A World of Fiction: Digital Collections and the Future of Literary History by Katherine Bode (review)

It has been three years since Katherine Bode electrified the digital humanities conversation by arguing (in an essay for Modern Language Quarterly) that close and distant reading were not opposites, but rather twins. She pointed out that both modes bracket off the historical, contextual details of literary production, treating texts as stable and selfidentical arrangements of words to be analyzed. For Bode, the missing counterforce is bibliography, which examines details of production and circulation to situate print as an unfolding event in time and space. That view animates A World of Fiction: Digital Collections and the Future of Literary History (2018), Bode's two-part monograph based on her edition of digitized historical Australian newspapers. The first half lays out a model of corpus-building for "data-rich literary history," one that uses book history, bibliography, and the principles of scholarly editing to curate digital models of literary systems (3). The second half deploys various digital humanities methods (bibliometrics, network analysis, and topic modeling) on a curated collection of nineteenth-century fiction as printed in Australian newspapers. As a sign of the strength of Bode's work, many of her concepts (for example, the importance of documenting and sharing data, and the emphasis on book history) have now been widely taken on board by the digital humanities community. For Victorianists, A World of Fiction is both a guidebook to digital engagement with the past and a nuanced report on Australian literary publishing in the nineteenth century.

Demonstrating the inadequacies of idealized distant readings by critics like Franco Moretti and Matthew Jockers, Bode offers a better path, one leading away from those who "present literary data and digital collections as pre-critical, stable, and self-evident" (20). Essentially, Bode shows the virtues of self-awareness when it comes to digital archives: they really should come with warning labels. She reminds us that all datasets are constructed, interpreted fields; that all data is "capta" (Johanna Drucker's term), shaped in the ways it is taken; and, moreover, that the print record is a very messy business (Johanna Drucker, "Humanities Approaches to Graphical Display" in Digital Humanities Quarterly 5.1 [2011], paragraph 3). As a result, both users and makers of digital collections have to engage in metacritical corpus studies, asking of both analog sources and their digital surrogates, where did this come from? By what processes of mediation and remediation was it framed? What is missing or obscured? And perhaps most tellingly, how is my analysis itself reframing or distorting the object or system of objects I am analyzing? To prompt answers to such questions, Bode calls for digital scholarly editions of literary systems, ones that can move analysis beyond studies of new literary production (tied to date of publication) and toward more complete models of historical contexts by mapping details of printings and reprintings across time and space.

There is a tension between this call for greater historical detail in digital models and Bode's assertion of the "explicitly interpretive and contingent" nature of such models (53). Scholarly editions are arguments—that is, they depend on interpretive decisions at various scales—and they are also firmly grounded in empirical observations. In Bode's terms, they are at once "hypothetical" and "stable" (53). But how interpretive, contingent, constructed, and conditional (even fictive) can a model or edition be without breaking loose from its historical commitments? Or to ask this another way, how do we know when we have reached the "pre-critical" ground of facticity upon which we can confidently build literary-historical arguments (20)? The more I was convinced by the first half of Bode's book, the more I found the second half presenting suggestive hypotheses rather than proven truths. And now I believe that is intentional: Bode wants us to see even her own striking claims about Australian literary history as new artifacts, bearing a real but indeterminate relation...