Madame Bovary (3 page)

What he is trying to achieve in this book, instead, is a style that is clear and direct, economical and precise, and at the same time rhythmic, sonorous, musical, and “as smooth as marble” on the surface, with varied sentence structures and with imperceptible transitions from scene to scene and from psychological analysis to action.

Though he did not write poetry himself, Flaubert complains in a letter to Colet, “What a bitch of a thing prose is! It’s never finished; there’s always something to redo. Yet I think one can give it the consistency
of verse. A good sentence in prose should be like a good line in poetry,
unchangeable
, as rhythmic, as sonorous.”

Yet Proust, in the course of his vehement response, in 1920, to a negative article about Flaubert, commented (admiringly) on what he called Flaubert’s “grammatical singularities,” which, he said, expressed “a new vision”; our way of seeing external reality was radically changed by Flaubert’s “entirely new and personal use” of the past definite, the past indefinite, the present participle, certain pronouns, and certain prepositions. He went on to talk about other singularities: Flaubert’s unprecedented manner of using indirect discourse, his unconventional handling of the word “and”—omitting it where one would expect it and inserting it where one would normally not look for it—his emphatically “flat” use of verbs, and his deliberately heavy placement of adverbs. But it was Flaubert’s innovative use of the imperfect tense that most impressed Proust: “This [use
of the] imperfect, so new in literature,” he said, “completely changes the aspect of things and people.”

The
imparfait
, or imperfect, tense in French is the form of the past tense that expresses an ongoing or prevailing condition, or a repeated action. It is most usually conveyed in English by “would” or “used to.” Expressing a continuing state or action, and thereby signaling the continuity of time itself, it perfectly creates the effect Flaubert was seeking—what Nabokov describes as “the sense of repetition, of dreariness in Emma’s life.” Thus, early in her marriage, Charles’s (tiresomely predictable) habits are described using a string of verbs in the imperfect: “He would return home late. … Then he would ask for something to eat. … He would take off his frock coat. … He would tell her one by one all the people he had met … he would eat the remains of the beef hash with onions … then go off to bed, sleep on his back,
and snore.”

While the imperfect, as agent of “background” description and habitual activity, was traditionally, before Flaubert, subordinated to the simple past tense, used to narrate finite action, with Flaubert, the habitual and the ongoing are foregrounded, and the division between description and action is blurred, as is the division between past and present, creating a sustained immediacy in the story. Even the speeches of the characters are often reported indirectly in the imperfect (as, for instance, in the mayor’s wife’s comment quoted above: “
Madame Bovary was compromising herself
”),
allowing Flaubert to slip seamlessly into a character’s point of view without abandoning the detachment of the third-person narration. The narration remains dynamic despite the fact that a large proportion of the book, in Flaubert’s view at least, is exposition or preparation for
action.

In a letter to Colet of January 15, 1853—sixteen months into the book—Flaubert worries about the amount of “action” so far: “I have now lined up five chapters of my second part in which nothing happens.” An exaggeration, of course—but he felt there was going to be a great quantity of exposition, or prologue, and then very little unfolding action, before the conclusion. This, too, had not been done before—to tell a story with so little action. He believed that those proportions were true to life: “A blow lasts a minute but is anticipated for months—our passions are like volcanoes: always rumbling but only intermittently erupting.” Yet he worried that the demands of aesthetics required something different.

If Proust calls
A Sentimental Education
“a long report” in which the characters do not really take part in the action, Flaubert calls
Madame Bovary
a “biography,” one that takes the form of an extended analysis of one woman’s psychology. But he believed that it could, even so, have the pace of action: “It also seems to me not impossible to give psychological analysis the rapidity, clarity, passion of a purely dramatic narration. This has never been tried and would be beautiful.” It would seem, in fact, that this was just the sort of action that really interested Flaubert: the subtle shifts of feeling created in a reader by description and by psychological analysis. “I maintain that images are action,” he says. “It is harder to sustain a book’s interest by this means, but if one fails, it is the fault of style.”

Many of Flaubert’s transitions are indeed imperceptible, while others are abrupt; at still other points in the novel, the narration suddenly makes a rapid advance, covering months or years in a paragraph or two. But there is a tight unity to the novel as a whole, arising not only from its extreme economy—in which every element serves more than one function—but also from its recurring words, phrases, images, and actions. A small sampling would be: butterflies (actual and metaphorical, as in the passages quoted above); constructions in layers (Charles’s schoolboy cap, the wedding cake, Emma’s nesting coffins); Emma’s intermittent attraction to religious faith; Homais’s quoted writings; Charles “suffocating” with emotion twice near the end of the book; the same phrase—
bloquer
les interstices
—used first literally, to describe “filling
the gaps” between Emma’s body and the sides of the coffin, and then figuratively, during the awkward last conversation between Rodolphe and Charles.

Particularly prevalent are recurring images involving water, the sea, and boats. These include the “skiffs by moonlight” in Emma’s convent reading, the gondola in her daydream of a future life with Rodolphe, the actual “skiff by moonlight” in which Léon and she go to the island each evening of their three-day “honeymoon,” and the gondola-shaped bed in the hotel room where she and Léon thereafter meet every week.

Most striking, however, is the repeated image of a sealed vessel (twice a carriage, once a coffin) in the tossing waves of a troubled sea. It appears first in a pompous speech by the official who takes the podium at the opening of the agricultural fair, as he pays homage to the “king … who … guides the Chariot of State amid the unceasing perils of a stormy sea.” Then, in the famous consummation scene in Part III of the novel, in which Emma gives herself to Léon during the prolonged ride through the city, the king is replaced by the driver of the hackney cab, steering (rather carelessly) “a carriage with drawn blinds that kept appearing and reappearing, sealed tighter than a tomb and tossed about like a ship at sea.” Here Flaubert has taken the speechifier’s mixed metaphor and added the simile of the sealed tomb. He then brings back the comparison at the end of the novel, as Emma’s nesting coffins,
hammered and soldered, are borne to the cemetery: “the bier moved forward in little jolts, like a boat pitching with every wave.”

Such is the tight construction of the novel, and the utter conviction of the detailed descriptions and psychological portraits throughout, that we compliantly ignore, most of the time, any passing questions we may have either about inconsistencies in the plot or about implausibility in plot elements, the most conspicuous being that Charles never suspects any of Emma’s betrayals, never notices the sound of the sand striking the shutters as he and she sit reading, never receives an anonymous letter from a busybody. (And how does he, deeply in debt by now, pay for Emma’s three coffins?) If space and time as handled in the novel are both “elastic,” as has been said by some critics, so is plausibility. And yet this is not a distraction as we read—it is barely noticeable. The requirements of psychology take precedence over plausibility and consistency in time and space, and the psychology is entirely convincing.

The Earlier Drafts

Flaubert worked from successive plans, following them, revising them. He wrote numerous drafts of every passage, often rewriting and perfecting it before cutting it out altogether—at one point he estimated that he had 120 finished pages but to achieve them had written 500. (He revised by cutting, whereas Proust revised by expanding.) In rewriting he would watch out for poor assonances, bad repetitions of sounds and of words (especially
qui
and
que
, which he occasionally underlined and apologized for even in his letters)—Zola remarks that “often a single letter exasperated him.”

He did not burn these early drafts but left them for us to pore over—approximately 4,500 pages in all, residing in the Municipal Library in Rouen. They are available to us in clearly legible form—even online (at
http://www.bovary.fr
)—because they have been transcribed by volunteers, under the direction of Yvan Leclerc, at the Centre Flaubert de l’université de Rouen, who have reproduced every rejected scene, every false start, every cross-out. The drafts are an invaluable resource to scholars and, of course, to translators.

Flaubert’s intensive cutting meant that occasionally a sentence or phrase was omitted that left a passage slightly ambiguous or puzzling, or simply left room for a (mistaken) assumption. What was in the little bottles held by the ladies at the La Vaubyessard ball in their gloved hands? Not perfume, we see from an earlier draft, but vinegar—which is much more interesting, though of course if we are conscientious, we can’t insert that information into the text.

Elsewhere, other puzzles are solved: Why was Charles, when a student, stamping his foot on the wall of his room while he ate lunch? The answer is in an earlier draft: to warm himself. Why was the church hung with straw mats? To protect the parishioners from the cold. Again, we won’t insert more information into the text than is there in the original, but now we won’t jump to mistaken conclusions either.

A more extended example of how Flaubert rethought one moment of a scene may demonstrate the fascination of watching him at work, and that is Charles at Emma’s graveside. From a careful reading of the final, printed version of the French, we may suspect that Charles did not in fact take the aspergillum that was held out to him by Homais, even if it is easy to
assume he did. And if we test our suspicion by looking at the early drafts, we see these progressive changes in the transaction, as Flaubert envisaged it: (1) Lheureux (not Homais) passes the aspergillum to Charles, he drops it; (2) Lheureux offers it to him, he refuses it; (3) Homais holds it out to him, he “does not want it.” In the final draft, Flaubert cuts out any reaction to the aspergillum on Charles’s part: Homais simply holds it out to him, and Charles does not explicitly refuse it, nor does he explicitly take it, instead falling to his knees in the earth
and throwing the earth into the grave by the handfuls. If we simply trust the words of the original and translate them as “held out to him” rather than “passed to him,” we will get it right, but it would be easy to get it wrong.

A Note on the Translation

“A good sentence in prose,” says Flaubert, “should be like a good line in poetry,
unchangeable
, as rhythmic, as sonorous.” To achieve a translation that matches this high standard is difficult, perhaps impossible. Of course, a translation even of a less exacting stylist requires millions of tiny, detailed decisions; many reconsiderations; the testing of one word or phrase against another multiple times. In the case of
Madame Bovary
, there are unusually many previous translations—I count at least nineteen into English—and it is intriguing to observe how differently previous translators have made these decisions.

In the second draft of my translation, I looked at ten others, eventually an eleventh, the most recent. As I made extensive comparisons, trying to arrive at good solutions in meaning, vocabulary, and construction, I came to know five or six of them quite well. The great variety among the translations depends, of course, on two factors: how each translator handles expressive English and how liberally or narrowly each defines the task of the translator.

Curiously, in the case of a writer as famously fixated on his style as Flaubert was, many of the translations do not try to reproduce that style, but simply to tell this engrossing story in their own preferred manner. And so the reader in search of
Madame Bovary
has a wide choice: Gerald Hopkins’s 1948 version, with added material in almost every sentence; Francis Steegmuller’s nicely written, engaging version, smoother than Flaubert’s, with regular restructuring of the sentences and judicious omissions and
additions (1957); the stolidly literal, sometimes inaccurate version by the very first, Eleanor Marx Aveling (1886), which caused Nabokov much indignation in his marginal notations but to which he resorted in teaching the novel; that version as revised (not always happily) by Paul de Man (or, rumor has it, by his unacknowledged wife), who chose to omit the italics, for example. There is
Madame Bovary
with fewer of those pesky semicolons, with serial “and”s supplied, with additional metaphors. There is a version in which Charles is made to sob on the last page, another in which he is made to say “Poor thing!” when his first wife dies. There is even Flaubert complete with the involuntary repetitions that he so disliked.

Perhaps Flaubert was mistaken when he believed that the success of the book would depend entirely on its style—since various of his translators over the years have composed deeply affecting versions that do not reproduce it. Yet he would not listen, but was infuriated, when Zola remarked that there was more to the book, after all, than its style.

It should be noted that painstaking as Flaubert was about certain features of the prose, he was quite casual when it came to others, particularly pronoun reference and capitalization. Where ambiguous pronoun references are not utterly confusing, they have been retained as he wrote them. As for his inconsistency in capitalization (as in the frequent variation of “Square” and “square”), I have also chosen to retain it. This inconsistency was apparently not the result of an editorial oversight, since the original French text went through numerous editions by different hands in which it remained, surviving even into the most definitive 1971 Gothot-Mersch edition. Evidently, either Flaubert did not care, as Proust believed he simply did not care about certain pronoun references, or, perhaps more likely, he capitalized instinctively, unthinkingly. In any case, since it is part of the experience of the French reader, I have let it stand.