|Scholars have increasingly turned their attention toward the ideologies and practices of slavery in the pre-modern Islamic world, but there is much left to investigate in this relatively new sub-field. While other participants in this roundtable will consider crucially important documentary and papyrological records of slavery, I consider ways to work with previously published literary sources from early Islamic history. Particularly, I consider the advantages and pitfalls of working with "big data," that is, compiling massive databases that may be analyzed computationally to reveal patterns invisible within any individual text. I will suggest the many collaborative opportunities that exist in creating such a massive database, using data gleaned from searchable text collections such as al-Maktaba al-Shamela and al-Jami' al-Kabir. This data could then be analyzed to reveal trends in the practice of slavery across time and place—for instance, what vocabulary words are used to describe slaves, who owns slaves, what is the percentage of male vs. female slaves, what jobs or tasks do various slaves perform, and how often are slaves manumitted? In addition to giving us a more complete picture of the practice of slavery across time and place, such data analysis would also open up opportunities for comparative analysis with scholars in other fields (medieval Europe, early modern China, etc.) |
However, this kind of big data has its limits, especially when using literary texts to populate the database—literary texts written or compiled by specific authors, in specific historical contexts, following specific genre conventions, and applying their own ideological lenses. Slavery did not just impact Islamic society in terms of trends and percentages, but by fueling debates about what it means to be Arab or non-Arab, moral or immoral, weak or powerful. I will end by suggesting ways to integrate big data trends and careful literary analysis of historical texts.