Humanistic Statistics and Beyond

I have been having a very interesting series of conversations with my graduate students about mathematics and statistics. In short, they are rather doggedly opposed to introducing any form of math into the work they do. In our discussions about Propp’s _Morphology of the Folktale_, for example, we looked at the pages of tabular material in the back of the book and I pointed out that it could easily be turned into a vector space model. Later, they seemed more amenable to discussing their own morphologies of various kinds of texts in a kind of pseudo-code of *if this … then that … else*, but when we turned to a discussion of statistics for inferring from their sample, they once again dug their heels in.

I want to be clear: I’m not faulting them. Our humanistic education has failed them. I could engage in some bombast here, but I think the really interesting thing is that nowhere is it written that we can’t use statistics to study literature, that we can’t attempt to quantify texts in interesting ways so that we can glimpse patterns perhaps not otherwise surfaced.

So I guess I am ready to have a conversation about what it is we should already know and what it is we should be making sure our undergraduate and graduate students know. There is, for example, no reason for me to teach basic statistics when that course is already available. Instead, I and my colleagues should be able to call upon a common, nay foundational, understanding of something like basic statistics as we explore ideas in upper division courses and graduate seminars.

Leave a Reply