Iraq

A Military Expert Analyzes Petraeus' Slides

| Wed Sep. 12, 2007 4:15 PM EDT

Thomas Mowle, an associate professor of political science at the US Air Force Academy, a former member of the strategy branch, Multi National Forces-Iraq (MNF-I), and editor of Hope is Not a Plan: The Iraq War from Inside the Green Zone, offers these observations about Petraeus' slides, with reference to his testimony. He emphasizes that the below are all his own assessments, not official ones.

Slide 1: Interesting that the PKK is included, though not the MeK (Mujahedeen e-Khalq) or Pezak (the Iranian offshoot of the PKK, also known as PJAK). No reference to the Kurd-Arab/Turkomen violence around Kirkuk. All the yellow boxes list threats... except "anti-AQI Tribal Success" in Anbar --unless that success is being counted as itself a threat to Iraq. Foreign fighters appear to come only from Syria, not Saudi Arabia.
Slide 2: Probably the most effective slide. Long view, since October 2004. al-Askariyah shrine destruction appears to NOT have sparked an overall increase in attacks, but is merely a point on a steady increase from about March 2005 through December 2006. Mid-February listed as "Baghdad Security Plan," with testimony saying "forces began to flow in January." Most reports I've seen suggest the "surge" began in March; this has been difficult to track down. "Surge" was complete in June 2007, at which point the # of attacks drops rapidly. Not clear what is meant by "attacks:" does this include all acts of violence in Iraq, or only those aimed at coalition or Iraqi government targets? If the latter, that could explain the lack of a spike after Samarra, and also the discrepancy between the timing of this drop in attacks and the drop in deaths noted on later slides.
Slide 3: Starts in January 2006. Civilian deaths are down since Dec 06, by 50% in Iraq and 75% in Baghdad. Trend since Jan 06 is up 200% in Iraq and up 400% in Baghdad. Most of the decline is from Dec 06 to Feb 07 -- during the surge period, deaths in Iraq are down 15% and in Iraq down 33%. Baghdad now represents a smaller share of the violence: about 70% at the peak, but now only 33%. That is about its "share" of the Arab population of Iraq, but suggests that violence may have been tranferred to other places.

Advertise on MotherJones.com

Slide 4: Maps start in December 2006, Graph in May 2006. The different starting dates are very strange. Looking ahead, other charts start in June 2006, October 2006, June 2006, August 2006, May 2006, January 2007, 15 June 2007, July 2007, and November 2005: 13 different sets of graphics, with 12 different start dates. Very odd. In any case, what are "clearly ethno-sectarian" deaths? Would AQI on cigarette-smoking clean-shaven Sunni Arab count? In maps of Baghdad, the demographic characteristics of neighborhoods never changes -- is this meant to imply there have been no changes, ignore such changes, or mere difficulty in coming up with a new map?

Slide 5: Lots more caches found this year. Is "cache" the best measure? Are these caches the same size as before, or are they smaller, meaning no more weapons have been captured? Does finding more caches suggest progress in finding them, or that there are more caches to be found?

Slide 6: IEDs: trend is good, corresponds to the surge... "Total IEDs = IED explosions + IEDs found + IED hoaxes?" What exactly is an "IED hoax?" Are there many cases of truly faked IEDs, or does this include situations where soldiers see something they think is an IED, but isn't? Could the decline in Total IEDs reflect only a decline in hoaxes, leaving the real categories unchanged?

Slide 7: Anbar attacks. We know this good news story. Peak shown for al-Anbar is October 2006. Attacks down 33% before the surge, and another 60% while the surge got up and running. Suggests that the success in al-Anbar has nothing to do with the surge.

Slide 8: Trends in Salah ad-Din, Baghdad, al-Anbar, and Ninewah. Why not Diyala? Would that graph look so good? Violence trends up while surge beginning, then down once complete, except in al-Anbar where the whole trend is down, and in Ninewah, where the downward trend is very minor. Measure is of violence, which = attacks + "murders events." What is a "murders event?" Why do the slides shift from deaths, to attacks, to level of violence?

Slide 9: High Profile attacks -- fits the surge story. Peak is in March 2007, then down except for suicide vests.

Slide 10: State of AQI. Almost impossible to comprehend without a story, which is not provided.

Slide 11: Lots of security volunteers, mostly Sunni.

Slide 12: # of "Fully Independent" battalions unchanged from April 2006-present. # of "Iraqi Lead with Coalition Support" battalions unchanged from September 2006-present. # of "Fighting Side by Side" unchanged over that same period. One would think that units would move steadily from level IV to level I over the nearly two years depicted. Possibilities: 1) It takes much longer to progress in capabilities than I think, so the lack of change is reasonable; 2) Units are progressing in capabilities, but once they become "fully independent" they get so cut up in battle that they regress all the way back to "fighting side by side." 3) Units are progressing to "fully independent," and then vanishing. 4) Not much progress is being made in developing independent capabilities for Iraqi Security Forces -- which is OK as long as they are not being counted upon.

Slide 13: Notable lack of dates after July 2008. Unclear what is going on with the stars and their location. If forces are drawing down, three possibilities: 1) ISF will backfill the capability as they become ready ("we stand down as they stand up"). Based on prior slide, that could be a while. 2) ISF will not be needed, because coalition forces will not withdraw until violence is down and stays down, and will not go back up later. 3) Coalition forces withdraw, ISF are not capable, and ...Probably not a victorious scenario.

"All of the above are my own assessments, not official ones -- but I stand by them as a professional."

Posted with permission.

Cheers,

Tom Mowle
US Air Force Academy

Get Mother Jones by Email - Free. Like what you're reading? Get the best of MoJo three times a week.