By Jessica Kane, University of Michigan

This collaboration was part of the WWP’s Teaching Partners program; for more information, see the digital edition created by the students.

Introductory literature courses typically seek to help students understand why stories matter, as well as introduce them to some of the tools we use for textual analysis. Close reading, one of the foundational tools for analysis, requires careful attention to a text’s language, patterns, and gaps – something students often find challenging. A previous course’s success encoding Eliza Haywood’s novella “Fantomina” (1725) prompted me to experiment with using textual encoding as a way to practice close reading and analysis in an introductory setting, with mixed results. The Haywood encoding provided a model for how to incorporate encoding into a literature course; this new iteration suggests some revisions of that model.

With no previous experience in textual encoding, my Fall 2022 ENG 152 students gamely took on the work of encoding Harriet Jacobs’ 1861 memoir Incidents in the Life of a Slave Girl. The text is the first book published by a Black woman in the United States, and recounts Jacobs’ own experience of slavery and liberation in an attempt to convince white Northern women to fight for abolition.

As with the Haywood encoding, my primary goal was to force students to read repeatedly and slowly via the encoding process. Identifying and encoding textual elements like people, places, and dialogue requires reading each word fairly carefully. As a result of this work, for example, students noticed that Jacobs often uses common nouns and pronouns (“my grandmother”, “the doctor”, “her husband, “they”, etc.) to refer to people rather than proper names (“Martha”, “Dr. Flint”, etc.). Several students then wrote about the ways in which the language of the text tends to foreground relationships rather than individual identities, and often emphasizes Jacobs as the protagonist by describing many characters in relation to her.

Students also practiced analysis via identifying themes in the text – something many found difficult because the same sentence or passage seemed to refer to more than one of our chosen themes. I typically resolved this difficulty by encouraging students to focus on whatever theme seemed most important or prominent to them, which was a useful lesson in the often-layered nature of analysis.

The project thus broadly achieved its goals, and I am proud of the students’ work. We encoded half the novel, and presented the work at a showcase for the campus community.

Useful changes

This version of a classroom encoding project increased the difficulty level for me in a number of ways compared to the previous version. It was a larger class (22 vs 7), a more introductory-level class (100-level vs 200-level), a class with more first-year students, and a class with far more non-majors and students fulfilling a general education requirement. I made several changes to the process to account for these differences, and would make several others in the future.

The most important change was one of time. A younger, larger, and less-experienced class needed slower teaching and more in-class work time. As with the previous class, Sarah Connell virtually taught us the basics of textual encoding. I often needed to request a pause during those workshops to give the students time to practice a given skill. I also needed to walk around the classroom more and check in individually with students. I reserved two full class periods for encoding instruction, and set aside part of each class period to do in-class encoding for almost four weeks. In total, students had nearly 12 hours of class time to learn about encoding and encode their chapter.

A second useful change was increased structure. This included things like having the students stop encoding five minutes early, note that day’s work under <listChange>, and save their work. It also included repeating and re-teaching particular aspects of encoding, to remind students of what they’d learned the previous day.

A third useful change was that I determined a long-list of possible analytical elements for the class to collectively narrow down, rather than building that list from the ground-up with them. This was an appropriate change given their level, and meant we could have those <interp> categories ready to go.

Some lessons learned

I also ran into challenges I had not anticipated, and would do things somewhat differently in the future. These changes would help nearly any class, but are particularly important for introductory classes with younger and less-experienced students.

1. Reserve appropriate loaner laptops ahead of time. Many students either borrow laptops from the institution or use a tablet or Chromebook in class. Tablets and Chromebooks cannot run Oxygen, and the loaner laptops were often slow and frustrating to use. The institution has newer loaner laptops, but these are usually all signed out at the beginning of the semester; our project began mid-semester, and thus we only had the slower leftovers.
2. Teach what we sometimes think of as “basic” computer skills, particularly shortcuts. Many students did not know how to undo via Ctrl-Z (PC) or Command-Z (Apple), nor were they familiar with other Ctrl- or Command- shortcuts. As a result, I spent a lot of time going over these and other shortcuts (Ctrl-Shift-Y to wrap the text within Oxygen, for example) individually. Many students were also unfamiliar with the norms of coding, such that they could not follow an instruction like “copy line 26 including the angle brackets, then paste it below” until I showed them how to do it step-by-step. A partial day teaching computer use – including the need to explicitly save one’s work when using a program like Oxygen – would be helpful.
3. Provide checklists. Students often became overwhelmed with the number of different elements to encode. At this level, it may have been more effective to instruct them to go through the whole chapter looking for just one element, then go through it again looking for a second element, and so on. Regardless of method, they would have benefited from a checklist or other process document with the different elements, since we really only used ten elements and six interpretive annotation categories. A checklist could also include the various steps to actually working with the text, like how to open it in Oxygen, adding to the <listChange>, and saving.
4. Related to #3, I would also build in time for students to quality-check one another’s work. A second student, working off the same checklist, could help to catch the elements their classmate missed. Quality checking is a useful skill in its own right, the process would be yet another chance to practice (re)reading carefully, and it would make my own “final” quality check much easier.

A possible lesson learned

A final lesson I have been pondering is whether to compound the complexity of the project by also teaching the students something like XQuery. Some of the challenge of textual encoding, particularly for students who are simply fulfilling a gen ed requirement, comes from the lack of stakes. We talked about how the encoded Incidents chapters would be available for future scholars to use, but ultimately this project isolated textual encoding from its ultimate purpose, which is to conduct research or answer a question. I think it would be challenging to incorporate even very basic XQuery or a very basic research question into an introductory English class, but it would allow us to use textual encoding as a tool to actually do something rather than simply an exercise. Were I to attempt this route in the future, I would probably also dramatically pare down our reading list to something like two or three texts, to make space for this particular manner of practicing literary analysis.

As before, many many thanks to Sarah Connell and the Women Writers Project, who continue to say “yes” to me.