THE death tolls of several recent wars may be vast overestimates, claims a review of studies of the conflict that devastated the Democratic Republic of the Congo between 1998 and 2003.
The authors say that the , which includes conflict-related deaths that occurred after the war ended, is at least twice the true number. They suspect that deaths due to the wars in Darfur, Sudan, and Iraq may be similarly inflated.
鈥淚f you keep coming up with tolls that are wrong鈥� you鈥檒l discredit the methodology,鈥� warns of Simon Fraser University in Vancouver, Canada, an author of the report published on 20 January by the .
Advertisement
鈥淚f you keep coming up with death tolls that are wrong鈥� you鈥檒l discredit the methodology鈥�
Mack鈥檚 criticisms focus on a (IRC) in New York. Researchers surveyed clusters of households in the DRC and extrapolated the results to give a national mortality rate. Mack says that the clusters were not representative of the DRC as a whole, and that the baseline pre-war mortality rate used was too low, causing the excess deaths due to the war to appear overly high.
of Columbia University in New York, who took part in the IRC research, says that while surveys of war-torn regions are inevitably less rigorous than ones in peacetime, the IRC figures broadly agree with other surveys.
Echoes of Iraq
The criticisms are reminiscent of the debate over deaths due to the US-led invasion of Iraq in 2003. A claimed that more than half a million people, or 2.5聽per cent of the population, died as a result of the conflict. Several aspects of the study have come under intense scrutiny, including the methods used to select clusters of households.
Roberts worked on both the Iraq and DRC surveys. He says that Mack fails to understand the difficulties of choosing cluster locations in the middle of war zones. The IRC team could operate only in less-violent areas, yet still suffered several safety scares, including four kidnappings. 鈥淚t is relatively weak science, but it was the best we could do.鈥�
Roberts adds that it would have irresponsible not to produce a national figure, as aid agencies would not have grasped the scale of the conflict. 鈥淚f we had not [extrapolated] we鈥檇 have committed a much more egregious sin than that of violating standard academic protocols.鈥�