No disagreement with your points re civilian mortality in Iraq (although I've already heard rightwing nutbars in the US claim that the Coalition bears no responsibility for deaths due to sectarian violence--never mind that we took the lid off with no plan to deal with it. We gave them freedom, by God, and it's not our fault that those brown people can't handle it!). However, re:

The Eisenhower (and another carrier group) are due to arrive in the gulf on October 21st.

The Yorkshire Ranter has a somewhat more sanguine take on the USS Eisenhower carrier group deployment. I don't remember seeing anything there about a second carrier group, though. Which one were you thinking of?

JBWoodford

Charlie, I wish you'd post more political opinions on your blog. If you think U.S. politics and policy look weird (OK OK downright evil) from where you are, it looks even worse here. President Clinton was impeached for "high crimes and misdemeanors" consisting of . . . well, nothing of much significance. Now, we've got a president, much of his cabinet, and members of Congress who are doing much worse, and all the talk is about how they've got to keep their majority in Congress or the Democrats will take charge, and we all remember how bad Clinton was.

(Remember Clinton? How the "family values" crowd screamed for blood? Where are they now?)

I live in the most Republican state in the nation, and I can tell you it's disturbing. My own values are more on the "conservative" side - i.e., I believe in a government of limited powers - but I don't see any so-called conservatives who even pay lip service to that ideal any more. Sad, and disheartening.

So, feel free to wax political on your blog. I won't complain.

Actually, recent polls show the Republicans are in trouble -- not because of the war (which half of Americans support more or less) -- but because of Foley and his naughty emails.

The truth is, most Americans don't really care how many Iraqis die. Most opposition to the war is based on the number of US deaths or the cost, as far as I can see. Most of the deaths are the fault of the Iraqis, the US has been remarkably restrained (look at what we did in Vietnam, for instance).

And personally, I think a civil war is better than oppression. The best thing we could do is split Iraq into three countries. Trying to create nations out of Colonial era administrative divisions doesn't work.

Your job, if you're voting in the upcoming election, is to decide whether you want to let a politician who cold-bloodedly ordered 655,000 murders in order to win his last mid-term election get away with the same trick twice, on behalf of his page-buggering, bribe-taking buddies.

Decided long since, at least for me, and most of the people I know. But then I live in a state that would be bright red (Republican), if it weren't for a whole bunch of us radicals who live in a city that, at least for a very long time, was one of the most liberal, and best places to live in the US (I'm speaking of Portland, which begins with 'P' and that rhymes with 'T', and that stands for "Toss the bums out".

The reason I qualified that statement about how good Portland is as a place to live is a local news article I saw a couple of days ago. For a long time, Portland has been a very tolerant place, especially by US standards. It was one of the few cities to attempt to change the climate of homophobia that's been rising in the last few years by performing gay marriages (until the state government stopped us). So I was shocked and saddened to find out that a local high school principal had decided to stop his school from performing a play in which tolerance towards homosexuality is openly advocated. The fact that the play re-enacts the rather gruesome murder of a teenager just because he was gay (a real crime committed only a few years ago), and that the play is primarily about why killing people you disapprove of is a bad thing, seems to have escaped the principal's attention in his zeal to protect his students from the "homosexual agenda".

But don't mind me. I'm just a foreigner, and my opinions don't count.

Not true. You don't get to vote in the election, true, but one of the serious flaws in the United Snakes' national character is the belief that other people's opinions don't matter. But then, we learned that from the English (we don't call you "wogs", we have other nasty names).

Charlie,

You're mistaken on several items.

I suppose there could be something in the Geneva Conventions on counting civilians killed (although I'm not aware of any), it would only make sense that this applies to the ones killed by U.S. action. I seriously doubt this could apply to those killed by insurgents.

The notion that "the occupiers are refusing to keep records of civilian deaths" is dubious. I can only guess that this comes from the quote atop a particular anti-American website where General Tommy Franks was answering a question about enemy body counts in Afghanistan. He only meant that we don't keep score that way, which is what the reporter's question was getting at. For all the noise about the Geneva Conventions, it appears to me that the U.S. and its allies are the only ones who do care about them. (Note that the far left has all but disavowed them.)

Besides that, the "occupation" ended a couple of years ago. Some might say that the U.S. still takes the blame for civilians killed by our enemies, but I think any expectation that we'd still get the job for tallying them takes this too far.

On the Lancet study, I'm sure you remember they published a six figure body count a couple of years ago, and it was disputed then. I'd question any study that polls Iraqis at a time when reporters have to hire local stringers to fetch stories from out of the green zone. Some of the Iraqis certainly have motives to inflate their body counts anyway.

As for an imminent attack on Iran, I think there's no chance of that as things stand now, and doubtful even if provoked. You might recall that Bush had been criticized for holding the retaking of Fallujah until after the 2004 elections. What you're suggesting would run counter to that.

Charlie: Yeah I notice that the same people who criticise the methodology for this study don't criticse similar studies (often carried out by the same people) when they fit their particular agenda. Where do they think that the numbers for the Congo, or Sudan came from? For that matter, what about the various estimates for people killed by Saddam?

Randy:

"The notion that "the occupiers are refusing to keep records of civilian deaths" is dubious."

Nothing dubious about it. They don't keep numbers, and have stated on several occasions that they don't think it is their job. If you think they're not refusing to keep records, then presumably the US military is keeping figures, in which case where are they?

"On the Lancet study, I'm sure you remember they published a six figure body count a couple of years ago, and it was disputed then."

It was disputed by people who were statistically illiterate with ludicrous arguments. Nobody (and I really mean nobody) who had any expertise in these kind of sampling, or understood statistics, criticised the report. The main criticism made by real experts (rather than random party flacks on CNN) was that the sample size was quite small (something acknowledged by the writers, who didn't have the resources to interview more people), which meant that the accuracy of the figure was quite low. This time the sample size is far larger, and so the number is far more accurate.

"I'd question any study that polls Iraqis at a time when reporters have to hire local stringers to fetch stories from out of the green zone. Some of the Iraqis certainly have motives to inflate their body counts anyway."

The survey was carried out by Iraqi doctors with experience of carrying out surveys. 80% of deaths were verified with death certificate, and there was no significant difference between those verified in this way, and those which weren't (so its unlikely that people were lying).

The authors of the paper, btw, who have carried out similar surveys in some of the most dangerous war zones in the world. There probably is nobody better qualified to design this kind of survey.

Article 34 of the

Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol 1) Adopted on 8 June 1977 by the Diplomatic Conference on the Reaffirmation and Development of International Humanitarian Law applicable in Armned Conflicts entry into force 7 December 1979, in accordance with Article 95

Implies records. It is not restricted to results of the actions of combatants. All of the Geneva Conventions are available to read on the Web, with commentaries and other remarks about them. Denial is not a satisfactory response.

I think the counter-revolution has succeeded. A pity.

(The Lancet has a more complicated reputation, it is good, but it also has been controversial - outspoken - from early on. Johns Hopkins which is where the paper has authors also has a good reputation.)

Cian,

I only assume the military is keeping figures of those killed in engagements with them. That doesn't mean I expect anything to be made public.

My point was not about whether or not the U.S. is counting civilian casualties, but whether or not the U.S. is "refusing to keep records" and I inferred from Charlie's post that some treaty requires this, and we're breaking it by not doing so. If there is some treaty obligation then I've never seen it cited.

If this survey was carried out by Iraqi doctors then that doesn't reassure me in the slightest. Making comparisons to those with death certificates is an interesting way of discerning honesty, but that doesn't reassure me either. I can just see these doctors explaining their ethical obligations to Al Sadr's goons.

It turns out that the previous study "just happened" to come out a month before the U.S. elections in October 2004. At this rate we might expect the next to appear conveniently a month before election day 2008.

But let's assume for the moment that the study is correct. Where are all the "anti-war" movement-types demanding an end to the civil war?

They're more than willing to march against the U.S., and certainly Israel, but what about stopping the fighting before another 655,000 Iraqis get killed? Does anybody care about that? Or do you think the Sunni and Shia will start shaking hands after the U.S. leaves?

Adrian,

The U.S. is not a signatory to Protocol I.

Even so, Article 34 is about respecting their graves. To read it in the way you're reaching for would imply that we need to look at every headstone.

Cian,

If you don't see my message yet, be assured (not that it really matters), I replied earlier. I think that's being held because it contains links. Either that or I went into the wrong section, in which case I humbly apologize to Charlie.

Charlie, you know what despots used to do to messengers bearing bad news. Nowadays they can't do that - its not the publicity so much as the difficulty getting blood out of that nice blue caropet - so they get their lap dogs to yap and yap and yap until everyone is so sick and tired of the noise that they cover their ears and so don't hear the bad news.

Which doesn't mean the bad news isn't true.

Before you place any faith in the Lancet study, Charlie, please consider this: the Iraq Body Count (www.iraqbodycount.com) states between 43,000 and 48,000 Iraqi deaths in reported and confirmed violent events since the invasion in 2003. This means that, according to the Lancet study, for every death by violence that's been reported, between 7 and 14 more deaths by violence have occurred that have not been reported. This, when the press in Iraq has been looking specifically for deaths by violence, ignoring all else. Does that strike you as plausible? What fraction of violent deaths due to "the Troubles" in Northern Ireland were reported at the time, and thus would have appeared in an "Ireland Body Count", compared to the true total?

Regarding a potential Iranian invasion, I point you at a comment I made on another blog. Rather than clog things up here, I'll just attempt a link to it here. In short, our military is stretched too thin, and enough neo-cons are aware of it for a third invasion to be doable without a draft. Some far extremists in our own government think a Draft is doable, but at this stage of the game any congress critter who votes in favor of one will REALLY get dinged in the polls. NOT an election-year issue.

That said, some big new Iraqi/Afghani offensive in the weeks before the election are well within reason.

As for the Lancet article, I heard a discussion with the authors today on NPR. There was some pointed questioning, but one quotation sticks in my mind. When asked about inflated numbers and how the researchers could validate their information, the guy said that the Iraqis asked about deaths could produce a death certificate in 80% of the cases. The audio can be found on the NPR.org site here, though I'm having trouble finding a transcript to quote from.

As for the Geneva requirements, we haven't been an Occupying Power since we handed over power in.. summer of whatever it was.

I can't understate the impact of a vote to reactivate the draft will have on American politics. Especially one that either narrowly fails or actually passes. These wars aren't quite popular right now, though the ambivalence factor is pretty high. Having the government force Our Young Men (and those Women who receive draft requests and show up of their own will) into battle brings the war MUCH closer to home than it has been. It won't be just military families facing the possibility of a tour-of-no-return, it'll be civilian families with zero military background doing the fearing. That ambivalence will gel into Opinion very quickly. That's why I think a third front is not feasable at this time.

The Iraq Body Count (which is actually at www.iraqbodycount.org) is a count of deaths which got mentioned in the Western Press. Personally, I would not be surprised for that to be a minority of the actual "excess deaths"; YMMV.

Where are all the "anti-war" movement-types demanding an end to the civil war?

It seems obvious to me that "the 'anti-war' movement-types" in question march when and where they think they have some hope of influencing events. You may note the lack of recent marches against the Iraq war in the US; I conjecture that this is because "the 'anti-war' movement-types" believe that the current administration will listen to them about as much as the various foreign fighters, Iraqi militias, and other assorted insurgents will.

It turns out that the previous study "just happened" to come out a month before the U.S. elections in October 2004.

Having submitted a paper for publication in a research journal, I can tell you that in my experience the timing of the process can't really be controlled that well. Even with only two reviewers it's a crap shoot how long the peer reviews take, what changes the reviewers will ask for, how long it'll take to make those changes, and what they may ask for after the first round. I've read that Lancet uses four reviewers, which just makes it more likely that at least one of them is going to ask for substantive changes. Do you know when it was submitted? Do you know if the authors had to make any changes? Do you seriously think that the authors have any control over the timing of the publication? If...IF the editors tried to time it that precisely I'll concede the point, but right now all I've seen is your speculation.

JBWoodford

Michael, first, it's hard to report from Iraq. IIRC, foreign journalists have been forced to rely almost entirely on Iraqi journalists, and that those people are under massive constraints. Please remember that Iraq is a place where the police set up roadblocks, and murder people for having a name which implies the wrong religious sect.

Second (I don't have the cite), it was claimed (by the Lancet article authors for the first article, IIRC), that, with the sole exception of Bosnia, there's a rule of thumb which states that fewer than 20% of deaths are reported in the media, in countries which are experiencing civil wars. If one stops to think of this, it's hardly surprising - reporting is lethally difficult, there's a constant drumbeat of killings, and many factions in, around, or outside of the government don't want things reported on.

Third, (again, IIRC), the Iraq Bodycount Project has very stringent requirements - two independent accounts of a death.

Randy: " Where are all the "anti-war" movement-types demanding an end to the civil war?"

Probably right next to the pro-war people, who are conducting those massive marches every month, demanding that the US also intervene in Dafur, the Congo, etc.

Randy, do you understand that (a) protests in the USA won't effect the behavior of the Iraqi combatans, (b) the potential protesters are well aware of that, and (c) that the US doesn't have any available method of bringing this war to a successful conclusion?

The Lancet study is a load of crap. It was a sample of exactly 547 reported deaths, extrapolated to extremes by the study's authors.

As a comparison, from Tim Blair:

"Let’s put Lancet’s number in perspective:

• It is larger than the total number of Americans killed during combat in every major conflict, from the Revolutionary War to the first Gulf War.

• It is more than double the combined number of civilians killed in the bombings of Dresden, Hiroshima, and Nagasaki.

• It is a larger number than were killed in Germany during five years (and 955,044 tons) of WWII bombing."

Not to mention, of course, that somehow the invasion of Iraq more than doubled the incidence of death from old age (apparently, the presence of US forces make Iraqis age at many times the rate they were aging at before, due to some top-secret time flow accelerator gadget) and quadrupled the incidence of death from heart failure and coronary disease, according to the report (I guess the Evil American Fast Food has started having an effect). Not to mention the increase from two violent deaths under Saddam to over 300 in the current sample...

In other words, you've been suckered by some extremely bad propaganda masquerading as science.

Cirby, you seem to have forgotten that the situation here is different- we're talking about what, a country of 20 million people, over a period of years, in a conflict involving lots of small arms. Look at Rwanda, Darfur, bosnia etc. Its amazing what you can do with lots of small arms. Comparisons to Dresden etc are pointless.

The germans in WW2 had bomb shelters, evacutations etc. Here we have what is in effect localised violence and murders. Not to mention that in WW2 there were more targets than just cities with people in them.

And do you have a cite for it not being standard methodology to extrapolate in this way in a study? Do you have a counter study, carried out using standard statistical methods, showing a different number?

And Andrew G, I'm so glad I dont live in the same country as someone such as yourself, who considers civil war, with its associated heavy body count and immense amount of pain, as better than a dictatorship.

Somebody has been suckered Cirby, but somehow I don't think its Charlie.

"The Lancet study is a load of crap. It was a sample of exactly 547 reported deaths, extrapolated to extremes by the study's authors."

No, it was a sample of 12801 individuals. This yielded 629 births, 80% of which were verified using death certificates. If you don't understand the difference, then you're really not in a position to judge the validity of the report.

"* It is larger than the total number of Americans killed during combat in every major conflict, from the Revolutionary War to the first Gulf War."

And its far smaller than the number of Vietnamese killed by America in the Vietnam war, or the number who've died in the Congo. Who died in the German invasion of the Soviet Union. Than who were exterminated in concentration camps. So what? What is this supposed to prove?

"* It is more than double the combined number of civilians killed in the bombings of Dresden, Hiroshima, and Nagasaki."

You're suggesting that two nuclear bombs would have similar affects to that of violent resistence to occupation, civil war and bloody lawlessness? I mean seriously? Gosh.

"It is a larger number than were killed in Germany during five years (and 955,044 tons) of WWII bombing."

And did the report suggest that all these people were killed by bombs? That this is the result of bombings? Again, what relevance has this got to anything?

"Not to mention, of course, that somehow the invasion of Iraq more than doubled the incidence of death from old age"

Not entirely sure where you got that from. I couldn't see any reference to that in the report. Non-violent deaths have increased significantly, but then the health care system has collapsed, the water system has collapsed, lack of fuel and electricity means people can't run air-conditioning - so how is this surprising?

"Not to mention the increase from two violent deaths under Saddam to over 300 in the current sample..."

Um. You do know that Saddam wasn't engaging in mass murder in the few years prior the invasion, right? That those killings happened earlier? You do know that the criminals who are kidnapping and killing people at the moment were locked up in jails? You do know that there was not a violent civil war going on? That there was no bloody resistance to the US going on in places like Fallujah?

Cian, you mean 629 deaths, not births.

So the Associated Press is a propaganda arm of the Republican party? Charlie, do you have any idea of how paranoid, hysterical and ignorant that sounds?

How far Left do you have to be before liberal news organizations like the AP look like fascists?

Several people have noted that the IBC only tracks reported deaths, and that these are typically 20% of actual deaths. But it's also worth noting the IBC only tracks civillian deaths directly attributable to US/allied military involvement. A US soldier shooting a civillian dead is counted by the IBC. However, deaths due to disease, poverty, the entire country's infrastructure being wrecked, not to mention shootings and bombings by non-Allied forces, are not tracked by the IBC. The Lancet study does include these, and compares them to pre-invasion figures. The effect is that while the IBC measures only the effects of military combat operations, the Lancet measures the effect of the invasion, toppling of Saddam, resultant instability, and subsequent mismanagement of the country by the occupiers. Combine this with the 5x multiplier from reported vs actual deaths, and the Lancet figure sounds surprisingly low to me...

And, Mr Duffy, 1. Charlie didn't accuse the AP of being a propaganda arm, he accused the CSIS of that -- wrt. the AP, he accused one specific article of being biased, and 2. The AP is only "liberal" relative to your own personal viewpoint.

Perhaps the article's author is biased, or perhaps the article is another victim of "unbiased" reporting having come to mean "reporting every claim equally, however dubious the source" -- such as a collection of Defense Department officials, investment bankers and oil company executives.

Randy, I didn't see your response before: "I only assume the military is keeping figures of those killed in engagements with them. That doesn't mean I expect anything to be made public."

1) The Lancet article surveyed the number of Iraqis who died, no matter what the cause. 2) What could the US military possibly gain from keeping such figures secret, if they were lower than the other estimates being made? The only reason to keep them secret (if they are being kept), would be because they are embarrassing in some way. 3) Such figures if they were kept would not be terribly accurate (they wouldn't include a large number of bomb/shelling victims - or bodies that were burried in some way). The Lancet methodology, as it happens, is far more accurate.

"My point was not about whether or not the U.S. is counting civilian casualties, but whether or not the U.S. is "refusing to keep records" and I inferred from Charlie's post that some treaty requires this, and we're breaking it by not doing so."

I believe the Geneva convention is quite explicit on the need for occupying powers to do this. I haven't checked this personally.

"If this survey was carried out by Iraqi doctors then that doesn't reassure me in the slightest. Making comparisons to those with death certificates is an interesting way of discerning honesty, but that doesn't reassure me either. I can just see these doctors explaining their ethical obligations to Al Sadr's goons."

This is pretty incoherent. I'm not sure what's "interesting" about the use of death certificates - it seems a fairly sensible and uncontroversial approach, but then I'm clearly not the expert on epidemological surveys that you seem to believe that you are.

Nor can I see why the doctors would have to explain anything to Al Sadr's goons (particularly in Sunni and Kurdish areas). Given that a huge number of the people killed seem to have been killed by either criminals, or Iraqi militia men (its all there in the report, perhaps you should read it), what possible interest would they have in inflating it.

"It turns out that the previous study "just happened" to come out a month before the U.S. elections in October 2004. At this rate we might expect the next to appear conveniently a month before election day 2008."

Even if this is true, so what? Are you really so contemptuous of democracy that you think its important that when Americans are voting they shouldn't know the full facts about Iraq before voting? If 600,000 people have died as a result of this, might that not be relevant?

"But let's assume for the moment that the study is correct. Where are all the "anti-war" movement-types demanding an end to the civil war?"

Um, come again? Are you suggesting that marching in New York/London/wherever is going to have an effect on militia men in Iraq? Or violent criminals in Iraq? Is this supposed to be a serious suggestion?

"They're more than willing to march against the U.S., and certainly Israel"

Do you not think that going on a demonstration against your own government might have slightly more effect than marching against fighters half way across the world? And Israel is financially supported by the US, so marching aginst (or indeed for) them in the US also makes some sense.

"but what about stopping the fighting before another 655,000 Iraqis get killed? Does anybody care about that? Or do you think the Sunni and Shia will start shaking hands after the U.S. leaves?"

Given that the violence has got worse and worse with the American presence, and wouldn't have happened without the US invasion, what purpose does our presence there serve? Things might well get worse. But things are getting worse anyway.

Daniel, the Associated Press is a wire service. What it puts out on the wire is whatever its affiliates put into it. Those affiliates come from a number of organizations, with different editorial and other policies. Because it's just a raw newsfeed you need to carefully evaluate the source and bias (overt or otherwise) of everything you read on it.

Clear so far?

And putting the word "controversial" in the first sentence of a news release about a peer-reviewed statistical study in a highly respected academic journal of record is a fairly good indicator that someone is trying to spin the piece.

Cirby: "Not to mention, of course, that somehow the invasion of Iraq more than doubled the incidence of death from old age (apparently, the presence of US forces make Iraqis age at many times the rate they were aging at before, due to some top-secret time flow accelerator gadget)"

Cirby, let me clue you in on something. When one is elderly, one experiences a myriad of afflictions. Modern medical care, good food, clean water, etcl. are very useful for mitigating these. Lack of same puts one into a position where life is nasty, brutish and short (to steal a quote). Do you actually not understand human biology 101?

"and quadrupled the incidence of death from heart failure and coronary disease, according to the report (I guess the Evil American Fast Food has started having an effect)."

To reiterate, it's amazing how deprivation works so well with other causes to kill.

Barry: I suspect you need to recalibrate your sarcasm detector. It appears to be giving false negatives ...

AP are, of course, one of the institutions mis-describing Foley as a Democrat. That's how credible their reporting is.

For those of you who desperately need the Iraqi death numbers to be so high, note that, overall, conditions in Iraq aren't anywhere near "desperate," or, for that matter, bad at all. They've got a strong, growing economy (you don't get that in a place where people are dying at the rate the Lancet study is pretending). The various world health groups note that, overall, people are getting better and more plentiful food than before the war.

Combat operations aren't that intense or common (only a few percent of the troops in country even fire weapons, and air ops are not anywhere near intense enough to account for that many deaths).

Of course, the Lancet study is a factor of ten or twelve higher than any other study that's been done, and nobody in Iraq has noticed the massive number of increased funerals (go read some Iraqi blogs, and be ready for some pissed-off Iraqis demanding to know where all of the dead people are, since they aren't seeing or hearing of them).

Finally, where are the cemetaries? Six hundred thousand dead in just three years? There should be a massive number of new cemetary plots, and nobody seems to be able to find them, either... (that silly "the press can't cover it" argument is pretty sad, since this is one of the most-covered wars in the history of the world - it's not the Cold War USSR).

Martyn,

As Charlie points out above AP itself doesn't report. It is a news feed of news from member organisations. Its initial purpose was to provide local newspaper groups with a means of sharing stories so that they could include news from outside the local area or country.

It was one of their member associations that reported Foley was a Democrat (as Fox News also appeared to do at least 3 times). Mind you most people who watch Fox News probably wouldn't vote Democrat even if the Republican Candidate was a known Murderer or a gay dead chimpanzee.

As with all content the quality of the output depends on the quality of the input. By the looks of it the idea of a News feed without a filter is still a strange thing for many people online to understand.

Cirby,

Where's you source that the Iraqi economy is growing. I've seen some pretty damning evidence that that is not the case in most of country (bar the Kurdish North).

Cirby, I'd love to live in your world. Really, I would. One in which combat operations aren't intense or common (there are, after all, only 500-600 attacks per week). It'd be lovely to be able to buy the panglossian reporting from the Pentagon that the rate of deaths in Baghdad is dropping -- but to get there, the Pentagon had to systematically exclude deaths due to bombs, mortars, rockets, and other mass attacks from their figures. (So if you're killed by a car bomb you don't count.)

The Iraqi economy shrank by over 50% from 2001 to 2003. Finding recent figures is hard work, but I'd be surprised if you could point to anything much more than a dead cat bounce in the past three years, driven by the high price of oil -- what oil they're still pumping, that is.

So Iraqis are better fed than they were back when we were systematically starving them before the war? Great: but starvation isn't what they're dying of.

Your assertion that combat operations aren't that intense or common because only a few percent of the troops in the country ever fire weapons is basically bullshit: in general, 90% of any military operation is the support tail, and guns are generally only fired by the combat troops tasked with being on the cutting edge. Air ops are irrelevant too -- policing by bomber didn't work for the RAF in the early 1920s in Mesopotamia, and it's not terribly effective in Baghdad today.

As for your cemetary argument, that's nonsense too. Iraq has a population of c. 24 million. You'd therefore expect a normal peacetime population turnover of about a third of a million per year. An increase of 70% over that isn't going to cause new cemetaries to be opened immediately -- your average cemetary runs for decades before anyone thinks about expanding it -- and indeed, huge cemetaries aren't inevitably associated with mass killing. (Go visit Auschwitz-Birkenau if you don't believe me.)

And Andrew G, I'm so glad I dont live in the same country as someone such as yourself, who considers civil war, with its associated heavy body count and immense amount of pain, as better than a dictatorship.

I rather hope I live in a country that would prefer civil war to oppression and tyrany.

Then again, I come from a country that thought the british monarchy - of all things - was an intollerable tyranny, and where half the country started a civil war over what they considered oppression from a rather weak central government.

As Patrick Henry said, "Is life so dear, or peace so sweet, as to be purchased at the price of chains and slavery? Forbid it, Almighty God! I know not what course others may take; but as for me, give me liberty or give me death!"

One thing to keep in mind is that the vast majority of the attacks are concentrated in a few regions -- Baghdad for instance. So it's entirely possible for the economy to be growing in large parts of Iraq, with a lower death rate than other parts where fighting is more intense.

Depending where you are looking you'd either get the impression that it's a horrible war or nothing major.

Andrew G: which is worse, a dictatorship that kills a few thousand people a year, or a civil war that kills hundreds of thousands (and is usually ended when another strong man comes to power)?

Hint: civil war tends to create oppression and tyranny (and a pile of skulls along the way) rather than overthrowing it.

In the case of the American war of independence, I'd have to call Patrick Henry a hypocrite: that war he promoted probably prolonged the institution of slavery in the American south for some decades. It was very good for the rich North American landowners (who had had the previous war fought in their defense at the expense of the British taxpayers, and then rebelled when asked to pay their share of the bill), but lousy for everybody else.

PS: Iraq's population is mainly urban. Those "few regions" where the attacks are concentrated are actually where 70-80% of the population lives.

(Now you'll have to excuse me -- but I have a job to get back to, namely writing the next [almost entirely apolitical, you'll be pleased to know] SF novel.)

It does look like we have representatives from two parallel worlds arguing, doesn't it?

A position from World A: the US is still the occupying power, whatever fancy dress parties are going on. The Hague Regulations define occupation as territory under the control of a hostile army. The presence of a central government with no authority over the occupying troops does not magically change this; the Hague Regulations emphasize de facto control. There is no status of forces agreement (and no way to try US criminals under Iraqi law, and US law chooses not to cover contractors), the Iraqi army and (I believe) intelligence services still report to the US and not the Iraqi government. Amnesty phrases it as follows: "The sole criterion for deciding the applicability of the law on belligerent occupation is drawn from facts: the de facto effective control of territory by foreign armed forces coupled with the possibility to enforce their decisions, and the de facto absence of a national governmental authority in effective control."

In addition, the Iraqi prime minister is not the man who was elected for the job - the US pressured him out because they were unhappy with him. This is not an independent government.

World One sounds like a much nicer place - maybe you can send some advisors from your Bush One to our Bush A to make our Iraq as successful as yours.

What's fascinating about Cirby is that despite attacking the Lancet study, he cites figures supporting his argument from god knows where and gathered god knows how?

How does he know the economy is growing strongly? Who Knows? How were the economic figures gathered? Who knows, Cirby surely doesn't? Was the methodology sound? Cirby doesn't know, Cirby doesn't care. How does Cirby know so much about what's going on in Iraq, despite the fact that most of it is now not covered by journalists (due to the extremely high level of violence, that he seems to missed, despite knowing all this other stuff about the place), and the bureacracy has broken down? Hard to say. He doesn't tell us.

given how unserious his response is, and that his criticism of the Lancet report largely comes down to him not believing the number could be that high (has he read the report? I kind of doubt it), it seems pretty clear that its Cirby who needs the number to be lower - much lower. All the dead Iraqis could be laid out in a field, and still Cirby wouldn't believe it. He's not interested in facts, he will never change his mind. So don't take him any more seriously than you would somebody who argued the earth is flat, or that evolution is impossible.

You're right. It's better to keep out of political commentary.

Charlie, good points. I'm just trying to point out that it's pretty engrained into the American national mindset that wars of liberation and civil wars to remove oppression are good things and worth the cost. Probably 2/3 of the country don't really care how many Iraqis die, especially if they perceive them as doing it to each other. Most of the objects here come from poeople feeling that too many Americans have died.

If it wasn't so expensive, the majority of people would probably be in favor of invading Iran, North Korea, Venezuela, and maybe Cuba. Luckily the US is pretty miserly when it comes to actually paying for wars (as you noted regarding the Revolution). We keep looking for cheap solutions that don't work, like embargos.

When I posted that rant I was expecting at least one troll to show up.

(I'm disappointed that I only appear to rate a C- grade know-nothing, though.)

As for my motives in posting ...?

First I tried letting it all hang out. I got trolls, they drove my blood pressure up, and it was all quite stressfull.

So I put a lid on it.

But then I found that keeping quiet has its own costs: because self-censorship makes you complicit in the atrocity, even if only indirectly and on a very small scale, and that didn't do my blood pressure any good either.

If the trolls get too much for me this time, I'll just switch on Typekey authentication, force all posters on the discussions to register, and then ban the trolls. But Cirby is nowhere near that point ...

But then I found that keeping quiet has its own costs: because self-censorship makes you complicit in the atrocity, even if only indirectly and on a very small scale, and that didn't do my blood pressure any good either.

I know the feeling, some times you have to speak out. I too feel angry by the death toll in Iraq. My anger is directed at those who cause the deaths, the insurgents and terrorists and religious fanatics in Iraq. Iraq is only one of two things that Bush has done that I agree with. (The other being tax cuts).

Andrew G: Yeah, you're right about the mindset that prefers fighting wars to leaving other people to wallow in tyranny.

Back during the cold war I was much more scared of Margaret Thatcher and Ronald Reagan than of the Soviets. The Soviets were rational: their country had been invaded by western powers four times in the preceding 200 years, and while they ran an unpleasant, dour, oppressive string of dictatorships, once Stalin was in his grave they were basically predictable -- their behavioural repertoire boiled down to (a) don't let the bastards invade the Rodina again, and (b) wait for True Communism to arrive and solve all the external problems.

Whereas Ronnie and Maggie were starry-eyed idealists who kept talking about freedom! and liberty! and the urgent need to destroy! the evil empire.

And I lived two miles from a large railway station and marshalling yard, three miles from one of the largest tank factories in Europe, and five miles away from a major motorway junction.

It tends to give you a pragmatic outlook on things ...

And that's part of why I was strongly against the Iraq invasion right from the start. Because, sure: Saddam was a complete bastard. But a quick look at the history of the region suggested that you need a complete bastard to hold Iraq together (at those times when they haven't had one, there's been civil war or terrorism instead), and one look at the troupe of clowns who were proposing the invasion suggested that these were the very last bunch of chancers you should ever go on a Magical Mystery Tour with. The proposed remedy was clearly going to be at least an order of magnitude worse than the disease -- even though I didn't expect it to be as badly mismanaged as it has turned out to be.

Charlie, science writer Malcolm Ritter (the author of the AP piece) is hardly a stooge for the Bush administration. I suggest you Google him and his works, expecialy his reporting on global warming. So yes, you do come off as biased, ill-informed and a little hysterical.

Now while wire services like AP rely on affiliates for their news feeds, you'll find that (outside of lonely Fox News and the WSJ) the American media is almost uniformly hostile to Bush and views the world through a liberal prism. Perhaps not the extreme off-the-edge-of-the-planet left wing view you seem to subscribe to, but biased enough to ensure that almost all AP news affiliates are safely anti-Bush.

All I know about the Lancet report is that the previous estimate was serously flawed with a too large of a margin of error to be meaningful, so I'l approach this report with suitable caution. So Ihave some questions concerning their methodology.

For example, was their sampling confined to the 4 provinces out of 18 that are actually experiencing violence? Did they then take these samples and extrapolate over all of Iraq including the peaceful and prospering Kurdish areas?

When polling the Iraqi people was there any overlap to the question of whether or not they knew of family members or neighbors that had died. In a tribal society like Iraq "families" can be very extensive and different Iraqis could have been referring to the same death. How was double count bias eliminated in the report (assuming it was)?

Were those questioned reliable? Certain groups (especially the Sunnis) have a vested interest in making the American presence in Iraq appear as bad as possible. Not so much because they are fighting oppresive imperialist overlords so much as they used to be (under Saddam) the oppressive imperialist overlords and would like to be top dog again.

Does the report break down how many were directly killed by American military action as opposed to terrorist attacks directed against Iraqi civilians by the likes of Al-Qaeda-in-Iraq. If so how do these numbers compare to the deliberate incineration of Japanese civilians in Tokyo fire bombings (100,000 dead in a single night raid) by American bombers or similar attacks by the RAF on German civlians in Dresden (60,000 dead)?

How does the Lancet death rate compare to the average annual death rate during Saddam's regime including judicial murder, genocide against the Kurds and other minorities, the war with Iran, etc. Some historical context would be helpful.

...even though I didn't expect it to be as badly mismanaged as it has turned out to be.

Yeah, it has been badly mismanaged. Not on the miltary/combat side -- if anything that went better than anyone thought. The problem was at the start, when Bush & co. had no intention of engaging in "national building" and no plan for what to do after Saddam was toppled.

Without local allies ready to take over like in parts of Afghanistan, you really, really, need a plan to engage in nation building. Every successful war has had one.

"Whereas Ronnie and Maggie were starry-eyed idealists who kept talking about freedom! and liberty! and the urgent need to destroy! the evil empire."

I seem to recall that Ronnie and Maggie actually did bring down the evil (what other adjective would you use to describe a regime that murdered 10s of millions) empire.

"But a quick look at the history of the region suggested that you need a complete bastard to hold Iraq "

Are you saying that Iraqis (and by extension Arabs) are brown savages incapable of understanding the white man's democracy? So why is it when a lefty says soething like this its "multi-culturalism" and when a conservative says such things it's "racism"?

Thats the point Andrew G- you said half the country preferred not to be ruled by the British. What about the other half? (It was interesting to find out that many people who fought for the gvt were actually descendents of Clansmen who had hated the gvt) To extend it further, what right has the USA got to decide for another country whether it will live in freedom or not? And I look forwards to their invasion of Burma, to free its people from the military dictatorship currently in power.

And Andrew, the point to bear in mind is that the Lancet study also found that 31% of post invasion violent deaths were caused by the coalition. You can also blame the coalition for taking the lid off things in Iraq, and letting it get as much out of control as it has done.

Cian, JB and Barry,

First, it has since come out that people at Lancet admitted to political timing. They do have an agenda but, in fairness, they're being clear about it. There's even a video of one of these guys giving a speech at a Stop The War Coalition rantfest.

I've also learned that the U.S. military has issued a report to Congress that includes some casualty figures. You may find a PDF on the web called "Measuring Stability and Security in Iraq." The figures aren't parallel, though, so it's difficult to compare.

IAC, I still maintain that the Geneva Conventions do not seem to have this requirement. I'm not a lawyer, nor am I an expert, but I've done some digging, and no one has yet come up with a direct reference. (These treaties are all online.)

Cian asked, "what possible interest would they have in inflating it." Every group in Iraq that thinks they can get the upper hand with the withdrawal of U.S. forces does have a strong interest in increasing these numbers.

As for the idea that "anti-war" types can't be expected to protest against terrorists, that might have been a reasonable argument at one point but we've seen too many opportunities for true idealists be squandered.

I had linked to images of those marching in Manchester, claiming it's for peace but clearly taking the side that uses children as human shields (a practice that violates the laws of war by anyone's definition). They can say what they want about Israel but it's clear to me that this "peace" movement doesn't give a damn about the Geneva Conventions. The use of children as human shields should never be acceptable. Those protesters had a moral duty but said nothing.

Yes, I do think they could make a difference. Hezbollah has a $100M budget, and it's supported by Iran who also have sympathies for the Sadrites in Iraq (who do want us out). Iran has great need for more good PR. Iran is itself allied to the likes of Hugo Chavez, who could certainly say something if he cared to.

No, that wouldn't directly stop the violent criminals in Iraq, pressuring the Sadrites could help bring some order that would change that environment. You could even say I'm wrong about whether it would help but you can't say they've tried.

One of the recurring arguments against U.S. policy is that it had allied itself with despots during the Cold War. Well, I'm making the same point here about the other side. I don't expect miracles but I do think people speaking out is a good first step.

You don't have to accept any of this. Just understand that the moral high ground has a very tough terrain.

And Andrew, the point to bear in mind is that the Lancet study also found that 31% of post invasion violent deaths were caused by the coalition. You can also blame the coalition for taking the lid off things in Iraq, and letting it get as much out of control as it has done.

I have a certain ammount of faith in my government and military that they aren't conducting large scale killings of innocent civilians. Most of those are insurgents, or unavoidable casualties from fighting insurgents.

Randy: I do not adhere to the doctrine that the enemies of my enemies are my friends. I do not support the Iranian government or Hezbollah, just because I despise the actions of the US administration. You're quite right that the moral high ground has rough terrain: but if you want to claim that terrain, a good starting point is to denounce violence everywhere, and to renounce it yourself except in strict self-defense. (Which the invasion of Iraq most certainly wasn't.)

You might want to remember that I'm not American, nor do I consider the USA to be a shining city on a hill or a beacon of liberty; it's just another goddamn meddling superpower, with no intrinsic claim to superiority other than that which it earns. And right now, your government is doing a really shitty job of convincing me it's superior to, say, that of Leonid Brezhnev.

Andrew G: I believe (for the reasons set out above) that your confidence in your government is very likely misplaced. (But then, the US military has a history of covering up its massacres of civilians very effectively, most of the time. Aided by a complaisant media environment in which only American deaths count.)

Andrew G: I believe (for the reasons set out above) that your confidence in your government is very likely misplaced. (But then, the US military has a history of covering up its massacres of civilians very effectively, most of the time. Aided by a complaisant media environment in which only American deaths count.)

True, we have had our nasty massacres from time to time. I think every great power has at some point.

But looking at the study it seems that there's a 5x-8x difference in the number of women to the number of men killed, and that most of the casualties are clustered in the 15-44 age group. This doesn't seem to be an indication of massacres of civilian populations to me, but rather combat deaths.

The study's results seem to confirm my belief that there is a three+ sided civil war going on in Iraq. Looking at the map, the highest death rates are clustered in the middle of the country, in the American zone -- the region where most of the Sunnis live. They're the ones who have the most to loose in the new government -- an unpopular minority with few resources in their territory.

Andrew, can you explain to me how the increased deaths from dysentery, cholera, and malnutrition in Iraq are unavoidable casualties from fighting insurgents? It's easy to find references, just Google for "dysentery Iraq", "malnutrition Iraq", etc.

Also, I'm not really clear on how a civil war between three authoritarian groups is superior to a dictatorship. I don't think the Freedom Sexy party is likely to come out on top here.

Andrew, can you explain to me how the increased deaths from dysentery, cholera, and malnutrition in Iraq are unavoidable casualties from fighting insurgents? It's easy to find references, just Google for "dysentery Iraq", "malnutrition Iraq", etc.

I believe the 31% mentioned in the study as being caused by the Coalition Forces refer only to violent deaths, not deaths in general. Numbers-wise they're saying that about 601,000 people died of violence caused by the war, while 'only' 54,000 more than usual died of other causes.

So, the Coalition caused about 200,000 deaths through violence, and the insurgents caused about 400,000. They don't assign blame for the 54,000 though I tend to think the blame lies squarely on the heads of the insurgents who sabotage efforts to rebuild the country.

Also, I'm not really clear on how a civil war between three authoritarian groups is superior to a dictatorship. I don't think the Freedom Sexy party is likely to come out on top here.

Sadly, you're probably right. Though I'd like a Whiskey, Freedom, & Sexy Party here in the US too. :)

All in all, I think the Kurds probably have a good chance of creating a democratic government. With the threat of Turkey and the Arab Iraqis as external threats they should have enough interal cohesion to avoid a Kurdish civil war between the two main factions.

Whoops, ignore that first para - I lost the plot that we were just talking about deaths by coalition fire, not all deaths.

Remember that Haditha was accepted as "unavoidable casualties" for a while, and probably would continue to be if there had not been video evidence to the contrary. There are also quite a number of accidental killings at traffic stops, and I question whether those were unavoidable.

Even granting the 31% as unavoidable, what about the rest?

For those craving more analysis on the methodology of The Lancet's report but too lazy to read the original .pdf, The Guardian (yes, right-wingers can wave their special Liberal Elite Media flags if it makes them feel better) has an article getting into it in at least a little more detail. A brief excerpt:

There was a sample of 12,801 individuals in 1,849 households, in 47 geographical locations. That is a big sample, not a small one. The opinion polls from Mori and such which measure political support use a sample size of about 2,000 individuals, and they have a margin of error of +/- 3%. [...] I have searched assiduously for the last two years, with the assistance of a lot of partisans of the Iraq war who have tried to pick holes in the study, and not found any.

Randy, The timing may be motivated by the elections, but so what? If the figures are accurate (and you've failed to show that they're not), then people should know about them before they vote. Whether they choose to act on them is their business. To argue anything else is to demonstrate a profound contempt for democracy.

I glanced through the document you mentioned, but couldn't find any figures on casualties (there were figures on attacks on infrastructure and what I presume were attacks on US soldiers).

Cian asked, "what possible interest would they have in inflating it." Every group in Iraq that thinks they can get the upper hand with the withdrawal of U.S. forces does have a strong interest in increasing these numbers.

The Lancet figures show that a majority of people were killed by either "insurgents", or criminals. Again, if the figures were faked by Al-Sadr, why would they produce figures that suggested they had killed as many as the Americans?

"As for the idea that "anti-war" types can't be expected to protest against terrorists, that might have been a reasonable argument at one point but we've seen too many opportunities for true idealists be squandered."

No, I asked what was the point of marching in the West to demand a civil war? What measurable outcome could it possible achieve? Would you seriously argue that either side in the civil war would take any notice? That they care what people in London, or New York, think? This is a silly argument.

"Pressuring the Sadrites could help bring some order that would change that environment. You could even say I'm wrong about whether it would help but you can't say they've tried."

Again, please explain how marching against the Sadrists in the west would put any pressure on them? What would this achieve? How would it work?

Cian

Charlie,

I wasn't including you as a member of that group in Manchester. Rather, I was making the point that the far left is taking sides. They clearly do support Hezbollah.

If there is a position for the center-left, it should be to call their acquaintances in the far left and ask them to defend their positions. I'm not asking for anything more than that. I don't even expect every blogger to confront this. I'm just noticing the volume is pretty low, and I can't help but think that our enemies have noticed this, too.

I agree on renouncing violence among rivals, even hated ones. It's doubtful that we can ever do the same with madmen.

I'm well aware of your thoughts on U.S. policy, and I don't begrudge you for having them. While I understand your wanting to away from politics, I agree with others who think it's good for you to be up front about your opinions. You're not the author of children's books. It would almost be two-faced for you to keep silent.

To Michael Brazier:

The cases of Northern Ireland during the conflict and Iraq are not comparable.

Even at its very worst, in the early 1970s, the situation in Northern Ireland did not begin to resemble the horror that is unfolding in Iraq today.

And there certainly never occurred in Northern Ireland the kind of breakdown of civil order that would impede the gathering of accurate mortality records, nor their reporting by the media.

Therefore the answer to your question 'What fraction of violent deaths due to "the Troubles" in Northern Ireland were reported at the time, and thus would have appeared in an "Ireland Body Count", compared to the true total?' is that the true total of dead in the NI conflict (3,500) is equivalent to the figures reported at the time.

There's even a book on the subject, Lost Lives, which lists, in detail, every single death that occurred as a result of the Northern Ireland conflict.

That you would introduce an inappropriate comparison with Northern Ireland into this debate indicates that you know very little about the present unpleasantness in Iraq: you definitely know nothing at all about Northern Ireland.

Cian,

"The timing may be motivated by the elections, but so what?"

Nothing wrong with that. I just think it makes this more partisan.

I haven't read the full report yet. I don't know if the figures are accurate, but my position (and probably yours) would be the same regardless. Frankly, the deaths of 1/10th that many should be troubling when there are a lot of innocents in that number. But it could affect my position only if I could believe an Iraqi iron hand could impose order in a way that didn't come along with a greater threat.

"Again, if the figures were faked by Al-Sadr, why would they produce figures that suggested they had killed as many as the Americans?"

I have no doubt that, if faked, they'd want it to be believable. I'll grant that it's a good question whether the enemies we have could be that astute.

"No, I asked what was the point of marching in the West to demand a civil war? What measurable outcome could it possible achieve? Would you seriously argue that either side in the civil war would take any notice? That they care what people in London, or New York, think? This is a silly argument."

Take a look at the types of people demonstrating now. Do you think George W. Bush cares what they think? Those photos come from a right-wing website. If they thought it could dissuade Bush, it might even have had the opposite effect. It's somewhat affirming.

Wouldn't you take comfort if the far-right (most of whom actually agree with you, BTW) were to demonstrate against your position?

Cian,

Sorry but I realize I didn't make that last point clearly -- even by the standards of someone who's trying to work on several things at the same time.

My contention is that they do care. Even an axis of evil needs allies.

It makes a difference to whoever is funding them. It makes a difference in recruiting. It makes a difference in whether they think they'll be recognized as a legitimate government after they gain power. And it makes a big difference in whether they can convince their followers that they can hold out long enough for victory. To them, such silence is a gift.

The number of Iraqi dead claimed by the Lancet report (655,000) exceeds the total German civilian deaths during five years (and 955,044 tons) of WWII bombing.

From The NY Times:

Robert Blendon, director of the Harvard Program on Public Opinion and Health and Social Policy, said interviewing urban dwellers chosen at random was ?the best of what you can expect in a war zone.? But he said the number of deaths in the families interviewed ? 547 in the post-invasion period versus 82 in a similar period before the invasion ? was too few to extrapolate up to more than 600,000 deaths across the country. Donald Berry, chairman of biostatistics at M. D. Anderson Cancer Center in Houston, was even more troubled by the study, which he said had ?a tone of accuracy that?s just inappropriate.?

Lancet's numbers are BS, pure and simple.

The numbers of dead in the Rwandan genocide twelve years ago exceeded both the numbers report in the Lancet study and the numbers killed in Germany by allied bombing from 1939 - 1945.

This was achieved largely with very simple technology - machetes.

The Iraqi factions are rather better armed. So I don't think a comparison with the dead in Germany six decades ago is adequate for the purposes of undermining the Lancet reports credibility.

Randy: f there is a position for the center-left, it should be to call their acquaintances in the far left and ask them to defend their positions. I'm not asking for anything more than that.

Sure, fine: "stop focussing on your own concerns and beat my enemies up for me instead. Otherwise I will have to assume that you are an [insert epithet here]."

Personally, I think you should start by denouncing the evils of corporate-owned sweatshops and demanding the compulsory nationalization of the means of production. After all, it's the only way to prove that you're not in bed with the neo-nazis, isn't it?

(Sorry, but I am not going to wear that hat: it is pointy and has a big "D" on the side.)

Charlie,

"and beat my enemies up for me instead"

Did I say "instead"? I wouldn't even ask for fifty-fifty.

In fact, I didn't even ask for that. I said, "ask them to defend their positions". You were, after all, asking us to defend ours.

"Nothing wrong with that. I just think it makes this more partisan."

Different argument. For the 2004 report, the editor of the Lancet (who did not write, or review, the report) stated that he persuaded the reviewers to get their reviews back to him as soon as possible (reviews usually take several months - mainly because academics are very busy, and so things get put off), due to the importance of the findings. He also stated that he felt it was important that if possible (reviews withstanding), the report should be published before the US elections.

However, to suggest that the writers of the 2006 report could organise a study of this size (and it is a massive sample size), analyse the data aiming to publish it through the peer review process in order to get it published one month before the election simply demonstrates your ignorance of how academic research is published.

An article is written by academics and submitted to the journal (the editor does not commission articles). The editorial board then find reviewers with relevant experience of the methodology/research area to review it. In this case four people. If, and most articles submitted to the Lancet are rejected, it makes it through this process it will then be rewritten to address the concerns of the reviewers and finally published. This process can easily take a year - I know people for whom it has taken two years.

You're claiming that the authors of this report (who are at one of the world's most prestigious medical schools, and who have excellent reputations for the quality of the work), the reviewers (who will be of the same standard) and the editorial board of the Lancet (which will include other eminant researchers, with reputations to lose) - would all be that biased is simply ludicrous. The Lancet is not the New Republic/Nation/NYT/Washington Times.

"But it could affect my position only if I could believe an Iraqi iron hand could impose order in a way that didn't come along with a greater threat."

So essentially you're saying that it doesn't matter how many people die, your opinion will remain the same. And the argument, as I see it, is that these are people who wouldn't have died if we hadn't invaded. At the very least we owe them a massive apology.

"I have no doubt that, if faked, [Al-Sadr] want it to be believable. I'll grant that it's a good question whether the enemies we have could be that astute."

No, I asked why would Al-Sadr fake data that made them look as bad as the US occupation forces? As a hypothesis this is ludicrous. I'm trying to get you to see how silly your suggestion was.

I find the idea that this data is faked, btw, utterly ludicrous. The interviewers only asked for death certificates after they interviewed people about deaths. The idea that Al-Sadr could have anticipated the interviews, and have prepared so thoroughly is a pretty wild conspiracy theory.

"Take a look at the types of people demonstrating now. Do you think George W. Bush cares what they think?"

Well in Britain, at least, I don't think (for the most pat) they were protesting to change Bush's mind. The idea is to try and change the population's mind so, come an election, things might change. And to make it absolutely clear that the government was doing this in their name.

"Those photos come from a right-wing website."

Um, well thanks for that.

"Wouldn't you take comfort if the far-right (most of whom actually agree with you, BTW) were to demonstrate against your position?"

The far right agree with me about what? Iraq? Not in Britain they don't. If you're talking about Israel (and I haven't actually stated my own position), the far right in Britain are very supportive of Israel (they hate Arabs more than they hate Jews, and they don't hate Jews if they live in Israel).

DOK, you miss the point. Lancet's figure comes to about 500 civilian deaths per day since boots hit sand in 2003. We have imbedded correspondents over there reporting on the violence. There is on-going documentation of what is going on there, daily reports. And the numbers have spoken over time, then this report comes up with numbers off the charts of anything witnessed by imbedded journalists or even foreign correspondents. Has even Al Jazeera been reporting 500 civilian casualties a day?

Sorry, but the Lancet numbers are BS - politically motivated BS that real staticians are shooting down.

D. O'Kane: If you read the blogs written by Iraqis, or by people residing in Iraq, you will not find any trace of a breakdown of civil order, as would be necessary to impede the collection of mortality records. What you do find is a situation broadly similar to that of Northern Ireland at its worst: namely, a functional civil society under assault from a variety of sectarian thugs, and defended (not well) by an army from a foreign country. Northern Ireland's mortality records were accurate during the Troubles; why would Iraq's mortality records now be suspect? The Lancet study is the only evidence presented so far that civil order in Iraq has broken down. Assuming the breakdown, and citing it as evidence for the Lancet study, is therefore circular reasoning.

The possible explanations for the factor-of-12 discrepancy between the reported mortality figures from Iraq and the Lancet study's estimate are: 1) civil order has disappeared in Iraq; or 2) the murderers of Iraqis are hiding most of the bodies so well that only 1 out of 12 of them have been discovered; or 3) the US government, the Iraq government, and the Western press are all colluding to conceal the true death toll; or 4) the Lancet study is in error, due to mistakes in calculation or conception. Of these, 2) and 3) are on a level with stories of abductions by aliens. 1) is contradicted by those currently in Iraq, as I mentioned before. Therefore I vote for 4) -- the more so because I know enough about statistical modeling to know how easy it is to get it wrong.

Cian,

I could accept that the work is high quality, and that I'm wrong about a lot of things, but I'll bicker on this point:

"You're claiming that the authors of this report (who are at one of the world's most prestigious medical schools, and who have excellent reputations for the quality of the work), the reviewers (who will be of the same standard) and the editorial board of the Lancet (which will include other eminant researchers, with reputations to lose) - would all be that biased is simply ludicrous. The Lancet is not the New Republic/Nation/NYT/Washington Times."

I had said that I consider 1/10th that many deaths to be troubling when it contains so many innocents. Would I lie if I believed it could bring some relief on that scale? Yes, I would. I think it would be wrong not to.

Please do not assume that I'm accusing Lancet of lying in that. (I understand the long-term importance of maintaining standards.) I'm merely wondering whether they're humanly capable of holding the same standards when someone gives them what they're looking for.

"So essentially you're saying that it doesn't matter how many people die, your opinion will remain the same. And the argument, as I see it, is that these are people who wouldn't have died if we hadn't invaded."

No, but I understand that you may hear others say that. I have a different opinion about the rationale for the war. What I'm saying is that giving in to the insurgents would have more perilous consequences.

"No, I asked why would Al-Sadr fake data that made them look as bad as the US occupation forces? As a hypothesis this is ludicrous. I'm trying to get you to see how silly your suggestion was."

A higher death toll only makes each side look like victims, and that's especially important for the worst of the insurgents. Shia deaths are attributed to Sunni, and Sunni to Shia.

After Dresden, German propagandists felt free to exaggerate the bombings' body counts.

"I find the idea that this data is faked, btw, utterly ludicrous. The interviewers only asked for death certificates after they interviewed people about deaths. The idea that Al-Sadr could have anticipated the interviews, and have prepared so thoroughly is a pretty wild conspiracy theory."

I'll concede that makes sense. There could be some steering for the sampling, but I'm not going to hang my hat on that unless it makes sense after I've read more of the report.

"Well in Britain, at least, I don't think (for the most pat) they were protesting to change Bush's mind. The idea is to try and change the population's mind so, come an election, things might change. And to make it absolutely clear that the government was doing this in their name."

That's a fair point as long as the guys with Hezbollah banners and t-shirts understand that Hezbollah is doing things in their name.

"The far right agree with me about what? Iraq? Not in Britain they don't. If you're talking about Israel (and I haven't actually stated my own position), the far right in Britain are very supportive of Israel (they hate Arabs more than they hate Jews, and they don't hate Jews if they live in Israel)."

I don't know enough about the far right in Britain.

In the U.S., the isolationist-right opposed the war, and the extreme-right (such as David Duke) is even more vehemently against it. They see this as a war for Israel. Duke is pretty popular in some Arab countries.

Posted by: daniel duffy: "DOK, you miss the point. Lancet's figure comes to about 500 civilian deaths per day since boots hit sand in 2003. We have imbedded correspondents over there reporting on the violence. There is on-going documentation of what is going on there, daily reports. And the numbers have spoken over time, then this report comes up with numbers off the charts of anything witnessed by imbedded journalists or even foreign correspondents. Has even Al Jazeera been reporting 500 civilian casualties a day?"

First, please read the report - it covers deaths from all causes, not just violence.

Second: http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2006/09/21/world/middleeast/21iraq.html&OQ=_rQ3D1Q26exQ3D1159588800Q26enQ3D450b0f4670d95964Q26eiQ3D5087Q250A&OP=4f8965acQ2FqnKwqQ3FgG31gghQ24qQ24eeQ5EqeQ2BqQ24Q3Bqng17Q3FqQ7CQ5DQ3FQ3F7KKP3hqQ24Q3BQ5D1P,tQ60hQ7C7 "A U.N. report says 5,106 people in Baghdad died violent deaths during July and August, far higher than figures from the city’s morgue."

http://www.alertnet.org/thenews/newsdesk/MAC145456.htm "Partial statistics compiled by the Health Ministry and issued by the Interior Ministry put civilian deaths last month at 1,089..." (for Sep alone)

and so on - do the reading yourself.

Think about how many reports of 'XX' people killed, or 'XX' bodies found. That's generally what we hear, because things are at the point where 'X' (single-digit deaths) are no longer newsworthy. And that's pretty much from just Baghdad, because reporting from Iraq is rather hard, and frequently lethal.

Something to consider: The study states that their investigators asked for death certificates for 545 deaths, and that certificates were produced for 501 of these. Iraq Body Count gives a maximum of confirmed violent deaths due to the occupation of 48,693 at the time I write; and the low end of the study's estimated violent deaths due to the occupation is 426,369. Therefore, if we accept that estimate, the probability that a death certificate has been issued for any given violent death in Iraq is ~11%. The question is: what is the probability, accepting the estimate, that investigators asking for certification of 545 deaths would find it for more than 500 of them?

You can calculate this with the binomial distribution, and I suggest that you do: with a probability of success in one try of 11%, the probability of over 500 successes in 545 tries is less than 10 to the -14th power. The investigators were astoundingly lucky, to find 501 death certificates for 545 deaths -- or, of course, the study's estimate is wrong. I trust you can all see that there are strong, non-partisan reasons to reject the estimate?

"Therefore, if we accept that estimate, the probability that a death certificate has been issued for any given violent death in Iraq is ~11%"

Except that it isn't, because they're completely different measurements. Body Count is a list of 'unarguable' deaths by direct military misadventure witnessed by two or more independent sources, and the Lancet report is an estimate of total deaths from all causes witnessed by friends and family.

As a general rule of thumb: "Any probabilistic intuition by anyone not specifically tutored in probability calculus has a greater than 50 percent chance of being wrong." -- Piattelli-Palmarini. Inevitable Illusions. 1994.

What gets me about the figures being thrown around is the precision. For instance. just upthread a figure of 426369 is quoted. This for an estimate derived from the statistical analysis of survey data revealing 545 deaths.

A difference of 1 death in the survey sample would be 782 deaths in the final result.

This doesn't invalidate the survey, or the calculations. It does look as though somebody used a set of standard processes, but I would suspect that modern epidemiology is full of this false precision in the final results.

False precision is not the same thing as inaccuracy. In this case, it's measuring a mile to the nearest millimetre, instead of the nearest metre. Either figure is still unmistakably different from a furlong. What we have is a huge difference, published in a peer-reviewed journal, and that huge difference, rather than the too-exact numbers, is what damns the current US Administration.

Michael Brazier

So Ulster was internicene warfare badly controlled by an invading foreign army.

Pardon me, pal, but exactly what universe do you live in? Ulster is part of the UK. The Republic wanted no part of the 6 counties. The inhabitants of Ulster have voted time and time again that they want to be citizens of the UK, not the Republic (however much the rest of us in the UK wish they would just sod off to the middle of the Atlantic and settle their feudal spat without involving the rest of us) You know what voting is, don't you? Its when the people put their X in a box to have their say in how they are governed. You might not like the result but in a democracy you live with it, if you're a democrat (small 'd') The army involved was the British Army, including quite a large number of Ulstermen in it. It was not an invading army. It was not an occupying army. It was sent in to try and keep the peace because the (highly partial) police either could not or would not.

There is no parallel between Iraq and Ulster. Any attempt to draw such a parallel undermines the credibility of anything you have to say on either situation.

As for censoring AP, nobody is asking for that. All I'm suggesting is that we are entitled to reasonable standards of journalism - like something approximating to the truth. Nobody expects that from Rupe, which is why the highest paid writers of fiction work for him. So everybody knows Foley is a Republican. How many other 'errors' slip through that you don't know are 'errors' and believe them because they're coming from journalists who we all know are saints and without any agendae of their own.

I'm unamused by the illusions on display here.

First, there's the grotesque statistical ignorance of those commenters who are trying to poke holes in the Lancet piece without recognizing precisely what it's tracking, or comparing it to the Iraq Body Count without taking note of the differences in methodology.

Then there are the interesting political illusions coming to light. Like the idea that embedded journalists will report even-handedly (clue: they're being embedded precisely so that they can be manipulated as adjuncts of the military psyops campaign), that Iraq and Ulster are similar (a better comparison would be post-Tito Yugoslavia), that the existence of quiet patches in, for example, the Kurdish region implies that the violence is confined to a couple of insignificant hot-spots (like, oh, 80% of the country) ...

It's remarkable.

But to summarize: the Lancet report isn't asserting that allied soldiers killed half a million Iraqis. It's simply an epidemiological study that compares the death rate in Iraq today, from all causes, with the death rate prior to the invasion, and demonstrates that lots more Iraqis are dying today than would have been expected by extrapolating the pre-invasion figures. And lots of them seem to have died of gunshot wounds and bombings.

Unless 80% of the death certificates in Iraq are forged, then the death rate is up significantly. Trying to weasel out of responsibility ("it's not our fault, we aren't an occupying power -- look at that puppet government on a stick!") doesn't change the picture. We broke it, and now they're paying with their lives.

"Then there are the interesting political illusions coming to light. Like the idea that embedded journalists will report even-handedly (clue: they're being embedded precisely so that they can be manipulated as adjuncts of the military psyops campaign)"

Mr Stross, you have a great deal to learn about American journalists if you think they are either pro-Bush, pro-military or easily manipulated. They have in additioal to their innate liberal bias an a cynical bias towards bad news (nobody reports a plane that lands safely). They don't ever report the good work done by American troops, that has to be found on blogs.

So if there were 500 or so dead Iraqis each day, every day since this began, they'd report it. The Lancet numbers don't pass the smell test. As for statistical ignorance of the commentators, please go back and read the statisticians quoted by the NY Times.

[sarcasm on] I know the NYT is a notorious right wing rag, but then you probably believe the Guardian to fascist mouthpiece. [sarcasm off]

Daniel, some reading about the history of psychological warfare might help at this point.

The key word in discussing embedded journalists is "embedded". Iraq is too damned dangerous for journalists to wander around without minders. The embedded journalists therefore get chauffeured around and shown what's going on by the US military. They see what they're shown, and they don't get to wander off and see what their minders don't want them to see.

The US military wouldn't be spending a considerable amount of money cosseting and chauffeuring journalists around if they didn't expect to get a return on the investment. Meanwhile, the treatment of non-embedded journalists is very different.

I'm getting my figures from the Lancet, and from a background that includes training in understanding medical statistics and reading epidemiology papers. (It's long enough ago that the memories have cobwebs on them, but I did at one point have a job that involved analysing and producing abstracts of medical reports.) The NYT and the Guardian are both secondary sources at this point. Trust the death certificates.

daniel - I'm not a professional source on this in any sense, but before you start accusing folk of needing to "learn about American journalism" you should perhaps learn some basic mathematics.

"...if there were 500 or so dead Iraqis each day, every day since this began, they'd report it." Try to remember - they'd report it if there were 500 deaths in one go. Lots of little death - not just from violent attacks are less sensational.

And (i have had to skim most of this so fogive me if I'm mistaken) but isn't that 500 deaths a day on average? If you actually go back up and read daniel, you'll note that on some days there were reports of 1000+ deaths in a single day.

It's all in the maths Daniel. You've just said that something is bullshit three times, when you've not pointed at any evidence of your own? Are you sure you can see where the BS really is?

Further to Serraphin's point - the violent bodycount in Baghdad has been running at ~100 day for quite a while now. Mostly it's people getting abducted, tortured, then executed and the corpses dumped, but of course there is the everpresent litany of car bombs and IEDs that don't get reported these days unless something notable happens. This general level of violence makes into enough background commentary about the Iraq situation in the British media outlets for it to be 'common knowledge', for me at least.

Baghdad comprises about 20% of the Iraqi population IIRC (and according to wikipaedia I do). So an estimate for excess mortality that averages to ~500/day for Iraq as a whole doesn't seem so utterly inconceivable that it should be rejected out of hand.

Regards Luke

Charlie, am I to understand tht non-embedded journalists (even Al Jazeera) have been reporting 500 civilian deaths per day all this time? Funny, I don't recall hearing that.

Mr. Lloyd's case is tragic, and the perp should be punished if guilty, but can you provide data showing a pattern of muder of non-embedded journalists by American soldiers as a matter of policy? Your statement that "the treatment is very different" implies such a policy as SOP.

The NYT and others are not secondary sources when quoting staticistians critical of the Lancet methodology.

I happen to be critical of the Lancet results based on historical standards. The Lancet report claims that Iraqi deaths exceed twice the number of German civilians killed by deliberate bombing of high density population centers from 1940 to 1945, including fire bombing of such cities as Dresden by the RAF, using more firepower than has been used in Iraq (the nice thing about smart weapons is that they greatly minimize collateral civilian deaths compared to the big dumb block busters of WWII and can accomlish the same mission with far less explosive power).

I call BS.

Unlike the Rwandan genocide, which left dead and mutilated bodies and skulls in piles for easy counting, physical evidence for 655,000 Iraqi civilians is conspicuous by its absence.

Again, I call BS.

Lancet previous efforts were less than accurate or statistically meaningful. Both efforts have been politically motivated.

Yet again, I call BS.

(I'm now up to 6 BSs in one thread, a new personal best.)

Daniel ...

Journalists do not -- unlike coroners -- have a duty to fill out death certificates and report every sparrowfall in a given country.

Being a journalist in Iraq right now is extraordinarily dangerous, as even a few seconds' research will demonstrate. So most of the coverage takes place at extreme range -- even Al Jazeera are very careful where they send their un-embedded reporters.

The deaths that make it into the news are typically large events that kill tens or more -- car bombings and similar.

Dragging the RAF/USAAF bombing campaign of WW2 in to make a point isn't helping you any; it's an apples-and-oranges comparison. Even brief research into that bombing campaign will tell you the situation is very different indeed: the bombing campaign was carried out with highly inaccurate weapons aiming primarily at factories, rather than guys wearing belts full of souped-up grenades walking into densely-occupied shops and mosques and looking for concentrations of civilians to attack. The suicide bomber is the poor man's smart bomb -- guided all the way.

Finally, the Rwandan genocide killed an estimates million people, give or take, in 100 days. It's hardly surprising that the bodies piled up. The situation in Iraq is estimated by the team reporting in the Lancet to have killed roughly half that number over a period of over 1000 days. So you'd expect less than a twentieth the number of bodies lying around -- assuming that they were left where they fell. (Which they are not.)

You're out by a factor of twenty. About par for the course.

Somebody who I trust (Daniel Davies) contacted both of the statisticians quoted by the NYT by email, and they told him they were either misquoted, or taken out of context. He's currently asking for permission to publish their emails.

FWIW, when I read the NYT article, neither seemed like very plausible quotes. One statistician seemed to quibbling over something quite minor, the other very clearly had been misquoted because the quotation as it stood made no sense.

The NYT did this for the last Lancet report as well.

What gets me about the figures being thrown around is the precision. For instance. just upthread a figure of 426369 is quoted. This for an estimate derived from the statistical analysis of survey data revealing 545 deaths.

A difference of 1 death in the survey sample would be 782 deaths in the final result.

This doesn't invalidate the survey, or the calculations. It does look as though somebody used a set of standard processes, but I would suspect that modern epidemiology is full of this false precision in the final results.

"Please do not assume that I'm accusing Lancet of lying in that. (I understand the long-term importance of maintaining standards.) I'm merely wondering whether they're humanly capable of holding the same standards when someone gives them what they're looking for."

First and foremost it isn't Lancet. It is: 1) The researchers who carried out the report (and who, being academics, will care most of all about their reputation among their peers. Even unconsciously falsifying data is a reputation destroyer). These academics, knowing how the previous report was received, have chosen to be very conservative in their interpretation of the data (its all there in the report, if you choose to read it). They've done their damndest to make the data rigorous. 2) The paper will have been reviewed by four, anonymous, experts. They will be experts in statistics and epidemology. If they think it isn't good enough, they can reject it. If they think there are flaws, then ask for them to be addressed. For a journal of the standard of Lancet (think Nature), this will be extremely rigorous. These reviewers will not have the same attachment to the data, that the researchers will have.

Secondly there are a number of procedures followed by statisticians and scientists designed to minimise the chance of you finding what you're looking for. its something that all scientists are aware of, and so you're taught to be skeptical of results that seem to be too good. Science is a far more rigorous process than you give it credit for - especially in elite circles (which this most certainly is).

"A higher death toll only makes each side look like victims, and that's especially important for the worst of the insurgents. Shia deaths are attributed to Sunni, and Sunni to Shia."

This argument makes no sense

"After Dresden, German propagandists felt free to exaggerate the bombings' body counts."

Unless you're suggesting those who compiled the data were propogandists for al-Sadr (unlikely, given that they were Iraqi doctors), this is irrelivant.

"That's a fair point as long as the guys with Hezbollah banners and t-shirts understand that Hezbollah is doing things in their name."

Hezbollah's doing nothing in their name. Hezbollah's doing things in the name of people who vote/support them in Lebannon. I don't think there's any great doubt about this. I think that you could make a case if you were talking about supporters of Israel.

"In the U.S., the isolationist-right opposed the war, and the extreme-right (such as David Duke) is even more vehemently against it. They see this as a war for Israel. Duke is pretty popular in some Arab countries."

This is an ad-homenim, by the way.

Mr. Duffy's rhetorical armamentarium is reminiscent of Nazi-sympathizer holocaust denier bushwa.

The statistic in question is the crude death rate, and a comparison of crude death rates in Iraq in 2002 and 2006. Unfortunately, because of the inadequacy of Iraqi provisional government record keeping (they didn't tabulate vital statistics after the March 2003 invasion), estimates for the value of the crude death rate for the later date must be obtained through the analysis of covariant proxy variables.

One way to tell the distribution of causes of death in the population of people who die is to sample the death certificates of a sufficiently large number of those people for categorical data on the frequency of various types of causes of death. For purposes of the study, we can partition causes of death into two categories: violent death and not violent death.

Then, we can calculate the sampling population necessary to estimate the population parameters:

N(0)=t^2*(p)(q) ---------- d^2 where

N(0) is the size of the sample

t is the z-score corresponding to the size (fraction of 1) of the upper tail and lower tails of the distribution of potential results from the sampling - the parts of the data outside of our confidence interval. The smaller we make the tails, the more likely our results derive from between the upper and lower bounds of our confidence interval.

For a 95% confidence interval, the upper and lower tails (assuming normal distribution) contain 2.5% each of the potential distribution, and, Student proved that 95% of all estimates at lie between -1.96 standard deviations and +1.96 standard deviations from the mean.

p is the standard deviation of the probability of violent death. Assumed to be .5

q is the the standard deviation of the probability of not violent death. Assumed to be .5 (1-p)

d is the error level acceptable. We like to estimate within 5% of the unknown true value.

So,

N(0) = (1.96)^2*(.5)(.5) ----------------- = 384 (.05)^2

To estimate to a 2.5% MoE, we find the sample size to be 1560.

The Lancet paper sampled 1800 families. Sounds about right.

To be continued.....

I just want to point out that all of the discussion here and elsewhere really does justify the AP's use of the word "controversial". AP doesn't seem biased so much as doing a good job of predicting that a large number of people would harshly dispute the results of the study.

I'd like to make a further prediction of my own: no matter how good the methods of the people doing the study, one side of the debate will automatically dismiss their study and anyone who quotes it as a baised loon. :)

And if you're thinking that something that follows the scientific method and well established methods of peer review can't be controversial, look at the century-long controversy over evolution in the US... (or EVILUTION as one friend used to call it).

"Body Count is a list of 'unarguable' deaths by direct military misadventure witnessed by two or more independent sources, and the Lancet report is an estimate of total deaths from all causes witnessed by friends and family."

The Lancet study estimates violent deaths, as well as deaths from all causes, and I used the "violent deaths" estimate. And Iraq Body Count lists every "civilian" (meaning non-Coalition) violent death in Iraq -- not just the Coalition's direct kills, but the terrorists' and militias' kills as well. I think those are comparable. If you disagree, I suggest you search the study's tables for an estimate that is comparable to Iraq Body Count's list, and run the same calculation I did.

I'll just mention that, if you want better than a 5% chance of over 500 successes in 547 trials, the chance of success in one trial has to be more than 90%.

Daniel Duffy, you "call BS" six times -- but apparently you haven't understood the material you're attempting to analyze. The Lancet study is not estimating violent deaths. It is not even estimating deaths due to military, or insurgent, or coalition, action. It is estimating total deaths.

By then comparing the total death rate to the total death rate of other periods of Iraqi history, some interesting questions come up.

Because the study measures total deaths, comparing it to the violent deaths in Germany during the bombing campaigns of WWII is irrelevant.

When you destroy transportation, sewage handling, clean water supply, and power, you increase disease, deaths from accident, deaths from taking too long to reach medical help, deaths assisted by malnutrition and dehydration, deaths due to heart trouble and strokes among old people in a climate that frequently reaches 120 degrees fahrenheit. I find it entirely credible that the overall death rate has risen immensely during our occupation.

A "smell test" is an extremely poor way to evaluate statistics. If all you've got to go on is your nose, stop wasting our time.

NYC Math Teacher, why are you running a calculation for the distribution of causes of death in Iraq? The Lancet study's estimate is of the absolute number of deaths. The authors weren't trying to establish odds that, given a death in Iraq, the death was violent. The sample size you've computed, 1560, is the number of deaths you'd have to examine to be 95% confident that the fraction of violent deaths in the sample is within 5% of the same fraction in the whole population. That doesn't have anything to do with the number of households sampled. A sample of households in Iraq measures the population of all households in Iraq, not the population of all deaths there.

Predictably, no-one in this debate has succeeded in moving anyone else so much as a nanometer from their initial position. Now that we've all vented some our spleen in the attempt, I'd like to lighten the atmosphere a little. Here's a link to a picture of the bumper stickers I recently put on my car, in an attempt to make my position clear in a different way.

Charlie, you'll understand when I say that I was thinking of you when I got these stickers.

There seems to be a bit of an argument here about whether the estimated 500 deaths a day is contradicted by the Iraq Body Count's approximately 50 deaths per day reported in the media.

The Iraq Body Count FAQ says "We ... do not update the data base [with new casualty figures] until we have located and cross-checked two or more independent approved news sources".

So, a question: does anyone know what the ratio between recorded deaths and deaths reported by independent and credible news sources is in the UK, say, or the US?

The UK's death rate, for example, is listed in the World Factbook as 10.18 per 1000 people. From the same source, the UK's population is 60,441,457, giving us 615,289 deaths per year. Does anyone know, or know how to find out, what percentage of those deaths were reported in two or more media sources?

Intuitively I'd tend to believe that less than 10% of all deaths are so reported. That would tend to correlate with the difference between the Lancet's figures (600,000 approx) and those of the Iraq Body Count (50,000 approx).

However, I could be entirely wrong.

Also, several commentators here comment that Iraqi blogs show a different picture of the country to that which is portrayed in the mass media. I think it might be very helpful to the discussion if you could give some examples of these blogs, as I'm not immediately familiar with them, and I think that may hold true of quite a few people reading here.

"...the bombing campaign was carried out with highly inaccurate weapons aiming primarily at factories, rather than guys wearing belts full of souped-up grenades walking into densely-occupied shops and mosques and looking for concentrations of civilians to attack. The suicide bomber is the poor man's smart bomb -- guided all the way."

You aren't making any sense Charlie. By your own description, the Allied bombing of Germany with its massive clumsy bombs should have killed far more civilians than your precise, "poor man's" smart bombs.

And if the study is not distingusihing violent from all other deaths why are you trotting it out with such fanfare as evidence of American "genocide"? Isn't that rather dishonest of you?

Michael - did the Lancet study extrapolate across all of Iraq including the peaceful 14 of 18 provinces especially the very peaceful Kurdish areas? If so, the same "methods can be used to show that Saddam had 600,000 WMDs (http://disturbinglyyellow.org/2006/10/11/lancet-600000-weapons-of-mass-destruction-in-iraq/):

"By extrapolating the fact that Hans Blix reported1 50 Al Samoud-2 missiles, 331 Al Samoud-2 engines, 70 deployed Al Samoud-2 warheads, 75 deployed Al Samoud-2 missiles, 40 warheads in manufacture, 32 warheads in training, 2 long range propellant casting chambers, 14 155mm chemically filled shells, 49 liters of 97% purity mustard gas, 500 ml of thiodiglycol chemical precursor, 18 122mm chemical warheads, 224.6 kg of growth media, and 40 vials of toxin standards, in Iraq just prior to the Coalition invasion, the Lancet estimates that Saddam Hussein indeed had 600,000 units of WMD at his disposal, by multiplying those findings by thousands."

Yeah I know its horribly unfair, but it is funny.

Andrew G, the controversy about evolution isnt about teh actual scientific results, its about some peoples reaction to them. Which is probably your point, but I felt I had to make it clear.
manages to stop going on a rant about ID/ creationism

Having had a quick read of the article itself, I can't identify any real problems with the numbers themselves. What I disagree with is your conclusion:

Your job, if you're voting in the upcoming election, is to decide whether you want to let a politician who cold-bloodedly ordered 655,000 murders in order to win his last mid-term election get away with the same trick twice, on behalf of his page-buggering, bribe-taking buddies.

To claim Bush "cold-bloodedly ordered" deaths caused by anti-democratic factions in Iraq (including those run by Iran and Al-Qaida) who incite violence, disrupt social order and retard the reconstruction of Iraq is hyberbole in the highest extreme.

The more worrying result is the upward trend in deaths. The question therefore is which party has the more coherent policy for solving this problem. And no, I don't know the answer since I haven't studied the exact positions of each side. NB, I'm a foreigner too, so I'm also easily ignorable.

Page-buggering is nothing new on both sides of the house so I hardly see how that is relevant.

Bruce: yeah, I dig the stickers. (Nice motor -- I drive an 850 :) On the subject of your initial sentence, that's precisely why I ration my posting of political pieces to a drip-feed.

Daniel: you missed my point completely. Which isn't exactly surprising given the previous direction of this thread ...

For an encore, I shall have to post my thoughts on creationism and the anti-womenabortion movement.

Cian,

"First and foremost it isn't Lancet.

I understand that. We've been referring to it as The Lancet's study since the first post. I think it's fair, as the report is trading on the Lancet's reputation.

I'm sure they do want the study to be rigorous. I have no doubt that the i's are dotted, the t's are crossed, and that all the math adds up.

If I may quibble, I was startled by this line at the top of page 7: "The circumstances of a number of deaths from gunshots suggest assassinations or executions. Coalition forces have been reported as targeting all men of military age.27,28"

This was in order to explain why the deaths tended to be men. But references 26 and 27 are based entirely on one incident where the Army charged four soldiers with murder of civilians. The soldiers claimed they were acting under orders. It's still in court, but even if true I see it as nothing short of moronic to make that claim given the context of this report. It's as though they've learned everything they know about the military from reading graphic novels.

None of that really affects the raw data, however, as it's in their "Discussion" section.

I am now more wary of the reliability of the survey teams. As I read it, the survey teams are the only ones who've seen the death certificates, and the protocol forbids them from making copies. Everything depends on the survey teams.

"Unless you're suggesting those who compiled the data were propogandists for al-Sadr (unlikely, given that they were Iraqi doctors), this is irrelivant."

Yes, I know they're doctors. No, they don't really have to be propagandists for al-Sadr in order for me to make this point.

Doctors are still human. I haven't forgotten the circumstances they're living under. They're subject to local pressures and prejudices, not to mention a horrific living environment for themselves and their kids. 87% of Iraqis want a timeline for withdrawal. 47% approve of attacks on U.S. forces (88% of Sunnis; 41% of Shia). That's the soup they're living in, and now they have an opportunity to toss a wrench into a U.S. election.

Why are Iraqi doctors immune? If you think they are then we'll just have to disagree.

"Hezbollah's doing nothing in their name. Hezbollah's doing things in the name of people who vote/support them in Lebannon. I don't think there's any great doubt about this. I think that you could make a case if you were talking about supporters of Israel."

I don't disagree with you wrt non-Israelis who support Israel. How would that not apply to non-Lebanese who support Hezbollah?

"This is an ad-homenim, by the way."

It might be if my intent had been to peg as an isolationist or a member of the extreme right. Instead, I'm only mentioning their position on the war while relating it to a hypothetical.

I'll add a point for those who take the report's conclusions seriously. The number of embedded journalists nowadays is often in the single digits. The lack of such reporting is not something I would use to dispute this report.

Embedded reporters weren't that good at clearing anti-American stories anyway.

Since moving people's positions in the debate now seems to be ruled out, I'll just say how pleased I am to see someone doing statistical analysis in a comments thread. So many times I've been tempted to do the same thing, but never quite had enough motivation to drag out the textbooks to check I'd done it right.

Also: something's gone wrong in a country if the government isn't counting dead bodies.

Hugh Hancock: "The UK's death rate, for example, is listed in the World Factbook as 10.18 per 1000 people. From the same source, the UK's population is 60,441,457, giving us 615,289 deaths per year. Does anyone know, or know how to find out, what percentage of those deaths were reported in two or more media sources? Intuitively I'd tend to believe that less than 10% of all deaths are so reported."

Would you believe that less than 10% of all murders in the UK are reported in two or more media outlets? The study does estimate violent deaths, and that estimate is over 600,000.

Daniel Duffy: Charlie linked to the study in his post, and I followed the link; if you do the same, you'll know as much as I do.

Neil Willcox: if the government isn't counting dead bodies, everything's gone wrong. That was Charlie's original argument: "this study proves the Iraqi government isn't counting the dead, therefore Iraq has collapsed!" That's also why direct testimony from Iraqis to the effect that Iraq has not collapsed is relevant.

Bruce Cohen: that Elder God has been running for the Presidency for more than 20 years, and the stars have never been right yet! Let's have the Unspeakable run instead, and give old squid-head a rest.

The more worrying result is the upward trend in deaths. The question therefore is which party has the more coherent policy for solving this problem. And no, I don't know the answer since I haven't studied the exact positions of each side. NB, I'm a foreigner too, so I'm also easily ignorable.

There are a number of people who are unhappy with the war, but are afraid that if the Coalition pulls out things will be even worse.

Right now I'm of the opinion that if the US and UK left, the killings would expand into full blown genocide. Then there would be even more trouble finding out how many died because there wouldn't be any Sunni Arabs left to ask.

There are only two ways out -- a decades long occupation until things calm down a little, or spliting the country. Neither the Shiite nor the Kurds are willing to let the Sunni rule the country again, and the Sunni are too afraid not to.

Michael Brazier asks:

Would you believe that less than 10% of all murders in the UK are reported in two or more media outlets?

Yes (depending of course on what you characterise as a "media outlet"). I see the figure for the recording year 2005-2005 was 859 homicides in England and Wales; that's a bit more than two a day. I don't assidously follow that sort of story, but my impression is that the number reported in the papers is more like one or two a week. No doubt all of the homicides were reported in the local small press, but very few will have reached nationally distributed media.

Michael, my own experience supports the idea that deaths in the UK mostly wouldn't qualify under the Iraq Death Count standard. I've lived mmost of my life in the middle of a triangle on the fringes of the distribution zones of four local newspapers. It's unusual for a death to be reported in more than one of them, even violent deaths, and for a death to be reported by some other source would make it big news, if it wasn't somebody of sufficient status to get an obituary in a national newspaper.

And, I wonder, does Ansible count as a news source? Or an alumni newsletter (lousy Ammericanisms, but I can't think of a similar label in British English).

No Randy, it is an ad-homenim. You said:

"Wouldn't you take comfort if the far-right (most of whom actually agree with you, BTW) were to demonstrate against your position?"

which suggests that there might be something wrong with a particular position because members of the far right demonstrated against it. Classic ad-homenim.

"Doctors are still human. I haven't forgotten the circumstances they're living under. They're subject to local pressures and prejudices, not to mention a horrific living environment for themselves and their kids. 87% of Iraqis want a timeline for withdrawal. 47% approve of attacks on U.S. forces (88% of Sunnis; 41% of Shia). That's the soup they're living in, and now they have an opportunity to toss a wrench into a U.S. election."

You are quoting verbatim statistics which were compiled far less rigorously about what Iraqis think (which again is a far harder datum to gather)? No doubt about these statistics, propoganda, influence, lies? Who gathered these statistics? Do you know? Were they gathered by experienced surveyors, as the Lancet statistics were? Or by people who had received a day's training and some danger money? Do you know, do you care? Do you understand why its hard to take your skepticism about the Lancet report terribly seriously?

And your theory that this might be their attempt to throw a spanner in the election is incredibly solipistic. The world does not revolve around the USA. Its just possible that being doctors (who may not have been Iraqi, btw), and not the cheesy political hacks you believe them to be, they think its important that the statistics about whose dying, and of what, is important. It's just possible that this is why they risked their lives going all over Iraq (which is probably the most dangerous country in the world at the moment by a considerable margin), just to influence a US election for congress? Have you any idea how ludicrous this sounds? When your reason for skepticism is that you believe that the survey was carried out by Arab doctors working for the Democratic party, its probably time to reevaluate your position.

I'll add a point for those who take the report's conclusions seriously. The number of embedded journalists nowadays is often in the single digits. The lack of such reporting is not something I would use to dispute this report.

There are almost no western journalists going out of the Green Zone these days, and even Arab journalists are not ranging very far (Patrick Cockburn from the Independent was one who did, and his stories were horrific). Journalists rely on reports from the Iraqi government, independent stringers and the US army. Journalists are providing a tiny, tiny, part of the picture now.

A side note on reporting:

A few weeks ago, there was a story, in the British media, from the British Army in Afghanistan, and the pictures were reported as being video shot by a soldier on his mobile phone.

I didn't hear of there being any big fuss about the footage, though it's possible some soldier got told not to do that again. I didn't see anything that seemed any different from the combat newsreel footage shot in WW2.

Michael Brazier, I think I may have put my point too concisely. I meant: if the figures from the Lancet are having to be determined by sample methods rather than full figures being reported by the Iraqi Statistical Office, then a fairly basic piece of information gathering isn't taking place. Either the Iraqi government has better things to do (for example is in a crisis and so information for long term planning has been placed low down the priorities) or is incompetent. This does not necessarily imply the crisis has become a collapse.

On a more general note, even if the Geneva Conventions don't explictly require records of deaths to be kept, I'd got the impression that any occupying force took responsibility for government of an occupied area. If America wants to be seen as a beacon of liberty does "We've complied with the minimum requirements of international law" sound like a good rallying cry?

Neil: The Iraqi statistical office isn't collecting full figures because their staff can't travel freely throughout the country and collate statistics. For some more coverage of the false controversy, see this Christian Science Monitor piece, or just look up Juan Cole's or Brad DeLong's blogs from my sidebar. Or Crooked Timber. Or indeed anyone else who's statistically literate.

Meanwhile, Zeyad of "Healing Iraq" does an in-depth number on the study here, culminating in this zinger:

One problem is that the people dismissing – or in some cases, rabidly attacking – the results of this study, including governmental officials who, arguably, have an interest in doing so, have offered no other alternative or not even a counter estimate. This is called denial. When you have no hard facts to discredit a scientific study, or worse, if you are forced to resort to absurd arguments, such as “the Iraqis are lying,��? or “they interviewed insurgents,��? or “the timing to publish this study was to affect American elections,��? or "I don't like the results and they don't fit into my world view, therefore they have to be false," it is better for you to just shut up. From the short time I have been here, I am realising that some Americans have a hard time accepting facts that fly against their political persuasions.

Now I am aware that the study is being used here by both sides of the argument in the context of domestic American politics, and that pains me. As if it is different for Iraqis whether 50,000 Iraqis were killed as a result of the war or 600,000. The bottom line is that there is a steady increase in civilian deaths, that the health system is rapidly deteriorating, and that things are clearly not going in the right direction. The people who conducted the survey should be commended for attempting to find out, with the limited methods they had available. On the other hand, the people who are attacking them come across as indifferent to the suffering of Iraqis, especially when they have made no obvious effort to provide a more accurate body count. In fact, it looks like they are reluctant to do this.

By the way, these same statistical methods were used to count civilians deaths in Darfur, but then I didn’t see anyone objecting to that.

This doesn't really prove anything to do with the reports, but I reccomend going to YouTube and searching for things liek "IED" "Baghdad" and "Iraq firefight". Any sort of Iraq & combat related keywords you can think of. It's pretty eye opening.

I can post some links here if people want.

Another set of analyses, "Following Up on the Lancet Study" and "The Iraqi Death Tally Study" at the ultra-geeky but meticulous site:

Good Math, Bad Math:

"Finding the Fun in Good Math, Squashing Bad Math and the Fools who Promote It."

Cian,

"You are quoting verbatim statistics which were compiled far less rigorously about what Iraqis think (which again is a far harder datum to gather)?"

If you look at reference 2 of the Lancet study, you'll see they found the Brookings Iraq Index worthy of use. That's where I got those numbers.

"Or by people who had received a day's training and some danger money?"

As opposed to the extensive 2-day training session the Lancet surveyers got?

I don't really see that that matters anyway. My only point in using that source was to say that conditions in Iraq are miserable, and I think we both agree that they are.

"And your theory that this might be their attempt to throw a spanner in the election is incredibly solipistic. The world does not revolve around the USA."

We're not talking about the world's opinion of the U.S. but I didn't think there was any doubt that the U.S. position in Iraq was important to Iraqis right now, regardless of whether they're hostile or not.

That said, world opinion isn't exactly uninterested in this report, and world opinion matters.

I'm still curious about this:

"Hezbollah's doing nothing in their name. Hezbollah's doing things in the name of people who vote/support them in Lebannon. I don't think there's any great doubt about this. I think that you could make a case if you were talking about supporters of Israel."

I had agreed with you on that last point. Americans who support Israel must certainly understand that they're (at the very least) a part of its support structure. Among other things, they allow Israeli voters to see that they're not alone when they vote for the leaders of a particular policy.

By the same token, Lebanese citizens can see that Hezbollah has the respect of all kinds of citizens in the western world, and that throwing the Geneva Conventions into the trash comes with little cost, if any at all. I see this formula as the same, and that's why I don't understand your reasoning that they're somehow different.

The same can be said about the Al-Sadr's Mahdi Army, who are surely watching their distant cousins in Hezbollah (or maybe it's Hezbollah watching Al-Sadr). It's germane to this discussion when you consider Charlie's post today re the excerpt from HealingIraq. The important thing isn't really about the report's accuracy. As I said a couple times before, 1/10th that number is horrific enough. When you look at the numbers close, you'll see that a huge proportion can be attributed to the insurgents' wholesale disregard for the laws of war.

In that sense, the dismal lack of outrage over (and some might say outright support for) such behavior is serious.

We've become a world where one side is being asked to comply with the Geneva Conventions, not just to the letter, but beyond even that. And the other side can ignore it all with no shame whatsoever. I could tell you what the inevitable results would be, but I don't have to. Just read the Lancet's study.

The Lancet surveyors had experience of doing survey research for academic purposes. Its there in the report. They were pros.

As I understand it your argument is:

The surveyors almost certainly invented the data They did this because:

They wanted the US to pull out of Iraq because they are Sadrists, or because conditions are really really bad (which rather supports the Lancet statistics, but never mind), or they hate the US, or something.

Anyway, its impossible for any Iraqi to compile accurate data, because they're so biased - and they're biased because the situation in Iraq is really, really bad and people are dying, security has collapsed. So the worse the situation in Iraq gets, the less we can believe statistics compiled there, and so obviously the lower the death count can't be that high.

Oh, and they're planning to influence the US congressional election, through a Lancet study, by making the Iraqi casualty rate look really really bad. Obviously they could have just come up with a really high number of people killed by Americans, but that would piss the Sadrists off, as they're macho hombres. So instead they made them the same number, and then threw another third called "unknown", just to cover all their bases (in case there were another bunch of murdering scum bags, who got pissed off that their ability to slaughter people in large numbers wasn't being acknowledged).

If you really believe that, then I don't think I can convince you otherwise. Not sure you'll find many takers though - its a little baroque.

I'll vote...but I consider it almost a waste of effort. Our county went with the Diebold voting machines. The ones that have been proven so often to be so easy to corrupt tracelessly, and yet which have been so frequently corrupted in a blatantly obvious manner that it's clear they don't CARE that the election is known to be a fraud.

Randy: occupation duty in a foreign country sucks. Doubtless it sucked to be a soldier in the Wehrmacht in France in 1942, when those lousy resistance fighters were showing such disregard for the Geneva conventions. Nevertheless, "but they're doing it, so it must be okay!" isn't a defense that stands up in court.

The allied forces in Iraq have backed themselves into exactly the position of every other occupying imperialist power through history. Under which circumstances I find myself compelled to agree with this summary (extracted from "Discourse on Colonialism" by Aimé Césaire).

Cian,

There are two facets to my reasoning:

"They wanted the US to pull out of Iraq because they are Sadrists, or because conditions are really really bad (which rather supports the Lancet statistics, but never mind), or they hate the US, or something."

Not really but close enough.

The fact that they're doctors doesn't impress me as meaning that they're above all that.

Nor am I impressed with those who made the report. I was appalled by those footnotes in the commentary I mentioned earlier.

"So the worse the situation in Iraq gets, the less we can believe statistics compiled there, and so obviously the lower the death count can't be that high."

That's another part, but it's not where I fault them. To the extent that this could be true I commend them for doing the best that they can.

One problem noted elsewhere was that for the numbers to work, given that these trends aren't straight lines, there could have been days when as many as 10,000 people were killed. I don't think polls could be taken in that kind of environment.

"Oh, and they're planning to influence the US congressional election, through a Lancet study, by making the Iraqi casualty rate look really really bad."

As bad as that sounds, you might want to remember that a terrorist's stock in trade is to terrorize. A poll like this becomes a multiplier. That makes these polls a very important target.

I'll acknowledge that it's certainly possible that the poll is accurate, or close enough, or simply better than what we've had before. If you look back to my first post, you'll see that it wasn't even a major part of my reason for posting.

The most important part to me was buried in this line: "For all the noise about the Geneva Conventions, it appears to me that the U.S. and its allies are the only ones who do care about them."

And that brings me back to the question you hadn't answered: Why do people who support Hezbollah not have to answer for the consequences, especially now that the Lancet has shown us the results?

Hezbollah are a Lebanese political movement and militia who came into being during the brutal Israeli occupation of South Lebannon. So that's probably why.

Charlie,

"Nevertheless, "but they're doing it, so it must be okay!" isn't a defense that stands up in court."

Perhaps you've got me mixed up with some other wascally right-winger. I am not saying that U.S. forces don't need to comply with the Geneva Conventions.

I am the one who's saying everyone should be held to it.

My previous comments were only that we should read the entire document as a contract and be able to use any loopholes as the writers in 1949 fully intended. They put all that text in there for a reason.

If you read the now famous common Article 3 (which doesn't apply to Iraq, but it's a good example anyway), you'll see that there are openings for the parties to make special agreements if they don't like those loopholes.

"The allied forces in Iraq have backed themselves into exactly the position of every other occupying imperialist power through history."

I disagree. Looking at the numbers, most of the deaths are caused by the insurgency, or the various factions at war with one another. It would be one thing if there was peace in Iraq and we didn't leave. We even rushed the issue of sovereignty and elections.

BTW: I've only read the first pages of "Why Orwell Matters" by Hitchens, as found in Amazon's sample pages, but he makes some of the same points about occupation.

Cian,

If you think so then maybe you'd better read Charlie's latest post again, where he said,

Nevertheless, "but they're doing it, so it must be okay!" isn't a defense that stands up in court.

Charlie,

Correction: "which doesn't apply to Iraq, but it's a good example anyway"

IANAL

I dunno if it could apply in Iraq in some ways. I thought common Article 3 was originally supposed to apply to civil wars, and I realize that perhaps now it might -- assuming anyone cared.

Hezbollah are a Lebanese organisation, based in Lebanon. They are not in Iraq. They are not fighting in Iraq.

1/3 of the violent deaths are the fault of the coalition 1/3 of the violent deaths are blamed on some kind of Iraqi (could be criminals, terrorists, or resistance fighters - which are different things. An Iraqi soldier killed by a resistance fighter is a legitimate target, legally). 1/3 of the violent deaths are assigned to unknown. So they could be anyone.

You are making too many assumptions about the figures.

There are also the excess deaths from the other causes, which are essentially the fault of the US as the occupying power.

Cian,

"Hezbollah are a Lebanese organisation, based in Lebanon. They are not in Iraq. They are not fighting in Iraq."

No, but no one who supports Hezbollah can ever claim to care about the Geneva Conventions. The types of numbers issued in this report are what they must come to accept as a natural consequence of the means they chose to fight their war.

"1/3 of the violent deaths are the fault of the coalition"

Not necessarily. If someone fights from within civilian housing or business areas then they're the actual cause of those deaths. The Geneva Conventions most certainly do allow legitimate forces to attack those areas under certain conditions. The fault is that of the insurgents.

"1/3 of the violent deaths are blamed on some kind of Iraqi (could be criminals, terrorists, or resistance fighters - which are different things. An Iraqi soldier killed by a resistance fighter is a legitimate target, legally)."

"1/3 of the violent deaths are assigned to unknown. So they could be anyone."

Killing the police only made those numbers rise. It may be a legitimate action, but the fault is theirs.

"There are also the excess deaths from the other causes, which are essentially the fault of the US as the occupying power.

As mentioned by someone else earlier, a lot of deaths were attributed to destruction of infrastructure. What I'll add is that to prevent rebuilding causes those deaths.

If you want to say that Saddam was the legitimate government and should be reinstalled, or that Al-Sadr's army represents the truly legitimate aspirations of the Iraqi people, then that's one thing. But we should at least acknowledge whether or not the laws of war should be respected. You can't say yes in one place, and no in another.

Or, if you think they can, then please say that. I'd like to see someone make that argument.

Charlie, I was understating too much; as well as being invisible to the Iraqi government, obviously the Iraqi Statistical Office will have been under the same constraints as everyone else in Iraq.

And I'd forgotten entirely about Good Math, Bad Math, which has plenty of statistics going on in comments threads.

Best to ignore every thing I've said on these topics, really.

The cross check for deaths is the number of new graves in cemetaries. Sure, the death squads may be putting a few thousands of people in unmarked graves but figure on 90% of the dead being buried in individual graves with family members wailing around them, every day for the last four years. Stross Russia was invaded four times in thirty years, in 1914 (Germany, Austria-Hungary, and Turkey), in 1918, Japan, America, Britain, France, Poland, Turkey again, and Rumania), in 1936 (Japan), and in 1941 (Germany, Rumania, Italy, Hungary, Slovakia, and assorted volunteers from the rest of Europe. I don't count Finland because they never went past their prewar borders). Cirby Germany lost more than 650,000 deaths to World War II bombing, counting increased deaths to stress as well as directly blown apart, smothered in bunkers, and burned alive, and far more than 6,500,000 deaths after the war to the kind of ethnic cleansing we are seeing in Iraq. Incidentally, heart attack deaths go up when you can't get medicine, air conditioning, and away from stress in general.

As far as "good news" from Iraq goes: there isn't any. We are building schools that no one can go to, hospitals that have no doctors, and anything else we build or repair becomes an instant target for someone. What I cannot believe is that our military (America's) has not made better plans or procedures for dealing with a battlefield like this. Urban warfare is inherently different than open field. I can only conclude that the errors come from the political powers that tell our military what to do. Donald Rumsfeld should resign. Then Bush should beg Colin Powell, a "real" soldier, to replace him or to help to pick a replacement.

And Tracskul, everyone we train as a police or military man in Iraq has an automatic target placed on his and his family's back. We want them to replace us, but they become targets by even allying with us. That sucks.

Just saying. People here are arguing about theoretical things, being on the ground there is a way different beast. I believe the Lancet figures because I know how peer-edited writing happens. If someone thought they were off mark, it would not have been published. These days, unfortunately, I am also willing to believe that if an experimental result is controversial, it would be kept under wraps too, especiallyi if it happens to be an American scientific publication. Because us criticising our Dear Leader is treasonous, y'know.

(no, I'm not a bitter American, no I'm not...I'm just tired of the crap.)

fracskul: Congratulations, you've lured me out of my retreat. Maybe you were being sarcastic when using the quotes in "Then Bush should beg Colin Powell, a "real" soldier, to replace him or to help to pick a replacement." I certainly hope so. I've never met Powell, or served under him, but my view of his statements and sctions over the last 15 years or so lead me to the opinion that he's not a real soldier, and not even much of a commanding general. By all I've seen he's what's called a "political" general, one who is more concerned with pressure from above than loyalty from below. Sure he's making noises of protest against Bush policies now, but he was willing to be Bush's stooge in front of the UN in 2002.

Fracskul & Bruce: Colin Powell appears to have taken the post-invasion occupation seriously enough that the State Department spent some tens of millions of dollars preparing detailed plans for how to occupy and administer Iraq ... only to have them torn up and thrown away by Rumsfeld's Pentagon, with backing from the usual PNAC crowd.

Presumably because plans for an occupation might have contradicted the party line that it was going to be a short, victorious war.

Desperately seeking salvation from a Man On A White Horse is usually the penultimate step on the road to defeat (conservative lane).

Randy: I'll take your swithering about Hezbollah -- a crude attempt to change the subject -- as a sign that your willing suspension of disbelief in the Bush administration's confabulation is cracking. Thanks, my work here is done ;-)

Charlie,

Don't take too much comfort in that. I could have made a similar post four years ago.

It's really the same subject. You're the one who brought up the Geneva Conventions, after all.

Regardless of the study and what any of us think, I don't think there's much chance of things changing in Iraq until after 2008.

It's sure to be a big issue in the 2008 election. Right now I'm hoping for either Edwards or Clark.

Sermon, meet choir.

Forget the Geneva Conventions.

The fucking idiots haven't even read Machiavelli, or they would, with great show of horror, have already tried the senior officers responsible for Abu Ghraib.

And it was an ancient trick when old Nick described it.

Dave,

The rule of law still applies in the military. They can't try the senior officers at Abu Ghraib without charging them with actual crimes. They can be disciplined in other ways, and many have.

If you still think there is a crime they personallly are guilty of then maybe you'd better take this quiz.

Yes, it's a pro-military website but I think you can bear it without getting the vapors. (You can hold your breath for ten questions.) Just skip the intro and scroll down to the questions to see how many you can answer.

The answers are below that, and they do link to references at real news sites.

I'm not saying it will change your mind, but at least you'll know that there is another point of view.

My reaction:

http://willscommonplacebook.blogspot.com/2006/10/more-than-600000-killed-in-iraq.html

Re: Abu Ghraib and rule of law.

Senior officers are responsible for what their subordinates do. That's why they spring surprise inspections, turn up with a bunch of aides and go through a camp or base or prison with a fine toothcomb, ripping new assholes en masse.

In the case of the US Army's rape and torture facilities in Iraq, either experienced senior officers didn't do their jobs in the singular case of Abu Ghraib or they turned a blind eye to what the rumour mill was saying about that hellhole. Either way those officers don't deserve their stars and a Court of Inquiry was the least they could expect. The UCMJ has, as I recall, a load of prosecutable offences for officers who fail in their duty to command subordinates properly.

Under this Republican administration instead, the buck stopped over in that bunch of E-4s as the Os covered each other's asses like tinted toilet tissue.

Robert,

Not exactly. There is at least one Lt. Colonel facing charges.

From the point of view of Niccolo Machiavelli, the important point is that, whether or not you put the senior officer there with the intent that he behaves that way, once you are rid of your enemies, you are very public about the trial and punishment of that senior officer.

If you were pulling that sort of trick with Abu Ghraib, you would want the shit to hit the fan.

The idea is that as soon as the brutality is officially brought to his attention, the Prince, however faked his shock at the revelation, is seen to put things right, protecting his new subjects against the depradations of his old subjects.

Machiavelli didn't use the phrase "hearts and minds", but he easily could have done. He distinguished between fear and hatred as motives, and urges that a Prince should not provoke hatred.

The Prince is on the web, and it isn't a long book. In this particular context, Chapter XIX would particularly apply.

Re: Bird Colonel on charges -- the responsibility for US military prisons in Iraq (which Abu Ghraib was at the time) is a star-rank position. A general was in charge. He was responsible for the torture, prisoner abuse and murder that happened under his command. He should have known; if he didn't know that's dereliction of duty. If he did know and let it pass then that's just as bad. If he was ordered to turn a blind eye to the abuse of Iraqis and others and did so, that's obeying an illegal order which is still an offence under the UCMJ.

Buck-passing in the US military wasn't always as blatant as it seems to be nowadays; that might just be the brass aligning their ethics with the current Administration who blame everybody else for events that happen when they're in charge.

One more thing: if there is any point of relevance between Northern Ireland in the bad old days and Iraq today, it's that partition of Iraq would be an even worse idea than partition of Ireland was.

there may or may not be an issue. Unfortunately, given the study's methodology, we DON'T KNOW. Look at the confidence brackets, for starters. Then read criticism of their methodology.

Contravercial dosn't even BEGIN to cover it, remembering that their "baseline" figure for civilian deaths is ALSO not universally agreed upon, at all!

Moreover, it's much the SAME flaws as the previous study. This looks to me as nothing more than political bias*, and I'd really like a properly conducted study.

This is deliberately NOT a comment on anything else.

(*Peer-reviewed only works when those peers don't mostly have the same strong expressed opinion, y'know)

Here's the raw data (via Brad DeLong)

"The Lancet study of deaths in Iraq. 47 neighborhoods. 1849 households. Among those households, 55 deaths per year (2 from violence) before the invasion. Among those households, 168 deaths per year (92 from violence) since the invasion. Scale up those sampling results to a population of 5 million households, and you have your 600,000 direct and indirect civilian casualties of war number."

Those are household studies. The researchers went out and spoke to people face to face and got the data from them directly, not from government databases or morgues or from telephone interviews. In 80% of the cases they counted the researchers saw death certificates for the deaths reported.

In the US a MORI poll using a 1000-person sample out of 200 million voters is expected to produce an error of plus or minus 3% when looking at polling predictions. This study had almost twice the sample size for a country with about five million households. The sampling was done to the highest standards and was collecting raw data (number of deaths, the date they occurred and the reason for the death) and not counting vague future intentions of voting preferences.

The 600,000 plus headline figure is extrapolated from these raw numbers but the raw numbers themselves are devastating to the arguments of the flag-wavers and wingnuts that try to ignore or downplay the devastation the invasion of Iraq has precipitated.

Robert,

General Karpinsky had to retire as a colonel after Abu Ghraib. If that, and all those other charges aren't enough for you then, fine.

I never said the U.S. military was perfect. But flawed as it may be, no other military does more at policing its own WRT the Geneva Conventions; very few do as well; and most don't even try. I take the Lancet study (accurate or not) as affirming this POV. It's hard for me to take all that pious concern over the Geneva Conventions seriously after that.

Dave,

"If you were pulling that sort of trick with Abu Ghraib, you would want the shit to hit the fan."

The shit did hit the fan. I've read most of the Army's Fay report, and they took everything apart. It went through much more than just what we saw in those initial pictures.

Keep in mind that senior leaders are limited in how much they can say. If Bush or Rumsfeld had said too much publicly then defense lawyers could cite command influence as unfairly pushing their prosecution.

But as I said, I don't think it's remotely as serious as what's going on now.

Randy, as you said Karpinsky retired as a colonel. He was not court-martialled for dereliction of duty as he should have been (his public defence was probably not going to be very pleasant).

His failings lead to the death of POWs under his care and the systematic year-long sexual abuse and torture of prisoners. If an Iraqi officer had perpetrated the same outrages on American POWs and prisoners then he would not have been permitted an honourable retirement and a pension, he would have been arrested by the US forces and publicly prosecuted as a war criminal. Instead a few E-4s got hard time and dishonourable discharges and the matter was swept under the rug. Anyone bringing up Abu Ghraib as a failure of senior command was accused of "not supporting our troops" and being sympathisers for the Saddam regime.

Your loss. Abu Ghraib has probably cost the US army several hundred dead soldiers over the past few years, but Karpinsky's pension is safe so that's OK.

All the painstaking efforts at assigning some of the excess deaths to American violence, some to "insurgents", some to disease, etc. are a waste of time. There is only one thing to remember: Colin Powell's Pottery Barn Rule. "You break it, you own it". Not said by a wimpy liberal, or a pinko comsymp, or an Al-Qaeda raghead. Said by the US secretary of state. " 'You are going to be the proud owner of 25 million people,' he told the president". And if one of them dies who wouldn't have under Saddam - why, on Dubya's head be it.

(Why do I put quotes around "insurgents"?) How can people who are fighting a foreign invader, in their own country, possibly be "insurgents"? Were the Free French "insurgents" when they fought the Germans, or the Vichy French government that collaborated with the Germans? I tell you this: the Germans treated the French people better than the USA has treated the Iraqis.

Robert,

First, Karpinski is a woman. IIRC, she turned around and gave the "not enough troops" defense.

I don't know that she can be blamed for any deaths of prisoners. I think there was only one who died under interrogation. Someone was prosecuted for that, and I believe that was under CIA authority.

Others were killed in the riot that preceded the events you see in the pictures, but that's because the guards ran out of rubber bullets. (That's usually not a problem for countries that don't stock rubber bullets.)

"If an Iraqi officer had perpetrated the same outrages on American POWs and prisoners then"...

If???? Where did you get the idea that Iraqi officers (or any other country since WWII) who took American POWs ever held to the Geneva Conventions at all?

Americans taken captive by Iraq during the first Gulf War were brutally tortured. Both women POWs were either molested or raped. The one who might not have been actually raped had been wounded in a helicopter crash. She had two broken arms and a bullet in her back, and the Iraqi still decided to remove her clothes and fondle her. You might not have heard about this because the rest of the world simply doesn't care. After all, if they're not going to raise the issue of the Geneva Conventions for Lebanese children being used as human shields, then they're never going to say a word about American POWs.

Americans are taught to understand that the Geneva Conventions will never protect them. As I said, we're among the few who do pay attention.

Robert,

I'm afraid the numbers mean nothing when they are based on a baseline which does NOT include realistic figures for the number of his own people who Saddam was killing, the Kurds who have been dying from he aftereffects of the gassing Saddam did on them back when... figures which are incorrectly quoted WITHOUT showing the error range, which is very high, and so on.

Then there are the methodology problems themselves, as in the last study the methods used are highly problematic. Essentially, the gathering methods are subject to bias in the collecting, in many cases the terrorists killed are double-counted, and so on.

Get a realistic study done, and I'd be interested. For myself, I am going to read at least three expert opinions of every Lancet study in the future before I believe a word of it.

No matter what, you can't make rational descisions based on bad data. Again, this is deliberately NOT a comment on anything else.

"Your loss. Abu Ghraib has probably cost the US army several hundred dead soldiers over the past few years"...

One further irony. If you took the quiz I linked earlier, and then looked at the answers, you'll find out that the military had the opportunity to sweep this all under the rug. The pictures were only released because they decided to prosecute anyway.

Andrew -- the Lancet's baseline was 18 months before the invasion, when the average number of deaths due to violence in the sample group was 2 a year. That's with the bastard Saddam totally in charge.

Afterwards, with George Bush in charge of the biggest best-armed militia of foreign fighters in Iraq, the death rate due to violence is 92 a year for the past three years in the same sample group.

Nobody's got hard numbers for deaths in Iraq during the first 80% of Saddam's murderous regime, but he's not in charge any more. There'a a popularly elected democratic government in power and people are dying violently in great numbers and the Southern Baptist Christian militia can't seem to stop it; indeed from the reports they seem to be doing a lot of the killing and destruction themselves.

Randy -- the Geneva Conventions aren't there to protect American troops from evil, they are there to prevent them from doing evil. It's not a quid quo pro. The US, having signed up to the Geneva Conventions, must obey them even if the other side doesn't. It's something America should be proud of, that it treats prisoners in its care with dignity and respect even if they are the greasiest bastards unhung. In this case and some others the US came up looking like Beria's NKVD on a bad night, and the nomenklatura skated while the proles got it in the neck. The Party triumphed again.

As for sweeping Abu Ghraib under the rug? The Iraqis knew about the torture and murder rooms the Americans were running in Abu Ghraib. The photographs taken by the sadists in US uniforms were circulating on the Web and on CDs being passed around various US military groups, being sent home to relatives and friends as keepsakes. Nowadays they ban prison guards from carrying digital cameras and phones on duty -- I wonder why?

You could buy postcards of some of those pictures in Baghdad -- the infamous one of the hooded man on a box, arms outstretched and connected to wires is known in Iraq as the Statue of Liberty. That's America's lasting legacy in the Middle East.

"Where did you get the idea that Iraqi officers (or any other country since WWII) who took American POWs ever held to the Geneva Conventions at all?"

There have not been any American POWs since 1945. Why? Simple - the USA has not been at war since 1945. Sure, it has conducted quite a lot of "police actions" and interventions. But those weren't wars - they couldn't have been, the USA never declared war on anyone. You can't be a prisoner of war if you were not fighting in a war. Those American prisoners were "enemy combatants", and we know how they should be treated: the President has told us.

Robert,

"the Geneva Conventions aren't there to protect American troops from evil"...

I never said it was. You mistakenly implied it's a benefit when you said, "If an Iraqi officer had perpetrated the same outrages on American POWs and prisoners then".

"The photographs taken by the sadists in US uniforms were circulating on the Web and on CDs being passed around various US military groups, being sent home to relatives and friends as keepsakes."

That may be the popular misconception now, but it's plainly not true. The photos that started the controversy (including that guy with the hood) were released by William Lawson, an uncle of one of the guards, in a stupid attempt to muddy the waters before his trial.

Sure, you can buy them now that they're out there, but those particular pictures were NEVER made public before 60 Minutes had them. We'd never have seen them if the Army had agreed to drop the charges against the guards.

Tom,

An act of war doesn't have to say "declaration of war" for it to be one.

Even if you want to quibble about some wars, I was talking about Gulf War I, and there's no doubt about that one.

If you look at the current war, and the definition of "enemy combatant", you'll note that the big Supreme Court case Bush lost a few months ago still didn't decide they were POWs as the Geneva Conventions defines them.

American military personnel captured in uniform most certainly do count under the third Geneva Convention. You can find it online.

Robert,

I've got other stuff to do but I just realized I missed a major opportunity with this phrase of yours:

"the Geneva Conventions aren't there to protect American troops from evil"...

Yes, and one big reason for the Geneva Conventions is to protect innocent civilians.

If you look back through this thread you'll see that I may have been the only person making this argument about the Geneva Conventions. It has utterly failed to protect those 600k+ who were (supposedly) killed in this war because the enemy isn't being asked to respect it.

The biggest and best-armed militia in Iraq has in fact signed up to the Geneva Conventions. It has certainly killed tens of thousands and probably hundreds of thousands of Iraqis.

We have met the enemy and they is us.

Not according to the numbers from whatever body count you care to look at. It's clear that the biggest killers don't have allies asking them to follow the Geneva Conventions.

BTW: The latest news is that the iraqbodycount.org site is disputing the Lancet's numbers, too. But they obviously have an intellectual property to protect.

I don't understand why some on this thread think that partitioning Iraq would be worse than the current situation. Since the original invasion became inevitable, I've thought it was a good idea - just not implemented due to the objections of the Turks.

Damian, the initial reason is that it'd involve the ethnic cleansing of one to a few million people. For example (so I've heard), the second largest Kurdish city in the world is Baghdad. Which is a city of five millions Sunnis, Shiites and Kurds. Given the vast slaughter that's already been done, there'd be no stable division except a 'clean' one - any bulges or islands would be rather hazardous.

Second, this involves screwing the Sunnis out of all the oil wealth, and they're probably not stupid enough to trust any 'federal' Iraqi government, or the USA. That gives them an incentive to fight.

Third, the Kurds (after expelling a bunch of Sunni Arabs and Turkmen) would have oil, which would have to be exported through (a) Sunni Arab land, (b) Turkey, (c) Iran, or (d) Shiite Iraq. This makes them rather vulnerable to pressure from several groups, all of whom hate Kurds. So the one group which likes the USA would have its balls in a vise.

Fourth, an Independent Kurdistan would pose a direct threat to Turkey. I've heard that Kurdish guerrilla attacks are up in Turkey. Given the US just having sliced their neighboring country in tres partes, it'd be a foolish, trusting Turkish government which wouldn't get rather proactive about defending itself.

Fifth, the Shiite Iraq would probably be rather closely tied to Iran, which would exacerbate current US headaches in the region. IMHO, they're headaches which could be resolved by competant diplomacy (the governments of the US and Iran have a lot of common interests, like no war), but we can't expect that for a couple more years.

Sixth, slicing up Iraq should make every other country in the region very, very nervous. As in 'acquire nukes' nervous. Please remember, in their history this is a common European imperial pastime, rearranging things on the ground to permit better control.

Seventh, and last, this would be sweet Al Qaida propaganda - 'the Americans came claiming to liberate, but raped/killed/pillaged/burned instead, and then set up even more puppet states'.

This thread's died, but just in case anyone runs across it...

A while up there I ask about the pro-invasion, happy-Iraqi blogs that several people mention.

No blog addresses have been forthcoming.

So I thought I'd supply a link to an Iraqi blog myself.

Here's Riverbend. I'd recommend that anyone - anywhere, actually - read her writing.

That's all.

Andrew Crystall:

"Robert,

I'm afraid the numbers mean nothing when they are based on a baseline which does NOT include realistic figures for the number of his own people who Saddam was killing, the Kurds who have been dying from he aftereffects of the gassing Saddam did on them back when... "

Incorrect - did you actually read the study? The pre-invasion baseline was derived from the same sample as the post-invasion figures.

"figures which are incorrectly quoted WITHOUT showing the error range, which is very high, and so on."

Kaplan (from Slate) didn't know his *ss from a hole in the ground with his 'wide confidence interval' comments. Please take Stats 101. As for many sources not quoting the CI, welcome to the world of journalist quoting statistics.

"Then there are the methodology problems themselves, as in the last study the methods used are highly problematic. "

Incorrect, as you'd know from reading the paper, or from the comment threds on Crooked Timber, and Deltoid, in Scienceblogs.com.

"Essentially, the gathering methods are subject to bias in the collecting,"

Yes - both up and down.

"in many cases the terrorists killed are double-counted, and so on."

Incorrect again; double-counting should have been very low, and there was a definite undercounting bias from skipping abandoned houses. Remember that ~80% of the deaths had death certificates asked for an produced, which would have allowed for most double-counts (which should have been few) to be eliminated.

"Get a realistic study done, and I'd be interested. For myself, I am going to read at least three expert opinions of every Lancet study in the future before I believe a word of it."

There've been numerous expert opinions on the Lancet Study, almost all of which endorse it.

"No matter what, you can't make rational descisions based on bad data. Again, this is deliberately NOT a comment on anything else."

This is the best data which has come to light. Note that the US government, with many opportunities to collect such information, has not. Neither has the Iraqi government. That is a definite sign. It's also been corroborated by the IBC project counts, which are at 50K civilians, dead by violence, whose deaths were covered by English language media. When the factors given in references 11,22-26 are applied, that gives a point estimate of 250K civilians, dead by violence. Note that that would be civilians, dead by violence. Not guerrillas, and not causes other than violence.

What is comes down to is that you haven't read the study, or haven't read it with comprehension, and don't know the basic stats needed to judge it. So at this point, I'll disregard your statements until three experts have validated them.

Fair is fair, isn't it?

Colin Powell is much more of a soldier than Rumsfeld. Rumsfeld was a Navy pilot for 3 years and never saw a day of combat. He never had a command. Rumsfeld was an arrogant bastard back when he worked for Nixon. He's a grinning ghoul. The sight of his face on TV makes me sick. Powell may have been Bush's "stooge" but he at least had the guts to resign after he realized how badly the Bush idiots were screwing things up in Iraq.Right now I'm just praying my country has the brains to kick the fucking Republicans out of office just enough to stop the rubber stamping of whatever moronic bullshit Bush comes up with next. And yes, I know most Democrats voted to authorize the use of force in Iraq. They were morons, too.

Oh, Andrew - rechecking the papaer reveals that deaths would not been double-counted, since the critieria for being a member of a household would have prevented that.

Again, you never did read that paper, did you? You just picked up on some right-wing lies, and figured that they'd work.

For those who are sincerely interested in some expert reviews of the Lancet article:

http://en.wikipedia.org/wiki/Lancet_survey_of_mortality_before_and_after_the_2003_invasion_of_Iraq

http://www.stats.org/stories/the_science_ct_dead_oct17_06.htm http://online.wsj.com/public/article/SB116052896787288831.html http://www.nature.com/news/2006/061016/pf/443728a_pf.html

I'm closing this thread for comments as of tomorrow -- the spambots are catching up.

(I may take to running open topics for free-form discussion if this sort of on-going debate recurs.)