The Weekly Standard blog reports that the Pentagon will elevate post-combat stability operations in importance to equal the status of the major combat operations phase.
The question remains of how many troops are enough to win such post-war insurgencies. The blog post recommends a RAND study as the starting point for this discussion.
I think the RAND study is terribly flawed in its major assumption that only US combat troops count when you do the math. When I do the math, I think we have enough troops to win in Iraq.
Another reason the RAND study is rather irrelevant for the real world is that for good or bad, we are unlikely ever to field an army capable of occupying and pacifying a hostile population with any type of resources to back them. Had we tried to conquer Iraq with all 25 million Iraqis hostile to us, with money and arms everywhere, we would have needed 500,000 troops of our own to fight them. And we would have suffered far more casualties than we have now.
The fact that we liberated 80% of the Iraqis from the other 20% that is still opposed to the new government doesn't seem to affect war opponents. Nor does the fact that our casualties are pretty low historically keep war opponents from seizing on every tragic death as a reason to quit. Can anyone honestly say our public would accept three times the casualties we've endured in Iraq to suppress an entire hostile population?
We simply must be in the role of a liberating force that can ally with the local people to help them fight any regime remnants. The math is pretty pointless if we forget this.