I am almost leaving after few intense days at the Second Moscow-Tartu Late Summer School on Digital Humanities, where I was invited to give a talk about my research on the cultural dynamics of emotion in fiction (see two papers here and here), and more recently in song lyrics (no paper yet – a blog post documenting the very beginning of the research is here).
Besides showing a picture of me as the cool researcher that is not scared to show swear words in his presentation, I decided to write a short blog post with the few thoughts that a quick encounter with a different, but related (to cultural evolution), discipline stimulated.
First of all, I had very good feedbacks about my talk, probably the kind I would have not had in a more social- or biological- oriented conference. Briefly, the main result I presented concerns a steady decrease in the amount of emotion-related words in English language fiction and song lyrics, a decrease entirely due to a decrease in positive emotions, while negative emotions remain stable (in fiction) or increase (in song lyrics). There were, of course, some expected charges of lacking nuance, but, in general, comments were constructive. One possibility is that negative words, especially dirty ones, are often used with a positive connotation. A sophisticated philological analysis of Nicki Minaj’s Anaconda, due to a conversation with Philip Gleissner, demonstrates, for example, how in the line:
This one is for my bitches with a fat ass in the fucking club
“bitches” and “f***ing” have a clear positive connotation, while, only two lines after, in:
Fuck the skinny bitches!
they have a negative one (notice I am being absolutely serious – the full text is here). This is indeed interesting, but it leaves open the problem of why this happens for negative words and not for positive ones, given the asymmetry I found in my data.
A related hypothesis implies a faster turnover for negative emotion words, somehow similar to Steven Pinker’s idea of euphemism treadmill. According to this, the increase in negative emotions words would be illusory and due to the fact that they are better represented in my list of contemporary words, while this would not happen for positive emotion words. Again, however, this does not seem to explain why the usage of positive emotions declines through time. Not surprisingly, given where I am, both explanations involve lexical or semantic changes (as opposed to cultural ones). Anyway, I have something to think about while travelling back home.
In general, it is comforting to see humanities people interested in computational methods and quantitative analysis (seeing literature graduate and undergraduate students using R/Python/LaTex is quite nice – also, for some reason, they are all virtuosi of tabbed browsing, but perhaps is a Russian thing…) but I noticed something that immediately reminded me of mainstream socio-cultural anthropology. Everybody is concerned with methodological problems, fancy visualisations, the right data, the right level of analysis, etc. and slightly less with what are the actual research questions. In my view, the latter should have priority and the methods, the kind of data, and so on, should come after or, at least, the two need to be related.
In the Birth of the cool paper, with Olivier Morin, for example, we decided to analyse a small data corpus (books that we selected from the Gutenberg project) not for a generic lack of depth of the Google Books corpus, but because we had a specific hypothesis (that the emotional decline could be explained by a demographic variable, i.e. authors getting older) we could not test with it (the hypothesis revealed to be incorrect). In sum, I have the impression that a way to advance the methodological debate in digital humanities would be to be clearer on what the research questions are, and, perhaps, in some cases, to be bolder, and actually formulating them!
Of course, my view is extremely limited and biased by my personal academic history. Not surprisingly, also, I believe that cultural evolution (ta-da!) could provide a theoretical framework in which some of these research questions could be formulated, and I look forward to more common projects. I know that at least the Tartu small group, which is likely to be the culprit of my invitation, shares this vision. Let’s see what will happen!