The U.S. lacks coherent and measurable metrics for success in the war there, making it difficult to know if our strategy is even taking us in the right direction
Residents watch at a U.S. soldier on patrol in Kunar / Reuters
Yesterday I launched my latest report, Measuring Success: Are We Winning? 10 Years in Afghanistan. The question of measurement is vital: it requires understanding the war itself, then our strategy for winning that war (which includes identifying an end state to work toward), and lastly how we look at the progress or not-progress we've made.
For an example of why this matters, I'd like to (reluctantly) draw attention to a brief blog post Marc Thiessen wrote for the American Enterprise Institute. He asks a simple question: Is Obama passing on opportunities to strike al-Qaeda in Yemen? Thiessen presents two sets of graphs, one showing the number of drone strikes that have taken place in Yemen and the other showing the number of drone strikes that have taken place in Pakistan.
The comparison he presents seems simple:
Bottom line: Either we do not have the intelligence to target AQAP [in Yemen] because we are no longer capturing the terrorists who can provide it, or we have the intelligence to target AQAP, but we are holding our fire. Either one is a damning indictment of the Obama administration.
That's a difficult statement to square with last week's death-by-drone of U.S. citizen-terrorist Anwar al-Awlaki (I discussed some of the implications of his death for the Washington Post, where Thiessen is a blogger). Moreover, it rests on a number of assumptions that don't fully add up:
- The U.S. could be using drones to target more terrorists in Yemen than they do, and the decision not to is deliberate (as he hints through a quote from Washington Post columnist David Ignatius);
- Yemen can and should host a program of equivalent size, scope, and fatality as that in Pakistan;
- Concerns about local attitudes and perceptions are immaterial to the goal of killing terrorists; and
- If only we captured (and, given Thiessen's past books, statements, columns, and advocacy) tortured more suspected terrorists, we'd have better information.
I do not say this lightly: we have a lot of information about Yemen. I can say this because I used to work on Yemen issues at the Defense Intelligence Agency. Our problem wasn't too little information, but too much: filtering and prioritizing the hundreds of pieces of data we'd encounter every single day is no small or simple task. So Thiessen's digs at how the intelligence collection, analysis, and targetting process work -- not just in Yemen but in general -- is coming from a position of ignorance.
Worse still, however, and relevant to my opening paragraph here, is the belief that a simple comparison of drone strikes in Yemen and Pakistan will tell you anything at all about either theater of operations. Yemen and Pakistan are very different places, with different threats to their stability, different populations, different histories, different cultures, and different terrorist groups preying off the different society's socio-economic inequalities. They are not comparable, at least in the basic 1:1 ratio Thiessen constructs.
So what does that mean for Afghanistan? The way in which we conceptualize, frame, and then judge the wars we conduct matters tremendously. In "Measuring Success," I tried to do that with Afghanistan. I went into it hoping to do a fairly anodyne report of "here's what this metric says, here's what this metric says, and here's a weighted answer of how we're losing."
What I found instead was that we just don't know what it is that we're accomplishing there. Simple numbers, like whether ISAF is right that violence is down or the UN is right that violence is up, don't actually capture the reality of the war.
War is complex, and it cannot be understood through simple bar charts and single metrics.
By looking at the war as a system, I realized something else: President Obama's strategy carries some real problems in terms of measuring its success. It has three basic components:
- Degrade al-Qaeda so they can never launch at attack on the U.S. from Afghanistan;
- Reverse the Taliban's momentum so they cannot overthrow the government; and
- Build the Afghan government and Afghan security forces' capacity so they can run things on their own.
Sounds good, right? But the first two components are defined by absence, which means they have already been achieved but they also can never be achieved (since a single recurrence would demonstrate mission failure). The third component is so vague and undefinable we have no way of ever knowing if we've ever reached it, except maybe in hindsight someday long after we've already reached it, or failed to.
Knowing how incoherent the metrics for the war's success are, we probably shouldn't be surprised at the other findings I discovered: the Taliban is fighting a fundamentally political war, while ISAF, despite much talk to the contrary, is still fighting a conventional military war. Taken together, these two wars should mean we should account for both military and political effects in the war.
But we don't. In fact, we measure most of the relevant metrics that would tell us if we're winning in Afghanistan either incompletely or not at all. So when we ask if we're winning, we haven't the foggiest idea how to answer or to find out.
If we are ever to make smart choices in foreign policy, we have to start
thinking in terms of strategy, not just of methods but of end states,
outputs, and measurements. That just isn't happening right now, whether
it's drone strikes in Yemen or the rapidly drawing down war in