Network News

X My Profile
View More Activity
Posted at 6:00 AM ET, 10/19/2007

Iraqi Civilian Deaths, Part II

By Michael Dobbs


Hillary Clinton and Gen. David Petraeus at Congressional hearing.
"Civilian deaths of all categories, less natural causes, have also declined considerably, by over 45 percent Iraq-wide, since the height of the sectarian violence in December."
--General David Petraeus, congressional testimony, September 11, 2007.
"The reports that you provide to us really require the willing suspension of disbelief. ... If you look at all the evidence that's been presented, overall civilian deaths have risen."
--Sen. Hillary Clinton, responding to Petraeus during his Senate testimony.

The Facts

Earlier this month, we attempted to descramble the claim by General David Petraeus that civilian deaths in Iraq have fallen by "over 45 percent" since last December. We also looked at Hillary Clinton's claim that "civilian deaths have risen" in Iraq. We refrained from a snap Pinocchio ruling on this subject because it is extremely complex, and we wanted to make sure we understood the arguments of both sides.

We have now spoken to various other experts, including the chief military statistician in Baghdad, Colonel David R. LaRivee, and feel more confident about issuing a verdict. But first, let's summarize the state of the debate.

The Petraeus claim of an over 45 percent drop in Iraqi civilian casualties rests on combining two very different datasets. The ochre bars in the graph below represent data collected and verified by Coalition forces in Iraq. The grey bars represent additional data collected by Iraqi authorities, primarily the Ministry of the Interior, which controls the police. (All data supplied by the U.S. military, which has attempted to eliminate duplicate reports.)

[Graph: Differing Iraqi Casualty Figures]

The statistics show a clear downward trend since last winter. The steepness of the decline, however, depends on which column you look at. Looking at coalition reports only, civilian deaths fell by about 12 percent from last December and slightly over 30 percent from last November. Factoring in the additional Iraqi data, civilian deaths fell by "over 45 percent" between December 2006 and August 2007. Thus the Petraeus claim, which formed a centerpiece of his testimony to Congress on September 10-11, and was prominently picked up by the White House as proof of the effectiveness of the "surge."

The credibility of the Petraeus "over 45 percent" claim depends on the degree of confidence you have in the Iraqi data and the methods used to combine the two datasets. This was the focus of our interview with Colonel LaRivee. (Petraeus's spokesman, Col. Steven Boylan, has already conceded that the December figures include "a large number of unverfied host nation reports.")

The colonel began by explaining that the U.S.-led coalition "catches most major events" happening around the country, "but we know we do not catch all of them." Say a bomb attack occurs in a part of Iraq where coalition forces are not present. In this case, U.S. statisticians rely on information passed up the Iraqi chain of command. Even in Baghdad, U.S. troops do not investigate many of the smaller incidents, such as isolated executions and murders, or small arms attacks on Iraqi units.

So how reliable are the Iraqi data? According to Colonel LaRivee, the Iraqi authorities are improving their data collection processes all the time, but the system is still far from perfect. There is "definitely a lag in host nation reporting." Some events that were reported as having taken place in December 2006 could well have taken place in November. There are sometimes duplicate reports of the same incident, leading to inflated figures.

Asked about Petraeus's "over 45 percent claim," LaRivee said that he would not "want anybody to read too much into the precision of that statement. [Petraeus] knows very well that the host nation reports have some problems with them. But the trends are moving in the right direction."

According to LaRivee, the August figures included several "high casualty" incidents, such as devastating suicide attacks on Yazidi villages in northern Iraq. By U.S. estimates, 326 people were killed in the Aug. 15 bombings of the Yazidi villages. (Preliminary estimates from the Red Crescent spoke of more than five hundred dead.) Had it not been for these mass suicide attacks, the decline in civilian casualties would have been even more pronounced. Most western news organizations based in Baghdad reported a fall in civilian deaths between August and September.

To sum up our findings, we agree with the Public Editor of the New York Times that there is good evidence that "civilian deaths from war-related violence have gone down since the end of last year." Like him, we are skeptical of General Petraeus's claim of an "over 45 percent decline" between December and August. As we noted in our last post, it is hard to explain why Iraqi-reported civilian deaths declined much more sharply than Coalition-reported civilian deaths over this period.

We should also deal here with a Council on Foreign Relations analysis that claimed that other organizations, including the London-based Iraq Body Count, the Brookings Institution's Iraq Index, and The Washington Post, have reported an even steeper rate of decline in Iraqi civilian death data than Petraeus. We have already established that the Council expert, Stephen Biddle, was using out-of-date Iraq Body Count data when he made his analysis. (The revised data shows a significantly shallower decline than that claimed by Petraeus.) Michael O'Hanlon of Brookings says he relied on Petraeus's own data for his 2007 analysis of Iraqi civilian deaths, so it obviously does not make sense to refer to Brookings as independent verification for Petraeus's statistics. The Washington Post data to which Biddle refers is an unofficial Iraqi health ministry tally of bodies delivered to morgues. The data excludes bodies buried directly by families, and is far from comprehensive.

The Pinocchio Test

After due consideration, we are awarding one Pinocchio each to the Council on Foreign Relations, Hillary Clinton, and General Petraeus, "for shading of the facts, selective telling of the truth," and/or "some omissions and exaggerrations."

Council on Foreign Relations

The Council relied on out-of-date data in its analysis of Iraqi civilian deaths and failed to correct the record on its own website, even after the mistake was pointed out.

Hillary Clinton

Claimed that Iraqi civilian deaths "have risen" without specifying her terms. She read talking points almost verbatim from a Democratic think tank, the Center for American Progress, which argues that the violence has not gone down. Her spokesman, Philippe Reines, told us earlier that Clinton was referring only to the months of July and August 2007 when she claimed that the death toll had "risen." He declined to respond to a question about whether the senator now believes that Iraqi civilian deaths have risen or fallen since last December. For our analysis of Clinton's promises to "end the war" in Iraq, click here.

General David Petraeus

We give the military credit for being willing to explain the admittedly confusing statistics after the event. We note, however, that Petraeus made no attempt in his Congressional testimony to qualify his dramatic claim that Iraq-wide civilian deaths have declined "by over 45 percent ...since the height of the sectarian violence." It would have been more accurate to have said something like "as much as 45 percent", or "our best estimate," or even "the trendlines are generally down, but we cannot be sure of the precise figure." Instead he succumbed to the public relations temptation of wanting to make the figures seem as rosy and as authoritative as possible.

By Michael Dobbs  | October 19, 2007; 6:00 AM ET
Categories:  1 Pinocchio, Gov Watch, Iraq  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: An Inconvenient Truth: Team Gore Responds
Next: Obama, Edwards, and the Lobbying Industry


Nice, balanced analysis that reflects the standards inherent in these public situations: everybody shades the truth for dramatic effect. Therefore, everybody deserves Pinocchios.

Kudos on coming out and actually criticizing Petraues where he richly deserves it. A military man in the service of politicians IS a politician and a rich target for criticism. Of course, you do know this makes you an active Al Qaeda collaborator in the eyes of the VRWC?

I'd add the following to your caption: "Clinton reflects on the inherent awfulness of cold Starbucks French Roast even with three Splendas. Petraeus conducts a brief recon, looking for dark roots."

Posted by: Judge C. Crater | October 19, 2007 8:09 AM | Report abuse

" Instead he succumbed to the public relations temptation of wanting to make the figures seem as rosy and as authoritative as possible."

Therein lies the problem. His role in the proceeding was source, not advocate. His performance was highly inappropriate.

Posted by: zukermand | October 19, 2007 9:50 AM | Report abuse

what did Senator Clinton's office say when you contacted them for clarification of her statement. While the General's statement may have been an exaggeration, or true, depending on what should set of data used, Senator Clinton's statement was an out and out falsehood. Unless you know of some reputable, verifiable data that shows deaths rising, you should have taken her to task much more than the military. Instead, you spend the entire article parsing General Petraus. Where is the love for Senator Clinton?

Posted by: Brian | October 19, 2007 10:41 AM | Report abuse

Generals follow their commander in chief regardless of right or wrong. Genl. Sherman followed orders and burned and looted everything in his path while marching to the sea, Genl. Westmoreland followed Johnson/Nixon orders in spike of his personal reservations (expressed publicly many years later)that Vietnam was a great waste of men and money, and now Genl. Petraeus.....? Nothing ever changes.

Posted by: A.Lincoln | October 19, 2007 10:44 AM | Report abuse

"She read talking points almost verbatim from a Democratic think tank, the Center for American Progress, which argues that the violence has not gone down."

So what? This very column pointed out last month that the AP showed a large increase in fatalities in 2007.

"succumbed to the public relations temptation"

Can you lick this man's boots any more carefully? No one cares about your psychoanalysts, Fact Chucker. The dude flat out lied, as per's prediction.

Posted by: HeavyJ | October 19, 2007 10:45 AM | Report abuse

Based on the coalition data, deaths have hovered around 1000 a month since December, (Pre Surge.) The only significant decrease was in June.
So based on the coalition data, the surge has not really had any effect.

Posted by: Ted | October 19, 2007 10:47 AM | Report abuse

I find it interesting that you spent so much time and effort fact checking the military. Even though the data shows a definite decrease in civilian deaths, you conclude that General Petraus deserves a Pinocchio. Meanwhile, there's no evidence that you fact checked Clinton. Could it be that you only fact check those that you disagree with?

Meanwhile, your own editorial board, on October 14, concluded that deaths, both civilian and US military, have decreased. In fact, they cite figures since the August cutoff you used that show even more dramatic declines.

If you want to be considered an authoritative source, you can't let your own political biases interfere with the fact checking. Please follow-up with Senator Clinton's office and let us know how she explains her statement.

Posted by: Eric | October 19, 2007 10:51 AM | Report abuse

The data on Iraqi civilian deaths lacks context. One of the reasons for the decline in reported deaths is that by this point, formerly mixed neighborhoods in Baghdad have been ethnically cleansed. There's no one left to be killed. If you consider it a success to have a capital city in which people are afraid to venture outside their ethnic neighborhoods and everyone is armed to the teeth, the surge is a huge success.

Posted by: Patrick Kiger | October 19, 2007 10:58 AM | Report abuse

What a crock of crap. You can't compare the number of attacks in the blistering hot Iraqi summer to the number of attacks in the winter and come up with anything meaningful. The real comparison is between the number of attacks in any given month and the number of attacks in the _same_ month one year ago. In every month the number of attacks per month has risen when compared in this method.

Moreover, since the Army's methods of data collection and analysis are classified, we can't rely on their data at all. Without knowing those techniques it is impossible to draw any conclusion based on this data other than to say that it is highly suspect.

Let it not be lost in all this bloody shirt waving that the guy who has the final say on what anyone does or says in the Pentagon is George W. Bush. Petraeus is his dog, these are Bush's numbers.

Posted by: DK | October 19, 2007 11:37 AM | Report abuse

This analysis seems too much one sided if somebody gave statistics base one some set of data and can be verified to be true then that person is technically right. Thus it totally mathematically false to give both conclusion the same weight, in any case mathematics is foreign for most journalist

Posted by: Tobi | October 19, 2007 11:46 AM | Report abuse

Actually DK, you are wrong. It would be interesting to know what data you are relying on to make your point. Since you speak with such certainty, could you let us know your source? Leaving the military statistics, which you obviously mistrust, let's look at data from Iraq Body Count, an independent organization that bases its casualty figures on multiple sources. They disagree with the military on several issues, but their data also shows a downward trend, month compared to month, for June, July, and August (2006 compared with 2007.) See my original post, and this graph, comparing the IBC data to the military data.

Posted by: The Fact Checker | October 19, 2007 11:55 AM | Report abuse

You conclude: deaths went down, maybe as much as Petraues said but probably not that much.

You decide:

Petraeus wasn't telling the whole truth and maybe exagerrating. 1 pin.

Hillary Clinton "explained" that she wasn't really talking about "overall deaths" and "all the evidence" when she said "If you look at all the evidence that's been presented, overall civilian deaths have risen." 1 pin too. all even steven.

Hey, dim bulbs... she was flat out wrong, cited no sources, and then hid behind a hack who made up nonsense instead of backing up the claim.

What sort of clown decides that it was pretty much a tie? Yep, an intellectually dishonest clown posing as a fair judge.

Posted by: Michael Korcok | October 19, 2007 11:56 AM | Report abuse

I am a bit nervous in trusting either of the two sources relied on here. Can you cross correlate with any other independent source BESIDES the US military (who of course wants to show progress) and the Iraqi government (unsure of their motives)?

What about those folks who are in neither government who are accessing civilian deaths. We know for sure that statistics by the military from the Vietnam War were quite fatally flawed in some significant cases.

Posted by: Anonymous | October 19, 2007 11:57 AM | Report abuse

Something is wrong in a war where the civilian deaths out number the compatant deaths. The only conclusion I can make is that this is a civil war and we need to get out.

Posted by: Fred | October 19, 2007 12:07 PM | Report abuse

Well like any war the real figures may take months and months to figure out. Any mass graves found in the near future, or far future, I am sure will be glossed over, the idea that deaths are going up or down is disgusting talk anyway. To the person that died THEIR death is all that matters. Having no right to be in that country(preemptive strike) to begin with makes any death in that country too many, and an upward trend that should never have happened.


Posted by: cartman | October 19, 2007 12:18 PM | Report abuse

Your bar graph is really helpful, thouh not perhaps in the way you intended. It demonstrates how unhelpful it is for everyone to focus myopically on various claims that "violence in Iraq is down [or up] by x percent since y [arbitrary date]," instead of asking how (if at all) it actually changes the situation in the real world. So I would invite everyone to simply LOOK AT THE GRAPH. It shows consistently high rates of civilian deaths (murders) on a continuing basis over the last year. Does anyone think it is meaningfully different to live in a country where "only" 1200 sectarian murders took place this month, compared to 1500 several months ago? Is political reconciliation any more possible in such an environment? Are these not the questions that really matter?

Posted by: Matt Clifford | October 19, 2007 12:44 PM | Report abuse

I'd like to have seen the following Iraq Body Count numbers of Iraqi civilians killed by violence in your article: 2003 12,000; 2004 10,500; 2005 14,000; 2006 27,000 and 2007 through about September 20,000.

They write: "It is important to place the events of 2007 in context. Levels of violence reached at all time high in 2006. Only in comparison to that could the first half of 2007 be regarded as an improvement. Despite any effort put into the surge, the first six months of 2007 was still the most deadly first six months for civilians any any year since the invasion."

(And then there is the recent survey of Iraqi households indicating a civilian death toll of 1.2 million since the invasion.)

Posted by: Bernice | October 19, 2007 12:51 PM | Report abuse

Once again, Dobbs has done a miserable job of "fact checking." It's embarrassing. First, how could Dobbs possibly claim to be fact-checking when he completely ignores the most authoritative studies conducted on civilian deaths, those from Johns Hopkins University's Bloomberg School of Public Health. The latest study done by experts at Hopkins last fall estimated that over 650,000 Iraqis had died as a result of the war, a figure far in excess of any published by either the Iraqi government or the American military. Clearly, the U.S. and Iraqi goverments are publishing figures that downplay the deaths occuring around the country in an effort to cover up the disastrous extent of the war on the Iraqi people. And it has been known for years that the body counts from morgues and journalists grossly under-count the true extent of deaths there.

Dobbs should have acknowledged how unreliable those figures were and at least made reference to the far more reliable Hopkins estimates. His failure to do so reveals his "fact checking" to be a sham.

Posted by: Connecticut | October 19, 2007 12:54 PM | Report abuse

Furthermore, Dobbs completely left out the seasonality of the fighting. It is just not possible to start with the traditionally heavy late-year casualty figures, then suggest that they are going down in the early part of the year. You have to compare deaths and casualties in July against previous Julys, June against previous Junes. Dobbs flat out blew it. He should not be permitted to publish "fact checks" until he learns how to fact check.

Posted by: CT2 | October 19, 2007 12:57 PM | Report abuse

And who had the best resources to get those numbers - Gen. Petraeus, but "succumbed to the public relations temptation of wanting to make the figures seem as rosy and as authoritative as possible." Isn't that just a nice way of saying he shaded the truth, or lied? He gets a whole lot more than a single Pinocchio from any "unbiased" evaluator.

Posted by: mkh1941 | October 19, 2007 1:11 PM | Report abuse

The man who told the truth, but perhaps not the whole truth, is rated the same as the woman who essentially told an out and out lie.

I've been reading these 'fact checker' things for a while now. In each one I've read, the conservative transgressor is held to much stricter account than the liberal transgressor.

The bias is obvious to the most casual observer, and diminishes the authority of this column. Too bad, because 'fact checking' is something we sorely need. But it must be objective and unbiased, to be worth a darn.

Posted by: anonymous | October 19, 2007 1:24 PM | Report abuse

We need to be suspicious of all the figures.
The Johns Hopkins study last year as pointed out be Bernice, showed 650,000 deaths. This is based on statistical sampling of the sort that is used by international organizations in countries where no reliable data are available.

There is another survey which also shows Iraqi deaths far higher than any of the report discussed by Dobbs.
Quoting from the LA times AP of sept 14 2007,
"According to the ORB poll, a survey of 1,461 adults suggested that the total number slain during more than four years of war was more than 1.2 million.

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?"

These 2 independent measurements put the total civilian Iraqi death toll at over ten times the amounts tallied by the Iraq Body Count organization using verified reports, and also many times higher than the official Iraqi government reports.

Truth is the first casualty of war.

Posted by: Eric A | October 19, 2007 1:26 PM | Report abuse

"Dobbs completely left out the seasonality of the fighting"

Maybe he left it out because there's no evidence for it (no reason to put it in). Is there any data that supports this "seasonality" theory? Trends for casualties in the war which I've seen have shown a pretty steady rise until this year, not rises and falls according to seasons.

Also, the "Hopkins study" isn't really relevant to this article, since it can't give anyone insight on trends during the surge. And before anyone gets carried way with this "reliable" study they should give this a read:

Posted by: Leslie | October 19, 2007 1:34 PM | Report abuse

Patrick Kiger had a good point - the deaths may be 'down' but it could also be due to the ethnic movement out of certain neighborhoods. Sad but true, there are less people to kill. I'd also like to the population figures on Iraq and how much those have dropped since the initial invasion. While the number of deaths have gone down I wonder about the percents.

Posted by: David | October 19, 2007 2:24 PM | Report abuse

There's something fundamentally unholy about sorting out who was more "factual" when the 2007 numbers are 20,000 dead and counting.

Posted by: jerseymissouri | October 19, 2007 3:08 PM | Report abuse

I didn't follow Clinton's exact wording all that closely in her exchange with the general, but in comparing their statements shouldn't we note that it was Petraeus's job to report, as accurately as possible, the situation in Iraq, and Clinton's job to raise questions if there were any doubts, contradictions, etc. in order to clarify what has been consistently muddy information (and disinformation) about our involvement in Iraq? In that case, I think we need to conclude that Petraeus DID NOT do his job, and that Clinton DID do her job. He tried to spin, she tried to dig more deeply into the contradictions of the evidence she had access to. It was NOT her job to come up with an accurate estimate of the casualties--how could she?

Posted by: Jess77 | October 19, 2007 3:15 PM | Report abuse

"The statistics show a clear downward trend since last winter."

This is like saying that spring retail sales in the United States showed a downward trend from last fall. Of course they do... retail sales are ALWAYS highest in the months right before Christmas.

Only someone completely ignorant of what is happening in Iraq would make the comparison you made. You're not a fact checker, you're just another shill for Hiatt and his bloodthirsty editorial board. Get a job at McDonald's where you can display at least a scintilla of integrity by giving people the correct change.

Posted by: paul lukasiak | October 19, 2007 3:27 PM | Report abuse

Eric A,
1.2 million does not pass the "Makes Sense" Test. From April 2003 to mid Sept 2007 is approximately 1,605 days. 1.2 million deaths equals an average of almost 748 per day or 31 deaths every hour, 24 hours a day, seven days a week, non-stop. And where do you suppose they are hidding those 1,600 mass graves with approximately 750 bodies in each one?

Posted by: HisRoc | October 19, 2007 3:39 PM | Report abuse

Having read a few of these "fact-checker" articles, I now know why so many articles in the media are so far off the mark: lazy fact checkers...
At least this one wasn't quite as obviously flawed as the one regarding "An Inconvenient Truth" which was based on not scientists' views, not peer reviewed sources, but a judge in a civil case... OMG.

Posted by: FrankD | October 19, 2007 3:54 PM | Report abuse

i wouldn't believe much of what petraeus says because he is a yes-man for bush. anybody who aligns themselves with him cannot stand alone apart from that evil dictator. so why in the world would i or anyone trust anything that a 'bushbaby' would say?!!!


Posted by: hemnebob | October 19, 2007 4:03 PM | Report abuse

Another question on the data from "Iraqi Government Bodies" is, which data from which government bodies, and when? Did they consistently use the same ones over the period? It seems not.

For example in January 2006 the chart above gives only 500. But the official figure for the Baghdad morgue alone that month was 1,068:

It would appear then that the Baghdad morgue is not one of the "Iraqi Government Bodies" included in the January 2006 figure.

The December figure in the chart is about 3,000. That's almost the same number given in the UN report for that month (2,914) which is combines "data from hospitals compiled by the Health Ministry and from the Baghdad morgue".

So it looks like Baghdad morgue *is* one of the "Iraqi Government Bodies" being included in December.

So which Iraqi Government Bodies are being included or left out in any given month across the period? By picking and choosing which ones to include and which ones to leave out in any given month, someone could pretty much make up any trend line they wanted.

Posted by: Leslie | October 19, 2007 4:57 PM | Report abuse

FACTS: Petraeus was incharge of all this equiptment..
AK-47 rifles: 110,000
Pistols: 80,000
Body armour pieces: 135,000
Helmets: 115,000

Gen. David Petraeus you lost all this equiptment or did you let the Carburton group steal it for the mercs. Petraeus cant watch these items, now he is incharge of real lives.

This is a boy who wants to work for the carburton grp. A boy who is watching real men die and he still continues to steal from and lie to America.


Posted by: 1-20-09 | October 19, 2007 5:30 PM | Report abuse

Truth may be the first casualty of war but the murderers are usually far left advocacy groups.

The Johns Hopkins Study is a joke that no one reputable will defend any more. The local pollsters were the same crew that generated huge anti-sanctions numbers for Saddam Hussein. Garfield recommended them to the Johns Hopkins people, in fact. The sample was so small that a little fudging here and a bit of bias there, and voila, 600,000 extra dead. Do some basic homework on the quality of research before endorsing garbage please.

The evaluation of the number of dead and the trends involved is not the problem here: he used the best information available and stayed away from the ideological nutcases on either side.

The problem is the decision to call it a tie when it clearly isn't.

The argument that Clinton's job is "to question" so it's ok for her to lie, intentionally or with gross negligence, should convince no one.

One more thing to remember. The number of civilian casualties should be compared to our best estimate of how many would have died if the US had not liberated Iraq. If the US had not gone in, Saddam and Sons would have kept power: they murdered about 750,000 Iraqis over the last 15 years of rule. A best estimate is that they would have kept the slaughter going at 50,000 Iraqis per year. Using the best estimate from IraqBodyCount, the DoD, the UN, and others, that means the US has saved, NET, about 125,000 Iraqi lives so far. Or do you just want to pretend that if the US had not gone in that Saddam and Sons would have found goodness and light somehow?

Huh. Your country has done good in the world. I know how awful that is for you to have to read...

Posted by: Michael Korcok | October 19, 2007 5:36 PM | Report abuse


Posted by: lonewolf | October 19, 2007 5:41 PM | Report abuse

To Matt Clifford:

Bravo! You captured the one and only salient point in all of this.

Be it 1500 or 1200, they are still dead, it IS a CIVIL war, and the IRAQIS pay a terrible and continuing price for our occupation. As tragic a sacrifice is the loss of one of our troops, it is equally tragic that 9 Iraqis lose theirs. No amount of "fact-checking" will bring any of them back, not the Americans nor the Iraqis.

Posted by: jade7243 | October 19, 2007 6:22 PM | Report abuse

All this data is relatively meaningless. As Joe Biden pointed out - this "decrease" means we are down to about "only" 1,000 attacks on US forces a week. The real issue is what our continued presence there means to Iraqis and to the situation in the greater ME. That is easier to put a finger on. As long as we stay, al qaeda has a justification for what it is doing in Iraq, at least to enough Iraqis to continue its presence. As long as continue to stay we are a recruiting poster for jihadists everywhere.
Bottom line, bush made a huge mess, and wants to make sure that the Democrat who follows him into office will have to clean it up. Then a decade or so from now, after a coordinated propaganda campaign the neocons will blame the failure of the war on Democrats. Facts have never ever stood in the way of conservative ideology.

Posted by: Brandon | October 19, 2007 6:37 PM | Report abuse

All of the organizations and newspapers that collect data on Iraqi casualties show a steady decline in civilian deaths in 2007. Petraeus however saying there was a 45% decline overlooks the work of the Pentagon and their quarterly reports to Congress on Iraq. In each of the reports issued in 2007, the number of deaths for previous months and years has gone up. The main reason given is that they are backlogged with stats, and don't get to them until months after a report is issued. On average the number of deaths have increased anywhere from 20-70%. There's no reason to believe that the Pentagon is caught up with their work, and won't revise their numbers upwards when the next report is issued to Congress. Petraeus was obviously trying to paint the most positive picture possible of the surge and took a number that might well be true for that week, a 45% decline, but one that most likely will not stand up after a couple months.

Posted by: motown67 | October 19, 2007 7:01 PM | Report abuse

People talk about DEADs as nothing more than a number. Put the label AMERICAN in front of that number and see how it feels. These numbers are brothers, sisters, mothers, fathers...;.....

Posted by: Sam | October 19, 2007 9:31 PM | Report abuse

Here's a piece a wrote at the end of Sep. 07 on the reporting on Iraqi casualties during the surge.

Is Violence Down Since The Surge?

When General Petraeus testified to Congress in September claiming that violence was dramatically down in Iraq, it set off a minor controversy. Some challenged his numbers, while others held them up to say that the surge was working. A analysis of military, think tank, and press reports show that the military is playing with the numbers, but that overall violence is down in Iraq since the surge.

Controversy Over The Exact Numbers ...

No one disagrees that Iraq set a new record for violence by the end of 2006. After the February 2006 bombing of a Shiite shrine, militias such as Sadr's Mahdi Army steadily increased their attacks upon Sunnis, reaching a crescendo in December. What happened in the country after President Bush announced the surge in January 2007 has been a matter of debate. In September General Petraeus claimed that violence was down 75% in Baghdad and 50% overall in the country. Several think tanks and newspaper articles challenged Petraeus' numbers.

What they found brought up several important questions about how the U.S. military collects and reports on violence in Iraq. Kirk Johnson who was the deputy director of the office in the U.S. Embassy in Baghdad from 2006-2007 that collected this information and now works for the Heritage Foundation said that U.S. reporting on average misses 35-65% of the violence. That's because it focuses upon sectarian violence, which is meaningless for homogenous communities like the Sunni west and Shiite south, it undercounts civilians because it doesn't have the means to collect all the information, and doesn't collect consistent information from areas under Iraqi control. Under Petraeus the military has been trying to refine their body counts, but intelligence officers in Baghdad who collect this information admit that many times it comes down to a subjective decision on whether a body found should be counted as a sectarian or regular murder. For example, if a body is found in an area that is known as a flashpoint between Sunnis and Shiites it is usually counted as a sectarian death. Likewise if it shows any signs of torture or being tied up it's counted as sectarian. However if bodies are found in a largely peaceful area, it is usually not included. Likewise an even more complicated occasion is when a body is found and there is no way to tell which sect it belongs to. The soldiers then have to make an educated guess as to the cause of death and whether to count it or not.

Another difference found was between what the Pentagon reported to Congress in September and what Petraeus presented that same month. A Pentagon report had a chart on average daily deaths and wounding of U.S. and Iraqi forces, and civilians, while a Petraeus' chart covered sectarian deaths per month. While the two had different kinds of information, the general trends should've followed the same pattern with the Pentagon report having higher numbers than Petraeus because it was covering more people. However, Petraeus numbers were much higher before the surge, and much lower since the summer.

A more disturbing finding is that the Pentagon has been systematically ignoring the greatest single cause of sectarian deaths in Iraq until September 2007. The Department of Defense is required to deliver quarterly reports to Congress on progress in Iraq. In 2007 there have been three reports in March, June, and September. Each one of these reports has revised its numbers upwards over the same months for violence in Iraq. The military admitted that the reason why violence went up in the September report from the previous two was because it was the first to include car and suicide bombings. These are the major causes of mass casualties yet they were never included before.

From this analysis it's pretty clear that the U.S. military has been systematically undercounting the violence in Iraq. Part of it is caused by having to collect information during a war with limited manpower. Another major cause has to be political pressure. It's hard to say that the U.S. is winning and that Iraq is moving towards a functioning democracy with so much death and destruction constantly being reported. There has been a huge drop in public support for the war at home as a result, and therefore it would only be natural to start playing with the numbers to defer criticism.

... But Trends In Violence Decreasing

Despite the controversy over the exact numbers and how they're collected in Iraq, the general trends show that violence is down since the surge. The Council on Foreign Relations released a report comparing reports by General Petreaus,, Iraq Body Count, the Associated Press, Reuters, the U.N., the Brookings Institutions' Iraq Index, the Washington Post, and McClatchy Newspapers. All of the reporting follows the same general trends: violence increased during 2006 with a high during the winter, and then decreased during 2007. Violence is still above 2006, but if the numbers continue to decline they will reach that level eventually. The one major difference is that four of the six sources record an increase in violence since June, while the military and Washington Post show a continued decrease. Whether this upward trend proves true will take until 2008 to determine because the numbers are always changing, and two months are a small fraction of an entire year.

What is now up for debate is the cause for this general decrease in violence in Iraq. When Petraeus reported to Congress he was careful to say that the surge has only been one factor for the decline. It's without argument that an increase in U.S. forces, especially with the new task of protecting the population has decreased violence in the areas where they have control, which is about 50% of the capitol. Another reason that has been brought up is the ethnic cleansing of Baghdad. With the majority of Sunnis having been forced out, there are simply fewer Sunnis to kill. Another probable cause for the decrease was the decision by Moqtada al-Sadr to stand down his Mahdi Army militia in January when the surge was announced. He made the tactical decision to stand out of the way of the new U.S. policy because he hoped it would focus on the Sunnis, and allow him to take over more of the capitol. The largest drop in violence actually happened from December 2006 to February 2007 before large numbers of additional troops were even sent to Iraq, so Sadr's decision and ethnic cleansing were probably major causes in this change. It's from a combination of all three of these factors that violence is down in Baghdad.


There will probably never be an accurate count of the violence in Iraq. Depending upon which source you ask, the numbers vary widely. For example, unofficially, the Iraqi Interior Ministry said that there was 2,318 civilians killed in August, up from 1,980 in July, while General Petraeus' report to Congress said there were only around 1,500 in August. The causes for these discrepancies are three fold: a lack of personnel, the difficulties of counting dead during a war, and the pressure to keep the numbers down during an unpopular conflict. Despite these problems, violence does appear to be declining for the first half of 2007, however, it's still above 2006 levels. Whether this trend will continue is an open question, especially because the surge will end by early 2008. While sectarian violence appears to be down in central Iraq, the country is always changing with Sunnis fighting Sunnis in the West and Shiites increasingly fighting Shiites in the South. How the U.S. deals with these will be the major issue in 2008 and beyond.

Posted by: motown67 | October 19, 2007 9:54 PM | Report abuse

And here's the sources:


Government Reports:

Department of Defense, "Measuring Stability and Security in Iraq," September 2007

Fischer, Hannah, "Iraqi Civilian Deaths Estimates," Congressional Research Service, 9/5/07

National Intelligence Council, "Prospects for Iraq's Stability: Some Security Progress but Political Reconciliation Elusive," National Intelligence Estimate, August 2007

Think Tank Reports:

Biddle, Stephen Friedman, Jeffrey, "The Iraq Data Debate Civilian Casualties from 2006 to 2007," Council on Foreign Relations, 9/25/07

Bruno, Greg, "Iraq Security Statistics," Council on Foreign Relations, 9/12/07

Cordesman, Anthony, "Iraq's Insurgency and Civil Violence," Center for Strategic and International Studies, 8/22/07

Johnson, Kirk, "Understanding Violence and Civilian Casualty Rates in Iraq: An Insider's View," Heritage Foundation, 9/10/07

Goldenberg Ilan, "More Fuzziness,", 8/30/07

Korb, Lawrence Biddle, Stephen, "Violence by the Numbers in Iraq: Sound Data or Shaky Statistics?" Council on Foreign Relations, 9/25/07

National Security Network, "Drop in Violence?" 8/30/07


Ambramowitz, Michael and DeYoung, Karen, "Petraeus Disappointed At Political State of Iraq," Washington Post, 9/8/07

DeYoung, Karen, "Experts Doubt Drop In Violence in Iraq," Washington Post, 9/6/07
- "What Defines a Killing as Sectarian?" Washington Post, 9/25/07

DeYoung, Karen and Tyson, Ann Scott, "Military Officials in Iraq Fault GAO Report," Washington Post, 9/5/07

Fadel, Leila, "Despite violence drop, officers see bleak future for Iraq," McClatchy Newspapers, 8/15/07
- "Security in Iraq still elusive," McClatchy Newspapers, 9/7/07

Glanz, James, "Civilian Death Toll Falls in Baghdad but Rises Across Iraq," New York Times, 9/2/07

Gordon, Michael, "Hints of Progress, and Questions, in Iraq Data," New York Times, 9/8/07

Hurst, Steven, "Violence lessens in Baghdad as it grows elsewhere," San Francisco Chronicle, 8/26/07

Lardner, Richard, "Defense agency chart shows scant progress," San Francisco Chronicle, 9/10/07

Michaels, Jim, "Major attacks decline in Iraq," USA Today, 8/13/07

Reid, Robert, "August particularly deadly for Iraqis," San Francisco Chronicle, 9/2/07

Sudarsan, Raghavan, "No Relief From Fear," Washington Post, 9/5/07

Susman, Tina, "Troop buildup fails to reconcile Iraq," Los Angeles Times, 9/4/07
- "U.S. defends sectarian death figures," Los Angeles Times, 9/13/07

Youssef, Nancy and Fadel, Leila, "What Crocker and Petraeus didn't say," McClatchy Newspapers, 9/10/07

Posted by: motown67 | October 19, 2007 9:56 PM | Report abuse

Unbelievable! Mainstream media finally makes the often difficult effort to take a stab at the truth and the facts, then buries the article way deep in the website. Wonder if it'll ever make print.

Great Job Fact Checker! Our country needs you.

Posted by: John Doe | October 19, 2007 10:29 PM | Report abuse

The bottom line is that if we did not invade Iraq, we are not responsible for one Iraqi death. Futher, HRC and all Republican candidates for president are promising at least another 5 years of war. How many more deaths are coming?

Posted by: TW | October 20, 2007 11:02 AM | Report abuse

More misleading commentary from The Fact Checker.

While it's correct that deaths were down from the "height of the sectarian violence in December," though by less than 45%, are we to ignore entirely that what was being reviewed in September was the surge, which was only starting in the end of January?

If you look at the Generals own data, the red bar above, deaths were higher in July and August than they were in February, at the start of the surge. Indeed, they were higher every month except June.

Posted by: Brian Blake | October 20, 2007 3:27 PM | Report abuse

Here's a breakdown of the sectarian murders that are reported by the Pentagon in their "Measuring Stability and Security in Iraq" reports to Congress from March 2007, June 2007, September 2007. With each report the number of deaths for the same previous months increase. For example, December 2007 the month with the highest level of deaths increased from approximately 1250 in the March report to around 2100 by September 2007. Given this constant increase the number that Gen. Petraeus quoted for the 45% decrease will most likely not exist by the next Pentagon report.

Sectarian deaths: March 2007 Report/ June 2007 Report / September 2007 Report
June 2006: 990 / 1000 / 1200
July 2006: 1190 / 1390 / 1600
August 2006: 750 / 900 / 1100
September 2006: 1190 / 1210 / 1250
October 2006: 1010 / 1600 / 1700
November 2006: 900 / 1410 / 1950
December 2006: 1250 / 1610 / 2100
January 2007: 1180 / 1500 / 1800

June 2007 Report / September 2007 Report
February 2007: 700 / 1200
March 2007: 600 / 1050
April 2007: 610 / 1050

September 2007 Report
May 2007 1050
June 750
July 1060
August 900

Posted by: motown67 | October 20, 2007 6:55 PM | Report abuse

"... You can't compare the number of attacks in the blistering hot Iraqi summer to the number of attacks in the winter and come up with anything meaningful. The real comparison is between the number of attacks in any given month and the number of attacks in the _same_ month one year ago. In every month the number of attacks per month has risen when compared in this method." -DK

Actually, a better way to adjust for seasonable variability in trends is to add the preceeding 12 months' data to the current month's data, and then divide by 13. This "statistical manipulation" controls nicely for seasonality, without adding bias to the resulting figures. Its biggest drawback is that it requires at least 13 months worth of data, which limits its usefullness in assessing early data, but not in assessing our current data. This is simple enough to do, but I don't have the raw data. Maybe the fact checker could have an intern do the math and show us the results.

Posted by: Anonymous | October 20, 2007 8:18 PM | Report abuse

At what point do we start to look at facts that make a difference. OK, maybe the number of deaths went down. Am I suppose to be happy about that? The issue is, why are we there, why are we spending my tax dollars and the tax dollars of my children, my grandchildren, my great grandchildren, etc. etc. on this foolishness.

We can't find a better target for the wealth of this country than to fund Halliburton, Blackwater, et. al. to continue their involvement in a war based on lies.

Is anyone thinking here, anyone at all?

Posted by: Dr. T | October 20, 2007 10:02 PM | Report abuse

Dr. T: You are exactly right. And we stand a better chance of securing Americans everywhere if we are strong on the inside first, instead of going abroad and fighting while growing increasingly hollow on the inside. Take all the money we give to the war and fund some kind of non-profit health care for Americans, or give every kid a square meal a day or a computer or a book - I'm sure we can think of something to do with allof these tax dollars other than give them to Halliburton and Blackwater. How embarrassing what we are doing in Iraq.

Posted by: Lisa | October 21, 2007 7:32 AM | Report abuse

"She read talking points almost verbatim from a Democratic think tank, the Center for American Progress, which argues that the violence has not gone down."

I thought this was a Fact Checker column?

Your tone ("talking points", etc.) sounds critical, but you don't actually seem to be trying to verify whether the Center for American Progress gave (and Clinton repeated) accurate information. If it was accurate, then it was -- your protest notwithstanding -- perfectly reasonable for Clinton to read it on the floor. If it was inaccurate, then you should say so and explain why.

Posted by: Crust | October 23, 2007 10:51 AM | Report abuse

Talk about chickening out. Why doesn't the Post just admit the obvious? Iraq has descended into such chaos since our occupation that no one has any clue what is going on over there anymore!

Posted by: Mark | October 25, 2007 4:02 PM | Report abuse

I'm appalled by the fact that you guys can't see whats right in front of you! 3 pinocs for clinton for sure.

Posted by: WOW. | November 16, 2007 6:04 AM | Report abuse

Hello, nice site :)

Posted by: Brin | December 2, 2007 12:15 PM | Report abuse

hoczrd mqlb zixcdao laskj ylekcvs iyth qylcxwfdj

Posted by: rnxhseatl wynumxp | January 2, 2008 9:55 PM | Report abuse

bhmdfoejv lnhgv tajscedoq aoljt xfbi qclw mkvqzytr

Posted by: dgpmer ykjlepxof | January 2, 2008 9:58 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2011 The Washington Post Company