Tuesday, March 21, 2006
How vs. What
I would like to start off our discussion of How Novels Think with some considerations that might seem both painfully obvious and annoyingly obtuse, or at least off-topic. Instead of taking up the specific details of Armstrong’s argument, I would like to isolate one of its presumptions. Actually, something even more basic than a presumption, and something almost universally accepted in literary studies. It is my belief, though, that remembering basic facts, and pointing out what goes without saying, can help keep our pursuits oriented.
I am entirely unfamiliar with Nancy Armstrong’s work, but was quite attracted by the title of this book, as it raises the possibility that literary artifacts have agency. And Armstrong explores this possibility throughout. For example, she almost always attributes arguments to books, rarely to their authors.
Whether or not one is interested in such a pursuit could be a matter of taste. Those taken by the idea of Modernist impersonalism might find if more intriguing than those enamored of Augustan rationalism. But the pursuit need not be an expression of partisanship. It might be considered as a means for ascertaining what degree the medium of thought (novels, poems, etc.) determines the kind of thought. This experiment would be in theoretical formalism, not just describing the features of literary form, but attempting to ascertain form’s function.
My primary desire when I took up Armstrong’s book was to see what if offered in the way of theoretical formalism. & as with any desire, mine has distorting effects. For me these distortions began before I even opened the book. When I read Armstrong’s title, I immediately considered it as making a distinction: how novels think vs. what they think. This of course is not a fair approach to any work. There’s the typical respondent at a talk whose response to the paper amounts to a selfish complaint: “you haven’t made the argument I would have made if I were up at the podium.” & I expect Armstrong would be wary of the how vs. what distinction, if for no other reason than how the how so easily bleeds into the what, and vice versa. It can seem a practically impossible distinction to make.
Still, pursuing the distinction can be informative, similar to pursuing (even if per impossible) the possibility of textual agency. & I believe Armstrong’s book has an actual answer to the question, though a humble one. & the humility, I will suggest, albeit cursorily, does not reflect on Armstrong as much as it does on the novel itself, or at least its general reception by literary critics.
When I ask to distinguish between what a novel thinks and how it thinks, what kind of answer am I looking for? I am more familiar with asking this kind of question about poems, but unfortunately I have yet to develop any sophisticated answers. But here are some possibilities:
Poems sometimes think that rhymes signal the completion of thought. Other times they think rhyme signals the continuation of thought. Even other times poems don’t think in rhyme at all.
Similarly, poems think with meter, or against it, or without it.
Some poems think all thought is someone talking. Some poems think that thought can only begin when the poem leaves off talking.
To get beyond poetry, think of plays: the way plays think would seem to work through the things people say to each other, or to themselves. Plus some rudimentary gestures and actions. For example, they can stab each other.
Isolating these features works best as a heuristic device, as a guide for understanding particular artifacts. None of the distinctions can be taken too far. It’s not as if the different forms express incommensurable thoughts.
Turning back on the novel, something like an omniscient narrator would seem to be an example of how novels think: some novels thinks as if everything can be explained or described.
Narrators play a small role in How Novels Think. Armstrong looks at what she calls “the affective organization of the textual material itself” (63). This organization essentially correlates to the social. Which leads us to how novels think according to Armstrong: they imagine kinds of people and kinds of society. A novel is a model, a diorama with moving parts.
Novels think by imagining certain kinds of people in certain kinds of situations. For situations read societies. And realize that kinds of people and kinds of societies are mutually constitutive.
These models regulate social relations, working to accommodate certain kinds of people, censuring others. Novels make these accommodations either by altering the people, or society, or both.
Novels make arguments by showing the results of different ways of being, individual and social. These results often take the form of various fates meted out to various characters. The results can also take the form of the reader’s judgment of the character, a judgment often informed by the character’s own defense of their actions, but ultimately determined by the novel’s “affective organization of the textual material.” To compare it to a trial, the reader may be a jury, but the novel is the judge who determines what the jury will see and what they should think about it, more along the lines of the British court system.
How obvious. None of this is Armstrong’s argument. All of these are premises for her, ones that she can be confident that almost all of her readers will follow. She does go on to make controversial claims, ones that many might not follow, as to the constitutive nature of the novel and the competition between genres. As I have already confessed, I am not presenting the specifics of her presentation.
But I would like to reconsider the premises. What possibilities do we forestall when we consider the novel as the essence of literature, and when we consider novels as essentially models of society? Though some may balk at Armstrong’s suggestion that such modeling is constitutive of individuals, most still read for models. Consider the recent discussion of Zadie Smith’s On Beauty. And the much bemoaned vogue of cultural criticism in English departments. Novelistic modeling is not synonymous with narrative, much less with the literary, something even those among the moaning sometimes forget.
Like you I was attracted to the book by the interesting title. I have no problem at all attributing agency of a kind to a text, and I hoped (projectively, no doubt) that Armstrong would be doing something that might make me more confident that writing about what texts require of their readers was more than just an intuitive reflex gesture.
I think, though, that the book is more about the lines along which novels think, rather than about how exactly they carry off that feat in the first place, given that they don’t have minds. That’s what you say at one point here, isn’t it Lawrence? Perhaps form and content of a novelistic thought can’t be separated.
Thanks for your remarks about how poems think, by the way.
That was a singularly unclear comment. I meant to speak to the “how vs what” idea and to say I understand Armstrong to be concerned with the “how” seen as a mobile and elaborating train of thought, perhaps not identical with the “what” if that means a static snapshot of content.
Then there’s how advertisements think, how songs think, how spiritual advisories and medical and self-help and astrology books think, how capitalists think.... As an unflattering comparison point, I’m currently reading Michael McKeon’s compendious The Secret History of Domesticity, which is at least provincial rather than parochial.
But let’s restrict ourselves to the title at hand. There’s a sense in which an artwork is a collaboration between artists and the work, and it’s also sensible to consider it a collaboration between the audience and the work. But figuring an artwork as a thinking subject in itself seems problematic except as parable. A novel isn’t artificial intelligence; it’s ELIZA.
Armstrong’s book derives its novelistic algorithms from just a few parings of the accepted canon. Even so, a more accurate title would have been “How Novels Debate”. It doesn’t resemble a manual of Intel processor instructions. Instead, it turns out to be the latest in the line of “Battle of the Books” stories. And not one of the livelier examples—a universalizing authoritative voice doesn’t doesn’t improve the form.
I agree with Armstrong, though, that it can be clarifying to consider twentieth-century psychiatric texts as narrative literature by other means.