For the HASTAC 2023 Conference on “Critical Making and Social Justice,” I propose a 90 minute workshop introducing participants to the “Queer Text Toolkit” (Qtt), a software package for computational literary research that is grounded in principles from Queer Studies. The project consists of two applications, “queer distant reading” and “queer text encoding,” which build from open source software and standards to offer an introductory, hands-on experience for quantitative text analysis and semantic markup procedures. As a digital component to my dissertation, “Since No Expressions Do: Queer Tools for Reading, Editing, and Archiving,” this toolkit draws from Queer Studies’s critique of heteronormative structures to problematize the reduction of gender, sex, and sexuality in literary text into computable formats. It reveals, in an introductory way, interpretive possibilities for experimenting with and reworking the limitations and constraints of text analysis and text encoding procedures. Here, users examine firsthand how digital formats, which collapse stylistic and formal expressions of gender, sex, and sexuality, can be redeployed toward creative, playful, and radical exploration. This exploration is guided by step-by-step instructions that demystify how computational processes (like text cleaning, for example) regularize and abstract textual data into workable units. Lowering the barrier of entry to using digital tools, the toolkit is aimed at an audience of humanist scholars at the beginning of their technical training, and would be ideally used in classroom and collective-learning settings.
The workshop will be facilitated by myself, the toolkit’s creator, who has experience leading over 15 technical workshops for my institution’s “Graduate Center Digital Initiatives” (GCDI). The workshop consists of three parts that balance instruction, individual practice, and group work over the course of 90 minutes.
First, I will introduce the “queer distant reading” tool, which is a command-line application that walks its users through text analysis procedures inspired by Judith Butler’s theory of gender performativity. Here, workshop participants will experience firsthand how the process of iterating over text, which is central to text analysis tasks, draws from Butler’s formulation of gender as a series of repeated acts that destabilize binary structures of gender. Participants follow application prompts to compute word patterns associated with gender markers, “woman” and “man,” and then to visualize the results of this analysis, such as in a network graph of expanded gender markers. The current version of this tool contains a command line app that allows participants to compute word similarity to “man” and “woman” in their text of choice, with an additional option to visualize the results in a network graph (see appendix for an example of this interface and graph visualization). Moving forward, this work will be expanded to demarcate and scaffold the stages of cleaning, adding options to retain information like pronouns, and add further functionality for visualizing the presence of gender markers in text. This work is currently being developed under my GitHub username “gofilipa,” in the repository, “qdr.”
For the second part of the workshop, I will introduce the “queer text encoding” tool based on the Text Encoding Initiative (TEI). This tool offers an interactive and beginner-friendly TEI workflow for “marking up” homoerotic content in text. The website interface, which enables users to tag homoerotic elements, encourages them to think productively about the limitations of discrete labelling protocols and how this work engages with Queer of Color Critique on destabilizing hegemonic power structures. The current version of the “queer text encoding” tool consists of a JavaScript-based web application containing a pre-transcribed and encoded manuscript, a portion of Oscar Wilde’s The Picture of Dorian Gray, which Wilde edited to remove suggestions of homoeroticism. By panning over the deleted text, workshop participants will see the pre-set labels for the homoerotic content in this text, which falls into four general categories: “intimacy,” “beauty,” “passion,” and ‘fatality” (see appendix for an example of the tool’s current interface). As I continue to modify the application, I will expand the interactive component of the application to allow users to enter their own custom labels for the homerotic content. This work is currently being developed in my GitHub repository, “dorian_encoded.”
For the final portion of the workshop, participants will have a chance to work in groups on a tool of their choice. I will scaffold this groupwork with prompts that encourage them to think about the pedagogical affordances for using their chosen tool. The wrap up the workshop, I will invite suggestions for further development and for publicizing the project.
