'Repertoire'', limited sets of routines that are learned, shared, and acted out through relatively deliberate processes of choice, cosntitutes a defining property of social movements. Nowadays, however, the famous traid/continuum ''violence, disruption, convention'', though extremely useful in preliminarily delimniting the subject-matter, ends up concealing more than it reveals. The reason is that the three key categories idenitfied become operative at a prohibitively high level of abstraction. When the matter comes to pinpointing and analysing concrete cases, they are both semantically fluid and denotationally opaque, containing far too many borderline cases and/or unpalatable combinations (what Sartori aptly termed ''strange bedfellows''). Violence is an extreme case in point, a veritable ''black box''. Prolifically discussed as it may be, we are still uncertain about its definiting properties, hence its concrete empirical contours. As a result, ''terrorism'' is typically conflated with ''contnetious disruption'', whilst both lack empirical specificity vis-à-is convention. The task of this paper is primarily conceptual: to specify defining properties of the main repertoire categories and tentatively suggest sub-categories. It goes without saying that the exercise involves the discussion of specific examples, both historical and contemporary.
