This article seeks to review the common assumption of a clear, largely endogenous, continuous and singular medieval period in Muslim history. The assumption implicates a romantic vision of history as one of continuity, and the possibility of writing the history of large-scale masses as one of singularity, and that the historical masses can quite satisfactorily be identified by transhistorical nominatives such as Islam, Hellenism, or the West. Correlatively, the histories of such masses are encoded according to the literary topos of rise and decline, à la Herder, Hegel, Renan, Becker, Toynbee and others. The author proposes that the conceptualisation of historical objects as Muslim or otherwise, and the consequent periodisation is possible only when shorn of romantic historism, and if the titles under which historical material is organised were to be subverted to preserve a definite distance between self-representation and subsequent construal on the one hand, and the actual workings of History on the other. It thus suggests that 'Late Antiquity' be taken to encompass the conditions that gave rise to Islam, which wouldpresent the Arab empires as points of arrival and the crystallisation of cultural, economic, demographic, religious, statist and other universalist trends that constitute Late Antiquity.