ABSTRACTThis review essay on Paul A. Roth's The Philosophical Structure of Historical Explanation follows Roth in accepting the notion of a “narrative sentence” (from Danto's Analytical Philosophy of History) and takes up Roth's view that a narrative explanation elucidates that which it constructs. I also analyze the “relating of events” within a temporal frame, but I come to different conclusions than Roth does: I deny any necessary connection between a narrative explanation and a narrative sentence and object to Roth's insistence on “nonstandardization” as a necessary feature of a narrative explanation. I argue that a “passage of time” is characteristic of all our descriptions, regardless of whether they are scientific, historical, or anything else. Like Roth, I draw on the work of W. V. O. Quine to address the pragmatic question of how large a “unit of empirical significance” is to be, and I show that different practical situations may require different practical decisions about what sizes of unit are required, with whole narratives being units of significance in themselves. Problems of the epistemic structure of narratives remain, and I argue that relating a passage of time requires a theory that enables us to preclude any temporal gaps. A theory of conceptual linkage of any stripe seems incapable of doing that.