More evidence that "churnalism" (media churning of
myths and dubious factoids, rather than checking the facts) occurs
not just in the mainstream media. Adbusters have published
which regurgitates many of the discredited myths (originating
in sloppy pieces from the likes of Medialens), over the Iraq death
count issue. Here's my email:
Sean Condon's article, 'A Question of Numbers', is so full
of basic errors and omissions, that I think it should be rewritten.
Here are some of the errors:
1. The Lancet 2006 study did not "find roughly 650,000
civilian casualties" (as Condon claims). It estimated roughly
650,000 excess deaths, including combatants. The survey made
no distinction between civilians and combatants.
2. Condon writes that Iraq Body Count (IBC) "only count
a civilian death if it has been reported in at least two English
media sources". This is incorrect and misleading. IBC monitor
around 70 non-western media sources daily, along with 120 western
sources. To take a given day (2/10/06), IBC compiled material
from agencies based in 7 countries - USA, UK, France, Kuwait,
Iraq, Germany, China. IBC utilises reports written or published
in the English language, although not always originating in
it - eg material from the foreign-language monitoring units
of Middle East media, or the major Iraqi media. Note also that
English-language wire services such as Reuters and AP (whose
reports are compiled by IBC) use reporters in Iraq who are predominantly
Iraqi and whose reports are frequently what local Iraqi media
rely on for daily news. http://www.iraqbodycount.org/analysis/beyond/state-of-knowledge/9
3. Condon writes that: "the number of civilian casualties
is estimated to be anywhere between 80,000 and 1.2 million".
Again, this is misleading. The 80,000 figure (approximately
IBC's current count) is not an "estimate". IBC have
always stated that they necessarily provide an undercount, since
"many if not most civilian casualties will go unreported
by the media". The 1.2 million figure is an estimate (by
ORB) of the total number of "Iraqi citizens [who] have
been murdered". In other words, this is not comparing like
with like. Note also that ORB have revised their estimate downwards
to "the order of 1,033,000" - Condon's "1.2 million"
figure is out of date.
4. Condon claims that "the Lancet study uses scientific
methods that have been proven accurate everywhere from Darfur
to Kosovo". In fact these methods (which were not designed
for violent conflicts) are not well-validated in war zones.
The co-author of the Lancet 2006 study, Gilbert Burnham, stated
in an email to me that because so few surveys of this type have
been conducted in conflict zones, and particularly in urban,
middle development level countries such as Iraq, there is no
standard for them. The notion that these methods have a "proven"
record of "accuracy" in war zones is a myth. To say,
as Condon does, that they've been "proven accurate everywhere
from Darfur to Kosovo" is simply incorrect.
5. Condon writes that journalists who quote IBC are "only
referencing their own reports, while creating the illusion of
a separate, reliable source". This is misleading and demonstrates
a lack of understanding of IBC's work (for one thing, their
data is drawn not just from media reports, but also from hospital,
morgue, NGO and other sources, wherever it can be integrated).
Moreover, IBC's data has received confirmation from two studies
which used methods similar to the Lancet study, but on a much
bigger scale and with better quality control - more on these
6. Omissions. Given that Condon's article is from the March-April
2008 Adbusters (although I can't tell when it was written) the
absence of any reference to the recent World Health Organisation
study is disappointing. This study used methods that were similar
to Lancet 2006, but on a larger scale (it surveyed 9345 households
in 1086 clusters, compared to the Lancet 2006's 1849 households
in 47 clusters). The WHO study estimated 151,000 violent deaths
compared to Lancet 2006's 600,000 violent deaths (over roughly
the same period). The WHO study's findings are sharply at odds
with Lancet 2006 over estimated violent deaths, and much closer
to Iraq Body Count (which records only civilian deaths; WHO,
like Lancet, includes combatants).
7. Another omission is the United Nations Iraq Living Conditions
Survey (ILCS), 2004 - another cluster-sampling study which dwarfs
the two Lancet studies (over 21,000 households surveyed). This
found a much lower number of violent deaths (in an overlapping
period – it estimated nearly 24,000 civilian deaths in
the first 13 months of the conflict) than is implied by Lancet
8. Condon refers to "the respected UK-based polling firm,
Opinion Research Business (ORB)" as providing the 1.2 million
estimate. ORB effectively subcontracted the Iraq polling work
to the Independent Institute for Administration and Civil Society
Studies. The person conducting the research, Munqeth Daghir,
is reportedly a self-trained pollster who began his polling
activities in 2003. Describing his early attempts at polling,
Daghir says, "I knew that Baghdad is distributed into nine
different areas, and how many citizens lived in each one. But
to tell the truth, I didn't know anything about the real random
systematic sample. We did it randomly by going to any house
we wanted to go to. So it wasn't a perfect sample."
9. ORB originally stated that their survey was based on "a
nationally representative sample." Later, they admitted
that the survey was "undertaken in primarily urban locations".
Given that about a third of Iraqis live in rural areas, this
is a significant omission, which ORB failed to disclose when
the results were first published. ORB conducted a follow-up
survey to correct this omission (after receiving criticism on
this point). As a result their estimated total was revised downwards
from the figure Condon quoted.