It’s that time of year, so I’ve gone ahead and posted my new syllabus for Introduction to Literary Text Mining. It’s still a work in progress and probably always will be. However I’m beginning to get a sense of the various contours/spaces of the field and the ways those can be taught to students.

The two major changes I made this year were to insist on learning R and remove the reading of a single novel. I now have weekly assignments where students learn how to implement pre-written scripts to perform different tasks (vector space models, topic modelling, network analysis, etc.). The aim is not only to familiarize them with R, but also to see how important “tuning” is — deciding on all those parameters that shape outcomes. I expect this to be very challenging (for instructor and student alike).

The decision to jettison the reading of a novel during the semester was difficult. You can only do so much. And the move between large and small scale just didn’t really work all that well. It was useful to have a textual example when we discussed certain problems (what is style or narrative point of view). But for the most part we just emoted away about Virginia Woolf and then switched back over to critiquing large-scale studies. I really want to figure out a way to get students as excited about dissecting texts at large scale as they get doing it line by line.

I’m toying with the idea of turning our Wednesday class more into a lab, where we work in groups to play with R scripts. I keep hearing that provostial refrain “active learning” in the back of my head and I am deeply dissatisfied with lecturing. But we also need to be candid about the unevenness of implementing active learning. It doesn’t just happen.

Anyway, my dream classroom is one where we work on a variety of projects collectively and share our findings along the way. That to me would feel special. We’ll see if we can get there.