ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

The Dead in Iraq and the War of Numbers

Two different studies on mortality in Iraq after the 2003 US invasion - by Iraq Body Count and Johns Hopkins School of Public Health - provide vastly different assessments.

Letter from America

The Dead in Iraq and theWar of Numbers

Two different studies on mortality in Iraq after the 2003 US invasion – by Iraq Body Count and Johns Hopkins School of Public Health – provide vastly different assessments.


mong the many wars taking placein and around Iraq – the Americanwar on (or of) terror, the warbetween the occupation forces and Iraqiinsurgents and civilians, the war relentlessly being waged on innocent civiliansby a wide array of armed forces, the strifebetween sunnis and shias, the war by mediagroups to gain the attention of the world,to enumerate only the most obvious formsof an escalating conflict that shows littlesign of diminishment – the war of numbersis shaping up as an important part of theconflict. The daily barrage of the dead issuch that a long-term perspective on casualties may appear to be something of aluxury. Since the American invasion ofIraq in March 2003, the independentorganisation Iraq Body Count ( has been maintaininga tally of “civilian deaths in Iraq thathave resulted from the 2003 military intervention by the USA and its allies”, and itscount includes civilians who have died at the hands of US forces, other coalition troops, Iraqi security forces, insurgents,and all other paramilitary organisations.As of October 27, 2006, the Iraq BodyCount gave a maximum total of civiliandeaths of 49,697.

Excess Deaths and Its Assessment

However, a recent and apparently credible study undertaken by the highlyrespected Johns Hopkins School of PublicHealth, whose researchers worked in collaboration with Iraqi medical doctorsaffiliated to the School of Medicine at Baghdad’s Al Mustansiriya University, hasestimated that through July 2006, therehave been 6,54,965 fatalities since the invasion.1 These are described as “excess deaths”, or fatalities above the preinvasion death rate. How is it possible that two widely cited figures on Iraqi fatalitiescould be so hugely discrepant? Which studyis more credible, or is it the case that, both studies being flawed, the “truth” lies somewhere in the middle? It surely matters howmany Iraqis have been killed since theinvasion, and not only because one studyfurnishes a number that is more than 10 times greater than the other number, andthe human thirst for truth cannot apparently be satisfied by both numbers. Itmatters because the larger number, if trueor at least a more reliable approximationof the truth, would suggest that an additional 2.5 per cent of Iraq’s population hasbeen wiped out in little over three years,and such an astronomical loss of lives cannot but have far-reaching demographicconsequences. The Johns Hopkins studynotes that 6,01,027 of the “excess deaths” occurred due to violent causes, and that the victims are largely male, aged 15-44 years.

A set of comparisons comes to mind:over the six years that the second worldwar lasted, the UK lost 0.94 per cent ofits population, even though London wassubjected to relentless bombing over thecourse of months, and China and the US lost 1.89 per cent and 0.32 per cent of theirpopulation, respectively. (The US onlyentered the war after the attack on Pearl Harbour in December 1941.)

The Korean war suggests a more chillingcomparison: though the precise number ofcivilian casualties is a matter of some dispute, with estimates varying from 1.8million to nearly four million fatalities, thematter-of-fact assessment of general CurtisLeMay, the architect of the fire-bombingof Tokyo in March 1945, casts a rathermore ominous light on the catastrophefacing Iraq: “Over a period of three yearsor so we killed off – what – 20 per centof the population”. Those who describeNorth Korea as a “rogue state” have notprobed whether the country’s presentdilapidated state might not have somerelationship to the brutalisation of its people by a foreign imperialist power. Iraq, onefears, is not far from being reduced to asimilar state of paupery.

Johns Hopkins Study

The Johns Hopkins study, expectedly,has attracted some criticism, and the American administration officials immediately dismissed it as “unreliable”. Ofcourse, “unreliable” here signifies nothingat all, unless it be the well known incapacity of officials in the Bush administration, for whom the example has been setby the commander-in-chief himself, toadmit that Iraq has been turned into aninferno. The authors of the study, who aremedical practitioners, cannot be accusedof being politically motivated, though infact, doctors and other health professionalshave every reason to feel aggrieved. Thehealth system in Iraq has disintegrated, andthere are documented instances of doctors and nurses being targeted by killers toprevent them from healing the woundedamong the enemy. There were 34,000physicians in Iraq before the invasion of2003, and already by mid-2004 some 2,000had been murdered while another 12,000 had emigrated. Even the most inexpensivemedicines and medical supplies, such asrehydration tablets, disposable needlesand plastic masks, are in acutely shortsupply. Some Iraqi health care specialistshave given it as their opinion that Iraqideaths could have been halved if adequatemedical facilities had been available to care for the wounded. On the other hand, the American fatality rate has never beenso low, and American soldiers who would have died of their present wounds in anyprevious war have been given a new leaseon life by the most advanced system ofemergency medicine brought to the battlefield and to military hospitals.

Iraq Body Count

Since the founder of Iraq Body Countand the authors of the Johns Hopkins studyhave alike expressed confidence in theirresults, it is instructive to understand whythey might have arrived at such differentconclusions. Iraq Body Count attempts toverify every report of fatalities, but, inwhat is doubtless a critical shortcoming,it relies entirely on online media reports“from recognised sources” and eyewitnessnarratives for its information. If a fatal incident has not been seen or witnessed,

Economic and Political Weekly December 9, 2006

the victim is unlikely to end up as a statisticand is consigned to utter oblivion. IraqBody Count’s researchers admit that theirtally does not account for all war-relatedcivilian deaths, but nonetheless they insistthat their mortality statistics still furnishthe most reliable data for Iraqi civiliancasualties. The Johns Hopkins study, incontrast, is described by its initiators as“the only population-based assessment offatalities in Iraq during the war”. Following on the heels of an earlier study in 2004,which was faulted for securing a samplingbase of only 1,000 families and having alarge margin of error, the 2006 survey wasmore exhaustive and used a standard cluster survey method ordinarily used to measuremortality in conflict situations. The surveymembers covered 16 of the 18 administrative districts in Iraq, and selected at random50 sites for their survey, and 40 householdsat each site. Every household thus had anequal chance of being included, and eventually data from 1,849 households wasincluded.

Cluster Survey Method

While the authors of the Johns Hopkinsstudy scarcely claim infallibility, their studyis accompanied by a robust defence of themethods they deployed to study mortalityin Iraq. The cluster survey method relieson random sampling, except that, in situations of extreme conflict where the listingof all persons or households becomes nearlyimpossible, it involves random selectionof clusters of people or households ratherthan individual people. They acknowledgethat researchers encountered numerous difficulties, having to overcome US military checkpoints, and occasionally somesuspicion in the homes that they visited.Each survey team consisted of two men and two women, all Iraqis, medical doctorsand fluent in English and Arabic. Eachresearcher was trained in the use of the questionnaire. The pre-invasion and postinvasion mortality rates were compared,and in every household where the deathof a member was reported to the surveyteam, the interviewees were asked to presenta copy of the death certificate of theirrelative. In 92 per cent of such instances,the study reports, “a death certificate waspresent”. Nonetheless, sceptics might feelentitled to entertain the usual doubts – the size and representativeness of the sample,the quality of survey techniques, the rapport between the researcher and the interviewee – about such sampling studies. Nordoes the thoroughness of the study preclude some obvious questions: for instance,could some who have been counted amongthe dead simply have left the country?Were household members likely to invent deaths, either out of bitterness or in the hope that this would entitle them to somecompensation?

Alarming Statistics

The findings of the Johns Hopkins studyare, to say the least, stunning. Though thewhole number of 6,54,965 excess deaths, of which 53,938 were due to non-violent causes, is the most alarming statistic toemerge from the study, some of the otherkey findings confirm the perception that theconflict in Iraq has sharply escalated overthe course of the last 18 months. The crude mortality rates tell their own depressingstory. In the 12 months preceding the invasion of March 2003, the death rate was

5.5 deaths per 1,000 per year; in the 12months following the invasion, it had risento 7.5 deaths per 1,000. From May 2004-05,it again rose to 10.9 deaths per 1,000, butin the subsequent 12 months, ending in June2006, it nearly doubled to 19.8 deaths per1000, or almost quadruple of the baselinecrude death rate of 5.5 deaths per 1,000 inthe pre-invasion period. At the present rate,over 900 people are dying from violencedaily, and over 50 per cent of the violentdeaths can be attributed to gunshots. Thesefigures, as the authors of the study realise,might not seem congruent with the figuresappearing in the newspapers and online mediareports. But, as they point out, the passivesurveillance methods to measure mortality,such as visits to morgues and reliance onmedia reports, have not been shown toidentify more than 20 per cent of the deathsin other major conflict situations such asKosovo, the Democratic Republic of Congoand Darfur.

Whatever one’s own political inclinations,profound ethical questions arise from suchstudies. In the midst of the disagreements, itis easy to forget that the violent death of asingle person diminishes everyone.Mohandas Gandhi understood this, not merely as something of an abstract idea, butas an ethical stance that forbids us from being seduced by the calculus of numbersand the attendant geopolitical considerations.In February 1922, at the height of the noncooperation movement which had been themost concerted opposition to British rulesince the rebellion of 1857-58 and by theadmission of some colonial officials had succeeded in paralysing British administration in various places, Gandhi invokedhis authority to put the entire movement intosuspension. Some 20 odd Indian policemenserving under the raj had been killed atChauri Chaura by angry crowds, ironicallymany of them acting in the name of theMahatma, and Gandhi was persuaded thatthe country was not adequately prepared tooffer non-violent resistance. To his many critics, who chafed at Gandhi’s authoritarianism, suspected him of subservience tothe regime of law and order, and felt enragedthat he had squandered the opportunity tobring the British to their knees, Gandhireplied unswervingly: With what wordscould he console the bereaved widows of the policemen, and with what countenancecould he look them in the eye and justifythose deaths as necessary in the name ofsome higher political good?

Yankee Individualism

The Iraq Body Count’s web site is prefaced in bold with a remark by generalTommy Franks, US central command: “Wedon’t do body counts”. More accurately, itis the Iraqi bodies that are not counted, sincethe bodies of dead American soldiers are, by contrast, a matter of obsessive concern.The death toll of American soldiers, listed by name, rank, regiment or division, andplace of origin appears across the countrywith numbing regularity in the New York Times, numerous other newspapers andweb sites, and public memorials. Even indeath, the American retains his, so to speak,rugged Yankee individualism. The Iraqisappear today, as they have in countlessother narratives, en masse, heaped together,always as part of a collectivity. One mustwonder whether it is dead Iraqis that arenow being fought over, or whether theterrain of conflict, as far as mortality inIraq is a question, has shifted to thequestion of what constitutes “science”.

That two studies, both grounded in whatappear to their researchers to be “objective”measures and animated by the concern tofulfil a moral obligation to the Iraqi peopleand to posterity, have come to such vastlydifferent assessments of mortality in Iraqsuggest that we must revisit, to invoke thephrase of the historian of science, TedPorter, our “trust in numbers”. Modern science replaced the certitudes of one agewith another set of certitudes, but as the present controversies over death and destruction in Iraq indubitably demonstrate, ouruncertainties are not likely to be resolved bymore accurate studies or more empiricalresearch. Not all narratives are equally compelling, but it also appears that one of themany insights to be gained from the disputesover mortality in post-invasion Iraq is thatin the onus to lead an ethical and politicallyaware life, science may not offer muchmore assistance than did religion.




1 Gilbert Burnham et al, The Human Cost of the War in Iraq: A Mortality Study, 2002-2006

Baltimore, Baghdad and Cambridge, Mass, 2006.

Economic and Political Weekly December 9, 2006

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top