Characterizing the discourse functions of linguistic expressions is surely one of the most difficult tasks in linguistic analysis. The starting point for any study of discourse functions is the examination of naturally occurring data; the limiting factor is the lack of well-developed theoretical frameworks for understanding language use. Still, a good descriptive study has lasting value, and empirical claims invite further analysis. Overstreet's study of the “general extenders” or something, and everything, and other members of this class makes a solid contribution on both fronts.