Review: “On The Perceived Complexity of Literature: A Response to Nan Z. Da”

In an article entitled, “On The Perceived Complexity of Literature: A Response to Nan Z. Da”, Fortis Jannidis refutes the following idea brought forth by Nan. Z. Da: “CLS’s methodology and premises are similar to those used in professional sectors (if more primitive), but they are missing economic or mathematical justification for their drastic reduction of literary, literary-historical, and linguistic complexity. In these other sectors where we are truly dealing with large data sets, the purposeful reduction of features like nuance, lexical variance, and grammatical complexity is desirable (for that industry’s standards and goals). In literary studies, there is no rationale for such reductionism; in fact, the discipline is about reducing reductionism,” (Fortis 1). Jannidis is the chair for literary computing and German literary history at the University of Wurzburg. His primary areas of study include the history of the novel, macroanalysis of the German novel, computational literary studies, and stylometry.  He serves as a co-editor of the Journal of Literary Theory and the series Narratologia, (Jannidis). His article is written in response to Nan Z. Da, a Dorothy G. Griffin associate Professor of English at the University of Notre Dame. Her research is primarily situated on comparative literature and critical theory, as well as nineteenth-century American and Chinese literature and literary history, (Marketing Communications

Da argues that quantitative methods cannot produce any useful insights into literacy texts, a claim which Jannidis cites through the quote utilized above. Jannidis argues that “the quality of specific studies cannot be the basis of an argument for the question of whether quantitative methods can be applied to literary texts with valid results,” (Jannidis 2). He takes issue with Da’s claim that computational literary studies has no ability to capture literature’s complexity, and explains that the reach of her analysis is not broad enough to truly support this claim on top of the fact that her argument contains a logical fallacy as she utilizes evidence that does not work to further her objective. Jannidis then goes on to further explain why complexity in computational literary studies is not lost as Da proposes – citing first and foremost that “it is hard to believe [the] argument that [literary studies] is singularly complex,” (Jannidis 3-4). It is this assumption made by Da that permeates the rest of her smaller claims, which Jannidis utilizes to breakdown her overarching argument that computational literary studies is unsuccessful in examining complexity. Jannidis then goes on to breakdown Da’s issue with statistics as well as selection bias and coda, continuing to utilize Da’s claim regarding complexity and explaining that Da herself simplified aspects of computational literary studies to a degree that scholars in this field do not. Jannidis ends his argument by examining a final claim by Da – that computational literary studies “cannot actually set out to answer the question it sets out to answer or doesn’t ask an interesting question at all,” (Jannidis 11). He goes on to cite the work of two highly respected DH scholars, Ted Underwood and Hugh Craig, explaining that much of this computational work is based and builds off of interested non-quantitative questions already addressed in literary studies, which contradicts the notion that these are uninteresting questions to all – they likely just seem that way to Da.

Jannidis’ argument is clear and concise. He breaks down the large components of Da’s argument, focusing primarily on these abstract claims that are near impossible to prove to be true or untrue as Da seeks to. His sources are well explained and integrated, citing studies done by other computational literary studies scholars and giving context to those claims through his introduction and analysis. A bulk of his work is dedicated to taking deconstructing the claims in Da’s argument, with cited evidence coming later in the piece, but this is a much needed background for the reader to understand exactly what claims Jannidis is examining and why. Overall, this is a successful breakdown of an argument that, as Jannidis explains, makes sweeping claims that even Da herself struggles to support. Jannidis’s analysis aligns with the work done by scholars like Ted Underwood in “Seven Ways Humanists Are Using Computers to Understand Text” or Michael Gavin in “Is there a text in my data? (Part 1): on counting words”. Both these pieces of scholarship seek to examine and legitimize the ways in which computational literary studies is successful in its approach of analyzing text and why it is necessary.

Overall, this piece was highly successful in teaching me ways to support my own claims that digital humanities and computational approaches to literature are neither unneeded nor uninteresting. Jannidis’ emphasis on closely breaking down sweeping claims and supporting his analysis with well supported research done by prominent scholars is a successful approach to informed criticism, which I hope I will be able to engage more with in the future.

Works Cited

Gavin, Michael. “Is There a Text in My Data? (Part 1): On Counting Words: Published in Journal of Cultural Analytics.” Journal of Cultural Analytics, Department of Languages, Literatures, and Cultures, 25 Jan. 2020,

Jannidis, Fotis. “On the Perceived Complexity of Literature. A Response to Nan Z. Da: Published in Journal of Cultural Analytics.” Journal of Cultural Analytics, Department of Languages, Literatures, and Cultures, 25 Jan. 2020,

Jannidis, Fortis. “Fotis Jannidis  .” Fotis Jannidis Research,

Marketing Communications, University of Notre Dame. “Nan Z. Da.” Department of English at the University of Notre Dame,

Underwood, Ted. “Seven Ways Humanists Are Using Computers to Understand Text.” The Stone and the Shell, 30 Nov. 2015,

Leave a Comment

Your email address will not be published.