Questioning the Need for and Utility of Digital Techniques
Questioning the Need for and Utility of Digital Techniques
Many humanities scholars feel that that the adoption of computation techniques in the humanities have been objectively unsuccessful. Although computational methods have applied to literature for over thirty years, they have not had a large effect on literary history. Writing in 1993, Mark Olsen of the University of Chicago stated: “Computer-aided literature studies have failed to have a significant impact on the field as a whole” (Olsen, 1993, cited in Underwood, 2017). Olsen traces this failure to the inability of computational methods to perform refined analysis of “subtle semantic or grammatical structures in single texts or the works of individual authors”, and that those methods cannot represent or process the complexities of language (Olsen, 1993).
Modern and Classical Languages scholar Jonathan Green of University of North Dakota, in a 2015 blog post, criticizes the overreliance on computation in the humanities, and raised questions about the entire project. He discusses an article in the New York times by Armand Marie Leroi, a professor of developmental biology at Imperial College in London, who presents the case that the humanities must become a more mathematical discipline if they are to survive.
Green states bluntly: “If we do in fact reach the point where the digital humanities expresses its results ‘not in words, but equations,’ where the ‘analog scholar won’t even know how to read the results,’ then the digital humanities will fail.”
A related issue is that the digital humanities may reach a point where words are not the end product of humanities scholarship, but rather equations and numbers. In that case, it is unclear whether the digital humanities will address the guiding questions of humanities scholarship or present their results in a manner accessible and comprehensible to the wider humanities communities. Green therefore challenges Leroi’s contention that the approaches of biology, mathematics, and computer science are directly mappable to the humanities.
As Green puts it: “It is furthermore not at all clear that the tools of evolutionary biology, where reproduction is the first imperative of the most basic building blocks of life, should apply to culture, where it is not.”
Algorithms currently support distant reading better than close reading, the type of reading most practiced by humanities scholars. Green summarizes the situation with the observation that mathematical, statistical and computational approaches are “dead ends” when applied to the humanities. He cites the example of semantic encoding of mediaeval texts in the 1970s and 1980s that were intended to facilitate searching by semantic collocations (idiosyncratic expressions or groups of words that are syntactically uncommon and do not appear with statistically frequency), in addition to text searches, and which Green has deemed a failure. However, he also describes corpus linguistics, in which languages (li