HOLBROOKE ON SUCCESS IN AFGHANISTAN: "WE'LL KNOW IT WHEN WE SEE IT."

Adam, myself and most of the folks interested in Afghan-Pakistan policy were at a briefing this morning with Special Envoy Richard Holbrooke and the bulk of his broad inter-agency team. Given the growing debate over what the U.S.' strategic interest in Afghanistan ought to be, this was an important chance to define how the administration will accomplish its strategic shift prior to next week's election. While the officials working on these policies are certainly very impressive, the overall impression from the event is that the government still doesn't have a handle on the end-state question.

Good news first: The various members of Holbrooke's team laid out what seemed like promising ideas, including eliminating poppy eradication efforts in favor of much broader support for Afghanistan's agricultural economy (8 in 10 Afghans are involved in agriculture), building a regional structure that creates incentives for a stable Afghanistan, and building government legitimacy and capacity after the upcoming election by directing more aid funding through the government and improving public communication infrastrcuture. As Adam tweets, however, the fact that each emphasized the need for their individual program to succeed if the whole project is to come off was disconcerting.

But on the pressing domestic question of how much longer we will be in Afghanistan and what kind of resources must be committed, there was not much news. Holbrooke said that the "military part of this struggle with American troops is not an open-ended event, but our civilian assistance is going to continue for a long time." The fact no one knows when our troops will leave Afghanistan suggests that the first part of this sentence is simply not true; until we know when U.S. armed forces are leaving, the military part of this struggle is most definitely open-ended. Holbrooke also said of success that "we'll know it when we see it," obliquely referencing the famously vague, quickly-dropped Supreme Court obscenity test. It was not comforting.

What was somewhat comforting was Holbrooke's clear understanding of the metrics question, how we measure success. I'd like to think it is a legacy of his time as a foreign service officer in Vietnam, when Robert MacNamara's infamously statistical approach to war failed in part because it did not understand the correct measures of success; the use of enemy body counts are perhaps the most prominent example. Today, Holbrooke emphasized the difference between inputs -- what his team is bringing to and doing in Afghanistan -- and outputs -- the actual results of those efforts. For example, the administration won't be focusing simply on how many Afghan troops are trained but also on how many missions they can handle on their own.

While the actual measurements of these outputs are unclear, an administration official tells me there are approxmiately 50 categories that will be used to understand the results of the new strategy. At least some of these measurements will be made public in a report to Congress that is due on September 24; data is already being collected for this report.

-- Tim Fernholz

[Updated: I originally had the title quote as, "You'll know it when you see it." Checking with other reporters after I published this post, it appears the correct Holbrooke statement was "We'll know it when we see it."]