EXERCISES IN STYLE, originally published in France in 1947, and in this wonderful, now reissued translation in 1958, comprises 99 ways of telling the same story (which is not a "story" at all). Underneath this surface lies a profound investigation into the nature of fiction, an investigation which should make even the most committed devotees of the Novel of Ideas dismount, at least for a moment, from their hobby-horses. The investigation is presented as a series of unmasked questions, none of which is answered. The whole is carried off with the brio and dazzling intelligence for which Raymond Queneau is justly famous.

A narrator on a Paris bus sees a young man with a long neck, and wearing a felt hat, quarrel with another passenger, whom he accuses of jostling him. Seeing an empty seat, the young man sits down. Two hours later, the narrator again sees the young man in the street with a friend who is advising him to have another button sewn on his overcoat. That is the story. We then read it 98 more times, each retelling deploying a different style, from word games and permutations, through slang, jargon, and cant, to narrational attitudes toward the subject matter, modes of formal rhetoric, and other linguistic tactics. It is an enormous pleasure, a book devoid of pomposity and fake wisdom, and if one's attention but skates over its surface, the delight the work gives may be reward enough.

However, as I have noted, Queneau is amusing himself while he tacitly attacks cherished convictions as to the way fiction works; it is a benign assault, precisely because it is comic and "light." But the questions raised are very serious indeed, and even the recognition of them, sans an attempt to come up with their answers will, I suspect, bring up short anyone who thinks that fiction is an imitation of reality.

For instance, the story as outlined above is supposedly the "real" story, the theme, so to speak, on which Queneau plays his variations. Fine. But who says so? Not Queneau. His first telling of the story is titled "Notation." His 16th telling is titled "Narrative." Is "Notation" realer than "Narrative"? In the first exercise we read that "The chap in question gets annoyed with one of the men standing next to him. He accuses him of jostling him every time anyone goes past." In the 16th exercise, we are given: "This individual suddenly addressed the man standing next to him, accusing him of purposely treading on his toes every time any passengers got on or off." The data have changed. How can we now tell what the real story is? What reality is this fiction reproducing? If some omniscient "Queneau" is writing these two exercises, telling us what he saw, why has he decided to adjust his vision? Perhaps the first exercise is a lie. Perhaps both are lies. Did any of it happen? If it is all a lie, or if one or the other is a lie, does fiction tender us any "truth" at all? Maybe all facts are inventions. Is fiction, which supposedly has a responsibility to its audience, supposed to do this? Perhaps fiction should be experienced as one experiences a tree: there it is.

If this minor contradiction (and there are dozens more throughout the text) calls into doubt the putative didactic and mimetic function of the fiction-writer, how are we to approach, say, Oliver Twist? Dickens might be as much at odds with the reality of Victorian London as Queneau is with the reality of a Parisian bus ride. Does fiction's value lie only in itself?

In the third exercise, "Litotes," we read (and I give the exercise entire):

"Some of us were traveling together. A young man, who didn't look very intelligent, spoke to the man next to him for a few moments, then he went and sat down. Two hours later I met him again; he was with a friend and was talking about clothes."

This, let us say, happened. All right. It is here rendered as classic understatement. But the the technique of litotes works only if we know what is being understated -- or does it? Suppose that we read it divorced from what we know, or think, the story to be. Is it self-sufficient? It isn't much different from: "Someone must have been telling lies about Joseph K., for without having done anything wrong he was arrested one fine morning" -- the celebrated opening of The Trial. Or from: "None of them knew the color of the sky" -- the first sentence of Crane's "The Open Boat," a sentence that is not considered mysterious or obscure, even though it is absolutely empty of information, is, in fact, a textbook illustration of litotes. We are beginning to see that fiction drained of data functions as well as fiction packed with it.

Each contradiction in the book raises new questions: What is a narrator? Is there a difference between a first-person narrator and a third-person narrator? Does a narrative point of view falsify or intensify the "real"? Can a false point of view be truer to the "factual" than an honest one? If the narrator is lying, what happens to his subject matter? What if he lies only part of the time?

Exercises in Style lays to rest (or should) the quaint idea that fiction is composed of two equal parts: Form and Content. It also calls seriously into question the generally accepted contemporary dictum that "form is never more than an extension of content." What it posits, in a great, bravura performance, is the joyous heresy that will not go away, despite the recrudescence of such esthetic nonsense as Moral Responsibility, Great Themes, and Vast Issues as the business of fiction, and that heresy simply states: Form determines Content.