Interesting discussion here. A couple of points.

It's important to remember (since the media and our chattering class have forgotten) that after the Taliban defeat in 2002 the priority in Afghanistan was DDR. While I would agree that more effort should have been made at preventing a Taliban resurgence, the problem of large, heavily armed militias was a serious one that had to be addressed. After all, it was warlordism that provided the space and popular support for the original rise of the Taliban in the 1990's.

Bob's world said:

The bigger issue is not one of city vs country; but rather one of a Afghan governance overly shaped and formed by Americans vice an Afghan governance truly shaped and designed by Afghans for Afghans.

In a land of militias, and weak central governance, we demanded they stand down militias and build a strong central governance and army to enforce its will. One has to ask, do the people of Afghanistan see this as their government representing them, or do they see it as one that is a puppet of the U.S? (Note, not asking how do we see it, nor what it is, but how does their own populace perceive it).
This is a very important point. I've been meaning to write something on governance for a while now - maybe I'll find the time. Anyway, I would argue that there are many in Afghanistan who do not want central "governance" no matter it's character. "Governance" is all the rage these days and is the policy that naturally flows from the view that a stable, functioning and legitimate government is a universally desirable end goal. And it's easy to argue that point of view because certainly from our perspective, dealing with legitimate governments is preferable to alternatives.

So what happens with societies who have no interest in governance - societies that have long existed under different notions of stability and legitimacy? Are we going to show them the way to the benefits of governance and globalization (whether they like it or not)? What happens when they refuse?

If and when they refuse (as is likely and already happening) then we will be set on a course of conflict with these societies as we try to implement our idealistic notions of how societies should be ordered. We are already seeing a lot of that in Afghanistan and its neighbors. We are trying to "sell" governance and key players are not buying. These players see benefits flowing from a central government as, at best, a prelude to the disruption of their societies and local power structures and the imposition of unwanted central authority and dependence on outsiders.

The point in all this is to suggest that policies designed to bring governance in any time-frame shorter than generational are likely to be counterproductive in these areas and will bring more conflict, not less, as the "ungoverned" resist our efforts.

Sometimes more isn't better. I would therefore argue that in many parts of Afghanistan we need to quit pursuing our paternalistic notions of governance and work with the local power structures to ensure their independence from all outsiders - Kabul and Taliban alike. The irony here is that in order to save a central government in Kabul, we may have to undermine its authority in some parts of Afghanistan - at least until it grows in capability and legitimacy enough to overcome the very real and long-standing hostility to central authority among many populations in Afghanistan. In trying to control the entire country, the coalition and Afghan government are overreaching. Better, IMO, to work on governance and legitimacy in those areas where that goal is realistically achievable over the short-to-medium term. In those areas where it is not, try to strike deals that guarantee local independence in exchange denying sanctuary and support to the Taliban and their allies.

Anyway, that's my theory.