INTRODUCTION

What can archaeologists contribute to the study of Greek slavery? First, we need to know just what it is we are studying. According to Peter Garnsey (1996: 1), ‘A slave was property. The slave-owner's rights over his slave-property were total, covering the person as well as the labour of the slave. The slave was kinless, stripped of his or her old social identity in the process of capture, sale and deracination, and denied the capacity to forge new bonds of kinship through marriage alliance. These are the three basic components of slavery.’ Some ancient historians dispute particular elements of this definition or merge chattel slavery into a broader category of unfree labour, but few ancient historians dissent strongly from Garnsey's phrasing. This immediately raises the issue that dominates this chapter: archaeologists face severe problems operationalising any plausible definition of ancient Greek slavery. Slavery was a legal category, driven by notions of property; and despite considerable ingenuity, archaeologists cannot dig up property rights. Moses Finley (1985: 25) emphasised this in his influential discussion of the complicated arrangements involving leases, labour and capital in Roman pottery kilns, insisting that ‘archaeological evidence or archaeological analysis by itself cannot possibly uncover the legal or economic structure revealed by the Oxyrhynchus papyri or the alternative structures in Arezzo, Puteoli, Lezoux or North Africa’.