« previous | MUCK HOME | next »
Today's Must Read
One week later, here it is: Gen. Petraeus' definition of sectarian violence.
Ever since the GAO report last week said it was "not clear" that the surge had contributed to a drop in sectarian deaths, Gen. Petraeus has been under pressure to explain his methodology. The GAO was agnostic on whether or not sectarian attacks had declined in recent months, citing that it required knowing a perpetrator's intent -- a task beyond the capabilities of the agency. But GAO was, at least inferentially, skeptical, noting that the broader pattern of attacks on civilians -- of which sectarianism is a proportion -- hasn't declined. And further reporting suggested problems with how MNF-I has tabulated sectarian casualties: one famous Washington Post story cited a senior intelligence official claiming MNF-I looks at where a bullet entered someone's head to divine sectarian intent.
Petraeus has disputed all of this. Yesterday, in Washington, Petraeus took a stab at an explanation. And in Baghdad, the Los Angeles Times reports, so did the U.S. military command, known as Multinational Forces Iraq, to combat the accusation that it's cooking the books to exaggerate the success of the surge. However, it's not exactly clear what that methodology tells us:
Stung by accusations that Army Gen. David H. Petraeus, the commander of U.S. troops in Iraq, had presented selective statistics during his testimony before Congress, the military released a statement here outlining its definition of sectarian violence: bombings, killings or other attacks committed by an ethnic group or religious sect against another, for purely sectarian purposes.
That seems a little circular. As I wrote last week, determining sectarian killings isn't a matter of determining intent. There's plenty of evidence from a body that a killing was driven by sectarian motivations. Victims of sectarianism "generally are males found without identification documents and shot execution-style. The bodies usually are blindfolded and bound at the wrists, and often bear signs of torture," writes the LAT's Tina Susman. It may be that MNF-I's methodology makes sense, and the GAO was unduly harsh. Or not.
Here's MNF-I's statement in full:
Multi-National Force-Iraq defines ethno-sectarian murder as a murder committed by one ethnic/religious person/group directed at a different ethnic/religious person/group, where the primary motivation for the event is based on ethnicity or religious sect.Ethno-sectarian violence is defined as an event and any associated civilian deaths caused by or during murders/executions, kidnappings, direct fire, indirect fire, and all types of explosive devices identified as being conducted by one ethnic/religious person/group directed at a different ethnic/religious person/group, where the primary motivation for the event is based on ethnicity or religious sect.
In our collection of data, a shot to the front or back of the head is not used to determine ethno-sectarian murder.
The number of ethno-sectarian murders has declined significantly since the height of the sectarian violence in December 2006. Iraq-wide, the number of ethno-sectarian deaths has decreased by over 55 percent, and it would have decreased much further if it not for the casualties inflicted by barbaric al-Qaeda bombings attempting to reignite sectarian violence.
It remains unclear why, as reported, the GAO, DIA and CIA have difficulty accepting MNF-I's definition of sectarian violence.

Comments (26)
Anonymous wrote on September 13, 2007 9:41 AM:I recall earlier in the week it was reported that "two intelligence agencies" approved the methodology of counting sectarian violence incidents. That's two out of a community of roughly sixteen.
So here we see that DIA and CIA, two of the most pre-eminent aren't the two.
So which ones are "the two?" Maybe Doug Feith's ex-CIFA organization? Maybe the Director of National Intelligence?
Karl Rove's the math lives!
david knowles wrote on September 13, 2007 10:05 AM:Enough of these silly patriot games. No matter how you slice it, Iraq is going nowhere fast. Obama has released his plan and it's a hell of a lot better than what we have going. Read about it here:
http://news.aol.com/elections-blog/2007/09/13/obama-has-seen-enough/
Anonymous wrote on September 13, 2007 10:17 AM:If sectarian violence is decreasing but overall violence is the same, that means criminal or other types of violence are increasing. Even if true, how is this a good thing?
JohnMcC wrote on September 13, 2007 10:31 AM:Only three things we need to know about counter-insurgency: 1. If you're counting bodies to determine if you're winning, you are losing. 2. If you are using artillery and airstrikes, you're losing. 3. If 60% of the population thinks attacks on you are 'justified', you've lost.
smr wrote on September 13, 2007 10:33 AM:You guys should go over and see how ridiculous the arguments are over at neoneocon.com are. unbelievable!
Bruce Rusk wrote on September 13, 2007 10:46 AM:There are several flaws in this methodology. One that comes to mind is this: the definition includes only "a murder committed by one ethnic/religious person/group directed at a different ethnic/religious person/group." But what about, for example, murders of people one the same group as collaborators? If a Sunni kills another Sunni for "collaborating" with the mainly Shiite Iraqi government, somehow this is not sectarian?
Steve wrote on September 13, 2007 10:48 AM:"If sectarian violence is decreasing but overall violence is the same, that means criminal or other types of violence are increasing. Even if true, how is this a good thing?"
Well it's not a "good thing", but here's the logic. If you reduce sectarian violence, specifically, then you have the potential to create a situation in which the sectarian parties can work out their issues. Presumably after they work out their issues, they can form some kind of coalition and then take on criminal violence.
Now that's the theory, but the reality is grimmer. A very strong possibility for a decline in sectarian specific violence is that Iraq is breaking apart along those lines. Sunnis in Shiia neighborhoods are either murdered or leave, and once they are all gone, sectarian murder rates decline. But of course.
Does that lend itself to a long term reconciliation, the entire point of the surge? No. It's like the beginning of a soccer game; it's just lining up the players on opposite sides of the field before the game begins.
KMM wrote on September 13, 2007 11:05 AM:"the number of ethno-sectarian deaths has decreased by over 55 percent, and it would have decreased much further if it not for the casualties inflicted by barbaric al-Qaeda bombings attempting to reignite sectarian violence".
Is not there a problem here? Why would Al-Qaeda bombings and casualties count as sectarian violence? Are they not an equal opportunity bomber? I can see the connection if the Al-Qaeda attacks caused the Shia to attack the Sunni or vice versa thereby increasing sectarian violence. But the staement clearly says that the casualties they were counting was caused by Al-Qaeda.
KMM wrote on September 13, 2007 11:09 AM:the number of ethno-sectarian deaths has decreased by over 55 percent, and it would have decreased much further if it not for the casualties inflicted by barbaric al-Qaeda bombings attempting to reignite sectarian violence.
How can that be? Why would Al-Qaeda attacks and casualties count as sectarian violence? They might inceite more violence...but the statement says the percent decline would have been greater except for causualties caused by Al-Qaeda.
Beau wrote on September 13, 2007 11:10 AM:Good point Bruce. another flaw is " for purely sectarian purposes". So if a Shite kills a Sunni for "sectarian purposes" who is also a business rival it is not "sectarian" violence?
It seems to me that we're have the clasic problem of "I can't tell you what it is, but I know it when I see it."
Betsy wrote on September 13, 2007 11:20 AM:I believe I heard on NPR this morning that the MFI does not count anyone shot in the head. They had a report by their reporter in Baghdad.
Bushie wrote on September 13, 2007 11:21 AM:If the number of sectarian targets is decreased, it's harder to kill in sufficient numbers to impress the MNF. So where in the stats is ethnic cleansing? Seems the Shia are driving the Sunni from their enclaves while the Sunni are driving Shia from Sunni enclaves.
david wrote on September 13, 2007 11:29 AM:How does the clarification of the definition even tangentially address the actual issue of the methodology of making the determination in each particular case?
paul wrote on September 13, 2007 11:30 AM:With a definition like that, no wonder they can make up any numbers they want. "Purely" sectarian is one enormous loophole, and "other group" is another. And of course, operationally speaking the definition is useless -- it provides no guidance at all about how to look at a body, the identity of a victim, or the identity of the people perpetrating a killing and decide whether which column to count it in.
It's a lot like Petraeus's refusal to answer whether there were any conditions whatsoever under which he would say the surge wasn't working.
Charlie wrote on September 13, 2007 11:38 AM:All of this seems to be beyond the point to me. Answer this simple question: Has the number of killings (regardless of motivation) decreased. Period. If ethnic violence is really down dramatically, then the total should at least decrease.
Granted, any society has a constant rate of deadly violence. I don't think it takes a PHD and a slide rule to understand that the monthly death toll in Iraq is not an acceptable societal norm. It is also not a leap of faith to say that the war is the major contributing factor in this.
So if the bodies keep piling up apace, regardless of the supposed intent of the killers, then the "security situation" has not improved as a result of the surge, not to mention an ongoing 4 year military campaign.
We need to stop losing reality in the minutiae. If your friends and loved ones continue to be shot, stabbed or blown up at an alarming rate, do you care the supposed intent of the attacker? No. Do you feel the "security situation" has improved? No. So why are we arguing about how the pencil pushers tally up the dead?
Dead is dean.
The surge has not been successful.
Middle Earth wrote on September 13, 2007 11:38 AM:From the moment Bush went into Iraq, Americans were screwed.
And that includes the Democratic Party, which is now being set up to take the fall. So the best hope for the Dems may be that Gen. Petraeus actually succeeds, over the next year, in significantly reducing ethnic tensions. It is a slim reed to hold onto, as they recognize.
I'm a severe skeptic on the likelihood of anything that looks like success in Iraq. But I don't think career public servants such as Ryan Crocker and David Petraeus are acting as partisan Republicans in their Iraq efforts. I think they both are sincere, experienced men attempting to retrieve what they can for America from Bush's catastrophe.
They may as well try, since the Democrats can't over-rule Bush and get the troops out, anyway. If the troops are there, they may as well at least be deployed intelligently, which is what Gen. Petraeus is doing. I wish them well in their Herculean labors. Because if they fail, I have a sinking feeling that we are all going down with them, including the next Democratic president. And their success is a long shot.
--Juan Cole
Batten down the hatches folks, it's going to get worse.
Tim Fuller wrote on September 13, 2007 11:40 AM:The reason for the perverse definition of success is that Das General Petrayusnik is thinking of his run for Preznit in 2012. Like everything else about this admin and it's fake war it's all politics all the time. Remember it's the fall season where the new shows..err strategies are rolled out.
Middle Earth wrote on September 13, 2007 11:48 AM:And it's going to get worse very soon:
1) Cheney has, since early 2001, been given executive authority over DOD. Take a guess what's up with the 'accidental' movement of tactical (read small) nuclear warheads from Minot to Barksdale AFB, a staging area for the ME? These warheads were mounted on missles, transported on wing pylons in violation of a long list of protocols. It has been reported that the President's personal code was used to authorize the transfer. What's going on here?
2) Two scheduled theater-wide exercises, NORTHCOM and TOPOFF, are being held during a period of extremely high threat. There's a lot of buzz on this. Could these exercises be the prelude to another paradigm-shifting national crisis?
3) Rumors circulating in DC that Karl Rove's departure resulted from losing the internal fight with OVP over Iran (who could have imagined that Turd Blossom was the voice of reason in this).
We ain't seen nothing yet, folks...
john mccutchen wrote on September 13, 2007 12:00 PM:"That seems a little circular. As I wrote last week, determining sectarian killings isn't a matter of determining intent. There's plenty of evidence from a body that a killing was driven by sectarian motivations. Victims of sectarianism "generally are males found without identification documents and shot execution-style. "
Strictly speaking Spencer you contradict yourself because you simply assert that it IS possible to INFER SECTARIAN INTENT from the body.
Paul Revere wrote on September 13, 2007 12:21 PM:Not to quibble but determining the intent of the killing is manifestly what it is all about
You've been scammed. The issue isn't sectarian violence. The issue is excess deaths during war as compared to deaths were there no war. Even Iraqi Body Count is going to under count that. If police are distracted chasing "terrorists" and fail to stop catch a "merely criminal" serial killer, that is still leads to excess deaths during war. This is why the Lancet's study should be considered definitive.
TomJ wrote on September 13, 2007 12:30 PM:If you read closely the definition above, 'ethno-sectarian murders' have declined, not 'ethno-sectarian violence'! It isn't clear if murders include civilian deaths or if ethno-sectarian deaths include only murders or any deaths caused by ethno-sectarian violence.
However, the bigger problem is using a single month of deaths as the highpoint.
What we should do is look at the figures from May-June when Republicans claimed that the surge really started. And year-to-year, or a rolling average would be much better than pure numbers.
illlich wrote on September 13, 2007 12:35 PM:post #3: "If sectarian violence is decreasing but overall violence is the same, that means criminal or other types of violence are increasing. Even if true, how is this a good thing?"
-- I couldn't have said it better myself.
I suspect that IF sectarian violence truly is decreasing, it's only because the various sects have simply tried to distance themselves from each other, becoming tired of years of neighbor-on-neighbor violence, combined with the fact that many are just leaving Iraq altogether, heading to Syria or Jordan or wherever. . . hmmm, maybe THAT'S the way to stop the violence, make EVERYBODY a refugee.
the MNF-I's logic certainly seems to be a classic case of "begging the question" a.k.a. circular logic: "we define sectarian violence as violence committed between different sects" (of course, how silly of me to ask, don't I feel stupid.)
anonymous wrote on September 13, 2007 2:20 PM:I'm sure it really matters to Iraqis whether they are killed by sectarian violence or non-sectarian violence.
Bottom line: civilian deaths have not decreased during the US occupation or in response to the surge.
Dead is dead.
Any decline in sectarian deaths may only mean that resources are being diverted from regular crime to mitigate sectarian violence resulting in an increase in non-sectarian violence and that if resources are rediverted to combatting non-sectarian violence that sectarian violence will again rise.
Dead is dead.
You can suppress sectarian violence in the short term with military action, but you can't eliminate it militarily or eliminate the causes - if you want permanent reduction in sectarian killings in Iraq then you will need a permanent force of 160,000 or more US soldiers.
Live (or die) with it.
Tim wrote on September 13, 2007 5:23 PM:If 'sectarian violence' is done by one ethic/religious group against a different ethnic/religious group, then Al-Qaeda (a radical Sunni organization) attacking the Sunnis of the Anbar Salvation Coalition is not sectarian violence. By the same token, Mehdi militia fighters blowing up Shiites isn't sectarian violence.
I'm beginning to see how this works.
Squeaky McCrinkle wrote on September 14, 2007 2:06 AM:I better understand where General Petraeus is coming from after seeing him on PBS New Hour.
He told a fairly wide-eyed Jim Lehrer that he carries the almost overwhelmingly heavy "Rucksack of Responsibility" for God's sake! And he's got to carry it while al-Qaeda is the "wolf closest to the sled", which I thought was a rather strange analogy for Summer in Anbar . . .
However, I am most curious to know the relationship between the "Rucksack of Responsibility" and the more famous "Cone of Silence", "Hour of Power" and "Sword of Damocles", not to mention the "Burden of Proof". Even Joe Liebeman can't lift that!
The General noted that "it's a big weight in this rucksack" and Iraq will be "a work in progress for a long time to come".
With men of this caliber I think we can certainly rely on that.
William Holt wrote on September 24, 2007 2:42 AM:Of course, I'm sure everyone is aware that comparing December 06 to the Summer of 07 is purely fraudulent.
Violence always goes down during the summer, ensuring that Petraeus September report would be able to spin the decreased violence as due to the surge.
As has been reported elsewhere, but not in the MSM, the summer of 07 had much less violence than December of 06, but had more violence than the summer of 06, or the summer of 05.
The man wants to be President. Admiral Fallon called Petraeus a "walking bag of ambition".