3 min read
Sometimes the NER worked really well, and sometimes not as much. I can imagine a range of applications for the tool that would be really compelling, and I’m really looking forward to thinking more deeply about the question of what is gained and what is lost by reducing — or maybe converting is a better term — a novel to a set of data. For me, using the NER on multiple texts either within the same genre or across genres to draw out particular distinctions or similarities would be really productive (obviously you’d need to clean the data pretty carefully if you wanted to draw real conclusions and advance theories about this), since I’m curious what you can tell using these kinds of tools/data about the ways in which texts adhere to, break with, or manipulate generic conventions. In general my experience with the NER was positive and made me feel a) like less of a failure when it comes to computer science after a disastrous CS21 experience, b) like I won’t fall behind in academia since I didn’t know what digital humanities actually consists of, and c) excited about the possibilities of doing exercises like these and seeing what the tools we’re going to use can tell us about novels/their history/the generic narratives associated with the texts.
W/r/t the actual exercise — I was particularly compelled by the NER’s failure when it came to cataloguing money in the book, since the first section of the novel is structured around/deeply concerned with money, status, power, etc. So, if someone who was just looking at the lists rather than reading the book was trying to understand that facet of it, it would be difficult. What relevant information does come through happened for me in the lists of names — there are lots of Biblical references, which aligns with how RC takes up Bible study pretty seriously during his island time — and the lists of dates and times, since these are basically what the novel has as a structure or plot, the passage of time, besides RC’s survival etc. His constant dating is reflected here.
In terms of what we can learn about lists from the book itself — beyond the NER we noted in class and briefly discussed that Crusoe is obsessed with cataloging and listing his belongings and that this might be sort of the framework for his proto-psychology, etc. I would love to explore more why exactly his lists and the tale of his survival is so compelling and creates narrative momentum — is it just the satisfaction of repeated problem-raising followed by closure/solution with material goods?
TL, DR: Great tool with hopefully big applications to generic questions I’m interested in; sometimes works, sometimes doesn’t, but is an exciting new way of converting the book; RC obsessed with cataloguing and listing and it’s somehow interesting, why?