Fred Kaplan's Take

Fred Kaplan, who offered up what I thought was an uncharacteristically blinkered and churlish dismissal of the first Lancet cluster-sample study of deaths in Iraq, has a more measured critical take on the newer study that raises two somewhat convincing points. One, as you may have seen on some other blogs is an argument developed by British academics Sean Gourley, Neil Johnson, and Michael Spagat who point out that the Lancet's household samples were located on "major commercial streets and avenues" rather than, say, back alleys. This significantly simplifies the logistics of doing the survey and for standard public health purposes works fine. For a war, though, it's at least plausible that the main streets might feature more violence than non-main ones.

Kaplan's other point is that the Hopkins account of Iraq's pre-war death rate (5.5 per 1,000 per year) is at odds with the UN's estimate (10 per 1,000 per year) and that if you take the UN's pre-war baseline, you wind up with more like 300,000 excess deaths than 650,000. I don't really know how to evaluate that dispute and I'm not sure anyone else does either. That said, this point from Kaplan seems to me like the crucial takeaway: "Let's say that the study is way off, off by a factor of 10 or five—in other words, that the right number isn't 655,000 but something between 65,500 and 131,000. That is still a ghastly number—a number that, apart from all other considerations, renders this war a monumental mistake."

He doesn't get explicit about this, but for my part the point would be that for a war allegedly justified at this point on what are largely humanitarian grounds to have any positive excess death rate is a scandal. A humanitarian intervention ought, on any reasonable view of the matter, save lives not increase the volume of death. The exact amount by which the death toll has gone up is sort of neither here nor there. On top of that, it's worth noting that there are other problems with the methodology that may be leading to an undercount. In particular, one assumes that in some instances entire households have been killed, but the Hopkins method isn't going to count any households like that.

Comments

the Lancet's household samples were located on "major commercial streets and avenues"

Or not.

Posted by Sven | October 22, 2006 3:04 AM

Powerline has done an interview with the main author of the Lancet study (Gilbert Burnham) and asked the question about the pre-war death rate. This is his response:

This was a ‘cohort’ study, which means we compared household deaths after the invasion with deaths before the invasion in the same households. The death rates for these comparison households was 5.5/1000/yr.What we did find for the households as a pre-invasion death rate was essential the same number as we found in 2004, the same number as the CIA gives and the estimate for Iraq by the US Census Bureau.

This answer perhaps is insufficient, but I do know that most of what folks have to say about this study is based upon simple ignorance. One must rise to the occasion and answer the data points of the study itself, not one's own intutions. One must show in detail that the methodology for the cohort study was insufficent and that the CIA's number is wrong. Otherwise one is talking out of one's ass, obviously a not infrequent occurence in American political discourse.

Posted by Steven Levine | October 22, 2006 3:08 AM

Agreeing that the civilian death toll is so incredibly high, regardless of the exact number, that the number itself no longer carries any additional moral weight ('but we only killed 100,000 people!" shall never be a legitimate excuse for anything), it;s worth pointing out that the body count is currently on the rise, according to virtually anyone with any firsthand knowledge, and, more disturbingly, we have no way to stop, or even slow it. No one has the leverage to make the kurds stop killing off non-kurds on their turf, the Sunnis have no incentive whatsoever to stop their insurgency short of measures that the Shiites would never in a million years be willing to accept, and the only thing that may bring the Badr and Mahdi Army fighters together is a common fight against the sunnis.

If I were more cynical, I'd suggest the only solution is to seed the country with Jews, since hating the Jews is the force that binds every other race, color, and creed together into a common band of humanity, at least for the past few thousand years.

Posted by jfaberuiuc | October 22, 2006 3:08 AM

I think Bush's actions rise to a level of criminality on par with some of history's nastier henchmen and despots. Killing tens of thousands of innocent civilians got many previous such madmen exiled, jailed, shot or hanged. If their were any justice in all this similar fates would befall Bush.

Posted by steve duncan | October 22, 2006 7:05 AM

WASHINGTON (Reuters) - The Republican congressman responsible for protecting minors online and calling for more severe penalties for internet predators, Mark Foley, made friends with a wide circle of teenaged U.S. House of Representatives pages, then singled out "hot" boys to write to. A former top aide to Foley said he told senior aides to GOP house leader Dennis Hastert about Foley's behavior three years ago, but Republican leaders said they did not know about the explicit e-mails sent by Foley before media reports. Four more former pages have now said they were sexually solicited by Foley. One said Foley sent him e-mails years ago when he was 16 asking about "my roommates, if I ever saw them naked" and said Foley hinted about a GOP job opportunity "because I was a hot boy". Two years later the page wrote Foley to ask about hotels in Washington. "You could always stay at my place. I'm always here, I'm always lonely, and I'm always up for oral sex," he quoted Foley as saying in reply. Another former page said he felt he had to flirt with Foley. Foley has said he is a homosexual and an alcoholic. "I didn't want to piss off a member of an institution that I really revered," the former page said. "I figured maybe someday I will want to be involved in Congress. I didn't want to make an enemy."

Posted by We The People | October 22, 2006 7:25 AM

WASHINGTON - The Republican congressman responsible for protecting minors online and calling for more severe penalties for internet predators, Mark Foley, made friends with a wide circle of teenaged U.S. House of Representatives pages, then singled out "hot" boys to write to. A former top aide to Foley said he told senior aides to GOP house leader Dennis Hastert about Foley's behavior three years ago, but Hastert said he did not know about the explicit e-mails sent by Foley before media reports. Four more former pages have now said they were sexually solicited by Foley. One said Foley sent him e-mails years ago when he was 16 asking about "my roommates, if I ever saw them naked" and said Foley hinted about a GOP job opportunity "because I was a hot boy". Two years later the page wrote Foley to ask about hotels in Washington. "You could always stay at my place. I'm always here, I'm always lonely, and I'm always up for oral sex," he quoted Foley as saying in reply. Another former page said he felt he had to flirt with Foley. Foley has said he is a homosexual and an alcoholic. "I didn't want to piss off a member of an institution that I really revered," the former page said. "I figured maybe someday I will want to be involved in Congress. I didn't want to make an enemy." After resigning, Foley is now collecting a government pension of over $30,000 per year. Current Page Board member Republican Heather Wilson of New Mexico covered up her husband's own brush with the law over "inappropriate contact with a minor" in 1996. Wilson, while Secretary of the New Mexico Children, Youth and Families Department, managed to cover up the incident involving the 16-year old boy and her husband. Wilson served on the Caucus for Missing and Exploited Children under chairman Mark Foley. Caucuses and boards devoted to exploited children, as well as other seemingly "youth development and protection" organizations, have been used by GOP and Religious Right wing groups to mask pedophilia and pederasty. Republican Sue Kelly of New York was Chairwoman of the Board from 1999 to 2001, along with former Chair Jim Kolbe of Arizona. Kolbe is under investigation by both the House Ethics Committee and the Justice Department for alleged inappropriate contact with two male pages during a July 4, 1996 camping trip to the Grand Canyon. Under Kelly's Page Board chairmanship, a page approached Kolbe in 2000 with allegations concerning Foley. Nothing was done. Kelly did not investigate allegations from a page in 1999 that Foley was stalking him, and another incident in 2000 when Foley showed up drunk at the Page dormitory and tried to gain entry. Kelly, like Missouri Republican Jo Ann Emerson, now want to make their past memberships of the House Page Board simply disappear. Kelly ran away from a televised debate with her Democratic opponent John Hall. Emerson will not respond to her Democratic opponent, Veronica Hambacker, about why Emerson lied when she said she was never a member of the House Page Board. Emerson was a member during Session 1 of the 106th Congress, during the time of the first allegations against Foley and three years after the incident involving her then Board chairman, Kolbe. In addition to Wilson, Kelly, and Emerson, other GOP members of the House and Senate are involved in the Republican child sex predator ring. The GOP for a number of years has used pipelines like the House and Senate page systems, political indoctrination organizations like the College Republicans, juvenile "diversion programs," and religious-connected political groups to groom young men for the predatory sexual appetites of older male Republican members of Congress. The Pennsylvania College Republicans, founded by current GOP representative Phil English with the support of current Republican Senator Rick Santorum, appears to be one such "recruitment" center. It is not just young men who are groomed by the GOP child predators. Republican Jerry Weller of Illinois, who is married to the daughter of Guatemalan ex-dictator and "fundamentalist Christian" Gen. Efrain Rios Montt, is rumored to have conducted an illicit affair with a 16-year old female page. This comes at a time when the United States and John Bolton, a supporter of Rios Montt, are attempting to sway the United Nations into electing Guatemala to the UN Security Council. Former CIA officials, well aware of the human rights abuses by past and current members of Guatemala's oligarchy, have leaked information on Weller, former CIA Director Porter Goss, and others to point out that the agency, as far back as the 1960s when Goss served as a Latin American CIA agent, routinely used Mexican and Guatemalan female teens to entrap U.S. and foreign political figures. The "outing" of right-wing Republican Weller, the son-in-law of Rios Montt and champion of Guatemala's oligarchy as a predator is no coincidence. It is meant to shine a light on this old blackmail network during a time when Goss' name and those of jailed Rep. Randy "Duke" Cunningham, Kyle "Dusty" Foggo, Brant "Nine Fingers" Bassett, and ADCS contractor Brent Wilkes have arisen again during a House Intelligence Committee investigation of Shirlington Limousine and the transport of young prostitutes of both sexes to "poker parties" at the Watergate and Westin Hotels in Washington, DC. House Intelligence Committee Chairman Peter Hoesktra, Goss's replacement, seems more interested in covering up this information than in discovering the truth.

Posted by We The People | October 22, 2006 8:23 AM

At least bold-spam in a relevant thread.

The gut-check response to the point figure isn't sufficient. But smart people with enough statistical knowledge to judge aren't looking at the point figure. The basic question is 'has the invasion made things in Iraq worse?' The answer is 'yes, and a whole lot worse'.

As for the humanitarian justification, it was declared inoperable by HRW in the spring of 2003. There was never any way in which invading Iraq had any humanitarian rationale. That's not to say there might have been a point when that applied, should another campaign of large-scale repression and murder have begun. But that wasn't happening.

Posted by pseudonymous in nc | October 22, 2006 10:05 AM

Tim Lambert, at his blog Deltoid, has pretty much skewered Kaplan's arguments. The first is completely incorrect, and the second does not stand up to scrutiny.

I would trust Lambert a lot more than I would trust Kaplan, since Lambert is a computer scientist who knows a bit more about statistics than does Kaplan (as, for that matter, do I).

Posted by raj | October 22, 2006 10:48 AM

Starting a war for humanitarian reasons is like throwing an orgy to promote chastity.

Posted by rea | October 22, 2006 10:49 AM

The broader point here is that you are unlikely to be able to publish a piece of epidemiology in the Lancet without overcoming at length a lot more demanding criticisms from much more knowledgable sources than Fred Kaplan. The fact that professors with statistical training can be found to offer insta-criticism of the results for Kaplan and others does not alter this fact. Much academic insta-criticism is not up to the standards of publishable critique of peer-reviewed articles.

Posted by otto | October 22, 2006 11:19 AM

Starting a war for humanitarian reasons is like throwing an orgy to promote chastity.

An orgy to promote chastity? Now there's an insane policy I can support!

The authors of the Lancet study point out the fact that there's a survivor bias in their work, and that the results could be understated:

"Families, especially in households with combatants killed, could have hidden deaths. Under-reporting of infant deaths is a widespread concern in surveys of this type," the authors say. "Entire households could have been killed, leading to survivor bias."

Posted by Carl | October 22, 2006 11:25 AM

I'm sorry, people who reside on streets that cross main streets (yes, that's who they actually surveyed) are more likely to be killed on the thoroughfares than (1) people who actually live on the thoroughfare and (2) people who live on streets parallel to the thoroughfare how?

Posted by calling all toasters | October 22, 2006 11:55 AM

A more "measured (and) critical" Fred Kaplan = dried shit.

Posted by calling all toasters | October 22, 2006 11:57 AM

Fred writes:

If the Hopkins researchers want to claim that their estimate is more reliable than the United Nations', they will have to prove the point. It is also noteworthy that, if Iraq's preinvasion mortality rate really was 5.5 per 1,000, it was lower than that of almost every country in the Middle East, and many countries in Western Europe.

Delong writes:

Open your CIA World Factbook and look for death rates. You will find--in addition to the 5.5 per thousand number for Iraq--numbers like these: Syria 5.0... Turkey 6.0... Egypt 5.2... West Bank 4.5... Iran 5.6... Middle Eastern countries, like all countries undergoing population explosions, have young populations and so thus (with modern public health in place) low peacetime death rates.

Mike says:

Fred Kaplan is a hack.

Posted by Mike | October 22, 2006 12:18 PM

Well it can't be the case that *any* positive death rate destroys
the case for intervention. If that was the case you could say that D-Day couldn't be justified on humanitarian grounds because France's civilian death rate in 1944 and 1945 was higher than it was in 1943. The same would apply to Kosovo comparing 1999 to 1998. Rather, humanitarian intervention has to be reasonaly certain of leading to a sustained and substantial improvement in the humanitarian situation after the initial military engagements.

Posted by rd | October 22, 2006 12:25 PM

If I were more cynical, I'd suggest the only solution is to seed the country with Jews, since hating the Jews is the force that binds every other race, color, and creed together into a common band of humanity, at least for the past few thousand years.

Jfaberuiuc: Jews? Maybe. Gay jews? Definitely.

Posted by max | October 22, 2006 12:29 PM

This Halloween I'm going as the Tree of Liberty. A bloody Tree of Liberty that cries out "Bloood".

Betcha didn't know that Jefferson inspired The Little Shop of Horrors (the fiction and the foreign policy).

Posted by md 20/400 | October 22, 2006 12:58 PM

These two paragraphs from the Iraqi blogger "Riverbend" should get more attention, I think.

http://riverbendblog.blogspot.com/

"We literally do not know a single Iraqi family that has not seen the violent death of a first or second-degree relative these last three years."

And:

"Let's pretend the 600,000+ number is all wrong and that the minimum is the correct number: nearly 400,000. Is that better? Prior to the war, the Bush administration kept claiming that Saddam killed 300,000 Iraqis over 24 years. After this latest report published in The Lancet, 300,000 is looking quite modest and tame. . . ."

Posted by gar | October 22, 2006 2:05 PM

All right, I've written a detailed response to Kaplan.

Posted by Tim Lambert | October 22, 2006 2:19 PM

Kaplan's a total hack. You can't pull your favorite pre-war death rate estimate out of some old report and subtract it from the Lancet estimate of post-war death rates. The two numbers were collected using totally different methodologies. The whole point of the methods used by the Lancet authors is to subtract the pre-war death rates stated by the *same* households in the *same* survey in response to the *same* questions as those used to get the post-war death rates. That way systematic biases caused by teh survey methodology and questions will tend to cancel each other out. Not all of them -- notably memory may be worse for older deaths -- but many.

Posted by MQ | October 22, 2006 6:23 PM

Kaplan's been a pretty consistent critic of the war, from before it was cool. "Hack" is pretty unfair. The guy is a professional writer and he doesn't understand statistics. Cut him a break.

Posted by JP | October 22, 2006 6:42 PM

Matt,

You give Kaplan too much credit. His current critique is about as innumerate and exagerated as his first one.

In Hopkins II, the investigators interviewed around 1800 households, comprising about 12,000 persons. The sample reported 2 violent deaths for the 14 months preceding the war, and 300 violent deaths for a 40 month period after the war. They found the overall death rate doubled from roughly 5.5 per 1000 pre-war to roughly 13.2 per thousand post-war. When you scale up the deaths from those 12,000 to the total population you come up with the 650,000 excess deaths estimate.

1. Kaplan claims the 5.5 per 1000 is much less than a UN estimate of 9.3 per 1000, which he implies is more accurate without any evidence. But why should that make a difference? If Hopkins study underestimated pre-war deaths it should have also under-estimated post war deaths. Moreover, estimates of mortality are based on survey methodology and the time period surveyed. Kaplan has no idea whether the UN figures are accurate. In fact they purport to be an average for the period 1995-2000. He has no idea how they were figured or for what time period. They were likely based on surveys from 1995 to 1999, during the worst of the sanctions, and projnected forward, with no better methodology than the Hopkins study. Both the CIA and the U.S. Census estimated mortality for 2002 of around 5-6% for Iraq and neighboring countries, which is the same as Hopkins for the 2002-2003 time period. Remember these are all rough estimates for different time periods

2. He claims Hopkins over-sampled households near "main streets" and this so biases the study that it is invalid. This is wildly over-exagerated even if there is some heightened risk of violent death for those frequenting main streets. First the study randomly selected a house on a street intersecting the main street. The house selected could have been quite far off the main street. It should be possible to estimate how much of the population was not near any "main street" by sampling some maps. My guess is it will be small, and the critics should undertake this. Secondly, all households have to travel on main streets, so the heightened risk of living nearby may be quite small. For the same time period, Hopkins I and Hopkins II came up with similar estimates of excess deaths, even though Hopkins I used GPS coordinates and tended to favor less populous areas. Hopkins showed that over the pre-war period the violent death rate has been increasing to that now it is 4 times as high. This trend is matched by the official Iraqi casualty figures which shows a simlar trend. This suggests that Hopkins is not oversampling violent areas in any significant way.

The main street effect, if it exists at all, is worth some further investigation. But Kaplan makes a wild exageration to claim that it is either proven or is of such a magnitude as to invalidate the study.

Posted by chew2 | October 22, 2006 6:51 PM

"...for a war allegedly justified at this point on what are largely humanitarian grounds to have any positive excess death rate is a scandal."

It's a scandal too, for a war waged ostensibly as revenge for 9/11. If just 3000 of the excess deaths are innocent civilians, or even 9000 if we prorate the 9/11 death toll over 3 years...

You see my point.

Posted by Grumpy | October 22, 2006 10:17 PM

He doesn't get explicit about this, but for my part the point would be that for a war allegedly justified at this point on what are largely humanitarian grounds to have any positive excess death rate is a scandal.

Matthew's problem is that this sentence is directly contradicted by this post above: "This suggests that any collapse of the Baath regime, whether or not precipitated by direct US military intervention, was likely to lead to a fairly bloody outcome..."

For purposes of determining whether the invasion is justified on humanitarian grouds, the relevant comparison is to what would have happened upon an alternate demise of the Baath regime, not to the pre-war deaths. After all, the Baath regime was not going to last forever.

Just as a though experiment, let's do a thought experiment. In Scenario A, the US invades in 2003 and there are 100,000 excess deaths for the next 5 years (that is, deaths in excess of the 2002 baseline). In Scenario B - what Matthew terms te "fairly bloody outcome" that would occur on a Baath collapse even without a US invasion - the US does not invade in 2003, but the Baath regime falls in 2006, and after fall of the regime there are 300,000 excess deaths (that is deaths in excess of the 2002 baseline) per year for the next 2 years. So after 5 years, under scenario A, there are a total of 500,000 excess deaths, while under scenario B there are a total of 600,000 excess deaths. Which would the "humanitarian" result? I would say scenario A. Now, of course, you can argue with the numbers, etc. So perhaps using different numbers you would still find that the US invasion is not the humanitarian result. But I think THIS is the way one needs analyze the problem (and, of course, this is just analyzing the number of deaths - any true analysis of the humanitarian nature of the war would have to look into many more variables, such as the fact that there is some semblance of democracy there now, etc.).

Posted by Al | October 22, 2006 10:56 PM

Well it can't be the case that *any* positive death rate destroys the case for intervention. If that was the case you could say that D-Day couldn't be justified on humanitarian grounds because France's civilian death rate in 1944 and 1945 was higher than it was in 1943.

D-Day could not be justified on humanitarian grounds, with regard to the population of France. It was justified on the grounds that the Allies were at war, for classic just-war reasons, with Germany -- namely, Germany was waging a war of aggression. Similarly, the bombing and invasion of Germany itself could not be justified on humanitarian grounds.

Humanitarian grounds are not the ONLY reasons for which wars may be justified. Wars which cannot be justified on humanitarian grounds may be justified on other grounds. The problem is that in Iraq, none of the other potential justifications seem to work, either.

Posted by brooksfoe | October 23, 2006 5:59 AM