This week’s excercise and readings regarding data collection and topic setting is illuminating insofar as it demonstrates the similarities between the Social Sciences and the Humanities. The notion that the humanities is somehow less technical or evidence-based than the social sciences is a tenuous notion, especially in light of Jockers’ writings regarding scale and theme. Indeed, Jockers’ creation of stop words and his appraisal of certain tools for use in the DH, isn’t explicitly DH, it’s statistical as well.
In the statistical fields, what Jockers is doing would be called “coding” and his attempt to illuminate certain patterns already existing in texts is extremely valuable in uniting the digital and the humane – the humane and social scientific.
In terms of my own research, topic modeling the interviews of Günter Grass, as opposed to his novels, seems like it would be extremely fruitful. Grass speaks of his novels consisting of German “tapeworm-length sentences.” Indeed, many readers of Grass are either amused or frustrated (or both) by these long sentences. Yet, in his interviews, Grass is much more reserved, speaking in very short, even fragmented, sentences. By running a topic model on Grass’ sentence structure, a comparative study could perhaps quantitatively indicate this distance. And through interpreting the topics of these sentences, perhaps show why.
