When Nitin Luthra received the 2023-24 Rhodes Doctoral Fellowship in the Computational Humanities, he decided to use the Rhodes at Duke University to design a unique interdisciplinary course. Merging the study of English literature with computational analysis, in what is commonly referred to as the Digital Humanities, this fall Luthra premiered “Enemy at the Gates: Reading Dystopian Literature” for Duke undergraduate students. The syllabus describes the course as a combination of humanistic close reading and coding analysis that introduces computational methodology to heighten our understanding of literature.
“What can we discover using computer software that can read thousands of pages at once?” Luthra asks.
This past semester, Luthra’s class read a string of novels – Lionel Shriver’s The Mandibles, J.M. Coetzee’s Waiting for the Barbarians, and Adania Shibli’s Minor Detail – focusing on dystopian tropes including authoritarianism, xenophobia, and othering based on race, ethnicity, nationality, etc., to name a few. In addition to their reading list, that includes not just literary but philosophical texts as well, Luthra’s students have been introduced to text-analysis digital software, which allows them to distant read using Voyant Tools, Orange Data Mining, and Python. Encouraged to take advantage of the slew of innovative technology at their fingertips, Luthra’s class analyzed texts using concordance, collocation, sentiment analysis, and topic modeling tools, in addition to traditional close reading.
When I ask Luthra about his interest in the intersection between STEM and the humanities, he says, “I think language is important everywhere. The English Department should not be confined to the study of novels and fiction. This is my endeavor to engage with other languages beyond natural languages, and try new methods to see what they yield compared to the more traditional modes of inquiry.”
“I have been dabbling with computational humanities ever since I joined Duke in 2021,” Luthra says. “My mentor Astrid Giugni and the Duke Libraries workshops have taught me to implement a number of the tools that I have been teaching – I started learning Python, and began natural language processing with Twitter, newspaper articles, and novels. Distant reading would be an amazing way to supplement and enrich close readings.”
The students in Luthra’s Dystopian class similarly affirm the success of this cross-disciplinary endeavor. While most of them are not students in the humanities, they have been studying both English and computational methods throughout the course of the semester in Luthra’s class. “It felt almost like a scientific paper,” says Judy Chen about the course’s second assignment, where students were required to write a distant reading analysis consisting of both manual close reading and computational textual investigation. “I don’t think I’ve ever had to write a paper like that before for an English class.” Tatum McKinnis agrees, commenting on the assignment: “It was interesting to do sentiment analysis for an entire book; being able to track it across an entire novel was very interesting.”
Using both VADER and SentiArt, many of Luthra’s students chose to analyze dystopian literature through the lens of computational sentiment analysis, where a text, or corpus, is uploaded into a software mining tool and each word is allocated a specific weight of sentiment. Where close reading allows us to systematically break down brief passages of a text, sentiment analysis does so for an entire corpus in a matter of seconds, assigning sentimental value in all primary ranges of emotion to a corpus. One can determine the breakdown of how positive or negative the language of a text frequents, or, using more advanced sentiment analysis tools, how prevalent connotations of fear, anger, surprise, or even disgust feature in a novel. And given the inherent subjectivity of any software design, the students also engaged in close readings of the novels in tandem with computational models.
“We’ve done close readings before, but looking for something over the entire course of the novel, which is impossible with the human eye, was interesting,” remarks Rohil Kanaparti. “Doing this with something even longer, possibly a really long piece where we don’t have the capacity to analyze every single word, would yield even better findings.” This conclusion is exactly what Luthra wanted his students to draw from his class. Not only can digital text analysis contribute to a broader understanding of English literature, but it can inspire careful reading practices with attention to the nuances of language across disciplines.
“What really drew me to this is that university departments and disciplines are often siloed very rigorously, where CS does one kind of labor, medical students do another kind, and humanities students another,” Luthra elaborates. “This division of labor is something I wanted to challenge. I want all of us to go beyond the traditional boundaries and methods that have been set for us and look at the other side of the aisle.”
Luthra’s dystopian literature course offers exactly what he aimed to establish: a third space where students are encouraged to delve into the connections between the humanities and STEM. For their final project, Luthra’s class will be advancing the research they have been conducting, creating either a conglomerate computational analysis of multiple texts using programming or a website using StoryMaps, a multimedia presentation software that designs interactive storytelling.
“Combining things that you might be more interested in with things you are not as knowledgeable about is always helpful,” Tatum says, reflecting on interdisciplinary study. In a class where the majority are STEM majors, it might seem difficult at first to engage an entire class in the study of literature. But Luthra takes this in stride, observing that, “The study of English, the critical thinking and close reading skills that we teach, are really important, and I want STEM students to take advantage of that. Every discipline should have these skills. This course is my effort to learn from students, and share what I know with them.” When designing his course, Luthra was clear that he wanted it to cater to both STEM and humanities students, expanding the knowledge of both groups. Where humanities students would be introduced to computational studies and programming software, STEM students would gain experience in literary analysis and exposure to great contemporary works of literature. “If someone didn’t want to take an English class at first,” Judy says, “but it included technology that they really like, this would really pull them into taking one.”
The cross-section between STEM and the humanities is larger than more people would assume. Using computational methods to study literature, while an emergent approach, holds the key to vastly improving our understanding of literature in a way we could never accomplish with close reading alone. Tools like Orange Data Mining, Voyant, and Python have the capacity to scan millions of texts in the blink of an eye, pulling data for readers to analyze in a way we have never been able to before. At the same time, when reading practices are plummeting across the board, such classes encourage students to develop close reading and critical thinking skills. Effective distant reading practices remain incomplete without close readings. Luthra’s course aims to share this emerging field with humanities and STEM students alike, changing the way we think about literature while also drawing us closer to it.
Author’s Note: Nitin Luthra has expressed that he will be seeking to offer this course again in future semesters, and I encourage any student curious about pursuing the exciting intersection between STEM and the humanities to register for his class.
