From the Inbox
In reference to CFR.org’s piece, “The Iraq Data Debate: Civilian Casualties from 2006 to 2007,” reader Glenn Kutler writes:
I commend Messrs. Biddle and Friedman for attempting to bring some clarity to the data concerning Iraqi civilian casualties. The analysis would be strengthened, I believe, by considering the following:
Donald Rumsfeld famously said in November 2003, “We don’t do body counts.” The U.S. military knows full well that such reports are suspect, highly charged and potentially misleading. The principal difficulty is independent confirmation. Although Rumsfeld was referring to enemy fatalities, the same concerns apply to civilians. The convergence of data from many sources since April 2007 is thus truly remarkable, and I believe probably means that we can use these numbers with confidence. I would be interested in the authors’ views on this subject.
The convergence of this data corresponds to the disclosure in a UNAMI report of the Maliki government’s policy that it was ending its practice of releasing its own counts of Iraqi civilian casualties. It would have been useful if the authors addressed this policy and examined the influence of the Iraqi government on casualty counts before then.
One thing seems clear: After the Iraqis stopped releasing data, news organizations were forced to go to other sources and apply their own journalistic standards to verify casualty reports. The impact of this journalistic effort needs to be addressed, as does the remarkable fact that the journalistic sources and the U.S. military numbers are so similar. Could the military be relying on journalistic sources, or vice versa?
The problem in identifying casualty trends appears to lie not with the 2007 numbers, but with the prior year data. The numbers before April 2007 are, as the authors indicate, very noisy, and the absence of convergence of the various sources confirms that the numbers are suspect. One reason, at least in iCasualties case, is that the totals reflect unusual casualty reports from the UN in August and September 2006 (which added 1,763 and 1,960 fatalities, respectively, to the totals as presented) and from the U.S. military in February, March and April 2007 (which added 1,483, 1,088, and 142 respectively). It is thus not surprising that in a comparison from summer 2007 to 2006 , or from late surge (after April 2007) to early surge, progress seems to be evident. But it is certainly curious that these types of inflated reports have been intermittent, and by causing prior period numbers to increase, they make current results look good.
In summary, looking back at trends in Iraqi fatalities before April 2007 is fraught with difficulties. But looking at data since then, especially in September 2007 (which is of course beyond the scope of the report) we can see an intriguing shift in the pattern of hostilities in Iraq:
September is on track to have under 1,000 Iraqi fatalities for the first time in over a year.
September is also on track to have the lowest level of U.S. fatalities in over a year. This pattern represents an apparent reduction in the intensity of hostilities in Iraq. This is especially noteworthy because September corresponds largely with the Muslim holy month of Ramadan. In Years Past, Ramadan has Produced U.S. Fatalities at Higher Average Rates (3 Per Day) Than Other Times (2 Per Day). During the Surge, this Pattern has been Reversed. During the Troop Surge Buildup, from February to June 2007, U.S. Fatalities Averaged 3.2 Per Day, And During the Surge of Operations from June Until the Start of Ramadan (9/12/07) U.S. Fatalities Averaged 2.9 Per Day. But During the First 14 Days of Ramadan, Through 9/25, U.S. Fatalities are down almost by half to near 1.5 Per Day.The question is: are the current declines sustainable results of the surge? Or have insurgents simply retreated with plans for high intensity reengagement later, or diverted some of their energies in other directions (see the New York Times article today entitled “Sunni Insurgents in New Campaign to Kill Officials”)?
By the time of Petraeus’ next report in March 2008, it may still be difficult to make a convincing assessment.
–Glenn Kutler
