A Clarifying Month
don't let those msm-reported civilian casualty figures fool you
By Charles Bird Posted in War — Comments (19) / Email this page » / Leave a comment »
First off, I'm getting the graphical information from Engram here and here, and the numbers are based on the Iraq Coalition Casualty Count which, from what I've seen, has the best available data. In the last three months, we've been at full manpower and our operations have been highly kinetic. Despite more troops and more action on the front lines, military casualties were relatively low in July and August.
In terms of civilian casualties, July and August are higher than June, but lower than the first four months of the surge operation (which basically started in February of this year). In the graph below, you can also see the trends since the Golden Mosque bombing in February 2006, where al Qaeda successfully triggered an upswing in sectarian violence. They tried to start a civil war, and it's still working in parts of Baghdad.
Read on...
But the above graph only tells part of the story. The coalition focused its efforts on Baghdad and the "belts", and that strategy has worked. Civilian casualties in Baghdad are way down.
Importantly, most of the decline in civilian casualties is due to the drop in execution-style killings, which primarily involves Shiite death squads targeting and slaying Sunni military-age males. With Muqtada al-Sadr announcing that he will stand his JAM militias down, I expect the extra-judicial killings will drop even further in September.
[The above graph was accidently left out of the original post--Ed.] With nationwide civilian casualties nearly unchanged in August and with civilian casualties in Baghdad lower, the obvious conclusion is that civilian casualties were higher outside Baghdad. Sure enough:
So why is this number higher? After all, we didn't reduce forces in the outlying areas. If anything, we increased our numbers both in Baghdad and its suburbs and exurbs. The answer is al Qaeda and its like-minded affiliates are still in play, and they're murdering civilian wherever they can:
Most of the difference between July and August is attributable to a single terrorist attack, the unfortunate victims being the pre-Islamic Yezidis. I picked the title of the post as "clarifying" because al Qaeda & Co. further clarified its major role. It bears repeating: al Qaeda chose Iraq as the central front in its war against us, and they remain public enemy number one in Iraq. Not far behind are Iranian-supported Shiite paramilitias. With al-Sadr standing down, the coalition will have an easier time targeting the "rogue" Shiite squads. Fred Kagan summarizes the make-up of al Qaeda in Iraq:
AQI, as the U.S. military calls it, is around 90 percent Iraqi. Foreign fighters, however, predominate in the leadership and among the suicide bombers, of whom they comprise up to 90 percent, U.S. commanders say.
Most of the suicide bombers are Saudi, which tells us something about that society. Most of the senior leadership is foreign born, and so concerned were they that they made up a fictional Iraqi leader (al-Baghdadi) out of whole cloth to hide that fact.
On the national political front, not much has changed, and we're going to fall short on the eighteen benchmarks. Five Iraqi leaders came together last week and worked an agreement on power-sharing, de-Baathification and oil legislation, but the various political parties were less than receptive. The oil legislation appears to have the best chance of passing in the near term, but the rest is going to take more time, perhaps quite a bit more time. So far, political advances are occurring at the local levels, and David Kilcullen (Australian advisor to the U.S. Command) has an excellent piece on how the tribes are moving the stakes.
The Iraqi Army is progressing, and national politics and the Iraqi National Police are faltering. Another concern is the corruption, both in the Iraqi government and the American, which is flat out maddening. The political timeline in DC is this month, and there are going to be a lot of heated exchanges in the coming weeks. As for me, I'm still reserving final judgment on the current counterinsurgency strategy until year end.
« We need more COIN in the Afghan realm — Comments (0) | "Oh, The Tide Is Turning!" — Comments (0) »
A Clarifying Month 19 Comments (0 topical, 19 editorial, 0 hidden) Post a comment »
HTML Help Central for Red Staters
I voted Harry Browne in 2000, but will never vote Ron Paul.
Mike Gamecock DeVine @ The Charlotte Observer
www.race42008.com
www.hinzsightreport.com
www.theminorityreportblog.com
"One man with courage makes a majority" - Andrew Jackson
Charles' post seems to have gone over your head with the celerity of a supersonic jet.
HTML Help Central for Red Staters
I voted Harry Browne in 2000, but will never vote Ron Paul.
As well as what our point was re: Democrats saying the Surge was a Failure before the Surge had even surged all the way, and before those Surged troops were doing anything Surgey.
HTML Help Central for Red Staters
I voted Harry Browne in 2000, but will never vote Ron Paul.
that if the troops weren't doing anything "surgey" back then, where do you begin the measuring for observable results? But peace, I don't want to pick a fight in your house. If I have misunderstood, then it will be apparent to all those who read my post - wouldn't be the first time I've been an idjit.
Call the first 4 months phase one. Petraeus was confirmed in late January and immediately started implementing the new strategy. We reached full manpower in June, starting Operation Phantom Thunder, so you can call that phase two. What I reject is your calling me duplicitous or saying that I "played around with the numbers". That is either a posting rules violation or you're bordering perilously close to it.
Surely "seems a bit duplicitous" is not a violation of the posting rules. I will happily stand corrected if I'm wrong.
And perhaps you have not been personally involved in stating something along the lines of "criticism of 'the surge' at this point is dumb/incorrect/politically opportunistic because it hasn't even started or isn't really in full effect yet." I don't know, thus you are free of any charge on my part. Sorry if I ruffled your feathers.
My point does stand, though, that if you are going to track success by starting with February metrics, then perhaps I should rethink my position, formed in part by postings by Streiff and others on this blog, that early critiscm was inapt. To me, it seems trying to have it both ways.
One way or the other, I hope you're right about the improvement in the situation. No need to bring the hammer after me.
Your implication is that I wrote something dishonestly, however "safely" you think you couched it.
The earlier criticisms of the surge strategy had to do with hard partisans like Harry Reid declaring--way back in April mind you--that it failed, two months before it had even reached full strength. There is nothing wrong with using February metrics because February is when the new strategy began. September is supposed to be the earliest time when Petraeus has seen enough results to report to Congress on the effectiveness of the plan.
You also are under a misimpression if you think that I was tracking "success". I specifically rendered no judgment on the success or failure of the plan, deferring my opinion 'til year end, but in full disclosure I do support the current strategy and have consistently written that it should have been implemented at least two years earlier.
__________________________________________________________
But the Hebrew word, the word timshel—‘Thou mayest’— that gives a choice. It might be the most important word in the world. That says the way is open. That throws it right back on a man. For if ‘Thou mayest’—it is also true that ‘Thou mayest not.'
extraordinarily superficial if you are commenting on the strategy at play rather than media coverage of the war?
I'm not really sure what one or two spectacularly successful attacks tell you about what is going on though I rather think not much.
Combining deaths from mass casualty attacks and extrajudicial executions also seems a lot like adding apple and oranges and coming up with strawberries. They are really very, very different things and symptoms of very, very different problems.
"A man can never have too much red wine, too many books, or too much ammunition." -- Rudyard Kipling
If you look above, you'll see one more graph, on extra-judicial killings, so the above graphs specifically split out EJKs and suicide terrorist attacks. Speaking generally, I can't see how it hurts to know the numbers and the nature of the casualties, just as it doesn't hurt to know the qualitative factors as reported by Kilcullen. If I were to add more figures, they include the numbers of Iraqi Army battalions at Level 2 status or better. With civilian casualties, it's relevant to see if the killings are done by paramilitias (which are indicators of civil war or sectarian strife) or by suicide bombers (which speaks more to the War on Terror).
to know something. But only rarely does incidental knowledge rolled together mean very much.
"A man can never have too much red wine, too many books, or too much ammunition." -- Rudyard Kipling
Is this analysis very accurate here? Seems like you're comparing apples and oranges here. Comparing casualties on a month to month basis implies that they don't have normal seasonal fluctuations, which I imagine they would. I volunteered as an EMT for years and one of the things we knew from looking at the numbers were that different parts of the year tended to have different amounts of injuries (more murders in the summer, more care accidents in spring, etc).
Has anyone done a comparison between this year and the other 4 years? That would seem to conclusively say whether or not "the surge" was actually affecting troop casualties one way or another (June 2006 versus June 2007 for instance). Taking a quick look on http://icasualties.org/oif it looks like troop casualties have gone up since last year, for the same time periods.
First, as you can see above, the Golden Mosque bombing resulted in an upswing of civilian casualties and ramping up of the violence since Feb-2006. Second, our strategy substantially different, starting last February. I don't see why you think it more valid to compare Jul-2006 to Jul-2007 when our operations have significantly changed between those two time periods.
Charles Bird said:
"I don't see why you think it more valid to compare Jul-2006 to Jul-2007 when our operations have significantly changed between those two time periods."
First, are we talking about civilian casualties or military? My comments have (and will be) strictly related to military casualties because with the huge discrepency in civilian casualty stats from different agencies (and the already shown inaccuracies in all of the collection methods used so far), trying to draw any sane conclusion about trends in civilian casualties is tenuous at best. Also, given the large exodus of Iraqis from Iraq it becomes even harder to accurately measure casualty rates for those who are "missing".
Second, the problem here is that your comment also negates the point you were trying to make, that is that the surge is having an affect. If you are saying that because the strategy is so different that doing historical comparisons is inappropriate, then the only logical conclusion is that you must start from the beginning of the surge. If that is the conclusion, than any reputtible analyst will tell you that a 6 month sampling period on a 4 year+ conflict is statistically irrelevant. At the very least you'd have to wait a year or so to truly analyze things. As a sidebar, since there have been several tactical changes in military operations since Iraq started, "the surge" shouldn't be viewed as somehow unique but rather the latest tactic being deployed. Seems to me that changes in tactics over long military conflicts is fairly common and doesn't change the fundamental trend of things like casualty analysis.
However, I don't think the situation of "the surge" is unique enough to negate doing a historical analysis, because the numbers are VERY clear that there is a seasonal relationship of casualties. To prove it to yourself simply divide every year we've been in Iraq into quarters (Jan->March, April->June, July->Sept, and Oct->Dev). Then average out military casualties in every quarter. The reason for using quarters is to normalize for small pertibations that occur as a result of large scale casualty incidents and to give a clearer overall picture of the cyclical nature of the data (also, quarters is a generally accepted way to analyze data like this).
The raw troop data (by quarter): http://thumbsnap.com/v/j0LMf3LN.png
graphical representation of troop data by quarter/year: http://thumbsnap.com/v/wTv5NaDI.png
average quarterly casualty data: http://thumbsnap.com/v/hNuqopvZ.png
A quick look at the data shows that indeed, the casualty data follows a VERY easily identifiable trend (with the exception of Q1-2003 because we don't have an entire quarters worth of data). The data follows a very predictable sign wave of casualty data that is best summed up by looking at the "average quarterly casualty data" graph I presented. In fact, the data follows some very simple rules (except for Q1-2003 for reasons already stated):
1. Q1 always shows a huge drop in casualties compared to Q4 of the previous year (Q1-2007 corresponds to the start of the surge).
2. Q2 always has more casualties than Q1
3. Q3 has as many if not slightly more casualties than Q1, but is always less casualties than Q2.
4. Q4 is always the quarter with the highest casualty rates
This is the basic message of my post. That no matter what the particulars of the timeframe were, the data over several years has shown a predictable trend from quarter to quarter. The implications for "the surge" are that in order to really judge its effectiveness it needs to be judged versus historical trends, not just month-to-month changes. Based on historical data casualty rates have actually increased this year over previous years, I make no fundamental judgement as to why, but the numbers are very clear.
One disturbing trend is that Q4 is starting, and it has historically been shown to have the highest casualty rates. I'm wondering whether or not September was intentionally chosen to decide on the efficacy of "the surge" because it was known that the casualty data from Q4 would be much less favorable than the Q3 data?
Finally, IMO it would be very illogical to attempt to negate incidents such as the mosque bombings, and their response, because those incidents are critical in gauging overall success of any tactic. To say, "if the bombings had not occurred..." is misleading. Kind of like talking about the Vietnam War without the Tet offensive or WWII without Pearl Harbor (not to say the mosque incident are of the same scale, but in all cases they resulted in an overall escalation of conflicts).
Unless I'm mistaken, the numbers for August involve deaths from the two helicopter crashes, making the numbers for the month artificially high. I don't know if it would be right to remove them, but as they were accidental and not "combat" deaths, including them makes it look as if the surge is having less effect than it really might be. You are probably better off keeping them in for the sake of consistency, but keep in mind that if there were four crashes instead of two, people would be using the same numbers to say that the surge has made things worse.
Combat or not, are not a good measure of progress. The "new" strategy involves many more offensive operations and much more exposure of smaller units on the streets. Compare the casualties of May and June, 1944. Were we "losing?"
Incidently, the dynamics of driving the organized enemy out of strongholds causes much civilian hardship as they attempt to establish themselves by force elsewhere. A strategic benefit to us in two ways as the insurgents are forced to expose themselves more and the locals are apt to resent the intrusion. Temporary civilian casualties in the new area are therefore not a good measure of insurgent success in that area.
I have been reading RedState postings on Iraqi operations pretty consistently since 2004, and have in fact been fairly convinced by better military minds than me here that criticism about "the surge" and its success was unfair and incaccurate up until the point that it actually went into effect, after we were fully ramped up this summer. Thus, maybe it was just a throwaway line, but playing around with numbers and now claiming the surge really started in February to show successes seems a bit, well, duplicitous.
But for the record, I hope you're right about the tea leaves in the numbers. We need success. I'm prepared at this point to accept failure if in fact that is what we're dealing with, but I hope you are right. And I agree with you that given the outlier attack on Yezidis that blowing up the civilian casualty number is, well, a bit duplicitous on the part of our media.