Though it doesn't claim as much, Dexter Filkins' article in Sunday's New York Times, headlined "Afghan Militias Battle Taliban with Aid of U.S.," may offer a clue to where President Barack Obama's strategic review of the war is going.
Or let's put it this way: If, in the coming days, Obama does decide to deepen America's involvement in Afghanistan, and if his strategy bears no resemblance to the approach Filkins describes, it is almost certain to fail.
Filkins, one of the most intrepid war correspondents, reports that special-operations forces have begun to help anti-Taliban militias in southern and eastern Afghanistan, where the insurgents are concentrated. These militias have risen up spontaneously in certain tribal groups, but U.S. commanders hope that they can use the example of these revolts "to spur the growth of similar armed groups across the Taliban heartland."
The interest, even excitement, in this development stems from two sources. First, it is reminiscent of the Anbar Awakening in 2006-07, when Sunni tribal leaders in western Iraq formed alliances with U.S. forces—whom the Sunnis had been shooting just months earlier—to beat back the bigger threat of al-Qaida.
Second, it has drawn high-level attention to a 45-page paper by Army Maj. Jim Gant, the former team leader of a special-ops detachment stationed in Konar province. The paper, called "One Tribe at a Time: A Strategy for Success in Afghanistan," recounts his experiences with organizing "tribal engagement teams" to help local fighters beat back the Taliban—and it spells out a plan to replicate these teams across the country.
One measure of the interest in the paper is that Maj. Gant, who was about to be redeployed to Iraq, has been sent back to Afghanistan instead to help set up more of these teams.
The premise of his paper is that Afghanistan "has never had a strong central government and never will." Rather, its society and power structure are, and always will be, built around tribes—and any U.S. or NATO effort to defeat the Taliban must be built around tribes, as well.
The United States' approach of the last seven years—focusing on Kabul and the buildup of Afghanistan's national army and police force—is wrongheaded and doomed. The tribal approach also has many risks. But the case for it, Gant argues, is this: "Nothing else will work."
There are signs that Obama has been mulling over something like Gant's strategy. At one of the seven meetings Obama has held with his national security advisers (the ninth, and perhaps final, session takes place tonight), he asked for a breakdown of which Afghan provinces could provide their own defense, which need our help, and to what degree. He also told ABC-TV's Jake Tapper, in an interview earlier this month, that he and his advisers were focusing on "not just a national government in Kabul but provincial government actors that have legitimacy in the right now."
A tribe-centered strategy may appeal to Obama in several ways. First, it keeps the Afghan people, not American occupiers, at the center of the operation. The U.S. soldiers live alongside the tribes, build trust, train them, supply them, gather intelligence from them, and fight with them. We are supporting players, not the lead.
Second, these teams of U.S. soldiers are small. As Gant puts it, the approach requires a lot of time—many months to gain a foothold, years to make the bonds stick—but not a lot of manpower.
If Obama is looking for a way to counter the Taliban and build Afghan security without sending all 40,000 troops that Gen. Stanley McChrystal has requested, this is one such way.
Third, the strategy makes military success less dependent on the political fortunes of Afghan President Hamid Karzai. Counterinsurgency campaigns work through local authorities; if the authorities are seen as corrupt, the campaign can't succeed. Karzai has promised reforms, which may boost his legitimacy among the Afghan people. But it he doesn't follow through, or if his efforts have scant effect, it won't matter so much with Gant's strategy, because the key authorities are the tribal leaders, not the central government in Kabul.
Gant has no illusions about the difficulty of working with tribes. He spells out the risks of getting enmeshed in internecine feuds. Several times during the 2001 invasion of Afghanistan, our guerrilla allies called in U.S. air and artillery strikes on what they said were "Taliban targets" but in fact turned out to be gatherings of rival tribes.
An explicit and essential part of Gant's strategy is to draw the individual tribal teams into a network of tribes—first across the province, then the region, then the nation—tied in to the Kabul government through a web of mutual defenses and the supply of basic services. He's less clear on the mechanics of how this "bottom-up" approach to national unity takes hold, but he recognizes that without it the Taliban can gain advantage by playing the tribes off against one another.
Nor does he contend that the Taliban can be countered by a tribal strategy alone. The officers who have been circulating Gant's paper, and discussing it in closed-door meetings, don't think it can be anyway.
Two weeks ago, asked about the continuing internal discussions on the subject, Secretary of Defense Robert Gates told reporters that President Obama was asking how to "combine some of the best features of several of the options" that his advisers had put on the table.
Obama is likely to announce his decision—on a strategy and on how many, if any, more troops it will require—soon after Thanksgiving. A key question to ask, in examining this mix, is how prominently it features the tribes