Counting Civilian Casualties   

Media organizations are relying on a little-known group that started in England for the best estimate of civilian deaths in Iraq.

In articles about the civilian toll of the Iraq war, Gen. Tommy Franks, former commander-in-chief of the U.S. Central Command, is frequently quoted as having said, "We don't do body counts." His words are often followed by the latest numbers from an organization that does: Iraq Body Count, a group that has tracked thousands of civilian deaths based on media reports.

What started as the personal project of a group of writers, professors and peace activists in the United Kingdom and the United States, all vehemently opposed to the war, has become the primary source of civilian casualty numbers for many national and international news organizations, including the Washington Post, United Press International and the Times of London.

Why Iraq Body Count? Mainly because there aren't many other sources to choose from.

In the absence of government data or comprehensive media studies, news outlets wanting to give some indication of how civilians have suffered have turned to this little-known band of volunteers.

The Brookings Institution's Iraq Index provides a monthly estimate, but it's a modified version of Iraq Body Count's numbers. Then there's the estimate of a Johns Hopkins research team whose report was published by the Lancet medical journal in October 2004. Using a random sampling technique, the researchers estimated about 100,000 civilian deaths.

No group besides Iraq Body Count provides a running tally. At press time, the organization reported a minimum of 22,850 and a maximum of 25,881 Iraqi civilian deaths resulting from the invasion and occupation of Iraq.

Brookings provides two ranges for deaths since March 19, 2003: a lower estimate of 12,700 to 23,000 and a higher estimate, which includes Iraqis killed by acts of crime, of 29,700 to 60,800. For the first range, Brookings uses Iraq Body Count's reports for the lower figure and calculates the higher number based on discrepancies between Iraq Body Count's figures during major combat operations and statements by the Iraqi Ministry of the Interior. Modifications aside, Adriana Lins de Albuquerque, senior research analyst at Brookings, says the group provides a good source of data, adding, "Their methodology seems sound."

For more than a year, the Washington Post usually has run a weekly two-column Iraq war graphic that lists both military and civilian casualty numbers, using Iraq Body Count's estimates for the latter. Assistant Managing Editor for Foreign News David Hoffman says the paper began including civilian casualty numbers because readers were asking for them. "We realize that there is no perfect source here," he says, but Iraq Body Count was the best choice because it uses published, archived information.

Founders John Sloboda, a psychology professor at Keele University in Staffordshire, England, and Hamit Dardagan, a freelance researcher in London, joined with friends, colleagues and acquaintances to form Iraq Body Count in January 2003. Sloboda and Dardagan decided to apply the methodology of a similar study of Afghan civilian deaths by Marc Herold, a professor at the University of New Hampshire (see "The Civilian Casualty Conundrum," April 2002). The Web site, www.iraqbodycount.org, made its debut in April 2003, shortly after the invasion.

Sloboda, Dardagan and their team make no effort to hide their opposition to the war. ("However many civilians are killed in the onslaught on Iraq, their death toll should not go unnoticed by those who are paying — in taxes — for their slaughter," the Web site states.) But to critics of their politics, they respond that their research can be verified. "If our methodology is transparent and public, then our political views are irrelevant. Any third party can assess the data for him- or herself," the founders wrote in a chapter of the book "The Iraq War and Democratic Politics."

Here's how it works: An Iraq Body Count researcher tracks civilian deaths reported by online news sources that range from the Associated Press and Reuters to Qatar-based Al Jazeera and Washington, D.C.'s Middle East Report, using comprehensive search engines. Deaths directly or indirectly resulting from coalition military intervention since January 2003 are recorded after they are confirmed by at least two credible news sources. To ensure that deaths are not counted more than once and to acknowledge the uncertainty of the numbers and the potential political bias of news organizations, a minimum and maximum number are recorded when there is a discrepancy.

The range is used if at least two news sources have published a lower number of deaths than other accounts of a specific incident. The results are reviewed by three members of the 12-person team.

Additionally, all news sources must meet certain criteria: Each Web site must be updated daily and available in English; articles must be archived with distinct URLs; the sites must be fully public and widely referenced in other media.

Considering the time and manpower necessary for papers to track civilian deaths, relying on an outside source is "really the only way of doing it," says Dita Smith, the Washington Post's assistant foreign editor for graphics, who organizes the paper's casualty figures each week. She and several researchers tried to track civilian deaths in wire reports for a week, but since some reports were duplicates and others were updated over time, she says, "it was simply too vague."

"We decided it would be better to go with an organization that does nothing but count bodies," Smith says. "It would be nice to have something more precise, but that's just impossible."

Indeed, Andrew Ross, the San Francisco Chronicle's executive foreign and national editor, doubted his newspaper could come up with a more accurate count, "given the extraordinary difficulties of reporting from the ground in a country, much of which is a 'no-go' area," Ross wrote in an e-mail interview. The Chronicle has published stories about the issue of civilian casualties infrequently in the past year, about four times, he says, in part because the figures aren't precise. The paper has cited Iraq Body Count numbers at least four times since the beginning of the war. Other sources' estimates are referenced as well, but "iraqbodycount.org is one of the few organizations out there, at least that I'm aware of, that [does] try to keep track in some fashion," he wrote.

Founders Sloboda and Dardagan say their numbers are likely a low estimate because of incomplete reporting. Still, Sloboda hopes his group, which released a more detailed report on deaths and injuries in mid-July, has set a precedent for tracking civilian casualties in future conflicts. "We would wish to see it as part of international law that all warring parties must always report to the U.N. Security Council with their best estimates of the numbers of civilians killed in any conflict in which they are a part," he wrote in an e-mail interview.

For perspective, the Post's Smith includes footnotes with the weekly casualty graphic, one citing the U.S. military's policy about civilian casualties and another citing the Lancet estimate.

Other editors feel the uncertainty of the numbers is reason enough to leave them out altogether. Chicago Tribune Foreign Editor Kerry Luft says his paper doesn't include exact civilian casualty figures regularly because the range of estimates out there is too large and it's difficult to determine their accuracy. "I think it's just much safer to say, 'Clearly the numbers are in the thousands and the estimates vary widely,'" he says.

But is an estimate better than none at all? William M. Arkin, a military analyst for NBC News and former senior adviser for the New York-based Human Rights Watch, calls Iraq Body Count's numbers a "vast underestimate" in light of the Johns Hopkins study, the incomplete results of a study known as Civic conducted by the late activist Marla Ruzicka and Arkin's own personal experience conducting civilian casualty assessments in Iraq, Afghanistan, Lebanon and Yugoslavia. He's working on a study titled "Why Civilians Die" for Harvard's Carr Center for Human Rights Policy. Since, in his experience, successful data collection is dependent on government-provided information, he questions the purpose of including Iraq Body Count numbers and doubts they have much impact on public opinion.

Using published, archived information does not necessarily mean the data are bias-free, Arkin says. "You're talking to somebody who opposed the war as well, and I don't question that faulty reporting is documented. That still makes it faulty reporting."

Arkin thinks the media should make it clear that these numbers come from an antiwar group and that they're not the official count. "I'm glad that [Iraq Body Count is] doing something that somebody else isn't doing, but it does not equal a body count," he says. "It merely equals the best body count that is available."