For more than a century historians and archaeologists have struggled to define Roman frontiers. Frontiers and borders have always been unwieldy notions as they belong to two different worlds and yet they are their own cultural universe. 1 Since separation is one of the most enduring themes in human history, one of the major challenges has been the negotiation of modern realities which constantly distort our understanding of what frontiers meant to the ancient Romans. Running the risk of falling into the trap of presentism, a number of scholars based their argumentation on implicit or explicit comparisons with modern frontiers. Nineteenthcentury efforts to redraw the map of Europe based on nationalistic views of political, cultural, and linguistic identity have prompted historians to rethink the function of Roman frontiers by projecting modern concepts onto ancient contexts. The ensuing debate has shaped two schools of thought. Scholars had to grapple with several frustrating questions: was the Roman frontier a linear barrier separating two worlds or was it an area of economic, cultural, and religious contact? What was the role of geography in sustaining imperial policy in frontier regions? Did frontier rivers unite or divide? Although the static frontier thesis has fallen out of favor with historians of the Roman Empire, the main goal of this chapter is to show that political conditions in Late Antiquity favored frontiers of exclusion and the Danube river will be used as a compelling example of a natural linear border which Roman emperors desperately tried to 1 The quest for a proper terminology has not yet yielded a universally accepted vocabulary. For the problem, see H. Elton, Frontiers of the Roman Empire (Bloomington, , 3-9. For an interdisciplinary approach, see L. Rodseth and B. J. Parker, "Introduction: theoretical considerations in the study of frontiers," in Untaming the Frontier in Anthropology, Archaeology and History, eds.