Like most everyone in the blogosphere (but see Barnett Rubin for a counterexample) my knowledge of and understanding of Pakistani politics is rather limited. What's more, it's clear that people who are knowledgeable about Pakistan disagree on the key points. One school of thought holds that Musharraf is a key bulwark against Islamist extremists whose power we need to help shore up. Another school holds that Musharrafism is actually rotting away the social foundations of moderate Pakistan. As Joshua Hammer put it:
And while the military aims to do the opposite, it is slowly destabilizing Pakistan. Eight years of usurpation of power by Musharraf have weakened secular parties, corrupted the judiciary, and implanted army men in every facet of civilian life. Pakistan’s population is now doubling every 38 years, creating severe social pressures. If the political process remains stunted, the Islamists may continue to gather strength until the country reaches a tipping point. “We are not going to collapse if Musharraf goes tomorrow; Pakistan will go on, insha’allah,” I was told by Mohammed Enver Baig, a senator with the Pakistan People’s Party. “But the 2007 elections could be a turning point for all of us. If the elections are not fair, don’t be surprised if next time—after five years—you come and see me, I might have a long beard myself.”
My best guess would be that this latter line of assessment is closer to the truth. The real policy problem, however, is simply that it's hard to know. Very few Americans have the sort of language skills and life experience that puts them in a good position to really understand Pakistan, and of course the Pakistani most capable of influencing the American elite's understanding of the situation are interested parties — Musharraf himself, other security officials, Benazir Bhutto and her circle, etc. — whose ideas and information are of questionable probative value.
The deeper problem, I think, is not so much that our understanding of Pakistan isn't as good as it might be, but that the country has put itself in a position where there's widespread consensus that we need to be trying to micromanage political outcomes in Pakistan. Within that context, there's a lot of disagreement, but the general trend is still to try to analyze the situation and then frame a policy as the thing to do if you want to bet that the analysis at hand is the correct one. It seems to me, however, that understanding and micromanaging Pakistani politics isn't something the United States is likely to be good at. The knowledge gap is sufficiently severe that the more we wade into trying to manipulate events there, the more likely it becomes that in fact we are the ones being manipulated and I don't think there's any way around that.
Rather than strategies for micromanaging Pakistani politics more successfully, what we need are strategies that don't require successful micromanagement and that try to avoid betting too heavily on any particular group or individual or interpretation of internal Pakistani events. Whatever the shake-out of the current crisis, we need to be prepared to deal with the resulting power structure on issues that are important to the United States, but I don't think it'll serve our interests to be too closely identified with whoever that is or to have us trying to pick the "best" candidate or course of events. Sheryle Gay Stolberg and Helen Cooper write in this morning's New York Times that "For more than five months the United States has been trying to orchestrate a political transition in Pakistan that would manage to somehow keep Gen. Pervez Musharraf in power without making a mockery of President Bush’s promotion of democracy in the Muslim world." This time, though, one can hardly chalk up their failure to the usual Bushian bumbling. That, simply put, would have been a really hard thing to do. And, indeed, "on Saturday, those carefully laid plans fell apart spectacularly." But it probably would have been impossible to keep them together.
The problem's not in the failure, but in the setting of the goal. Or, rather, in the setting up of a grand strategy that gets us stuck in that vise. We need approaches that don't depend on our ability to successfully pick and choose who comes to power where since our efforts to do this seem to have a noticeable pattern of making things worse.