Neither the traditional meaning of "insurgency" (which emphasizes the non-legal-belligerent status of the insurgents, and their threat to legal authority) nor the various US Army/DoD definitions imply that the armed force being used needs to take the form of guerilla warfare. The root of the word (from French and Latin) is simply to "rise up" against something.

While this may appear to be a case of how many definitional angels can dance on the head of a pin, I think it is important. Insurgent groups have a panoply of operational approaches to choose from: acts of terrorism, guerilla warfare, massive public protests, rapid urban insurrections (usually in conjunction with subversion or political neutralization of key military garrisons). Because of opportunities, resource constraints, and other factors they may emphasize some techniques at some times, and others at others. It is rather like deciding whether you'll seize an objective through heliborne insertion, a rapid armoured thrust, or a long battles of attrition.

Indeed, we see movements often shifting techniques. The best example of this were the debates within the Sandinistas between the GPP (Guerra Prolongada Popular), Proletarian and Tercerista factions, emphasizing rural guerilla warfare, urban insurrection, and a mix of approaches respectively. However, they can also be found with the Bolsheviks, the various Iranian revolutionary groups pitted against the Shah, the PLO shift from semi-regular armed struggle to intifada-like civil and violent protests. Moreover, virtually all urban guerilla (terrorist) movements see this as a first step towards later full-scale guerilla (and later semi-regular) warfare.

Usually widespread guerilla warfare characterizes the eventual overthrow of a regime. Not always, however, as the Russian Revolutions (both), the overthrow of Shah, British withdrawal from India (and several other decolonizations) and even the end of apartheid in South Africa (if one codes the transition as a de facto regime surrender, which it was) illustrate.

Given this, I think it is potentially problematic in many cases to carve off insurgency-as-solely-guerilla-warfare as distinct analytical category, just as it would be to carve off "wars in which people used helicopters"—especially when the folks on the other side don't think in those terms.

While a lot of this overlaps with what Wilf says, I do think there are important differences between guerilla wars fought by insurgents (who need to derive the majority of their resources from the local population, and in which the objective is regime change) and other irregular warfare, since the domestic political and socio-economic arenas are important in different kinds of ways.

To anticipate Wilf's probable objection: yes, I know it is all about degrading the enemy's military capabilities and breaking their will. On the other hand, croquet, hockey, rugby, and the annual Gloucestershire Cheese Wheel rolling competition are all about scoring more objectives than your opponents do—and I'm not sure how far that commonality gets us if we're trying to win one of them

CavGuy, I do understand what you're saying about having to make do with the datasets out there, and sympathize. I too am a little taken aback that Lyall and Wilson won't let others look at their dataset.