Monday, November 01, 2004

One Hundred Thousand cont'd

Oh dear, I was waiting. Rebuttal. Thank you Charles WT for providing this valuable opportunity. You can't possibly be telling me you place any faith in all those partisan links? Let me begin with Shannon Love's post. I will use the Q A format, two answers for each question.

Q; Shannon observes; "Most other sources put total combined civilian and military deaths from all causes at between 15,000 to 20,000". Shannon then asks "The Lancet study is a degree of magnitude higher. Why the difference?"

A 1; Most other independent sources do not have the resources to conduct an in depth study to the degree that The Lancet has undertaken. Specifically, most other sources have not recently conducted studies in field. The last in field study was very near the beginning of the conflict. There have been many more occasions for loss of life during conflict since those early days. It is no surprise that the death rate has increased since that first and as far as I am aware only other in field study.

A 2; Most other non-independent sources although possessing the resources to conduct comprehensive casualty studies, do not have the inclination to do so. In fact most other non-independent sources possessing the resources to conduct comprehensive casualty studies have an interest in suppressing such data.

Q; Shannon calculates that; "100,000 deaths over roughly a year and a half equates to 183 deaths per day" and then speculates "With that many people dying from air strikes every day we would expect to have at least one or two incidents where several hundred or even thousands of people died" before asking "Heard of anything like that?"

A 1;
183 deaths per day spread throughout areas of conflict in Iraq does not neccesarily amount to mass incidents of hundreds or thousands of deaths. Let us take six well known areas of conflict in this war.

  • Baghdad. Major civilian center subject to air raids.
  • Basrah. Another major civilian center subject to air raids.
  • Fallujah. Subject to "insurgency containment".
  • Tikrit. Saddam's "home town". Subject to severe military "operations" during the hunt for Saddam.
  • Mosul. In Northern Iraq/Kurdistan. Subject to notable conflict near the wars beginning.
  • Ramadi. Of particular focus in recent planned strikes.

I have mentioned only six areas of conflict and already if one divides 183 by those 6 mentions one has a result of only 30.5. A result of 30 and a half casualties per day spread over only six cities. When one looks at the number of towns marked on this map showing CPA presence in Iraq - oop, map diversion. Take a look at that CPA map. CPA North, CPA South, CPA Central and CPA Baghdad. The whole bloody country is occupied by the goddamn CPA. Is that a map of Iraq or is it a map of some new occupied territory named ah, "CPA"?. But as I was saying. When one looks at the number of towns marked on that map and divides 183 by them all, one can see how incrementally easy 183 deaths per day can accumulate in a war-torn country.

A 2; Well yes Shannon, with that many people dying each day you could expect at least one or two incidents where several hundred or even thousands have died. In fact, you can find those incidences recorded right here, on Iraq Body Counts site. Isn't that good of them to have been so diligently keeping track of things. Sorted by largest entry first, a few Iraq Body Count excerpts;

19 Mar 2003 -
09 Apr 2003
- 19 Baghdad hospitals - - 1482

4 Apr 2003 -
31 Aug 2003
- Violent deaths recorded at Baghdad city morgue - 60 per cent from gunshot wounds 1214

20 Mar 2003 -
24 Apr 2003
- Municipality of Baghdad - - 778

20 Mar 2003 -
06 Apr 2003
- Nassiriya - various 633

05 Apr 2004 -
30 Apr 2004
- city of Falluja 'insurgents', incl. those responsible for the killing of four US security contractors air attacks and ground battles 572

20 Mar 2003 -
20 Apr 2003
- Hospitals in Najaf, Karbala, Mosul, Samawa, Madain, Diwaniyah, Kut, Tikrit - various 484

01 Sep 2003 -
30 Sep 2003
- Violent deaths recorded at Baghdad city morgue - Over 55% by gunfire 362

Please note these are only a few excerpts from Iraq Body Count. Iraq Body Count provides a minimum and maximum figure where different media sources report different figures. I have quoted the conservative minimum estimate. Larger entries are sourced from hospital morgues. It is reasonable to consider that most dead bodies are buried rather then taken to hospitals, hence it is highly likely that these reports are only an "iceburg tip". A time delay can be expected between the occasion of conflict and the gathering and accumulation of bodies in morgues, hence the date spans on these items. Also worth noting that some hospital casualty records were deliberately confiscated and destroyed early in the conflict. Sorry this sounds so cold and grim but that's the way war is and I see no reason to sugarcoat it.

Oh, one other little thing. With that many people dying each day it is surprising that anyone might be left standing to report such incidences at all. Probably there are a fair few unreported incidences where either a journalist was not present to record it or no one was left alive to report it or someone was left alive but they had getting the hell out of there more on their minds then finding a journo to report too.

Only two paragraphs through Shannon's obviously bogus discourse, and so much more to answer. However, one step at a time. To be continued, probably.


#11/01/2004 05:52:00 pm Assalam Aleikom Blogger Bruno

Gentlemen and ladies, that is what is commonly referred to as "whuppass". Glad to see our host can dish it out.

Personally I feel that 100000 is probably a tad high ... but on the other hand, the US military claims it is killing 1500 + insurgents a month ... and given that the guerillas are still active, who *did* they shoot?

#11/01/2004 09:45:00 pm Assalam Aleikom Blogger CharlesWT


You're welcome.

"You can't possibly be telling me you place any faith in all those partisan links?"I don't place much faith in anything without proof, partisan or not.

If you have the time and energy to dig through all the comments made at the links I gave, you will certainly find plenty of arguments for and against the study.

#11/01/2004 10:51:00 pm Assalam Aleikom Blogger Michael

There's a kind of "folk" interpretation that confidence intervals have on account of their use in opinion polls. Many people are under the impression that if one candidate has a 3-point lead over the other in a survey with a 3% margin of error (usually from a 90% confidence interval), it means that they're most likely tied. It doesn't. The interpretation which is closer to its technical meaning is that there's a 5% chance of a tie or reverse lead. Likewise, the 95% confidence interval in the study says that there's a 2.5% chance that the actual toll is under 8,000. How rosy is that?

But the researchers have themselves to blame for undermining a valuable study by enouraging the media to misrepresent it in a sensationalistic way. E.g., "... we're quite sure that the estimate of 100,000 is a conservative estimate, said Dr. Gilbert Burnham." That's technically consistent with the paper, which goes to some lengths to consider ways in which the estimate could biased upwards. But it has no bearing on the confidence of the estimate. What the team should have done was present several confidence intervals and explain what they tell us. Statistics are there to communicate uncertain findings, not to bedazzle the unwashed, even if for the best of reasons. Instead, now anyone who wants to resolve the discrepancy between their expectations and this number in favor of sounder sleep will have a pundit on their TV screen ready to explain how this is "junk science."

The study also quite deliberately refrains from researching the circumstances of violent deaths, unlike their occurrence. They don't even report the breakdown of responsibility attributions for Falluja (71% of violent deaths in the survey) separately from the rest of the country. The message of the paper is to urge the military to investigate and record civilian casualties, which is not quite what one reads in the press. All of this might be misleading for humanitarian reasons if tomorrow Americans were voting on whether to kill some more Iraqis. On the balance, it doesn't look to me like the brightest page in the history of either scientific or journalistic ethics.

#11/02/2004 10:23:00 am Assalam Aleikom Anonymous Anonymous

ok, some iraqis what? these are the same hate-filled idiots that were dancing in front of the tv camera on 9/11 and shooting their guns into the air like fools and loving the fact that america was attacked and people were dying. and i am supposed to fill sorry for these people now? what upsets me is that more than $1 of my tax money is going to feed these jerks. at least the people in afghanistan and the kurds are civilized and deserve our support. which US candidate wants to exterminate the entire arab population? i'll vote for him.

#11/02/2004 04:49:00 pm Assalam Aleikom Blogger Bruno

Right on anonymous!

Tell me, do you also recruit for Al Qaeda in your spare time, or is your only hobby trying to persuade others that all Americans really are as moronic, arrogant and self inflated as yourself? Because, if they were, then I'd also be dancing in the streets on 9-11.

#11/04/2004 11:04:00 am Assalam Aleikom Blogger emigre

This comment has been removed by a blog administrator.

#11/04/2004 02:01:00 pm Assalam Aleikom Blogger emigre

This comment has been removed by a blog administrator.

#11/04/2004 02:04:00 pm Assalam Aleikom Blogger emigre


If you had even a shred of credibilty you might have actually read the study instead of just making stuff up. That way you would know the figure of 100,000 civilian deaths excludes data from Fallujah. If the researchers had been able to obtain data from Fallujah the figure may have been higher.

Because the Lancet study is such a responsible study with such high ethical standards interviewer safety was always a priority. In minimising danger to data collectors in field the study excluded areas deemed too dangerous and violent, areas where interviewers might have been at risk and become part of the death toll themselves.

One can only imagine what the researchers might have found had they been able to obtain data from those more dangerous areas.

#11/07/2004 06:35:00 pm Assalam Aleikom Anonymous Anonymous

The Lancet report has already been debunked in many newspaper articles.

Usually a study like this takes at least six months and is subject to rigorous academic proofing. This report appears to have been rushed through in much less time and without any proofing at all.

Much mention has been made of the timing of the report, and it appears that the report was politically and not scientifically motivated.

It would be interesting to compare this figure to a study of the number of civilians killed by the former regime, and the civilians killed by the insurgency , however this has not been even taken into consideration by the Lancet's clearly flawed report

#11/09/2004 07:53:00 pm Assalam Aleikom Blogger Papa Ray

If the figure of dead is 100,000 or 600,000 it is still in line with past history of conficts. In fact it is too low.

I predict that the figure will be over one million dead in Iraq before this "small event in history" is finished.

If it is any confort, more people than that have died in automobile accidents in the USA in the last century.

This post is mine

Papa Ray
West Texas

#11/10/2004 02:33:00 pm Assalam Aleikom Blogger emigre

Dear Anonymous

I can only conclude that (like Michael) you have not read the Lancet Report in full either, otherwise you would have jumped on the red herring in my reply (to Michael).

Herring; "If the researchers had been able to obtain data from Fallujah".

Researchers were able to (and did) obtain data from Fallujah, although it was not used in calculating the 100,000. Findings from the Fallujah cluster sample accounted for two-thirds of all deaths reported by interviewees.

That a cluster sample which could have scewed results was excluded in calculating civilian casulaties only validates the integrity of the researchers. Study findings were submitted for peer-review, revision and editing at the beginning of October and published at the beginning of November. That is a damn sight more time then any blog commenter takes to check their claims.

#11/12/2004 12:48:00 pm Assalam Aleikom Anonymous Anonymous

i really dont care how many poor pathetic iraqis lost their lives. many of the deaths might have been from all the car bombs and stuff from terrorists. but who cares? over 1,000 americans have died in iraq and nobody says anything about that. and besides, 1 american death is equal to 1million iraqi deaths in my eyes, so we have a lot more killing to be done.


Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours? Weblog Commenting by