Worst. Article. Ever.

Well, I haven't done an exhaustive study. But this article on counting casualties in Iraq is embarassingly bad, even by the standards of cursory reprints of press releases. As it happens, I've been writing about casualty counts in Iraq (so I'm afraid you're going to be hearing much more about this over the next month or so), and I'm flabbergasted by its bizarre omissions. I don't think that any credible person who has spent any time on the debate would be satisfied with this particular attempt to deal with it.

For starters, it repeats nearly uncritically the results of a survey from ORB which purports to find a million casualties in Iraq. Pretty much everyone I've spoken to regards their numbers--which utilize a murky methodology on a very small sample, and then publish bizarrely tiny uncertainty estimates on a survey in a bloody war zone--as a bit of silliness. It fails to mention that the largest survey using the best regarded methodology, which was just released by the WHO earlier this month, found about a tenth as many deaths.

But the really strange thing is that it drags in the first Lancet study without mentioning the second.

Medical journal The Lancet published a peer-reviewed report in 2004 stating that there had been 100,000 more deaths than would normally be expected since the March 2003 invasion, kicking off a storm of protest.

The widely watched Web site Iraq Body Count currently estimates that between 80,699 and 88,126 people have died in the conflict, although its methodology and figures have also been questioned by U.S. authorities and others.

ORB, a non-government-funded group founded in 1994, conducts research for the private, public and voluntary sectors.

It's as if he's unaware of Lancet II--which unawareness is really, really, really hard to achieve, because it's the main thing you get on either a google search or a Nexis of Iraqi casualties, or any plausible variant on those terms. Even Lancet I's authors would say that the study, which was published in 2004 on the cheap and had a very small sample, has been superceded by the larger, better funded, and more recent Lancet II, which found 601,027 violent excess deaths (654,965 total).

Comments

Eh, what's a power of ten here or there when you've got an axe to grind?

Posted by anony-mouse | January 31, 2008 5:29 PM

After the travesty that was the headline on her last post about Iraq war casualties, I'm surprised Megan's headline didn't read: 'No Iraqi's Died In Iraq War'.

Posted by RICKM | January 31, 2008 5:59 PM

I much prefer your method of research, Megan, which is to automatically trumpet the lowest public number, in a way which understates even that result.
Once you pick up a 2x4 you just can't put it down, can you?

Posted by brad | January 31, 2008 7:22 PM

Hmmm...I haven't seen Megan "trumpet" only the lowest number, nor have I seen her claim that no Iraqis have died.

I'd like to see more people covering this issue.

Posted by ed | January 31, 2008 8:03 PM

"utilize a murky methodology"

That's for sure. ORB's first press release from September claimed then that it was a "nationally representative sample".
http://www.opinion.co.uk/Newsroom_details.aspx?NewsId=78

(ORB seems to claim this reflexively for all its press releases.)

But then they came back later saying it wasn't any such thing. Now they say it basically had no "rural" coverage and they've done new interviews in "rural" areas. Supposedly we're to believe this now makes it "nationally representative", as it wasn't when they first claimed it was.

First, they never define what constitutes an "urban" or "rural" area for the purpose of their poll. Second, if the first effort (which still provides most of the deaths in their "revised" estimate) was actually based on random selection there's no way they would have just happened to miss all the "rural" areas. Obviously it was not a random sample, which means extrapolating it to an estimate like this is groundless. Additionally, the poll gives no indication at all about how respondents were selected within these non-randomly selected locations.

So neither the location selection, nor the respondent selection within these locations, have any credible claim to randomness.

Also, since when are opinion pollsters credible sources of mortality estimates? Has ORB ever produced a mortality estimate? Do they have any idea of the problems in doing such a thing? I don't think so. This seems like their first attempt, and they don't seem to know what they're doing (as their laughably small error margins they claim here, among other things, attest).

Also it seems that nobody directly from ORB actually did any of the work. They contracted it out to some Iraqi firm which was started in 2003 by someone named Munqith Daghir, a self-trained pollster who doesn't appear to know much of anything about random sampling:

"I knew that Baghdad is distributed into nine different areas, and how many citizens lived in each one. But to tell the truth, I didn't know anything about the real random systematic sample. We did it randomly by going to any house we wanted to go to. So it wasn't a perfect sample."
http://www.opinion.co.uk/Documents/Polling%20in%20Iraq.pdf

For good measure, he was also quoted in a National Journal article called "Iraq's Slippery Polls" as saying he started his polling work "as a patriotic job that could help in putting more pressure
on the government, or the invasion forces to withdraw."

So we have a crude opinion poll here masquerading as a scientific mortality estimate, conducted by someone who doesn't seem to know what he's doing and who has an agenda to use his poll results to drive US forces out of Iraq.

And from this we get a ludicrous estimate of "1 million" Iraqis killed.

Are we all surprised? Well, i'm certainly convinced!

Posted by paul | January 31, 2008 8:16 PM

I guess you missed this three weeks ago, ed.

Posted by brad | January 31, 2008 9:11 PM

ed-

Megan claimed that only 150,000 Iraqi's died, which is off by about a factor of 3. It was a bs lie and she knew it.

Posted by rickm | January 31, 2008 10:07 PM

It is important to get the numbers right but it is perhaps even more important to attribute the deaths properly. A very large proportion of the Iraqi civilians that everyone runs around referring to as having been "killed in The Iraq War" or (slightly more neutrally) "casualties in Iraq" were, in fact, killed either by other Iraqis or by foreign Muslims who travelled to Iraq to murder them. This is not a trivial point (at least, it shouldn't be!) yet peoples' continued refusal to acknowledge it leads to a surreal and infuriating public debate in which only the U.S. is seen as an autonomous actor. It is almost as if the six million Jewish holocaust victims were never referred to as anything other than "casualties of the U.S. war against Germany".

So I'd be happy to provisionally accept a larger estimate of the number of people who have been killed in Iraq since 2003 if a certain self-styled 'war critic' faction would be willing to wake up and actually acknowledge by whom many of them were killed, which to this point they have not.

Posted by Sonic Charmer | January 31, 2008 11:04 PM

Megan, I've read this post almost five times and have know idea what your point is. At first, given the title worst this or that ever, I thought you'd disagree that a million Iraqis have died since GWB saved them.

Then you criticize the study's methods using one of the all time great rhetorical devices that none of your friends agree with the one million figure, it's silly.

But then you end the post by trotting out what you call Lancet II which shows 600,000 deaths.

Why, then, is one million figure silly, but the 600,000 figure accurate?

It sure sounds to me like you want more Iraqi deaths, which wouldn't suprise me given your blood lust during the run up to the war itself.

Posted by Mortimer Madler | January 31, 2008 11:07 PM

600,000 is more than enough.

Posted by gerontion | February 1, 2008 12:56 AM

A million dead Iraqis? Please. That does not even pass the smell test. This idiot study no more requires exhaustive research to debunk than would a claim that 10,000 people were murdered in Boston last year. Those numbers are just way the fuck off.

Posted by American War Correspondent | February 1, 2008 4:06 AM

Is there no appreciation of the role the USA and UK have played in destablising Iraq? Tens of thousands of civilians have died as a direct result of US bombs, hundreds of thousands of innocents due to a decade of sanctions.

To blame fellow Muslims for most of the murder and carnage in Iraq is to hide from the conditions that made all this possible.

Is it a million, is it half a million or 250,000. Does it matter? There's little doubt that the USA used the excuse of 9/11 to invade a country for its oil.

Shame on those Americans (and Brits) who use smokescreens such as debates about the number of deaths to detract from the responsibility UK/US has for so many innocent deaths.

It was a poor article but at least Reuters bothered to mention it, unlike most of the Western press who take their cue from the politicians.

Posted by AM - United Kingdom | February 1, 2008 6:08 AM

...blame fellow Muslims for most of the murder and carnage in Iraq...

I'd say it's fair to place quite a bit of blame on the fellow Muslims who are actually pulling the trigger.

Posted by CatCube | February 1, 2008 6:30 AM

AM:

To blame fellow Muslims for most of the murder and carnage in Iraq is to hide from the conditions that made all this possible.

Seems to me one should blame 'fellow Muslims' precisely if and when those 'fellow Muslims' actually committed the murders - nothing more and nothing less. To do anything else is morally retarded, it displaces blame from where it belongs and sets up a perverse incentive structure according to which people who hate the U.S. see that they can stick-it to us, and Bush, by going to Iraq and murdering some Iraqi.

The current rules according to which any human killing any other human inside the boundaries of Iraq after 2003 is seemingly tallied as having been 'killed by the US', 'in the Iraq War' are bizarre and sick and, incidentally, do the Iraqis no favors.

Posted by Sonic Charmer | February 1, 2008 6:55 AM

I'm flabbergasted by its bizarre omissions. I don't think that any credible person who has spent any time on the debate would be satisfied with this particular attempt to deal with it.

You're flabbergasted because you're treating an article in the MSM as if were meant to be part of a credible debate among informed people rather than meant to be a piece of propaganda aimed at those who aren't paying close attention.

Posted by Ralph Phelan | February 1, 2008 9:08 AM

I think you are being harsh.

Look at the by-line. It is Reuters. Although they do credit the reporter and editor, they are named as an afterthought. It is wire service notification. The reporter is just notifying readers of the study's existance. The NY Times put the inappropriate headline there. If the headline were "New study claims 1 million dead in Iraq", the article would read differently.

The notice could certainly have been better, listing the WHO and Lancet II studies claims, but it was certainly not intended as a critical analysis of the data. That's not how news works. First reports are simply about existance - "This story exists!". Analysis comes later, if the first story sparks any interest.

Posted by Njorl | February 1, 2008 9:48 AM

> Is there no appreciation of the role the USA and UK have played in destablising Iraq? Tens of thousands of civilians have died as a direct result of US bombs, hundreds of thousands of innocents due to a decade of sanctions.

Right... the Iran-Iraq war was benign in comparison. And in the 90's, we should've waited to see what other invasions Saddam might've attempted. Had we not stopped the growth of his military development (including nuclear weapons research) and checked his desire for acquiring nearby sovereign-state oil assets, the region could be more stable now and less of a threat to global policy.

Posted by solarlux | February 1, 2008 10:11 AM

That's not how news works.
Yeah, the way "news" works is that when a liberal or leftist organization publishes a tendentious study Reuters and the New York Times reprint their press release above a reporter's byline.

I don't think she's being to harsh on Reuters - I don't think that's physically possible. I do think she's being way too naive about them.

Posted by Ralph Phelan | February 1, 2008 10:12 AM

I can't remember where I saw it, but apparently the WHO methodology suffers from a downward bias in its casualty estimates. In addition, the WHO results and Lancet 2 do agree, if you do some extrapolation, since Lancet 2's figures refer to excess deaths assuming that their data on the pre-war baseline is accurate, while the WHO top-line figure is deaths due to violence. You can infer from the report that the relative risk of death in WHO = relative risk of death in Lancet 2 when comparing the pre-war to the post-war period. In other words, the conclusions are the same! Wars are bad, they increase the risk of death quite a bit. The rest is accounting.

Posted by Martin | February 1, 2008 11:13 AM

'Yeah, the way "news" works is that when a liberal or leftist organization publishes a tendentious study Reuters and the New York Times reprint their press release above a reporter's byline.'

I went to reuters.com. Searched for Iraq and read the first story on casualty numbers. It had the inflammatory headline of:

"FACTBOX-Military and civilian deaths in Iraq"

The numbers it used were the IBC numbers (Civilians Between 80,744 and 88,173) - a source that admits to gross undercounting of civilian casualties.

Goddamn anti-American islamofascits liberal bastards!

Posted by Njorl | February 1, 2008 12:10 PM

AM:
“To blame fellow Muslims for most of the murder and carnage in Iraq is to hide from the conditions that made all this possible”.

Sonic Charmer
Seems to me one should blame 'fellow Muslims' precisely if and when those 'fellow Muslims' actually committed the murders - nothing more and nothing less. To do anything else is morally retarded, it displaces blame from where it belongs and sets up a perverse incentive structure according to which people who hate the U.S. see that they can stick-it to us, and Bush, by going to Iraq and murdering some Iraqi.

The current rules according to which any human killing any other human inside the boundaries of Iraq after 2003 is seemingly tallied as having been 'killed by the US', 'in the Iraq War' are bizarre and sick and, incidentally, do the Iraqis no favors.

Sure the Muslims pulling triggers bear most of the responsibility, but does that really allow us to wash our hands so neatly? Should we really ignore the responsibility of creating the conditions? I suggest you compare the number of deaths due to civil war in Iraq in the five years before Mar 03 (maybe 10,000) and the five years afterwards (200,000+). It might be instructive.

For another example, a guard goes nuts at an insane asylum, releases the inmates, and kills a few doctors. Is he responsible only for the deaths of the doctors, or does he bear some of the responsibility for the murders committed by the escapees? After all, those deaths would not have happened except for the actions taken by the guard.

Posted by Donald Clarke | February 1, 2008 12:22 PM

"I can't remember where I saw it, but apparently the WHO methodology suffers from a downward bias in its casualty estimates."

You remember incorrectly. It said this and then raised it's estimate to account for it.

"In addition, the WHO results and Lancet 2 do agree, if you do some extrapolation, since Lancet 2's figures refer to excess deaths assuming that their data on the pre-war baseline is accurate, while the WHO top-line figure is deaths due to violence. You can infer from the report that the relative risk of death in WHO = relative risk of death in Lancet 2 when comparing the pre-war to the post-war period."

No, they don't agree. Lancet estimated 600,000 deaths from violence. WHO estimated 150,000 (even after adjusting upward substantially, which Lancet did not do).

"if you do some extrapolation" = if you make up your own "excess death estimate" for the WHO where they didn't give one. The only place the WHO study says anything relevant to this is where they state that the pre-war numbers would tend to suffer from more downward bias than more recent years, which means, according to about the only thing WHO does say about it, the "extrapolation" you're doing to get an excess deaths estimate would produce an inflated number.

And even making up that inflated excess deaths estimate it still comes out lower than Lancet by over 200,000.

The two studies (or any others) might agree on some broad thing like "war is bad", but by this standard everything can be "the same", no matter how drastically they differ.

Posted by paul | February 1, 2008 1:25 PM

AM: Why no blame for a certain Iraqi named Hussein for that "decade of sanctions", let alone the deaths from it due to his gaming of the Oil-for-Food regime and desire to use the very conditions in Iraq that he created as a weapon against the sanctions regime?

Or for his actions in invading Kuwait, a "fellow Muslim" state, and starting the entire process? How about starting a war with Iran, also a "fellow Muslim" state?

The United States and The West are not the prime mover, responsible for all actions and reactions and all states of being in the world; to argue as if they are is to deprive the world's other nations - including and in this case especially the Muslim ones - of their accurate status as actors on the world stage, rather than merely puppets and victims of the One Motive Force.

(The implicit idea that "stability" is the state to be aimed at, and "destabilizing" is bad in-and-of-itself, is a trap that I won't critique in detail here, but suffice it to say I reject it entirely.

Just as Iraq was "stable" under the Ba'athist stranglehold, Europe could have been "stable" under the Fascist or Communist boot for decades or centuries; that does not suggest that fighting against any of them would have been unwise or immoral.

No such authoritarian, totalitarian State has any moral legitimacy or deserves any respect for its "sovereignty", and the obligation the world has to the people of such a State is to liberate them from it, if possible, or failing that to encourage its downfall from internal pressures.)

Posted by Sigivald | February 1, 2008 1:29 PM

Sure the Muslims pulling triggers bear most of the responsibility, but does that really allow us to wash our hands so neatly? Should we really ignore the responsibility of creating the conditions?

If you want to discuss the responsibility for creating the conditions, then we need to back up to the 1970s, and revisit a certain policy called "Twin Pillars" under which the United States determined that stability was paramount, and the US would support two entities in the Middle East under the guise of keeping things "stable". One was Saudi Arabia, the other was Iran. After Iran's government fell in the revolution, support was switched to Iraq, and a certain up-and-coming strong man named Saddam Hussein. The US egged on the Iraq-Iran war in order to weaken Iran, and continued to support Saddam afterward without looking too hard at things like brutalization of the populace until the first time he acted outside of US oil designs: invading Kuwait.

Gulf II was, without a doubt, one of the modern world's uglier Undo commands. But don't get the idea that responsibility starts there, or some sort of recent mistake; responsibility for the current state of things started with Cold War mistakes from 35+ years ago.

Posted by anony-mouse | February 1, 2008 1:56 PM

Njorl 12:10 pm

When I say that Reuters is biased and lazy, it doesn't mean that biased will trump lazy 100% of the time. Sometimes lazy wins.

Posted by Ralph Phelan | February 1, 2008 1:56 PM

Donald wrote: "For another example, a guard goes nuts at an insane asylum, releases the inmates, and kills a few doctors. Is he responsible only for the deaths of the doctors, or does he bear some of the responsibility for the murders committed by the escapees? After all, those deaths would not have happened except for the actions taken by the guard."

Come on - are you really equating Muslims to inmates at an insane asylum? People are kept in insane asylums because they are a danger to themselves and others. Your analogy indicates you believe that Iraqis and the foreign fighters that have flooded Iraq since the war began should be kept under strict authoritarian control because they lack the sanity to make proper moral decisions about killing people.

that's an odd statement, to say the least.

Posted by RMH | February 1, 2008 2:04 PM

Actually, I kind of like Donald's theory that Arabs are, essentially, insane and need to be kept locked up, and the various dictatorial regimes that the U.S. cozies up to are the asylum keepers, and that instituting "democratic" reforms in an Arab country would be like letting asylum inmates elect the guards.

And the World Court ordering Israel to tear down its fence would be the equivalent of some federal judge ordering deinstitutionalization. Although no one ever suggests that someone in that position should be held responsible for the resulting homicides.

Posted by y81 | February 1, 2008 2:20 PM

AM United Kingdom,

Is there no appreciation of the role the USA and UK have played in destablising Iraq? Tens of thousands of civilians have died as a direct result of US bombs, hundreds of thousands of innocents due to a decade of sanctions.

So what allegedly superior alternative policy do you propose we should have followed? Lifting the sanctions and allowing Saddam to continue his record of genocide, gross human rights abuses, and military aggression? Let's review: He first invaded Iran, causing the Iran-Iraq War. Estimated casualties: 1 million to 1.5 million. Later, he invaded Kuwait, causing the Gulf War. Estimated casualties: 100,000 to 200,000. He also threatened wars against Syria and Israel, and actually attacked Israel with missiles. He repeatedly engaged in military aggression toweards his neighbors even after his defeat in the Gulf War. He had also murdered and tortured hundreds of thousands of his own people in Iraq. There was no indication that this record would come to an end. There was no indication that Saddam was at risk of being deposed by internal Iraqi forces. In fact, he was grooming his equally thuggish sons to take over when he eventually retired from power. He was a genocidal maniac who for two decades had been a continual source of violent conflict and instability in a region of the world vital to the interests of the United States and the global economy.

Posted by Mixner | February 1, 2008 4:53 PM

Megan, having a bit of problems with my post containing links?

Posted by Barry | February 1, 2008 5:19 PM

Is it a million, is it half a million or 250,000. Does it matter? There's little doubt that the USA used the excuse of 9/11 to invade a country for its oil.

Why do people still believe that we invaded Iraq for the oil? No effort has been made to take the oil. In fact, it looks like money from the sale of oil will go back to Iraqis, either directly or indirectly.

How many excess deaths were caused by the US invasion of Europe during WWII? I suspect that a lot more people died in WWII than would have died under Hitler if we'd left them alone. Same goes for the Civil War. Wars kill lots of people.

The UN is responsible for a lot of the problems in the Middle East. The UN enabled the Oil for Food scam and the UN culture of corruption made it easy for Saddam to game the system. The UN is also responsible for keeping Palestinians locked up in barely livable camps and for turning a blind eye to UN refugee camps being used as bases for terrorism.

During the 1970's we were focused on a much more dangerous enemy... Iraq and the Middle East were a proxy battlefield of the Cold War against the USSR. What we did then may not have been the best thing, but our leaders were not trying to build a stable, peaceful Middle East, they were trying to contain the USSR. It's pointless to criticise our Middle Eastern foreign policy out of the context of the Cold War.

I do find it hard to believe that the death rate in Iraq was so low before the war. We keep finding mass graves in Iraq. And we know that Saddam killed whole villages just to test chemical weapons. Actually, it doesn't matter... even if the death rate were low, much of it would have been directly a result of Saddam's murderous actions. And his sons were worse, from what I heard.

Saddam may not have had WMDs, but he had programs that could have been restarted at a moment's notice. Making chemical weapons doesn't take very long if you have a chemical plant. Making biological weapons probably wouldn't take very long if you have the resources of s government. His nuclear weapons program may not have been very far along, but he would have started it back up as soon as he could.

We defeated Iraq in the Kuwait war but he never quit fighting. Continuing the sanctions and letting them continue to fire on our planes wasn't helping. If Iraq had accepted the terms of surrendor and actually cooperated, none of this would have happened. As it was, we had a choice between continuing the status quo (corrupt oil for food, sanctions, continued low-level fighting) or invade and end it.

And Saddam DID have contact with Al Qaida and other terrorist groups. They made no effort to hide the fact that Iraq paid $25,000 dollars to the families of suicide bombers in Israel. They allowed terrorists to use their land for safey and training.

If we'd done a WWII style invasion with carpet bombing and all, a lot more Iraqis would have died. Our soldiers try not to kill civilians, but it's hard when the enemy hides among them. It sucks that so many innocent Iraqis have died. I wish we could have managed to only kill bad guys. But war isn't nice like that.

And part of the problem is that we screwed up. The invasion wasn't handled very well. Our strategy wasn't what it should have been. We've learned and changed. That happens in war as well as every other human endeavor.

I still find it very hard to believe that an average of 500 Iraqis died by violence every day since the invasion over and above the "normal" death rate. Though I'd like to deduct dead terrorists from that number...

Posted by Earnest Iconoclast | February 1, 2008 5:33 PM

Why do people still believe that we invaded Iraq for the oil?

Probably because a hackneyed ephitet is the first resort of a lazy and ignorant mind.

Historically speaking, the US has used oil interests as a primary cause for meddling in ME affairs. The consequence was a tendency to produce terrorists. It was only at 9/11 that we discovered that the terrorists were becoming more sophisticated and violent than anything we had previously given them credit for, and this in spite of Clinton applying the classical leftist solution for eight years: attempting to back down the levels of US military capability and treating terrorism as primarily a law enforcement problem.

Bush, after intially drifting in an isolationalist direction, responded to 9/11 by calling war on all that had worked to produce it. And thus lazy, ignorant minds saw the US go from Afghanistan into Iraq, and assumed that since US interventions in the mideast were traditionally about oil, this one must be also. And out pops the ephitet.

Posted by anony-mouse | February 1, 2008 6:17 PM

To be fair, the war in Iraq is about the oil in the sense that Saddam (or his sons) would have most likely attempted to conquer his neighbors. The resulting war would have either ended with him (or maybe Iran) dominating the region, which would have threatened a significant chunk of the world's oil supply and given the ruling party a lot of power or it would have ended in a war that devastated the region and reduced the oil supply, causing massive economic harm to the rest of the world, mostly in places that are too poor to find alternatives.

Until the world has a better energy supply than oil, what happens in the Middle East matters to the rest of the world. And a power-hungry nutjob with delusions of grandeur is a threat to many people. A serious reduction in the oil supply would cause economic harm which would end up with actual people dying from lack of resources. And most of those people would be in poor countries. The US has other options for drilling for oil and for using other power sources and the money to suck it up to some extent.

But that's not what the "No Blood for Oil" people mean... they no doubt believe that tankers are shuttling the oil to the US as fast as possible.

Oh, I have to go, my share of the oil is here...

Posted by Earnest Iconoclast | February 1, 2008 7:23 PM

Donald Clarke:

Sure the Muslims pulling triggers bear most of the responsibility, but does that really allow us to wash our hands so neatly?

Who's arguing that we can "wash our hands"? To repeat myself: one should blame 'fellow Muslims' precisely if and when those 'fellow Muslims' actually committed the murders - nothing more and nothing less. This says nothing about the actual casualties of the war (as opposed to the casualties of terrorist and mafia-style violence), of which I most certainly do not suggest we can "wash our hands".

Should we really ignore the responsibility of creating the conditions?

I reject this notion that we "created the conditions". What "conditions" did we create which somehow forced terrorists to travel to Iraq to blow up marketplaces and mosques? What about the mentally-retarded female bombers used today by terrorists: did we somehow "create" "you-can-use-female-mentally-retarded bomber" conditions? No we did not.

Nothing in the laws of physics somehow compels terrorists to attack Iraqis as a consequence of the 2003 U.S. invasion of Iraq. They are doing this of their own accord, because they want to. There are no objective "conditions" one can point to which force or make them do it. To say otherwise - again - displaces blame from where it properly belongs, on people who plot mass murder. As Sigivald points out, when you speak this way you deny the people doing these things status as autonomous actors. You are treating the rest of the world - all non-Americans - as, in effect, automatons who only react mechanically to what we do or don't to.

Did Bill Clinton "create the conditions" which allowed Timothy McVeigh to murder 150+ people in Oklahoma, therefore deserving some of the blame? Can you answer this question decently and be consistent?

For another example, a guard goes nuts at an insane asylum, releases the inmates, and kills a few doctors. Is he responsible only for the deaths of the doctors, or does he bear some of the responsibility for the murders committed by the escapees?

As RMH points out, this is a problematic and perverse analogy. Pray tell, in your mind who played the role of the 'doctor' in the Iraq 'asylum'? Saddam Hussein, it would seem. Bizarre analogy.

More specifically, the comparison is simply inapt; killing doctors at an insane asylum and deliberately releasing criminally-insane patients is an irredeemably evil act of which no decent people can approve. Whatever you think of the Iraq invasion, reasonable people disagreed about its merits and it was not an unalloyed evil deed with no positive benefits: Destroying an evil and megalomaniacal autocratic regime is simply not an unalloyed evil thing to do. (I can't believe I had to say that.) It is certainly not akin to murdering a doctor. But as y81 alludes, it is somewhat telling that doctor/mental patient is the analogy that sprung to your mind when thinking about dictators who lord over Muslims....

Posted by Sonic Charmer | February 1, 2008 7:26 PM

I have to agree. That was a terrible article.
The best estimate for the period through June 2006 is 150,000. This was published recently in the New England Journal of medicine. The Lancet 2 study that claimed over 600,000 (through June 2006) is not a credible study.

I'm a little suprised that Reuters standards are so low. I hope this is an exception, not the rule, and that they will issue a correction. As it is, an ignorant or biased reporter uncritically passed along a howler of a "study" with absolutely no effort at checking its veracity.

Posted by Tim | February 2, 2008 7:42 AM

Let me start by saying that finding out what is actually happening in Iraq is very difficult, and that actually gathering that information requires admirable courage.

That said, I think that those that try and make sense of what’s going on there need to display at least a minimal level of introspection, humility, and recognition of how limited our information is.

In September 2007, the British polling survey firm ORB issued a survey from Iraq that suggested that more than a million Iraqi citizens had died violently since the invasion. The ORB website used a less neutral term: “murdered”.

ORB’s core competency seems to be the familiar western opinion survey by random phone interview. That isn’t actually very relevant to doing a cluster sample mortality survey in a war zone. And the survey work itself was done by an Iraqi firm, IIACSS, that didn’t exist before 2003, founded by an Iraqi with apparently only limited formal training in survey methodology.

This may not be as much of an issue in surveying public opinion. Similar opinions may be broadly spread through either the Iraqi population as a whole or regional subsets. If so, exactly where you sample may not affect results too seriously.

Violent mortality, on the other hand, can be lumpy. It’s pretty clear that some parts of Iraq are a lot more violent than others, and even when you get down to the local neighborhood some blocks or even houses can be can be lot luckier than others, or the reverse. Mortality surveys are very sensitive to even small sampling errors.

ORB doesn’t seem to have provided a lot of added value in terms of oversight and quality control. They originally reported that the survey was based on “a nationally representative sample.” Later, they admitted that the original survey was “undertaken in primarily urban locations” Given that about a third of Iraqis live in rural areas, this is a significant omission, and ORB failed to disclose the choice when the results were first published. They should have. When they did a follow up survey and sampled more rural areas, they reduced their initial estimate of violent deaths by about 200,000.

Their latest press release on the study indicates that it covered "112 unique sampling points". That is, it was a cluster sample of the sort used by other researchers in Iraq. It wasn’t the 2,163 independent observations you’d get if you did that number of random phone interviews. ORB calculated their margin of error as though it was that number of independent observations, not the much larger margin of error for a cluster sample of that number of households. This not only grossly understates the level of uncertainty in the estimate, but makes you wonder how well they understand this sort of work.

And even when theoretical sampling error is correctly calculated, that doesn’t include other sources of uncertainty. Researchers make subjective decisions on when to skip areas because of security or other issues, and when to do follow up interviews of clusters missed for these or other reasons, violating a truly random sample. Survey teams may curbstone, or invent responses, particularly when the risks of carrying out the survey are as real as they are in Iraq. There may be errors in tabulating the data.

The survey was originally published with a glaring error in Baghdad’s religious composition undetected. This does not speak well for ORB’s diligence in checking for possible error, bias or fraud.

Collecting the number, age and sex of household members is a powerful tool for checking the plausibility of the sample in a mortality survey. It is unfortunate that neither the ORB survey nor Burnham et al did so, since the absence of this data limits the credibility of the information gathered at considerable personal risk by the survey teams.

Follow up visits by supervisors is another powerful check of survey accuracy. I think it’s important for those publishing survey results in Iraq to disclose if this was done, as well as the number of sampling points and what factors, if any, were used in weighting the raw data.

Posted by Will McLean | February 2, 2008 12:00 PM

To me, a telling point is the reluctance of al-Reuters and the NY Times to let go of the badly flawed Lancet Studies, simply because they fit the anti-Bush, anti-American narrative too well to let them be discarded.

They crop up again and again, and even after being academically discredited against peer-reviewed rival studies as being off by a factor of ten - are now seen EMBEDDED as base ASSUMPTIONS - in reporters articles.

Whereas studies that are favorable to US interests, like the early electric generation capacity study of 2003 and the study that charted police brigades "stood up" in 2004 -were dead and properly buried by media within weeks.

The conclusion I have is that the authors of the Lancet studies (one who later ran as a liberal antiwar Democrat for Congress) and the MSM were complicit in making supposedly objective science and statistical analysis into a political weapon in the 1st study...then ignoring all objective criticism of the 1st study and creating the 2nd study primarily for it's political impact on the American and EuroLeft.

Posted by chris ford | February 2, 2008 1:40 PM

Chris, you really ought to check out what the studies really say. It's been widely booted about after all that the studies _agree_ with each other. Look up what a confidence interval is some time, and what it means when confidence intervals from two studies overlap.

I assure you that it's not some sort of 'liberal math'.

Posted by ScentOfViolets | February 2, 2008 6:58 PM

Chris, you really ought to check out what the studies really say. It's been widely booted about after all that the studies _agree_ with each other. Look up what a confidence interval is some time, and what it means when confidence intervals from two studies overlap.

The studies do not remotely "agree" with each other. The mere presence of an overlap in the 95% confidence intervals of two studies is not "agreement." If you narrowed the confidence intervals, even the overlap would disappear.

Part of the problem with sample-based studies is the use of a single point value to report their estimates. Lancet II is routinely reported as estimating 650,000 excess deaths, even though the confidence level that that number is even roughly correct is very low. The use of the midpoint value of a CI as an estimate is just a statistical convention. The 95% confidence interval is 393,000 to 943,000. You could also cite any other number in that range as an "estimate."

Posted by Mixner | February 2, 2008 8:14 PM

"Why do people still believe that we invaded Iraq for the oil? No effort has been made to take the oil. In fact, it looks like money from the sale of oil will go back to Iraqis, either directly or indirectly."

While there are some who naively believe that we invaded Iraq in order to physically steal their oil, it is none-the-less true that we did invade Iraq because of oil.

US foreign policy has always sought to keep the sources of imported oil diversified. Saudi Arabia is an extremely tempting target for domination by its much more populous and industrious neighbors. Iran and Iraq kept each other at bay for years, until we destroyed Iraq's military in the first Gulf War. After that, we were reduced to safeguarding Saudi Arabia directly with our own military. This led to a predictable backlash, weakening the friendly Saudi Royal family against the radical religious element.

In an attempt to solve this, we decided to try to reinstitute a balance between Iraq and Iran by destroying and rebuilding Iraq as a power that we could work with.

If it were not for the oil, we wouldn't be there. It is accurate to say that we are using our military primarily to ensure that we can obtain more oil for less money in the future, but no, we are not stealing it.

Posted by Njorl | February 3, 2008 2:42 AM

I am leary of imputing any one motivation - or any one set of motivations - to the people responsible for getting us into a war with Iraq. All you really need to know was that there apparently was no good stated reason, that the war and subsequent occupation has been a disaster, and that this is all to the tune of well over One. Trillion. Dollars.

Not a sign of good stewardship.

Posted by ScentOfViolets | February 3, 2008 10:56 AM

I see SOV is trotting out "the studies agree" argument again- this grossly strains the meaning of the word "agree". As Mixner has pointed out, just because there is some overlap between the confidence intervals does not mean that the studies agree. If one could state that as a reasonable argument, then any overlap would mean the studies agree, even if it was overlap of a single fatality.

The two studies give very different ranges for all confidence levels 95% and lower, thus they do not agree.

What one can say is that we don't know for certain which study is most valid, and it is possible that both studies were well conceived and executed attempts at estimation based on sampling. The weight of the evidence today suggests that Lancet II was in error for some reason. Based on the distribution of the assigned causes of death, I am wagering that the error arose from fabricated data; data fabricated by people who probably didn't think it prudent for their own well-being to collect actual data (Iraq is a dangerous place, and I can tell you honestly that I would have refused to perform such sampling for that very reason). I think they made up the data, and since violence is such dominant factor in the lives of Iraqis, made up data that heavily overweighted death by violence.

Posted by Yancey Ward | February 3, 2008 1:11 PM

As far as I'm concerned, the Iraq War was never about WMDs. That was just an exercise in PR and legalism. The war was about removing from power a genocidal maniac with a 20-year record of mass murder and military aggression in a part of the world of vital importance to the United States and the global economy.

I also think SOV's claim that war has been a "disaster" is absurd. The sanctions had killed hundreds of thousands of Iraqis by the time we ended them in 2003. If they had continued until now, they would likely have killed hundreds of thousands more. If we had ended the sanctions without military intervention, Saddam would have restarted his war machine, a machine he had already used to start two major wars and numerous smaller conflicts. It'll be a decade or more before anyone is in a position to properly evaluate the long-term effect of the war on Iraq, the United States, the middle east, or the world as a whole.

Posted by Mixner | February 3, 2008 1:25 PM

"The sanctions had killed hundreds of thousands of Iraqis by the time we ended them in 2003. If they had continued until now, they would likely have killed hundreds of thousands more."

I agree with you overall, Mixner, but wanted to point out that it wasn't exactly the sanctions themselves that killed so many Iraqis. It was primarily Saddam Hussein's distortions and abuse of the sanctions, such as using his money to build palaces and bribe French and Russian officials rather than actually trying to get needed supplies for his people.

Remember when Iraq under the sanctions was caught smuggling a large shipment of infant formula out of the country, presumably to sell it? But of course if Saddam had actually tried to help his own people, then there wouldn't have been as much pressure on the international community to drop the sanctions because they were supposedly killing babies.

Before Sept. 11, one of the first diplomatic efforts George Bush made when he came into office was to try to improve the sanctions program to make it easier to bring in food and medical supplies (but more difficult to bring in military supplies). Bush even got the Chinese to agree not to veto it, but the Russians refused to see a more efficient sanctions regime that would help the Iraqi people. And we know who paid the Russians and French to oppose that change.

Posted by Ann | February 3, 2008 2:42 PM

Pearl Harbor – 2400

Valley Forge – 2000 (six months about 10 a day)

Fredericksburg - 2000 (4 days, about 500/day)

Gettysburg – 8000 (3 days, about 2666 per day)

Normandy – 245,000 (81 days, about 3000 per day)

Battle of the Bulge – 35,000 (41 days, about 900 per day)

Posted by jj mollo | February 3, 2008 5:21 PM

Long post – sorry, did not get back to my computer over the weekend

Donald wrote: "For another example, a guard goes nuts at an insane asylum, releases the inmates, and kills a few doctors. Is he responsible only for the deaths of the doctors, or does he bear some of the responsibility for the murders committed by the escapees? After all, those deaths would not have happened except for the actions taken by the guard."

RMH, Y81 and Sonic Charmer, all pointed out my analogy had problems.

Come on - are you really equating Muslims to inmates at an insane asylum? People are kept in insane asylums because they are a danger to themselves and others. Your analogy indicates you believe that Iraqis and the foreign fighters that have flooded Iraq since the war began should be kept under strict authoritarian control because they lack the sanity to make proper moral decisions about killing people.
that's an odd statement, to say the least.
Posted by RMH | February 1, 2008 2:04 PM

You are right. My apologies to any offended Muslims. I was not intending to liken all Muslims, let alone Iraqi Muslims to inmates of an insane asylum where they are kept to protect themselves. A better description would be inmates of a stereotypical hellhole prison, who have been brutalized by the guards. We free the inmates from the guards, but instead of trying to reintegrate them into a structured civil society, we just say “Good luck, you are on your own” and point them to the horizon. We are then surprised when looting, rape, and murder break out.

Maybe a better example would be to use the case of a the man who shouts “fire” in a crowded theater. Even though he does not actually trample anyone to death, doesn’t he bear some of the responsibilities for the deaths in the ensuing panic?

Sonic Charmer also wrote

Nothing in the laws of physics somehow compels terrorists to attack Iraqis as a consequence of the 2003 U.S. invasion of Iraq. They are doing this of their own accord, because they want to. There are no objective "conditions" one can point to which force or make them do it. To say otherwise - again - displaces blame from where it properly belongs, on people who plot mass murder. As Sigivald points out, when you speak this way you deny the people doing these things status as autonomous actors. You are treating the rest of the world - all non-Americans - as, in effect, automatons who only react mechanically to what we do or don't to.

Lets see. There were how many car bombs in Iraq before we invaded? IIRC, the number is 0. Do you really believe that all of those Iraqis and foreign terrorists would be blowing themselves up if we were not in Iraq? Yes, they choose whether or not to respond to our actions. They see an Islamic country invaded by the West and a certain proportion respond as effectively as they know how. I think we both consider them misguided and their tactics evil, but they don’t think they are misguided. They think they are lawful defenders of Islamic lands and we are the new crusaders.

The things that really annoy me about Iraq are not that I am a worshiper of stability at all costs, but that the Bush administration has so incompetently pursued its pro-democracy strategy in the Mid-East that it has probably set the progress towards democracy back by years. Every other dictator can point to Iraq and say, “See what you get from democracy, chaos, car bombs, dead people everywhere”. For that matter, Iraq has probably served to nurture the next generation of terrorists the way Afghanistan nurtured Osama. I would really like to be wrong about that. We will see over the next 10-20 years.

Did Bill Clinton "create the conditions" which allowed Timothy McVeigh to murder 150+ people in Oklahoma, therefore deserving some of the blame? Can you answer this question decently and be consistent?

Sure, Timothy McVeigh was a nut acting in the US. I don’t recall Bill Clinton invading Oklahoma, thus creating conditions of anarchy in the streets.

Let me try another example. Say someone (Chinese, Martians, whoever) invaded California. Americans, along with some Canadians, and Mexicans rallied to California to try to fight the invaders. Are you telling me that their decision to go to California to fight is completely independent of the invader’s decision to invade? After all, “nothing in the laws of physics somehow compels terrorists to”, and you can bet that the invaders would call them terrorists.

More specifically, the comparison is simply inapt; killing doctors at an insane asylum and deliberately releasing criminally-insane patients is an irredeemably evil act of which no decent people can approve. Whatever you think of the Iraq invasion, reasonable people disagreed about its merits and it was not an unalloyed evil deed with no positive benefits: Destroying an evil and megalomaniacal autocratic regime is simply not an unalloyed evil thing to do. (I can't believe I had to say that.) It is certainly not akin to murdering a doctor. But as y81 alludes, it is somewhat telling that doctor/mental patient is the analogy that sprung to your mind when thinking about dictators who lord over Muslims....

I agree that destroying an evil and megalomaniacal autocratic regime is not an act of unalloyed evil. However, how you destroy it matters a lot. For example, we could have destroyed Saddam’s regime by nuking Iraq into glass and killing everyone of its 26 million odd people. It would have destroyed Saddam and his regime. However, I think (and devoutly hope most agree) you would have had to look long and hard for the alloying bit that kept it from being a completely evil act.

My problem with Iraq is that we completely failed to plan for what happened after the regime was toppled. AFAIK the administration plan was that Chalabi would come in with a few hundred armed followers and run the country with us. What we did was something akin to executing radical surgery on a patient without having a plan or even a clue about post-operative care.

Question for everybody. Given the choice of statements to defend in a debate, which do you think you would have an easier time defending, A or B.

A) The US was justified in invading Iraq.

B) The US occupation of Iraq was well planned and executed.

Is there anyone who would choose B as easier to defend?

Posted by Donald Clarke | February 4, 2008 4:53 PM

The study is described so clearly that it seems elegant, simple and air tight. In reality, it's not much different than the Lancet studies. A random sample of households is nice in principle, but how do you get such a sample. Are they using a cluster approach? If so, then each cluster is a measurement unit, not each household, and you have to know the statistical frame, that is, the universe of possible clusters before and during the study period. That frame might change. Likewise, even if you have a true random sample of households, there is very likely to have been consolidation and redistribution of households. The households with the least resources and mobility are more likely to be interviewed since others may be temporarily out of the country, for instance. The largest households are more likely to stay put. These households are also more likely to be victims. Did the survey ascertain whether the demographic profile of households was the same as before the war? Was there any attempt, at least, to investigate the non-responders? Maybe families that have lost members are more likely to speak their piece. I hope, at least, that they addressed the "Main Street" bias.

There is also the issue of handling false data. Do they know what percentage of these households provide them with incorrect responses? None? That would be quite an achievement. What checks do they have to make sure that the survey workers were doing their jobs? Not necessary? Have there been efforts to address the question of political bias that surfaced in the Lancet studies?

The fact that they needed a re-do to get the rural vs. urban ratio correct makes me wonder if there are any other ratios that have been neglected. Did they at least account for the fact that there are fewer buildings in Iraq than there were before the war? Fewer buildings, families consolidating, households traveling, families hiding with other families, families moving out of the city to rural areas, flexible definition of households -- all these things could have a tremendous impact. I notice that the report says nothing about Confidence Intervals or statistical significance. Maybe the Lancet people felt that they were burned by those statistical niceties.

Posted by jj mollo | February 4, 2008 6:32 PM

The things that really annoy me about Iraq are not that I am a worshiper of stability at all costs, but that the Bush administration has so incompetently pursued its pro-democracy strategy in the Mid-East that it has probably set the progress towards democracy back by years.

Yes. The simultaneously saddest and funniest part is, the Onion was much more accurate in its prewar prediction than, say, Bill Kristol (whose predictions are also portrayed in the counterpoint section of that same link).

Posted by Derek Scruggs | February 4, 2008 11:36 PM

Donald,

It seems that I make a distinction between "creating the conditions" and "doing an act which contingently led to.." whereas you don't. "Creating the conditions" carries with it (for me) a connotation of inevitability, of near-justifiability, on the part of the people who are reacting to the "conditions". Example: a woman who gets raped because she was wearing a skirt. The rapist says "I saw her skirt and...". So I suppose you could say she "created the conditions" which led to her rape. But I don't think she bears responsibility for the rape and few people talk this way. Rape is not in any sense a normal, logical, or jusifiable response to seeing a woman in a skirt; a skirt is not somehow intrinsically a "condition which leads to rape" - or at least, shouldn't be thought of in these terms, because in a sense it *concedes the rapist's point*.

Likewise: bombing IRAQI CIVILIANS is not a normal, logical, or justifiable response to the fact that there are U.S. soldiers inside the boundary of Iraq. A big problem I have with the remainder of your post is your broad-brush characterization of the terrorists' motives - "to fight the invaders". How on earth is bombing a marketplace full of Iraqi civilians "fighting the invaders", and even if that's what the terrorists think they are doing, what on earth requires us to play along with their delusion? It seems like there is a failure to apprehend the true nature of this violence. It is simply too pat to look at each and every act of violence in Iraq and think (1) "insurgent!" and (2) "fighting the invaders". This gives them entirely too much credit and I refuse. Seriously, in your mind are all acts of violence in Iraq automatically "insurgency"/"fighting the invaders" simply by virtue of the fact that there are U.S. soldiers stationed in the same country?

This is what is wrong with your Chinese/Mexican analogy as well; to fix it would require something like:
-Chinese invade California
-Some Mexicans travel to California and start murdering CALIFORNIANS, and organizing/funding/extorting underground bands of poor and/or hothead Californians to do the same.

And of course you, enlightened foreign observer, look on and say merely:

"They're trying to get the Chinese out. This proves the Chinese should get out."

I have a big problem with this misdiagnosis; it is far too simplistic and betrays an inability or unwillingness to see the violence for what it is. (For one thing: even if the Chinese get out, will the Mexicans? Doesn't it matter?) Anyway, no one's reading this thread anymore so I'll leave it there....bye,

Posted by Sonic Charmer | February 5, 2008 7:30 AM

I like the Chinese/Mexican/California analogy. That is a better comparison than the "they're fighting the invaders!".

I don't see Americans fighting an invasion by blowing up malls and grocery stores full of civilians. I see Americans attacking the enemy military, either directly or indirectly.

Posted by Earnest Iconoclast | February 5, 2008 3:35 PM