Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

September 14, 2007
By: Kevin Drum

VIOLENT DEATHS IN IRAQ....A British polling company recently surveyed 1,461 adults in Iraq and asked each one, "How many members of your household, if any, have died as a result of the conflict in Iraq since 2003?" Based on the results, they say that 1.2 million Iraqis have died violent deaths in the past four years.

The methodology here is nowhere near as detailed as that of last year's Lancet study, which produced a figure of about 650,000 war-related deaths in three years (and probably would have produced a number of about 1 million if it had been extended into 2007), but at first glance it certainly seems to support the notion that the violence rate has been far higher than usually reported. However, here's a second glance:

According to its findings, nearly one in two households in Baghdad had lost at least one member to war- related violence, and 22% of households nationwide had suffered at least one death. It said 48% of the victims were shot to death and 20% died as a result of car bombs, with other explosions and military bombardments blamed for most of the other fatalities.

Hold on. 20% of the deaths were from car bombs? That's 240,000 deaths. Since the average car bomb kills about 7-8 people, this poll is suggesting there have been nearly 32,000 car bomb attacks in Iraq since 2003.

Roughly speaking, that's 20 car bombs per day, compared to official estimates of 2-3 car bombs per day. And while overall death counts are necessarily fuzzy, car bombs are big public events that usually get reported fairly reliably in the media.

So....I dunno. As I recall, there was a similar criticism of the Lancet study on this particular point, and I'm not sure how it got resolved. I know some of my readers have delved pretty deeply into the Lancet controversy, so maybe they can help out in comments. Overall, though, unless you think that car bombs have been massively underreported, which is harder to believe than it is for death counts in general, this result suggests that household self-reporting of violent deaths in Iraq may be prone to exaggeration.

Alternatively, there are way more car bombs in Iraq than we think. Brrr.

Kevin Drum 11:59 AM Permalink | Trackbacks | Comments (72)

Bookmark and Share
 
Comments

Were the questions phrased to ask specifically about deaths *in* the household? It could well be that people are reporting deaths in the extended family and those are being misinterpreted. That would boost the numbers significantly.

Posted by: Randy on September 14, 2007 at 12:08 PM | PERMALINK

Extended family in Iraq may include hundreds of people. My father used to talk about a bus accident that ended up with medical claims from more people than there were seats on the bus. This sounds similar.

Posted by: Mike K on September 14, 2007 at 12:11 PM | PERMALINK

What Randy said.

Posted by: David W. on September 14, 2007 at 12:12 PM | PERMALINK

It's a fair question. We could speculate endlessly about it (and I'm sure we will), but I don't see how we can resolve this without further data.

It sure would be nice if there was a real effort to discover how many Iraqi deaths have resulted from the invasion, but so far as I can see, the U.S. armed forces haven't bothered to investigate this, and wouldn't have much credibility if they did. Other organizations have a hard time looking at it due to the horrific security situation.

Posted by: jimBOB on September 14, 2007 at 12:14 PM | PERMALINK

That the poll basically corroborates the Lancet study seems pretty compelling. I could see there being exagerration in self-reporting, though, along the lines of what Randy suggests. It's possible, isn't it, that the methodology for determining the proportions of cause-of-death was flawed (maybe murders that target individuals make people nervous, and once you ask them about cause of death, they get disturbed by the interview and decide to say the death was just a car bomb instead of the execution-style killing it actually was. No one wants to be connected with someone who people are/were out to get).

Posted by: Swan on September 14, 2007 at 12:14 PM | PERMALINK

From LAT article:

ORB said it drew its conclusion from responses to the question about those living under one roof: "How many members of your household, if any, have died as a result of the conflict in Iraq since 2003?"

It reads as if the pollers specified the definition of household to exclude extended family NOT living under one roof.
Whether respondents understood, and drew that distinction in their answers, we can't say for certain. But on its face, Randy, they did in fact phrase the question so as to exclude overcounting of extended family members NOT living under one roof.

Posted by: kenga on September 14, 2007 at 12:18 PM | PERMALINK

Does the average car bomb really kill 7-8 people? Rather than, say, 20 people? If they're more lethal than you think, the undercount wouldn't be too high...

Posted by: ajay on September 14, 2007 at 12:21 PM | PERMALINK

It could be, also, that it's become popular in Iraq to cover up executions by paying of the local coroner / police, and then performing executions when car bombs go off.

The bereaved gets a phone call: It's the coroner's office or the police. "You know your wife [or whatever] who went missing yesterday? Well it turns out she got killed in the explosion in the market." But she was actually killed because she pissed people off. They didn't want people to know they went after this particular victim for some reason, though- or maybe they just think they can keep American aways / keep them from investigating pr pursuing their group by keeping them from thinking people are performing assasinations in that area. So they try to cover it up as a random death that was part of an act of terrorism, rather than a targeted killing.

Posted by: Swan on September 14, 2007 at 12:22 PM | PERMALINK

problems here do not affect the lancet study, which asked for, and in most cases received, a death certificate.

Posted by: kid bitzer on September 14, 2007 at 12:22 PM | PERMALINK

From ORB themselves- the question specifically said "Please note that I mean those who were actually living under your roof." So some people might be double counted if they move around a lot, but there shouldn't be that much overcounting.

Posted by: SP on September 14, 2007 at 12:25 PM | PERMALINK

"Since the average car bomb kills about 7-8 people"

Where does that statistic come from? It seems to me that car bombs in markets and at bus stations usually account for many many more deaths per incident. Conversely suicide bomb casualties may range from maybe 40-50 down to zero (suicide bomber dies, and only a few bystanders injured)

But car bombs - I don't recall reading about any car bombs where the Iraqi death toll is not at least 10. Again car bombs are different from IED's which mostly target American GI's and where the casualty rate generally maxes out at the number of soldeirs in a Humvee.

If, as my gut tells me the average carnage in a car bomb is 50, then we are talking about 4800 car bombs.

Posted by: Paul on September 14, 2007 at 12:26 PM | PERMALINK

Wasn't the picnic grounds that Mcain and pony show blown to bits after his visit killing somthing like 200 people,Then I thought there was a big bomb in Mousal last year that did about the same amount of damage 7-8 people per car bomb is way to low even for an average.

Posted by: john john on September 14, 2007 at 12:26 PM | PERMALINK

A pet peeve. We get hardlines that, let's say, 7-8 people die and 40 are injured. We never hear about the disposition of the 40 wounded. We don't get conditions for them either. They presumably go to the Iraqi hospitals, which have what recovery rate?

Many may die long after the bomb goes off. We'd never know.

Posted by: Mudge on September 14, 2007 at 12:27 PM | PERMALINK

There's all kinds of craft and deception that goes into crime, assasinations, and law enforcement that your average ultra goody-good law-abiding person doesn't even think of. It's not as if people go out and commit executions and crimes in a totally unthinking way in which people don't do anything else (think how much care you put into your profession, running your schedule, doing a project, etc., and how many tricks you know and use). You don't just decide to kill someone and just go and kill them and that's it- at least, not nearly all of the time. The extraneous behaviors that are part of deception, and so on, can skew these kinds of stats. You need to find a way to explain the lures people throw out to keep the wolves away before you read the stats. Iraq is a place where stats from self-reporting mean a lot less in the US, because Iraqis lived under a dictator, and now Iraq is a semi-lawless quasi-warzone.

Posted by: Swan on September 14, 2007 at 12:28 PM | PERMALINK

Kevin,

Two questions come to mind. My recollection is that the Lancet study reported on the number of excess deaths, the total number of deaths beyond what would be expected had pre-2003 conditions held constant. That number includes a great many causes of death outside of violent death (increased infant mortality, malnutrition, death from lack of medical care, increases in heart attacks due to stress, etc.) If that recollection on my part is correct than this study's number of 1.2 million violent deaths is actually much larger than the Lancet's number and so the individual sub-categories are also large, not surprisingly.

Also, while the above discussion about extended family may well have been covered by the questions, I still wonder how they managed to have each death counted only once. The poll asked each adult how many members of his or her household had died from violence, but depending upon how you contact them (cell phone for example) you could still get multiple people reporting on the same death, especially if Iraqi households tend to have a lot of people in them.

Posted by: MSR on September 14, 2007 at 12:29 PM | PERMALINK

It is very difficult, methodologically and practically, to do these surveys even under the best circumstances. Even if the death toll were only half of what is estimated, it suggests the Lancet study is conservative, it would still rank as one of the most costly modern wars based on national population. If you regard the first Iraq war and the years of sanctions as a first phase of American engagement with Iraq and add in all the displaced persons you have the outline of something quite the opposite of liberation. You have a human horror story that ranks against any in the murderous 20th century. Tyranny is preferable to this fate.

Posted by: bellumregio on September 14, 2007 at 12:31 PM | PERMALINK

Mudge is right, and that was going to be my next point. The question asks about people who died as a result of the war, and car bombs reported in the media only report the death toll from that day. Probably lots of weak (elderly or diseased or just not psychologically hardy enough) people who are injured in attacks never recover and ultimately die.

Posted by: Swan on September 14, 2007 at 12:33 PM | PERMALINK

OT: The confirmation process [for Zubkov] lasted less than two hours and underscored the Kremlin's control over the Russian political process. Pro-Putin lawmakers praised Zubkov, posed easy questions and gladly accepted his responses in rote exchanges that sometimes evoked the airless atmosphere of Soviet-era Communist Party meetings.

And in America between 2001 and 2007 . . . all confirmation processes lasted less than two hours and underscored the White House's control over the American political process. Pro-Bush lawmakers praised the nominees, posed easy questions and glady accepted their responses in rote exchanges that sometimes evoked the airless atmosphere of Soviet-era Communist Party meetings.

Posted by: anonymous on September 14, 2007 at 12:34 PM | PERMALINK

Are some deaths being double counted by this methodology? Let's say they interview my household and ask me the grandfather. I report son's son (my grandson) was killed. But then they interview another family, that of my son's wife, and they also report a dead grandson. Of course it is the same child. But now counted twice?

Posted by: dr.steveb on September 14, 2007 at 12:36 PM | PERMALINK

MSR, there were probably safeguards in the study to make sure multiple respondents from the same housedhold weren't included. They probably thought of it and used a computer database of the respondents' addresses to check.

Re: my 12:33 comment: Especially in an impoverished country like Iraq.

Posted by: Swan on September 14, 2007 at 12:36 PM | PERMALINK

Can be that the number of dead per bomb is greater than we see at the press? Some car bombs were really destructive, I guess that the number of dead that we see immediatly after the attack is an underestimation.

However, that can make for some diference at the data, but certainly not all the diference.

João Carlos

Posted by: João Carlos on September 14, 2007 at 12:53 PM | PERMALINK


the physicist enrico fermi was very good at closely estimating the answers to very hard
probs. given very little data.

probs. where you are asked to closely estimate the answer to a veryhard prob.where you are given
very little data are sometimes called fermi probs.

wikipedia has a fairly good (fairly long) article on fermi probs.

Posted by: wschneid25 on September 14, 2007 at 12:57 PM | PERMALINK

I wonder if part of the discrepancy on the car bomb numbers is that the media reports the number killed right away, but may not count the people who die a day or week or a month later as a result of injuries from the car bomb. Whereas a person would (rightly so) answer a survey like this in the affirmative if a family member died days after the car bomb went off.

Posted by: Michael Patrick on September 14, 2007 at 12:58 PM | PERMALINK

I have to agree with the general feeling that deaths from car bombs go underreported in our media, both from initial estimates of people being killed outright being too low and from wounded people dying hours, days, or even weeks later from their wounds. If you're only counting the initial reported death toll, you're probably getting a pretty severe undercount.

Posted by: Mnemosyne on September 14, 2007 at 1:00 PM | PERMALINK

Just did a google search of car bombs,The first page alone brings the avg. way higher then 7-8 People per More like someone said above, 50 or better on AVG.

Posted by: john john on September 14, 2007 at 1:01 PM | PERMALINK

usually get reported fairly reliably in the media

Doubtful.

The car bomb wounded probably die very slow, agonizing deaths. There is not a lot of quality healthcare in Iraq, if any now. The wounded have no access to antibiotics or even pain killers. The most traumatized victims lie in the heat and beg for death, which pleases a significant portion of Americans, but still not enough for this kind of carnage to receive daily reporting from American media. Continuous reporting about the dead and suffering in Iraq fatigues overweight Americans, who prefer to always have positive super-fantastic attitudes. American media knows reporting on all of the death in Iraq whittles down our great corporate inspired positive attitudes, so they do not do it.

Posted by: Brojo on September 14, 2007 at 1:02 PM | PERMALINK

The statistic of 7-8 killed is from a study by the Iraq Body Count group.

After reading their study, Lancet's and this recent account, it really matters not how many have been killed by Bush's War. When the first loss of life was suffered following our invasion across the Kuwaiti border, it was TOO much.

The increasing numbers just make this obscene Jingoistic adventure by vain glorious Shrub all the worse and his supporter's words more despicable in their utterance.

Posted by: thethirdPaul on September 14, 2007 at 1:04 PM | PERMALINK

For one thing, let's say you want to kill somebody, but you're worried about revenge killings. That's a pretty big threat in a country like Iraq. Everybody has a side, and every side has guns and bombs. So you cover up that the killing was a targeted killing.

Or let's say you work for America or Iran affecting politics in Iraq (to whatever degree covertly or openly), and you want someone dead. You don't want anyone to trace your work back to you, even if you have no reason to think anyone's looking out for you, and there's just a chance of people stumbling across discovering what you're doing. Then you want to cover up the assasination and make it look like it wasn't a targeted killing. Same thing as above, people will wonder, "Well who could have wanted Hajib dead?" and once they start thinking of that, they'll start looking at you.

If you ever watch that show Oz on HBO, those guys are always trying to think of ways to cover up their hits on each other, or to find out who wanted them hit. That's what like half the show is about.

Posted by: Swan on September 14, 2007 at 1:06 PM | PERMALINK

I think it's pretty unlikely that the ave. car bombs could kill 40-50 people. Note than on average, in addition, 100-300 or something like that would have to be wounded.

I think it's a lot more likely that people go missing and then the police tell the relatives that they wandered by the car bomb the other day.

Posted by: Swan on September 14, 2007 at 1:18 PM | PERMALINK

Answers?

"Car Bomb" may be broader than we think, including suicide car bombs, remote detonated car bombs, or explosives thrown at/near cars.

Many "car bomb" attacks may actually be much smaller, killing only two or three civilians, and hence not making it into media reports.

"Car bomb" may be safer to report to pollsters than "sectarian assassination", "kidnapped and tortured", or "shot by security forces".

Agree with Madge (and others). Severe injury is a death sentence in Iraq.

Posted by: EthanS on September 14, 2007 at 1:28 PM | PERMALINK

Even if you disregard the car bomb deaths altogether we're still looking at a million war realted deaths. Either way it's horrific and among the other many Bush administration crimes, this has got to be a standout considering that we were given, after the fact, that at least our presence was going to stop the atrocities of the ancien regime.

Posted by: digitusmedius on September 14, 2007 at 1:39 PM | PERMALINK

And one other point (sorry if someone else pointed this out above): we're usually given just the immediate number of deaths the day of the bombing and almost never hear of those who died days or even weeks later due to their injuries. That alone could lift the number of dead by 50%, esp. considering what I believe would be the highly degraded state of medical care in Iraq.

Posted by: digitusmedius on September 14, 2007 at 1:42 PM | PERMALINK

Fewer Households Today

Easier answer: the estimate is too high because, while the percentages are correct, the total number of households in Iraq has fallen sharply. If the researchers are not accounting for this, their final estimates will be way off.

Lots of families have had fathers or mothers killed. Survivors (especially widows with children) would move in with their relatives. Millions more are displaced, many living with relatives. Thus the number of existing households drops.

Older sons and young couples are less likely to move away from home and established safe places to start new households. Creation of new households has been essentially suspended.

Result: far fewer households than in 2003, and the ones that remain are more likely to have been victims of violence. When multiplied by older data on numbers of households, the estimates are way off.

Posted by: EthanS on September 14, 2007 at 1:51 PM | PERMALINK

Support for fewer households

Iraqi men report employment at 51%, women at 16%.
Iraqi men report household employment at 65%.

Okay, maybe the 16% of women who work have unemployed husbands... BUT

Iraqi women report household employment at 83%!

Ergo, many Iraqi women are living in households where women outnumber men. Given the gender-differences in death tolls, that makes sense. It suggests that households have been consolidated. And note that the pollsters did not ask about household size...

Full Results

Posted by: EthanS on September 14, 2007 at 2:01 PM | PERMALINK

The polling organization says number of households based on 2005 census figures. There may be some fall in number of households by the mid-summer polling date, but less than using "start of war" as a starting point. For more, see http://www.opinion.co.uk

Posted by: TedW on September 14, 2007 at 2:08 PM | PERMALINK

Believing that the average car bomb in Iraq kills 40-50 people is like believing in the Tooth Fairy or leprechauns.

Posted by: Swan on September 14, 2007 at 2:11 PM | PERMALINK

A pet peeve. We get hardlines that, let's say, 7-8 people die and 40 are injured. We never hear about the disposition of the 40 wounded. We don't get conditions for them either. They presumably go to the Iraqi hospitals, which have what recovery rate? Many may die long after the bomb goes off. We'd never know.

A good point. The mass casualty headlines that most people see in the papers typically account for only those who die on the scene -- those who die days, weeks or even months later are not counted.

Posted by: Stefan on September 14, 2007 at 2:36 PM | PERMALINK

Yeah...

  1. Iraqi families are large and closely attached.Deaths from injuries past a day aren't usually counted.
  2. Colloquial use of 'car bomb' includes truck bombs, IEDs, suicide bombers.
  3. Car bombs are easier to remember and probably less traumatic than remembering the police shot someone in the head.
  4. Perhaps more deaths are explained to families as 'car bomb' than as torture and dismemberment.

Posted by: Crissa on September 14, 2007 at 2:42 PM | PERMALINK

You do realize that for the average of less than ten dead per car bomb, there has to be over twenty 'car bombs' that kill no one, for the one 'car bomb' that killed 200?

Posted by: Crissa on September 14, 2007 at 2:52 PM | PERMALINK

the IBC study accounted for later deaths from injuries.

most car bombs cause relatively few fatalities. a few massive ones cause large numbers of fatalities (those are going to be the ones showing up in a google search).

no, there haven't been 230,000 car bombs in Iraq (in fact the first couple years after the invasion there were very few altogether).

throw in collateral damage from IEDs and you get a couple thousand more fatalities.

inaccurate self-reporting has been an inherent problem with Iraq data all along...it's just not reliable (it's a problem in any middle eastern country and the conditions in Iraq only make it worse)

Posted by: Nathan on September 14, 2007 at 2:55 PM | PERMALINK

Well, this is really gruesome, but ... what if there isn't any uh, evidence of a person to *count* after a car bomb, and the family doesn't figure it out till much later. Brrr.

Posted by: Doc at the Radar Station on September 14, 2007 at 2:56 PM | PERMALINK

inaccurate self-reporting has been an inherent problem with Iraq data all along...it's just not reliable (it's a problem in any middle eastern country

We've noticed it's a problem in Nolita, too.

Posted by: shortstop on September 14, 2007 at 2:57 PM | PERMALINK

Why is everyone talking about car bombs?

The real survey question should have been "How many members of your household, if any, have died by American hands?"

I'd would not be surprised if killings by Americans far outnumber the Iraqi-on-Iraqi totals, both in the initial 'warfare' stages and the current 'anti-terrist' phase. (which mostly means 'Kill people who oppose the occupation'.

Another dirty little secret.

Posted by: Buford on September 14, 2007 at 3:26 PM | PERMALINK

I'd would not be surprised if killings by Americans far outnumber the Iraqi-on-Iraqi totals, both in the initial 'warfare' stages and the current 'anti-terrist' phase. (which mostly means 'Kill people who oppose the occupation'.

Another dirty little secret.

IBC looked into that, too. In the first three years (2003-2005), counting the original invasion, Coalition forces accounted for 37 percent of civilian victims. In what IBC calls "Year Four" (March 2006-March 2007, referred to on their site as "the worst for violence"), Coalition forces accounted for 739 civilian deaths out of 29,435.

Sorry to disappoint.

Posted by: harry on September 14, 2007 at 3:41 PM | PERMALINK

from the way the question was worded
(and translated to/from Arabic)
Car bombs could have included road-based IEDs.

Posted by: davidn on September 14, 2007 at 3:56 PM | PERMALINK

***

Posted by: mhr on September 14, 2007 at 4:23 PM | PERMALINK

One check on these numbers would be to compare results on other questions with similar ones from the ABC/BBC poll from last week. If percentages are roughly the same, then we can at least rule out the idea that the ORB poll had a bias in its sampling techniques (or at least, rule out that it had a unique bias). Of course, even if they match, that still leaves open the possibility that violent deaths are reported less reliably than other things, but at the very least it would be nice to know if the random selection of sampled households was consistent from poll to poll.

Posted by: JD on September 14, 2007 at 4:37 PM | PERMALINK

The poll was conducted by Opinion Research Business, or ORB, and their website has a summary of the findings.

From there you can follow links to files containing the raw data, which include a PDF and an Excel sheet.

Note that the standard margin of error based on the sample size is ±2.4%, so the actual estimated number of dead ranges from 1,191,286 to 1,249,873.

The summary also notes that this number exceeds the number of dead in the 1994 Rwanda genocide (that was 800,000), and that almost one in two households in Baghdad has lost a family member, which is the highest rate of any region in the country.

The raw data also includes results about the numbers of injured and refugees, which are apparently not reported in the LA Times. The summary at the ORB site says that more than a million have been injured, and that nearly half of the millions who have left their homes have left the country; most of these have gone to Syria.

Posted by: Buckeye Hamburger on September 14, 2007 at 5:20 PM | PERMALINK

Well, The Trouble with Harry and Nathan like to use the IBC to discount the work of Lancet and the newest results.

However, why do you not use the actual words of the IBC concerning the Lancet report? The IBC concludes: "Do the Amercan people need to believe that 600,000 Iraqis have been killed before they can turn to their leaders and say 'enough is enough'? The number of certain civilian deaths has been documented to a basic standard of corraboration by 'passive surveillance methods' surely already provides all the necessary evidence to deem this invasion and occupation an utter failure at all levels."

Sorry to disappoint, but nice cherry picking. anyway.

Posted by: thethirdPaul on September 14, 2007 at 5:41 PM | PERMALINK

Ah. The "truthiness" argument. Doesn't matter if our facts are crap, the basic idea is what counts. "Cherry picking," which at least involves selecting certain actual facts over others, is still preferable to just pulling numbers out of your ass.

Where I come from, reality counts for something. Even Kevin points out that the survey makes no sense on the face of it, or when checked against almost every other reliable source. But hey, a million dead Iraqis sounds so much more impressive when you write about it.

Besides, if you can artificially inflate that number high enough, at some point you might manage to get past the number of Iraqis Saddam killed through oppression and warfare.

We do still remember Saddam, don't we?

Posted by: harry on September 14, 2007 at 6:26 PM | PERMALINK

Ah. The "truthiness" argument. Doesn't matter if our facts are crap, the basic idea is what counts. "Cherry picking," which at least involves selecting certain actual facts over others, is still preferable to just pulling numbers out of your ass.

The IBC calls their facts "passive". This and the Lancet study showed their work. That you disagree with said work doesn't mean THEY were pulling numbers out of their asses. It means they have done research that your sorry ass can't understand, won't review and dismiss out of hand because someone else's number (gleaned from newspaper reports) says something else (which THEY admit is an incomplete and passive method).

Even Kevin points out that the survey makes no sense on the face of it, or when checked against almost every other reliable source.

You people said the same thing about the Lancet study. You have NO numbers save for the IBC's which you cite, no methodology to counter that of those who have done actual studies and no fucking clue whether the numbers are real or not.

What other reliable source have you checked ace? The US says they don't do body counts. The Iraqi government doesn't either.

It's really pathetic you can't own up to the fact that regardless of the precise number, you've supported a war that has killed, wounded or displaced literally MILLIONS of people. And for what? Assholes like you.

Posted by: Jay B. on September 14, 2007 at 7:08 PM | PERMALINK

"... Coalition forces accounted for 37 percent of civilian victims. In what IBC calls "Year Four" (March 2006-March 2007, referred to on their site as "the worst for violence"), Coalition forces accounted for 739 civilian deaths out of 29,435".

Holy shit... we spent a quarter trillion dollars on shock and awe and only accounted for 40% of the dead? Can we get our money back?

"Coalition forces accounted for 739 civilian deaths "

Ya sure. Hmmm... What's a 'civilian'?

Posted by: Buford on September 14, 2007 at 8:27 PM | PERMALINK

Kevin,
Kirk A. Johnson served 13 months as the chief statistician for Ambassador Crocker and concludes that the numbers for violence are probably gross underestimate.

More here

Posted by: Steve J. on September 14, 2007 at 8:50 PM | PERMALINK

harry,

You sound like a criminal defense lawyer representing someone accused of murder for a driveby firing of a Glock into 15 people standing on a front lawn.

You say, "No way, your honor, my client only saw three and wasn't it a drug house anyway?"

Posted by: thethirdPaul on September 14, 2007 at 9:45 PM | PERMALINK

Isn't IBC from media reports? In that case, their estimate of car-bomb deaths would be subject to the same issues as the rest of their estimates. It pretty much establishes an absolute lower bound.

Posted by: paul on September 14, 2007 at 10:07 PM | PERMALINK

Speaking of the disrupted households, destruction and misery Bush brought to Iraq, riverbendblog.blogspot (aka Baghdad Burning) posted September 6th that she left Iraq
via neighboring Syria, her previous post having been as far back as April.

This kind of says it all--

"...I'm trying to rid myself of the expectation that armed people in black will break through the door and into our lives. I’m trying to let my eyes grow accustomed to streets free of road blocks, hummers and pictures of Muqtada and the rest…"

There she is, an anonymous person blogging, living it. It must simply be a most dismal existence. Very sad.


Posted by: consider wisely always on September 15, 2007 at 4:29 AM | PERMALINK

So, Harry, if you add together the number of deaths caused by Saddam with the number who died since Dubya came to town, do you get a number that will make YOUR head explode?

Posted by: Kenji on September 15, 2007 at 4:55 AM | PERMALINK

When Saddam counted his dead, over a longer period of time, he needed mass graves to get rid of the bodies. Where are the mass graves now, if the numbers are ten times what Saddam counted?

When the Lancet counted they claimed that most of the dead were from coalition air attacks, so if now most of the dead are shot to death or car-bombed...? WTF?

When the Lancet counted they suggested that they used death certificates to certify deaths. Why didn't they ask the certifying agency about their numbers? If there is enough civil society to issue certificates, can't they get a more accurate number by asking how many certificates went out?

Congratulations to this blog for having sensible reality-based commenters who know that both sides have an agenda to paint the worst or best possible image.

It doesn't mean things don't suck over there, but it surely doesn't help to cherry pick the very worst points of view every time, which is too often the case in this highly emotional, highly fractured debate.

Posted by: Undertoad on September 15, 2007 at 8:16 AM | PERMALINK

The death of Sheik Abu Risha means the tiny progress in Al Anbar province is now gone, according to Patrick Cockburn, who has actually spent time in the real Iraq (outside of the Green Zone), unlike “Peaches” Petraeus. Read more here. Petraeus has become the military equivalent of Paris Hilton, whose celebrity far exceeds his accomplishments.

Posted by: The Conservative Deflator on September 15, 2007 at 10:09 AM | PERMALINK

Where are the mass graves now, if the numbers are ten times what Saddam counted?

If you followed the news, you would have seen daily reports of bodies turning up, shot and/or tortured. Add to that other bodies dumped and not found, and you don't need to look for mass graves.

When the Lancet counted they claimed that most of the dead were from coalition air attacks

I don't think this is so. I think they merely asked about excess deaths. Causes could be anything from violence to disease caused by malnutrition.

Why didn't they ask the certifying agency about their numbers? If there is enough civil society to issue certificates, can't they get a more accurate number by asking how many certificates went out?

Because comparing excess deaths to deaths from all causes is an apples-to-oranges comparison which would tell you nothing.

Posted by: jimBOB on September 15, 2007 at 12:14 PM | PERMALINK

Uh actually the U.S. Military stopped counting car bombs as violence in Iraq, smart bombs are not violent either. Didn't see this on the news!

At least 43 Iraqis were killed and 20 more wounded FRIDAY. Among the incidents was a suicide bombing in Baiji that killed or wounded over two dozen people. A roadside bomb killed four GIs in Diyala province today. Three GIs were wounded in combat in Garma; usually, GI injuries do not get reported in the media. Outside a Baiji restaurant/checkpoint, a suicide truck bomber rammed his vehicle into two parked police cars killing 11 people, mostly policemen, and wounding 16 others. In Baghdad, six dumped bodies were recovered; three of the corpses were beheaded and found in Doura. Outside Baghdad, gunmen killed three farmers who were guarding their small village. Gunmen shot three people dead near Suwayra. In Kirkuk, gunmen stormed a home and killed an Iraq army soldier. Also, a policeman was shot dead in the al-Wasiti neighborhood. A woman was killed and her daughter wounded during an armed attack in Hawija. A roadside bomb in Fallujah killed one policeman and wounded two others. In Balad, the bodies of a judge and a police officer were found. The body of a former Iraqi Army major general was found yesterday in Mosul; he had been kidnapped on Monday. At the central Hilla home of a senior army official, gunmen clashed with guards, killing one and wounding a second. Three gunmen were killed and three others arrested after they were caught planting a roadside bomb in Khalis. U.S. forces killed four suspects and detained five in Yusufiya. Another gunman was killed in Tarmiyah. At a raid on a mosque in Garma, U.S. troops killed three gunmen. Three GIs were also wounded.

Posted by: Post American on September 15, 2007 at 12:48 PM | PERMALINK

As Post American demonstrates, just another day at a Farmer's Market in Indiana.

Posted by: thethirdPaul on September 15, 2007 at 12:53 PM | PERMALINK

If you followed the news, you would have seen daily reports of bodies turning up, shot and/or tortured. Add to that other bodies dumped and not found, and you don't need to look for mass graves.

We can't learn everything from the news. Sometimes we learn from what is not news.

Now, look, it's very simple. Put on your critical thinking cap and follow me.

A human body takes approximately 8 months to completely decompose.

Longer in the desert.

So, either there are mass graves, or the streets are full of stinking decomposing corpses causing all kinds of secondary diseases.

If either case were present in Iraq, you would have surely seen it on "the news".

Posted by: Undertoad on September 15, 2007 at 1:36 PM | PERMALINK

I believe the poster at 1:36 PM left a Y off his username.

Posted by: thethirdPaul on September 15, 2007 at 1:47 PM | PERMALINK

If there were 20 car-bombs a day, it seems like there basically wouldn't be any cars left in Iraq at this point.

Posted by: Adam Kotsko on September 15, 2007 at 2:06 PM | PERMALINK

Iraqis might interpret IEDs as "car bombs". The roadways are littered with IEDs and no doubt Iraqis are being killed everyday by them. The killing isn't reported in the news because only a small number of passengers are killed by an IED.

Posted by: Jim on September 15, 2007 at 2:11 PM | PERMALINK

For those who are using IBC figures - their criteria is (a) deaths of civilians (b) by violence (c) with at least two english-language media reports.

That leads to a *lot* of undercounting. IMHO, the later tolls of wounded dying from a car bomb would get less media attention, so they should undercount those most severely.

The original survey in the Lancet discussed this issue, among others. They said that experience showed that the reporting rate for deaths went down to 20% in civil wars, as the death rate climbed. Applying that to the IBC figures means to multiply by 5, just to account for criterion (c).

Posted by: Barry on September 15, 2007 at 3:58 PM | PERMALINK

We can't learn everything from the news. Sometimes we learn from what is not news.

IOW I'm going to ignore what you said because, well, BECAUSE!

A human body takes approximately 8 months to completely decompose.

Not necessarily, if there are predators present.

either there are mass graves, or the streets are full of stinking decomposing corpses

Iraq is over 400,000 square kilometers. Assuming a half million corpses, that's only a bit over one per square kilometer. Try again.

Posted by: jimBOB on September 15, 2007 at 5:15 PM | PERMALINK

harry: We do still remember Saddam, don't we?


more americans died in iraq AFTER saddam was captured...

than in all his decades in power...

and since his execution...

that number has risen...

Posted by: mr. irony on September 15, 2007 at 5:30 PM | PERMALINK

Let's just say the figure is a bit off (and I'm not saying it is), I think the number of deaths in Iraq, since the (illegal) US invasion and subsequent (illegal) occupation, is far far far too many to ever be conscionable. Would you want to live there?????

Posted by: Edwin on September 15, 2007 at 6:48 PM | PERMALINK

"Assuming a half million corpses, that's only a bit over one per square kilometer. Try again."

Not to defend our dear little toady friend, but you know that's not quite correct. The violence is concentrated. Of course, with concentrated violence, e.g., in Baghdad, comes the ability to deal with large quantities of dead people.

"Mass graves" only make it into the news when they are a surprise and nobody knew about them. Mass graves caused by hundreds of people being killed every day aren't a surprise and aren't news.

Posted by: PaulB on September 15, 2007 at 10:24 PM | PERMALINK

"Would you want to live there?????"

No, nor die there. I think that's one of the reasons that the Iraqi reaction to Bush's speech about how "normal" everything is there was so intense. They know better. As if they really care whether the violence is "sectarian" or not. Dead is dead.

Posted by: PaulB on September 15, 2007 at 10:26 PM | PERMALINK




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly