There’s a certain serendipity to being a graduate student. While I cannot strictly rely on this method of gathering information, I find that as I get more interested in a topic, it begins to turn up in all sorts of interesting places. Partially, this is the fruition of my own admittedly messy efforts at keeping track of people and institutions doing what is technically called “interesting stuff” and it doesn’t hurt that I inflict my interests on my friends and they, in return, send me articles that look promising.
However, sometimes you just wake up in the morning and there’s a random article on Brain Pickings entitled thus:
(For those of you unfamiliar with Maria Popova’s well-curated blog, Brain Pickings, I highly recommend checking it out. It is fast becoming my number one source for aforementioned interesting stuff on the internet)
This article struck a chord with me for two reasons. The first is that she agrees with something I said two days ago, which is that academics in the digital humanities in particular have the aesthetic sensibilities of a molting parrot dipped in green goo (I may be paraphrasing here). For those of you wondering where I said it, I will direct you to the joint project I am currently working on with two of my colleagues for a class about the Digital Humanities and new ways of knowing. We’re keeping track of our work here: Playful Visualization at Work/Working Visualization at Play and I highly encourage all of you to check it out if you’re interested in what I’m up to academically or if you want to boost my ego and make me feel less like I’m shouting into the abyss of the internet (which tends not to answer back, except with the occasional cat .gif).
Anyway, before I wandered off onto that self-promotional tangent, I was talking about ugly graphs and how this particular publication is guilty of crimes against the common color wheel. But that’s not the point. And here we come to the second reason this post resonated with me. If you can get past the utilitarian ugliness of the graphs and actually look at what they’re saying, then you find that the authors have done something quite interesting. Rather than looking at the books themselves, they had readers rate the characters from 202 different 19th century novels on several different scales and, using those scales, investigated how readers’ quantified responses to those traits reflected a global pattern in how we relate to these fictional people.
It’s kinda brilliant. It’s very evolutionary-psychology based, so you have terms like “mate selection” being thrown around instead of “marriage” but that’s what happens when the authors are all interested in creating an evolutionary psychology study of literature. But just the idea of looking at people’s responses as a way to gauge what’s going on with characters and recognizing that we now have the infrastructure and technology to do that in a meaningful (if ugly…sorry, I can’t help it) fashion is awesome.
And then they get to the conclusion.
The emotional responses of readers to characters mirror the
adaptively conditioned responses that people make when
assessing other people as potential partners for social
Reading literature simulates the experience of emotionally
responsive social judgment.
That simulation has an adaptive value in cultivating the
emotionally charged social responses of readers, organizing
their social attitudes, and thus ultimately in regulating their
So let’s deal with this on a sentence by sentence basis.
First sentence: I’m assuming that the short PDF version just didn’t bother providing citations for the data on this and that the book fixes this oversight. To be fair, it fits in with what I hazily remember of my junior year class on (evolutionary approaches to) Human Sexuality so I’ll take it. I wish we had details for it, especially because I’m not so sure if the way we respond to antagonists is as reflective of the way we respond to the villains of the novel as we might wish.
Second sentence: Not…necessarily. There’s currently an argument as to whether, when we read, we judge independently of the character whose point of view it is or whether we are drawn into their consciousness so to speak. There’s good evidence for both sides (There was an article wandering around my facebook page earlier this week about this, let’s see if I can find it…aha! Losing Yourself in a Fictional Character Can Affect Your Real Life). So an equally compelling argument is that books train us in the act of emotionally responsive social judgement and we become trained to judge like out favorite characters. There’s a problem with that argument as well and I’ll get to that in sentence three.
Third sentence: In order for this conclusion to be true, I would need to see evidence that people who read 19th century novels are better able to regulate their social behavior. There’s no…actually, wait, there’s conflicting evidence on the subject. Some experiments show that readers are better in social situations, while others show that readers get their social fix from reading and perform worse in social situations because there’s less impetus to succeed.* If the simulation does help readers in social situations, it has adaptive value. If it doesn’t, then perhaps it has another form of value or it’s merely piggybacking on our adaptive need to socialize–empty relationship calories, to paraphrase Steven Pinker. Carroll, one of the authors, argues against this last position in his book Literary Darwinism, but his broad argument boils down to “there’s no way something this integral to our lives could be an accident.” There are instances where I think the data really points to certain elements of literary engagement having adaptive value, but I just don’t think there’s enough evidence to say so definitively, which is what they’re doing here.
Now why do I care? Especially since I have said (though possibly not here yet) that I don’t believe that analytic investigations of literature need to produce scientific conclusions. In fact I believe they shouldn’t; they should be allowed to roam free in as many directions as literary interpretations do. If this study was simply having fun with the conclusions and speculating about adaptive value and the psychology of reading, I would love it. And if it meandered into issues of classification and thinking about crowd-sourced literary analysis, I would love it even more. Perhaps the actual book does; I want to check it out anyway.
But this pdf version doesn’t. It takes on an aura of scientific validity and pretends to be science without backing up its claims the way science should. And the one thing I would hate to see literary analysis become is a pseudoscience.
*I apologize, but I cannot remember which meta-data study was responsible for this conclusion. I also can’t remember which science magazine linked to it. I’ll try to hunt it down later, but as this is a blog post rather than an academic article, it’s less urgent.