This thesis examines the relationship between language and Roman identity, defined in the broadest sense as the political and cultural practices involved in being Roman. It focuses on evidence of multilingualism: Roman identity was defined through opposition and interaction, and it is at points of contact that these debates are cast into relief. It looks predominantly at evidence of Greek-Latin bilingualism, but also considers evidence of other languages to consider how their relationship to Roman identity differs. It combines historical and sociolinguistic approaches to multilingualism. Understanding bilingual language practices requires close sociolinguistic reading of evidence to understand how languages interrelate and analysis of the historical factors and contexts that determine language choices and their social, cultural and political implications. The thesis responds especially to the use of bilingualism as a model for Roman cultural relations, arguing that a closer engagement with sociolinguistic terminology and with linguistic evidence is necessary if we are to use language and bilingualism as a way into broader issues of politics and identity. Language is simultaneously a model for identity that works across ancient and modern thought and a central part of this identity. It frequently plays into other markers of Roman identity and a range of themes and concerns surrounding it including integration, migration and citizenship. The thesis examines three case studies in detail: the different layers of bilingualism in Suetonius’ biographies; Greek in the graffiti of Pompeii; epigraphic and literary evidence for different languages in the city of Rome. These case studies demonstrate the politics of language in different types of practice and at different levels of society: the thesis argues that the overlaps between them are greater than has sometimes been appreciated. The case studies also show that the boundaries of Roman identity did not develop in a progressive or linear fashion but were continually defined and redefined through ongoing processes of absorption and rejection.

Primary funding from a Peterhouse Research Studentship 2014–17. Additional funding from Peterhouse Gunn Studentship 2017–18 and a Classics Faculty Sandys Studentship 2016.