Wednesday, May 27, 2009
Graphs, Maps, Trees and Breeding
I’m going to write my response to Jenny Davidson’s Breeding as a kind of post-script to our Moretti event of yore. (Which I’m supposed to be turning into an actual book if life ever stops getting in the way. Troubles, troubles.)
Anyway, let’s say the following is a post-script to this post I wrote for that event. Executive summary: I was pretty jazzed about Moretti. It seemed pretty good.
I’m also going to bounce off some things Jenny herself said at the time.
For about a year after I wrote that post it seemed to me literary scholars I bumped into, attended conferences with, or corresponded with, were buzzing about Moretti (quite independently of our humble little online get-together). He seemed to be enjoying a moment of academic celebrity. Fine by me. But it surprised me how much of the buzz was negative. I was struck, in particular, by one panel discussion I attended at which it was more or less agreed by various participants that scholarship and pedagogy of literary history are, at present, mutually ill-suited. (I am providing my own gloss on their agreement, if memory serves. But I’ll withhold names, in case memory does not serve. Maybe this is just me talking to myself here.) On the one hand, you need a set of texts that will provide you with sufficient evidence to pronounce intelligently—justifiably—on such subjects as ‘the nineteenth century American novel’. On the other hand, you need a set of texts to fill out a 12-week syllabus for an undergraduate course of that title. There isn’t any one set of texts that can do both jobs.
Of course it isn’t so surprising that the most sophisticated scholarship goes beyond what can be crammed into an undergraduate semester. But there is more to the point, it seems to me. There seems to be a tendency for good undergraduate pedagogy to recapitulate bad (as opposed to merely incomplete or preliminary) historiography. The teacher finds him or herself proceeding as if ‘the nineteenth century novel’ (pick your suitably broad subject) is the sort of thing that is at all likely to show up through the lens of, say, eight novels to be read. Reading a small number of novels and writing a few interpretive essays can be a fine and enriching way to spend a few months. But it’s not the same kind of enriching activity as studying the novel historically, with scholarly rigor. In a sense no one really thinks otherwise. So tension between pedagogy and historiography is not just tension between for-students simplification and for-scholars sophistication. It is tension between certain notions of value and certain standards of validity. You are asking people to use one method—close reading—to arrive at answers that ought require a different method—distant reading (to use Moretti’s term.) The value of literature, on the other hand, is substantially derived from close reading activity (broadly conceived—you can be more hermeneutic or erotic about these virtues of closeness, as you like.) But, again, validity of many sorts of scholarly claims demand more ‘distant reading’ methods.
So how do you proceed, if value and validity tug against each other, rather than going together, as one would have wished? (If all this is terribly wrong, and there is no tension, then please correct a poor philosophy professor who doesn’t know what goes on in English departments, apparently.)
Sitting in the audience, thinking my own way through it this, I was innocently waiting for someone to point out that Moretti might help. Yet three out of four of the scholars on this panel were very negative about Moretti when his name (inevitably) came up. They were almost (not quite) dismissive. The fourth seemed impressed mostly by the newness; seemed heartened that there was this a new thing some scholars were enthusiastic about. Yes, fine. But surely there is more to be enthusiastic in Moretti than enthusiasm itself.
Let me put the point another way. Rereading the contributions to our event, which I have recently done, I am struck by contrasts between respondents who were otherwise in basic agreement. I say Moretti is good because he is imitable, reproducible, scalable. Sean McCann approves of Moretti, but predicts he will be largely un-imitable. Tim Burke writes: “Of all the odd things I’ve heard in recent years, one of the oddest would be that there are objections in principle to the research paradigm that Franco Moretti describes in Graphs, Maps, Trees.” Last but not least, Jenny Davidson quotes Elif Batuman regarding Moretti’s “irresistible magnetism of the diabolical.” She says she is “at once seduced and rendered wary” by this distinctive quality, which she associates with the ‘heresy’ of quantification.
How is it possible that an author can seem, at once, moderate and agreeable to the point of utter, unassuming unobjectionablility, yet also ‘diabolical’, ‘heretical’?
If the answer is that literary scholars take the undesirability of quantification for granted, whereas everyone else takes its desirability for granted, the literary folks are flat out of luck. Everyone else is right. But I’m quite sure that’s not the whole story. (I hope it’s not even a significant part of it.)
I think these divergent responses are largely due to divergent ways of taking characteristic “grand flourishes” (Davidson’s term) Moretti makes on his own behalf. Davidson objects, rather mildly, that Moretti poses as a revolutionary, but seems to be reinventing the wheel. “As a scholar working in the field of eighteenth-century British literature and culture, I find Moretti’s work around these questions fruitful but its distinctiveness or originality somewhat overstated. You don’t have to be a heroic scientific pioneer and experimentalist to uncover the patterns in long-forgotten British novels of the eighteenth and nineteenth centuries.”
‘Distinctiveness or originality somewhat overstated.’ I would have said: surely highly distinctive even if the results are not so original. Moretti’s methods will mostly help by acting as an independent check on what we think we already know. And yet: that’s very exciting and new. Davidson continues: “Ruth Perry’s recent Novel Relations, for instance, reaps the rewards of a lifetime of reading eighteenth-century fiction to discern patterns (the rise of the novel of the second attachment, the trope of the cri de sang) that are as illuminating as Moretti’s graphs about literary and social history.” But the concern is obviously going to be that a lifetime spent reading such-and-such a body of material may, after all, reap the reward of a distorted view. Should I trust what Perry has to say or not? If some sort of independent, quantitative confirmation or disconfirmation were possible, that would be extremely welcome.
At the time of our event, I think I was more or less reading Moretti’s grand flourishes as just flourishes. I was enchanted by the humbler possibilities. Whereas Davidson was a bit more inclined to take them seriously, hence regard them as potential hubris. Rereading Moretti, I am inclined to split the difference. But this post is a supposed to be about Jenny Davidson’s Breeding: A Partial History of the Eighteenth Century. Let’s move on, then eventually I’ll circle back.
Davidson’s is a good book, I think. I learned from it. But I am quite sure it is exactly the sort of book that could use Morettiesque methods to complement it. There is a tension in Davidson’s methodology, and even more so in her characterizations of that methodology. She makes her own problematic gestures of distinctiveness and originality, you might say. When we see this, Moretti’s humble uses look even more useful. (Not that Davidson’s looks less useful. Complementarity is symmetrical. I’m trying to praise evenhandedly, not take backhanded swipes.)
Here is Davidson’s characterization of what is going on in her book: “I wanted, medium-like, to make these pages a sort of parliament, an auditorium in which the voices of actors in and commentators on the story of heredity in the eighteenth century can be heard” (7-8). And: “The great value of this kind of swerve away from the straight-and-narrow of the historical method—what makes it worth the risk—seems to me to lie in the counterfactual or path-not-taken traction it offers on ideas” (8). She takes as models W.G. Sebald and Roland Barthes and feels generally confirmed in her faith in, “the writerly approach, an essayistic or discursive mode that prefers not to participate in all of the disciplinary practices of history or criticism proper” (12). Yet, as she writes near the end of the book, “I offer these thoughts in a spirit diametrically opposed to the deliberate amateurism of David Denby’s Great Books: My Adventures With Homer, Rousseau, Woolf, and Other Indestructible Writers of the Western World. I want to keep all the intensity and the precision of academic writing, and the virtues of specialization, but to make what I write at least potentially open to readers in other disciplines, or in other walks of life” (198).
I think the tension here is fairly obvious: if you are deliberately refusing to fulfill disciplinary expectations; if you are giving up specialization for the serendipitous joys of a swerving, cross-disciplinary jaunt; if you are foregoing a sober, academic style for a more writerly, novelistic one; then whatever makes you better than David Denby can hardly be the fact that you are a properly disciplined, specialized, academic writer, whereas Denby is not. (I have not read Denby and so have opinion whatsoever about the specific case. I really am interested only in the general methodological question.)
And there is another problem. Davidson’s against-the-disciplinary-grain posture is not fully credible because, frankly, this posture is the disciplinary grain at the present time. This sort of post-Greenblattian-Auerbachian-Sebaldian-Barthesian, New Historicist, or post-New Historicist ‘touch of the real’ essayistic-academic style is, if not predominant, then at least familiar, very commonly met with. It’s normal. On the one hand, this makes perfect sense: people are attracted to this style for all the reasons Davidson says she is. So they write in this style. But, on the other hand, it can’t fully make sense for writing against the disciplinary grain to be the disciplinary grain.
‘Against the grain’ is contrarian wisdom; in touch with all those confining, facilitating, secure grooves of disciplinarity, specialization, so forth—yet leaping above and beyond all that. Losing touch yet keeping in touch. Inspired yet rigorous. Solidly argued yet boldly leaping. These are the contraries we all strive to synthesize. I feel the impulse myself, of course. It’s no mystery why people want this. But how do we seriously propose to get it? What’s the disciplinary formula for indulging in counter-disciplinary irregularities on a regular, disciplinary basis?
Let me put it another way: just as the plural of anecdote is not data, the plural of synecdoche is not historicism. (I don’t even know what the plural of synecdoche is, come to think of it.) The basic concern about the sort of historicism Davidson practices is that it is argument-by-synecdoche. We are given a sample of ‘voices’. How do we know the sample is valid? Here is the first paragraph from Moretti’s “Graphs”:
what a minimal fraction of the literary field we all work on: a canon of two hundred novels, for instance, sounds very large for nineteenth-century Britain (and is much larger than the current one), but is still less than one per cent of the novels that were actually published: twenty thousand, thirty, more, no one really knows—and close reading won’t help here, a novel a day every day of the year would take a century or so.
I know what Davidson will reply. Is my question: why isn’t your book 4,000 pages long? If so, then the answer is: because the publisher would never stand for it. And you wouldn’t read it if it were.
I just read Scott’s post, praising Davidson for really letting her chorus of sources speak as much as possible. That is, incorporating very generous chunks of primary material, which is very welcome. Like Scott I appreciated it. It is clear to me that writing this kind of thing is basically a good thing, and there is no way to write this kind of thing in a fundamentally different or better way than Davidson has done it. But this doesn’t solve the problem Moretti points out. There just isn’t any way to solve that problem with this sort of production - not without making it 4,000 pages long and unreadable (and unwritable as well, but who’s counting at this point.) Davidson is not taking a dusty old ‘These Are The Canonical Works’ approach. She’s got a much more idiosyncratic, interdisciplinary set of voices she is conducting in chorus. But the methodological problem is the same as with the dusty old Canon. It all seems very plausible, and certainly the details are fascinating. But how can I trust that the overall picture is really adequate? That sounds like a weirdly flat-footed question, with undue suspicion on top. It’s not that I don’t trust it, certainly not that I suspect Davidson of having some agenda or angle or bias. But what are we really doing here, methodologically?
Now it would be just as fair to flip the problem over and say: how are you going to quantify the intellectual history of ‘Breeding’, in a Morettiesque fashion. A nightmare of a project. But I’m sure some rigorous cuts could be taken at it. Some attempts to corroborate Davidson’s erudite appreciations and odd-angle bringings-together. (It’s not that I’m sure this case is in such particularly desperate need of it, but this is the case i am talking about as a case in point.)
I guess what I’m saying is that I would like there to be Moretti-type projects and Davidson-like projects. (So far so agreeable with most people, I guess. Although there is this weird hostility to Moretti in some circles.) And I would like there to be a bit more forthcomingness on the following score: Moretti-type stuff without Davidson-type stuff is, in a sense, pointless. Distant reading without ever any close-reading. No one likes that. (So much so obvious.) Less obvious is that Davidson-type stuff without Moretti-type stuff is, in a sense, undisciplined. Which is not to say: un-talented, unworthy, uninspired. Just that whatever good there is in it, intellectually, does not derive from participation in a disciplinary framework. (All that stuff I already quoted about the ‘writerly approach’ comes back at this point, you see.)
Have you noticed that I haven’t actually discussed what Davidson says about breeding? I’m just noodling at the meta-methodological level. (Stupid philosopher tricks.) As I said, I enjoyed the book and think I learned from it. But I am, in some sense, unqualified to criticize the specific historical picture that emerges, so it would seem to follow that I’m rather unqualified to praise it as well. I’ll leave that to others - Scott Kaufman, for example.
I don’t worry much about any critic’s claims of originality, since most of us spend so much time rereading (and re-teaching) a remarkably small number of canonical works, or incorporating other people’s criticisms into our own writing.
I do think you’re onto something about the tension between conventional literary pedagogy and literary history. One is essentially pre-packaged and -digested into a single semester, and the other could fill many lifetimes. Frankly, novel classes are the hardest to align with historical research, because they take up so much time for students to read and comprehend. But my conclusion has been that it’s better to read fewer novels, and have them spend more time doing their own research.
There seems to be a tendency for good undergraduate pedagogy to recapitulate bad (as opposed to merely incomplete or preliminary) historiography.
Conversely, at least in my experience, good historiography does not translate well to good pedagogy. At least not popular pedagogy. (cf. Sam Wineburg)
It is surely possible to look at a small representative sample and get a good picture, without having to check against a vast, exhaustive survey for validation. After you’ve read a certain number of books on a topic, you stop getting new points of view and start seeing views you’ve already seen before, repeated over and over again, with, at best, minor variations. At that point, you can surely be reasonably confident that you’ve sampled the field quite well. Can’t you?
I like your post, John! Very thought-provoking. Quick response: I think my disavowal of historicism is somewhat disingenuous, though that is not to say it’s not strongly felt - I was on a history search committee once where I was extraordinarily struck by the difference between the kinds of lines of questioning pursued by historians wanting to “spot-check” an argument and what your usual English department search committee would ask (i.e. “what was going on in France and Germany as opposed to Britain? What would the story be like if you told it in terms of political theory, sociology, culture?”, with a kind of comparative/"correction" bent).
I am also tempted to recount an anecdote that is mostly just revealing of my own personal preferences. I got asked a few years ago to write a “[Fill in the blank of academic publisher] Introduction/Companion to the Eighteenth-Century Novel.” I turned it down partly because I already had a couple books I was trying to finish, and it wasn’t a good time. It also didn’t really fit with my approach - I am almost always teaching cross-genre, so that it seemed to me counterintuitive to write a book with that sort of goal (introduce undergrads to the body of material) without putting, say, Richardson and Fielding in the context of Restoration comedy/Scriblerian satire etc. But my overriding reason was my sense that such a project should be properly Morettian, and that I could not face the notion of reading so many random prose fictions! And my recent sense that I WOULD be willing to reconsider doing such a project comes from my newer conviction that actually I would strongly prefer to write the non-Morettian version - that the interesting things I might have to say about the novel in the eighteenth century come much more clearly from, say, 12 “major” novels than from a well-proportioned account of everything. So that in the end, based on my interest in telling a story about _technical innovation in narration_, I would really be wanting to write something that has more in common with Genette’s narratology (which at least rhetorically relies on a relatively small canon of examples) than with either the Perry model (lifetime of reading) or the Moretti one (deliberate and rhetorically provocative adoption of “distant reading").
I have a lot of things I want to respond to here, so I apologize in advance for the long comment. The first thing is a meta-comment on the Moretti, which I’ve come to think about a little differently since the event. (Didn’t I participate in it? I remember writing something, but I don’t see it up there. Not important.)
Looked at a certain way, even though no single person does what Moretti did—not even Moretti himself, if memory serves—but the discipline does. What I mean is that I read hundreds of terrible turn-of-the-last-century novels and popular scientific texts and characterized them as X. (I’ll speak hypothetically for the moment.) Then I read some secondary literature, some of which incorrectly characterized a set of terrible turn-of-the-last-century novels that shared 50 percent of its titles with mine as Y, and some of which correctly characterized it as X. I added my voice to the devotees of X, who will, in the long run, be proven correct and have created a substantial body of secondary literature demonstrating their correctness. In other words, this:
Some attempts to corroborate Davidson’s erudite appreciations and odd-angle bringings-together.
Would happen first with the reviews, and second with works that cite, challenge, and build upon Breeding. It’s not Moretti-style systemic, but neither is it unrigorous. Moving on…
She’s got a much more idiosyncratic, interdisciplinary set of voices she is conducting in chorus.
The idea that you can analyze a work within a particular discipline becomes fuzzier the farther back you go. I’ve told the story, haven’t I, of William James’s class on Herbert Spencer? One semester it’s in the biology department, the next in the new political science department, the next in the new English department, the next in the new philosophy department, the next in the new school of medicine. I’m oversimplifying, but you see my point: outside of novels, it’s difficult to say whose discipline is the rightful owner of a given work, and because every cultured person was conversant with and the original audience of what we would now call scientific texts—e.g. Darwin’s Origin—calling the project of an English professor that analyzes the Origin interdisciplinary doesn’t make much sense. (Shorter: “Can’t we just call everyone ‘intellectual historians’ and be done with it?")
Davidson’s against-the-disciplinary-grain posture is not fully credible because, frankly, this posture is the disciplinary grain at the present time.
“Yes,” but also “No.” I don’t think it’s possible to write as Jenny did in an article, and I don’t see many—in fact, I can’t think of a single one—that even try to. Journal editors may allow for a single Greenblattian flourish in the first paragraph, but after that it’s back to business as usual, and that might not be an altogether bad thing: Jenny’s argument can’t be made (or made well) within the confines of an article. It needs space to develop of the sort only manuscripts can provide.
(More later. Students are actually showing up for office hours now that they have a paper due. Funny how that works, ain’t it?)
Hi Jenny, glad you liked my little response.
Scott: “The idea that you can analyze a work within a particular discipline becomes fuzzier the farther back you go.”
I think we are more inclined to see it as obvious that narrow disciplinary border-patrol is pointless, but we can have a bit of a blind spot for the fact that merely crossing the borders a lot is not automatically going to be better. This takes us back to the good old days of argument about ‘everything studies’. (I’ll have more to say later, I think.)
For me, the issue is not the kind of work Moretti is doing; it’s the specific way he’s doing it. I think you’ll find that a number of academics with a “negative” view of Moretti hold that view because of the way he argues. The best example is his work on Sherlock Holmes.
1. Why did the Holmes stories “survive” when so many did not?
2. It must be because of the clues.
3. But other writers also used clues.
4. So it must be because of the decodable clues.
5. But only 4 of the 60 Holmes stories have decodable clues!
6. So “why did Doyle’s touch desert him” at this crucial juncture? (notice that the question has changed! the right question would be: “so why did I bring up clues in the first place?")
7. Answer: he had to choose between decodable clues and a brilliant hero capable of solving non-decodable clues.
A more economical way to rewrite this argument would be as follows:
1. Why did the Holmes stories “survive” when so many did not?
2. Because Sherlock Holmes is a really great character.
This seems true, although alas perhaps not entirely earth-shattering (just ask the “Baker Street Irregulars,” who started meeting in the 1930s to talk about their favorite hero).
Notice that this more honest approach also allows authorial intention back in (one of the key claims in the article is that authors work by “trial and error,” flailing about blindly; but if that’s true, the question about Doyle’s “touch” makes no sense).
I guess I take seriously the “grand flourishes,” and I suspect a lot of other academics do too, which might account for at least some of the underwhelmedness out there. (A fantastic example: http://www.nplusonemag.com/adventures-man-science.) Given what damage literary scholarship did to itself in the 80s and 90s, we need to recapture the reputation we used to have for caring about whether what we say is true, before we went down the rabbit hole of charismatic work with uncertain argumentation behind it. There’s a part of Moretti—the part I like and admire—that deeply wants that too. But this isn’t, I suspect, the part that gets people excited about his work.
Yes, I raised that objection in earlier comments, and Christopher Prendergast raised it more formally in New Left Review. Unfortunately, Prendergast ill-advisedly used the word “gentleman” rather than “signature character,” to which Moretti responded that Holmes and other serial detectives are not literally “gentlemen.”
Of course the stereotypical hero of a classic puzzle mystery is a gentleman, in order to justify the detachment with which they sashay into and out of social traumas; Holmes (and some earlier and later heroes) instead achieve this dreamlike (reader-like) state through a sort of consumed-by-the-Muse pathology. In contrast, writers and readers who insist that detectives must suffer some level of emotional or moral scuffing from their experiences are notoriously less interested in clues that “play fair.” But since no commercially successful clue-driven series has existed without a more-or-less impervious identification figure, leaving that out of the research is like mapping the incidence of dysentery without considering water supplies.