The Washington Post Undercounts Iraq Deaths

The Washington Post's
weekly Saturday feature on "Iraq War Casualties" has consistently
listed a "maximum count" of Iraqi civilian deaths that is dramatically
lower than the likely civilian death tolls assessed through surveys of
the Iraqi public.

In the most recent edition of this feature (10/25/08) which the Post has been publishing as a chart in the Saturday newspaper since August 2, the Post offers a "maximum count" of 96,719 Iraqi civilian deaths. Yet as the Post itself acknowledged in a footnote to its chart on June 15, 2007, there are studies that put the Iraqi death toll much higher: A 2006 survey
by Iraqi physicians and overseen by the Johns Hopkins Bloomberg School
of Public Health estimated over 600,000 killed at the time.

The source the Post cites for its "maximum count" of Iraqi civilian deaths is based on casualty reports from the group Iraq Body Count, which bases its figures
on violent civilian deaths that are reported in media outlets and, when
possible, by other NGO and official sources. While the group's figures
represent a serious effort to document reported Iraq deaths, they are
much lower than the death tolls assessed through surveys of the Iraqi
public--the standard method for assessing casualties of large-scale
wars or disasters.

Both the 2006 Johns Hopkins study and an earlier study conducted by Johns Hopkins (both published in the peer-reviewed British medical journal Lancet, 10/28/04, 10/11/06) estimated a death toll several times larger than that of Iraq Body Count; the more recent Lancet estimate found 601,027 "excess" deaths from violence in Iraq. A more recent survey
conducted in August 2007 by the British polling firm Opinion Research
Business (ORB) estimated 1.2 million excess violent deaths in Iraq
since the U.S.-led invasion. And an investigation by the U.N.'s World
Health Organization and the Iraqi Health Ministry found, as the Washington Post reported (1/10/08), that "151,000 Iraqis died from violence in the three years following the U.S.-led invasion of the country."

These estimates do not distinguish between civilians and combatants,
but given that the U.S. government estimated that the U.S. and its
allies had killed 19,000 insurgents by September 2007 (USA Today, 9/27/07),
it's clear that civilians make up the vast bulk of the deaths found in
these surveys. And these surveys are all at least a year old; the WHO
survey in particular was conducted before the most violent extended
period of the war.

It's hard to see why you wouldn't include both civilian and combatant
deaths when attempting to measure the effects of a war. But even if the
Post wanted to count only civilians, the surveys indicate that 88,000-96,000 is almost certainly a very serious underestimation.

Why, then, does the Post opt to
refer only to the lowest figures available for Iraq casualties? And why
does the paper use the misleading label "maximum count" to refer to the
96,719 deaths recorded by Iraq Body Count? Iraq Body Count used to
label the top of its range of reported deaths as its "maximum" number,
but no longer seems to do so; its website notes,
"Gaps in recording and reporting suggest that even our highest totals
to date may be missing many civilian deaths from violence."

Given that over a year ago, the Post mentioned the Lancet's death toll
estimate of over 600,000, it should stop referring to a figure six
times lower as the "maximum count." If the Post insists on relying on
the Iraq Body Count's admitted underestimate of Iraq deaths, it should
convey the statistical differences between the different estimates with
a sentence to the effect of, "Household surveys in Iraq suggest likely
Iraqi death tolls 2 to 10 times greater than this count."

Instead, the paper seems to be opting to use the lowest range it can find.