Tooling Up for Digital Humanities

Early forms of text analysis relied on paper technology and were incredibly labor intensive. While there were examples of Parisian monks creating concordance lists for the Vulgate bible in the thirteenth century, a more modern forerunner was Father Roberto Busa’s Index Thomisticus, which started in the 1940s as an attempt to analyze the complete works of Thomas Aquinas. Busa linguistically annotated and indexed all of the words by hand on index cards, which were later converted to analogue form and published in the 1970s.

The quantitative turn in history of the 1960s and 1970s also opened the field to more macro-level analysis . Historians began counting and measuring, from accumulating probate records of New England towns, for instance, to doing bigger-picture analysis of census data. But with the new wave of quantitative analysis emerged a strong backlash, as many humanities scholars critiqued the social scientific methods for what they saw as a failure to draw humanistic meaning out of data (see, for example, the controversial arguments concerning the profitability of slavery in Stanley Engerman and Robert Fogel’s book Time on the Cross).

In the realm of literature, forms of text mining have recently been characterized more broadly as a form of “distant reading”, a term coined by Franco Moretti to describe a macro-level analysis of literature. Distant reading allows analysis of a larger sample size, which can complement a more traditional analysis relying on a close reading of a small number of texts. While traditional methods often involve extrapolation based on a small sample size, mining a broader swath of data offers an opportunity to find wider larger, more representative patterns. Moretti argues that a broader scope allows scholars to analyze the “great unread” of non-canonical literature so often ignored by literary critics.