First, Leites and Wolf is a classic. The old Rand Corp set the bar high for think-tanks, and they provide great work on small wars. When I have a chance, I'll see if I can find their piece on the Malaya Emergency for posting. In terms of insurgency as a system, one failed assumption Laites and Wolfe made was assuming that the system is closed. In practice, as with many human endeavors, the system is open, and this distinction allows intevention points when accurately defined. An insurgency holds an information advantage against the host nation (they can see us, we can't see them), but they need people, guns, money, and silence from the populace in order to thrive. With that said, on to Surferbeetle's points...

Quote Originally Posted by Surferbeetle View Post
In Surferbeetle's ideal world a single school is part of a larger educational system or chain in which long-term concerns about an adequately educated populace, academic standards, teaching standards, dependable funding, building codes, safety codes, etc. have a place. Schools are a resource intensive activity which require timelines greater than 12 months.
Intuitively, I agree with you if I understand you correctly (shore up the rear/protected areas); however, this action neglects the safe-havens, denied areas, and hinterlands. Given enough time unhindered, the insurgency can continue to grow in their protected areas to the point where they can mass effectively against the host nation (Pakistan last summer was getting close). So, what do we do with the denied areas while Surferbeetle is revamping the protected areas? Five current options:

1. Big Army/Big Hammer Approach. Clear, Clear, Clear
2. Jim Gant "One Tribe" Approach. Bottom-up FID/micro-COIN
3. Greg Mortenson Approach. Build schools in the hinterlands. Over time, education trumps grievances and the insurgent's message fails.
4. Man-hunting. UAVs, airstrikes, hyper-infantry raids
5. Do nothing.

The first three options have high potential pay-offs, but come with high risks/costs. The fourth option may disrupt over-time, but the results are hard to measure. The fifth option is very dangerous IMO.

Mike