Thursday, October 20, 2005
Occasional Valve commenter Bill Benzon has a very interesting, long review essay in Entelechy (whose 'about' page informs me it is a journal of 'Darwin-touched ideas.') The books reviewed are: Graphs, Maps, Trees: Abstract Models for a Literary History, by Franco Moretti; and The Literary Animal : Evolution and the Nature of Narrative, eds. Jonathan Gottschall and David Sloan Wilson.
I'm interested in part because a while back Tim Burke discussed Moretti's project in a way that makes me want to know more - although I haven't taken the additional step of actually reading Moretti. (But if I start writing posts about him, eventually I'll have to.) Burke is impressed by the systematic quantification, and the paucity of work like this to date. "Moretti is perfectly right that if you take any given chronological slice in any given modern nation, our knowledge of the total range of what was published (just to stick to books) is actually strikingly absent, and I strongly suspect that there are many surprises to be found in a more systematic, quantified account of that range."
Benzon has high praise for Moretti, and sees in him the virtues Burke does. He aligns him with quite conventional (old-fashioned even) 'history of ideas' productions - he mentions Lovejoy's The Great Chain of Being; then explains how he goes beyond:
What Moretti adds to this standard practice is very simple, yet crucial. He draws tree diagrams to depict the structure of a lineage. However obvious this may seem, drawing trees is different from imagining them in your mind. Imaginary trees are vague and fuzzy. Drawings must be definite, and that definition forces you to think a little more precisely about your material. Beyond that, trees — and more generally, diagrams where points are connected by lines in any configuration — can be subjected to mathematical analysis; such mathematical objects are important in many branches of science, including biology and cognition. Thus Moretti is laying the empirical groundwork for more sophisticated analytic work.
Taken together these two chapters — “Graphs” and “Trees” — allow us to ask interesting questions and, as I have said above, the “Graphs” chapter may have identified a new phenomenon in literary evolution, the generation-long succession of genres. Moretti’s work is important because it is empirical and descriptive. In the specific context of this review, the work is important because it gives strong evidence of cultural processes at work. Moretti is not proposing a middle way between biology and culture as an abstract theoretical construct, as an object of intellectual desire. He is patiently mapping that territory.
Still, Benzon is approaching from a different angle than, say, Burke. Benzon has high hopes for "a vigorous naturalist literary criticism" - as he puts it near the end of his review. (If you want to know more about Bill, I asked him to write himself a little bio and I've tucked it into the tail-end of this post.) Yet naturalism, in Benzon's sense, is neither here nor there with regard to what interests Burke; at least potentially. (I'm not totally sure, but the possibility is what is important.)
Let me make the point philosophically, then more in terms of practical criticism. In the NY Times profile Burke was responding to, Moretti is quoted thusly: "My little dream is of a literary class that would look more like a lab than a Platonic academy." The irony here, if I may say so, is that Plato's academy is in fact the original home for this sort of labwork. (Recall Plato's project for his house astronomers: mathematics to express movements of the wanderers [planets]. No one had asked for that before, but it turned out to be a pretty fruitful research program. What Moretti is looking for are mathematical movements of a different class of wanderers - literary genres; unruly things, or so you might have thought.)
Maybe this is picky apologetics on Plato's behalf. But maybe it helps to see that there isn't any necessary link between championing empiricist naturalism and an impulse to systematic quantification. Or, maybe better, there is an ambiguity between: approaching in mathematically rigorous fashion, on the on hand; on the other, driving towards explanations that emphasize our status as animal - yet cultured - inhabitants of the natural world. (We may not deny this is what we are, yet think that's not the right level to be explaining at.) My impression, in fact, is that Burke's approval of Moretti's method is independent of any desire to push through cultural history to an improved understanding of biology/culture (though I don't suppose he'd mind if you handed it to him on a platter.)
Not all interest in the study of cultural history is an interest in cracking the crib of nature/nurture down whatever middle proves best. (What about Moretti? What does he want? I don't know.)
Here is a more practical critical line to the same point. Maybe. Not long ago I read a paper - Eric S. Rabkin, "Science Fiction and the Future of Criticism", PMLA (2004 119.3) - reporting rather curious results of a qualitative-quantitative collaboration between a mathematician, a population biologist and a literary critic.
"We have been reading a representative sample of American science fiction magazine short stories. In developing categories for the genres (or subgenres) into which they fell, we found it indispensable to distinguish between genre form and genre content. With fourteen genre forms and sixteen genre contents, we were able to code at least ninety-seven percent of all stories encountered" (p. 465-6). Interesting results emerge, in particular, along an axis of contrast between stories only printed once and stories reprinted more than twice. Some types of stories get published in bulk but relatively seldom republished; others are seldom published but, when they are, tend to get republished. "It seems as if, to get a science fiction story printed at all, one is best advised to write an alien contact-alien story [contact is the form; alien the content]; however, if one hopes to make a lasting contribution, one is best advised to write a dystopian satire. Why is that?" (p. 472). Actually, if I read the charts right, it looks like the best way to get republished is to write a philosophical tale of exploration. (I may not be reading right. There are bars that stop part way between 90 and 100, etc.; one has to eyeball the line between, e.g. 96 and 98.) It breaks down like so:
alien contact-alien stories total (97); reprinted (4) - rate approx. 4%
dystopian satire total (12); reprinted (4) - rate approx. 33%.
philosophical tale of exploration (18); reprinted (8) - rate approx. %44
[The data set consists of 1,959 stories published in American SF mags from 1926-2000.]
To quote Pee Wee Herman: "What the significance? I don't know!" But it warms the cockles of my Platonic heart to hear, not just that philosophical exploration is a good long term investment, but that there are patterns among the wanderers.
It is not obvious that anything about this data speaks on behalf of solutions to any biology/culture conundra, let alone forging any middle ways. (I'm not saying it obviously doesn't, just that it doesn't obviously.) If I had to guess? The explanation for the differences is that crap alien contact stories are often fun, but crap dystopian satire is a drag, and crap philosophical tales of exploration are nails across the chalkboard. And crap doesn't tend to be reprinted? All the same, I don't think the fact that bad alien contact is fun, whereas bad philosophy is intolerable, is exactly a cultural fact about America between 1926-2000; nor a function of what our ancestors were up to on the veldt. Don't get me wrong: it's not that I don't see how cultural/biological explanations could go: it could be that reprints tend to be selected by those with an interest in raising the profile of SF, cultural capital-wise. Perhaps philosophy and satire seem more high-toned? It could be that alien contact stories tend to have fighting in them; and the monkey in the reader likes fighting; and the monkey in the reader is rather easily satisfied with cheesy fight scenes. That would be an ev psych hypothesis as to why alien contact tends to enjoyable even when crappy (hence oft published but seldom republished.) All the same, I'm not sure a thoroughly old fashioned evaluative critic - an aesthete - wouldn't be able to make just as much out of this data. Suppose I am writing about what tends to make certain sorts of SF stories go well, and others badly. I might make something out of all this data, without making it anything about us being animals or creatures of culture. True, maybe I wouldn't manage to make much. Is it obvious rigorous naturalists, or theorists of culture, will be able to make more? (Make of it what you will.)
That was a bit of a digression. Back to Benzon. At the end of his review he writes:
I feel strongly that the Darwinians must moderate — if not altogether eliminate — their irritating attacks on postmodernist thought. It is not that I think things are fine in postanalytic–demodernist-psychoconstructionist theoryland. I do not. That is why I turned to the cognitive sciences years ago (e.g. Benzon, 1976; Benzon and Hays 1976). But sticking your tongue out and making hex signs — even the most sincere and earnest ones — is not helpful. This kind of activity, while a common feature of intellectual warfare, does little to win over the thinkers you oppose, who fully anticipate your magical gestures and are prepared with their own counter magic.
What is worse, this kind of warfare tempts you to lower your intellectual standards. Being different from Them is easy. Being deeper and truer is not.
I agree with that, probably. (If I knew more about this literature, I suspect I'd keep agreeing.)
I notice Frederick Crews wrote the intro to The Literary Animal. This reminds me that one of the mock-papers in his Postmodern Pooh is "Gene/Meme Covariation in Ashdown Forest: Pooh and the Consilience of Knowledge", by Renee Francis. She is also the author of "Rodent Nesting Hygiene in The Tale of Mrs. Tittlemouse" and "Monkey Business: Curious George and the Rhesus Mischief Reconciliation Routine". Her dissertation was "Pat the Bunny and the Opposable Hominid Thumb".
In her paper, Francis launches a vigorous, triumphant attack on Derrida and Foucault before settling into the far superior 'empirical' approach she favors. Ahem. (Her diagram is priceless. You really should see it.)
Well, you get the joke even without the funny diagram. But allow me to extend it.
The issue, as Benzon says, concerns the quest for consilience. As E.O. Wilson defines it: "Literally a jumping together of knowledge by the linking of facts and fact-based theory across disciplines to create a common groundwork for explanation." This is the happy middle way, the higher synthesis . Contributors to volumes with titles like The Literary Animal all want it, clearly. But, as Benzon says, wanting isn't having. The risk, per Crews' parody, is that you will find you - personally - can't spell consilience without the 'silly'. "Literally, a 'making silly' of facts and fact-based theory across disciplines by passing off groundless leaps of faith between them as explanations in terms of a common groundwork." [Insert fake Kierkegaard quote, uttered pseudonymously by Johannes de Consilientio.]
This is in fact a far more extreme accusation than Benzon levels against anything in The Literary Animal And it is the very opposite of what he says about Moretti. Nevertheless, let us pause to consider the possible worst case scenario, because Benzon's points orbit around its avoidance.
Why might it happen? The rigorously naturalistic explanation is given in A.R. Wallace's classic study, "Protective Memicry in Academics".
We now propose to deal with this part of the subject more fully, in order to explain what is meant by "protective memicry" - perhaps the most interesting and the most wonderful of all the phenomena of display among academics. It is only among the teeming forms of life of overspecialized departments that the best cases of memicry are to be met with ...
Memicry is the term applied to the phenomenon presented by certain academic views which, being themselves indemonstrable, and being associated with disciplines which are criticized by numerous enemies, obtain protection by their close resemblance to some of the shiny and technical species which are free from attack on account of their rigor and empirical solidity. In most cases it is not a general but a special resemblance which serves this purpose ...
As Benzon observes in his review:
The Literary Animal is tendentious in a way that Maps, Graphs, Trees is not. While both books announce a break from current critical practice, Wilson and Gottschall do so with klieg lights tracing designs in the sky and trumpets blaring triumphal fanfares. And they do so in the name of science, while Moretti never mentions science, yet some of his results are as solid as anything in The Literary Animal and arguably more important.
Did I mention I haven't read it (so how could I be accusing it of bad things?) And, again, Benzon's point isn't that The Literary Animal is bad, only flawed. He thinks there is a real risk here - one that merits scrupulous avoidance.
Seeking consilience is, you might say, close kin to 'doing Theory', from a certain point of view: the latter seeks a marriage of poetry and philosophy. The danger (which perfectly well may be avoided) is that the synthesis will not be higher but lower: pseudo-intellectual kitsch. Consilience, with regard to literature, seeks a marriage between hermeneutic methods and quantitative methods (you might say). The danger, again, is achieving lower synthesis: some sort of question-begging confirmation bias. Bad ad hominid arguments - just-so stories about how our Weltanschauung is our Veldtanschauung. (I trust we all know how evolutionary psychology can go bad.)
Benzon does not accuse the contributors to The Literary Animal of anything so dire. (They're just my favorite puns, so I can't make myself not make them.) But he does accuse them of two less low syntheses. One: failing in the leap from evolutionary psychology to literature, so that the results are of interest, perhaps, but not of literary interest. Two: settling for taking easy pot-shots at extreme 'social constructivist' positions that probably could have been defeated without a research budget.
Let me try to pull some threads together. Benzon writes:
Moretti has a nose of the empirical and I think he should be encouraged to do the work he finds most compelling and interesting. I would like to see at least half a dozen more studies like the one on genre succession in the British novel and 20 or so trait studies comparable to those on clues and on indirect discourse. Beyond this, if Moretti wants to collaborate with computer scientists in developing tools for this research — I am thinking, for example, that the tools of corpus linguistics would be useful in tracing trait genealogies through large bodies of texts — I would be happy to entertain proposals of that nature.
But also: "In any event, I have a taste for high-risk research that promises significant intellectual gains and am less interested in dotting i’s and crossing t’s."
Doesn't this run potentially against the grain of a taste for Moretti?
Isn't one of the points Benzon is making about Moretti that he isn't taking high-risks, i.e. going out on explanatory limbs? (Isn't this how he avoids any threat of confirmation bias, on behalf of pet speculations?) He's just being quantitative and systematic? And doesn't Benzon himself admit it isn't really clear what the significant intellectual gains are from these purely descriptive maps, graphs and trees? I connect this thought with the point I hint at, above. The point is: maybe Moretti is not really giving us a middle way in the sense of 'some new and better answer about biology/culture'. He's only giving us a middle way in that he is combining literature and mathematics without sinking into some morass of confirmation bias? Or maybe he's using methods that biologists have pioneered, but they won't therefore necessarily lead to biological explanations - i.e. explanations in terms of our natures as literary animals, nor answers in terms of middle ways between biology/culture? (I haven't read Moretti. I don't know. But I'm curious, so I'm asking.) I do appreciate that Benzon is pushing a research project, and there is certainly nothing wrong with hoping one middle way will turn into the other.
I like the stuff Benzon has to say about oral storytelling vs. the written word. I have a few philosophy of language bones to pick. But never mind that. I've said enough. And let me add: this post was written rather in haste. I reserve the right to edit at leisure if I've said something really stupid. Here is Bill Benzon's short autobiography:
I’ve been pursuing the relationships between the “newer psychologies” and literature for over thirty years. I did my undergraduate work at Johns Hopkins in the 1960s, where I studied literature with D. C. Allen, Don Howard, Hillis Miller, and above all, Earl Wasserman and Richard Macksey. I didn’t attend the ‘66 Structuralism conference, but was on campus at the time and got the flavor in various courses.
I also studied developmental psychology with Mary Ainsworth. She introduced me to the work of John Bowlby and to the research literature in primate ethology. I learned about Chomsky from James Deese and eventually found my way to the stratificational grammar of Sydney Lamb.
My interest in neuroscience started, I believe, with Arthur Koestler's The Ghost in the Machine, where I learned of MacLean's notion of the triune brain. Eric Lenneberg's The Biology of Language was a strong influence, especially his discussion of the importance of rhythm in organizing articulatory activity. And then Karl Pribram published an article on vision in Scientific American in 1969. That introduced my to his notion of neural holography, which has remained with me to this day -- and which is more or less continuous with Walter Freeman's neuro-dynamical theories (Freeman studied with Pribram). I saw a rough analogy between neural holography and something Levi-Strauss called the totemic operator (in The Savage Mind) and that sealed it. I had to learn more about the brain.
I did my graduate work in English at Buffalo from 1973-1978 where I studied with David Hays in the linguistics department. Hays had been one of the pioneers of computational linguistics, opposed Chomskyian linguistics, and did cross-cultural work as well. For all practical purposes I did my dissertation under him. It was entitled “Cognitive Science and Literary Theory.” I had an odd notion of cognitive science, however, placing it at the intersection of behavior, neuro-anatomy, computation, ontogeny, and phylogeny.
I’ve been at it ever since.
Did someone mention lower synthesis?
E.O. Wilson’s _Consilience_ is just wrong, unless I’ve misunderstood it, because he seems to think that sciences that study more complex, historically embedded objects can be derived from sciences studying less complex, ahistorical objects. It doesn’t work that way; no understanding of chemistry will let you understand what’s going on with the biological functions of DNA, no understanding of biology is going to let you understand the specifics of culture.
I’ve not read Wilson’s book, so I don’t know his full-dress account of consilience. Here’s a paragraph from his Forward to Literary Animal:
“Confusion is what we have now in the realm of literary criticism. The naturalistic
(“Darwinian”) literary critics have an unbeatable strategy to replace it.
They do not see the division between the great branches of learning—the natural
sciences on one side and humanities and humanistic social sciences on the
other—as a fault line between two kinds of truth. They do not consider it a line at
all but rather a broad expanse of mostly undiscovered phenomena awaiting cooperative
exploration by scholars from both sides. This conception has the enormous
advantage that it can be empirically proved to be either right or wrong or,
at worst, unsolvable.”
Later on he says:
“The mind is a narrative machine, guided unconsciously by the epigenetic
rules in creating scenarios and creating options. The narratives and artifacts that
prove most innately satisfying spread and become culture. The societies with the
most potent Darwinian innovations export them to other societies. In the process
of gene-culture evolution, genes affect which scenarios and memes are created,
and the cultures thereby generated affect which genes survive. The long-term
interaction of genes and culture appear to form a cycle, . . .”
It seems pretty bland to me. & no hint of the problems of reductionism.
As for Concilience, I don’t think E.O. Wilson ever said “that sciences that study more complex, historically embedded objects can be derived from sciences studying less complex, ahistorical objects”, but then I don’t have the book on hand to double check.
But as for DNA, I’m not sure what you mean about chemistry not allowing you to understand its biological function. Surely it does. If you mean that no one could have figured out the structure of DNA merely from the laws of chemistry, I see your point, but then that point is not Wilson’s.