Network Theory Circa 1800: Charles Brockden Brown'sArthur Mervyn

Abstract

This essay investigates an early instance of “network theory” in order to argue that such theories did not, as most scholars suggest, emerge exclusively in the digital age. Charles Brockden Brown's Arthur Mervyn (1799–1800) attempts to theorize the information networks of the early American republic by comparing the spread of information to the spread of yellow fever. Unlike other novels that focus on the spread of contagious disease (such as Dickens's Bleak House), Arthur Mervyn refuses to trace a clear path of transmission from person to person. Instead, the randomness of the fever's spread gives Brown a stark, dramatic way of visualizing the unpredictability built into all such chains of transmission. Rather than a mark of nostalgia for an earlier age of “face-to-face” communication, this interest in casual and ephemeral channels of communication (and indifference to print) is a mark of the text's modernity. What Brown recognizes in his analysis of emergent networks is the power of the city to reorder social connection, enabling individuals to bypass official sources of information, play a role in the process of transmission, and become (often unwitting) participants in a transformed public sphere. In Arthur Mervyn, the yellow fever epidemic works as a fantasy of exposure, an impossible kind of social transparency that ultimately serves as a map for comprehending the mysterious workings of a “connected age.”

Measuring Information Propagation in Literary Social Networks

“…With the rise of sociological approaches to narrative, work in literary criticism has increasingly turned to the ways in which authors depict social networks in their texts. This includes critical attention to both network topologies, such as understanding characters and their structural relationships with others (Levine, 2009), and information flow, such as theorizing the representation of disease and gossip (Levine, 2009;Margolis, 2012;Spacks, 1985). Much computational work in NLP has arisen to support the former line of research, including extracting social networks from text , predicting familial relationships (Makazhanov et al, 2014), and modeling the interactions between characters (Iyyer et al, 2016;Chaturvedi et al, 2017).…”