This project aims to do a distant reading of five hundred years of Chinese history, spanning the twenty-six emperors and five-hundred and forty years of the Ming and Qing Dynasties. To do so, I will conduct machine-learning based topic analysis of the court records of the Ming and Qing Dynasties, called the Veritable Records or Shilu. This analysis is intended not only to reveal the broad patterns of Chinese event history from the perspective of the court, but also to serve as the basis of an investigation of the changing institutional behaviors and conceptual trends of the imperial bureaucracy.
In this experimental study, I will examine both the medium and the message of Chinese court documents in order to present a perspective on the Chinese past that is nuanced in its breadth rather than its depth, and that can serve as the basis of further study. The aim of this project is neither a narrative based on a small subset of these court documents, nor is it simply a work of archiving. Instead, I aim to "read" the entire corpus, to examine the overall pattern of events as well as the pattern of event coverage by the imperial bureaucracy. The bias of the Veritable Records is generally considered "well-known," foregrounding issues of national and political importance. I argue that to truly understand this bias, we must examine it systematically. To do so, I will use LDA Topic Modeling to do a complete reading of the corpus, presenting an analysis of both the general flow of events and the changing nature of their coverage by the imperial bureaucracy.
Back to Networks and Network Analysis for the Humanities: Reunion Conference