Like Jason, I’m impressed by the Pillar-Nagl debate in the National Interest. Like Jason, I think both gentlemen make excellent points; I find myself agreeing with both to some extent. But, where Jason engages with Pillar in his questioning of the purpose of US involvement in Afghanistan, I diverge. Instead, I, like Nagl, recognize that not all conflict zones are created equally.
It seems that Pillar implicitly argues that the United States is only able to engage in one conflict zone at a time, that if it is involved in Afghanistan then it cannot deal with security threats in Yemen. Or, conversely, that the Underwear Bomber having operated from Yemen indicates that Afghanistan is no longer relevant to US national security concerns.
While US resources are finite—and its freedom of action has been severely curtailed since the 2003 invasion of Iraq—it is objectively not the case that United States is so tied up in Afghanistan that it cannot address other, smaller security threats. Various responses, including both military special forces and civilian covert attacks, have been used over the last year to address al-Qaeda branded security threats in Pakistan, Yemen, and Somalia.
It is also not the case that the Underwear Bomber’s non-Afghan point of origin is indicative of Afghanistan no longer being a threat to US national security. Abandoning Afghanistan presents at least two distinct threats. First, eight years of war in Afghanistan should have taught us above all that conflict separation based on such thing so arbitrary as the Durand line is nonsensical. The notion that abandoning Afghanistan will not have regional consequences is insane. There is no reason that, should Afghanistan collapse, the Pakistani Taliban will not begin to use Afghanistan as a safe haven as the Afghan Taliban have used Pakistan since 2002. It should be remembered that the Pakistani Taliban have a real beef with the state of Pakistan—they, unlike the Afghan Taliban, would actually like to see Pakistan collapse. Of course, the collapse of Pakistan is a nightmare scenario.
Second, while the hijackers that actually carried out the attacks were not Afghans, the freedom of action al-Qaeda enjoyed there certainly facilitated those and other attacks. Though al-Qaeda is a diffuse operation (diffuser now than it was in 2001), having a base of operations is useful to the organization. Were that not the case, bin Laden would not have relocated from Sudan to Afghanistan. Were that not the case, then it would be mere coincidence that so many of the terrorists that launched or attempted to launch attacks in Europe since 9/11 spent time in either Afghanistan or Pakistan.
Though the Underwear Bomber’s plot was hatched in Yemen, it simply does not follow that Yemen is the sole threat to the United States currently. Yemen is merely the most recent example of a phenomenon that we might take to be a natural law at this point: weak or failed states afford non-state actors space in which to incubate, train, and launch attacks. The degree of intervention required to ensure (so far as that is possible) US national security is a fact-specific question. In Afghanistan, there is not yet a state to which the territory can be left, the US ought remain engaged until such a state exists—hopefully the Marja operation and the rounding up of Afghan Taliban leadership is a step in the right direct (there are many people that think it is not, however). Yemen, however, presents different facts and different requirements. There, the state, though weak, exists. Working within the state’s apparatus to disrupt al-Qaeda in the Arabian Peninsula is the best course of action. Additionally, as Nagl notes, the ease of access US military assets have to Yemen via the Arabian Sea also affords the US an ability to strike from a stand-off position which it does not enjoy in Afghanistan.
So, to answer Jason's question, we are trying to save ourselves. Long-term policy making is hard. It is harder still to execute. But it is absolutely necessary.