Before You Read: A Reconstructed Literary History in Reading by Numbers

To enhance our understanding of overarching trends and paradigms in literature, digital humanists have long argued that we must go beyond minor expansions of the literary canon and embrace quantitative methodologies. Matthew Jockers, for instance, critiques the close reading practice of drawing “conclusions about literary periods from a limited sample of texts” and asserts that “generalization from the specific can be particularly dangerous when the texts examined are not representative of the whole”  [Jockers 2013, 47]. Along these lines, Katherine Bode examines Australian literary history from a new vantage point in her 2012 book, Reading by Numbers: Recalibrating the Literary Field.

One of the strongest points of Bode’s analysis is her refusal to employ simple binaries to discuss quantitative literary studies. She does not position distant reading as a replacement for careful textual analysis, but as one tool that allows us to explore new questions and old assumptions — not as positivists, not as all-knowing, objective, bean-counting gods — but as humanities scholars who decline to use only half the tool chest available to us. Far from dismissing the salient criticisms of distant reading, she provides a nuanced and thoughtful response to critics like Katie Trumpener, Robert Tally, Gayatri Spivak, Rachel Serlen, and Jonathan Arac. For example, she includes Trumpener’s argument that distant reading reduces “books to mere commodities…as if the book’s content (and the irreducibility of authorial style) was virtually irrelevant” and Arac’s claim that synthesizing texts this way suppresses individual reader interpretation via “covert imperialism” [Bode 2012, 10–11]. In response, Bode acknowledges that there is indeed a complex “rhetoric of objectivity and truth” that surrounds computational methods; however, she asserts that it’s impossible to completely avoid power inequities in research as language itself, whether used in qualitative or quantitative scholarship, is inherently imbued with hierarchies [Bode 2012, 12].

Therefore, Bode is careful to put her own research into context, caveating that no database can hope to capture every scrap of the written word and no analysis can represent the whole truth of publishing history (especially since her corpus was compiled through archiving practices, which select some entries and omit others.) Instead, she is clear that her mission is not to unequivocally defend one method of analysis or one epistemological view, but to uncover new patterns in literary history. Thus she maintains that we need to think of numeric interpretations in the same way we think of text, recognizing:

…them as a form of representation and, as such, to explore how they operate and the ways in which numbers accrue authenticity and authority. Like language, numbers provide an imperfect and mediated way of accessing the world; but in the absence of any perfect or unmediated access, they are tools we can use in our attempts to understand and investigate the literary field.  [Bode 2012, 12]

As an explanation of her methods, in the chapter “Literary Studies in the Digital Age” Bode introduces her corpus of study: an online archive called AustLit created through collaboration between the National Library of Australia and over one hundred researchers. The archive includes hundreds of thousands of works and authors and is updated with new titles frequently [Bode 2012, 20]. Over four years, Bode studied the database and updated her records every six months to account for new entries. Ultimately, her findings are crystallized in four content chapters and twenty-four empirical graphs.

Although she is not overly explicit about her day-to-day methods in the book, she notes that she relied on a process of “modeling” to generate her findings and conducted a “form of distant reading”  [Bode 2012, 19]. In other words, she examined cross-sections of AustLit bibliographic data to test hypotheses about readership and publishing. Then, based on the models and graphs she built, she identified patterns and explored emerging points of interest in greater depth. Though the book’s “Notes” section provides more information on methods and URLs within AustLit, a more overt explanation would have been a welcome addition to the book.

Also of note, Bode’s form of exploration might not mesh with everyone’s definition of “distant reading.” In fact, her book title might be a misnomer if one expects a book about macro textual analysis or teaching a computer to identify linguistic genre markers (e.g. Moretti’s Graphs, Maps, and Trees). A more appropriate title might be Trends in Historical Publishing or The Numbers Behind Reading. In essence, Bode’s methods constitute more of a study of the market for Australian literature over time than actual computer-aided reading. Her introduction explains her process as “mining, modelling, and analyzing data”  [Bode 2012, 1]; in other words, she does not look at the text of thousands of books via computational practices; rather, she examines aggregate trends in metadata and economic markers which show authorship, publishing, and readership circulation during certain historical periods in Australia.

As a demonstration of the types of information that scholars can uncover with quantitative methods, Bode's chapter “Beyond the Book: Publishing in the Nineteenth Century” examines the publishing history of Australian authors in the 1800s. She investigates the relationship between Australian authors and British publishers to explicate the perceived marginalization of local publishing. Whereas one hypothesis might hold that there was a one-way stream of publications from colonizer to colonized, Bode's construction shows a much richer history of give-and-take between Australians and Britons with a particular focus on local colonial publishers. For example, she explains the financial challenges associated with printing books in Australia given the innately smaller readership base in this colonial era. As a result, Australian publishers had to diversify their publishing models to make ends meet, and many authors considered local publishing to be “the avenue of last resort” due to its meager economic rewards [Bode 2012, 33]. Publishing their work through British publishers offered colonial authors more stability, the possibility of more reprints of their work, and a certain cultural esteem.

In response to these challenges, many Australian authors turned to serialized publication, publishing their work in local newspapers instead of novels. In looking at sales figures and other data, Bode makes a case that scholars have not examined periodicals closely enough to comprehend the impact they had on the authorial culture of the time. In particular, her studies of AustLit show that serial publication became a significant mode of production a decade earlier than previously thought [Bode 2012, 37]. She asserts that the rise of serial publications actually predated the technologies of the 1870s that were thought to be the genesis of this mode. Additionally, she points out that the broad regional circulation of local editorial content meant that the readership of local authors was much broader than previously thought.

Bode then jumps ahead in history in “Nostalgia and the Novel: Looking Back, Looking Forward” and examines beliefs surrounding the “golden age” of Australian publishing (circa the 1970s and 1980s). In this section, she critiques the pervasive narrative that this era represents the peak of local Australian publishing, spurred by an increase in government funding for publishing activities. In contrast, Bode’s data indicates that local publishing was at its highest directly after World War II and declined during the “golden age” (Figures 4 and 5). This directly contradicts the belief that publication of Australian works was almost entirely the purview of Britain prior to the 1970s. Bode speculates that much of the discrepancy between concrete publishing figures and perception is due to nationalistic desires that paint certain historical periods in nostalgic terms. She posits that this underlying nationalism elevated certain works and ideologies and effectively discounted other prevalent genres (like pulp fiction). Therefore, this selective social memory has led to an inflated view of publishing during the golden age.

Another area of marginalization that Bode explores is that of gender in publishing. In “Recovering Gender: Rethinking the Nineteenth Century,” she challenges various scholarly perceptions of female authors. For example, she demonstrates that women actually represented a larger percentage of the works published in the nineteenth and early twentieth centuries than previously thought (Figure 15). Her work indicates that early female authors were a significant component of the publishing market — not eclipsed and ignored writers, lacking in agency [Bode 2012, 108]. Mapping back to the earlier discussion of periodicals, this chapter also shows that women were more likely to be published in periodicals than men in Australia between 1860 and 1899 (Figure 16). Additionally, although colonial male authors were published in novel form more often, in certain time periods, female authors actually exceeded male authors in terms of books published in Britain (Figure 18). These are just a small sample of Bode’s complex set of findings which redefine many existing conceptions of gender in publishing.

Finally, Bode examines macro trends in authorship and gender from 1945 to 2009 (Figure 19) in “The Rise of the Woman Novelist: Popular and Literary Trends.” This section reveals that males dominated the novel publishing field up until the 1990s when female novelists surpassed them. This supports existing feminist literature which trumpets the growing prominence of female publishing in the late twentieth century; however, Bode calls for a more nuanced understanding of such trends that does not rely on a binary of pre- and post-female liberation. She points out that much of the growth in female publications corresponded to an increase in popular fiction genres like romance (written by women for women). Ironically, many scholars consider this type of literature to be counter to female agency and not emblematic of societal liberation [Bode 2012, 152]. Thus this increase may signal publishing growth as driven by a popular market and not by a political movement. This is yet another example of the trends and corresponding hypotheses that Bode advances regarding Australian publishing

Overall, Reading by Numbers is a stimulating book that brings new quantitative methods to bear on longstanding questions and assumptions in Australian publishing history. Bode effectively demonstrates the value of these methods in real-world contexts without over-asserting their primacy or authority. Instead, she argues that applying numbers to literary history is one meaningful form of interpretation that is situated and malleable, given the ever-changing nature of literary records and the subjectivity of the researcher’s decision-making process. Her findings help us better understand how publication format, gender, and colonizing power affected publishing in Australia and how local authors developed their own identities alongside Britain’s influences. Applied more broadly, this type of research data can have important implications for literary scholars: enabling us to form a richer historical picture beyond specific privileged works, helping us to spot changes in the macro publishing timeline, and empowering close readers to know where to zoom in and look more carefully at shifts and anomalies in history.