Text Mining and the Conceptual History of the "Republic of Letters"

All that we know about the early modern Republic of Letters, from the heterogeneity of its membership to its continued significance in the learned world of the sixteenth, seventeenth, and eighteenth centuries, is based on traditional historical research: the close-reading of historical documents. Yet, as the number of primary sources shared online keeps growing, it is time to discover how computational approaches can advance our understanding of this complex community. In my paper, part of my research for the SKILLNET project, I will explore the use of digital text mining in the study of the conceptual history of the ‘Republic of Letters’, investigating if and how the distant reading of a large corpus of letters can trace key concepts that relate to a sense of commonality and to the ideal of sharing knowledge. Looking at the frequency and spread of words, I am mapping the main ethical notions which held this learned community together. Did these notions change over time or differ according to, for example, region, language, or religion? Can they tell us if this knowledge society represented a utopian idea, detached from religious and political concerns, or if we should explain its long-term significance in relation to its pragmatic value, allowing scholars to share their thoughts and texts with each other? Exploring these questions, my paper will touch upon the construction of my principal dataset, comprising of more than 80,000 early modern letters, and discuss the complexities of conducting experiments with a historical and multilingual corpus.