Models for Narrative in Digital Humanities

[This is a reading reflection for HIST946: Digital Humanities with Professor William G. Thomas during the Fall 2011 semester. This week’s reading was Franco Moretti, Graphs, Maps, Trees: Abstract Models for Literary History. You can find related posts here.]

The main argument put forth by Franco Moretti is compelling: instead of reading texts we should mine them for data. Moretti suggests a new way of analyzing literature. Instead of close reading that is usually the domain of the humanist scholar, researchers can think of reading as macro- and micro-reading. Most humanist work is micro-reading, a close reading of the text and the use of anecdotal evidence to support arguments. Moretti argues that tools now exist for us to step back to a macro-reading that enables scholars to spot patterns and notice things that may otherwise be missed with micro-reading.

He suggests three tools: graphs, maps, and trees. When we visualize data through graphs, we can spot basic trends (in his case, the visualizations of publication history). Data like this does not tell us why, it only says what is happening. So, Moretti suggests reasons for the observed data by arguing for the genres of 19th century British texts. The conclusions as he admits are not new; what is new is visualizing extensive sets of data makes a strong and less anecdotal case – a sort of empirical proof that the data supports. Mapping includes the charting of characters, events, locations, and relationships in the texts. The final tool, trees, presents a hierarchical visualization of large amounts of data, similar to flowcharts and genealogies.

Moretti’s methodologies offer a lot for digital humanists to think about. He calls for a “quantitative approach” relying on the repositories of information and moving away from reading individual texts and getting in return concepts. By stepping back and engaging in the modeling of topics instead of close reading, broader themes emerge. The models Moretti suggests does not eliminate reading, but rather suggests new ways of reading. The graphical visualization of non-graphical data (texts, in this case) allows scholars to present multiple dimensions of data and query whether relationships exists among data. As Moretti admits, quantitative research only gives data, not interpretation – visualizations give scholars an idea of when something happened, but answering why means confronting the evidence differently (p. 9).

However, the digital humanities are not wedded to the specifics of scientific rigor. Scientific visualization tends to rely upon statistics, while humanistic data must be adjusted for visualizations since the data relies on objects that are non-numerical and subjectively quantified. Humanists certainly have no lack of material to visualize, everything from letters, books, sentences, and phrases, to shifting relationships, emotions, and spatial chronologies. The ability to visualize material reduces its complexity, allowing scholars to communicate and argue visually rather than textually. The visualization of remarkable levels of digital information helps communicate the patterns of the past and reveals new ways to think about events.

Moretti makes an important point early on when he argues that “problems without a solution are exactly what we need in a field like ours, where we are used to asking only those questions for which we already have an answer” (p. 26). Instead of standing with a canon or a small set of theories, Moretti says scholars need a broader framework for examining works of literature. Digital humanities and the tools that exist shift analysis away from the fragmentation and specialization so common in humanistic inquiry today and causes digital humanists to think more broadly about events, people, patterns, and causes. Although close reading and specificity are important, scholars should resist the fragmented examination of texts and, with tools for visualization and data processing, widen their lens for a wholly different perspective.