Recently, we had a talk about word clouds and what they were useful for and, if I’m honest, my personal conclusion was ‘not much’. I did, however, enjoy playing with the program by taking the text of novels from the Project Gutenberg1 site and putting it into the generator3 to see if the cloud matched my own ideas about the themes. I tweeted about this (it was a quiet evening) and, after being told off for calling it ‘hipstery rubbish’, was given some extra homework; an article by Paul Fyfe called ‘How to Not Read a Victorian Novel’4.
As well as making a good argument against reading lists for students (all arguments against this are good arguments), the article spoke about an experiment run by a professor in which students were asked to write an essay on a book they had not read. This part of the experiment does not seem all that revolutionary. However, as well as acknowledging that the students were not going to read these long Victorian novels, Fyfe gave the students instructions on how to gather knowledge about their given texts without reading them. The first method was to do exactly as I had been doing: downloading a plain text version of the novel from Project Gutenberg and making a word cloud.
After playing with data and going through various different forms of mucking about with internet tools to see what came up, the students had to stop and think about what they had done. This was the step I had been missing. As it happens, turning the books into ‘data’ to be manipulated in this way gave an entirely different and entirely useful insight into the architecture of the novels, both in terms of prevalent semantic themes and, by dividing up the plain text of the novel into sections, by allowing word choices over time to be compared.
The main thing I took from this was a reminder that academic curiosity can in fact be valuable in an exam culture, and that following your nose and playing with your food are two very valuable habits when it comes to studying literature (or anything else). I’d recommend to anyone to read the article, or at the very least to have a go at making some word clouds from novels.
1 http://www.gutenberg.org/ Projec t Gutenberg is dedicated to uploading books to the internet as soon as the works enter the public domain. They provide the usual ebook formats (mobi, epub) as well as pdf and plain text versions. 2
2 Footnotes like this are better than Harvard. I don’t need to explain why because it’s 2013.
3 The generator I used was Wordle: http://www.wordle.net/
4 This article can be found at: http://t.co/8tlK0RZTBP