The Iraqi War Dead
John Hopkins, MIT and Al Mustansiriya University researchers' study (PDF) in The Lancet conflict epidemiology ignited controversy that won't fade away by calculating that 392,979 - 942,636 Iraqis died between March 2003 and June 2006, since the US and UK led coalition invasion. Adding to the confusion, last week two science journals each published their own take on the Lancet study. Science's analysis was called, "Iraqi Death Estimates Called Too High, Methods Faulted", while Nature titled their report "Iraqi Death Toll Withstands Scrutiny". "Toll" or "Estimate", and why such universal disagreement?
Researchers collected data from 1849 households containing 12,801 individuals in 47 cluster points, talking to representative households then extrapolating this information to the whole population. The cluster sampling methodology is well known in epidemiology. Yet critics of the study questioned how the scientists conducted the surveys, whether they conducted them as they said they did, and if the baseline death rate was accurate. They questioned the sampling methodology and perceived biases.
Tthe researchers said that had they gathered 10 times the number of cluster points, the range of the death toll would be three times smaller, however pursuing such precision would have been even more dangerous. Science notes that neighborhood survey methods were "destroyed to prevent them from falling into the wrong hands", which the journal said "infuriated" critics. The John Hopkins team points out that many deaths were verified with death certificates. Obviously, counting deaths in a war zone is difficult, but Science quoted one economist who charged, "It is almost a crime to let it go unchallenged".
The highly emotional response to the study was almost universal. President George Bush vehemently dismissed the report in a press conference. His denial could have been taken for a maladroit blunder. Republican allies shuddered last week when he said that the current Iraq situation is tantamount to the Tet offensive¹ -- it was a mistake to conjure up the "V" word. But his dismissal of the study as "not credible" was deliberate. Predictably then, the National Review called the study "cooked up" and 'motivated by the upcoming election'.
Other criticism seems less about national politics. The site Iraq Body Count (IBC) produced a point by point evaluation of the study. According to their count, deaths to date were between 41,000 and 49,000 -- similar but less than what the Iraqi Ministry of Health counted. Many Iraqi bloggers weighed in of course, some with interesting insight about the methodology and their own estimates.
Naturally, the study stimulated commentary from residents of the country who live with death everyday, as well as from epidemiologists, journalists, statisticians who have professional opinions to share. But thousands of random citizens also weighed in. Responding to a Wall Street Journal article ("655,000 War Dead?"), readers from Germantown, Maryland ("I'm not a fan of..the Iraq war...or President Bush"), and Bedford, New Hampshire wrote letters authoritatively describing their own estimates, based on "plain old common sense" and "simple arithmetic".
Conflict Epidemiology Conflict
All of this arguing makes little sense. No one knows for sure how many died during the Peloponnesian Wars, or as a result of the World Wars, the Asian wars or the current wars. In the Vietnam War, counts for the number of deaths throughout the war and after the war were wildly incorrect. You can pick any number, and it's represented in the pool of statistics. In 1988 Encyclopedia Britannica said that "thousands" of Vietnamese were killed. Today people generally agree that as many as 4 million South Vietnamese and 4 million North Vietnamese deaths can be attributed to the war. However, some say that only 1-2 million were killed.
Why the fuss over the research in Iraq - were the numbers in the media that much more comforting? Do people believe that whomever is left of the media in Iraq -- the bedraggled, kidnapped, shot at, embedded or cloistered in the Green Zone denizens -- have a more accurate bead on the death toll than epidemiologists interviewing residents? And why all the concern with this epidemiology, when people readily accept other research that has a much more immediate effect on their lives? They eat and drink and take medicine; sometimes capriciously; because they at least by default accept the epidemiology and clinical trial results behind their choices.
Obviously, this isn't only about how to tally casualties. Nature, quoting Jana Asher of the Science and Human Rights Program of the American Association for the Advancement of Science in Washington DC, said that such disputes over the numbers were unusual for the field of conflict epidemiology and were fueled by political questions about US reasons for entering Iraq. There is no doubt that politics plays a role in many reactions to the numbers . Governments manipulate perception about deaths during wars according to what it calculates the public will bear.
Casualties in countries that were invaded used to be offhandedly labeled "enemy" deaths. During the war South Vietnamese and American officials exaggerated "enemy counts", while North Vietnamese underestimated their own losses. The strategy of U.S. military commanders was to bomb to the enemy's "breaking point". Many in the US accepted the propaganda -- if the US could inflict the maximum deaths on the enemy, the North Vietnamese would forfeit the war. In one TV appearance in the protracted public relations imbroglio, Secretary of State Dean Rusk marveled at the "resistance and determination"¹ of the North Vietnamese.
During the first Iraq war General H. Norm Schwarzkopf famously made US policy about death counts clear: "I have absolutely no idea what the Iraqi casualties are", he said, "and I tell you, if I have anything to say about it, we're never going to get into the body-count business" (NYT Feb 3, 1991). This was a blunt new tactic to combat more accurate (and grisly reporting) --convince people that a few "collateral" deaths were an acceptable part of the precise, technology driven military tactics.
Certainly, some citizens might choose to believe government figures although they would also need to accept inconvenient inconsistencies, for instance that the US government prohibits caskets from being shown, promotes torture for information gathering, but is not motivated to lie about the death toll. But are people apoplectic about Lancet's new report because they were habituated to previous reports of 30,000 to 150,000 deaths? Is 600,000 deaths a political crisis, but not 150,000 deaths? I think it's cynical to think the only reason for all the public strife is politics.
Conflict Epidemiology: Promising New Field
While politicians try block access to grim and grisly details of war, accurate casualty reports could help keep us more informed. If the governments' public policies wrack havoc, we should be accurately informed on the extent of that havoc -- it's not only our right and our politics, it's our reputation and our conscience. It's not just about the methodology or the politics. The discrepancies rattle us. There are people who simply care whether 100,000 or 300,000 or 600,000 people died -- that's a lot of people, a lot of lives. The world is a closer place. Perhaps globalization influences people's perceptions of war fatalities as much as it does business, making wars and their dead not so far away, so removed. As Margaret Drabble wrote:
"Between" "12 and 32 million killed" is a phrase that cannot exist [...] What kind of history what kind of mathematics is this, what has happened to those spare tens of millions? Unnumbered, unburied, will they haunt the earth for ever, will they ever find a resting place? Do they not jostle us, do they not stifle us, are we not kept awake at nights by their...cries? Margaret Drabble (1991³).
The ongoing debate is about many things such as the politics surrounding the invasion, concern on all sides that public opinion will turn one way or another, growing realization that the government blatantly obfuscates the facts, the troubling reality that peoples deaths, in fact their lives, go miscounted, uncounted and discounted, as well as questions about whose to blame for it all. This raises the stakes for official death counts, and the demand for epidemiologists. Nature predicts that "the study will be a key publication in the growing field of conflict epidemiology."
As such, scientists can attribute some optimistic news -- "a growing field" -- to the burgeoning military-industrial complex4;. Nature reports that the field has expanded in the past 15 years because of better methodology [and more conflict?]. Apparently there's no end in sight. Quoting Jana Asher the journal notes that the numbers could be used be truth and reconciliation councils who could add quantitative data to personal accounts to help them punish war criminals and help countries come to terms with history.
In addition to reconciliation councils, there are immediate needs for the studies. People make up all kinds of reasons why the US needs to stay in the Iraq. Bush may want to get rid of the "insurgents", but Democrats and Republicans both think they're being reasonable when they argue "humanitarian" reasons for staying in the war. The US should stay in Iraq, because if US left, the country would break into civil war, they often say. David Brooks said it again last Friday on the PBS NewsHour. We can't leave, Iraq, he said:
"...the fundamental debate that we've had for the past year, that we all want to get out, but if you get out and there are 200,000 people killed in a genocidal civil war, what does that leave you? And so that's always been the series of bad choices we have.."
It sounds so reasonable. Truly, most experts will admit that there doesn't appear to be and "good", option, and that any option might have a "bad" outcome. But the US won't *prevent* 200,000 deaths by staying -- especially without a better strategy. No matter what you think of the decisions to invade Iraq, whether you believe that the US caused all the casualties, how many casualties you believe there are, or what you understand about the real situation in Iraq5, too many people have already died. We may not know whether 100,000 or 900,000 people have died, but we don't believe it was only 30,000. Would the study have been more useful if it were more precise? Probably, but then there might have been more dead too.
--------------------------
1. Perhaps he thought the host, George Stephanopoulous, was asking about his change of strategy...from the gut, to the tête?
2. Secretary of State Dean Rusk, NBC-TV interview, July 2,1971, in John E. Mueller,"The Search for the "Breaking Point" in Vietnam, The Statistics of a Deadly Quarrel" International Studies Quarterly, Vol 24 No. 4. Dec. 1980
3. Drabble, M. 1991 The Gates of Ivory. London: Penguin Books. Quoted from in 'Between one and three million': Towards the demographic reconstruction of a decade of Cambodian history (1970-79), Patrick Heuveline Population Studies, 52 (1998), 49-65.
4. The new field offers excitement, world travel, the opportunity to meet new people in foreign lands. Timid scientists marching around their labs might shudder, but the danger can be tempered. For the Lancet study, US based researchers met their Iraqi colleagues only twice during the study, in Jordan. It's vital work with invigorating peer interaction. You will be assailed, dismissed, challenged refuted, and attacked (in more ways than you think). So maybe it's not for those who want to be **rockstars**, but it is critical work.
5. Rory Stewart wrote an interesting perspective on the coalition challenges in The Prince of Marshes.
Leave a comment