Combining Test Tubes and Poetry

Digital humanities professor Brian Croxall discusses the importance of translating undergraduate research to the humanities field.

Image from

If you Google “undergraduate research,” you’ll find pictures of students dressed in lab coats, looking intently at test tubes or petri dishes, pinching pipettes, or pointing at computer screens. It’s all very scientific. One of BYU’s newest faculty members, Brian Croxall, is challenging this collective perception of what undergraduate research looks like. Together with his students, Croxall is adopting the practices of these lab-coat-wearing researchers by seeking answers to questions that have never been asked before—but instead of peering into microscopes, his students are peering at book pages. At a Humanities Center Colloquium, Croxall discussed two research projects he is conducting in his digital humanities classes, providing an example for how to, as he says, “combine poetry and test tubes to bring real research in the classroom.”

In the field of digital humanities, scholars are boldly crossing disciplines to find new ways to analyze literature. According to Croxall’s definition, “Digital humanities is the use of computational tools or approaches to find patterns in humanistic production, when those patterns are then used for interpretive purpose.” In other words, digital humanities offers a new way to interpret literary texts, film, art, history, and more by incorporating technology. Digital humanities has gained attention over the last decade, including the formation of BYU’s Office of Digital Humanities, and Croxall is among scholars who are on the forefront of these new methods of research.

Croxall says that studying digital humanities was alluring because “it seemed to open up a new world of methods and questions that could drive not only my own research but also my work with students.” While attending this year’s annual university conference, he felt inspired by BYU president Kevin Worthen’s remarks encouraging professors to reject a dichotomy between teaching and research. With Worthen’s counsel in mind, Croxall confronted the question: What does undergraduate research look like for the humanities?

“In the humanities, it turns out we don’t necessarily make it easy for our students to ask questions that generate new knowledge,” he continued, “I think the key is that we need to ask our students to do real research. We need to give them questions we don’t know the answers to.” To test his hypothesis, Croxall has introduced students to real research questions through class assignments—without knowing the answers beforehand.

One class assignment was aptly titled, “How to Not Read Hemingway,” because, as Croxall joked, “No matter how much someone has written, you’ve always got time to not read it.” In the project, the class read fewer than fifty pages of Hemingway’s short stories and then digitized the entirety of Hemingway’s works (including a full collection of short stories, nine novels, and six nonfiction works) to create a full Hemingway corpus. From the corpus, the students formed groups and conducted analyses of his work using text analysis software called Voyant. One group observed that Hemingway’s word density in his posthumous works is 60% higher than the word density in works published while he was alive, and they conjectured that the difference could be caused by different editing procedures or perhaps editors who couldn’t quite capture Hemingway’s between-the-lines style. Another student used the Hemingway corpus to train a predictive keyboard to write a Hemingway-style short story.

Image from

Another long-term class project was inspired by a letter from the Special Collections at Emory University written by the poet laureate of England, Carol Ann Duffy. The letter expressed Duffy’s belief that her her fifth book of poetry, The World’s Wife, was different enough to warrant sending it to a different publisher. The students, taking a cue from Duffy’s own explanation, decided to see if this fifth book was empirically different from Duffy’s previous works. Using the same basic research process as the Hemingway project, the students found that although all of Duffy’s poetry has the same linguistic signature, readers can feel a distinction between The World’s Wife and her previous collections because the words are employed to different effect.

Croxall notes that digital-driven research is not meant to replace established humanities research methods. “I want to be clear that traditional humanities research is valuable, and so is traditional humanities pedagogy,” he said. “Most often, the process trumps the product: that’s what learning is about.”

However, opening a space in the class for yet-unanswered questions can help students to get excited about the learning material. Discussing the results of the Hemingway project on his classroom, Croxall said, “I’m confident that my students found and chased down things that Hemingway scholars have never thought to ask before. They found new things that they were interested in, and that I know people haven’t published on. And the class was energized, I think, as we worked to find out something that I simply didn’t know the answer to.”

Morgan Lewis (B.A. English ’18)

Morgan covers events for Humanities Center in the College of Humanities. She is a senior pursuing a degree in English with minors in editing and digital humanities.