Excavations in the Roman Southwest, in what is now Portugal, Spain, and southern France, have brought to light a diverse corpus of sculpture that was displayed in late antique villas, ca. 250-450 CE: monumental reliefs, mythological statues and statuettes, and portraits, of both private individuals and imperial personages. Although the production of some statues is roughly contemporaneous with the late antique villa in which they stood, many pieces were antiques, carved centuries earlier. But the diversity within and across these statuary assemblages, and the concentration of such finds in villas of Hispania and Gaul, have not been given a full treatment in modern scholarship. Rather, sculpture is interpreted generically as evidence for the elite status of a villa owner, leading to problematic assumptions that all statuary collections, and late-Roman villas more broadly, evince an indistinguishable class of “elite” Romans. This dissertation invites readers to reconsider the narrow but well-established definitions of the late antique elite, through analysis of the many differences that are readily apparent among villas and their material assemblages. I argue that material culture can and should be interpreted as evidence of the regionalism which played an important role in late antique society, and thus I treat sculpture as a window into the social practices and peer polity negotiations among villa domini.