August 2008

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

« The Democrats and the Middle East | Main | "Third World development? What for?" »

September 19, 2003


David Gillies

Any simple-minded calculus of human life is morally dubious at best, and repugnant at worst (and I'll be the first to admit I've erred in the past in using it as a justification for 'our' sons-of-bitches killing people in order to stave off a far worse hypothetical calamity should 'their' sons-of-bitches have prevailed). But Herold and his acolytes are, as you say, doing little more than a naive zero-sum calculation of civilian casualties. Why don't we play them at their own game? As has been pointed out in this post and elsewhere, the removal of the Taliban and Saddam has dropped the death rate in Afghanistan and Iraq by huge amounts (the much-touted Chomskyite prediction of millions of dead Afghan refugees has been proven to be as much of a fantasy as the 'brutal Afghan winter', and if there's one thing that's sure, it's that Uday Hussein's industrial shredder has been standing idle since the occupation of Baghdad). Let's run the odometer backwards. How many 'innocent, defenceless civilians' today live and breathe because the regimes under which they toiled have been overthrown? The anti-war crowd's refusal to acknowledge this question goes beyond simple-minded cultural relativism to embrace a truly noxious racism in which any action by the West is morally suspect, and any action by the non-West is excused by reference to the West's much-touted iniquities of the past. If an incinerated New York stockbroker equals a Baghdad greengrocer dismembered by an errant bomb, then that same greengrocer alive and kicking due to the Tikrit thugocracy's fall is an entry in the credit side of our moral balance sheet. If we take the IBC people at their word, then the invasion of Iraq has resulted in a _negative_ number of civilian casualties. If nothing else, this serves to point out the vacuity of their position.

A significant factor in the unhinged nature of the anti-Bush sentiment that infests the fringes is ignorance. I had the opportunity to tear a strip off a young American lady in the pub the other night. She averred that George Bush was every bit as bad as Saddam Hussein. I pointed out (with a few beer-fuelled expletives) that until a mass grave containing the corpses of 1200 children was found outside Des Moines she was, not to put to fine a point on it, talking drivel. She retracted her allegation, but I think she was a) surprised to encounter someone of a differing viewpoint and b) not fully in command of her brief (she had never heard of Halabjah, for example). Herold and his ilk, by virtue of their positions in the academy, have no such excuse.

Howard Shaw

If you agree with Jean Bethke Elshtain that "the United States must investigate every incident in which civilians are killed" then is it not a little surprising that we are still only able to "estimate that between 1000 and 2000 Afghan civilians were killed in the US-led campaign"?

Colin MacLeod

One other argument that I think most of us understand intuitively, but that we too infrequently make explicit, is that the primary objectives of war in Afghanistan and Iraq, as of any just war, is the prevention of future calamities. Regarding casualties, the question isn't how many died on September 11, 2001, but how many more would be expected to be die in the future both from direct acts of terrorism and, inescapably, amidst the wars of succession that would likely follow a general US retreat from its security role in the Mideast and elsewhere.

If a "body count" comparison has to be made, then the proper one would be between the casualties and other costs of a war fought mainly on our terms versus those, including indirect casualties of misrule and oppression, of a war fought mainly on theirs.

Chris Brooke

Since this post presents itself as being inspired by my comment on your earlier post, perhaps I should respond.

My reference to "smears" was intended as a description of part of the argument of your post, not as an "allegation" in any strong sense. Glancing at the OED, I see that the verb "to smear" is defined as "to attempt to discredit (a reputation...)", which seems to me to be a fair description of what you were trying to do in your previous post, and one which you could happily accept but for the word "smear"'s inescapably pejorative connotations.

What I was trying to do in those opening remarks was to suggest that when we removed the ad hominems against the individuals involved in the IBC for their associations, connections and professional background, the actual substantial critique of the IBC's methods which you offered was really quite weak. And since I take the view that it's the quality of their methods which really matters, not their friends, associates, areas of scholarly expertise or interests in rock music, I wrote what I wrote to try to refocus attention on the question of methods. It's clear from this post that you still think these other issues matter quite a bit. On that point, we continue to disagree.

What's the nature of the association of the IBC with Herold? I'm not entirely sure. But one thing seems clear, that the methods being used by the IBC have been drawn up partly in order to meet the more cogent objections directed against Herold's earlier Afghanistan project -- and learning from previous errors is usually considered a good part of scientific practice. Safeguards against double-counting have been introduced through the ways in which incidents are identified in the database, for example, and while you report in this post the criticism that Herold was then making too many poorly-grounded subjective judgments in his interpretation of the media reports, the methodology of the IBC -- including the minimum/maximum tallies -- is designed to eliminate the need for the analysts to make judgments of this kind. As I've said already, the rules aren't ideal. But they aren't worthless, either.

One of the frequent charges against the IBC is that they're trying to produce as large a tally as possible. But if the IBC wanted to make their figures larger, there are lots of things they could have done to do this. In line with the British Government and the BBC (cheap shot, cheap shot), for example, they could have been content to rely on single sources for the information in their database. They could have upped the numbers of civilians they count as belonging to a "family". They could have broadened the range of media sources of evidence they were using for their count. They could have accepted evidence from print sources which don't appear on the internet. They could have included reports of fatalities which aren't directly linked to military action (deaths due to epidemics facilitated by the collapsing infrastructure in parts of Iraq, for example). They aren't doing any of these things, and that seems to me to be worthy of remark, and of approval.

For the record, I also think the project has been making some bad mistakes, which do seem to be the result of a bias induced by its political stance. This page, for example -- -- documents a specific incident in their database and suggests that the fatalities had nothing to do with coalition military action. It's cases like these which tend to discredit the IBC in my eyes, at least, not because of the counting rules being used or any worries about whether too much official Iraqi information is seeping into the database (as I said last time, if that's the worry, just deduct 130 from the counts you come across), but because their inclusion in the database suggests that the project's own practice isn't living up to its carefully-stated aims. Which is a very great shame, and diminishes a worthwhile project.

marc w.

I think a clear indication of the political nature of the IBC effort is their counting of 'resistance' bombings and failed attacks on US troops. While they're being somewhat upfront about these attacks, they're all counted in the same statistic, and almost everyone who quotes the IBC tally thinks they're talking about civilians killed by the US/UK.
They've counted ammunition dump explosions, attacks meant for US troops, a car bomb or two, etc. You're right, Chris, that they could inflate the count if they wanted. But by doing just enough to be called 'rigorous' by their ideological allies, they get a lot more than shock value: they get legitimacy.

Squander Two


How do you reason that a criticism of someone's professional background is an adhominem attack? If someone who has qualifications and experience in the fields of history and archaeology writes a book about neurophysics, it's perfectly reasonable to criticise him and his book on the grounds that he doesn't know what he's talking about. That's not an ad hominem attack.

Of course, if a bunch of neurophysicists all say that, surprisingly enough, the book's pretty good, then fair enough and well done. In the case of IBC, though, experienced statisticians haven't endorsed the work; they've rubbished it.

Chris Brooke

Squander Two:

A question: who are the "experienced statisticians" you've come across who have "rubbished" the IBC? I've read lots of criticism of the IBC on right-wing (sorry, Oliver!) blogs, some of it well-reasoned, but the authors of this stuff are no more experienced statisticians than I am. So if there is intelligent, dispassionate or academic criticism of the IBC project out there by professional statisticians (or demographers), I'd be interested in knowing where to read it.

A reply: Since Oliver went well beyond pointing out that the IBCers weren't statisticians to give considerable detail about the particular academic interests of the IBCers, I took the subtext of his remarks to be something like this: "C'mon, these guys are interested in rock music: why should we take anything they say about Iraq seriously?" That seemed to me to have an ad hom flavour to it. If they've got a well organised project and some intelligent methods (something we continue to disagree about), then maybe we should take it seriously -- even if their only interests are golf, strangling animals and masturbating.

If the historian writes a bad book about neurophysics, after all, the thing to do is to point out the mistakes about neurophysics in the book rather than to observe that the author trained as an historian...

Squander Two

>> "who are the "experienced statisticians" you've come across who have "rubbished" the IBC?"

Oliver Kamm, Iain Murray, and me. A quick bit of googling could find you many more, I have no doubt. I don't think I've ever met a statistician who wouldn't laugh derisively at IBC's use of the phrase "most of the time", as highlighted by Oliver.

Just to be clear, I supported the war for political reasons and would still have done so if the real body-count had been ten times what IBC say it is. I would have absolutely no ideological or political problem with IBC's claims being completely correct. If I want to criticise their anti-war stance, I'll do so in the same way I criticise any other enemy of liberty. When I criticise their statistical methods, it's because, as a maths & logic graduate and former professional manipulator of statistics, I can spot deliberate number-fudging a mile off and think it's a particularly disgusting way to fool people, particularly over an issue this important.

>> "If they've got a well organised project and some intelligent methods (something we continue to disagree about), then maybe we should take it seriously -- even if their only interests are golf, strangling animals and masturbating."

I agree absolutely.

Chris Brooke

Squander Two:

We agree about more than we disagree.

Re: "Most of the time": That's a good point, but it's one about the Afghanistan project, not the IBC. As far as I can tell the methodology of the IBC was designed to avoid this problem: the only numbers that enter the IBC database are numbers in the reports they use, not any guess (sorry, "judgment") they make from rival and conflicting reports.

Daniel Calto

One thing missing from the "body-count" reductionism (regardless of the whopping inaccuracies of the IBC's methodology) is the lack of accounting for the value of liberty for the people who actually survive the war. The number of Iraqi civilian dead may or may not exceed the number of people killed in 9/11. How does one account for the lifting of brutal political repression and the increase of political and civil freedoms for close to 50 million people in Afghanistan and Iraq?

For anyone who has seen the effects of such repression on the spirit of people under the yoke, even indirectly, they are profound, and soul-killing. My wife lived through the Cultural Revolution, and even though she was a little girl and thus spared its worst effects, its pernicious consequences continue to affect her well into her adulthood and maturity, in lack of ability to trust, estrangement from her parents, and lack of hope. I am very happy that her nieces and nephews will face a far milder set of consequences than she had to, even growing up in a country that one would classify as "still unfree".

Cervantes said it best:

"Liberty is one of the most precious gifts heaven has bestowed upon Man. No treasures the earth contains or the sea conceals can be compared to it. For liberty one can rightfully risk one's life."

Squander Two

>> "That's a good point, but it's one about the Afghanistan project, not the IBC."

So it is. Oops. Er... carry on. Nothing to see here.

Phil Rodgers

To comment on your update, regarding the forwarded correspondence: I think the IBC participant has become confused by your blog layout, rather than by living in a parallel universe, as you suspect. His email clearly refers to the comment beginning "Once we remove the smears by association..." rather than the immediately preceding comment beginning "Great post, and badly needed". However, he has erroneously attributed the "smears" comment to Franco Alemán, rather than Chris Brooke, thinking that the "Posted by" lines refer to the following, rather than the preceding article. This is an easy enough mistake to make when scanning through a long sequence of comments. Perhaps you could make the "Posted by:" line appear before the horizontal rule, rather than after it, to reduce the chance of this happening?

The comments to this entry are closed.