This panel offers a series of new computational approaches to understanding the forms through which authors of the eighteenth-century produced, mediated and circulated knowledge. The Enlightenment was a period of expanding information (Siskin 2016). But information--raw data--by itself is indistinguishable from noise: to be understood and acted upon, it must be systematized, catalogued and abstracted (Shannon and Weaver 1949). Faced with a wealth of information beyond the capacity of any one human mind, Enlightenment thinkers sought methods by which to order it, to make it comprehensible and, more importantly, useful. In other words, data had to be mediated into a tractable form before it could be considered knowledge (Yeo 2001).

Eighteenth-century knowledge projects such as Chambers Cyclopedia, the French Encyclopédie, the Encyclopedia Britannica, and even Johnson’s Dictionary, can all be understood as experiments in mediation. All were attempts to communicate information in a structured fashion so that each fact, each fragment of data, could be understood to be part of a greater whole that could be grasped, understood and rationalized by the user.

The complexity of these knowledge systems and their contingency on the limitations of human reading has hidden their shape from scholarship. In recent years, however, their re-mediation into digital formats has rendered these hidden shapes of knowledge newly visible. Visualized, for example, as a networked system of links, the renvois of the Encyclopédie becomes a legible system of disciplinary knowledge (Heuser, Algee-Hewitt and Bender, 2016, Blanchard and Olsen, 2002). As a network of shared quotations, the Dictionary reveals an implicit critical apparatus that informed Johnson’s choices (Algee-Hewitt, 2019). And Chambers Cyclopedia, visualized by cross-reference, uncovers the fuzzy boundaries and interdisciplinary links that characterized eighteenth-century knowledge (Rudy, 2019). In each of these cases, the computational remediation of the knowledge object allows scholars to reconstruct the original mediation experiments that informed the eighteenth-century knowledge project. The ability of the computer to abstract information, to quantify connections, and to link from the objects themselves into the wider world that they sought to represent, allows us to reevaluate the relationship between information and knowledge that lay at the heart of the Enlightenment project and to understand the ways in which period thinkers sought to communicate this relationship to the larger reading public.

Each of the papers on this panel adopts a different computational strategy to explore the hidden forms of communication that are embedded in the different Enlightenment strategies of mediating knowledge. Each also seeks to understand a different kind of knowledge object. Yet all are united in their use of computational methods to reveal hidden patterns of ordering information in the eighteenth century. The full complexity of the patterns that exist in both single objects, such as an encyclopedia or dictionary, and in the very discourse surrounding how things can be known (and who can know them), are often hidden from those who originally created them. Lacking the information processing capability of the computational models in these papers, the original authors of the eighteenth-century were often unaware of the interplay between the explicit patterns they adopted, and the epiphenomena that occurred when these configurations interacted, producing patterns of interference across the period.

The first paper on the panel, Mark Algee-Hewitt’s “Decentering Knowledge”, explores the radical reorganization of knowledge in William Smellie’s original Encyclopedia Britannica. The text of the first edition offered readers a confusing hybrid between a dictionary that defined individual terms and an encyclopedia that sought to teach users larger concepts. In this paper, Algee-Hewitt combines network models and text analysis to argue that the Britannica was both at once: a hyper-linked document of hubs and nodes, whose patterns of text reuse between the longer articles and shorter definitions reveal a proto-digital document that anticipated a digital environment even as it left its contemporaneous readers baffled.

In “Mapping Knowledge and Mapping Rome in Piranesi,” Jeanne Britton reconstructs the “visual epistemology” (Drucker 2014) of Giovanni Battista Piranesi’s intricate images. At once maps, architectural renderings, text documents, and artistic renderings, Britton reveals the ways in which Piranesi’s images function as an interface to the layered knowledge that they contain. As the original printed images are remediated as an interactive, online, environment in Britton’s project, they reveal the intricate web of connections between the visual and textual, the geographic and the intellectual, and the factual and imaginary. Britton’s project not only returns attention to a key figure of the eighteenth-century, but demonstrates the extent to which the interface was a key technology in Enlightenment systems of knowledge.

Finally, Seth Rudy and Colin Jennings move beyond the individual work as they explore linked sites of abstracted knowledge in “At One View.” Drawing on Gale’s Eighteenth Century Collections Online (ECCO) corpus, Rudy and Jennings locate the nexus of image, text, and mind that characterize Enlightenment knowledge production in the repeated use of the phrase “at one view.” Across texts as diverse as histories, travelogues, and novels, the phrase functions as a signpost for a specific kind of Enlightenment rationality which seeks to render complex systems legible as objects of comprehensive knowledge. By locating the phrase (and its constituent words) within a series of increasing complex vector space models of the corpus, the authors not only are able to trace the evolution of the phrase (and the mediating concept it signifies), but also are able to better understand the implicit connections between its individual instantiations.