President-elect Barack Obama has repeatedly said that he will focus more effort on the deteriorating situation in Afghanistan. Current plans call for an additional 20,000-30,000 troops to be sent to Afghanistan next year, and Secretary of State Robert Gates wants to double the size of the Afghan army and intensify U.S. training efforts. Fred Kaplan at Slate reports that "we finally have a strategy," though he wisely cautions "that may not be enough."
Here's why I'm worried, too. Take a look at the recent RAND Corporation study of U.S.-led peacekeeping operations, which contains in-depth case studies of the occupations of Germany, Japan, Bosnia, Kosovo, Haiti, Somalia, and Afghanistan. The success stories (Germany, Japan, Bosnia, and Kosovo) all featured a far more intensive commitment of men and resources than anyone is talking about in Afghanistan. In Germany, we had 100 U.S. troops for every 1000 people in the local population; in Kosovo and Bosnia, we started off with roughly 20 NATO soldiers per 1,000 locals. Somalia and Japan began with about five soldiers per 1,000 locals and then drew down (Somalia was a failure; Japan an obvious success).
By contrast, the post-conflict occupation of Afghanistan began with 14,000 U.S. and allied troops for a country with a population of nearly 30 million. That works out to well under 1 peacekeeper for every 1,000 Afghans. There are now about 63,000 U.S. and NATO troops in Afghanistan and the planned reinforcements will bring that total up to somewhere around 90,000 next year. But for a country that now contains over 32 million people, that's still fewer than three soldiers per 1,000 population. To get to the force levels achieved in tiny Kosovo -- which is only 11,000 square kilometers as opposed to Afghanistan's total land area of roughly 650,000 sq. km. -- the United States and NATO would need to put over 600,000 troops in the field. And nobody is proposing to make that sort of commitment, even if we could. Afghanistan has also received much lower amounts of aid per capita than the successful postwar occupations.
Of course, rules of thumb like these should be used with caution, as the context of different occupations varies considerably. One could argue that the insurgency we are fighting is confined to certain areas of the country, so counting the whole population overstates the problem. Adding the Afghan Army into the mix can bring the force-to-population ratios up to a more encouraging level, particularly if the army proves to be loyal and if efforts to expand it succeed. One could also argue that smaller amounts of aid per capita can still have a disproportionately large impact, because Afghanis are already so poor and it doesn't take much outside help to make a big difference. Supporters might also argue that the tactical success of the "surge" in Iraq shows that additional troops deployed in the right way can have big positive effects.
On the other hand, the difficult nature of Afghan terrain and the lack of a well-developed transportation network means that higher ground force-to-population ratios are needed to achieve lasting stability. That problem is compounded by the porous border between Afghanistan and Pakistan, which gives Taliban and al Qaeda insurgents a safe haven from most of the NATO contingent. Moreover, the lack of local infrastructure makes it harder to deliver aid cheaply and thus reduces the amount of money that actually touches the lives of ordinary Afghanis. Finally, Afghan tribes and warlords have a long history of alignment and realignment, which means that the cohesion of the "national" Army and the loyalty of tribal forces may not be something on which we can rely. And as my new colleague Rory Stewart points out here, adding more troops in the past has "had a negative political impact on the conservative and nationalistic communities of the Pashtun south and allowed Taliban propaganda to portray us as a foreign military occupation."
Add all this together, and it's easy to see why the U.S. commander in Afghanistan, General Bantz Craddock, said last week that the United States and NATO will have to keep a large force in Afghanistan for "at least ten years," and maintain a presence there for "decades." But Craddock's implicit recommendation should not be accepted uncritically: saying the United States should stay there for "decades" begs the question of whether it is in our interest to commit lots of blood and treasure toward the creation of a unified Afghan state. And that's a question of national interest and overall grand strategy, not just a matter of military operations or counter-insurgency tactics.
Bottom line: there is a very real possibility that escalating our commitment in Afghanistan will not succeed, which means we will need a Plan B. I hope we won't, but I also hope somebody in the Obama administration starts working on one now, so that we don’t have to improvise one on the fly. Proposals welcome.
SHAH MARAI/AFP/Getty Images
Stephen M. Walt is the Robert and Renée Belfer professor of international relations at Harvard University.