10-19-06, 9:13 am
George W. Bush, the expert statistician, apparently considers the numbers not “credible,” and believes that the researchers’ methodology is “pretty well discredited.” Somehow, I doubt that his aides explained to him what a clustered random household survey is, let alone log-linear regression – somehow, I doubt he has any aides who know.
His assessment was corroborated by epidemiological expert General George Casey. Other dismissive hacks included Anthony Cordesman, who understands military issues but little else, and Michael O’Hanlon of the Brookings Institute.
On the other hand, no actual expert in epidemiology who was asked has dismissed the findings (and, obviously, it passed peer review in the Lancet, the premier British medical journal), and Human Rights Watch has said that this is probably the best result out there.
It is true that the numbers are vastly different from other numbers you see out there. Iraq Body Count, which combs media reports thoroughly and hospital and morgue reports sporadically, and only counts civilian deaths, counts 44,000-49,000; the Los Angeles Times, which did a rather better job, looking at statistics from the Baghdad morgue, the Iraq Health Ministry, and other agencies, estimated that at least 50,000 Iraqis had died due to violence. Unlike the IBC, which viciously attacks anyone who comes up with higher numbers, the LA Times story acknowledged that its number was likely a serious undercount.
The reason for these strong divergences is simple. Suppose, say, you were trying to estimate the number of rapes in the United States. Even if you did a very thorough job combing police station and court reports, you would come up with a tremendous undercount. Surveying a random sample of people would likely give you a much more accurate estimate. In the United States, death – especially violent death – doesn’t work the same way. Virtually all of it gets reported and logged by some government agency; even here, though, trying to estimate deaths due to inadequate health-care from media and hospital reports would be a hopeless and ridiculous task.
In Iraq, by contrast, if you live in a small town in al-Anbar province and your family got hit by a U.S. airstrike, you bury them by sundown and go on. You don’t hit the AP speed-dial on your cell-phone and get a story in the papers; you don’t go to a useless government agency, even if you could find one; you don’t take dead people to a hospital. And you don’t get counted.
As Les Roberts and collaborators explain in their article, “passive” methods based on reports or monitoring how many bodies show up at a given location usually give dramatic undercounts in Third World countries wracked by violence – during the worst years in Guatemala, newspaper reports captured bout 5% of total deaths.
The survey has very high margins of error, because, although 1849 households were sampled, they had to be sampled in a total of 47 different randomly chosen clusters. Even the low end of the estimate, however, is just shy of 400,000, far greater than any other number yet produced. The high end is over 900,000.
The truth is that this study is the standard by which all the other numbers should be judged. Its conclusions are stunning -- one out of every 40 Iraqis, 2.5%, have died since the occupation started. In the 15-year Lebanese Civil War, about 150,000 people died. Iraq has 9 times the population, but it’s only been three years.
Of equal import, 31% of violent deaths were caused by the occupying forces – 186,000. Even during the period June 2005-June 2006, 26% of violent deaths were caused by the occupiers. The occupiers continue to be a major part of the problem, not, as recent news reports seem to suggest, an ineffectual part of the solution.
From Empire Notes
--Rahul Mahajan is publisher of the blog Empire Notes and occasionally teaches at New York University. He has been to Iraq twice and reported from Fallujah during the siege in April 2004. He has been published widely, including in USA Today, Newsday, the Baltimore Sun, the Times of India, and the Jordan Times. His first book, 'The New Crusade: America's War on Terrorism' (April 2002, Monthly Review Press), has been described as 'mandatory reading for anyone who wants to get a handle on the war on terrorism.' His second book, 'Full Spectrum Dominance: U.S. Power in Iraq and Beyond' (June 2003, Seven Stories Press), is a wide-ranging look at the war on Iraq, the plans of the Project for a New American Century, and the Bush administration's imperial policies in practice since 9/11. He also does a weekly commentary, posted on his website and also aired on KPFK's Uprising Radio, and gives frequent radio and television interviews. Mahajan is a long-time activist and currently serves on the Steering Committee of United for Peace and Justice and the board of Austin's Third Coast Activist Resource Center. He can be reached at
