In this paper, I seek to delineate the build-up of the Italian slave population. My parametric model revolves around two variables: the probable number of slaves in Roman Italy, and the demographic structure of the servile population. I critique existing estimates of slave totals and propose a new ‘bottom-up’ approach; discuss the probable sex ratio, mortality regime and family structure of the Italian slaves; and advance a new estimate of the overall volume of slave transfers. I argue that the total number of slaves in Roman Italy did not exceed one-and-a-half million, and that this population had been created by the influx of between two and four million slaves during the last two centuries B.C.

The article investigates the social profile of Roman funerary epigraphy, focusing on Ostia and Pompeii, and reconsiders the predominant role of freedmen in this material. Comparing the epigraphic behaviour of decurions and freedmen, it concludes that the ‘epigraphic habit’ was not uniformly adopted throughout Roman society; different classes used inscriptions in different ways and for different purposes. The epitaphs do not therefore reflect the overall composition of the Roman population as much as the particular concerns and aspirations of individual social groups and categories within it.