Friday 19 February 2010

Literary Science

[Readers: My apologies for the lateness of this post. RM]

Although we take it somewhat for granted that our readers are interested in the science of fiction, this enterprise should probably not escape question. Why should we study literature and film with a method of observation borrowed from physics and chemistry? Can a science of fiction be anything but reductive? Do we cheapen our experience of art by placing it under a microscope? Or, more accurately, by placing tiny parts of it under a microscope and attempting to generalize to the whole. These are not trivial questions, and a number of valid, and even opposing, perspectives on this issue can be easily defended.

For myself, I believe in the scientific study of fiction because I believe in science. Basic science, specifically, or science for science's sake, with no necessary reference to some predicted application. In other words, I believe that we should be making careful observations of our world in order to better understand it, and that scientists have long ignored many aspects of our world that seem important: such as fiction.

Another reason why I believe in the science of fiction is that I believe in fiction. I believe it is a powerful force in the lives of many, if not all, of us. Moreover, I detect a troubling trend in our culture whereby science is becoming over-valued relative to art. Both are ways of shedding light on truth and neither should be privileged over the other. One way to restore a balance, I believe, is the scientific study of art. If fiction, as art, can be demonstrated to have important consequences that are measurable and reliable, our culture will begin to pay more attention to art and its role in our lives. This, at least, is my hope.

Jonathan Gottschall wrote a very interesting article on the necessity of a science of literary fiction for the Boston Globe, which appeared on the front page of the 'Ideas' section. It is a concise and compelling argument, but only one side of this debate. I would be interested to hear our readers' thoughts on this issue, including the many possible counterpoints to the points raised by myself and Dr. Gottschall.

Bookmark and Share

13 comments:

graywave said...

No counterpoint from me, I'm afraid. I read your blog because I wholeheartedly support the idea that we should understand ourselves and that the best tools for the job are those of science.

I have known people who argue that applying science can only "analyse all the beauty out" of a subject. It's not an argument I have any sympathy with. Milton's poem on his blindness will always be beautiful, no matter what we know about how our brains respond to it. The stars will always be beautiful no matter what astronomers tell us about them.

Knowing does not preclude feeling. Sometimes each enhances the other.

Jim Murdoch said...

I have an O-Level in Applied Mechanics. There are plenty of people who don’t and yet they drive cars. Do I drive better because I drew pictures of all the different kinds of internal combustion engine in a school jotter or because I know what the coefficient of friction is? Probably not but it’s nice to be aware that there’s a reason behind what I’m doing. I’m the same with writing. I know about sentence construction and figures of speech but I never think, Oh, I could slip a bit of onomatopoeia in there. I write naturally just as I drive naturally but I’d like to think that I don’t do either without thinking. There is both an art and a science to driving and the same applies to writing. They compliment each other. You might evens say that art is enjoying science.

Lt. Cccyxx said...

Very interesting - thanks for the link. I definitely think that scientific methodology can be useful in interpreting literature and our responses to it. As a (hard) scientist, I always wondered how literary scholars could so freely make pronouncements about the human condition without any training in science (especially psychology and the social sciences like anthropology, sociology, history).

Gottschall's piece was a little lacking for me because it presumed the reader understood the way literary scholars work/think now. I would also have liked to see more specific examples. Here's a question we have about literature, here's the scientific methodology we can use to answer that question in a new way, here's what happens when we apply it, here's how it is testable, etc.

One last observation is that the temptations of evolutionary psychology will be great in any scientific literary endeavor. But if everything becomes an adaptation, we've shifted paradigms but not gotten any more sophisticated than the current "blank slate" Gottschall talks about.

Nicholas Tam said...

The scientific study of fiction, in particular the cognition of fiction, need not rule out the existing streams of literary theory and interpretation.

Let's begin with an admittedly simplistic distinction. From a scientist's perspective, we want to make claims about fiction that are a) stable and b) causal. We want to find reliable patterns of change that emerge over large sample sets. With language change we've been able to extract a lot of those patterns into formal linguistics; with reader response we've been able to locate cross-cultural similarities in brain activity.

From a literary scholar's perspective, we are less interested in causal explanation than we are in representation and association - finding likenesses at an incredibly high level of abstraction. This doesn't tend to lead to stable claims, because it's trivially easy for the creative human mind to intentionally generate counterexamples. Authors subvert existing cultural/symbolic associations all the time and create new ones as they go; here there appears to be an unconstrainable element of free will.

It is not clear to me that the second set - that of individual choices by authors or readers, independent events that emerge from independent experiences, can be systematized into the kind of claims that an empirical methodology can extract. If we somehow find a way to precisely correlate neural activity with extremely high-level units of experiential abstraction, we might make a few steps. In the meantime, I do think literary scholarship (and scholarship in the humanities broadly) is producing valuable claims that add to our understanding, even if they are arguments rather than stable, cumulative truths.

I see two ways, then, to incorporate literary studies into a scientific framework. First, we can consider literary studies itself as part of our object evidence. The fact that scholars read fiction in the context of a theoretical schemata informed by other texts is interesting; the development of new theoretical schemata is also interesting.

Second, literary studies can proceed in the direction of non-causal scientific activity, like the classificatory systems developed by chemists and natural historians. Nowadays we don't think of these sciences as epistemically equivalent in strength to mathematized, systemic disciplines like physics and genetics: they seem to be ways of constructing an interface for us to engage with facts rather than locating facts themselves. I think this is acceptable, although it has largely fallen out of fashion in English departments as a consequence of the postmodern suspicion of categorical narratives.

As Gottschall correctly notes, the dominant paradigm among many literary scholars is a premise of total socio-historical constructivism and a wholesale rejection of biological determinism of any sort. Were they more scientifically informed their position would be less extreme. At the same time, I harbour very strong doubts that empirical methods will give us a complete picture of fiction until we have a better account of the symbol-manipulation that occurs in the mind.

Keith Oatley said...

Thank you very much for these extremely interesting comments, graywave, Jim, Lt. Cccyxx, and Nicholas. I agree with what you have all written. I think that for certain people in the field of literary studies to make claims while being contemptuous of the idea that there might be evidence for or against the claims takes only one step forward, when two would be better. The first step is conceptualization, but the second is scientific investigation.

One sign of reconciliation, perhaps, is that among literary theorists, the Russian Formalists such as Roman Jakobson and Victor Shklovsky are now well thought of; they were founders of a scientific approach to literature, which at the same time was fully committed to literature as such.

Bill Benzon said...

Nicholas Tam is very much on point to indicate the importance of non-causal scientific activity. Yes yes yes. He mentions classification, but even prior to classification we have description -- actually, the two go hand-in-hand. Charles Darwin, to give only one obvious example, benefited from several centuries of descriptive work by students of flora and fauna. Without that descriptive work, and the classification that followed from it, his work on relationships between species and, ultimately, their origins, would have been utterly impossible.

Students of literature, whether they are humanistically or scientifically inclined, do not have descriptive control over the texts that they study. Oh, sure, we know that sonnets have 14 lines and certain rhyme schemes, and that this text is a tragedy and that a comedy, and so forth. But that's just beginning. This lack of descriptive control hit me like a hurricane early in my career when I decided to examine "Kubla Khan." What I found is that this poem (that, apparently, was the product of an opium vision) had an elaborate and highly symmetrical structure that was utterly without notice in the literature. Some critical aspects of this structure emerged when I decided to treat commas, semi-colons, colons, and periods as delimiters functioning the way parentheses, brackets and braces function in mathematical expressions (or computer code). Once I decided to do this, structure just tumbled forth. It was easy.

But it took me a long time to make that decision. Why? Because doing such a thing wasn't part of the standard interpretive drills I'd been taught. Well, those standard interpretive drills are blind to lots of things. They are all but blind to the actual examination of formal features, though there no lack of talk about form. (I suppose the prior sentence is a little too strong, but it's too strong in the right direction, so I'll let it stand.)

If the profession had been on its toes, my findings would have been debated (I first published those results in 1985, with more detailed descriptions in 2003) and the most conservative of them would have been incorporated into "handbook" level material on "Kubla Khan." Instead, that descriptive work has all but been ignored. And, beyond the facts of authorship and publication, there is very little in the way of handbook level material on any text.

Continued in next comment.

Bill Benzon said...

Continued from previous comment.

Another example. Keith mentions the Russian Formalists. One of the things they pointed out is that the order in which events appear in a narrative often differs from the order in which they actually happened (in the fictive world). Stern's Tristram Shandy is a key text for this bit of descriptive observation. Given this distinction, between plot (order in narration) and story (order of happening), one could, for any text, lay out the story and show how it is mapped onto the plot. In fact, this is almost never done for any major text. As far as I'm concerned, it should be done; it's one aspect of gaining descriptive control over our materials.

Now, if this sounds tedious, yes, it is. If it also sounds pointless, that's a different matter. If you're greedy for the "meaning" of the text, then, yes, it's probably pointless. Alas, that's been the state of literary studies for well over half a century, greedy after the meanings of texts. But if you want to lay the foundations for deeper work on how texts function in the mind, then this tedious descriptive work is a necessary preliminary.

I say more on these issues in my essay on literary morphology, though I don't quite focus on description. Finally, I note that the work Gottschall is best know for, large-scale cross-cultural studies, rests on descriptive work, though he frames his work as theory-testing (which it is). Each of the texts in a study must be coded with respect to certain variables (e.g. the presence of romantic love). Such coding is a form of description.

Bill Benzon said...

Here's an essay by one David Michael entitled "Towards a Literary Science":

http://www.perplexicon.net/2010/02/towards-a-literary-science/

It too takes off from Gottschall.

Keith Oatley said...

Thank you very much, Bill, for these comments. Sorry to be a bit tardy in replying. I'm scrambling, here, to keep up. I've been thinking about your very interesting article on Literary morphology, and I am sure you are right that this kind of description of poetic texts is absolutely necessary. I have thought a lot about the sonnet form, and your analysis adds to what I know. In Shakespeare's Sonnet 129, the quatrain structure is certainly emphasized in the way you say. It seems to me that what this kind of structuring does is to enable juxtapositions to be made (in Jakobson's sense): as you say, in this sonnet, between desire, consummation, and shame. So that after them can come the generalizing couplet, that involves all human beings.

Keith Oatley said...

Thank you very much, Bill, for these comments. Sorry to be a bit tardy in replying. I'm scrambling, here, to keep up. I've been thinking about your very interesting article on Literary morphology, and I am sure you are right that this kind of description of poetic texts is absolutely necessary. I have thought a lot about the sonnet form, and your analysis adds to what I know. In Shakespeare's Sonnet 129, the quatrain structure is certainly emphasized in the way you say. It seems to me that what this kind of structuring does is to enable juxtapositions to be made (in Jakobson's sense): as you say, in this sonnet, between desire, consummation, and shame. So that after them can come the generalizing couplet, that involves all human beings.

Bill Benzon said...

No problem, Keith.

Specifically, on Sonnet 129, I've been thinking that the appropriate grouping might be:

((4 + 4) + 4) + 2

Now, just why I think that, well, that's certainly more than I can articulate here, and maybe more than I can articulate in any circumstances. The idea is "compression" of emotional energy. So, first we've got 8 bars (two quatrains), then "compress" the energy into 4 bars, and then compress and transform it into two, the concluding couplet.

The real question about the sonnet is, why? Why 14 lines, rather than, say, 16 ((4 + 4)(4 + 4)), or 12 (4 + 4 + 4)? That is, 14 lines doesn't parse directly into multiples of 2, or some simple combination of 2s and 3s. There must be some reason for the attractiveness of the form. & that new guess at the structure of 129 may provide a clue.

The larger point, however, is simply that this kind of question simply isn't on the table. The specifications of the sonnet are simply taken at face value when, in fact, they present a problem and, I suspect, a deep one. Ultimately the answer will probably involve the properties of the brain.

I would expect that the real benefit of serious morphological work will come when we have lots of cases worked out and can start comparing them.

Keith Oatley said...

Thank you very much, Bill, for your further thoughts on the sonnet form. I think your idea of structure as shared and as reflecting something about the brain is absolutely right, and that this has been neglected. And I agree, too, that there is an interesting question as to why this particular 14-line form is so resonant.

I too find the sonnet form fascinating, in part because of the close relation it enables between syntax and semantics. In its original form, invented, so I read, in the court of Frederick II, in Palermo, Sicily, some time between 1220 and 1250 BCE, there is, as you know, a turning point (between lines 8 and 9, that is to say between the octave and the sestet). You point out that with the Elizabethan sonnet there is another turning point between lines 12 and 13. So perhaps the basic, resonant, structure is: 8 + 4 + 2. This is not just a compression of emotional energy, though I agree it includes this, but a narrative structure that enables an exposition (long enough to be complete), a transformation (only half its length because the exposition has established essentials), and a conclusion (only half the length of the transformation because it becomes an abstraction or generalization).

Bill Benzon said...

...an exposition (long enough to be complete), a transformation (only half its length because the exposition has established essentials), and a conclusion (only half the length of the transformation because it becomes an abstraction or generalization).

I like that formulation, Keith.

In the case of 129, the first transformation glosses lust as madness (through the simile of the baited animal), while the second one has a rather different, and restorative, character.

Related Posts Plugin for WordPress, Blogger...