... if you have just got some job in Afghanistan for which you want to quickly bring yourself up-to-date with the situation there.
If you want to have a better understanding of that situation than what the security briefs offer you by dryly looking at trends of security incidents in the vicinity of where you're setting up shop.
If you are ready to read in French.
Then this study is compulsory reading:
Sébastien Pennes: L'Insurrection talibane : guerre économique ou idéologique? Politique étrangère, 73 (2008):2, pp. 345-358.
Pennes had worked with some humanitarian organisations before joining UNAMA in 2005. He has an exceptionally good insight into what is going on in places like Badghis or Faryab, rarely reported on in the media. If you want to read more about the neglected north, this article is something you shouldn't miss. But it wouldn't do this study justice to say that that's why it is valuable. It is valuable for a thousand more reasons.
Pennes has an argument similar to Giustozzi's, made in the latter's book, Koran, Kalashnikov and Laptop, but it somehow gives me the impression that his insight justifies much more clearly the significance of the local-conflicts-explanation regarding how the insurgency could be reignited and made spread subsequently. Giustozzi, for his part, worked a lot from sources some of which he was at a difficulty to cite for reasons one can imagine, which made his conclusions seem methodologically weaker even while they always seemed quite plausible. Pennes, however, writes with accuracy and confidence about the areas he knows more of (and he doesn't write much of areas he doesn't know that much of). And the results are the same or very similar, so it's time to take everything very seriously indeed. (You know, us, bloggers, working through the open-source debris, we were making similar points, but we could have never hoped to be able to come to conclusions with nearly as much authority as those doing extensive research or experiencing everything on the ground.)
The point is that the insurgency is very creatively looking for footholds and safe passage routes all over Afghanistan's neglected rural areas, where resource-scarcity and subsequently rampant local conflicts and a predatory mode of economy provide opportunities to do that. Outlining this argument with much clarity, Pennes then proceeds to take a balanced view of the role of ideology. While he says that local stabilisation will require more (and more effective) aid, spent according to the real local priorities, he notes that in the end economic means may serve to create the way to securing the Afghan government's legitimacy in shaping the country's future. But meanwhile he is also very critical of the endeavour to create the kind of centralised government that has been allowed to create disruption on the peripheries without being able to offer much in return. More transparency and a more legitimate operation of central control would require, for example, that provincial shuras be given the chance to exercise effective oversight over provincial governors' spending decisions, as well as, possibly, at least minor funds that they could dispose of themselves.
Oh, I shouldn't forget - there is an Uruzgan-related hint in the article as well - my raw translation included here:
"Save for the great centres of the insurgency organising attached to the production of drugs (Helmand, north of Badghis), a radical madrasa (south of Ghazni, Kunar) or a logistical node of the guerrillas (Uruzgan), the majority of the areas said to be under the control of the Taliban are in fact but paths of safe passage for the more battle-hardened guerrillas." (pp. 348-349.)
Just in order not to misrepresent Pennes' arguments, he doesn't make much of the claim of parts of the country that are insurgency-free being under government control, either. If anything, that is the gist of his study (deliberately forgetting about his otherwise equally important points regarding the disorganised political party scene and everything else I haven't yet come to mention): the Afghan peripheries (think of the Afghan Gap if you feel an urge to put this in Thomas Barnett's way) are just left to fend for themselves in most cases.