Charles Tilly (2003) Politics of Collective Violence, and people who should have read him.
Charles Tilly was a prolific writer in history, sociology, and political science, and I find myself returning repeatedly to his work on violence and the state. His 2003 book, The Politics of Collective Violence lays out the most useful analytical scheme for thinking about the different kinds of organized violence that afflict society, and his 1985 essay, "War Making and State Making as Organized Crime" should be taught along with it in staff colleges, in the same way that doctors and lawyers are introduced to the debacles and failings of their professions. We often have exaggerated ideas about our social utility, but if we understand the dynamics of violence and self-interest more clearly, professions can do something to improve this over time.
What has been bringing me back to Tilly has been the echoes of his work in new thinking, including a recent collection by Barkony and de Guevara (2012) on the Microsociology of Violence, in which almost every chapter cites Tilly (2003). More often, however, I find myself reading new or recent work whose authors should have read Tilly, perhaps did read him, but inexplicably don't cite his insights, even when they seem to be close to similar ideas.
Kilcullen's (2013) Out of the Mountains: The Coming Age of the Urban Guerrilla is an excellent piece of work, particularly chapter three on competitive control, but Tilly isn't mentioned. I think this is a good case of those farther from the front understanding more than those on the front lines, although the reverse is perhaps more often true.
Berti's (2013) Armed Political Organizations: From Conflict to Integration is another superb set of case studies - Hamas, Hezbollah, and IRA - in the context of a very useful organizational framework. But I think it could be improved by reference both to the state-making functions described by Tilly (1985) and the forms and functions of collective violence articulated in Tilly (2003).
I think the difference between the authors in Barony and de Guevara (2012) on the one hand and the authors Kilcullen and Berti on the other is that the latter are more practically oriented; they are more concerned with pragmatic action than scholarly theory, and have read less theory. But for me, Tilly's work and its implications reinforce an old soldier-scholar's observation: the most practical thing in the world is a good theory, because it gives you a useful way of seeing the world. (Thank you, Ken Eyre).