This dissertation presents a new research methodology called accidental narration. Accidental narration occurs whenever references or comments embedded within the artefacts of a dominant culture serve to unintentionally expose, animate and/or narrate the physical presence, social interactions, and/or social successes of a subordinate or minority person or group. For the purpose of this study, accidental narration is achieved when the archived documents of elite white males who ran the British slave trade reveal incidences of Afro-Caribbean slave women and poor Irish women either speaking out, acting out-or in a few cases, displaying in their lifetimes-a measure of social success that mimics the wealth and lifestyles of members of the plantocracy. In pre-emancipation Barbados in particular, both groups of women existed at the very bottom of slave society and they had no access to education and publishing. Thus, very few of their personalized narratives exist today. This dissertation employs accidental narration to challenge the practice of approaching women's narratives from the consciousnesses of slavery's elites, and suggests instead that researchers target the elites' unconscious recordings to unearth plausible, non-patriarchal female 'voices'.

This thesis deals with slaves. More precisely, it deals with the slave-run agricultural estates in Italy during the period of Roman imperial expansion. The main point to be addressed is that of the relationship between its two main genders: adult male and female agricultural slaves. Whilst scholarship has maintained for almost a century now that male slaves played a significantly more important role in this period than female slaves, this thesis will argue that their economic and social contributions were at least equal. It will further to this argue that the traditional view is largely based on a highly biased and discriminating attitude towards the role of women in the world of work, and on a more general disregard and subsequent unbalanced valuation of women's contributions. The prevalence of male slaves on Roman agricultural estates is traditionally attributed to their availability for purchase as a result of Rome's intensive warfare. Furthermore, the various labour tasks usually associated with agricultural slaves are typically regarded as male labour domains, especially work in the fields, be it for grain, wine or olive production. To start with, this thesis will question the narrow range of productive activities that were carried out at these estates. By suggesting through examination of the evidence in a non-traditional way the regular occurrence of productive activities that are typically regarded as female labour domains, especially wool and textile production, the door is opened for a fresh look at the evidence for female labour on agricultural estates, ranging from epigraphic material for the management staff, to passages in the literary sources, and finally the application of demographic and economic models that support the propositions derived from the study of the ancient evidence. Although this thesis title may suggest a descriptive focus on the female slave, it is in fact merely one of analysis: this thesis does not strive to explain the various tasks carried out by female slaves, nor does it aim at the compilation of whatever evidence there may be for female agricultural slave labour. Rather, it aims at questioning a preconceived model of a male-female-relationship that, in current imagination, has huge repercussions on other significant aspects of Roman history. By creating a picture that encompasses slave family life (based on female reproductivity) and high female productivity, traditional views of chattel slavery, based on social deracination and total loss of any liberties, are questioned together with views of economic activity that leaves the Italian (servile) countryside virtually free of a female element.