Megan on the body count: Seems reasonable
By Thoreau
I have been tough on Megan McArdle, but her article on the body count in Iraq seems reasonable. She reports that some of the higher estimates may be off, and those who want to can bash her for that, but she makes the important point that even if the body count is “only” 80k or 100k or whatever, that’s still pretty freaking horrific. As somebody who has been opposed to it from the start, I have to say that my case against the war doesn’t really depend on a particular body count, or at least not the most gruesome estimates. Even if the very lowest figure were taken, that would still come to 16,000+ deaths per year in the 5 years since the invasion, or the same body count as 5+ 9/11 events per year. If America were to suffer 5+ 9/11 attacks every year, we would call that a complete failure of security policy, and our society would degenerate into something even uglier than we’ve already witnessed. I think that’s the perfect indictment of our policy in Iraq. If anybody wants to know why the Iraqis can’t achieve stability, it’s because in the chaos and the power vacuum they’ve been subjected to violence that would unravel even our society. (Just look at what 9/11 did to our society, multiply by 30 or more, and you get the picture.)
So I have no problem with her reporting on the difficulty of getting reliable numbers, and nowhere in the piece do I find any hint of apologia for bloody chaos. Whatever issues I may have with some of her blog posts, this piece of professional writing displays none of the pathologies that I’ve bemoaned in the punditocracy at large.
EDIT: Some have called her points into question. I admit that I don’t know anything about the methodology in these sorts of things (although she sets off my BS detector once or twice on a second look). She could be right, she could be wrong. My main point here is that, whatever other flaws one might find in the article, pro-war apologetics is not one of them. Anybody who says that even 100,000 dead is ghastly and unacceptable is no longer a war apologist. That’s all.
Comment by Barry —
March 27, 2008 @ 4:52 pm
Start with Tim Lambert’s post on Deltoid (over at Scienceblogs). The category is:
lancetiraq
Then go to Daniel Davies blog (he’s a newer regular poster on Crooked Timber):
Here is the search string
then go to Crooked Timber:
using this search
But first, stop trusting somebody who’s proven to be hard-core, professional, serial liar.
Thoreau, we’ve all spent several posts jumping on Megan. Prominent among things noted is that she’s (a) wrong, (b) in a particular way, (c) with extremely dishonest stated reasons.
Comment by Thoreau —
March 27, 2008 @ 5:49 pm
For all I know the studies that she casts doubt on may be completely correct and she may be completely wrong. Be that as it may, she doesn’t attempt to do anything nefarious with her arguments. She doesn’t try to say “See? Only 80k dead! What’s so bad about that?” She states that the bloody chaos making such research difficult, plus the fact that the “low” estimate is still ghastly, is a pretty bad indicator of how things are in Iraq.
I’m giving credit where credit is due. Whatever might be said of her blogging on Iraq, I don’t find anything nefarious in this researched, edited, and paid article.
Comment by matthew hogan —
March 27, 2008 @ 6:00 pm
I havent checked it out (Megan’s thing) but as someone who has opposed the Iraq mess from Day minus one through today and believes that the ugliness may be underreported, I’ve always found, for reasons related to solid metrics, that the high casualty figures like Lancet are greatly bunk. Iraq Body Count is a good place to build a reasonable inquiry around.
Comment by Glaivester —
March 27, 2008 @ 6:16 pm
One hting also to remember isthat most of the casualty counts are net fatalities (aka “excess deaths”), not gross fatalities. That is, when one says that “well, only 100,000 died. If Saddam were in power, more would have died,” one needs to remember that these counts already attempt to take into account how many would have died under Saddam. (Of course, you can always argue that the estaimtes are wrong, but most people ignore the fact of these estimates’ existence in the first place).
Comment by lemuel pitkin —
March 27, 2008 @ 6:44 pm
You know, if I cared what Meg McArdle thought, I would read her.
I’m beginning to think maybe Henley needs a blog of his own.
Comment by Thoreau —
March 27, 2008 @ 6:53 pm
Oh, we all go through our obsessions. Remember when Mona was constantly blogging about Michael Ledeen?
Comment by Uncle Kvetch —
March 27, 2008 @ 7:01 pm
The “obsession” with McArdle extends way beyond this blog, though, which makes it all the more baffling, not to mention annoying. Is it because she calls herself a “libertarian” (albeit one who didn’t realize until only recently that the US military couldn’t build a Western-style democracy in Iraq from scratch) that other self-identified libertarians feel compelled to engage her?
Comment by dsquared —
March 27, 2008 @ 7:09 pm
Seems but isn’t. It’s more of a piece with the whole whining “I was was right to be wrong; you were wrong to be right” stuff that was all over her blog last week. Specifically, it is an attempt to defend, post facto, her rubbishing of the Johns Hopkins team back in 2004 who were giving the first early warnings that this was turning into a horror story.
Her specific conclusion from that article, as in, what does she actually say we should do differently, is that battlefield epidemiology “probably shouldn’t be done”, because hey, those estimates have wide confidence intervals and our national debate is better off without such information. That is not harmless at all my friend, is it? It’s the ideology of “we don’t do body counts” that cost us at least two years of mainstream opposition to this war.
And also, in context, the sting in her final paragraph where she accuses people who defended the Hopkins researchers against hack critiques of “not wanting Iraqis to be alive” is really rather disgusting. Be a bit sharper here, Thoreau mate.
Comment by dsquared —
March 27, 2008 @ 7:17 pm
Oh, and of course the people who paid for The Atlantic magazine might care about the fact that the article is riddled with statistical errors and criticisms to one-sided as to be totally misleading (I mean come on, Thor, when you saw in the first paragraph that a number which is obviously the point estimate of a quantity estimated with a large confidence interval is an “absurdly precise” number, didn’t the residual mathematician in you stop and think “how the hell is any integer more ‘precise’ than any other?”?)
Comment by Thoreau —
March 27, 2008 @ 8:00 pm
Maybe I shouldn’t blog an article that I read in a hurry. Still, I think that even after all is said and done it’s hardly a piece of war apologia. Although I admit that I haven’t followed her history of arguments concerning body counts, so maybe there’s a subtext of “I was right to be wrong” that I’m not getting.
Comment by Tim Lambert —
March 27, 2008 @ 9:06 pm
She’s been writing inane anti-Lancet posts ever since the the Lancet study was published. This is just more of the same, this time pretending to some sort of knowledgeable, objective observer. Now I’m going to have to write a post about her crap.
Comment by Donald Johnson —
March 27, 2008 @ 9:08 pm
The following part is just factually wrong. I don’t mean in the sense that I “know” what the true death toll is–I mean she doesn’t get her facts right regarding the studies she cites–
“It covered basically the same time period and used similar statistical techniques, but with a much larger sample and more-rigorous interview methods. It found that the Lancet study’s violent-death count was roughly four times too high. This has a familiar ring to it. A smaller study, released by the Johns Hopkins team in 2004, had been quickly contradicted by a larger UN survey suggesting that it had overstated excess mortality by, yes, about a factor of four.”
There are several mistakes and misleading statements packed in there. First, the larger UN study didn’t say the earlier John Hopkins study had overestimated the death toll by a factor of four. As Tim Lambert showed, when you compare like with like, the two studies gave roughly the same result.
Second, that earlier John Hopkins study happens to agree with the recent New England Journal of Medicine study on death rates in the first 18 months of the war. They both say about 120 violent deaths per day in that time period. So all those critics of the 2004 study (which probably includes Megan) have been discredited by this later study and she doesn’t bother to tell her readers. It’s true that this later study gives a violent death toll number four times lower than Lancet2, but an honest, competent reporter would have also said it supports the numbers in the earlier Lancet1 paper. But not Megan.
Third, the IBC number isn’t an estimate. IBC is rather cantankerous about people who say the true number is much higher than theirs, but even they admit that their number is anything but an undercount.
Fourth, and here we get into vaguer territory, what “more rigorous interview methods”? This latest study didn’t ask for death certificates and didn’t focus on deaths and furthermore, informed the people they interviewed that they worked for the Iraqi government. That’s commendable honesty, but does it occur to Megan that telling people living in a war zone that you work for a government which runs death squads might not be the best way to get honest answers?
So she can’t get her facts straight and she doesn’t give any information that might cast doubt on the lower figure. She should stick to her area of expertise, if anyone can identify what that might be.
Comment by Donald Johnson —
March 27, 2008 @ 9:09 pm
“but even they admit that their number is anything but an undercount.”
I should edit my posts, I suppose. What I meant is that even IBC admits their number is an undercount.
Comment by Thoreau —
March 27, 2008 @ 9:14 pm
I did not evaluate the article for rigor on the first reading. I evaluated it on whether it was a piece of pro-war apologia. I see no such thing in there.
It may very well be a flawed analysis (I’ll read it later for that) but it isn’t pro-war apologia.
I will say this much against her: If the statistical reasoning in there is deeply flawed (haven’t evaluated it for that, I will soon) then I do wonder how she became an economics writer.
Comment by Avram —
March 27, 2008 @ 9:20 pm
If anybody wants to know why the Iraqis can’t achieve stability, it’s because in the chaos and the power vacuum they’ve been subjected to violence that would unravel even our society. (Just look at what 9/11 did to our society, multiply by 30 or more, and you get the picture.)
I don’t know if that necessarily follows. American traffic fatalities amount to more than ten 9/11s per year, and have for quite some time, and it hasn’t turned is into an Iraq-style mess.
Surely Iraq’s problems are a complicated matter. The divided society, no history of liberal democracy, the presence of foreign troops, the cobbled-together corrupt government, the inability of the government to provide services, and other things that I can’t figure out how to boil down to half-a dozen words. And, yeah, the ongoing violence too.
Comment by Donald Johnson —
March 27, 2008 @ 9:27 pm
I don’t think you should get defensive, Thoreau. People aren’t attacking you (I don’t think). They’re pointing out that you’ve missed the flaws in her piece. That’s not a sin. I find it really depressing, though unsurprising, that this article was researched, edited, and paid for.
Comment by sglover —
March 27, 2008 @ 9:57 pm
Even aside from the issue of good faith (which disqualifies McArdle straight away), is there any reason to believe that she’s got even minimal competence to assess mortality statistics?
If you’re gonna say that she’s got an MBA, I’m reminded of what a former girlfriend of mine told me about her MBA classmates (her undergrad degree was in EE): They think that if they can an integral, they’re geniuses. I’m very sceptical that MBA’s have the quantitative skills of most undergrad science students. Oh, and there’s another thing that’s dimmed the luster on the vaunted MBA…..
Comment by sglover —
March 27, 2008 @ 10:05 pm
I think it’s pretty much of a piece with Sullivan bashing — which there’s never enough of. It’s got absolutely nothing to do with their pathetic attempts to be provocative or contrarian, because neither of them has the ability or imagination for that.
No, the reason why Sullivan and McArdle get so much attention is that they’ve attained prominent positions MILES beyond any talent that either of them has ever exhibited, and the outfit that gave them their perches is a magazine that a lot of us used to remember pretty fondly.
Comment by Mark Holden —
March 28, 2008 @ 1:02 am
@Avram: I think the original analysis is absolutely right. It’s not just the number of deaths that’s important — it’s also the nature of the deaths. And here, it seems to me that the comparison between “9/11 events” and the bloodshed occurring in Iraq is entirely apt.
@all: It’s also important to note that America is much wealthier than Iraq, and has a much larger population. So while I agree entirely that 5+ 9/11 events per year would be highly destabilizing to America, I think the difference — how much more destabilizing they are for a much less developed country like Iraq — could stand greater emphasis.
Comment by Just a Quick Question —
March 28, 2008 @ 10:20 am
Continuing on Mark’s point, the estimate of Iraq’s population is something like 27 million, about halfway between Texas and California. So, another way of looking at it is 4 9/11 events annualy for Texas, New York and Florida, 6 for California. etc.
Based on population alone, it’s closer to a 9/11 event in the US, weekly.
With no end in sight.
Remind me again, what was her argument for listening to the people who thought this was a good idea?
Comment by Keifus —
March 28, 2008 @ 11:13 am
So McArdle’s an MBA? No wonder she seems to think she’s a member of the master class. (Though to be fair, I hardly visit, and maybe that’s not her usual thing.)
An MBA vs. engineering pile-on is just too delicious to resist. Back in the happy undergrad days, I perceived the business school as nothing more than a fallback for those students (some of them my friends) who weren’t really bright enough for engineering. Little did I know at the time that (a) teh business school is, evidently, a huge part of my alma mater’s (RPI’s) income model, and (2) I’d be plagued with these nincompoops as my bosses for the forseeable future of my career. My best friend, a good engineer, left the field for the “upward” path of a marketing big shot. He more than doubles my salary now and gets these things called “bonuses,” but decently, doesn’t pretend it’s the same intellectual challenge.
I sometimes think I should have gone into bartending. It wouldn’t have been very satisfying, but it would’ve been a hell of a lot less depressing.
Comment by lemuel pitkin —
March 28, 2008 @ 12:16 pm
I admit that I don’t know anything about the methodology in these sorts of things
Many of us might find this a reason to refrain from posting on them.
Comment by Thoreau —
March 28, 2008 @ 12:18 pm
lemuel-
That’s why I didn’t endorse her analysis of the methodology, and only spoke to other aspects of the article.
Comment by Tim Lambert —
March 28, 2008 @ 2:25 pm
Here are my comments on McArdle’s piece.
Comment by Barry —
March 28, 2008 @ 4:46 pm
Comment by Thoreau —
“lemuel-
That’s why I didn’t endorse her analysis of the methodology, and only spoke to other aspects of the article.”
Thoreau, somebody writes something whose methodology you don’t know enough to criticize. However, you seen (and have just discussed with others) that same someone’s extensively repeated examples of willful dishonesty in another field over several years. Dishonesty which, from other recent conversations, you are aware that that someone is rewarded for (cf. Will Roger’s saying, “it’s hard to get a man to understand something, if his paycheck depends on his not understanding it.”).
*My* first thought would be to not trust that someone.
Comment by Thoreau —
March 28, 2008 @ 4:50 pm
I said that whatever she’s doing in that article, it isn’t pimping for the war. That much I can infer despite limited expertise.
Although perhaps the thread title is not quite appropriate to that point.
Comment by Thoreau —
March 28, 2008 @ 4:51 pm
FWIW, guys, I find this site kind of interesting:
http://firemeganmcardle.blogspot.com/
Comment by Barry —
March 28, 2008 @ 5:27 pm
Comment by Thoreau —
March 28, 2008 @ 4:50 pm
“I said that whatever she’s doing in that article, it isn’t pimping for the war. That much I can infer despite limited expertise.”
Remember back when the Iraq Body Count Project was at 30K, and the right-wing of the blogosphere thought that was too high, and the administration, as well?
What changed? The Lancet survey came out, and suddenly 30K dead became the hard authoritative number. Why?
Overton Window.
Remember ‘good news from Iraq’, and the theme among the right that there were no real problems? That changed, to ‘flypaper’. Violence became good. Why?
Overton Window.
Or, in other words, ‘walking slowly backwards’.
Remember one of Bart Simpson’s favorite comments? ‘I didn’t do it/nobody saw me/you can’t prove anything’ (quote from memory).
The last few years have been rich display of the right being wrong, and just coming up with new reasons and justifications, all the while like Megan has just been doing heaping scorn on the people who were proven right.
I was watching an episode of ‘Sex and the City’ a couple of years ago, with a friend who was getting her English Ph.D. At one point Carrie, after moving to Paris (i.e., leaving ‘The City’) loses a cheap necklace which said ‘Carrie’ (i.e., losing her name).
I turned to my friend, and asked her if she’d assign SatC episodes for essays. She replied, ‘no, the symbolism is far too easy’.
The same thing applies her, for students of politics and propaganda. The march of folly, ignoring the current truth while admitting past truths and still dissing the truthtellers is incredibly strong.
Comment by Thoreau —
March 28, 2008 @ 5:33 pm
She may very well be wrong for all sorts of reasons, and by all means go after her for that. My only point is that it isn’t war pimping because she says that even the lowest figures proposed are still unacceptably high.
There are lots of mistakes that she might be making in that article. War pimping isn’t one of them. That’s my only point. Comprende?
Comment by Thoreau —
March 28, 2008 @ 5:39 pm
BTW, if you think that saying somebody is not a war pimp amounts to praise, then we’ve sunk pretty low in this day and age.
Comment by Donald Johnson —
March 28, 2008 @ 6:17 pm
The problem, though, is that in some sense she is a war pimp. She’s saying that look, those antiwar people are still stupid and you shouldn’t listen to them, because they believe those wildly inaccurate Lancet estimates. It’s more of this “they were right for the wrong reasons” crap.
Plus, she’s incompetent. Lancet1 and the NEJM paper are in close agreement for the first 18 month period, which means that if she were intellectually consistent, she’d be saying that the violent death toll might be much lower than Lancet2 claims. but Lancet1 is supported by this new study that she thinks is better. (Though actually it has problems of its own.)
That would mean that all the critics of Lancet1 back in 2004 would have to admit they were probably wrong about the death toll and the dirty fucking hippies were right, at least at that point. She either couldn’t figure that out or she didn’t want to admit it.
I think, Thoreau, you felt a little uncomfortable bashing her so hard in the previous posts and so you wanted to find something about her worth praising. That’s understandable. So you latched onto this article as something that wasn’t so bad. But it’s just more of the same old Megan.
Comment by Just a Quick Question —
March 29, 2008 @ 7:45 am
I think Barry @ 28 has hit the nail on the head. The main point of the article was that we shouldn’t be trying to count the number of dead Iraqis. It seems like the comment that if it’s “only” 150 K dead is bad, was made to bolster her point that over-estimates of the dead are worse than acknowledging any casualties.