1. the people who did it used the same method as they and others have done for figures in the Congo, etc. without much demurral. In the light of previous discussion on Bosnia, Croatia and Kosovo, I guess the numbers are ok when one wants it that way, but not when not.
2. as for being widely discredited -- I think crookedtimber.org did an extensive take on it and pretty much rubbished those who tried to discredit it, plus wasn't there an item in highered.com which covered it and pointed out that the oft-quoted washpost article cited someone who subsequently retracted that quick, off-the-cuff comment.
the 'sin' of the Lancet paper was simply being published at the wrong time, a misjudgment perhaps of the principal author.
on the other stuff in the logictimes.com piece -- not sure it's so much statistical as sociological, since the author hinges quite a bit on the gender differentials, implying that all those men who died were, well, not civilians. perhaps saddam might want to apply the same technique, esp in light of the previously, and once again?, anointed allawi's claim, and the current govt's plea to be allowed the use of more 'robust' measures?
whatever, while ANSWER and the like might have something to answer for, would it make the invasion and occupation ok if it could be validated that few civilians died? and what if it turns out the other way? or, is the criminality of the invasion and occupation independent of the actual number of casualties, the latter simply a compounding matter?
kj
On 11/28/05, Michael Pugliese <michael.098762001 at gmail.com> wrote:
> http://www.logictimes.com/civilian.htm
> I'll forward this around for comments by those far better able to
> judge the statistical analysis.