on description

I've been thinking about description lately, in part because I'm currently writing a novel filled with description, but prompted mainly by reading several lazy and tiresome book reviews (no, I'm not going to cite them here) in which the reviewers, although they discuss apparently quite mundane (and, from the excerpts provided in the reviews, quite mundanely-written) books, spend much of their space praising the descriptive capabilities of the writers being reviewed while offering no sense of the criteria by which these reviewers judge "good" "description."

John Gardner, in his (in)famous treatise The Art of Fiction, considers it the task of the writer of realist fiction to

present, moment by moment, concrete images drawn from a careful observation of how people behave; and [s/he] must render the connections between moments, the exact gestures, facial expressions, or turns of speech that, within any given scene, move human beings from emotion to emotion, from one instant in time to the next.

Gardner, of course, termed this method the "vivid, continuous dream," in which the writer must above all convince the reader (through "proofs—in the form of closely observed details") while doing nothing to rupture the "dream" created by the fiction—i.e., that we are somehow experiencing what we read. "[O]ne of the chief mistakes a writer can make is to allow or force the reader's mind to be distracted, even momentarily, from the fictional dream," Gardner observes. Despite the naivete of this description, and Gardner's apparent conflation of "human beings" with "fictional characters" or "linguistic constructs," his instruction remains the essential template for so-called realist fiction. In fact, it's basically a restatement of Poe's comments on Hawthorne:

A skillful artist has constructed a tale. He has not fashioned his thoughts to accommodate his incidents, but having deliberately conceived a certain single effect to be wrought, he then invents such incidents, he then combines such events, and discusses them in such tone as may best serve him in establishing this preconceived effect. If his very first sentence tend not to the outbringing of this effect, then in his very first step has he committed a blunder. In the whole composition there should be no word written of which the tendency, direct or indirect, is not to the one pre-established design. And by such means, with such care and skill, a picture is at length painted which leaves in the mind of him who contemplates it with a kindred art, a sense of the fullest satisfaction.

William Gass, friend and frequent antagonist of Gardner's positions on fiction (particularly Gardner's ideas about morality and fiction), was asked, during a 1978 debate with Gardner, his thoughts about, as the moderator put it, "this vivid and continuous dream." Gass's reply:

It's rather imaginary. In music, let's say, the motion of the work comes from the performance. That's also true in the theater. So if there's an interruption, or your mind goes blank or someone rattles a bag, you miss something and that's too bad, it's lost. In reading fiction, however, the motion that moves the text comes from the reader. Now the writer can indicate or try to indicate how that motion should go and at what rate. But I don't think that anyone writes a book now supposing that the reader will sit down and read two hundred pages in a dream. He's going to, in fact, stop, brush a fly off his nose, go back to the first page, read it over, skip, look around for the juicy parts. The book is more like a building which you're trying to get someone to go through the way you want them to.*

Later in the exchange, Gardner claims

I think language exists to make a beautiful and powerful apparition. [Gass] thinks you can make pretty colored walls with it. That's unfair. But what I think is beautiful, he would think is not yet sufficiently ornate. The difference is that my 707 will fly and his is too encrusted with gold to get off the ground.

Gass replies, "There is always that danger. But what I really want is to have it sit there solid as a rock and have everybody think it is flying."

It's easy to dismiss Gardner's metaphor of prose taking flight, and obvious to say that narrative, as Gass points out, is an act of illusion (or of ruining those illusions). I'll simply note that Gass's own descriptions are of course the sort that make readers pause, that rupture any dreams the fiction may encourage as they read, whether the language of the description is simple (I recall an old MFA cohort pointing out to me that "The Pedersen Kid" contained almost no words over two syllables) or ornate. Gardner's desire for an essentially passive reading experience hinges on the idea (again, from Poe) that description must serve a very specific singular function; Gass's observations about readerly agency suggest that any orchestration is a matter of a reader's docility in being led. (To be fair, most readers these days are, it seems, much happier taking the scripted tour than in exploring the "building" of a book themselves.)

But in returning to the question the lazy reviews prompted for me about description, and "good" description in particular, I have to agree with Gass; the lines of books that stay with me are always of the dream-rupturing variety, those over which I pause in my reading—to work out an image, to enjoy a rhythm or sound, to marvel at diction. Given the prose cited as impressive in the reviews, as well as my own experiences with editors and other readers, most people, even though they prefer the passive Gardner model of reading, still like some minor verbal thrills as they read; the question thus becomes one of degree, and taste, with a sort of middle-class (or Puritan, or WASP, or whatever social group you like) disdain of too much showing-off. And someone such as Updike or Bellow (i.e., someone telling a straightforward narrative with occasional linguistic flourishes) seems far more likely to be granted leniency in these matters than, e.g., Gass or Christine Schutt or Noy Holland, in all of whose work narrative is often anything but straightforward.


*(Gass elaborated on this architectural metaphor of reading in a 1979 interview with G.A.M. Jannsens: "...in reading I don't go continuously, I break off, I take up the book a week later, I go back and forth. I am like I would be when I went through a building: I am putting the pieces together to compose the building which exists ontologically all at the same time, and which I can only know experientially one at the time, and therefore I can only conceive or conceptualize the way it actually exists; I can have an idea of how this house exists.")