I’ve been listening to an audiobook version of Susanna Clarke’s Jonathan Strange and Mr. Norrell, an entertaining fantasy set in early nineteenth-century England, and enjoying how the book thinks about magic, in two different ways. First, simply the variety of the magical inventions; second, the way it integrates the essentially unbelievable aspects of the story—the magic—into what otherwise presents itself as a realistic situation, complete with pseudo-scholarly footnotes about the magical texts and personalities. It’s historical fiction, with recognizable figures (e.g. the Duke of Wellington, King George) and situations (the war with France), but it blends these conventions with a matter-of-fact presumption of the existence of magic, so that the characters going about their daily lives in London or at war in Spain take it for granted that magic and magicians are real. The book pulls this off quite well, and part of the pleasure of reading is the amusement in recognizing how it renders such implausible material as if it were completely ordinary.
It struck me this morning, as I was chuckling over it again, that in effect what I’m admiring is how the book sets itself a certain arbitrary constraint—presume the existence of magic—and then finds ways to accommodate that constraint while telling a persuasive story. At a certain level of abstraction, this is formally the same as a novel written with the rule that it must not contain a single instance of the letter e, or a poem written according to some pre-ordained scheme of rhyme and rhythm. Of course, the constraint this novel observes is, like the sonnet, say, more familiar than choosing to omit a single letter, the result of generations of readers and writers agreeing to observe it as part of a recognizable set of conventions. I think what makes the arbitrariness stand out more in this case is the way this book mixes conventions: it’s not a typical fantasy novel, replete with the faux-medieval social structure, norms, and artificially inflated language, for example, that characterize so much of that genre. Instead, it mostly follows the conventions of historical fiction and then inserts one element (magic) from that other genre; as a result, the conventionality itself is highlighted. In reading books which observe the conventions of a single genre more completely, if they are successful, we tend to ignore the artificiality of the conventions, accepting them as the frame of the picture, and slip into the world they create wholeheartedly. Of course any experienced reader is aware of them on some level, and will take note of them, even admiring the skill with which they are handled—but the conventionality as such is not part of the foreground, as it were. Here, it is.
But while it’s more noticeable in a book like this, it’s true of any piece of writing which observes the conventions of a genre—which is to say, of any piece of writing. This example, as I say, highlights the arbitrariness of the convention, the idea that it’s an artificial constraint imposed on the work for no particular reason, but which then governs its construction and reception. That’s what made it stand out and what appealed to me. The example of OULIPO and other similar experiments pushes our recognition of that arbitrariness to an extreme, casting an interesting retrospective light on the artificial conventions or constraints that have evolved more gradually, with less conscious control, through the interactions of whole generations of readers and writers. (I owe this understanding to Charles Hartman, who discusses and works with the idea in a variety of contexts in several works, including Free Verse: An Essay on Prosody, Jazz Text: Voice and Improvisation in Poetry, Jazz, and Song, his poem Tambourine, and elsewhere.)
The other thing—if it is another thing—that appealed to me was thinking about the fantasy genre, in particular, in this way. OULIPO and the whole meta-genre, so to speak, of “constraint-driven writing” are conventionally(!), in my mind at least, associated with a certain sort of intellectualism, a very cerebral approach to reading and writing, to aesthetics as a whole. Part of the pleasure and play of this kind of work derives from an ability to step back from the reality-effects of any given genre, to perceive them as artificial objects, to experience a kind of secondary enjoyment in recognizing how they manipulate us, rather than in the first-order experience of simply being manipulated. Fantasy, on the other hand, is—again, conventionally, again, in my mind at least—associated with a certain romanticism in one’s approach to reading and writing; a desire for escape, obviously, which is at odds with the impulse to analyze and play with the means of escape; and, as a corollary, with an acceptance of genre conventions as given. So it was interesting to think about the received literary convention of “magic” as essentially the same as an arbitrary constraint like, say, Calvino’s decision to write a book that consists entirely of first chapters.
Of course, the Calvino example highlights some of the problems with the rather Procrustean dichotomy I’ve sketched out here, since a number of his books include elements of both that OULIPO-style of constraint-driven writing (e.g. the chapter structure of Invisible Cities) and of something that, while not fantasy of the Tolkien or George Martin variety, has some kinship with it. Certainly the self-conscious exoticism or Orientalism of Invisible Cities is recognizable to a reader who, like me, had absorbed a great deal of that stuff in the fantasy genre, even while the magical realism of its descriptions hints at fantasy of a different type, or from a different direction. And of course Cosmicomics and t-zero draw heavily on the conventions of science fiction, while playfully subverting them. But Calvino is so obviously playing, OULIPO-fashion, with genre. Again, to return to the spark that flared this morning as I was listening to Jonathan Strange…, what appealed was the unconventional perspective on more conventional genre writing as also, if somewhat less self-consciously, structuring itself by means of an essentially arbitrary constraint.
Now that I’ve got this far—in fact, before I’d gotten this far—I’ve become acutely aware of the difficulties and complexities of the topic. For one thing, genres are always in a state of flux, always in the midst of a historical development, and readers and writers are constantly adjusting them, playing with them, questioning and critiquing them, putting them to new uses, and so on. To quote a lovely passage from Leslie Marmon Silko’s Ceremony, “But long ago when people were given these ceremonies, the changing began, if only in the aging of the yellow gourd rattle or the shrinking of the skin around the eagle’s claw, if only in the different voices from generation to generation, singing the chants. You see, in many ways, the ceremonies have always been changing” (126, Penguin paperback edition). Thus, a writer we may come to think of as defining the genre—say, Raymond Chandler with the hard-boiled detective novel—does so precisely because their work transforms the genre, demonstrating new ways to deploy its conventions. (In the article I’ve linked above, Chandler is quoted as saying, “I obeyed the formula because I honestly liked it, but I was always trying to stretch it…”) Any sufficiently sensitive reading of any work will always include an awareness of its arbitrariness. The dichotomy I relied on, between writing that is self-consciously aware of its own conventions and works to make the reader aware of them as well, and writing that takes those conventions for granted and expects readers to—such a dichotomy is, of course, itself quite artificial and ultimately unsupportable.
This touches on all kinds of issues in literary theory which I am now going to avoid.
Another set of issues has to do with the epistemology of genres, so to speak. At one point I wanted to challenge my own assertion that conventions are always arbitrary or artificial, and so went looking for the most extreme counter-example I could think of. The two that occurred to me right off were the scientific report and the shopping list. I thought I would focus on the structure of scientific reports—highly conventional but also quite utilitarian, seeming to pose a challenge to the claim of arbitrariness. But as soon as I started thinking about them, it seemed to me that it would be trivially easy to demonstrate that their conventions are quite arbitrary: they vary by discipline, even by journal; it’s easy to imagine alternative structures that would be just as functional, and so on. Yet questions remain. True, they vary by discipline, but why? Aren’t the disciplinary details themselves driven by the demands of the subject matter? And then you get into questions about how academic disciplines slice up reality, and whether that is arbitrary or not. By now we’re deep into debates about the relation between reality and how we represent it to ourselves and each other. In short, it became impossible to avoid the idea that generic conventions necessarily involve truth-claims—hence my reference earlier to their “reality-effects.”
And I didn’t want to go there. I just wanted to note the pleased surprise I felt at recognizing a familiar idea in a different context, and unpack some of the associated ideas.
Thinking about all this has got me wanting to read Barthes again…