In Fall 2022, 15 students enrolled in an English 290 course on “Fanfiction.” The course aim was to contextualize contemporary fan writing in a longer history of intertextual writing, and to bring in quantitative methods to examine digital fanfiction texts. Approximately half of the students were majoring or had one of their majors in the home department of English, and the other half came to this course from disciplines like engineering, biology, and computer science (with a portion of them double majors between a science and a humanities discipline). The course allowed for an engagement with canonical texts and close reading methodologies perhaps unfamiliar to the students of the sciences, and exposed humanities students to basic coding and data analysis. Those that were double majors were able to blend their skill sets in service of this class. Students of all disciplines had previous knowledge to bring to the table, and the classroom allowed for collaboration and interdisciplinary research integrating quantitative and qualitative methodologies.
Digitally written fanfiction is an inheritor of much longer traditions of writing from sources, and students read canonical texts to contextualize such writing practices. After reading selections from The Iliad, students then read and considered how Virgil appropriated Aeneas for his Aeneid, and how Dante likewise included figures from myth as well as Virgil himself in the Inferno. Students also read Jean Rhys’s Wide Sargasso Sea against Jane Eyre as a case study.
Alongside these traditional literary texts, students also learned about media fandom origins (it all began with Star Trek!), changes in fandom over time, and situated their own fan practices within this framework. Since digital fanfiction is written at an astounding scale, with hundreds of thousands of stories written across many fandoms, the course required students to engage with digital humanities methodologies in order to properly work at scale with these texts.
For the duration of the semester, students selected their own contemporary fandom of interest (ex. Harry Potter, Gilmore Girls, Genshin Impact, One Piece, Euphoria) to explore. Beginning journal assignments asked them to read fanfictions in their fandom of choice, reading popular and recent works to qualitatively begin to understand their fandom. They reflected on the kinds of stories that were popular, comments from other fans on the stories, and tropes they began to see as unique to their fandom.
Look at the comments on stories you’ve read and are reading. Find at least one that has many comments on it. Fanfiction, of course, relies on a passionate community. What kind of conversations are people having in the comments of stories? Is the tone generally the same across different stories in the fandom, or are they catered to the specific story? What kind of reactions do fans have? If it’s a multiple chapter fic, are there a lot of people commenting across chapters? What about the author, are they present in the comments? Please include the link to at least one of the stories whose comments you look at.
Students did not just read stories—they wrote their own fanfictions in their fandom as well, practicing the kind of writing they were analyzing. Students were encouraged to experiment and write any kind of story they wanted, and reflected on the process in a reflection paper, which was the graded element. In the reflections, almost all students found this to be harder exercise than they thought it would be, and they achieved a greater respect for the writings they were working with. These qualitative assignments gave students a grounded understanding of their fandom, and helped to motivate sharp and insightful research questions as quantitative tools were introduced.
Students learned simple command line prompts in order to use a ready-made scraper from GitHub to extract stories and metadata for analysis from the popular fanfiction site Archive of Our Own. Once each student had a data set, a variety of quantitative tools were offered to help students explore the data they collected. These included natural language processing with DocuScope, network analyses with Gephi, and visualizations of data with Tableau, in order to explore each student’s interest in their respective fandoms. With close readings and findings buttressed by quantitative data, they completed a final project that answered a research question they were interested in exploring. Students looked at the popularity of characters over time, agency of characters in fanfiction as compared to agency in a piece of media, the popularity of posting following particularly exciting TV episodes airing, and more.
The following visualizations are a selection of student work and are available for your exploration (all shared with permission from the students.) Full student project sites can be found here.
This course was derived from my own work within the digital humanities. Select chapters of my dissertation as well as a project examining popular readers employ computational methodologies to study readers at scale.
The StoryGraph Project:
In Fall 2023, at The Association for the Study of the Arts of the Present (ASAP) conference, I presented data from my ongoing project (the eventual findings from which I plan to compile into an article for publication) examining the new popular book reviewing site, The StoryGraph.
A decade ago, Lisa Nakamura wrote on Goodreads, calling the site a new “social valence of reading.” Goodreads, since acquired by Amazon, now boasts over 125 million users; but given rising uneasiness with Amazon’s labor exploitation, environmental impact, and data and privacy issues, readers have begun to utilize the Black-woman owned app The StoryGraph for a more ethical place to track reading. The app offers many analytics, including the ability to tag and find books based on “mood.” When reviewing, readers are also prompted to answer questions, including whether there is a “diverse cast of characters,” challenging readers to reflect on diversity, broadly conceived. The social valence readers encounter on this site, then, differs from Goodreads in its data-inflection and social conscious. This kind of data can allow scholars to begin to explore at scale the influence diversity may have on ratings, the moods that accompany reading, and how reception differs between Goodreads and The StoryGraph.
The subject matter of the fanfiction course emerged out of my research on a dissertation chapter titled “Somebody’s Story: Writing the Real Person as Character in Digital Fanfiction.” While my students were encouraged to think broadly about more “traditional” fanfiction writing, my dissertation chapter is more incisively interested in how fans blur distinctions between real actors and the characters they play in the genre known to fans as “Real Person Fiction.” My theoretical framing of how fans come to write this kind of story, and the larger implications this kind of writing has for popular conceptions of literary character, also necessitated large-scale analysis of fan writing.
I employed BookNLP, a natural language processing tool (more complicated and computationally demanding than the DocuScope tool I employed in the classroom) to read a dataset of approximately 8,000 fan stories from the fandom Supernatural and the parallel stories about the actors who play the characters on Supernatural. These findings helped me to examine the degree to which fans write about the body of the actors and the body of characters, and the intersections this has with the explicitness of stories. The use of data in this chapter for a large corpus of born-digital texts supplements my more traditional dissertation chapters, all in service of looking at the influence of the digital on how we read and write literary characters.