Violence in Iraq and and army to focused on COIN?

I just finished reading Linda Robinson's Tell me how this Ends about Petraeus in Iraq. I also just read an interesting question and answer with COL Gentile. He makes several interesting arguments. The first is the the military is overly focused on COIN, the second is what Petraeus implemented wasn't actually that much different than what was happening before and thus other factors that led to the drop in violence. He also claims achieving victory in Afghanistan is impossible.

His first argument holds some water with me but the truth far more nuanced than that. The military probably is too focused on COIN training and most junior and mid level officers know it best. The important thing for the Army as a whole is not whether we are focused on fighting COIN or conventional, but whether we have the flexibility to learn and adapt quickly to whatever kind of war our civilian masters and the enemy throw at us. We usually start with an army ill suited for fighting the war we find ourselves in. Once we have reduced our presence in Afghanistan and Iraq, we should definitely build up our conventional capacity since we are unlikely to be fighting any large scale counterinsurgencies.

After reading Linda Robinson's book, I am certain his argument that Petraeus didn't change much is weak. Most of the military early on had the belief that not enough of the Sunni officer corp and Ba'ath party had been taken care of in the initial invasion. As such, no one addressed the fundamental political problems driving the insurgency. The most significant one was the Sunnis feeling like they had no place in the new Iraq.

What Petraeus did with Crocker was to craft a strategy that addressed the fundamental political issues. Petraeus focused on creating reconciliation from the bottom up while Crocker with Petraeus help promoted reconciliation from the top down. Reconciliation meant providing security for the population against sectarian attacks, getting reconcilables to stop fighting or switch sides, and then marginalizing the irreconcilables. This included both Sadr's men and the Sunni insurgency.

His third argument holds some water but at the same time, he seems to be thinking to narrowly. He says:

"Does anybody really think that Afghanistan, a ravaged, ethnically divided country of 25 million with 72 percent illiteracy and little history of centralized rule, can be turned into a real state any time soon, on a budget that US. taxpayers can support?"

The first criticism I have is that Afghanistan's ethnic divisions are not nearly as significant as those in Iraq. There isn't as much of I'm going to kill you because you are different. Second, he is setting the bar to high when he says "real state." Victory in counterinsurgency usually means reducing levels of violence to acceptable societal levels on terms favorable to security forces. In Afghanistan, it is clear we are not going to create a Jeffersonian democracy, Afghan standards are not very high. As one of my friends jokingly said "As long as they have fire they're well off." Thus, as long as we are able to create local security with a central government only strong enough to prevent ethnic or tribal conflicts, we can go home.