by Ethan B. Kapstein
Download the Full Article: Military Metrics
How do governments know whether they're winning or losing a military campaign? That question is devilish enough in the context of conventional wars with pitched battles, as conflicts often take surprising twists and turns en route to their endgame. It was more than sheer bravado that led Charles De Gaulle, who knew a thing or two about military operations, to declare in June 1940, "France has lost the battle, but France has not lost the war."
Precise knowledge of a conflict's progress is perhaps even more difficult when it comes to the counterinsurgencies now being fought in Afghanistan and, somewhat more surreptitiously, in places like Yemen. How do military leaders and policy-makers ascertain if they are "winning the hearts and minds" of the local population? What are the indicators of success?
Military history suggests that generals and public officials have often looked at the wrong data—the wrong metrics—for information and insight about what's really happening on the ground. The Vietnam War provides a poignant example (Nagl 2002; Kilcullen 2010). As late as the summer of 1974, a study group from the U.S. House of Representatives boldly asserted that "it is unlikely that the North Vietnamese can win a military victory" and it shared the view of the American Ambassador to Saigon, Graham Martin, that South Vietnam was now on the verge of an "economic 'takeoff' similar to those which have occurred in South Korea and Taiwan." The congressional group drew this conclusion from the lopsided difference in military casualties between North and South Vietnamese forces—the infamous "body counts"—which cast doubt on the ability of Hanoi to sustain the constant pummeling much longer. Needless to say, Saigon would fall to the North within nine months of that study's publication, with Ambassador Martin departing by helicopter from the U.S. Embassy's rooftop.
Download the Full Article: Military Metrics
Ethan B. Kapstein is a Non-Resident Senior Fellow at the Center for a New American Security and a professor of public policy and business at the University of Texas at Austin. A retired naval reserve officer, he has served as an Academic Advisor to the Counterinsurgency Advisory and Assistance Team at ISAF Headquarters in Kabul. His most recent book (with Nathan Converse), is The Fate of Young Democracies. The views expressed here are strictly his own and do not represent the opinions of any organization with which he is or has been affiliated.
About the Author(s)
Comments
Mr. Kapstein,
As someone who recently re-deployed from RC-East in Afghanistan working in the CJ35 as an Assessments guy, I thoroughly enjoyed your article and share many of the frustrations that you mention. Would like to offer some comments, thoughts, and observations.
The ISAF "Assessment process" was as ambiguous as you describe. Prior to the ISAF Joint Command (IJC) standing up in October of 2009, we at RC-East received little guidance or requests for input / data from ISAF, and it was unclear how they were assessing progress. When IJC become FOC, they implemented an assessment process based solely on subjective input from the BCT commanders in what they nominated as Key Terrain Districts in each of the Regional Commands. As you know, subjective metrics, while important, dont trend or translate well in-between RIP/TOAs. To assess progress based entirely on subjective metrics is, nor will ever be, optimal. And while simple to use, not easily, in fact impossible, to defend in light of rigorous analysis.
I would offer, in reference to your comments about atmospherics, that they can be relevant, and when taken in context with other supporting information, do not represent simply a "hodgepodge" of data. Surveys done professionally, asking the right questions, and interpreted with mathematical exactitude by qualified individuals can give commanders a "feel" for what the population is thinking, and while a certain amount of sufficing and error is unquestionably present, I believe that this is an important aspect of assessing progress. Ideally, one would look at subjective input from commanders on the ground, atmospherics, and objective metrics, to include your economic / investment indicators that you mention in your article, (though we at the RC level never had access to any of the data that you present) creating a virtual "three legged stool" of how we measure progress.
Mr. Kilcullen has offered some well thought out metrics of progress, specifically in an insurgency, and the one regarding "IED tips" is one of the best. And, yes, it is unclear whether this metric is being collected, though it certainly should be. The biggest problem lies at the tactical level and the entry fields in the CIDNE database, providing no efficient method of conducting a search for the data, assuming its even there, without sifting through thousands of entries by "hand".
Look forward to reading another article. Take care.
MAJ Tom Deveans
Student, ILE
11 JUL 11
Tom,
I agree with you and the author and have been attempting to influence the situation for some time to no avail. A couple of related issues for your info as you head home.
One, ISAF attempted to conduct an effects based assessment back in 2005/06 in concert with its Campaign Plan which included monthly operational mission data as well as questionnaires from 14 PRTs. Unfortunately, this early attempt was too heavily peppered with MOPs and not MOEs. The lessons learned were not acted upon appropriately and the results were worse.
For example, the Afghan PRT Executive Steering Committee created a working group of UN/IO/NGO/ISAF and CFC-A participants to try to identify indicators of success (such as the economic measures discussed in this article) but this failed because the various communities failed to reach consensus. The concept was to collect indicators and share the data from which each group could assess and analyse according to their own requirements. This effort ended in the 'too hard' box.
Secondly, the number of monthly measures ballooned (I term an effects explosion) from 100 to over 600 measures, which meant everything and nothing was being collected or measured. I coined my own EBO term, Effects Blurred Operations, which foreshadowed the GEN Mattis memo.
Despite this, Canada attempted to create an effects assessment mechanism known as the Effects Dashboard which strengthened these processes and attempted to place emphasis on effects measurement at the operational to strategic level in support of quarterly mission assessments. Additionally, we have had quarterly polls (conducted by Afghans) within Kanadahar province for the last 5 years collecting data on Afghan perceptions across the mission space and in all districts (across demographics). Interestingly, even with consistent data describing Kandahari perceptions (be it security, ANSF, education, etc), I would have J5 Effects Officers reject the polling data, unless, it agreed with their previously held views.
So, collecting and assessing the effects data is insufficient unless we have a culture and operationally sensitve/responsive system of planning and conducting operations and lessons learned. Then perhaps we can contend with the application of COIN/FID/SFA operations successfully.
Great article; also one that hints not-too-subtly at a basic lack of understanding the situation. After a decade one might think that something more than local insight at the company/ODA level would materialize.
In Afgh. as in Iraq we have set the implicit standard for success as pacification. Iraq was beginning to seem like a victory (until last month's 15 KIAs) because of a reduction of violence, not because we 1. eliminated the enemy or 2. because the GoI was all of a sudden an effective entity. While I'm on board with Ethan Kapstein's view on metrics, I would add a simple note on context. Economic metrics and "property rights/law-n-order/stability" frameworks fall under the topic of governance. Until we start seeing our occupations as first and foremost exercises in governance, especially in those crucial first couple of years we will not be the "premier benevolent occupiers" we aspire to be. Ironically, this was not lost on MacArthur and his cadre in their administration of occupied Japan and Korea. USAMGIK (United States Army Military Government in Korea) is a premier example of how-to-pacify.
In our 60-yr quest to repeat our post-WWII successes as occupiers we are wholly ignorant of how we did it. Not only that, we seem to feel that we are above the burdens of governance and rush like teenagers in love to turn over administration to local champions. The economic metrics that are championed in this article are just as useless in the long run unless we understand that to impact them we must govern, as we have in the past.
Concur with Bob T in a limited sense: ISAF (I Suck At Fighting, or I Saw Afghans Fighting) does indeed lack focus on effects. As I sit in my camp chair in Bagram waiting to fly home after nine months on a PRT, I can say, at least in Khost, effects are not being measured. Then yesterday I met a fellow Civil Affairs brother who brought up this same issue with me and replied: we paid Afghans to conduct surveys. Yes, we bent the rules (FAR/CERP) and yes, we got 900 responses. 900 responses in one province!!!!!! We did this in Fallujah and Ramadi for 2nd MEF; I know, I helped set it up. Why can't we do it here?
Metrics: we need them. And they need to indicate effects, not performance. GEN Mattis did DoD a great disservice with his kill-EBO memo, and set the stage for the spending spree we are currently on.
tom
This is behavioral economics and measuring response to positive incentives as indicators of military success. I agree that this is a valid approach. However, data is not easily obtainable in 3rd world, semi-permissive environs - especially where hawalas are present and unregulated. In Southern Afghanistan, the bank of poppy underwrites much more 'investment' (good as well as bad) than any regulated financial institution.
I completely agree that security is FUNDAMENTAL. The author uses 'perceived security' but lets not fool ourselves into thinking that we can 'IO' people into believing they are secure when they are not.
Highly enlightening article! I've always thought that if any country is to move forward they need a semblance of a middle class- until we lift trade barriers, Afghanistan will struggle to acquire one.
The body count fallacy is also interesting- in the end I believe the U.S estimation of NVA/VC dead was actually LOWER than the final tally revealed by Hanoi after the war.The fact is, by 1973 the NVA were gravely worried about the prospect of launching another offensive with American air power still in theatre. But there worries were unfounded: the regular RVNAF recruit felt he didn't want to die for his rich, corrupt masters: there was no middle class to aspire to, and therefore, for many ARVN, nothing to fight for. As the Vietnamese said at the time, "the roof leaks from the top down."
Let's hope it doesn't go that way in Afghanistan...