2006 October 11 Wednesday
Johns Hopkins Estimates 650,000 Dead In Iraq

The official death toll reports from Iraq probably drastically underestimate the real number of deaths from the US invasion and its aftermath of continuing war.

As many as 654,965 more Iraqis may have died since hostilities began in Iraq in March 2003 than would have been expected under pre-war conditions, according to a survey conducted by researchers at the Johns Hopkins Bloomberg School of Public Health and Al Mustansiriya University in Baghdad. The deaths from all causes—violent and non-violent—are over and above the estimated 143,000 deaths per year that occurred from all causes prior to the March 2003 invasion.

The estimates were derived from a nationwide household survey of 1,849 households throughout Iraq conducted between May and July 2006. The results are consistent with the findings of an October 2004 study of Iraq mortality conducted by the Hopkins researchers. Also, the findings closely reflect the increased mortality trends reported by other organizations that utilized passive methods of counting mortality, such as counting bodies in morgues or deaths reported by the news media. The study is published in the October 14, 2006, edition of the peer-reviewed scientific journal, The Lancet.

“As we found with our previous survey, the majority of deaths in Iraq are due to violence—although we also saw a small increase in deaths from non-violent causes, such as heart disease, cancer and chronic illness. Gunshots were the primary cause of violent deaths. To put these numbers in context, deaths are occurring in Iraq now at a rate more than three times that from before the invasion of March 2003,” said Gilbert Burnham, MD, PhD, lead author of the study and co-director of the Bloomberg School’s Center for Refugee and Disaster Response. “Our total estimate is much higher than other mortality estimates because we used a population-based, active method for collecting mortality information rather than passive methods that depend on counting bodies or tabulated media reports of violent deaths. Though the numbers differ, the trend in increasing numbers of deaths closely follows that measured by the U.S. Defense Department and the Iraq Body Count group.”

They estimate over 91% were killed by violence and that coalition forces were responsible for about 31% until July 2006 when deaths from coalition forces declined to 26% of the total. That's probably a sign that the sectarian killings have increased.

The invasion more than doubled the death rate in Iraq.

According to the researchers, the overall rate of mortality in Iraq since March 2003 is 13.3 deaths per 1,000 persons per year compared to 5.5 deaths per 1,000 persons per year prior to March 2003. This amounts to about 2.5 percent of Iraqi’s population having died as a consequence of the war.

One of the justifications for invading Iraq was to stop Saddam Hussein from killing Iraqis. Oops.

Update: Here's what I've long wondered about the casualty rate reports from Iraq:

  • What percentage of those killed show up as visible dead bodies to even have a chance of being collected by police and morgues?
  • Has the government crumbled so far in some parts of the country that there are no morgue employees left to collect bodies? In some areas the police and elected officials have fled. So why not the morgue employees? Do they get paid? The Iraqi army has a had time getting paid. And so maybe the morgue employees in some areas have given up working.
  • How much political pressure is there to publically report lower death figures than are known within the Iraqi government? The huge underreporting of deaths in Baghdad in August 2006 (revised upward by a multiple a week or so later) suggests this happens at least some of the time.
  • How many bodies aren't accounted for because artillery or air strikes obliterate some unknown number of insurgents and civilians in buildings?
  • Do any car bombs obliterate some victims so that an accurate death count is not possible?
  • Are Shia deaths or Sunni deaths more underreported?

Update II: Steve Sailer has an extensive post on the plausibility of this report. Steve points to the text of the study that shows most of the Iraqis polled who reported family deaths were able to provide death certificates.

The study population at the beginning of the recall period (January 1, 2002) was calculated to be 11 956, and a total of 1474 births and 629 deaths were reported during the study period; age was reported for 610 of 629 deaths, sex reporting was complete. During the survey period there were 129 households (7%) that reported in-migration, and 152 households (8%) reported out-migration. Survey teams asked for death certificates in 545 (87%) reported deaths and these were present in 501 cases. The pattern of deaths in households without death certificates was no different from those with certificates.

The death certificates suggest that parts of the Iraqi government have the documentation that could be used to measure the death rates. In theory one could go to an Iraqi city government administration building, go into the archives of death certificates, and see how many death certificates there are for various time periods. Or do Iraqi bureaucracies issue death certificates that they do not keep copies of?

Update III: One problem with the latest study is that the assumption it uses for the pre-war mortality rate might be too low.

The first issue here: Iraq's pre-war mortality rate. The first Johns Hopkins study from 2004 pegged it at five per every 1,000 population, based on what those interviewed recalled. This one was 5.5/1,000.

But UN reports had suggested Iraq's crude death rate was higher than this in the 1980s and '90s. It was in at least the 6.8/1,000 range and rising, which would make the difference between normal deaths and what the researchers called "excess deaths" brought about by the war quite a bit smaller.


This one roughly 40 households in each of 50 sites and as a result the confidence continuum has narrowed considerably to between 426,369 and 795,663 — which is still quite a range.

If we take the low end of the confidence interval we are still left with a high increment in the death rate. Worse, even those who argue for a lower death rate admit the death rate rose substantially this year.

Update IV: Daniel Davies says the numbers do add up.

The results speak for themselves. There was a sample of 12,801 individuals in 1,849 households, in 47 geographical locations. That is a big sample, not a small one. The opinion polls from Mori and such which measure political support use a sample size of about 2,000 individuals, and they have a margin of error of +/- 3%. If Margaret Beckett looks at the Labour party's rating in the polls, she presumably considers this to be reasonably reliable, so she should not contribute to public ignorance by allowing her department to disparage "small samples extrapolated to the whole country". The Iraq Body Count website and the Iraqi government statistics are not better measures than the survey results, because one of the things we know about war zones is that casualties are under-reported, usually by a factor of more than five.

And the results were shocking. In the 18 months before the invasion, the sample reported 82 deaths, two of them from violence. In the 39 months since the invasion, the sample households had seen 547 deaths, 300 of them from violence. The death rate expressed as deaths per 1,000 per year had gone up from 5.5 to 13.3.

Talk of confidence intervals becomes frankly irrelevant at this point. If you want to pick a figure for the precise number of excess deaths, then (1.33% - 0.55%) x 26,000,000 x 3.25 = 659,000 is as good as any, multiplying out the difference between the death rates by the population of Iraq and the time since the invasion. But we're interested in the qualitative conclusion here.

How can a survey of such a large number of people come out drastically wrong? Davies is right. 12,000 people is a large number of people to survey.

Update V: What other method would be more accurate than a survey? I agree with Richard Garfield that under the circumstances a survey is the most accurate method available for measuring death rates.

"I loved when President Bush said 'their methodology has been pretty well discredited,' " says Richard Garfield, a public health professor at Columbia University who works closely with a number of the authors of the report. "That's exactly wrong. There is no discrediting of this methodology. I don't think there's anyone who's been involved in mortality research who thinks there's a better way to do it in unsecured areas. I have never heard of any argument in this field that says there's a better way to do it."

Do you trust lower official figures from a dysfunctional and corrupt Iraqi government? Recall that an employee of the Baghdad morgue reported in August that the real death toll the morgue saw was 3 times the death toll reported for Baghdad by the Iraqi government.

Update VI: The Iraqi woman who writes the Baghdad Burning blog says she sees so many deaths per family among the families she knows that she finds the Johns Hopkins results plausible.

For American politicians and military personnel, playing dumb and talking about numbers of bodies in morgues and official statistics, etc, seems to be the latest tactic. But as any Iraqi knows, not every death is being reported. As for getting reliable numbers from the Ministry of Health or any other official Iraqi institution, that's about as probable as getting a coherent, grammatically correct sentence from George Bush- especially after the ministry was banned from giving out correct mortality numbers. So far, the only Iraqis I know pretending this number is outrageous are either out-of-touch Iraqis abroad who supported the war, or Iraqis inside of the country who are directly benefiting from the occupation ($) and likely living in the Green Zone.

The chaos and lack of proper facilities is resulting in people being buried without a trip to the morgue or the hospital. During American military attacks on cities like Samarra and Fallujah, victims were buried in their gardens or in mass graves in football fields. Or has that been forgotten already?

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. Abductions, militias, sectarian violence, revenge killings, assassinations, car-bombs, suicide bombers, American military strikes, Iraqi military raids, death squads, extremists, armed robberies, executions, detentions, secret prisons, torture, mysterious weapons – with so many different ways to die, is the number so far fetched?

We know the Iraqi government will deceive about death rates. We also know the Iraqi government barely exists in some parts of Iraq. Surely the death rate is underreported. The question is only by how much?

Update VII: Richard Miniter interviewed Johns Hopkins Professor and study leader Richard Burnham on how his team collected the information for this study and why he thinks the results are valid.

Burnham: 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.

Death rates are a function of many things—not just health of the population. One of the most important factors in the death rate is the number of elderly in the population. Iraq has few, and a death rate of 5.5/1000/yr in our calculation (5.3 for the CIA), the USA is 8 and Sweden is 11. This is an indication of how important the population structures are in determining death rates. (You might Google ‘population pyramid’ and look at the census bureau site—fascinating stuff.)

The fact that his measured 5.5 rate is close to the CIA 5.3 rate for the pre-war period is an indication (though not conclusive) that at least for the pre-war period his sample was representative and his method of data collection was sound.

Burnham says most facilities (e.g. morgues) are not reporting their mortality information to the central government and the government is manipulating the data it does get.

PajamasMedia: You write that an active survey is more accurate than a “passive” system of counting media reports, morgue reports or other lists of the dead, which are often grossly incomplete in a war zone. This seems reasonable. To make sure people weren’t making things up, you teams received death certificates some 80% of the time. Also reasonable. So why are the active death figures an order of magnitude higher than the passive counts?

Burnham: The difference depends on the proportion of the passive-reporting facilities whose reports on death tolls reach some central tabulating body. Our information is that not many facilities are reporting, and what is being reported is often being manipulated.

I fully expect far less than complete reporting by the local government units. I also expect not all bodies to even make it to morgues or hospitals. I also expect provincial governates and the central government to cook the books. Plus, reporters can't even get out to most of the places where people are dying unless they are embedded with US troops. So reporters can't get the story of what is really happening with deaths.

Update VIII: A January 2008 report from the Iraqi government and World Health Organization estimates only a quarter the number of deaths that the Johns Hopkins study found.

A new survey estimates that 151,000 Iraqis died from violence in the three years following the U.S.-led invasion of the country. Roughly 9 out of 10 of those deaths were a consequence of U.S. military operations, insurgent attacks and sectarian warfare.

The survey, conducted by the Iraqi government and the World Health Organization, also found a 60 percent increase in nonviolent deaths -- from such causes as childhood infections and kidney failure -- during the period. The results, which will be published in the New England Journal of Medicine at the end of the month, are the latest of several widely divergent and controversial estimates of mortality attributed to the Iraq war.

The three-year toll of violent deaths calculated in the survey is one-quarter the size of that found in a smaller survey by Iraqi and Johns Hopkins University researchers published in the journal Lancet in 2006.

Here from the WHO press release on the January 2008 study:

The estimate is based on interviews conducted in 9345 households in nearly 1000 neighbourhoods and villages across Iraq. The researchers emphasize that despite the large size of the study, the uncertainty inherent in calculating such estimates led them to conclude that the number of Iraqis who died from violence during that period lies between 104 000 and 223 000.

"Assessment of the death toll in conflict situations is extremely difficult and household survey results have to be interpreted with caution," said study co-author Mohamed Ali, a WHO statistician who provided technical assistance for the survey. "However, in the absence of comprehensive death registration and hospital reporting, household surveys are the best we can do."

"Our survey estimate is three times higher than the death toll detected through careful screening of media reports by the Iraq Body Count project and about four times lower than a smaller-scale household survey conducted earlier in 2006," added Naeema Al Gasseer, the WHO Representative to Iraq.

The study found that violence became a leading cause of death for Iraqi adults after March 2003 and the main cause for men aged 15-59 years. It indicated that on average 128 Iraqis per day died of violent causes in the first year following the invasion and that the average daily violent death toll was 115 in the second year and 126 in the third year. More than half of the violent deaths occurred in Baghdad.

You can read the full text of this January 2008 report in the New England Journal of Medicine: Violence-Related Mortality in Iraq from 2002 to 2006.

Share |      By Randall Parker at 2006 October 11 11:00 PM  Mideast Iraq Costs

Your Image Here said at October 12, 2006 1:30 AM:

Randall, I understand your point in using that study. The problem is the source of those figures did a similar ''study'' in 2004 that was THROUGHLY discredited as poor statistics resulting in a grossly exaggerated figure.
Do I agree that Iraq is a disaster, a bloodbath and the ''goals'' the US is trying to accomplish there impossible? Absolutely. But citing such a dubious source to advance our arguments undercuts the points we are trying to make. Our opponents can readily sieze upon those dubious statistics as 'proof' that we are merely 'inflating the numbers' to advance a flawed argument.

Septeus7 said at October 12, 2006 4:53 AM:

I suggest you read this http://chronicle.com/free/2005/01/2005012701n.htm, http://timlambert.org/category/fumento/, and http://timlambert.org/category/lancetiraq/ .

Real Scientists have agreed that the Lancet study was not seriously flawed and only the innummerates like Moran the Moron and Mental Fumento say otherwise with no qualification or evidence.

Lawrence Auster said at October 12, 2006 6:48 AM:

This 600,000 figure is ludicrous. That's the number of dead in the American War between the States, with vast battlefields covered with the dead. There obviously has been nothing approaching that in Iraq, even with the ongoing almost daily terrorist killings. The idea that the Coalition forces have killed 200,000 is also ludicrous. Mr. Parker does not help advance intelligent criticism of Bush's disastrous Iraq policy by reporting "straight" such an absurd, and, I'm sure, politically motivated, finding.

Bob Badour said at October 12, 2006 8:24 AM:


Comparing combatants killed on the battlefield with total mortality is specious. I have no idea what motivates you to forward such a specious argument, and I wonder whether you project your motives onto others.

I find nothing ludicrous about a death rate of 13 per 1000 per year in a country during a time period that saw massive infantry battles, sectarian death squads, and massive loss of basic infrastructure such as utilities, hospitals etc. According to the study the total mortality rate in Iraq is only 62% higher than the total mortality rate in the USA. Assuming 'ludicrous' means inaccurate by at least a factor of 2, are you suggesting the actual mortality rate in Iraq is only 80% that of the USA?

Prior to the conflict the total mortality rate in the USA was 44% above that of Iraq due to Iraq's younger population in spite of Saddam's tyranny. Is it ludicrous to believe the death rate in Iraq rose from something less than the rate in the USA to something above the US rate?

As one might expect during a time of war, young males in Iraq are dying at greatly elevated rates.

Do you also forward the argument that the birth figures reported are similarly ludicrous at 2.3 times the number of deaths?

John S Bolton said at October 12, 2006 9:32 AM:

These numbers are exaggerated, but if they are believed by the third wotld and the moslem in paticular, they will do us a world of good.
If you kill thousands of Americans, we will kill hundreds of thousands of yours.
The base of comparison could have been Saddam's propaganda to make it look like he was doing a good job on medicine and public health, when he wasn't.
If it were that way, though, it still is not such a spectacular bubble-gum handout, to leave public health at the level which Saddam found actually acceptable.

Lawrence Auster said at October 12, 2006 9:44 AM:

Interesting that the first thing Bob Badour does is to question my motives for disbelieving the study. I wonder why he instantly assumed that I have bad motives, rather than that I am just confused?

I agree that if the figure is stated in terms of death rate statistics, as Mr. Badour has done, it doesn't sound so impossible. I also arrive at a figure of 13 per 1000 per year total deaths (see my calculations below).

However, these aren't just generic deaths but violent deaths. According to the study's estimates, 91 percent of the 655,000 additional deaths were violent. That's 596,050 violent deaths in 3 1/2 years. That is comparable to the 600,000 soldiers who died in the four years of the the American Civil War. Morever, a very high percentage of that 600,000 in the Civil War died of disease, not wounds, which would mean that a significantly higher number of Iraqis have been killed by violence since March 2003 than the number of American soldiers killed in battle in the Civil War.

That is the basis of my skepticism about this study.

Very few Iraqis died in the invasion of Iraq—most of the divisions, including the Republican Guard divisions, just dissolved and went home. Contrary to what Mr. Badour says, there have not been massive battles. There have been lots of skirmishes and endless terrorist attacks, many of them killing scores of people at a time, but there has been nothing to suggest that the carnage would have equalled or surpassed that of the 1861-1865 period in the the U.S. Even the big battle of Fallujah in November 2004 killed far fewer insurgents than expected because the enemy (as I predicted at the time) just left town to continue the fight elsewhere.

As for the death rate, since readers may wonder how Mr. Badour derived his figures, here are my own calculations, the result of which jives with his figure on the total death rate.

The study says that as many as 655,000 more Iraqis may have died since hostilities began in Iraq in March 2003 than would have been expected under pre-war conditions, which was an estimated 143,000 deaths per year.

If we break down 655,000 for the 3 1/2 years since the invasion, that's 187,142 additional deaths per year since the invasion, on top of the base level of 143,000 deaths per year, or a total of 330,142 deaths per year in a country with approximately 25 million people. 330,142/25,000,000 = .01321 or 13 out of one thousand.

John S Bolton said at October 12, 2006 9:51 AM:

The plundering and near-collapse of the Iraqi medical system would be expected to result in excess deaths, especially in terms of ordinarily non-lethal attacks becoming murders through decline in quality of emergency response; but not to the extent these politicized researchers say has occurred.

No-PC-wars said at October 12, 2006 9:52 AM:

"Comparing combatants killed on the battlefield with total mortality is specious."

Study says that 650K died in ADDITION to a number of deaths projected from pre-invasion

Study says 650K are dead in battles, low level civil wars, general lawlessness, drop in living conditions, etc.
Auster compares 650K with American Civil War deaths. I don't know if Civil War figure includes non-battlefield deaths due to the war itself.

If Civil War figure includes those deaths, figures can be compared, if it does not, they may or may not be compared depending on fraction of non-battlefield deaths.

In either case, what difference does it make?
If Iraq war advances OUR (USA) strategic interests, would you object to a million Iraqis dead?

Iraq war is fought for an irrelevant (how in hell Shia ala-Iran democracy can be useful for us? For existing example see Palestinian "democracy")
and may be unachievable objective. Will you support that war if only 100 Iraqis dead?

If war is needed, kill as many as needed for victory. If war is not needed, don't do it.

Iraqi war is a grandiose failure of American political system that put in power a country bumpkin president, gave
access to power to a group of deranged ideologues (neo-cons) and produced basically anti-american opposition with no solutions of their own.

Picking a wholly irrelevant number does not contribute to understanding.

Lawrence Auster said at October 12, 2006 11:36 AM:

If we accept No-PC War's premise, which I do accept, about a justifiable war versus an unjustifiable one, and that waging a war for jihadist democracy is not worth a single life, then the numbers matter a great deal. If 600,000 people lost their lives for a deranged neocon brainstorm, that's obviously a lot worse than if 100 people lost their lives for it.

However, proving or disproving a point about Bush's war is not why I objected to the study showing 600,000 dead. I objected to it simply because the figure seems extremely unlikely to me. Apparently such a simple reason for registering an opinion is beyond several of the commenters here.

John S Bolton said at October 12, 2006 1:23 PM:

There's also a question of the war aims being not really as stated, in which case the value sign of the numbers of increased deaths could be different.
We do not call WWI or WWII a defeat for the allies on the grounds that the rights of neutrals were no more respected afterwards than before.

Stephen said at October 12, 2006 2:16 PM:

John said: If you kill thousands of Americans, we will kill hundreds of thousands of yours.

But Iraqi's have only killed thousands of Americans during the invasion and subsequent occupation (well, 2,700 based on US military reports). Are you proclaiming some god given formula of 1 dead US soldier = 240 dead Iraqi's (military or civilian) in reprisal? If so, could I suggest that its a trifle insane??

Bob Badour said at October 12, 2006 4:14 PM:


What's even more interesting is I assumed nothing about your motives. I stated quite plainly that I do not know what motivates you, and I only mentioned motivation in the first place to immitate you. When one removes the specious pretense of empiricism from your original post, all that's left is an ad hominem argument suggesting the scientists biased the results for political motives. Unless one also wants to count the argument that the figure is ludicrous because it is absurd and presumably absurd because it is ludicrous.

When one considers that the population of Iraq is comparable to the population of the US during 1860's, that the total mortality in Iraq compares to the combatant casualties during the civil war merely confirms how much bloodier the civil war was. The closest figures I could find for mortality in the US during the civil war are census data from 1870, 1880 and 1890. The PDF file titled Vital and Social Statistics in the United States at the Eleventh Census 1890. Part III. suggests that total mortality in Iraq closely resembles total mortality in late 19th century America.

Comparing violent deaths with battlefield deaths is similarly specious as your earlier comparison. Doing so would discount all the suicide, lynchings etc. that I am sure followed the civil war especially in the south--not to mention tribal warfare with the Plains Indians, Jesse James, Wyatt Earp et al and all those other legendary gunfights in the American West during that time.

One wonders, though, whether a car accident, drowning or other death by misadventure counts as a violent death.

As for the death rate, since readers may wonder how Mr. Badour derived his figures, here are my own calculations, the result of which jives with his figure on the total death rate.

I simply read the death rate as reported by the scientists who conducted the study. I am sure they would take comfort to know you verified their ability to perform simple addition, division and multiplication. Do you still contend that their figures are "ludicrous" given the figures are not much different from post-civil war mortality rates? Or have you softened your opposition to them in any way? It seems you say they are not quite as ludicrous as you original thought, but not very clearly so I could be mistaken,

Bob Badour said at October 12, 2006 4:25 PM:


I wonder whether you have any empirical basis for your opposition to the figures from the study. After their 2004 paper, the general consensus among the scientific community and among the peer reviewers was their methods are sound and valid. The research team who conducted the surveys and who tabulated the results comprise a wide range of political views regarding the war. I find it unlikely that one or even two members of the team could usurp the process sufficiently to bias the result to an absurd extent.

Partisan hacks sqawked about the 2004 study and threw lots of sophist rhetoric against it, but then again, that's what partisan hacks do.

Stephen said at October 12, 2006 4:33 PM:

The figures appear to be the most reliable estimate available. If anyone can find statistically better estimates, please link to them...

As for whether the figures sound reasonable in a subjective personal bias sense, my word is as good as anyones, and I think they do sound doable (ie military killings + civil war killings + crime killings + degraded infrastructure deaths (hospital access, potable water etc)). But of course, preferring subjectivity over scientific method doesn't advance this particular debate either way.

Bob Badour said at October 12, 2006 5:04 PM:


I agree about advancing the debate; however, to have a meaningful debate one has to get past the visceral and knee-jerk reactions.

The initial reactions of some appear to be to dismiss the study entirely and even to chastise Randall for pointing to it. I think that sort of dismissal is a mistake, and I strongly believe that self-censorship a la political correctness doesn't advance anything intelligent -- criticism or anything else.

Do the scientists say the exact number must be 650 some odd thousand more deaths? I strongly doubt it. Might 500 thousand lie within the range of highly plausible too? I bet it would. Whether the number is 400,000 or 650,000 is pretty much irrelevant to the observation that whatever the exact number, Iraqis have no reason to think of the US as liberators who improved their lives. The average Iraqi on the street has every reason to think the US turned the country into a primitive, dangerous shithole where constant fear of imminent death is more rational now than under Saddam.

TD Larkin said at October 12, 2006 5:06 PM:

Lawerence said: I objected to it simply because the figure seems extremely unlikely to me.

So you just have this "feeling" that the number is wrong? I wonder what facts that feeling is based on?

Consider this: the only independent reporting we have on Iraq comes from members of the media who (at great risk to their own personal safety) remain largely clustered in the protected areas of Baghdad. When was the last time you saw a media report from Ramadi, or Fallujah, or Basra? When was the last time you saw a realistic body count from the morgues in those cities?

Statistical sampling is the only way to accurately estimate the extent of the civilian casualities in Iraq. Media reports don't come anywhere near to reflecting reality. Surely, if 50 bodies are discovered in a day there are probably at least 50 more that aren't. Wouldn't it make sense that the people committing these atrocities would be dumping the bodies in mass graves and burying them when they had the chance? Of course it would.

The counts from the morgues and the Iraqi government also aren't reliable because many Iraqi families are simply going to bury their dead without taking them to a morgue or reporting them to the government. Do you really expect a Sunni family to report the death of a family member to the Shiite-dominated police force that has been infiltrated by death squads? Come on.

And you've missed the point of the Lancet study. The real point is that the death rate is significantly greater than it was before the war. It's now 13.3 compared to 5.5. A sample of 12,801 individuals in 1,849 households, and in 47 geographical locations is plenty large enough (unless you want to invoke the "OJ" defense about statistics).

People need to wake up. Iraq is becoming the new Darfur.

(Vicente) Fox Hound said at October 12, 2006 5:48 PM:

[quote]Consider this: the only independent reporting we have on Iraq comes from members of the media who (at great risk to their own personal safety) remain largely clustered in the protected areas of Baghdad. When was the last time you saw a media report from Ramadi, or Fallujah, or Basra? When was the last time you saw a realistic body count from the morgues in those cities?[/quote]

I'm willing to believe that in and around Baghdad is pretty much where most of the violence in Iraq takes place. If this is "sectarian violence" (civil war), then the violence would be in areas where the population is mixed Sunni and Shi'ite, like Baghdad, Baquba Samarra, and Mahmoudiya. I would not find it hard to believe that the media being in Baghdad makes the city a better setting for violence. I'm not saying Ramadi and Fallujah are safe places. They're not at all. Any westerner there would be kidnapped immediately and killed. They are no-go areas. But there's less large scale violence in those cities because there are no opposing sects.

Bob Badour said at October 12, 2006 7:27 PM:


In the 2004 survey, the mortality rate in Fallujah was so much higher than the rest of the country the sample was excluded as an outlier. I think you are focussing on a red herring while missing the true import of the study.

The overall mortality rate is significantly higher under US occupation than it was under Saddam. Combine this with the fact that the electricity seldom works etc. and I doubt many Iraqis think the US helped them by toppling Saddam.

Lawrence Auster said at October 12, 2006 7:41 PM:

To TD Larkin, I didn't say I had a "feeling," I said it seemed extremely unlikely, based on rule of thumb and common sense. And because the figures of war dead in the Civil War are so similar to these figures, that gave an apt basis for comparison.

Further, the media have not been covering up violent deaths in Iraq. For heaven's sake; every day's news for the last three and a half years has featured the latest atrocity in Iraq. And it seems to me that there simply has not been the scale of deaths to add up to anything like 600,000 dead from violence. There has NOT been a Civil-War scale major war in Iraq for the last three years, there has been an ongoing insurgency and counterinsurgency with lots of terror attacks, a large number of mass terror attacks with 20, 30, or 40 people being killed at a time, and lots of murders and assassinations, but nothing that could add up to 600,000 killed.

To Bob Badour, I will just say that a person who on the basis of no evidence suggests that another person is expressing an opinion not because he honestly believes it, but because he has some hidden partisan motive, and then turns around and denies that he was charging a hidden motive, is contemptible.

Randall Parker said at October 12, 2006 7:55 PM:

Fox Hound,

Some of the Shiite areas have lots of violence due to fights between rival tribes/gangs/militias. Look at Basra as the classic example. I've written posts about Basra violence and lawlessness. See Going Down In Basra Southern Iraq and Basra And Ramadi Big Iraq Troublespots. The rival civic improvement associations are trying very hard to improve the neighborhoods for their benefit by killing anyone who stands in the way of their using the government to improve the cash flows of their own groups.

So, no, the areas with both Shiites and Sunnis are not the only areas with considerable amounts of violence.

Also, Basra did have a lot of Sunnis even though it was deep in the Shia heartland. The second of those two links is from June 2006 and quotes a source that says Basra went from 40% Sunni to 15% Sunni since the US invasion. Well, what happened to that 25%? Some fled. How many died? Similar things happened to the Shias and Kurds of Fallujah. How many died before fleeing.

Randall Parker said at October 12, 2006 8:10 PM:


I do not know how accurate this survey is. But I really believe that the Iraqi government's figures for death tolls understate what is going on. I read a great deal on Iraq and come across reports on intentionally inaccurate death toll reports from the Iraqi government. The August 2006 Baghdad death toll flap (where the official death toll understated the eventual revised figure by a factor of 3) and the US military's weird classification of deaths during that episode made me think we can't trust the US military's death figures either. If they are willing to deceive then how much have they been deceiving us that we didn't catch?

Can a lot of people be getting killed in Iraq? Well, let us see. They have a barely functional police force. So no control on normal killings by garden variety killers. Plus, their own government supports groups that run death squads against Sunnis. The Sunnis have their death squads too. Even in Shia areas the Shia groups battle it out for supremacy with violence at an intensity that makes 1920s Chicago gangsters quite tame by comparison.

I've used the term Hobbesian to describe Iraq. One thing different about Iraq as compared to a Hobbesian state of nature is modern weapons. Machine guns can spew out hundreds of rounds per minute. Primitive Hobbesian humans couldn't kill at the rate that modern humans can manage.

I am aware of the previous Johns Hopkins study that had a much larger confidence interval. They improved their sampling and got down their confidence interval. The figures they report seem hard to believe because the numbers seem shocking. But I'm not ready to dismiss them just on that basis. Iraq is shocking.

(Vicente) Fox Hound said at October 13, 2006 12:37 AM:

I'm specifically referring to Iraq 2006. I'm not saying there's no violence elsewhere in the country. Iraq is a shithole. The violence is worse now than ever. There's violence almost everywhere. But Baghdad is center stage. It houses 20% of the entire country's population, and has large Sunni and Shi'ite populations. It is in Baghdad where a Sunni car bomb or two kills 25, 30, even 50 people EVERY DAY, and Shi'ite death squads respond by torturing and killing 50 of Sunnis and dumping them into the Tigris. Iraqis in Ramadi spend ther time attacking Americans and those helping the Marines. 400 assassinations took place in Basra over a two month span. And what, at least 250 people are killed in Baghdad each week?

John Thomas said at October 13, 2006 5:24 AM:

What I find missing from the debate is that rightly or wrongly, the USA has invaded Iraq and now governs Iraq.
Therefore, the USA has a moral responsibility towards the Iraqi people, who are all under American suzerainty, to maintain the peace, and to defend the life, limb and property of the Iraqi people is far as is humanly possible - anything else is dishonorable cowardice and spineless bare-faced treachery of the highest order.
To put it bluntly, the Iraqi people are America's charge, in the same way a legal parental guardian has the charge of children.Any attempt to side-step or weasel away responsibility is gross moral turpitude, in the same way as an act of perfidy on a sworn treaty is.
In failing to disarm and pacify Iraq, American has failed the Iraqi people.

Bob Badour said at October 13, 2006 5:59 AM:

John Thomas,

I am on record predating the Iraq war, when I supported the invasion, as knowing that war by half measures is evil. It is always evil. By that measure, Bush and Rummie are evil men who squander national treasure and much more importantly young American lives.

If Bush is right, he will burn in hell for eternity. He certainly deserves to.

We lost the war the day after the statue came down when anarchy broke out.

Bob Badour said at October 13, 2006 6:57 AM:


Clearly, I have offended you, and for that I apologize. I am sorry.

I ask: Do you think your charge of hidden political motive against the scientists is any less contemptible? Do you think attacking their motives did anything to foster intelligent criticism, intelligent debate or anything else intelligent?

Apparently, you seem to believe you have some remarkable insight into the motives and intentions of other men. I do not believe any man has any such insight, which is why I plainly stated I do not know what motivates you. I expressed curiosity about one hypothesis regarding your motives; however, I admit the hypothesis is not falsifiable. With regard to your insight into my motives and intentions, I can say with absolute certainty you have none at all.

When I mentioned motives, I merely aped you, and unlike you, I did so without charging you with anything and without expressing any sort of certainty. It was my intent to expose your sophistry as sophistry, and I had no illusions about you finding that pleasant. While it was not my intent to hurt your feelings or to evoke an emotional response, I admit your response was perfectly predictable collateral damage.

Randall has spoken highly of you. You can do better than sophist rhetoric and demands for political correctness. So do better. Hopefully, your comment above was just a momentary lapse.

Mark said at October 13, 2006 9:39 AM:

You write here that 12,000 individuals is a big sample. But individuals are not the unit of analysis, households are, i.e. if 6 individuals in one household report that a member of the household died, that is not six bodies but one. 2000 households is still a big sample, of course. But given that it's a war zone and both respondents and interviewers have hundreds of motives and means to non-randomly select, and lie, among many other methodological reservations, I am amazed that anyone puts any stock at all in the findings. Which are just not believable on the face of it. I mean, I understand why slavering communist westophobes like Daniel Davies would put stock in it, but not why normally sceptical Parapundit does. I am in agreement with the general consensus of the host and commentators here that, whatever the value of the original intervention to remove Saddam , it is time to clear out of Iraq. Doesn't mean we have to take enemy propaganda at face value. When the Lancet reports in October 2008 that everyone in Iraq is now dead, will you be just as keen to believe them?

crush41 said at October 13, 2006 1:01 PM:

Iraq is a fecund, youthful 'nation', so a death rate roughly 65% that of the US pre-invasion isn't implausible.

What bugs me is that an estimate of 50,000 non-violent deaths per year comes to a death rate of less than 1 per 1000 (population 27,000,000 million) annually. I find that estimate stunningly low.


As one who enjoys your commentary, I'm not sure why you're going after Lawrence Auster, who has been a thoughtful critic of the Iraq debacle. The problem with surveys like these is that the tendentiousness or at least questionability of their accuracy/methods serves to discredit those in opposition to the 'stay the course' mantra. I agree with Randall's sentiment just as I did with his post regarding homoegeneity and happiness, but want to advise caution in running with things that are apocryphal.


Even if said estimates are true, Iraq is still birthing three people for each one that dies! Is that enough to scare the Islamic world out of ticking off the US?

William Wixted said at October 13, 2006 4:43 PM:

John Thomas said: "Therefore, the USA has a moral responsibility towards the Iraqi people, who are all under American suzerainty, to maintain the peace, and to defend the life, limb and property of the Iraqi people is far as is humanly possible - anything else is dishonorabe cowardice and spineless bare-faced treachery of the highest order."

By this logic we could still be fighting in Vietnam this very day. After 30 years of involvement in Vietnam and 250,000 dead American soldiers would you still be saying the same thing?

What's your limit when it comes to Iraq? We've been pounding the place for 15 years now (going back to the first Gulf War), we've lost 3,000 and squandered hundreds of billions of dollars. We've dropped the equivalent of six Hiroshima-sized nuclear bombs on the country in conventional weapons and we still can't control the damn place.

Iraq is a no-win situation. What we need is an exit strategy and leaders like Richard Nixon with the courage to carry it out. Iraq is slowly draining the life-blood out of this country. We've got to stop the bleeding somehow.

TD Larkin

Randall Parker said at October 13, 2006 5:32 PM:

The youthfulness of Iraq is one of the factors that makes a high death rate from civil war and insurgency possible. Teenagers and guys in their 20s are a lot more willing and able to kill and risk their lives than older folks are.


Is the survey approach really questionable? If so, why? Some possibilities:

1) The interviewers systematically lied.

2) The interviewees lied or exaggerated or had bad memories.

3) The research paper writers fabricated the aggregated results when the collected up the individual interview records.

4) The researchers chose unrepresentative locations or unrepresentative households. Intentionally or unintentionally?

Well, if the survey is off which of these factors is the major cause?

Lawrence Auster said at October 13, 2006 6:44 PM:

I suppose Mr. Badour has never heard of academic and media organizations in this country skewing data to make the Bush administration's policies including its Iraq policy look bad. My thought process was very simple. The figure of 600,000 violent deaths seemed incredible on the face of it, and it makes the Iraq disaster seem like an even greater disaster than it is. It is a perfectly reasonable inference that the study, like so many other reports, is skewed to make things in Iraq look even worse than they are. I don't know that that's true; but it certainly is a reasonable possibility.

Somehow in Mr. Badour's mind, that statement of mine, which I made about a study group in the news, not about a fellow participant in this discussion, gave him the right to accuse me, a fellowo participant in this discussion, of having some partisan motive for making it. Then when I called him on that improper statement, he dishonestly denied that he had suggested that I had a ulterior motive, then he turned around again and went through the motions of apologizing for having offended me, then he turned around yet again and justified what he had said.

By the way, what motive could I have? I have been a consistent and harsh critic of the Bush Iraq policy for the last three years. What ulterior motive would I have for attacking a finding that makes Bush's policy seem even worse? The only likely motive is that the finding in fact seemed incredible to me.

What's happened here is simple. Mr. Badour gratuitously questioned my motives; I called him on it, and ever since then, unwilling to retract his initial gratuitous charge, he's been thrashing dishonestly all over the place.

Bob Badour said at October 13, 2006 7:50 PM:
I suppose Mr. Badour has never heard of academic and media organizations in this country skewing data to make the Bush administration's policies including its Iraq policy look bad.

Making a charge of academic impropriety is extremely serious. I would find it contemptible for anyone to make such a charge without a single whit of evidence. Do you have any evidence whatsoever?

We have already established that a 13 per 1000 per year total mortality rate is entirely believable. The US had a similar mortality rate not much over 100 years ago when it similarly lacked effective electricity, water purification, sewage disposal and in the aftermath of a war.

As it happens, plenty of kneejerk idiots made identical baseless accusations after the 2004 survey. As a result, the scientists increased their sample size to narrow the confidence range--and in case anyone hasn't figured it out, they did so at their own personal peril. All evidence suggests these scientists conducted as valid a measurement as is possible under the circumstances. Nobody at any point in time has provided a single shred of evidence that the methodology was flawed or that anybody ever manipulated the measurements. Both studies were peer reviewed prior to publication and the 2004 survey was further examined very carefully by scientific peers in response to the reactions of partisan hacks' baseless accusations. Those who have the greatest knowledge about measuring mortality in areas of unrest concluded that the 2004 study was entirely valid and represented the best scientific method available executed to the highest levels of ethical professionalism. I see no reason whatsoever to suspect the 2006 survey was in any way inferior to the 2004 survey.

Shame on you. While Randall has spoken highly of you in the past, I conclude you are an idiot who totally lacks any semblence of intellectual honesty. I am no longer curious about the hypothesis regarding you projecting your motives and methods onto others. You have confirmed the hypothesis beyond any reasonable doubt. You have contributed nothing to this discussion but fallacious reasoning, the basest of sophistry and demands for others to toe some politically correct line. You don't like to believe the Iraq war increased total mortality in Iraq by somewhere between 400,000 and 800,000 deaths. Tough. Too bad. The 95% confidence levels strongly suggest that's exactly what happened regardless of your mindless incredulity.

Anyone can read what I wrote on its face and confirm what I said for themselves. Apparently, you cling to your lunatic belief in some magical insight into the motives and intentions of others. That makes you as nutty as Bush.


John S Bolton said at October 14, 2006 1:01 AM:

If only ad hominem can be used in support of this study's high estimates, that is an indication of the unreason which underlies it.
Leftist scholars who make common cause with the moslem are known to have betrayed every principle they ever pretended to stand for.
To take at face value, the assertions of moslems about the injuries they say they have suffered at the hands of the infidel, when they are known to regard it is as perfectly honorable to lie when they are presenting such grievances, indicates an approach which is actually scandalous as research to be published in a scientific journal of renown.
Next perhaps, Nature will publish articles describing as valid data, Palestinian claims to have been subject to large-scale atrocities at the hands of the Jews.
Taking moslem witnesses as reporting in good faith, would soon contaminate and destroy cultural advancement.
We're seeing the beginnings of this process, and it has to be subject to elimination of the assumption of good faith.
Thomas Jefferson said:' I'd rather believe two yankee professors would lie, than that stones fall from the sky.'
That was in the day when scholars had much to fear if they were caught making misleading interpretations of doubtful reports; today they have tenure guaranteed by a lie-loving government.

Stephen said at October 14, 2006 1:14 AM:

Changing the topic for a moment, I wonder what the effect will be on Iraqi society over the next generation? Since 1980, Iraq death tolls would be:

500,000+ Iran war
100,000~ Gulf War v1
400,000~ Gulf War v2 (just to pick a number)

1,000,000 Iraqi's dead in 25 years, the vast majority being males of military age. That must have some real effect on society. On the one hand it will reinforce the once fading practice of surviving males taking multiple wives etc. On the other hand, in the west after WW1 there was a similar man shortange and that caused the crucial suffragete movement.

Interesting to see which way Iraq goes...

ziel said at October 14, 2006 5:49 AM:

Randall, I think there's a pretty good chance that the sampling was unrepresentative, since it wasn't really random. They didn't actually sample 12,000 - they didn't even sample 1900 households - they sampled 47 locations (because it's just too dangerous there, ironically, to do it right). I don't know how they did their calculations (why all the data and calculations cannot be provided on-line is beyond me) but I'm not sure that poor sampling of neighborhoods would even show up in the variances if the calculations were done as if each household (I assume not person) were randomly selected. There's no excuse for the details not being provided. I haven't asked, though, I must admit.

Lawrence Auster said at October 14, 2006 8:07 AM:

Well, Mr. Badour has certainly sized me up. I am "an idiot who totally lacks any semblence of intellectual honesty." I "project my motives and methods onto others." (Hmm, what _are_ these motives and methods of mine that I project onto others?) I engage in the "basest of sophistry," and all I've contributed to this discussion is the "demand for others to toe some politically correct line." (But what _is_ that politically correct line? All I've done is express my opinion about the figures and I've strongly objected to Mr. Badour saying, without any logical basis for it whatsoever, that I had some ulterior motive for expressing my opinion.) Also, my doubts of the figures show "mindless incredulity," and I have a "lunatic belief" in some "magical insight into the motives and intentions of others." Also, I am "as nutty as Bush," and "pathetic."

Yes, certainly Mr. Badour has made a devastating case against me. He certainly knows how to argue in an objective and persuasive way.

And all this because (1) I expressed my incredulity at the idea that 600,000 people have been killed by violence in Iraq over the last 3 1/2 years, and (2) I said that the finding may have been motivated by a desire to make Bush look bad, just as innumerable other wild statements about the Iraq war over the last few years have been motivated by the desire to make Bush look bad. The only thing that I can see objectionable in what I have said here is my initial passing comment that I was "sure" it was a politically motivated finding. That was overstated. Nevertheless, I do believe that it's a reasonable inference that a desire to make the Iraq situation seem as bad as possible could be a factor in these fantastically high estimates of violent death.

I will close by returning to my original, commonsensical point: I have not seen any pictures of Shiloh-type battlefields in Iraq, with acre after acre carpeted with dead bodies. And that is the way you get at 600,000 dead. I could well be wrong. But even Mr. Parker said that the figures "seem hard to believe."

Bob Badour said at October 14, 2006 9:25 AM:
If only ad hominem can be used in support of this study's high estimates, that is an indication of the unreason which underlies it.


Your argument would have merit if its premises were true. However, the support for the survey's results comprise far more than ad hominem. See for example the reasoned support for the 2004 study already provided by septeus7 above. The survey itself as published by The Lancet stands on its own. (PDF warning)

On the other hand, if only ad hominem and other fallacies can be used to discount support of the survey's measured result, that is an indication of the unreason which underlies those discounting it. Do you agree?

What's more, I don't support the survey or address the argument against the survey by noting Lawrence is an idiot. I conclude he is an idiot because his argument so totally lacks merit, because he so willfully ignores observable facts and because he expects others to accept as fact the fantasies of his own imagination.


You seem to suggest that somebody must be hiding things because the data are not available online. Much of the data are online both in the PDF linked above and at the Lancet's fulltext version of the article (requires free registration.) What's more the contact information for the researchers is available in the survey, and I am sure one can request their raw data from them for further analysis. I don't know whether the researchers have already published their data at the Johns Hopkins Bloomberg School of Public Health because a key file server seems to be down at the moment. Perhaps, we can find out when the server comes back online.

I respectfully suggest the broad ranges given in the report already reflect potential error due to reasonable adjustments to the sampling method for personal safety. The scientists themselves certainly do nothing to conceal or to discount those adjustments and their potential effect on the results.

Key things to note: Every measure of mortality in Iraq--not just this survey--points to the "Liberation" of "Iraqi Freedom" making Iraq a much more dangerous place and to the violence increasing over time. The trend is clear: Things are getting much worse in Iraq. The longer the US stays, the worse it will become.

perroazul del norte said at October 14, 2006 9:39 AM:

Comparing violent deaths with battlefield deaths is similarly specious as your earlier comparison. Doing so would discount all the suicide, lynchings etc. that I am sure followed the civil war especially in the south--not to mention tribal warfare with the Plains Indians, Jesse James, Wyatt Earp et al and all those other legendary gunfights in the American West during that time.- Bob Badour
I have no point to make re the larger issue, but this is a truly strange paragraph.No one with even a passing familiarity with US history would think that lynchings, the Indian Wars and Old West gunfights combined added up to anything more than a very trivial percentage of the deaths related to hostilities that occurred during the Civil War. And since the US was a deeply Christian country in 1865, I don't think think there were an extraordinary number of suicides.

Lawrence Auster said at October 14, 2006 11:20 AM:

Folks, here I go again with my idiocy and my total lack of intellectual honesty.

In an article on the Iran-Iraq war at Global Security.org (see link below), I find this:

"Iran acknowledged that nearly 300,000 people died in the war; estimates of the Iraqi dead range from 160,000 to 240,000."

Now the Iran-Iraq war lasted for nine years, involving many major battles with horrible loss of life. Yet according to this article Iran suffered 300,000 deaths, and Iraq suffered between 160,000 and 240,000 deaths.

That's a major, conventional war, lasting nine years, with many large infantry battles involving mass casualties.

Now does anyone believe it is plausible that Iraq in the last 3 1/2 years has lost TWICE what Iran lost in the nine-year-long Iran-Iraq war and TWO AND A HALF times what Iraq itself lost? Is the figure of 600,000 killed by violence remotely believable, short of much more convincing evidence than this study offers? I say it is not—except for people who are not just opposed to Bush's policy but who are willing to believe any story making Bush's policy look bad, no matter how outlandish. And I say further that it is a _reasonable_ inference (not a certainty, but a reasonable, rebuttable inference) that the people who did this study belong to that group.


Randall Parker said at October 14, 2006 12:43 PM:


I'm going to ignore the attacks on each others' characters. I think it reflects poorly on those who make them. Worse, it distracts from an argument centered on the facts on the ground in Iraq and what we can know or infer about those facts.


On the Iraq-Iran war as compared to the current conflict:

1) The Iraq-Iran war was a classical "front" war. The civilians away from the front lines were not threatened or killed.

2) The Iraq-Iran war was a lame front war. The Iraqis, for example, liked to shoot off lots of artillery and didn't want to go running across no-mans land. They kept their heads down. They weren't real aggressive. The mullahs in Iran started recruiting kids to serve as human wave attackers. That killed more. But apparently they didn't recruit a lot of kids since Iran only lost a few hundred thousand.

3) The Iraqi approach was typified by Iraqi air force pilots who dropped their bombs from high and ineffective altitudes because they didn't want to risk getting hit by triple-A fire.

4) If Iran and Iraq only lost a half million over 9 years then that is not major. We lost a half million in the US Civil War in about 3 or 4 years (more American losses than all other wars Americans have fought in btw - we kill each other better than anyone else can). The US population was only 20 million at that time.

5) In Iraq now it is a lot easier to kill than it was in the Iran-Iraq war because so many more people can do the killing. In just about any town in Iraq young guys can get away with killing rival tribes, rival religious sects, and so on.

6) The range of places in Iraq where sectarian killing can be done is a lot larger than is generally thought. As I mention above, Basra was 40% Sunni even though it is near the Iran border. So far from Baghdad in the heart of Shia land the Shias could kill Sunnis. Now most of the Sunnis have left Basra or been killed. What drove out a quarter of Basra's population? That's a huge displacement.

650,000 seems like too a large figure to me. But I'm not absolutely certan that the figure is too high.


As Steve Sailer has pointed out (click thru on my link to him in the post above) one can just as easily see reasons why an unrepresentative sample would be biased toward underestimating the deaths as overestimating them. Most obviously, which neighborhoods are most likely to miss getting sampled? The most dangerous ones. Who wants to go into them?

Also, the sampling is per family or household. That requires at least one member of a household to still be alive to answer questions. But what happens when an entire household gets wiped out? No one is left to report on the deaths in their household.

We should expect a lot of lying about cause of death. No, Akbar didn't get killed because he was trying to plant a bomb and it went off on him accidentally. No, Ahmed wasn't shooting mortars at the local American base. So then does the family misreport the cause of the death or not mention it at all?

Randall Parker said at October 14, 2006 1:03 PM:


Calling someone an idiot does NOT help the quality of the conversation.

Calling someone an idiot satisfies a need on your part at the expense of the quality of the conversation.

Calling someone an idiot does not convince any readers. It actually causes people to discount what you say.

John Bolton,

I'm not making ad hominem attacks on those who doubt the accuracy of the study results. Try not to judge a study's validity by the insults of anyone who either sees it as valid or sees it as invalid. There are more serious minds on both sides.

I'd also add that there are people like myself who are willing to entertain the possibility that the death toll in Iraq really is that high but who are not convinced due to the limits on the study. But note one thing: The biggest cause of limits on this study is the fact that Iraq is such a violent place that conducting a study like this in Iraq is extremely difficult. The level of violence is high enough to cause a quarter of the population of Iraq's second largest city to flee that city. The police chief of that city says he doesn't trust his own police.

Bob Badour said at October 14, 2006 1:33 PM:
Hmm, what _are_ these motives and methods of mine that I project onto others?

I answered that question ages ago: "I have no idea what motivates you to forward such a specious argument." The specious argument in question being the specious comparison of total mortality in Iraq over a 3 year period with combatant deaths during the American civil war. One could just as legitimately pretend to discount the figure by noting it is much higher than the number of casualties of the Hiroshima bomb and by noting a lack of news coverage of cities in Iraq flattened by fission devices. Specious comparisons are, well, specious.

Just as I am the authoritative source regarding my motives and intents, you are the authoritative source regarding your motives and intents. I have stated directly what my motive and intent was when I mentioned your motives: to mimick you and to highlight your sophistry. Did you mention motives? Did I turn around and do the same in my reply? Is it at least possible I aped you? You accuse me of dishonesty because I stick to that explanation of my motive and intent; however, intellectual honesty forces me to stick to that explanation because I was there at the time and I directly observed my motive and intent. You further object to factual descriptions of what I wrote that anyone can verify for himself.

Have you stated your motive and intent? Not directly. However, you did say: "Mr. Parker does not help advance intelligent criticism of Bush's disastrous Iraq policy by reporting [this] finding." Am I wrong to infer you expressed concern regarding the effect on a political argument you advance? Is Iraq policy a political matter? Yes. Is criticism of Bush's Iraq policy political speech? Yes. Are you on record as advancing the opinion that Bush's Iraq policy is a disaster? Yes. What _is_ your motive for continuing to make a specious comparison of total mortality in Iraq with soldier mortality during the civil war?

But what _is_ that politically correct line?

To censor valid measurements some might find shocking or surprising regardless of their likely accuracy. Duh.

Mr. Badour saying, without any logical basis for it whatsoever, that I had some ulterior motive for expressing my opinion.

Why do you put words in my mouth instead of replying to what I actually say? Beating up straw men is not particularly convincing. As near as I can tell, your own description of your motive was political and not ulterior at all. To be ulterior you would have had to conceal it and not reveal it like you did. As I said earlier, in my opinion, you have confirmed the hypothesis that you project your own failings onto others--not just motives but other failings too like your lack of intellectual honesty. The only ulterior motives were the political motives you projected onto the scientists who risked their lives to make the measurement, which you called "an absurd, and, I'm sure, politically motivated, finding." You even went so far as to express certainty about those ulterior motives.

my doubts of the figures show "mindless incredulity"

Actually, the argument that the figures are ludicrous because they are absurd and absurd because they are ludicrous shows "mindless incredulity". You are now advancing the similarly compelling argument that the figures are fantastical because they are ludicrous and absurd, which only tells me you stick to mindless incredulity even in the face of overwhelming empirical evidence to the contrary. Or do you likewise suggest that late 19th century total mortality in peacetime America was fantastical because it was ludicrous and absurd? After all, the figures are very similar.

If 13 per 1000 per year is fantastical in Iraq, presumably 15 per 1000 per year in 1880's America was fanstastical too. N'est pas? Do you similarly accuse the US Census Bureau of exaggerating total mortality in 1880's America for political ends?

Yes, certainly Mr. Badour has made a devastating case against me. He certainly knows how to argue in an objective and persuasive way.

Thank you. You are too kind.

The only thing that I can see objectionable in what I have said here is my initial passing comment that I was "sure" it was a politically motivated finding. That was overstated.

Apparently, you have ignored most of the objections. I found all of the following objectionable:

  1. your ad hominem agument (the scientists had political motives)
  2. your petitio principii argument (the numbers are ludicrous because they are absurd)
  3. your shooting the messenger (your attack on the scientists who brought you the news)
  4. your misleading vividness (battlefields strewn with corpses)
  5. your package deal fallacy (some people have made up figures to discredit Bush so the scientists who performed this survey must have made up their figures)
  6. your appeal to consequences (most people's initial shock will impede receptiveness so the figures must not be true)
  7. your pretense of empiricism (comparing deaths of soldiers in the civil war to total mortality in Iraq)
  8. your straw man argument (overstating my position)
  9. your association fallacy (some people make up numbers to criticize Bush so every number critical to Bush is made up)
  10. your argument from incredulity (you don't want to believe the figures so they must be false -- even absurd)
  11. your wishful thinking (ignoring empirical results that indicate things are worse than you want them to be)
  12. your appeal to ridicule (your assertion that the empirical results are absurd and contrary to common sense)
  13. your agumentum ad populum (many people will resist belief in that level of mortality so intelligent criticisms will censor those empirical results)
  14. your argument from repetition (your repeated unsupported accusations directed at me)
  15. your proof by assertion (you merely assert the empirical results are ludicrous)
  16. your style over substance argument (I insult you because you ignore the all empirical evidence and substantive criticism -- not because my arguments lack empiricism and substance)

Finally, I found your appeal to motive objectionable (the scientists must have had political motives because you disagree with them.) Apparently, you agree that appeals to motive are objectionable because when I aped your argument you objected quite vigorously.

While I don't find your sophistry at all convincing, I am duly impressed by your remarkable ability to cram so many fallacies into so few words. Bravo!

Regarding your retreat back to your nonsensical position, in WWII, neither the urban guerilla warfare, the insurgent resistance movements nor the aerial bombing resulted in battlefields strewn with bodies. That's because none of those styles of warfare concentrates large numbers of soldiers on relatively small battlefields for relatively short periods of time. Similarly, the total mortality in Iraq is spread over an entire country and 365.24 days per year instead of a handful of battlefields on a handful of dates. The lack of battlefields strewn with bodies, while vivid, in no way suggests the survey results lack validity or merit.

Bob Badour said at October 14, 2006 2:01 PM:


The quality of the conversation was already so low I didn't see how an insult could hurt it either.

Certainly, there is some uncertainty surrounding the exact number of deaths, which the scientists who collected the survey acknowledge and report in their findings. Does it make any difference whatsoever whether "Iraqi Freedom" increased violent mortality by 400,000 deaths or 800,000 deaths? One thing remains certain: "Iraqi Freedom" made Iraq a much more dangerous place for Iraqis. Would that observation change if the increase was actually between 200,000 and 400,000 violent deaths? Not just no but hell no.

What is even more important is the agreement among all mortality estimates that the rate of mortality is increasing with time--and quite rapidly. Will mortality continue to accelerate? How much higher can mortality go before we do have 'battlefields strewn with mangled bodies' ? At what point will they kill off enough young men that mortality decreases?

Bob Badour said at October 14, 2006 2:23 PM:

perroazul del norte,

The American Civil War and the conflict in Iraq are very different conflicts. The Civil War was characterized much more by large armies lining up on either side of battlefields and trying to kill each other en mass. The conflict in Iraq is characterized much more by intertribal murder and by 'hit and run' insurgency.

My point was not that intertribal murder was a huge contribution to mortality in the Civil War, but that it did contribute to total mortality of the Civil War, and that the figures Lawrence cites omit those deaths entirely. Similarly, the conflict in Iraq is not totally without battlefields; however, most of the deaths do not come from battlefields.

Paraphrasing you: "No one with even a passing familiarity with current events would think that all the massed battles combined added up to anything more than a very trivial percentage of the deaths related to hostilities that occur during the current conflicts in Iraq."

It is specious to compare a subset of the violence of one conflict with all of the violence of a very different type of conflict.

Rockin' Warrior said at October 14, 2006 3:11 PM:

"This 600,000 figure is ludicrous. That's the number of dead in the American War between the States, with vast battlefields covered with the dead."

That's a lame argument, Lawrence. The population density of the USA in 1861 was much, much less than the population density of Iraq in 2006, especially considering the highly populated Iraqi cities. Moreover, Civil War soldiers never had access to so-called "bunker buster bombs" and daisy cutters that could rain mass death down on civilian populations, as the US can currently do and indeed does regularly in Baghdad and Anbar.

The 600,000 number sounds just about what I expected. As Randall said, actual reported deaths in a war like this generally undercount by 5, and it's only by solid data analysis, trips to the morgue, and examinations with families using independent surveys that one can obtain something approaching an accurate picture. In fact, since the Iraq War has so thoroughly devastated Iraq's public health system-- especially running water, electricity and the water treatment facilities necessary for modern sanitation-- a very, very large proportion of Iraq's dead are going to be young children who perish from infant diarrhea and similar diseases. Remember, what devastated Rome more than anything else after 476 A.D. was the destruction of the Roman aqueduct system-- this was what actually killed the Roman cities, by precipitating the sort of public health crises that devastate entire generations.

I've worked in the data analysis field (and I'm generally politically conservative). But from a cold, detached perspective, the Hopkins study is top-notch. It would be a shame to bury such a fine accomplishment of scholarship in political bickering.

Rockin' Warrior said at October 14, 2006 3:18 PM:

John S. Bolton wrote,

"If you kill thousands of Americans, we will kill hundreds of thousands of yours."

And who was it that killed "thousands of Americans," John? Somehow, I don't remember any Iraqis among the September 11, 2001 hijackers, and ANY link between Saddam (and the secular Baathists) with Sept. 11 has been utterly discredited. IOW, Iraq had nothing to do with 9/11, yet it was the Iraqis who were attacked and murdered by the US government. So the message is not "kill thousands of Americans, and we'll kill hundreds of thousands of yours" but "kill thousands of Americans, and we'll go kill hundreds of thousands from somebody else's innocent tribe and country while we leave the real perpetrators alone."

You sound like another one of those chicken-hawk losers (and yes, I'm using that term) who like to talk tough but are too cowardly to actually go and fight yourself. This stupid Iraq War has ruined the Conservative movement in the US before we could accomplish squat-- call me a Paleoconservative if you like, but I can't see what in the world is "conservative" about launching a war of aggression against a country that didn't hit us on 9/11, not even slightly. If you believe in the Iraq War so much, I suggest you walk to your nearest Army recruiting office and sign on the dotted line-- we are losing there in large part due to a manpower shortage, after all. Otherwise, you're just talking out of your ass.

Rockin' Warrior said at October 14, 2006 3:34 PM:

crush41 wrote: "Even if said estimates are true, Iraq is still birthing three people for each one that dies! Is that enough to scare the Islamic world out of ticking off the US?"

Again, Crush, you fall into the trap of equating "War on Iraq" with "retaliation for 9/11." The Iraqis had nothing to do with 9/11, so there's no lesson here whatsoever. That is, the bloodiness of the Iraq War delivers no useful message whatsoever-- if anything, it's telling Iraqis (and civilized people in general), that even if you're innocent of a massive offensive attack, and have nothing whatsoever to do with such a bloody act of war, it doesn't matter. The US will let the real culprits go free, while using 9/11 as a pretext to go and kill Iraqis for greedy objectives.

A better example of a "just retaliation" would be e.g., what the Chinese did to the Mongol heartlands after the Mongols were kicked out of China in the 14th century. The Mongols had killed tens of millions of Chinese and brutalized dozens of other peoples before the Mamluks and the Japanese defeated them, and so the native Chinese were understandably pissed when they got their chance to deliver some hurt to the Mongols. The Chinese went into the Mongol capital and burned it down, while scuttling other Mongol territories, looting the Mongol cities and tribal settlements (which were themselves filled with loot from China), and introducing a pacifistic form of Buddhism to shut down the Mongols as a political force-- this was what ultimately doomed Genghis Khan's dream of a Eurasian Empire. The Chinese actually were merciful in context, and didn't kill the Mongol civilians by and large or even Mongol soldiers en masse, but they permanently broke Mongol power. It was a proportionate response against a clear and accurately identified enemy to deliver a clear message-- "mess with us like that again, and we'll take you down, even if it takes a century to do it."

With the War in Iraq, the US has done the opposite and blundered unimaginably. We didn't hit those who hit us-- we hit an outside third party that was not involved in the 9/11 attacks and even more ironically, was if anything a US ally against our common enemy (al-Qaida, which had also targeted Saddam and the Baathists). We haven't demonstrated that we avenge attacks against us-- we've demonstrated that we're too clueless and stupid to actually shoot straight.

crush41 said at October 14, 2006 4:32 PM:


The survey method seems to be the best we're going to get. I don't like to speculate on things when I'm cognizant of my severe limitations, but dishonesty by those being surveyed stikes me as most likely. Most Iraqis want us out, and the Shia stand to benefit from our withdrawal, so why not inflate the numbers? The more damage we do (or are perceived to have done) to Iraq, the more the international community feels pressured to help through wealth transfers to the country and the more PC our 'strategery' must become. Iraqis have incentives to inflate the numbers.

What is the death rate in the Kurdish north? With around 5 million Kurds, that would make central and southern Iraq numerically comparable in total population to the US during the Civil War.

You speculate that family members of some of those that got whacked while doing bad things would instead report the death as something more benign. That would understate the number of violent deaths and overstate the number of non-violent deaths. But the number of non-violent deaths is almost certainly understated even with that artificial inflation, a point that makes me most suspicious.

Breaking down the non-violent deaths annually yields a death rate of about 1 in 2000. If we assume that all of those non-violent deaths are suffered by the 800,000 Iraqis over the age of 65, that still yields only 1 in 56 people 65+ dropping off each year, in a country where the average life expectancy is only 69 years. Of course, it's not exclusively old fogies (they only represent 3% of the population) who are dying non-violent deaths. So that would suggest the non-violent deaths of the old occur at a rate substantially lower than 1 in 56. It just doesn't make sense.


Iraq is still going to have a very high population of male youngsters in proportion to the total population. Even if these likely exaggerated results are true, for every Iraqi that dies three more are born.


John's not insinuating culpability for 9/11. I think he's pointing out that the results probably scare other leaders of the Islamic world by letting them know if they do something we don't like, and we happen to invade, their countries are going to suffer from massive loss of life and chaos. So they better do what we say. He's making a Machiavellian observation, not a moralistic one.

crush41 said at October 14, 2006 4:35 PM:


I'd say the Chinese reaction was less than just--it was too lenient. The Mongols, who created virtually nothing and prospered almost exclusively through plundering, regularly 'accepted' surrender and then when let into a town, enslaved the healthiest men, took the women they wanted, and killed the rest!

Randall Parker said at October 14, 2006 4:42 PM:

Just how high is the death rate in Iraq? How many families are losing members to the killers? Iraq's Sunni VP just lost his 3rd sibling killed since spring 2006.

BAGHDAD, Oct. 9 — Men wearing military police uniforms broke into the house of the brother of Iraq’s Sunni vice president on Monday, chased him onto a neighbor’s roof and shot him in the head, killing him, Iraqi authorities and witnesses said.

Amir al-Hashemi was the third sibling of Vice President Tariq al-Hashemi to be killed since spring. His death underscored just how deeply Baghdad has sunk into lawlessness, particularly in its religiously mixed neighborhoods, and was similar to the politically motivated assassinations that have plagued Iraq since the American invasion.

A bomb in a parked car exploded at nightfall on Monday in a crowded market area in Shaab, a predominantly Shiite neighborhood, killing at least 13 and wounding 46, a police official said. It was the first large bombing in the capital in almost a month, and brought the number of Iraqis killed in violence on Monday to 18.

In addition, Iraqi authorities said they found 57 bodies in eastern and western Baghdad.


Killings of politicians have become grimly familiar, but the one on Monday stood out. The killers wore what looked like official uniforms, enabling them to surprise and overwhelm Mr. Hashemi’s guards. They then seized at least seven neighbors who witnessed the attack, including the neighbors’ children and an elderly bakery worker. As of Monday night, their whereabouts were still unknown.

How many bodies don't they find?

Some people do not find an estimate of hundreds of thousands dead from the post-invasion violence as credible. I am thinking we really do not know how bad it is there. Never mind that we have 140,000 troops there. We do not have the means to measure the real level of violence. But parts of the government are hunting down and killing Sunnis - some specific targets and others just because they run across them.

Randall Parker said at October 14, 2006 4:53 PM:


This is not a valid excuse:

The quality of the conversation was already so low I didn't see how an insult could hurt it either.

This amounts to arguing that if you decide someone else has lowered the level of quality of conversation then you feel free to indulge your own desire to carry out conversation at that low level.

I try not to react irrationally and counter-productively to initial emotional response or responses that do not provide sufficient justification. I prefer coming back with reasoned arguments. I figure the people I'm arguing with have the capacity (and Larry does have considerable intellectual capacity and knowledge as I've told you) to flesh out the reasoning in their minds and present rational arguments.

You are factually wrong to call Larry an idiot. Furthermore, your tone of conversation is counter-productive and makes you look bad to other readers here. Do you want to make yourself look bad?

Rockin' Warrior said at October 14, 2006 4:56 PM:

I agree with your assessment of the Mongols, crush41-- I have little respect for them. The Chinese, Japanese and even the Arabs (at the time, who were still amazed enough by the Greeks and independent-enough thinkers in their own right to resist the Inquisition-like anti-intellectual mentality that would soon overtake them) were busy creating advanced civilizations and advancing knowledge, culture, even science and technology. The Chinese were the inventors of paper after all, as well as the navigation tools that Europe found so useful later on-- our Scientific Revolution had a lot to do with the tools diffusing from the Chinese.

Yet the Mongols contributed almost nothing, just pillaged and plundered their way through the more advanced civilizations around them. This is why they left so little cultural imprint, and why their enemies eventually turned the tables on them and defeated them militarily.

On this statement: "John's not insinuating culpability for 9/11. I think he's pointing out that the results probably scare other leaders of the Islamic world by letting them know if they do something we don't like, and we happen to invade, their countries are going to suffer from massive loss of life and chaos. So they better do what we say. He's making a Machiavellian observation, not a moralistic one."

I agree it's a Machiavellian observation, and my point is, it fails as badly as a Machiavellian observation as it does as a moralistic one. If the message is that, "If we get attacked, we're gonna use that as an excuse to beat up on anybody we don't like," that doesn't exactly help us to achieve objectives in the region-- it only encourages just about everybody to gang up against us for their own protection, since they can't trust us to be anything remotely resembling a fair arbiter. IOW, the US, in invading Iraq, is seen to have embraced the might makes right doctrine-- there was no moral justification for invading Iraq, only a Machiavellian one. The problem with this, of course, is that others can play that game too.

This is a big part of why Iran is pushing so hard to obtain a nuke, since in a "might makes right" world, only such a dangerous deterrent speaks loudly enough to resist an invasion, at least in the minds of the Iranians. That may be a perfectly rational response-- I don't know. But clearly, Iran has seen that (a) Iraq, which lacked any WMD and had a weak military, was invaded and brutalized by the USA for no good reason except the greed of its government elites and their cronies, while (b) North Korea, an even more repugnant regime which *does* have WMD, has been basically left alone. It's not so hard to see why Iran has drawn the conclusion that it did.

This, of course, does not mean in any way that I look forward to Iran getting nukes-- that prospect scares the crap out of me. The point is, I doubt that Bush would try anything to stop it at this point, even limited air strikes-- though IMHO the best thing to do would be just a 1 on 1 talk and a little humble pie. This "we refuse to negotiate with a rogue state" is the worst of both worlds, failing to engage Iran while simultaneously encouraging its nuclear proliferation plans. In any case, Bush feels that he is so weak, his first instinct right now is probably abject capitulation to Iran, thus punting the problem for the next President to solve. If I had to be bet money on the Bush policy toward Iran now, it would be meek acquiescence.

Randall Parker said at October 14, 2006 5:00 PM:


Are you talking about the non-violent deaths in the 650,000 incremental deaths? Or are you talking about the total non-violent deaths? These guys are talking about an incremental increase in deaths of 650,000. The pre-existing death rate accounts for the deaths of old people from old age. So is your calculation using the total or the increment?

If your calculation is yielding a total death rate of old people that is too low then I'm surprised and think that makes their figures suspect. But I am thinking they didn't make that mistake. Are you making a mistake or are their figures for old age deaths really too low?

So what is your calculation?

Randall Parker said at October 14, 2006 5:02 PM:


Another question I have: What percentage of all deaths were reported to the Iraqi government under Saddam Hussein? Did the poor villagers just bury their old dead folks without reporting their deaths? I'd love to know the rate of death certificate issuance under Saddam and today.

Randall Parker said at October 14, 2006 5:09 PM:

Rockin' Warrior,

There was no Machiavellian justification for invading Iraq that furthered US interests - at least not a justification that was correct.

North Korea's regime was left in place because it had the ability to rain massive amounts of artillery shells down on Seoul South Korea. WMDs were not necessary to protect North Korea. Even now North Korea does not have the ability to deliver a nuclear warhead into South Korea. North Korea might that capability in a couple of years - but not quite yet.

Yes, Bush screwed the pooch so badly with Iraq that this pretty much guaranteed Iran's ability to get nukes. The US is overstretched in Iraq. Both Iran and North Korea feel emboldened as a result. I think nuclear weapons proliferation control is pretty much dead at this point.

crush41 said at October 14, 2006 5:20 PM:

Wow, I don't know why I had that 50,000 as a total in my head. No, my calculation was based only on the suggested incremental increase and therefore meaningless. Sorry for the confusion. How embarrasing. Geez.

Bob Badour said at October 14, 2006 6:10 PM:


I was just about to hit Post when I saw your retraction. I am posting it anyway because I think it puts the numbers in perspective.

Check out Tables 2 and 3 in the PDF file of the report in The Lancet. You will see that about 45% of the total post-invasion deaths are non-violent, that 92% of the post-invasion deaths of the elderly are non-violent, and that the non-violent deaths have remained relatively constant somewhere between 4 per 1000 per annum and 7 per 1000 per annum over the study period.

Assuming the pre-invasion mortality continued post-invasion without any change, the Iraqis would have 430,000 bodies to deal with from non-violent causes in any case. One would not expect to find those bodies strewn across battlefields and would not question the number due to any absence of body-strewn battlefields. Similarly, the US disposed of about 2.5 million bodies last year without a single body-strewn battlefield and over eight million bodies since the invasion of Iraq based on 8 per 1000 per annum and a 300 million population.

The survey finds total mortality increased dramatically post-invasion and that most of the increase was due to violent deaths. Thus the Iraqis had to dispose of an additional 650,000 bodies over and above the 430,000 they would have disposed of anyway (that nobody in the US would have noticed in any case) with about 600,000 of those additional corpses dying from violence. The survey sampled 2 violent deaths in the year prior to the invasion and 300 in the 3.5 years since. That's about a 50-fold increase.

Is it possible there were about 4000 violent deaths per year in Iraq the year prior to the invasion in a population of about 30 million? Does that seem at all credible to you? Do you believe a 50-fold increase is possible (even likely) in the post-invasion anarchy complete with skirmishes, death squads, aerial bombardment etc.? Because a 50-fold increase would give about 600,000 dead bodies from violent causes--with or without body-strewn battlefields.

What does "about 600,000" bodies mean? It means somewhere between 400,000 and 800,000 with a higher probability near the middle of that range.

Also check out figure 4 on the second-last page of the report.

crush41 said at October 14, 2006 7:11 PM:


Man, how careless of me. I was at work and the pdf wouldn't open so I just read the article. Shoulda kept my mouth shut but misread the article and the low number of non-violent deaths stuck out at me so I ignorantly blathered without checking myself (why I describe myself as audacious).

But that sort of slight increase, not far from what had been reported pre-invasion (as you point out), makes sense given the infrastructure deterioration and level of insecurity in many of Iraq's cities, which is having some detrimental effect on the population's health.


My apologies for throwing up something that'll just confuse people who stop somewhere in the middle of the comments.

Bob Badour said at October 14, 2006 8:18 PM:
Do you want to make yourself look bad?

Since you have known me, have I ever stood out as a person who cares how other people see me? Especially online?

John S Bolton said at October 15, 2006 1:08 AM:

I still think doubt is reasonable for extraordinary claims, but that a figure in the low hundreds of thousands for additional deaths is not so unreasonable.
I don't believe that Iraq caused 9-11, but that moslem killers did, and their fellows are taking the excess deaths.
If the military wanted me, after hearing what I say about the 'religion of peace' and the chances for democracy in a moslem country, their indoctrination must have gotten a lot easier than one would have guessed.
The Islamic faith is a military indoctrination, commanding atrocities for so long as there are any infidels, and it is full of mysticism.
This is why I believe they are best dealt with by casualty ratios mounting up very high against them, even 100:1 as in the 19th century on many occasions.
What the moslems believe is being done to them can be as important as what is actually occurring.
The study will most likely be believed by moslems all over the world as representing, not a high estimate, but a low one.
They will beieve that deliberate policy from DC set them up to take these casualties, and they will feel it as their own loss.
This is all to the good for us, because the rational parts of the world will doubt, while the believers whom we ought to have intimidated, will be.

John S Bolton said at October 15, 2006 2:01 AM:

A point that has not been raised is that the Iraqi collaborators could be under death threats against themselves and their relatives, if they did not generate results sufficiently high to serve Islamic propaganda about how they're so hard done by.

Bob Badour said at October 15, 2006 7:22 AM:


I direct you back to your comments about ad hominem. Have you read the published report? (PDF warning) The researchers themselves acknowledge the threat of death to the researchers and acknowledge the ways the researchers deviated from a purely random sample for their own safety. I believe they were more concerned they would be murdered for cooperating with Americans than for failing to forge results.

On one or two occasions, the surveyors chose a different location to survey from the one picked at random due to perceived danger. Three of the samples were discarded due to miscommunication and at least one of those because the surveyors surveyed a location in a different district when they moved the sample point due to concerns for safety.

Other faculty at the Iraqi university that collaborated in the survey have been murdered, and I suggest that the researchers from that university are already at risk of murder just for being who they are. The researchers themselves acknowledge the surveyors may have responded to subconscious signals to avoid or to choose one house over another; although, the researchers are unaware of any specific bias.

As others have explained, there are as many reasons to believe the survey might underreport deaths as overreport them.

I don't fully understand your resistance to this survey, and I don't find the reasons you give at all convincing. Whatever your reasons, if you can accept that--since the invasion--there has been an increase in deaths numbering in the low hundreds of thousands rather than in the mid hundreds of thousands, aren't the qualitative conclusions the same? That--rather than liberating Iraqis--the occupation has ruined their lives and killed their friends and family?

What did the Iraqis ever do to us to deserve such abuse?

Randall Parker said at October 15, 2006 10:52 AM:


You do care what other people think of you. But you have a stronger urge to satisfy your immediate urge to lash out than you do to earn and keep the respect of others. You are rationalizing why you think your behavior is alright with you. If you looked harder at yourself I do not think you'd find that rationalization holds up.

As for your:

That--rather than liberating Iraqis--the occupation has ruined their lives and killed their friends and family?

You seem to be arguing with a strawman version of Larry. He's quite the critic of the war and the Bush Administration's conduct of it.

As for your:

I don't fully understand your resistance to this survey,

Yes, there are reasons why the survey might underreport deaths. But there are reasons the survey might overreport them. We have no way to check whether the bigger biases are toward overreporting or underreporting.

You are taking the stance that you are defending the more reasonable position. But we really do not know how high the rate of killing is in Iraq. Modesty for one's claims is in order, not insult-fests.

Bob Badour said at October 15, 2006 11:25 AM:
"That--rather than liberating Iraqis--the occupation has ruined their lives and killed their friends and family?

You seem to be arguing with a strawman version of Larry. He's quite the critic of the war and the Bush Administration's conduct of it.

I wasn't arguing with Lawrence. I was asking John a rhetorical question.

You are taking the stance that you are defending the more reasonable position.

I am defending the more reasonable position, and I have been very clear that we do not know the exact number of deaths. Are you not arguing against a straw man version of my position?

ziel said at October 16, 2006 6:05 AM:

For the record, I requested the raw household data from the lead study author. He responded that they cant' release it at this time due to confidentiality concerns. He thinks they might be overcautious at this time and in time may release it, but they're not ready to yet. I certainly accept what he's saying at face value, but does make fully evaluating the study difficult. At any rate, 200,000 or 400,000 or 655,000 are all numbers of deaths far too high to comprehend in the world I live in.

Bob Badour said at October 16, 2006 8:30 AM:


Yesterday, I thought about the issue of the raw data, and it occured to me the researchers might need to restrict its distribution. It seems one of the important issues people would want to examine is the randomness of the sampling or any potential sampling bias. That would mean the researchers could not effectively hide the locations of the people surveyed. It is at least conceivable that some groups might target all of the houses in the survey areas for some kind of retribution for cooperating with the survey.

I respectfully suggest it is not the world you live in as much as it is the human mind that cannot comprehend large numbers. Do you really comprehend a population of 26 million?

I currently live in a rural county with a small population: West Prince, Prince Edward Island. According to population statistics I found from 2001, West Prince has a population of about 14,000. That number is more comprehensible.

If we scale the 650,000 violent death numbers from 26 million Iraqis to 14,000 West Prince residents, we get about 350 over 3.5 years for a population the size of West Prince. That's 100 deaths a year or about 2 per week. Opening last week's West Prince Graphic to page 11, I find 2 funeral announcements, which is about about typical.

Now, West Prince is a peaceful--even bucolic or pastoral--area with an older population. If I imagined a different, younger West Prince where the Acadians and later British settlers were duking it out with death squads and IEDs while the Mi'kmaq set up an independent enclave on Lennox Island, could I imagine an additional 2 deaths per week? Yes, I could. I would have no trouble comprehending or believing that.

Mark said at October 16, 2006 11:32 AM:

A lot of people are reporting a 'feeling' that this result just can't be right. I share that, but acknowledge the point of the people who don't accept feelings as reliable, even if the impressions are formed by intelligent sceptical people with broad historical knowledge. So instead, take a look at the Iraq Body Count analysis of the implications of the Lancet numbers:
E.g., among many others, if the Lancet is right then 'Some 800,000 or more Iraqis suffered blast wounds and other serious conflict-related injuries in the past two years, but less than a tenth of them received any kind of hospital treatment'.
The IBC people extrapolate a bunch of similar implications, and they have a feeling that they are bogus. They might be true - there might be 'bizarre and self-destructive behaviour on the part of all but a small minority of 800,000 injured, mostly non-combatant, Iraqis'. But I have a feeling they are as utterly bogus as they first appeared.
And it is okay to suspect the findings on the grounds that the people that produced them are extreme anti-Bush Liberals, and the grounds of the timing, and the grounds that the editor of the Lancet is a prime and proud useful idiot for the jihad. It's more than okay - anyone that didn't suspect the findings on those grounds is foolish. The left has never been shy to subordinate truth to the higher mission they have appointed themselves - they know best what the world needs, and inconvenient facts in the way of pushing us there are irrelevant to their zeal.

John S Bolton said at October 16, 2006 2:53 PM:

From Dr John Jay Ray's Dissecting Leftism site: What....
" explains why the study appeared in a medical journal rather than a more statistically sophisticated journal is that there was NO VALIDATION of the survey results. That your survey-takers might just sit down under a tree and "make up" their "interview" results is a routine peril and it is routine to take precautions against it -- usually by going back on a later occasion and checking with the alleged respondents a proportion of all interviews handed in. Just the awareness that a sample of the respondents will be re-interviewed tends to keep the interviewers honest -- though not always so, regrettably. So the results reported in the Lancet study have no credibility at all and must be regarded as garbage."

Randall Parker said at October 16, 2006 4:58 PM:


Maybe the ratio of deaths to injuries is a lot higher than the IBC people are assuming. Look, the Iraqi medical system isn't well supplied. Lots of doctors have fled the country. Sunnis are afraid to go to Baghdad hospitals because Shia death squads go into the hospitals and kill Sunnis.

Randall Parker said at October 16, 2006 5:47 PM:


I do not know how to interpret this sentence:

At any rate, 200,000 or 400,000 or 655,000 are all numbers of deaths far too high to comprehend in the world I live in.

Are you saying it is hard to grasp such numbers? Or that they are too high to be believeable? What do you mean by "comprehend"?

Cody said at October 17, 2006 8:25 AM:

I don't really have a dog in this fight, but I would like to point out that the Iraq Body Count (IBC), an anti-war organization that seeks to quantify the number of civilian deaths in Iraq, seems to have sided with the skeptics on this issue.

Currently, IBC reports that civilian deaths to be between 43,937 (min) and 48,783 (max).

Jake said at October 18, 2006 3:16 PM:

My apologies if this simple arithmetic has already been presented earlier in this thread ...

654,965 war related deaths/3.5 years/365 days/year = approximately 510 deaths/day

thats an average of 510 deaths per day ... every day ... for 1278 days

Bob Badour said at October 19, 2006 7:09 AM:


And that works out to one daily death for every 843 square kilometers, which is like a square 29 kilometers wide by 29 kilometers long. (18 miles by 18 miles) Granted, the deaths won't distribute evenly.

(I am preparing a longer reply to some of the links provided a couple days ago. I caught a headcold and I am just waiting for my head to clear a bit first.)

Lawrence Auster said at October 19, 2006 2:46 PM:

I'm not surprised to find out that the methodology used in the Johns Hopkins study is grossly, scandalously flawed. Steven E. Moore (not the Stephen Moore of the Cato Institute), writing at Opinion Journal, explains that the study used vastly fewer "cluster points" than were called for in a random study of this size. "With so few cluster points," he writes, "it is highly unlikely the Johns Hopkins survey is representative of the population in Iraq." He then goes on to explain that when he spoke with the director of the Johns Hopkins study, he found to his astonishment that the study had not asked demographic questions of the respondents, which is necessary in such studies to determine if there is a representative sample of the population.

This is, in short, a bogus study, having approximately the same credibility as, say, one of those network news report from the 1980s saying that 40 percent of American children were suffering from hunger.

To anyone exercising a modicum of common sense, the figure of 655,000 extra deaths in Iraq over the last 3 1/2 years would have seemed extremly unlikely on its face. The reason some people believed it was that they wanted to believe it.


Bob Badour said at October 19, 2006 8:01 PM:
Steven E. Moore (not the Stephen Moore of the Cato Institute), writing at Opinion Journal, explains


You accused the researchers of obvious political motivation in your ad hominem outright dismissing the survey. To attack their methodology, you cite a political consultant employed by the Republican Party. If alleged political motivation supposedly renders the survey invalid, doesn't receiving one's income from the governing political party render one's criticisms of the survey invalid?

Bob Badour said at October 19, 2006 8:09 PM:

Oh, and who has the self-declared mission of putting a positive spin on the US military role in Iraq. Not that his obvious political motivation necessarily discredits his argument, but if we use the standard of your prior argument, I wonder why you bothered.

Bob Badour said at October 20, 2006 7:54 AM:


I apologize for taking so long to get back to your comment about Dr. John Jay Ray's remarks [he types while coughing up a lung.]

Dr. Ray writes a polemic comprising mostly ad hominem. He makes three specific criticisms--one of which you noted in your comment above:

1) Dr. Ray complains that no validation was done to prevent the interviewers from just making up the results.
2) Dr. Ray dismisses the 98% response rate as an amazingly high implausibility.
3) Dr. Ray calls the figures about the death certificates a lie.

Taking these in reverse order, Dr. Ray provides a link where he calls the death certificate figures a lie, but he provides no proof of the assertion whatsoever and neither does the polemic he points to. In the end, that assertion is nothing more than a backhanded ad hominem accusing the researchers of lying.

Dr. Ray's own research confirms that high response rates are possible and depend largely on cultural factors. Given that the COSIT ILCS study linked from the IBC press release provided by Mark and Cody reports a similar 98%+ response rate, it is at least possible Iraqi culture--similar to the Indian and white South African cultures--leads to high response rates. Then again, Dr. Ray's research also suggests that any feeling of compulsion on the part of the responders can eliminate any benefit of a high response rate.

Finally, Dr. Ray makes proud claims that he performed many of his door-to-door cluster analysis surveys himself. See for one example the 1985 paper linked above. One wonders then whether Dr. Ray validated his own surveys to prevent himself from making up the results. Assuming he did not, one must consider what conditions allow one to forego such checks, and I assume those conditions relate to trust. While perhaps a flaw, it is conceivable Les Roberts had almost as much trust in the four interviewers used as Dr. Ray had in the one interviewer he frequently used. Dr. Ray has no hidden motive--his weblogs including Dissecting Leftism are overtly political. If his criticisms were less political and resorted to fewer fallacies, I might give his hyperbole regarding this flaw greater weight. While it is a flaw, I don't necessarily accept his assertion that the flaw renders the results "garbage".

One of Dr. Ray's ad hominems relates to the report noting that the four interviewers were all medical doctors fluent in english and arabic, which he calls gilding the lily. While I don't think medical school necessarily prepares one as an interviewer, one would have to suspect the report as trying to conceal bias if it failed to disclose such a common background among the interviewers. While I have no evidence to support the assumption, I think it is fair to assume the four doctors are instructors or researchers at the medical university in Baghdad that participated in the survey rather than general practitioners or surgeons pulled away from active practice as Dr. Ray suggests.

I see that Dr. Ray posted an update where he linked to the IBC press release and to the political consultant employed by the Republican Party to put a positive spin on the Iraq war and more-or-less said "Yeah, and that too."

Mark and Cody,

I again apologize for the delay.

First, the IBC press release makes a specious extrapolation to assume 800,000 wounded never sought medical treatment. Given the damage to infrastructure and the exodus of many doctors, any seeking medical treatment would likely have sought it within their tribe or from their fellow insurgents, and many might have simply bled out and died for lack of treatment. I doubt the death squads pulling adult men out of their homes for torture and execution left many wounded.

The IBC summary also makes the specious assumption that violent deaths are necessarily caused by coalition forces, when much of the current carnage is clearly the result of inter-tribal murder.

The IBC press release offers the COSIT ILCS Survey from 2004 as a contradiction of the Johns Hopkins survey from around the same time period. However, the two surveys actually agree as one would expect. The ILCS 2004 survey sampled more households and more clusters so one would expect a narrower confidence interval. One would also expect the ILCS 2004 survey to occupy some range entirely within the confidence interval of the Johns Hopkins survey from 2004, and it does.

The question is: If the ILCS survey were repeated right now (and perhaps it is already in process, I don't know), would it still occupy a range within the confidence interval of the latest Johns Hopkins survey? If so, one would expect it to offer a narrower confidence interval somewhere over 400,000 excess and violent deaths.

What the ILCS Survey does contradict, on the other hand, is the IBC's numbers giving a minimum count that exceeds the IBC's maximum count at the time.

The ILCS Survey noted underreporting of mortality with respect to infant mortality, and in fact, repeated that part of the questionnaire. (Page 50 of the Analytic Report-English) One wonders whether there was a similar underreporting of mortality among military aged men -- especially if the family members knew the men were insurgents. The ILCS 2004 also noted a relatively large number of dead fathers of children born after the 1991 Gulf War, which is exactly what one would expect from elevated levels of mortality among military aged men due to the current war. (Table 40 on page 54 of the Analytic Report-English).

Finally, the IBC notes that the reported civilian deaths they record are reported by an average of 6 news reporters. I think this suggests the news reporters get their leads from the same sources more than it suggests comprehensive reporting.

On the other side, The Lancet's editor Richard Horton has appeared recently at political rallies etc. like the recent Time To Go Demonstration on September 23 revealing that he at least does have a political motive. This might explain some of the flaws of presentation like reporting a calculated figure to 6 digits precision and then rounding it to 3 digits precision when the confidence interval suggests less than 1 digit of accuracy. (6+-2)*10^5 is different from 601,000 for instance and reporting 601,000 amounts to misleading vividness.

In summary, in addition to the flaws already noted in the report, two additional flaws have been identified: 1) a lack of repeated questionnaires to validate the interviews and 2) a lack of demographic data to validate representativeness.

After the 2004 survey, the researchers increased their sample size to narrow the confidence interval. If they have to repeat this survey in another 2 years, hopefully they will address the demographic data issue. While it would not be reasonable to demand it of a survey in a place like Rwanda, it might improve confidence in a place like Iraq -- assuming one can trust the available census data.

In an audio interview of Les Roberts, he suggests it should be easy to verify the increase in deaths simply by interviewing the people responsible for graveyards asking them whether they have noted a four-fold increase in business since the invasion. Likewise, a number of Iraqis participate in the blogosphere, I wonder whether they might share their experiences. If hundreds of thousands are dead by violence, almost everyone in Iraq would know a couple people killed. What I have read, suggests to me that most Iraqis do know multiple people who have died; although, that observation is certainly subject to "Man bites dog" bias.

Bob Badour said at October 20, 2006 11:37 AM:

As a follow-up to my suggestion regarding Iraqi bloggers knowing anybody who died, I thought I would look up Salam Pax to see if he had anything to say about The Lancet report.

Salam Pax has not been very active, and his second-last blog suggests he knows of a couple or three violent deaths. His previous blog just sort of stopped in 2004 just before the violence really started to ramp up according to The Lancet. The April 6, 2004 blog has some few things to say whether anybody could hide violent deaths on a massive scale.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright ©