the beginning of [the Baghdad security plan, known in Arabic as] Operation Fardh al-Qunun.”

The GAO instead suggests substituting a measurement of total civilian casualties. And there the news isn’t so good: the GAO reports that the “average number of daily attacks remained about the same over the last six months.” A graph, based on data from the U.S. military command in Iraq, shows that from February to July — the lifetime of the surge so far — average daily attacks on civilians hover at a bit higher than 20 per day (click to enlarge):

Now, Petraeus can make a case that the GAO isn’t being fair to him. The GAO is measuring civilian casualties, and by extension sectarian violence all throughout Iraq, while he’s talking about a reduction in sectarian violence in Baghdad alone. Similarly, there are proxy measurements available to tally sectarian violence: binding bodies together are hallmarks of sectarian death-squad killings, as are close-in murders with implements like drills (a favorite of Shiite militias). Often militias or insurgents will leave a victim’s ID in plain view as a way of indicating the victim was killed for his or her background. Certainly attacks on mosques count as sectarian violence.

It’s true that none of these measurements are exact, but the GAO suggests that measuring sectarian violence is a matter of mind-reading. Petraeus will have an opportunity next week to explain how his command tallies sectarianism, and Congress will have an opportunity to inquire how sound his measurements are.

But that’s not the only instance when the Pentagon foisted dubious data on the GAO. According to a critique written by General Mark Kimmitt, the deputy assistant secretary of defense for the Middle East, DOD wanted the GAO not to attribute a decline in readiness for Iraqi Army battalions to “sectarian and militia influences.” GAO refused, citing the Pentagon’s last quarterly report to Congress, which itself attributed the readiness decline to sectarianism and militia infiltration. (We highlighted the Pentagon’s proclivity for quietly revising its quarterly assessments last week.) Indeed, retired Marine General Jim Jones, who led an independent inquiry into the Iraqi police, will report to Congress tomorrow about pervasive sectarianism in the force.

Kimmitt wasn’t done. GAO reported that only 65 percent of Iraqi security forces actually report for duty at any given time. The Pentagon wanted that number raised to 71 percent, citing “accuracy.” Again the GAO refused to change the number, on the grounds that “we do not have documentation for the new figure.”

When I inquired last week about the basis for measuring given security trends — in particular sectarian violence — I was bounced from offices between Baghdad and Washington, and ultimately got no reply. If Kimmitt’s reply to the GAO is any indication, the Pentagon is banking on the prospect that investigators will take its data on face value, no matter how poorly supported or dubious it appears.

More on the guts of the GAO report soon.