Although commercial activity was one of the central features of Roman urban centres, the identification of commercial space in the archaeological record is not always straightforward. Identifications are routinely made through the application of Latin nomenclature to particular architectural typologies, almost inevitably leading to interpretations of space influenced by both textual and modern analogies, a practice which can be most clearly demonstrated by the so-calledtaberna. Using thetabernaas a case-study, this paper explores the issues of Latin nomenclature and textual analogy; architectural typology and modern analogy; and material evidence and functional space, demonstrating the difficulties of identifying specific functions for so-calledtabernaein the archaeological record. It argues that a much wider variety of spaces should be viewed as potentially commercial, since commerce — and especially retail — could take place almost anywhere, and suggests ways in which we could use the ancient evidence to look for commerce beyond thetaberna, demonstrating the commercial potential of streets, porticoes, arcades, open spaces, and even private houses, since the separation of commercial and domestic activities in the ancient world was minimal.