There's a striking paragraph near the top of Fred Kaplan's latest column that I've seen quoted on a bunch of progressive blogs, but his more important point comes deeper into the piece explaining the problem with the idea that success in working with locals in Anbar Province against al-Qaeda is a promising stepping-stone to nationwide stability:
But in these alliances, we're dealing with tribesmen who are cooperating with us for a common goal. It is not at all clear on what basis these various local Sunni factions can be stitched together into some seamless security quilt—or why, because they've agreed to help us kill jihadists, they might suddenly agree to stop killing Shiites, compromise their larger ambitions, redirect their passions into peaceful politics, and settle into a minority party's status within a unified government.
Alliances of convenience rarely outlive their immediate aims. Josef Stalin formed an alliance with the United States and Britain for the purpose of defeating Nazi Germany. But once the war was over, he had no interest in integrating the Soviet Union into the Western economic system.
Kaplan notes that this idea appears to have come to the administration via Steven Biddle, a very sharp analyst, who thinks his own plan has "maybe one in 10" chance of generating "something like stability and security in Iraq." You'd have to be out of your mind, really, to adopt a military strategy whose author thinks the odds of failure are overwhelming unless the alternative was something like national extinction. In many ways, I feel like the requirement of "serious prospects of success" is the most unfortunately overlooked aspect of just war doctrine. Unfortunately for us, the country has George W. Bush on hand so a strategy that you'd have to be out of your mind to adopt is precisely what we're going to get