Sadly, no surprise
Iraqi death researcher censured
This study of civilian deaths in the Iraq War has been known to be junk virtually since its publication, although the anti-war left constantly cites its ridiculous estimates as established truth. Now the primary investigator won't cooperate with an inquiry into how the research was done.
It's a deep embarrassment, or should be, to both The Lancet, where it was published, and to Johns Hopkins, where Burnham works. The method of extrapolation used was simply preposterous, and it produced preposterous results. As murderers always discover, death leaves a body, and it takes extraordinary effort to make a body disappear. When the extrapolations yielded death estimates that were half a million (!) more than the body count, it should have waved a profusion of red flags to the Lancet reviewers. Sadly, they were more interested in having the journal make a political point, thus rendering yet another formerly trustworthy source untrustworthy.
[ Aside: You can find a similar pattern of politically correct junk science in CDC studies of gun deaths in the US, and in Michael Bellesiles, (History Prof. at Emory), being awarded Columbia's Bancroft Prize for a study of gun ownership in early America, then having to return it and resign his position in disgrace when his research was exposed as fraudulent. Bellesiles downfall was due to the efforts of a lay expert, Clayton Cramer, not to professional historians policing their own. On the former, see http://www.reason.com/news/show/30225.html , for the latter http://en.wikipedia.org/wiki/Michael_A._Bellesiles ]
Science in the service of grinding some ideological axe is always an ugly sight. In a free society, science and news reporting are the two areas that must be kept free of bias; as we swim through social-political space, they are the only reliable anchors we have. It's no accident that they are the very first things every authoritarian regime, left or right, sets out to corrupt.
Byron
This study of civilian deaths in the Iraq War has been known to be junk virtually since its publication, although the anti-war left constantly cites its ridiculous estimates as established truth. Now the primary investigator won't cooperate with an inquiry into how the research was done.
It's a deep embarrassment, or should be, to both The Lancet, where it was published, and to Johns Hopkins, where Burnham works. The method of extrapolation used was simply preposterous, and it produced preposterous results. As murderers always discover, death leaves a body, and it takes extraordinary effort to make a body disappear. When the extrapolations yielded death estimates that were half a million (!) more than the body count, it should have waved a profusion of red flags to the Lancet reviewers. Sadly, they were more interested in having the journal make a political point, thus rendering yet another formerly trustworthy source untrustworthy.
[ Aside: You can find a similar pattern of politically correct junk science in CDC studies of gun deaths in the US, and in Michael Bellesiles, (History Prof. at Emory), being awarded Columbia's Bancroft Prize for a study of gun ownership in early America, then having to return it and resign his position in disgrace when his research was exposed as fraudulent. Bellesiles downfall was due to the efforts of a lay expert, Clayton Cramer, not to professional historians policing their own. On the former, see http://www.reason.com/news/show/30225.html , for the latter http://en.wikipedia.org/wiki/Michael_A._Bellesiles ]
Science in the service of grinding some ideological axe is always an ugly sight. In a free society, science and news reporting are the two areas that must be kept free of bias; as we swim through social-political space, they are the only reliable anchors we have. It's no accident that they are the very first things every authoritarian regime, left or right, sets out to corrupt.
Byron
posted by Byron | Wednesday, February 04, 2009
0 Comments:
Post a Comment
<< Home