Reaction to the AAPOR press release

I asked Mary Losch (chair of AAPOR’s Standards Committee) to comment on my previous post

I have read your entry and would note that the links you provided did
not supply the questionnaire items but rather a simple template (as
noted in the heading). The Johns Hopkins report provides only
superficial information about methods and significantly more detail
would be needed to determine the scientific integrity of those methods
– hence our formal request to Dr. Burnham. The Hopkins website
refers to data release but, in fact, no data were provided in response
to our formal requests. Included in our request were full sampling
information, full protocols regarding household selection, and full
case dispositions — Dr. Burnham explicitly refused to provide that
information for review.

We do not provide public reports of the investigations but if there
are other specific questions that I could answer, I would be happy to
try to do so.

It is more than a little ironic that AAPOR has censured Burnham for not fully disclosing the data behind his study, while themselves failing to fully disclose the basis for their serious charge.

In response to a question from Mark Blumenthal she added:

we requested the survey instrument (including consent information) and it
was not provided. The template did not appear to be much beyond an outline and
certainly was not the instrument in its entirety.

I don’t think that Losch or the AAPOR have disclosed, specifically, what is missing from the information that Burnham has provided. I stand by my earlier statement that their press release was misleading. I checked their website and they’ve now put their press release up, but provided no supporting information. They’ve also linked to a paper by David Marker published in their journal which implicitly contradicts their press release. Marker clearly felt that Burnham had published enough information for him to evaluate their mthodology, writing:


Burnham et al. attempt to estimate the number of excess Iraqi war dead throughout the country using fairly standard survey methodology, for which they are to be commended. We have examined four specific methodological factors: coverage errors, correct probabilities of selection, migration, and training and control of interviewers. Coverage provided by the first stage of sample selection (the sample of administrative units) appears to have been complete (other than the exclusion of area considered as too violent to allow household interviews as well as the exclusion of the three sampled clusters for the reasons the authors indicated). The coverage at the second stage appears less than complete. In addition to the exclusion of some deaths mentioned earlier (short term household members), there may have been systematic exclusion of some types of households or housing units. The extraordinarily high response rates reported suggest this as a possibility. It should be noted that for the two governorates where data were not collected, they used an underestimate of no excess deaths, choosing to err on the side of understating mortality in these unknown cases. This level of coverage represents a major improvement over other reported estimates.

Tim Parsons responded to my query on behalf on the Johns Hopkins Bloomberg School of Public Health:

The level of civilian mortality in Iraq is a controversial subject. Questions have been raised regarding the findings and methodology of the 2006 Iraq mortality study conducted by Dr. Gilbert Burnham and published in The Lancet. The Johns Hopkins Bloomberg School of Public Health takes any allegation of scientific or professional misconduct very seriously. It believes that the correct forum for discussing the reported findings of the Lancet study and the general methodology that led to those findings is in the regular exchange of views in the scientific literature. The Bloomberg School of Public Health has undertaken a review of the study to determine if any violation of the School’s rules or guidelines for the conduct of research occurred in the conduct of the study. That review is nearing completion and the School is unable to discuss the results at this time.

The American Association for Public Opinion Researchers (AAPOR) chose to criticize Dr. Burnham for failure to fully cooperate with the organization’s review of his 2006 study. The Johns Hopkins Bloomberg School of Public Health is not a member of the organization and does not know what procedures or standards were followed in reaching the decision regarding this study and therefore is not in a position to comment on the decision.

I guess we’ll have to wait for the Johns Hopkins report.

Megan McArdle adds to the mountain of her errors on the Lancet studies with this:

I happened to be writing my story just as the World Health Organization study that was highly critical of Burnham, et. al. was released. Les Roberts, who had become the public face of the team, was making frankly lunatic claims on the radio that the two studies basically agreed, even though the introduction to the WHO study specifically said that their results made it very unlikely that Burnham et. al. had been correct. This claim was so unusual that when I asked neutral conflict epidemiologists, they patiently explained that I couldn’t possibly have heard Roberts correctly, because no one with half a brain would ever have said that.

McArdle does not seem to have understood what Roberts was saying was in agreement: the excess deaths in the Lancet study (about 650,000) and in the IFHS study (about 400,000).

Comments

  1. #1 Robert Shone
    February 5, 2009

    David Marker, in the study that Tim Lambert links to, also says the following:

    “A few years ago, 35 leading survey researchers issued a consensus statement on how to minimize interviewer falsification of data (AAPOR 2003). This statement has been endorsed by the American Association for Public Opinion Research and the Survey Research Methods Section of the American Statistical Association. They listed eight factors that could affect falsification rates. Inadequate supervision, poor quality control and off-site isolation of interviewers were three of those factors that are present in this [Lancet] study. The remaining five factors (training on falsification, interviewer motivation, inadequate compensation, piece-rate compensation, and excessive workload) are harder to assess in this situation due to the limited information available on these topics.”

  2. #2 James Hovland
    February 5, 2009

    I want to know if AAPOR has substantiated the claims that 6 million Jews were killed in the Holocaust. Is that accurate? And, were the methods use to deduce that number scientific? Is it even OK to question those figures? Is it true that in Germany it is illegal to question those numbers? And, if so, does that qualify as having explicitly refused to provide that information for review? Also. Is AAPOR as aggressive and critical in their research of public opinion when it is pro-western? Or is that allowed to slide because it fits their public opinion agenda?

  3. #3 bi -- IJI
    February 5, 2009

    > We do not provide public reports of the investigations

    So AAPOR thinks it has the right to demand Burnham to disclose his data and methods, where Burnham is not a member of AAPOR, and where the right to demand information is based on a secret investigation initiated by a secret complaint.

  4. #4 per
    February 5, 2009

    so far, we have:
    40 interviews per day per interviewing team for months on end, with full informed consent; at 20 mins per interview, do the math !

    >98% returns; described as “extraordinary” by many professionals

    they won’t release the original data, survey forms, etc

    a widespread array of academics severely criticising the methodology

    An independent professional body censures the researchers for failing to release data and methods

    But despite this rather surprising concatenation of interesting events, Tim stays behind the study all the way.

    the longer you keep this up, the more embarrassed you are going to be.

    yours
    per

  5. #5 Robert Shone
    February 5, 2009

    Let’s not forget one of the important claims made by the Lancet authors:

    “The sites were selected entirely at random, so all households had an equal chance of being included.” (Burnham et al, 2006)

    Remarkably, researchers are still waiting for an account of the procedures used in the attempt to accomplish this result. It goes to the heart of the study. Perhaps this is what Losch had in mind when talking about the “protocols regarding household selection” which Burnham has apparently failed to provide?

  6. #6 bi -- IJI
    February 5, 2009

    AAPOR:

    > We do not provide public reports of the investigations

    Hmm, let’s repeat that:

    > We do not provide public reports of the investigations

    and again:

    > We do not provide public reports of the investigations

    and yet again:

    > We do not provide public reports of the investigations

    Hahahahah……..

  7. #7 per
    February 5, 2009

    AAPOR’s richard Kulka

    When researchers draw important conclusions and make public statements and arguments based on survey research data, then subsequently refuse to answer even basic questions about how their research was conducted, this violates the fundamental standards of science….

    this is a fundamental standard of science. Perhaps bi and Tim could address this fundamental standard, or explain their own alternative standard. Tim will no doubt be at pains to point out that the authors have not even released their questionnaire, or their informed consent forms…

  8. #8 wildlifer
    February 5, 2009

    How do they determine the difference between “civilian” and non-uniformed combatant? How do they screen for disingenuous answers? Like claims the deceased were civilians when in fact they were not?
    I don’t think any survey can be trusted without complete background checks on each of the deceased. IOWs it’s unknowable.

  9. #9 Tim Lambert
    February 5, 2009

    Sorry per, by “the authors” above are you referring to the authors of the Lancet study or of the AAPOR press release?

  10. #10 bi -- IJI
    February 5, 2009

    per:

    > “When researchers draw important conclusions and make public statements and arguments based on survey research data, then subsequently refuse to answer even basic questions about how their research was conducted, this violates the fundamental standards of science….”

    Sorry, but these are nothing but insinuations with no evidence. Or, to put it another way, with as much evidence as Saddam Hussein’s WMD.

    * * *

    Who the heck are the AAPOR anyway, and why are they on the Internets tubes? Hmm… a whois reveals their snail-mail address to be shared with goamp.com. The goamp.com home page reads thus:

    > Applied Measurement Professionals, Inc. (AMP) provides certification organizations, government agencies, professional associations and private industry with innovative assessment and management solutions. With more than 25 years of experience and over 100 clients representing a wide range of professional occupations, AMP is dedicated to providing our clients with “Technology That Works and People Who Care.”

    > AMP is a private stock corporation located in the Kansas City metro area. AMP was incorporated in the state of Kansas in 1982 as a wholly owned subsidiary of the National Board for Respiratory Care (NBRC). In 1996, AMP acquired Logic Extension Resources (LXR), which now serves as the company’s LXR Division.

    Hmm… hmm… hmm…

  11. #11 bi -- IJI
    February 5, 2009

    To confuse matters further, they also list another snail-mail address which is shared with nbrc.org.

    Hmm… hmm… hmm… hmm…

  12. #12 Jeff Harvey
    February 5, 2009

    Wildlifer, If you think this distinction is applicable, shouldn’t it be up to the aggressors to count the number and status (civilian or combatant) of the dead as a result of their invasion? Why don’t you think the US or UK showed any interest in carrying out a detailed survey of the death toll in Iraq as a result of their combined aggression? Or, for that matter, of any wars waged? And is such a distinction between ‘civilian’ and ‘combatant’ deaths valid anyway? I am sure that many ordinary Iraqi civilians became combatants as a result of the fact that they don’t necessarily like being occupied by a foreign aggressor, particularly one that has murdered their loved ones. Given that the invasion violated just about every standard of international law (as well as the United States constitution), it was illegal anyway and thus so-called combatant deaths are also a crime.

    But back to my original point: it should be obvious why the US and UK did not carry out a census of the death toll as a result of the invasion. First, they didn’t care. Second, it the actual total was accurately known, at least to within a few thousand, it would fuirther demolish the notion that the US and UK are great respectors of freedom and human rights. But as long as the actual total is very vague, then any estimate generated from any source is open to doubt. This enables the western corporate media to downplay the idea that the invasion resulted in utter carnage, and maintains the myth of the ‘basic benevolence’ of the west.

    The major aim here of defenders of unilateral US aggression is to sow doubt as to the actual death toll in Iraq. As long as there is doubt then it is possible, in public realtions terms, to ‘manage the outrage’. Actually, there are similarities with respect to the campaign waged by certain sectors and individuals to donwplay the theory of anthropogenic climate change. The sceptics know that they will never win the scientific argument, but they don’t have to: so long as they can sow enough doubt amongst policy makers and the public, then nothing will change. If the true death toll as a result of the war in Iraq was known, then I am sure that there would be repercussions. So the governments, along with much of the corporate mainstream media, do everything they can to obfuscate the truth. They routinely attack scholars whose work suggests that the death toll was enormous, because, as I said above, it demolishes much of the myth that our governments pursue humanitarian agendas. They have agendas all right, but they are vastly different.

  13. #13 hardindr
    February 5, 2009

    per,

    Please note from Mark Blumenthal’s blog post on AAPOR’s censure of Bernham:

    The AAPOR censure does not involve Burnham’s methodology and renders no opinion on the substantive conclusions of the Lancet study. Instead, it focuses entirely on disclosure, or rather on Burham’s failure to disclose “essential facts about his research.”

  14. #14 bi -- IJI
    February 5, 2009

    > An independent professional body

    Given what I found above, I think I can now ask the good old question again:

    Independent… of whom?

  15. #15 David Kane
    February 5, 2009

    Tim writes “I don’t that Losch or the AAPOR have disclosed, specifically, what is missing from the information that Burnham has provided.”

    1) I think you want “don’t know that” instead of “don’t that”.

    2) I can help with this! AAPOR want Burnham to disclose the actual survey used. He has not done so. (The National Journal published some sort of survey template but a) It was never clear whether this applied to the 2004 or 2006 survey and b) No Lancet author has ever confirmed that this was the survey used in 2006. Just because The National Journal thinks that this is the survey does not mean that it actually was.

    Although I (and other researchers) would able to see some of the household-level data, I (and, I think, all other researchers) were never allowed to view the actual survey (either in English or in Arabic).

    AAPOR standards require that survey questions be made public. Burnham et al have not done so.

  16. #16 bi -- IJI
    February 5, 2009

    The AAPOR, the AMP, and the NBRC also share snail-mail addresses with another group, the National Association of EMS Physicians.

    This “independent professional body” sure has a ton of ‘dependencies’, eh…

  17. #17 Robert Shone
    February 5, 2009

    Bi, with respect, this continual paranoid rambling is making this “science” blog read more like the Medialens message board.

  18. #18 stewart
    February 5, 2009

    In other words, the AAPOR has no standing in this, and never did. However, the Johns Hopkins report is going to be the strongest possible source, as they have authority and expertise in this situation.
    I look forward to reading the report when it comes out, almost as much as the reactions to it.

  19. #19 bi -- IJI
    February 5, 2009

    Robert Shone:

    > Bi, with respect, this continual paranoid rambling is making this “science” blog read more like the Medialens message board.

    Just the facts, sir. It’s always interesting when an “independent” non-profit organization doesn’t disclose that it’s run from the same address as a for-profit company and a ton of other “independent” organizations. I’m sure one can come up with a perfectly good explanation for this, but the interesting thing is that nobody’s even bothered to give one.

  20. #20 bi -- IJI
    February 5, 2009

    Something nice and chunky from the AMP, the company that’s situated at the same snail-mail address as the AAPOR.

    Coincidence? I report, you connect the dots.

  21. #21 sod
    February 5, 2009

    AAPOR standards require that survey questions be made public.

    is it typical for english papers to publish arabic questions?
    what sense would it make?

    AAPOR want Burnham to disclose the actual survey used. He has not done so. (The National Journal published some sort of survey template but a) It was never clear whether this applied to the 2004 or 2006 survey and b) No Lancet author has ever confirmed that this was the survey used in 2006. Just because The National Journal thinks that this is the survey does not mean that it actually was.

    is it typical to question whether questions published were really used?

    and why do you guys question the questionaire? do you have ANY reason for this?

  22. #22 Bud
    February 5, 2009

    Bi, that link returns nothing for me.

  23. #23 per
    February 5, 2009

    the Johns Hopkins report is going to be the strongest…

    this is very simple. Scientists release data, subject to practicalities. It doesn’t take Johns Hopkins’ imprimatur to validate this very simple principle.

    What possible reason can there be for refusing to release the questionnaire and informed consent forms ? It takes two minutes to scan to pdf and put on the web- so what possible reason is there not to ?

    Anonymising data and putting it out for reanalysis is something that happens on a regular basis for lots of epidemiology studies. So why not here ?

    There are already several reasons for wanting to examine this data. 98% returns- which Tim quotes as “extraordinarily high response rates “. Undertaking 40 questionnaires per day, with full informed consent, in the fast time of 15 minutes each, gives a ten-hour day, and no time for lunch, toilet breaks, or moving from house-to-house. Examining the data might enable a greater certainty that this data is credible. And then again, it might make it quite understandable why they don’t want to release the data.

    per

  24. #24 sod
    February 5, 2009

    Anonymising data and putting it out for reanalysis is something that happens on a regular basis for lots of epidemiology studies. So why not here ?

    NO! typically a poll will state what sort of data will be published. you wont give away more data, because you are NOT allowed to do so!

    is this your level of “expertise” on this subject?

    data from “anonymised” questionairs or even cluster data is highly problematic, as it might be pretty simple to figure backwards!

    this is very simple. Scientists release data, subject to practicalities. It doesn’t take Johns Hopkins’ imprimatur to validate this very simple principle.

    this is very simple. scientists share data, if they want to share it. being friendly and cooperative yourself helps. being hostile mostly doesn t.

    What possible reason can there be for refusing to release the questionnaire and informed consent forms ? It takes two minutes to scan to pdf and put on the web- so what possible reason is there not to ?

    it takes about two minutes, to write down WHY OH WHY you want to have the questionaire? what part of it do you have doubts about?
    how good is your arabic anyway?

    There are already several reasons for wanting to examine this data. 98% returns- which Tim quotes as “extraordinarily high response rates “. Undertaking 40 questionnaires per day, with full informed consent, in the fast time of 15 minutes each, gives a ten-hour day, and no time for lunch, toilet breaks, or moving from house-to-house. Examining the data might enable a greater certainty that this data is credible. And then again, it might make it quite understandable why they don’t want to release the data.

    the response rate is not that unusual. if you had ever taken a look at the questionaire, you wouldn t think that it takes 15 minutes to fill the form.
    releasing more data had absolutely ZERO effect on the people attacking its credibility.
    http://tinyurl.com/bhsr8p

  25. #25 Sortition
    February 5, 2009

    > so far, we have: 40 interviews per day per interviewing team for months on end, with full informed consent; at 20 mins per interview, do the math !

    With two teams, 47 clusters, and one cluster per team per day, the data collection would be over in less than 4 weeks.

  26. #26 Megan McArdle
    February 5, 2009

    Well, if I didn’t, I was in good company, because several prominent conflict epidemiologists flatly told me that this was not a valid interpretation of the results, including one of the few researchers who had access to both datasets, Olivier Degomme. They weren’t trying to get Les Roberts; they simply thought that I must have misunderstood. I worked with Roberts’ remarks in front of me, and very clearly asked whether the excess deaths being close to the lower bound of the violent deaths meant that they substantially agreed. They said no, that wasn’t the way it worked. As, of course, did the authors of the WHO paper.

  27. #27 elspi
    February 5, 2009

    Cliff notes version of this:

    AMP is a propaganda machine and AAPOR somehow shares an address with them.

    And we have these “I am become death; destroyer of irony meters.” quotes from AAPOR:

    “Dr. Burnham provided only partial information and explicitly refused to provide complete information about the basic elements of his research”

    “We do not provide public reports of the investigations”

    How can the credibility of the Lancet denialists get any lower?

    Any day now, I expect to see them on TV flogging “natural male enhancement”.

  28. #28 sod
    February 5, 2009

    total deaths between the lancet and the WHO are in good agreement, as Tim wrote above.

    if you disagree with that statement, you disagree with reality.

  29. #29 Information
    February 5, 2009

    McArdle does not seem to have understood what Roberts was saying was in agreement: the excess deaths in the Lancet study (about 650,000) and in the IFHS study (about 400,000).

    I’m struggling to understand this claim, so please help me out.

    Per the UN/IFHS study found at (http://www.who.int/mediacentre/news/releases/2008/pr02/en/index.html), 151,000 Iraqis were estimated to have died between 2003 and 2006. That isn’t the 400,000 mentioned. Further, in the UN’s FAQ section concerning their study the specifically address the differences between the Lancet study and the UN/IFHS study:

    Q: Which figure is more reliable: the new survey or the 2006 household survey?
    A: The 2006 household survey shows a very different trend than that seen in the new survey and the Iraq Body Count, with increasing numbers of deaths per day, rising from 231 during 2003-2004 to 491 during 2004-2005 and 925 during 2005-2006. The biggest difference between the 2006 household survey and the other two sources is in the figure for the third year. Most of those deaths occur in six high mortality governorates outside of Baghdad, while in the Iraq Body Count and the new survey, most deaths occur in Baghdad.
    The difference between 925 and 126 violent deaths per day is very large. To reach 925, the Iraq Family Health Survey would have to have missed more than 80% of deaths detected in the smaller survey. This is highly unlikely given the much larger number of clusters and households visited in the new survey.

  30. #30 sod
    February 5, 2009

    151,000 Iraqis were estimated to have died between 2003 and 2006. That isn’t the 400,000 mentioned.

    you are confusing violent and total excess death cases.

    the questions over household deaths are on page 16 of the questionaire. (do you spot the trouble with this?)

    http://tinyurl.com/dgloll

  31. #31 QrazyQat
    February 5, 2009

    My god, bi–IJI, there’s also the Association of Polysomnographic Technologists, Citizen CPR Foundation, and the Oak Park Homes Association! You’ve uncovered a major conspiracy all right; all over our USA we find these “businesses” and “organizations” clustered together in “buildings”. Somne call them “office buildings”, but we know better.

    Watch the skies!

  32. #32 Tim Lambert
    February 5, 2009

    Megan:
    >I worked with Roberts’ remarks in front of me, and very clearly asked whether the excess deaths being close to the lower bound of the violent deaths meant that they substantially agreed.

    Which is not what Roberts said. The surveys disagreed about the number of violent deaths, but the number of excess deaths from the IFHS was inside the confidence interval for excess deaths from Lancet 2.

  33. #33 stewart
    February 5, 2009

    per and others:
    Scientists share data at times, subject to constraints and other issues. I have access to some data under a written agreement that states what I am going to do with it, that I will share the results of any findings, and not release to other parties. I am involved in several research projects that have involved trying to contact various researchers for information beyond what has been published. Sometimes we get it, often we don’t. I wish we always did, but tough – I can’t compel anyone. So tough – discriminate between science and propagand.

  34. #34 per
    February 5, 2009

    NO! typically a poll will state what sort of data will be published. you wont give away more data, because you are NOT allowed to do so!

    in fact, you do not substantiate that statement at all. You do not (cannot ?) point to any undertaking that the authors gave to the participants, so you cannot show that the authors are bound.

    Yet another disappearing bit of evidence. Strangely, many other epidemiology groups manage to organise independent review and assessment of their confidential data. What is so “magic” about this study, that they cannot ?

    the response rate is not that unusual.

    I merely guess that you are not a professional epidemiologist or pollster ? The issue is that many professionals seem to think that it is extraordinary (see Tim’s quote above) and e.g. this post from 2006

  35. #35 David Kane
    February 5, 2009

    Tim mentions talks about the “number of excess deaths from the IFHS” Of course, IFHS did not publish an excess deaths number. This is just the sort of stuff that Roberts is guilty of.

    Now, it is possible for you to make a bunch of assumptions that the IFHS authors did not make and derive an excess death number from the raw IFHS data. That is a fun exercise. But that has little to do with whether the IFHS paper — that actual words that they published in the NEJM — agrees or disagrees with L2.

    IFHS published an estimate and confidence interval for violent deaths. L2 published an estimate and confidence interval for violent deaths. Those two confidence intervals do not overlap. Indeed, the L2 estimate is four times larger. Someone is wrong.

  36. #36 slickdpdx
    February 5, 2009

    I would understand the secrecy (to some degree) if we were talking about research that might lead to a profitable patent or was subject to some kind of national security consideration of privacy concerns. Is there anyone who would care to explain the reason for secrecy in the Lancet/Iraq study (for the data that has been requested by its peers and other critics?)

  37. #37 Zarquon
    February 5, 2009

    If the survey authors release the raw data it may lead to the identification of individuals who participated in the survey. No one needs to know this. If someone does not like the survey’s methodology (which has been published) they are free to commission their own.

  38. #38 bi -- IJI
    February 5, 2009

    QrazyQat:

    > My god, bi–IJI, there’s also the Association of Polysomnographic Technologists, Citizen CPR Foundation, and the Oak Park Homes Association! You’ve uncovered a major conspiracy all right; all over our USA we find these “businesses” and “organizations” clustered together in “buildings”. Somne call them “office buildings”, but we know better.

    > Watch the skies!

    Well, given that the AAPOR and AMP,

    1. in addition to being clustered together in one building (in Lenexa, KS), also
    2. share the address of another building (in Olathe, KS), and
    3. the administrative contact for the AAPOR site also goes to an AMP e-mail,

    I think the relationship between AAPOR and AMP goes beyond “independence”.

    Extraordinary proof requires extraordinary evidence, and it does seem something extraordinary is going on here (even if I don’t quite know what it is).

  39. #39 bi -- IJI
    February 5, 2009

    Bud:

    > Bi, that link returns nothing for me.

    Sorry… it seems the archival took a while. Do try again.

  40. #40 dhogaza
    February 5, 2009

    Ah, AMP is a lobbying firm.

  41. #41 Tim Lambert
    February 5, 2009

    [Robin Meija](http://www.motherjones.com/commentary/columns/2008/11/witness-after-math.html):

    >More than two years later, the Iraq study remains mired in controversy. But other recent findings suggest that Roberts and Burnham were on the right track. In the summer of 2006, the World Health Organization conducted a large family health survey along with Iraq’s Ministry of Health, interviewing about five times as many people as Roberts and Burnham had, and in a more distributed fashion. In August, Mohamed Ali, a who statistician, reported his preliminary results to colleagues at a Denver statistics conference: Nearly 397,000 Iraqis had died because of the war as of July 2006.

    >That number falls at the low end of Roberts and Burnham’s confidence interval, which ranges from roughly 393,000 to 943,000. But while epidemiologists and statisticians are still pondering questions raised by differences between the two surveys, there’s no longer much doubt among them that Iraq’s civilian casualties number in the hundreds of thousands.

  42. #42 David Kane
    February 5, 2009

    I stand corrected! And, ironically enough, I actually organized the panel at which Dr. Ali presented that result. (I don’t recall the exact number but I have no reason to dispute Robin’s reporting.)

    We will see what happens when IFHS publishes their estimate. The main feedback that I saw Dr. Ali get about those numbers concerned the underestimation (because of recall bias) of the pre-war mortality. But, of course, Ali and his team are experts, so I am sure that whatever estimate they come up with will be reasonable.

    But, anyway, none of that makes the L2 and IFHS numbers more comparable because the central (almost only) critique of L2 has always revolved around their violent death total. No one disputes that their non-violent excess death estimate is plausible.

    And, you can’t just try Roberts trick of claiming that the violent deaths in IFHS were mischaracterized as non-violent in IFHS because the gender balance of the deaths is radically different.

    And we still have the issue of the different sorts of adjustments made by IFHS and not made by L2.

    But, Tim is right that, given Dr. Ali’s presentation, it is fair to say that IFHS presents a draft estimate of excess deaths which is within (barely!) the 95% confidence interval of the L2 estimate for total excess deaths.

  43. #43 Gaz
    February 6, 2009

    This may be off topic but has anyone worked out how many Iraqi soldiers were killed in the Iraq wars, or is it a matter of male+in uniform+Iraqi = don’t matter?

    Even if it worked out to one soldier per ton of bombs it would surely add up to well over 100,000.

    That’s a lot of heartbroken mothers.

  44. #44 Tim Lambert
    February 6, 2009

    David, the IFHS and Lancet surveys had different methodologies. It is absurd to argue that adjustments made to the IFHS survey should be made to Lancet numbers. Especially when you find the resulting number implausible.

  45. #45 Bruno
    February 6, 2009

    [david kane] “none of that makes the L2 and IFHS numbers more comparable because the central (almost only) critique of L2 has always revolved around their violent death total. No one disputes that their non-violent excess death estimate is plausible.”

    May I submit that the discrepancy in ‘violent deaths’ may lie with the identity of the pollsters and the interviewed Iraqi’s reaction to that? Given that Iraq was in a grip of a vicious sectarian war, it could well be that the respondents modified their answers (possibly for all three studies) in order to please whoever they thought the pollsters were.

    For example, the IFHS survey was carried out partly by the Iraqi Health Ministry, which was run by Sadrists at the time – who were also heavily involved in the sectarian fighting. If a Sunni family were interviewed, and their son, for example, had died violently, they might want to hide that fact lest they be suspected of being involved in the sectarian fighting and set themselves up as targets. (It is notable that there were areas deemed “too dangerous” for these pollsters to visit in Baghdad, and very likely these were hardcore Sunni areas where the pollsters as Shiites risked being killed themselves.) One supposes that the opposite might have occurred in the case of Burnham et al’s study, too.

    As for the study authors not disclosing more detailed information on who specifically was interviewed, it is understandable for the same reasons: those polled would run the risk of becoming sectarian targets. Imagine how difficult it would be for further pollsters in Iraq if it became known that the details of previous interviewees were released. At the moment, for example, Iraqi translators are living in terror that their details will be released to the Iraqi authorities by the US military, for this very same reason.

    I think it is hard to imagine the kind of fear that pervades these communities in Iraq, and this will certainly have some sort of effect on surveys.

  46. #46 Mike H
    February 6, 2009

    “Which is not what Roberts said. The surveys disagreed about the number of violent deaths, but the number of excess deaths from the IFHS was inside the confidence interval for excess deaths from Lancet 2.”

    If I may, they disagreed “violently” about the number of violent deaths, Tim. Let’s be frank here. Similarly, L2 and L1 are highly contradictory in terms of violent death. This notion that the IFHS and the Lancet studies are somehow corroborative, so long as their respective violent death components are ignored, is ridiculous.

    These studies were undertaken because there was an armed conflict going on in Iraq. It is widely understood that people die in armed conflicts either directly from violence (bullets and bombs) or indirectly from the collateral effects of the violence on society. Any survey purporting to accurately detect the excess death toll from a conflict must be able to accurately detect the excess death toll of the two subgroups. Obviously, some latitude should be granted for accuracy beyond these two largest, most basic subgroups of death, but there are absolutely no grounds for giving a death study a pass on accuracy for the two main subgroups. The differences between the mechanism of death in each major subgroup (violent and non-violent) are too marked to allow any greying of the two whatsoever. Studies that are in sharp disagreement over the disposition of deaths from natural causes and accidents and those caused by violence are highly contradictory of one another, and call into serious question the accuracy of one or both of the studies.

    To put this another way, studies cannot be said to corroborate one another if said corroboration requires the swapping in and out between studies of tens of thousands of deaths, even hundreds of thousands of deaths, between violent and non-violent causes. No one knows the precise number of excess deaths that have occurred in Iraq since March 2003, but there is a precise number. Every one of the victims had an identity, and every one died from a specific cause. Hundreds or thousands of deaths extrapolated from a car bomb in Karbala from one study cannot be balanced off in some form of corroborative equation against hundreds or thousands of extrapolated excess heart attack deaths in Nineveh in another study. In effect, that’s what you’re doing, Tim. The same person cannot die twice, from two different causes, and in two different parts of the country.

  47. #47 Jeff Harvey
    February 6, 2009

    Mike H, You shoot yourself in the foot when you casually remark, “These studies were undertaken because there was an armed conflict going on in Iraq”.

    Where to begin dismantling this simplistic remark? First of all, there wasn’t merely an ‘armed conflict’ going on; the truth is that the ‘conflict’ was hugely disproportionate in terms of technology. The second point is that the ‘conflict’ should be termed ‘aggression’, because that is what the ‘conflict’ was: aggression on the part of the US and its proxies. I am amazed how those who somehow defend the invasion always resort to Orwellian termsd like ‘operations’ and ‘conflicts’. The ‘conflict’ was an act of aggression that violated international law. The result. whatever end of the scale it is measured by, was carnage. Why not call it what it was, and stop decorating it up?

    Second, as I said in an earlier post, of course there is a precise number of deaths caused by the war, either directly or indirectly, but why didn’t the invaders show any interest in carrying out their own detailed survey? As I said before (a) they didn’t really care about the death toll, because they were pursuing an altogether different political agenda, and (b) a detailed number would cast a pretty horrible light on the outcome of the aggression. As long as the figure remains in ambiguity, any figure, particlarly those that project almost genocidal totals, can be dismissed as heresay.

    How many people perished in Iraq as a result of the sanctions of mass destruction between 1991 and 2003 that resembled a medievel siege? Two senior UN officals called the sanctions ‘genocide masquerading as policy’ but the corporate media ignored them almost completely. One of them, Hans von Sponeck, recently wrote and published a book on the effects of the santions on Iraqi civilians, and it was ignored, as to be expected. Again, the body count under the sanctions is exceedingly unclear (one million? Half a million? 250,000?), hence it can also be summarily dismissed by the west.

    Finally, I think the studies were undertaken not because there was a ‘conflict’ (ugh, that word again) going on in Iraq, but because the invasion was highly controversial, illegal under international law and the US contitution, and because a number of critics had argued beforehand that there could be an enormous humanitarian cost. Whatever the critics of the OMB and Lancet surverys say, the critics were correct on all counts. Iraq society was utterly devastated by the war and the death toll was horrific. Moreover, all of the arguments used by Bush, Blair and co. to justify the invasion were found to based on a series of epic lies. What is there left to defend?

  48. #48 bi -- IJI
    February 6, 2009

    Shorter Mark H: We don’t know, therefore we know.

  49. #49 Bruno
    February 6, 2009

    [mike h] “To put this another way, studies cannot be said to corroborate one another if said corroboration requires the swapping in and out between studies of tens of thousands of deaths, even hundreds of thousands of deaths, between violent and non-violent causes.”

    Did the IFHS (NEJM) study ask for death certificates to corroborate its data? The Gilbert et al / Lancet study did.

  50. #50 Information
    February 6, 2009

    you are confusing violent and total excess death cases.

    Then the author is comparing apples to oranges.

    The Lancet study attempted to determine excess violent deaths per this release:

    We estimate that as of July, 2006, there have been 654,965 (392,979–942,636) excess Iraqi deaths as a consequence of the war, which corresponds to 2·5% of the
    population in the study area. Of post-invasion deaths, 601,027 (426,369–793,663) were due to violence, the most common cause being gunfire.

    To compare the two studies and claim that the results are similar is simple dishonesty.

  51. #51 bi -- IJI
    February 6, 2009

    Shorter (Mis)Information:

    I’ll now purposely ignore Burnham et al.’s figure for total eaths.

    I’ll then make a bogus comparison between IFHS’s figure for total deaths and Burnham et al.’s figure for violent deaths.

    Then I’ll point out that my bogus comparison is, well, bogus.

    Profit!

  52. #52 sod
    February 6, 2009

    nice catch bi :)

    to quote Tim s original post above: (last paragraph)

    McArdle does not seem to have understood what Roberts was saying was in agreement: the excess deaths in the Lancet study (about 650,000) and in the IFHS study (about 400,000).

    The differences between the mechanism of death in each major subgroup (violent and non-violent) are too marked to allow any greying of the two whatsoever. Studies that are in sharp disagreement over the disposition of deaths from natural causes and accidents and those caused by violence are highly contradictory of one another, and call into serious question the accuracy of one or both of the studies.

    this is FALSE and RIGHT at the same time.

    basically comparing subsets (like violent dead) is problematic, because you lose NUMBERS. the best comparison is between the biggest numbers you got.
    imagine two groups are polling people for a rare event (accidential death). their result for totals is pretty close. but their results for a subset (people killed by lightning) is wide apart.

    on the other hand, there IS an easy explanation for the difference:
    one study did a very short interview about ONLY death cases. the other is a twenty pages interview, with death on page 16. one team of interviewers were independent doctors. the other was done by members of the iraqi health ministry, a group RESPONSIBLE for a significant amount of the death cases, they were polling…

    just by chance, one of the two groups confirmed their death numbers by looking at death certificates….

  53. #53 David Kane
    February 6, 2009

    sod: You think that the Lancet interviewers were “independent doctors?” That’s not true. They worked for an Iraqi hospital run by the Iraqi government. That, obviously, does not make them bad people, but their pay checks were written by the Iraqi state, just like those for the IFHS interviewers.

  54. #54 sod
    February 6, 2009

    David, i see a difference. don t you?

    HOUSEHOLD QUESTIONNAIRE
    HOUSEHOLD INFORMATION PANEL
    We are from the Ministry of Health and the Central Organization of Statistics & Information Technology. We are conducting a
    survey on the health of families, women and children. I want to talk with you about this subject because we believe that family
    health is the base of community health. This survey will take some of your time.

    this is how they introduced themselves. might sound threatening to a Sunni, actually….

  55. #55 David Kane
    February 6, 2009

    Excellent point. How did the Lancet interviewers introduce themselves?

    ….

    Oh, that’s right! You don’t know. The Lancet team has never released the script that the interviewers were supposed to follow. This is example #334 of how the IFHS conducts itself and in professional manner while the Lancet authors have not.

    Obviously, the fact that the IFHS interviewers introduced themselves in such-and-such a way might create problems. But it is impossible to evaluate the extent the of possible bias in comparison with the Lancet papers unless we get some transparency about what the Lancet interviewers said.

  56. #56 Jody Aberdein
    February 6, 2009

    Just to get this clear in my mind: how exactly would you be able to work out the extent of bias if you knew the Lancet team’s manner of introduction?

  57. #57 QrazyQat
    February 6, 2009

    bi — IJI, you seem to have dropped your claim of their nefarious connection to the National Association of EMS Physicians. So when you said:

    I’m sure one can come up with a perfectly good explanation for this, but the interesting thing is that nobody’s even bothered to give one.

    I pointed out that a perfectly sane explanation is that businesses and associations often group together in office buildings. Sane people think of this possibility instead of jumping immediately to paranoid claims of conspiracy due to the sharing of addresses, and note that to get to the bottom of this oh so scary conspiracy, one might think of going to the site of AAPOR, AMP, or NBRC, where they apparently forgot they’re supposed to be some clandestine operation and point out in numerous places that Applied Measurement Professionals is AAPOR’s management firm, and that both are a wholly-owned subsidiaries of the NBRC. My god, an organization with differtent holdings! That’s scary!

    Even more scary connections are revealed when you look at their site (don’t they realise they should be hiding this stuff or their conspiracy will fail?). Why they’re sponsored by the American Association for Respiratory Care, the American College of Chest Physicians, the American Society of Anesthesiologists, and the American Thoracic Society. Scary stuff kids!

    Robert Shone is right when he points out that your posts on this, uncovering through diligent whois detective work what could be found by going to the site of the organization and reading what they say about themselves, makes you look like a loon; given those posts and their tone I’m not sure I agree with what I read as his connotation that this would be an inaccurate characterization.

  58. #58 Mike H
    February 6, 2009

    Where to begin dismantling this simplistic remark? First of all, there wasn’t merely an ‘armed conflict’ going on; the truth is that the ‘conflict’ was hugely disproportionate in terms of technology. The second point is that the ‘conflict’ should be termed ‘aggression’

    Jeff, I’m not interested in engaging in a semantics exercise in order to satisfy your desire for a polemical exchange. Further, I’m not interested in having yet another long winded discussion over the case for war, let alone the sanctions period of 1991 – 2003. If you have something specific to my comment about the statistical comparisons of the various studies, then I’ll be happy to respond in detail.

  59. #59 Barton Paul Levenson
    February 6, 2009

    Gaz writes:

    This may be off topic but has anyone worked out how many Iraqi soldiers were killed in the Iraq wars, or is it a matter of male+in uniform+Iraqi = don’t matter?

    Even if it worked out to one soldier per ton of bombs it would surely add up to well over 100,000.

    That’s a lot of heartbroken mothers.

    I assume you’re not saying here that this many people were killed because of American bombing? Because there hasn’t been a lot of American bombing since the very first phase of the war. Most of the bomb fatalities have been from suicide bombers and IEDs.

    To clarify my position — I have no doubt that hundreds of thousands of Iraqis have died due to the war, and the figure may well be 600,000. I haven’t examined the details and can’t give an informed opinion. I just want to point out that a high casualty total does not ipso facto mean we should not have invaded, though it might well mean we should have done a better job of it. A lot of people died in World War II, but that was still a war better fought than unfought. And again, I’m not endorsing the Bush administration or their motives, though I’m sure somebody here will accuse me of being a neocon.

  60. #60 Information
    February 6, 2009

    Shorter (Mis)Information:

    How clever! An insult against my chosen screen name. I’m quite certain something even more devestating is percolating inside that head of yours.

    I’ll then make a bogus comparison between IFHS’s figure for total deaths and Burnham et al.’s figure for violent deaths.

    Wait a second here. The IFHS specifically warns against using their study for any extrapolation other than excess violent deaths. Per the study:

    The non-violent mortality rate increased by about 60%, from 3.07 deaths per 1000 people per year before the invasion to 4.92 deaths per 1000 people per year in the post-invasion period. This was not further addressed in this analysis, which focused on mortality due to violent deaths. Further analysis would be needed to calculate an estimate of the number of such deaths and to assess how large the mortality increase due to non-violent causes is, after taking into account that reporting of deaths longer ago is less complete.

    Are you seriously going to accuse me of dishonesty when I simply mistook the poster’s reliance on a fictional figure for a figure that the study’s authors actually expressed confidence in? Such sarcasm is usually indicative of an individual who is usually immune to challenges to their authority. You’re either extremely arrogant or incredibly insecure. Maybe both.

    Futher, using the numbers utilized by the UN/IFHS study, I’m still a little confused as to how you arrived at the 400,000 number. Per their figures (3.07/1000 pre-2003, 4.92 post-2003, 28m population over a period of 38 months), the difference is ~160,000. Added in to their estimate of ~150,000 violent deaths leaves us with ~310,000 excess deaths. I’m being sincere when I ask for guidance on where my calculations are off and how the figure of ~400,000 is determined.

  61. #61 Barton Paul Levenson
    February 6, 2009

    Jeff Harvey, predictably, writes:

    that is what the ‘conflict’ was: aggression on the part of the US and its proxies.

    Al Qaeda wasn’t being aggressive? You do that Al Qaeda in Iraq A) was responsible for a huge number of deaths, and B) was largely made up of non-Iraqis who came in to fight the US? For that matter, were the Sunnis being aggressive when they massacred Shi’ites, or vice versa? How about the Kurds? Was Iran being aggressive when it shipped arms to people fighting the Americans? Or sent in troops, as some people allege it has done?

    Portraying this as a simple war of Iraqi good guys against American/UK bad guys is not just wrong, it’s grossly ignorant.

  62. #62 Barton Paul Levenson
    February 6, 2009

    Jeff Harvey, still channeling Noam Chomsky, writes:

    How many people perished in Iraq as a result of the sanctions of mass destruction between 1991 and 2003 that resembled a medievel siege?

    Do you think it would help if Saddam had used the oil-for-food money and spent it on food instead of arms and building programs for the military?

    The money was there. It could have been used to feed people. But Saddam was kind of like Kim Jong Il — he always took care of the soldiers first, and didn’t really care what happened to anybody else. But I forget — in Jeff’s world, only the west can be a bad guy.

  63. #63 dhogaza
    February 6, 2009

    Al Qaeda wasn’t being aggressive?

    Not in Iraq.

    A) was responsible for a huge number of deaths, and
    B) was largely made up of non-Iraqis who came in to fight the US?

    The fact that Al Q made an appearance in Iraq post-invasion supports a claim that we invaded due to Al Q aggression pre-invasion?

    Interesting.

    Portraying this as a simple war of Iraqi good guys against American/UK bad guys is not just wrong, it’s grossly ignorant.

    Jeff did no such thing. I thought Christians were supposed to be honest.

  64. #64 Mike H
    February 6, 2009

    Did the IFHS (NEJM) study ask for death certificates to corroborate its data? The Gilbert et al / Lancet study did.

    Bruno, regardless, we still have stated figures for excess violent death from these studies. I haven’t seen any compelling evidence that the IFHS violent toll is an undercount because respondents could have been intimidated into characterizing violent deaths as non-violent

    basically comparing subsets (like violent dead) is problematic, because you lose NUMBERS. the best comparison is between the biggest numbers you got. imagine two groups are polling people for a rare event (accidential death). their result for totals is pretty close. but their results for a subset (people killed by lightning) is wide apart.

    The problem with your analogy, Sod, is that your subset (lightning) is exceedingly rare. The major subsets of death, violent and non-violent, are not. Every death fits into one or the other, without exception. Unlike death subsets further down the chain, there should be few errors recorded when determing whether a reported death is violent or non-violent, and any such errors would likely be intentional on the part of the survey respondent. I don’t believe intentional misleading was a significant practice by those surveyed.

  65. #65 Mike H
    February 6, 2009

    Shorter Mark H: We don’t know, therefore we know.

    Bi–IJI, Implicit in your sarcasm is the assertion that you and the other defenders of the Lancet studies do “know.“ If L2 and the IFHS corroborate each other, then what have you come up with in terms of an actual excess violent death toll? Did you simply split the difference between the two, or have you employed some other formula for resolving the huge disparity between the two studies?

    The truth is, we really don’t “know” with much accuracy, and the Lancet studies are particularly less than helpful in that regard. I’ll give you an example to illustrate that.

    In December 2004, when the first Lancet study was being debated here, I contacted Les Roberts and asked for a breakdown of the bombing deaths and the excess violent deaths not attributed to the coalition. Included in the information Roberts provided me was the location of every excess violent death the study recorded, except for the 3 shooting deaths attributed to coalition soldiers:

    http://scienceblogs.com/deltoid/2004/12/lancet11.php

    “Yes, all 12 non-coalition violent deaths happened outside of Falluja. (1 Kut, 1 Thiqar, 1 Karbala, 7 Baghdad, 1 Diala, 1 Missan, Note Baghdad is about 3-7 times greater in population than these other Governorates so the rates are not so different)

    Bombing deaths:

    Thiqar
    M5, M2, F22 (one family)
    Thiqar (different village)
    M27

    Missan
    1mo. & 6mo. in same households (often there are multiple sons with wives under the same roof — interviewer did not record the gender of the infant)

    Falluja
    10 girls<12 years, 13 boys<12, M14, 25 adult males, 3 adult women (adult defined as 15--59). “

    As you can see from Roberts’ reply, the violent death toll was heavily weighted to the southern Shiite governorates and Baghdad, with more than half the violent deaths in L1 occurring (Falluja cluster excluded) in the south. It is generally understood from reporting on the ground that the southern Shiite governorates have seen much less violence than Baghdad and the mixed Sunni/Shiite governorates in the north-center of the country (Anbar, Diyala, Ninevah, and Salahaddin). L1 does not reflect that reality at all. If you look at Roberts’ figures, we know the location of death for 18 of the violent deaths recorded in the L1 survey, and only 1 of them occurred in Diyala, Ninevah or Salahaddin. We don’t know where the last 3 violent deaths (from coalition gunfire) occurred, however it seems highly unlikely that all 3 of these deaths happened in these 3 provinces. Even if, for the sake of argument, we allow that all 3 of these shootings occurred in these 3 high violence governorates, we’re still left with an overall extrapolation of violent deaths in these 3 provinces of less than 12,000.

    Then along comes L2, and it provides a very different geographical distribution of violent death from its predecessor. I don’t have access to the data from L2, but I have corresponded with someone who does, and he has given me some of the numbers he extrapolated for violent death from governorate to governorate. For the same 18 month period covered by L1, L2 extrapolated 32,000 - 34,000 violent deaths in Diyala alone. Ninevah, 18,000 violent deaths and 36,000 for Salahaddin, for a total in these three governorates of 86,000 violent deaths. As I pointed out earlier, the violent death toll from L1 for these governorates can’t be any higher than 12,000, and may well be much less, as low as 3,000. My source for the L2 data advised that the extrapolated violent death toll for Diyala alone, to the end of the L2 survey period in July 2006, was in excess of 100,000.

    However, that isn’t anywhere near what the media reported in Oct 2006, when Agence France Presse ran this wire service story about the carnage in Diyala:

    http://findarticles.com/p/articles/mi_kmafp/is_200610/ai_n16933422

    Iraq province loses 9,000 to sectarian killing

    Sat Oct 21, 8:38 AM ET

    “The bloodsoaked Iraqi province of Diyala has seen 9,000 of its citizens killed and at least 31,500 forced to flee since the fall of Saddam Hussein, its police chief has said.

    Defending the scope of an aggressive new security operation, Major General Ghassan Adnan al-Bawi told lawmakers Saturday that his force was dealing with a massive campaign of sectarian cleansing in the killing fields north of Baghdad.

    In the three years and seven months since the US invasion, Sunni and Shiite death squads in Diyala have been battling it out for one of Iraq’s most fertile and religiously mixed areas.”

    The link provided does not allow for the complete article to be read without registration. This article was widely available when I saved it in 2006, however now, not so much. I have the entire text if anyone wants to read it.

    9,000 dead in a province of 1.4 million people is still a very ugly number, and that number went up considerably in Diyala in 2007 and 2008, when many were calling it the most violent place in Iraq. Still, it is completely out of whack with L2’s extrapolation.

  66. #66 Mike H
    February 6, 2009
  67. #67 Mike H
    February 6, 2009

    Third try;

    Diyala

  68. #68 sod
    February 6, 2009

    In August, Mohamed Ali, a who statistician, reported his preliminary results to colleagues at a Denver statistics conference: Nearly 397,000 Iraqis had died because of the war as of July 2006.

    the 400000 number has been confirmed by the WHO, as written by Tim above.

    Every death fits into one or the other, without exception. Unlike death subsets further down the chain, there should be few errors recorded when determing whether a reported death is violent or non-violent, and any such errors would likely be intentional on the part of the survey respondent.

    would you know, whether a person in your household has died over the last 4 years?
    on the other hand, could you have another idea about the cause of death, than even the doctor? or could you be tempted to give a different explanation for the death, when asked about it? (because of shame for example?)

  69. #69 Information
    February 6, 2009

    the 400000 number has been confirmed by the WHO, as written by Tim above.

    I’m sorry, but that is a reference to an article in Mother Jones, not a reference to anything written in the study. Shall I repost the study’s wording?

  70. #70 bi -- IJI
    February 6, 2009

    QrazyQat:

    > one might think of going to the site of AAPOR, AMP, or NBRC, where they apparently forgot they’re supposed to be some clandestine operation and point out in numerous places that Applied Measurement Professionals is AAPOR’s management firm, and that both are a wholly-owned subsidiaries of the NBRC.

    OK, OK, I take back those claims of mine.

  71. #71 z
    February 6, 2009

    ” on the other hand, could you have another idea about the cause of death, than even the doctor? ”

    did some work on traffic accidents (in the US) a few years back. official cause of death listed in the government mortality records was pretty much useless, as a large proportion of the folks in a severe car crash, then hospitalized, then dead without ever leaving hospital were listed as cardiac failure. that may well have been the proximate cause of death, but most laypersons would consider that an indirect cause secondary to having your thorax crushed. (the effect is even more pronounced with death by illness rather than trauma)

  72. #72 Gaz
    February 6, 2009

    Barton (#59) – “I assume you’re not saying here that this many people were killed because of American bombing? Because there hasn’t been a lot of American bombing since the very first phase of the war. Most of the bomb fatalities have been from suicide bombers and IEDs.”

    I know the surveys being debated here do not concern deaths during the “very first phase”.

    The point I’m making is really that there’s an awful lot of focus on civilian deaths, women-and-children deaths, and deaths during the occupation rather than the war proper.

    I just don’t accept the implicit value judgment in all this that if a 19-year old girl gets blown up by a roadside bomb it’s a tragedy but if her little brother was conscripted into the army then vaporised by a 1000 lb bomb or buried alive in the sand by a bulldozer tank it’s just, hey, c’est le guerre, baby.

    Is someone killed during the “very first phase of the war” somehow less dead than someone killed six months later? How much less is the life of a conscript (Google: Saddam amputation conscription) worth compared to that of a civilian? Was the “Highway to Hell” in 1991 not that much of a big deal, morally speaking? Is there a formula or a lookup table or something?

    I know it’s important to find out what’s going on now so we can assess the success of the occupation of Iraq but surely the judgment must include the price paid in terms of lives lost in the “very first phase”.

    When it comes down to it, the idea that the lives of soldiers, especailly enemy soldiers, are worth less than other people’s lives is what makes war possible. I try not to get sucked in by it.

    Sigh.

    I have more chance of understanding GCMs than this weirdass moral calculus.

    I know this is off-topic. I just think there’s something generally wrong with the topic.

  73. #73 dhogaza
    February 6, 2009

    I have more chance of understanding GCMs than this weirdass moral calculus.
    I know this is off-topic. I just think there’s something generally wrong with the topic.

    In a way it’s not off-topic. What you’re seeing here is very similar to what you see at climate audit. Claims of fraud and incompetence on the part of individual researchers in order to spread FUD about facts that certain people find politically inexpedient. CA et al want to cast doubt on the notion that scientific claims regarding AGW are true. Kane et al want to cast doubt on claims that the Iraq war has cost the lives of a very large number of civilians.

  74. #74 Gaz
    February 6, 2009

    dhagaza: “Kane et al want to cast doubt on claims that the Iraq war has cost the lives of a very large number of civilians.”

    My point is that even if they fail to do that they’ve at least succeeded is drawing attention away from the countless thousands of young men in uniform who’ve been slaughtered.

    As a handy test, whenever you see the term “civilians”, substitute the term “white folks” and see how it sounds.

    So, doing that with your comment: Kane et al want to cast doubt on claims that the Iraq war has cost the lives of a very large number of white folks.

    Sounds pretty bad doesn’t it, because we know white folks and non-white folks should be valued equally.

    I just happen to think those young men in various armies and non-white folks should be valued equally.

    I know it’s not fashionable, but there you go.

  75. #75 David Kane
    February 6, 2009

    dhogaza writes: “Kane et al want to cast doubt on claims that the Iraq war has cost the lives of a very large number of civilians.”

    Oh, ye of little faith! I (and every single one of the academic Lancet critics I know) believe that the Iraq War has caused the death “a very large number of civilians.” Happy?

    We just think that the data underlying the Lancet studies is wrong.

  76. #76 dhogaza
    February 6, 2009

    My point is that even if they fail to do that they’ve at least succeeded is drawing attention away from the countless thousands of young men in uniform who’ve been slaughtered

    Well, no more than about 10,000 Iraqi military personnel are thought to have been killed during the invasion. Even if you only accept lower figures than Lancet II you’re looking at most something like a 5% as many uniformed soldiers killed vs. post invasion deaths. The real figure is probably considerably lower.

    If you’re talking about military and police killed during our occupation, it’s a similar “around 10,000″ figure – including recruits who’ve not completed training or even sign-up.

    The focus on the much larger group of deaths rather than your “white men” seems justified.

  77. #77 Gaz
    February 6, 2009

    dhogaza, certainly you could justify the focus on civilian deaths on the basis that there have been so many more of them then military deaths but I firmly believe it would be the same if it was the other way around.
    I watched in 2003 as everyone mourned the accidentally bombed civilians but even the trendiest bleeding heart liberal wouldn’t bat an eyelid when a bunch of troops was blown to smithereens.
    (A digressive point – sure “only” 10,000 or 15,000 or so were killed according to most estimates but the Coalition of the Willing was prepared to kill many times that number to achieve their goal. The relatively low number was the result of good luck or good planning, but it was not guaranteed by any means.)
    Anyway, my point is that soldiers just don’t seem to matter that much to people – it’s as if there’s a tacit consensus that a war where only soldiers were killed would be just fine. This doesn’t just apply to the Iraq war or just to wars the USA is involved in.
    I wouldn’t want anyone to stop counting the dead civlians on my account though.

  78. #78 sod
    February 7, 2009

    I (and every single one of the academic Lancet critics I know) believe that the Iraq War has caused the death “a very large number of civilians.” Happy?

    not yet. do you believe in the IFHS study of 400000 excess deaths since the invasion? (and seriously more up till now!!)

    why don t you spread your confidence in that number via the right wing blogosphere?

    that would make me seriously happy!

  79. #79 Mike H
    February 7, 2009

    would you know, whether a person in your household has died over the last 4 years? on the other hand, could you have another idea about the cause of death, than even the doctor? or could you be tempted to give a different explanation for the death, when asked about it? (because of shame for example?)

    One might mistake the stroke death of a family member for a heart attack death, but I wouldn’t expect a respondent to have any difficulty deciding whether a member of the household died from a stroke, or was blown to pieces by a car bomb.

    As I mentioned earlier, the temptation for a survey respondent to deliberately falsify cause of death when asked does not strike me as a widespread occurrence. In any case, there seems to be some serious cherry-picking as to the frequency of this alleged practice whenever this discussion arises, depending on which survey is being discussed.

  80. #80 Barton Paul Levenson
    February 7, 2009

    dhogaza writes:

    Al Qaeda wasn’t being aggressive?
    Not in Iraq.

    dhogaza, there is an organization called “Al Qaeda in Iraq” which has been in Iraq since about 2003 and has caused a great deal of death and destruction. Google for the full phrase.

  81. #81 Barton Paul Levenson
    February 7, 2009

    dhogaza continues:

    The fact that Al Q made an appearance in Iraq post-invasion supports a claim that we invaded due to Al Q aggression pre-invasion?
    Interesting.

    You are commenting on a claim I never made. Read what I wrote again.

    Portraying this as a simple war of Iraqi good guys against American/UK bad guys is not just wrong, it’s grossly ignorant.
    Jeff did no such thing. I thought Christians were supposed to be honest.

    It looked that way to me. I thought thoughtful posters like you were supposed to not jump to conclusions about the honesty of others without making some effort, however feeble, to check that honesty out. Do you really think I was deliberately misrepresenting Jeff, assuming that’s what I did? What the hell is wrong with you?

  82. #82 Barton Paul Levenson
    February 7, 2009

    Gaz, a consensus has developed over the past couple of centuries, at least in the west, that soldiers should try to avoid killing civilians if possible. Yes, it’s sad when a soldier dies, too. But the conclusion from “all deaths matter the same amount in all ways” must be either “let’s go to total pacifism” or “let’s kill whoever we need to to achieve our goals.” Since most people disagree with both of these, the distinction between military and civilian deaths will probably continue.

  83. #83 dhogaza
    February 7, 2009

    dhogaza, there is an organization called “Al Qaeda in Iraq” which has been in Iraq since about 2003 and has caused a great deal of death and destruction. Google for the full phrase.

    We invaded in 2003, and telegraphed that invasion for about six months previous.

    Again, you can’t use a reaction by AlQ to our aggression to claim that AlQ aggression was justification for our attack.

    AlQ wasn’t active in Iraq until we invaded.

    I googled. First hit:

    “Al-Qaeda in Iraq (AQI) is a group playing an active role in the Iraqi insurgency.”

    Insurgency. AGAINST OUR OCCUPATION.

  84. #84 Eli Rabett
    February 7, 2009

    Mike, bombing continued over Iraq at a rate that would cause multiple fatalities, besides which the US and Brits used air support at an even higher rate. Eli doubts deaths from air strikes with machine guns would be differentiated.

  85. #85 Mike H
    February 7, 2009

    Eli:

    Sorry, I’m not sure what you’re trying to establish with your post. Are you addressing the issue of violent death/non-violent death subsets, or the violent death disparity between L1 and L2 in the 3 Sunni-populated governates?

  86. #86 wildlifer
    February 7, 2009

    Wildlifer, If you think this distinction is applicable, shouldn’t it be up to the aggressors to count the number and status (civilian or combatant) of the dead as a result of their invasion?

    Not really.

    Why don’t you think the US or UK showed any interest in carrying out a detailed survey of the death toll in Iraq as a result of their combined aggression? Or, for that matter, of any wars waged? And is such a distinction between ‘civilian’ and ‘combatant’ deaths valid anyway? I am sure that many ordinary Iraqi civilians became combatants as a result of the fact that they don’t necessarily like being occupied by a foreign aggressor, particularly one that has murdered their loved ones. Given that the invasion violated just about every standard of international law (as well as the United States constitution), it was illegal anyway and thus so-called combatant deaths are also a crime.

    No it didn’t. We were in a cease-fire with Saddam dating back to ’91– one which he continuously violated during the Clinton administration. That was enough to take him out. I didn’t need the excuses of shrub’s administration.

    But back to my original point: it should be obvious why the US and UK did not carry out a census of the death toll as a result of the invasion. First, they didn’t care.

    Exactly. It doesn’t matter.

    Second, it the actual total was accurately known, at least to within a few thousand, it would further demolish the notion that the US and UK are great respecters of freedom and human rights. But as long as the actual total is very vague, then any estimate generated from any source is open to doubt. This enables the western corporate media to downplay the idea that the invasion resulted in utter carnage, and maintains the myth of the ‘basic benevolence’ of the west.

    War is hell. I don’t need the media to tell me shit and people get blown up in a war.

    The major aim here of defenders of unilateral US aggression is to sow doubt as to the actual death toll in Iraq. As long as there is doubt then it is possible, in public relations terms, to ‘manage the outrage’.

    It was war. What does it matter if it was 10,000 or 10,000,000?

    Actually, there are similarities with respect to the campaign waged by certain sectors and individuals to downplay the theory of anthropogenic climate change. The skeptics know that they will never win the scientific argument, but they don’t have to: so long as they can sow enough doubt amongst policy makers and the public, then nothing will change. If the true death toll as a result of the war in Iraq was known, then I am sure that there would be repercussions. So the governments, along with much of the corporate mainstream media, do everything they can to obfuscate the truth. They routinely attack scholars whose work suggests that the death toll was enormous, because, as I said above, it demolishes much of the myth that our governments pursue humanitarian agendas. They have agendas all right, but they are vastly different.

    It’s a stupid argument to have and a lesson I thought we learned in Vietnam. Body counts don’t mean anything.
    But to expand on my original point. We know Iraqis will lie through their teeth (Baghdad Bob). We can’t verify how many were or were not actual regular army troops who died, and are being posthumously “discharged”. Nor do we know how many deaths are attributable to sectarian violence (quite a bit it seems). But I do not believe at anytime our forces intentionally targeted civilians.

    (I fixed your typos. You’re welcome.)

  87. #87 Eli Rabett
    February 7, 2009

    Sorry Mike, it was Barton who was speculating that bombing from aircraft had stopped.

    OTOH, might one inquire of David Kane if he was the one who complained to the AAPOR?

  88. #88 P. Lewis
    February 7, 2009

    He has indicated otherwise, Eli:

    And, for those who care, I am not a member of AAPOR and did not initiate the complaint.

  89. #89 dhogaza
    February 7, 2009

    It was war. What does it matter if it was 10,000 or 10,000,000?

    I love it when people pooh-pooh the deaths of large numbers of other people.

    War’s hell! What does it matter of 6,000,000 Jews were killed by the Germans!

    Such a moral person, this wildlifer.

  90. #90 wildlifer
    February 7, 2009

    I love it when people pooh-pooh the deaths of large numbers of other people.

    War’s hell! What does it matter of 6,000,000 Jews were killed by the Germans!

    I love it when people think everything, no matter how unrelated, is the equivalent to Hitler.
    I wasn’t “pooh-poohing” anything. I’m just not shocked that people actually die in a war.

    Such a moral person, this wildlifer.

    Thank you.

  91. #91 Mike H
    February 7, 2009

    No problem, Eli.

    While I’ve got your attention, what are your thoughts with regard to my comment #65, and the huge regional discrepancies in violent deaths between L1 and L2?

  92. #92 dhogaza
    February 7, 2009

    No it didn’t. We were in a cease-fire with Saddam dating back to ’91– one which he continuously violated during the Clinton administration. That was enough to take him out. I didn’t need the excuses of shrub’s administration.

    Take him out, and it wouldn’t matter if we killed 10,000,000 innocent Iraqis in order to do it.

    I love it when people think everything, no matter how unrelated, is the equivalent to Hitler

    From a moral point of view, “unrelated” is an empty claim.

  93. #93 wildlifer
    February 7, 2009

    Take him out, and it wouldn’t matter if we killed 10,000,000 innocent Iraqis in order to do it.

    And saved exponentially more current and future lives in doing so.

    From a moral point of view, “unrelated” is an empty claim.

    Sorry, I must have missed the part where we marched them into gas chambers. You have a cite?

  94. #94 sod
    February 8, 2009

    OTOH, might one inquire of David Kane if he was the one who complained to the AAPOR?

    in the first post in the first topic of this subject, he said that he neither is a member, nor did make the complain.

    funny enough, i believe him. even though he didn t make public all his memberships and neither confirmed his claim with a public oath.

  95. #95 Barton Paul Levenson
    February 8, 2009

    dhogaza writes:

    Again, you can’t use a reaction by AlQ to our aggression to claim that AlQ aggression was justification for our attack.

    Again, I never claimed that it did. Again, you’ve jumped to a conclusion without taking the time to read what you were critiquing.

  96. #96 Gaz
    February 8, 2009

    Barton (#82), could I respectfully ask you to re-examine the logic of this statement:

    “Yes, it’s sad when a soldier dies, too. But the conclusion from ‘all deaths matter the same amount in all ways’ must be either ‘let’s go to total pacifism’ or ‘let’s kill whoever we need to to achieve our goals.’”

    That just doesn’t follow from what I’ve said (even allowing that I’d claimed all deaths mattered equally, which of course I didn’t). It’s a very un-Barton-like argument, I must say.

    Wars result in death and injury. That cost must be weighed up in deciding whether – and how – to fight a war. It is simply not logical to extend that argument to either absolute pacifism on one hand or militarism unfettered by concern about killing on the other.

    My argument is simply that the cost in terms of dead soldiers, in particular the dead soldiers on the other side in lop-sided contests like the Iraq wars, is not given enough weight, and often no any weight at all, these days.

    The idea that the lives of soldiers, particularly enemy soldiers, don’t matter much seems to enters the war-making decision as an axiom.

    Well, it’s not an axiom, it’s a value-judgment even if it is part of a consensus.

    Let’s put it this way.

    Say, you think the invasion of Iraq was justified for some reason – WMD’s, getting rid of a tyrant, restoring the ecology of the marshes, whatever.

    Now let’s say we knew in advance that a million Iraqi civilians would die as a result.

    Would it be worth it in order to achieve the objective?

    If not, where would the cut-off point be? 100,000? 20,000? 5,000?

    Implict in all the discussions about of surveys of civilian deaths is the idea that at some point the price may not have been worth paying (or, more correctly, not worth forcing the Iraqi people to pay).

    Now, let’s ask this question.

    How many Iraqi military deaths would have been worth it?

    A million? 500,000? 10,000? 1,000?

    Where is the cut-off point?

    Is it different from the cut-off point for acceptable civilian deaths? If so, why?

    How is the decision made? By whom? On what basis?

    By the way, Barton, I disagree with this statement from you: “Gaz, a consensus has developed over the past couple of centuries, at least in the west, that soldiers should try to avoid killing civilians if possible.”

    Well, maybe that would apply for the past 50 years, but if you go back a couple of centuries you’d have to include World War 2.

  97. #97 dhogaza
    February 8, 2009

    And saved exponentially more current and future lives in doing so.

    You’ve got some sort of evidence to back up this claim? Start with *current* lives, please, since you’ve used the ‘and” word. I assume the exponent involved is greater than 0 or 1?

    Details, please.

    Oh, and while Godwin’s law is cool, I can’t help but point out that nazi germany justified slaughtering jews with a similar argument, i.e. that eventual jewish domination of the world, if not checked, would lead to Horrifying Consequences For The Master Race and other christians.

  98. #98 dhogaza
    February 8, 2009

    And, BPL …

    Jeff Harvey, predictably, writes:
    that is what the ‘conflict’ was: aggression on the part of the US and its proxies.

    Followed by your response:

    Al Qaeda wasn’t being aggressive?

    This suggests that aggression on the part of the US was triggered by, justified by, or somehow related to aggression by AlQ.

    If you meant this to be a meaningless, pointless deflection from the issue, my apologies for misunderstanding your point. Not that I think such a meaningless, pointless deflection is any more effective at countering Jeff’s point.

  99. #99 wildlifer
    February 9, 2009

    You’ve got some sort of evidence to back up this claim? Start with current lives, please, since you’ve used the ‘and” word. I assume the exponent involved is greater than 0 or 1?

    Details, please.

    Are you entirely ignorant of Saddam’s history?

    Oh, and while Godwin’s law is cool, I can’t help but point out that nazi germany justified slaughtering jews with a similar argument, i.e. that eventual jewish domination of the world, if not checked, would lead to Horrifying Consequences For The Master Race and other christians.

    You sound just like the right-wing reactionaries, and just as revisionist.

  100. #100 Barton Paul Levenson
    February 9, 2009

    dhogaza writes:

    Jeff Harvey, predictably, writes: that is what the ‘conflict’ was: aggression on the part of the US and its proxies.
    Followed by your response:
    Al Qaeda wasn’t being aggressive?
    This suggests that aggression on the part of the US was triggered by, justified by, or somehow related to aggression by AlQ.

    No, it doesn’t suggest any such thing. It’s a complete and total non sequitur.

    Let me try and spell this out so even you can understand it.

    I understood Jeff to be saying, “The only source of damage in Iraq was American/UK aggression.”

    I was saying, “No, other parties also caused a lot of damage.”

    Do you get it yet?

    If you meant this to be a meaningless, pointless deflection from the issue, my apologies for misunderstanding your point. Not that I think such a meaningless, pointless deflection is any more effective at countering Jeff’s point.

    See above.