After quoting a reader's comment last night, I signed off by saying, "Another way to put this 'nationalism without a state' is of course 'tribalism,' a great blight domestically and internationally."

The reader says that he meant to convey the opposite. An emphasis on tribalism, he says,

is precisely not what I was trying to say. I understand the notion of comparing tribal affinities and relationships for stateless nationalism, but the central point is that a relatively coherent and enduring "national" identity exists independently of any combination among tribal affiliations. The basic way that I see this is by a simple question: do Afghans with connections with well-defined tribal communities, whether large (i.e. Pashtun, Uzbek, etc.) or small (i.e., locally specific with strong familial ties), still see themselves as Afghan?

My point, if I may, is to argue that the fundamental misjudgment with counterinsurgency is the idea that counterinsurgents can and must embed in insurgency-prone target societies from the bottom up with the expectation the individual "sites" (as in villages, towns, clans, etc.) can be coopted individually and that the aggregate will create the premise for collective action... but that this is no substitution whatsoever from a concerted study and understanding of the context in which these events will play out.

Another reader writes to say:

I am not sure it is helpful to look at tribalism in this way in the Afghan case. Afghanistan became a problem for us after the Taliban managed to subdue the entirety of the country and begin building a centralized state. I tend to think that one of the big challenges for the U.S. effort in Afghanistan is coming to the realization that polyarchy (i.e. tribalism) is a fairly stable end state for the region. Indeed that efforts at state building may be counterproductive. It's likely better to leave the Taliban to compete in a decentralized, federal (i.e. tribal) Afghanistan than leave them the opportunity to take over a well-defined and powerful state apparatus.

Noted in both cases. Again, as from the start, I am explicitly not an authority on Afghan circumstances. I do, though, stick with my view that "tribal" style thinking, which I'll define below, is a surprisingly big drag on human betterment, but perhaps I should not have complicated things with that unexplained theme.
Tribal thinking: Viewing all issues a priori by who supports them -- our friends? or our enemies? -- rather than on their merits. Simple example in U.S. politics: your party endorses a policy, like an anti-deficit commission. Then the other party decides to support it -- so the original supporters switch to opposing it, simply because it's now the other side's issue. Corollary outlook: whatever hurts my enemy is good for me, even if it hurts me too. Obviously this is a kind of thinking that applies much more broadly than to "tribes" as usually discussed -- Ashanti, Uzbek, or whatever. I'll leave it at that for now.

Update. another reader adds this suggestion:

Seeing the exchange regarding Nationalism and Tribalism prompts me to recommend "Afghanistan, A Cultural and Political History", by Thomas Barfield. The recently published book provides a very good picture of the area and put to rest some misconceptions I had about the history and people of that area. I use the word "area" because the country is more a construct of colonial border drawing than the evolution of a discrete nation.