The Three-Speed Problem in Digital Humanities Pedagogy

A Problem of Speeds

During graduate school I frequently rode to campus on an old three-speed bicycle to teach. The bike worked well enough except for one problem—its aging components would seize up when shifting gears at high speeds and fling the rider to the road. I attended school in mountainous central Virginia, so this mechanical issue meant I spent a lot of time on the ground. When I finally arrived on campus each day, bruised but ready to teach, I found my students dealing with similar issues of speed. No matter the topic, a familiar request was, “Could you slow down?” A student might feel bored one moment and challenged in the next, but at the same time, their neighbor might experience the same feelings in reverse. This is the crux of what I call the “three-speed problem”: for each student in the room, at some point, the class will be moving too quickly, too slowly, or at just the right speed. The stakes of speed in the classroom might appear less extreme than being tossed from a moving bicycle, but if our pedagogies do not accommodate them, they can result in very real feelings of isolation, shame, and breakdown for students.

Teaching never meets with a uniform audience. In no situation, in no field, will all students possess the same backgrounds, training, and ability to process new information. Few would argue with these statements, and yet, as I will discuss in this essay, the tensions they point toward remain unresolved problems for digital humanities (DH) pedagogy in particular. DH teaching often asks students to engage in complex, interdisciplinary skills and methods that might be far afield from their disciplinary training. Humanists, our typical students, can often feel frustrated in these circumstances as they lack familiarity with DH questions, methods, or even modes of instruction. At the same time, DH students with specific methodological training (in computer science, for example) might pick up new material much more rapidly and experience different challenges. Within a DH classroom, the instructor is always and inexorably going too quickly for some, too slowly for others, and at just the right speed for a few students. The ways in which an instructor responds to this tension strike at core questions of DH pedagogy: what we imagine it to be, who it is for, and how we can empower our students to engage in new methods.

A reflexive response to the three-speed issue might be to increase control, to assume that reception, difficulty, and skill levels are things instructors can regulate for our students. In some cases, this happens quite explicitly, when instructors attempt to narrow their audience by appending enrollment prerequisites to course or workshop descriptions. Doing so frames DH pedagogy as a question of skill acquisition, exacerbating assumptions about the degree of experience necessary to count as a digital humanist before students even get in the door. Furthermore, this attempt to signal and structure the course for a particular audience rarely works perfectly, especially since participants are often imperfect arbiters of their own backgrounds and frequently miss or misinterpret these caveats. In other words, while prerequisites will turn away some, they still result in a classroom with students of various skill levels.

If we cannot control the speed at which students take in our material, teachers of DH should embrace this diversity of experience as a fundamental part of DH instruction. After all, our responses to the three-speed problem can make it difficult for students to manage their inevitable frustration with the material and end up preserving accidental hierarchies of skill within the classroom. In short, pedagogical tensions around speed are entwined with questions of access and belonging in the classroom. In what follows, rather than attempt to exclude at the door, I suggest teachers of DH should embrace a more flexible and inclusive pedagogy that gives more control of the classroom over to students so that they can shape their own learning experience. To make this case, I propose tying together different theoretical strands from DH pedagogy to empower students in three areas: content, delivery, and outcomes. I will describe how low-tech approaches to DH instruction, in particular, can more capably promote student agency as a means of addressing inclusivity in our instruction. Rather than focusing on skills, I use this approach as a case study to argue for an approach to DH teaching that centers concepts and inquiry-based learning. No pedagogy can be all things to all people. But by recognizing and addressing the inevitability of difference in the classroom, we can begin to invite more people into the conversation.

Digital humanities classrooms are especially prone to student frustrations with course material and the manner of its delivery. The interdisciplinary nature of the subject means that students typically approach the material with more diverse backgrounds. Furthermore, the “digital” component of a DH subject might appear to imply a learning experience that would be quantitative and wholly logical, but in fact, the presence of technology can exacerbate the negative feelings frequently associated with difficult learning. I have found these feelings to be all too salient in my own courses on humanities programming: my students often struggle with increasingly palpable frustration while working with introductory computational logic that can appear hyperrational and childishly simplistic next to their complex humanities questions. In the next moment, though, the material might seem to hurtle forward and leave them behind.

The appearance of DH courses as sites of straightforward, uniform learning experiences is often exacerbated when students look at a course description to determine whether they are qualified for enrolling. While it might appear necessary to offer restrictive language about prerequisite skills and difficulty levels in course descriptions to signal the course experience to come, these markers are imperfect tools by which to structure a course. Paige Morgan has identified the subjective nature of qualifiers like easy or difficult when discussing digital tools, in particular, and whether a student has fulfilled the basic prerequisites for a course is similarly subjective and contextual: a student simply may not be able to determine whether they have fulfilled the basic requirements for enrolling.1 With a course sequence taught on a single campus, it might be easier to gauge whether students have completed a prerequisite course, but a great deal of digital humanities training occurs outside curricular contexts and without hard requirements for enrolling.2 Without such limitations, when a course suggests, to use a hypothetical but common example, “prior experience with programming is helpful but not required,” it is left to the students to parse their own background and experience and decide whether they will be able to work with the course.

While a reflective approach to failure is often a critical part of DH teaching, approaching the classroom without attention to the three-speed problem can risk a more debilitating and unproductive sense of failure for our students. Since, as I have argued elsewhere, frustration in DH courses is inevitable, a course that describes itself as requiring no or limited prior experience is likely to force students to look inward when they encounter difficulty.3 Allison Carr theorizes failure as a concept that can take isolated incidents and internalize them: “When I talk about failure as an affect-bearing concept, I’m talking about the myriad ways failure sticks to people and marks them as failure, metonymically remade in the image of their shortcomings.”4 As Carr notes, failure can become a part of a person’s identity, and failure in DH courses, if not handled properly, can lead students to believe they do not belong. Descriptions and prerequisites might help forestall some instances of complete mismatches between a student’s preparedness and course materials, but they can only do so much. The desire to offer an open, inclusive classroom free of the need for prior experience is one born of good intention. But DH instructors need to be careful that our introductory courses do not set students up for failure by suggesting that learning will be easy regardless of their backgrounds. Instructors need to match the promise of an inclusive classroom offered by these messages with an approach to teaching that can sustain them. We need a pedagogy that can respond reflexively to the needs of a diverse group of students of all levels and backgrounds. This orientation cannot be enacted at the door or in a course description. Instead, the most responsive pedagogy is one that allows the students to structure their own experience based on their needs, their intuitions, and their expertise (as is feasible in different classroom contexts and institutional structures).

Student Agency in Content, Delivery, and Outcomes

To return to my bicycle—the problem that frequently injured me was not the presence or absence of gears. Rather, the issue was that the rider could not shift between the different settings when they needed to. And therein, I think, lies one potential response to what I have identified as a three-speed problem in DH pedagogy. Rather than further attempting to control or direct the flow of a class, I would advocate for a pedagogy that gives controls over the class back to the students. This suggestion might appear abstract, however, so in what follows I offer specific examples drawn from low-tech approaches to DH teaching that emphasize student agency in three areas: content, delivery, and outcomes. Taken together, these approaches can shape a more inviting and inclusive method of teaching that allows students to self-navigate the difficult process of learning DH, to better manage the three-speed problem of DH pedagogy on their own terms.

Content is often the first element discussed when putting together any new workshop or course, but this material can be approached in ways that encourage or discourage student agency. DH pedagogy is often framed around platforms or tools as a means for exploring new ideas in the classroom. While platforms and tools are necessary elements of DH research and teaching, the ways in which they are introduced to students can sometimes overemphasize the importance of any one tool. John E. Russell and Merinda Kaye Hensley refer to this tendency as buttonology, “software training that surveys different features of an interface in an introductory manner.”5 Doing so might result in students very well versed in a single platform but ignorant of the underlying issues that it addresses. More specifically, Russell and Hensley suggest that pedagogy aim to move beyond buttonology and instead offer deeper experiences beyond the interface itself. In the context of my argument, an overemphasis on buttonology as a means for instruction necessarily maintains authority in the hands of the teacher: the instructor knows the platform and will share it with the students who are unaware of its workings. In Pedagogy of the Oppressed, Paulo Freire describes this sort of student–teacher relationship as the banking model of education, where the students are empty vessels ready to absorb the wisdom of the teacher only to later reproduce evidence of their learning.6 Students cannot take control of their own learning if the instructor remains the sole expert in the room. Instead, the three-speed problem persists as they are subject to whatever pace is being set for them.

Beyond buttonology, I would argue, lies a pedagogy that encourages an inclusive pedagogical environment by focusing on the underlying concepts behind an interface or a digital method. When called on to teach text analysis, for example, I often begin with pencil and paper as a means of offering an inclusive introduction to the concepts and methods underlying popular analytical tools. Rather than introducing Python packages as a means of explaining sentiment analysis in practice, I will first introduce the concept of quantifying sentiment: we could quantify the distinction between the words bad and good as −1 and +1. I then turn the participants loose in pairs to measure the sentiment of a few paragraphs using the same approach by hand. Doing so allows the students the freedom to intuit many of the underlying issues and complexities with the method before ever approaching Python. And wrapping up the lesson by introducing a Python package or tool for implementing the same activity using the computer can show the students where to go next. By displacing the actual means for implementing the method, the students are free to explore the activity at their own pace and control the speed of their own learning.7

Low-tech approaches to digital humanities research like this one create a more inclusive environment by unsettling expectations about how DH knowledge is delivered to students as well as assumptions about the expertise necessary to participate in a DH discussion. And, as is frequently the case with universal design, “if an environment is accessible, usable, convenient and a pleasure to use, everyone benefits.”8 Students of all backgrounds benefit from such an approach, but the instructors do as well. A teacher new to network analysis, for example, might be wholly unable to execute network analysis in Python. But they are, in many cases, capable of learning the concept and then designing a low-tech approach to it. For example, in the Scholars’ Lab, we ask our graduate student fellows to design speculative low-tech DH workshops for each other as a means of exploring DH pedagogy related to their own research interests. Chris Whitehead, a recent student, used the exercise to develop an introduction to digital text analysis using string and rope.9 The assignment asked Whitehead to learn about network analysis, but it also required him to step into the role of teacher and instruct his peers and other library staff in these methods. While a graduate student like Whitehead might be especially ready to take on this role as expert, grounding the terms of classroom engagement in low-tech activities offers new opportunities to increase student agency for a range of types of students. Workshops grounded in buttonology tend to rely on direct instruction: the instructor possesses information about the interface that the students do not, and the workshop will likely involve walking them through the interface. On the other hand, low-tech workshops like those I described earlier are more flexible in that they can more readily incorporate inquiry-based approaches to pedagogy. In other words, because they lower the technological stakes of the activities, the instructor is less called on to directly share information about particular digital methods or tools in the form of lecture, however interactive or hands-on that lecture might be. Instead, the students can develop their own sense of the concept by interrogating the digital method together at a speed they direct.10

Rather than think about DH teaching as an opportunity to share what we know, we could instead reframe it as a shared exploration with students. An inquiry-based approach like this addresses the three-speed problem, in part, because it replaces the singular forward momentum of direct instruction—you do not know something and now you do—with a more organic classroom experience that students can control themselves. To return to my pencil-and-paper text analysis workshop, I tell the students very little about the activity or the underlying idea of sentiment analysis beyond the beginning parameters: you can quantify the emotion in a text, and you could do so by assigning numbers to words that seem to have positive or negative sentiment. From there, their own exploration of the method guides the development of the lesson. I often frame their group activity by saying that they will inevitably discover problems with the activity. When they do, they should raise the question to the group for conversation: these problems are actual signs of good, humanistic thinking that we should discuss. Before long, a student might raise the point that words function in context—not individually. Very bad has a different polarity from bad. Or another student might intuit the need for common word banks of sentiment-laden words after noticing discrepancies in how different groups mark the same passage. The students ask questions of the activity together, and the ensuing discussion gets as far and as deep as I ever would want in a lecture. By freeing the students to find their own ways to the issues at hand, though, the activity allows each group to explore the concepts at their own pace in a more inclusive way. I have successfully taught the same workshop for high school students, undergraduates, PhD students, and faculty members. Each of these very different groups has proved able to handle difficult discussions regarding machine learning, computation, affect, and language. Because the activity is framed as a group discussion rather than a lecture, the pace and tone of the course can more easily match the needs and level of the students. They drive the conversation, and while they might not learn any Python in this particular workshop, the students will be better equipped to think critically when they do take the next step of implementation.

On an even more basic level, we can develop a more inclusive pedagogy for our students by changing our attitudes toward the outcomes of DH teaching. Increasingly, conversation in digital humanities pedagogy has emphasized productive reflection on failure as a means for engaging with new and challenging tools. Whether by encouraging “screwing around” or engaging with failure as an epistemology, these approaches ask us to rethink our approach to outcomes in digital humanities research and teaching.11 These interventions encourage us to think about how the process of doing DH, even if it results in null results or imperfect outcomes, can prove useful upon reflection. A reevaluation of outcomes in this context can help students see temporary failures not as shameful reflections on themselves but instead as a necessary and useful part of the learning process. Furthermore, low-tech approaches to teaching grounded in student-driven conversation rarely fail, because they are low stakes: the only real basis for success is that the conversation happen at all. If anxiety over success is at the core of students’ experience of the three-speed problem, shifting the terms of what achievement looks like can help to introduce a level of comfort back into a process that might otherwise feel like it is out of their control.

The three-speed problem is experienced from both sides of the classroom, just in different ways. When I traveled to campus each day on my three-speed bicycle, I thought about my teaching as much as my gears. At the same time that my students might have been worried about their ability to keep up, I was worried, as a teacher, about the consequences if too many students fell behind. For all the critical discussion of failure as it pertains to pedagogical outcomes, I think more work is to be done to reframe failure as an institutional and professional concern for the teachers themselves. After all, the drive to use prescriptive course descriptions to ward off the three-speed problem in DH pedagogy might be founded in instructors’ own fear of professional failure, in the fear of what might result from an unmanageable classroom with wildly divergent skill levels. If we reframe DH pedagogy to be about the exploration of a classroom experience as it unfolds in dialogue with students, we may be able to mitigate these fears and open up new pedagogical opportunities. Jentery Sayers has referred to his own work in the area of low-tech DH pedagogy as “Prototyping as Inquiry,” an idea that suggests that low-tech approaches to digital research and teaching “foreground processes of study.”12 This is as true for students as for the instructors. A low-tech approach to digital humanities teaching can counteract the three-speed problem by developing a space where students and teachers are equal collaborators in an emergent pedagogy in process. This focus on praxis can help to rediscover a space in which students have more control over the process and instructors have less fear about the outcomes. Instead of framing our DH classrooms around skill acquisition, we should focus on helping students to discover themselves as empowered practitioners and teachers of digital humanities. Instead of focusing on what we have to give students, we should focus on what we can learn together. The three-speed problem will not go away, but together we can retake a measure of control over it.

Notes

1. 1. Morgan, “Consequences.”

2. 2. While institutions like the Digital Humanities Summer Institute (DHSI) and Humanities Intensive Learning and Teaching (HILT) might offer suggestions for preliminary courses or course sequences, there are generally no hard requirements before a student can enroll in a course. DHSI offers some courses as foundational and others as potentially benefiting from a particular sequence (see Digital Humanities Summer Institute, “Course Offerings—DHSI”). For example, eleven of the 2020 DHSI’s fifty-two courses were marked as suitable for a first foray into digital humanities. At HILT, only one course (on natural language processing and machine learning) from the 2019 offerings mentioned any prerequisite experience as an expectation for the course (see Humanities Intensive Learning and Teaching, “Courses—HILT”).

3. 3. Walsh, “Frustration Is a Feature.”

4. 4. Carr, “In Support of Failure.”

5. 5. Russell and Hensley, “Beyond Buttonology,” 588.

6. 6. Freire, Pedagogy of the Oppressed.

7. 7. For more on this approach to teaching workshops, see Walsh, “Text Analysis Workshop.” For an entire course that employs this concept-based approach to text analysis, see Walsh and Horowitz, “Introduction to Text Analysis.”

8. 8. Center for Excellence in Universal Design, “What Is Universal Design.” For more on universal design as it could and should intersect with digital humanities research and teaching, see Williams, “Disability.”

9. 9. Whitehead, “String Theory.”

10. 10. I have successfully carried out versions of these activities with high schoolers, undergraduates, graduate students, and faculty members. While some of these learners might be more equipped to step in as expert discussants than others, I find that activities like these can nicely unsettle these distinctions. The discussions work best, even, with a mixed audience of various backgrounds.

11. 11. For “screwing around,” see Ramsay, “Hermeneutics of Screwing Around”; and Harris, “Play, Collaborate.” For failure as epistemology, see Croxall and Warnick, “Failure”; see also Graham, Failing Gloriously and Other Essays.

12. 12. Sayers, “Low-Tech Approaches.”

Bibliography

1. Carr, Allison. “In Support of Failure.” Composition Forum 27 (Spring 2013). https://compositionforum.com/issue/27/failure.php.
2. Center for Excellence in Universal Design. “What Is Universal Design.” https://universaldesign.ie/What-is-Universal-Design/.
3. Croxall, Brian, and Quinn Warnick. “Failure.” In Digital Pedagogy in the Humanities: Concepts, Models, and Experiments, edited by Rebecca Frost Davis, Matthew K. Gold, Katherine D. Harris, and Jentery Sayers. New York: Modern Language Association, 2020. https://digitalpedagogy.hcommons.org.
4. Digital Humanities Summer Institute. “Course Offerings—DHSI.” https://dhsi.org/course-offerings.
5. Freire, Paulo. Pedagogy of the Oppressed. Translated by Myra Bergman Ramos. New York: Continuum, 2000.
6. Graham, Shawn. Failing Gloriously and Other Essays. Grand Forks, N.Dak.: Digital Press at the University of North Dakota, 2019.
7. Harris, Katherine D. “Play, Collaborate, Break, Build, Share: ‘Screwing Around’ in Digital Pedagogy.” Polymath 3, no. 3 (Summer 2013): 1–26. https://doi.org/10.17613/h3fe-tz76.
8. Humanities Intensive Learning and Teaching. “Courses—HILT.” http://dhtraining.org/hilt/courses/.
9. Morgan, Paige C. “The Consequences of Framing Digital Humanities Tools as Easy to Use.” College and Undergraduate Libraries 25, no. 3 (August 2018): 211–31. https://doi.org/10.1080/10691316.2018.1480440.
10. Ramsay, Stephen. “The Hermeneutics of Screwing Around; or What You Do with a Million Books.” In Pastplay: Teaching and Learning History with Technology, edited by Kevin Kee, 111–20. Ann Arbor: University of Michigan Press, 2010. https://doi.org/10.3998/dh.12544152.0001.001.
11. Russell, John E., and Merinda Kaye Hensley. “Beyond Buttonology: Digital Humanities, Digital Pedagogy, and the ACRL Framework.” College and Research Libraries News 78, no. 11 (2017): 588–91, 600. https://doi.org/10.5860/crln.78.11.588.
12. Sayers, Jentery. “Low-Tech Approaches to Digital Research.” February 26, 2019. https://jentery.github.io/lowtech/.
13. Walsh, Brandon. “Frustration Is a Feature.” Pedagogy 19, no. 3 (October 2019): 519–24. https://doi.org/10.1215/15314200-7615485.
14. Walsh, Brandon. “Text Analysis Workshop: Four Ways to Read a Text.” walshbr (blog), September 21, 2016. https://walshbr.com/blog/ways-to-read/.
15. Walsh, Brandon, and Sarah Horowitz. “Introduction to Text Analysis: A Coursebook.” http://www.walshbr.com/textanalysiscoursebook.
16. Whitehead, Chris. “String Theory, or: Let’s Explore Social Networks with String!” Scholars’ Lab (blog), November 7, 2019. https://scholarslab.lib.virginia.edu/blog.
17. Williams, George. “Disability, Universal Design, and the Digital Humanities.” In Debates in the Digital Humanities, edited by Matthew K. Gold, 202–12. Minneapolis: University of Minnesota Press, 2012. https://dhdebates.gc.cuny.edu.