As in all branches of science, from subatomic physics to astronomy, order of magnitude can have great importance in literature. A consideration of size and scale in literature, moreover, soon provokes opposing claims. One might claim that all literature tends to the miniaturized condition of haiku and the maxim. And one might claim that all literature tends to the condition of the roman fleuve, the overflowing epic of everything. Even Aristotle was concerned about size and warned us in the Poetics that the dimensions of a work should not exceed our capacity to grasp its beginning, middle, and end. Many writers have produced unified works that may strike us as inordinately long: Lady Murasaki, the authors of the Roman de la rose, Tolstoy, Musil. But it is primarily in respect to Marcel Proust’s In Search of Lost Time that we work out our problems and our impulses concerning order of magnitude.

Until his late thirties, the Paris dilettante and apparent snob published mostly social gossip, short stories, and translations. Then, during a period of seemingly helpless hesitation between writing a novel and writing philosophical criticism, he grasped that he could do both by having the narrative lead up to the philosophy. In 1912, he sent out the first of two projected volumes. By the time the full novel appeared in 1927, five years after his death, it had grown to thirteen volumes. To read Proust in that paperbound, unannotated form requiring a papercutter, as was necessary for three decades, carried the challenge of exploring an unmapped continent without a reliable compass. In 1954 the French Pléiade collection brought out the full novel in three compact volumes with indexes and résumés and notes. Random House consolidated Scott Moncrieff’s translation into two large readable volumes in 1934. In these editions one can, with attentive reading, perceive the underlying shape of the novel sustaining its great length. In subsequent years, scholars produced a series of substantial biographies, from George Painter’s in two volumes to the most reliable and discerning of all, the 950-page opus of Jean-Yves Tadié. It is due out soon in English. Meanwhile Proust’s correspondence has appeared in twenty-one volumes.

When Proust’s novel fell into the public domain in 1987, three Paris publishing houses were ready with new editions that had been in preparation for several years. They all carry the same basic 3,000-page text with few variations. The differences lie in packaging and presentation. Laffont-Bouquins chose to publish three fat volumes prefaced by elaborate historical and biographical materials. Garnier-Flammarion produced ten pocket-sized volumes competently edited by Jean Milly. The new Pléiade edition, published by the original copyright holder, Gallimard, made the boldest, most ambitious, and most expensive bid to claim the market. In a combination of editorial, literary, and commercial decisions, Gallimard proposed to influence the way we read Proust and, to some degree, the way we approach all great literary works.

The difference is immediately visible and palpable. The Pléiade edition of A la recherche du temps perdu grew from three compact leather-bound volumes totaling 3,500 pages to four bulging volumes totaling over 7,000 pages. The added pages include expanded apparatus and, most prominently, newly introduced sketches, drafts, and variants in tiny print. These esquisses add up to some 1,250 pages, equal to almost half the length of the novel itself. Eight inches on the shelf instead of three. Is there someone somewhere who believes that this elephantiasis is appropriate to Proust’s novel? Three thousand pages are daunting enough. What past mistakes or what new discoveries explain this more than doubling of the product we are to pay for and hold in our hands? Other shorter, far less expensive, and fully satisfactory editions in French and English await one in the bookstores. Nevertheless, Proust scholars working with the original French have generally elected the new 7,000-plus-page Pléiade as the authoritative version to be cited.

The prime mover and defender of the faith in this context is Jean-Yves Tadié, author of the superb Marcel Proust: Biographie (1996) and of two earlier studies of Proust’s novel. His scrupulous scholarship and immense knowledge of Proust’s life and writings qualify Tadié to be the current dean of Proust studies throughout the world. When he began on the new Pléiade edition, he was powerfully drawn toward what is known as genetic criticism—the study of the evolution of a work out of earlier outlines and drafts and sketches into its (presumably) final state. In Europe and the United States today, genetic criticism has expanded rapidly and opened up lush new fields for scholars, particularly studies of major modern authors with well-furnished archives. Most of Proust’s working papers and drafts became available at the Paris Bibliothèque Nationale in the Seventies. They provided a scholarly bonanza. As head editor of the Pléiade volumes, Tadié persuaded himself, along with a crew of young scholars he had recruited and the responsible committees at Gallimard, that the new edition should contain extensive selections from Proust’s working drafts as well as enhanced editorial apparatus.


And thus it is Tadié who, at the end of his 100-page general introduction, provides the justification for this remarkable expansion. Traditionally, the highly respected and widely sold Pléiade collection (a complete library of several hundred titles representing world authors, classic to modern, and including its own encyclopedia and other reference works) has presented an author’s work or works in a readers’ edition along with a brief introduction and basic notes. A few titles (the second edition of Baudelaire and the Sartre volumes) have somewhat enlarged the apparatus. But Tadié’s Proust shifts the undertaking to a new order of magnitude, in which the actual text of the work occupies less space than the apparatus surrounding it. Many a miniature painting hangs in a frame much larger than the tiny picture it contains. We accede to this summons of our attention to something so concentrated. But Proust’s In Search of Lost Time cannot pass for a miniature. Like Monet’s waterlily series, it needs only a modest frame. Yet between the 1954 and the 1989 Pléiades, the editor has enlarged the frame from 500 to almost 5,000 pages. How are we to deal with this example of physical and intellectual gigantism?

Proust is an exceptional author. The intelligent general reader will need some guidance. Other new French editions, and increasingly those in English, provide that guidance without intrusiveness. Therefore, we must carefully examine the sheer dimensions and specific content of the Pléiade phenomenon for the implications it contains concerning the status of a literary work and the future of literary studies.

The sweeping argument Tadié builds in the last six pages of his general introduction turns in part on an ambiguity of the French word oeuvre as used in the context of literature and art. It can mean a single, clearly defined, unified work—sonnet, tragedy, novel, painting—by a single author or artist. It can also mean the entire corpus of a writer’s work, a lifetime’s accomplishment. (In the latter sense, it is often given the masculine gender.) Tadié observes that in certain mechanical and superficial ways the last three volumes of Proust’s novel (composed of seven with different titles) were left unfinished at his death. Then he adds emphatically: “But if In Search of Lost Time is unfinished in some details, it is in no fashion an incomplete work [une oeuvre incomplète].” Four pages later in closing, he states that in this new edition the reader will be able to reconstruct Proust’s entire “oeuvre,” meaning everything he wrote. This subtle shift from one meaning of oeuvre to another cannot be ignored. Tadié beckons us away from the universally recognized 3,000-page novel to consider a very different entity: Proust’s entire output. Do we wish to follow him?

Tadié begins this closing section by recalling Proust’s discussion in The Captive of the “retroactive unity” projected back over their earlier works by certain nineteenth-century writers such as Balzac. Even though Tadié claims that this discussion “defines Proust’s poetics,” he goes on to demonstrate the opposite. Proust did not find his “circular structure” after the fact. “Deciding from the start to link his first and last chapters,” Tadié writes, “Proust avoids the incompleteness of the great nineteenth-century books.” Tadié argues that the special effectiveness of Proust’s structure lies not in the fact that he imposed it retrospectively but that he waited until the end to reveal it.

Suddenly, in the middle of a paragraph, Tadié insists that what Proust tells us about the order and structure of his novel has led to “the principles that direct the present edition.” Tadié cites without transition an earlier abandoned novel to introduce Proust’s taste for unpublished writings and concludes: “For the amateur, nothing that has fallen from Proust’s pen, above all if it has to do with the novel, is indifferent.” In one sentence, Tadié has vaulted from the unifying structure of the novel to the mass of materials omitted from it in order to create that unity. A bit later, Tadié tells us how much affection Proust felt for the word esquisse, meaning sketch or draft.

Proust’s three fictional artists are now summoned to testify in favor of successive, stacked-up, unfinished inédits of the kind Tadié wants to include in his edition. The inédits, Tadié tells us, resemble the built-up layers of glaze seen by the writer Bergotte in Vermeer’s patch of yellow wall in the View of Delft. (But glazes are integral to the final painting, not preliminary trial runs separate from the work.) The editor of the composer Vinteuil’s posthumous septet, Tadié argues, makes sense of his “indecipherable notations…by presenting the successive layers which, once unfolded, permit us to understand the composition of the work, the depth of its material.” (But in Proust’s novel, Vinteuil’s editor, rather than laying out all the prospective versions, winnowed them down to the firm line of the septet’s unity.) And the painter Elstir sometimes preferred an early sketch to the final product. Now Tadié feels confident enough to make his final declaration of principle about his editorial method, about the true status of literature, and about the final meaning of the word oeuvre.


The oeuvre, daughter of time, does not assume its full relief unless one superposes its different stages, does not reveal its full depth unless, from the “ground plan,” one descends to the crypt of the cathedral. It is a great privilege to be present at the birth of a work. The drafts must therefore not be considered frozen and lifeless, but read as Swann listened to the motifs of Vinteuil’s sonata: “Swann listened to all the scattered themes which would contribute to the composition of the phrase, like the premises of a necessary conclusion, and he witnessed its birth.” Thus casting on the full set of Proust’s published works and on the even more considerable mass of his unpublished writings a “retrospective” gaze, similar to the gaze the novelist himself cast on Pleasures and Days, the prefaces to Ruskin’s works, Jean Santeuil, Against Sainte-Beuve, on his articles, rough drafts, and letters in order to compose In Search of Lost Time, thus the reader reconstructs the work [oeuvre]—in time.1

In other words, the oeuvre you have undertaken to read in buying these four volumes has become something far vaster than the novel Proust wrote and published. In order to locate its mysterious genesis, you must take on his entire written output. The proliferation of sketches and drafts, which the author eliminated in order to obtain his circular structure and to maintain a certain order of magnitude, the editor now salvages and reintegrates, not into the text proper of the novel but as close to it as possible. This editorial decision and the shaky arguments justifying it constitute a far-reaching redefinition and redistricting of literature. The author’s decisions are now virtually overruled. The boundaries of a work of literature disappear by fiat. The search for a creative process extending backward in time and coterminous with the time of the author’s biography usurps the unity of a work as a finite, free-standing whole.

In arguing so insistently against the inclusion of extensive esquisses and elaborate editorial notes in the new Pléiade edition, I am not arguing against their publication separately in specialized collections annotated by scholars for scholars. Having taught and written about Proust for over thirty years, I find many of Proust’s drafts absorbing for all the reasons Tadié supplies and for further reasons concerning the evolution of Proust’s style. But such sketches can be distracting. At least four esquisses in volume three of the Pléiade edition concern an alluring “girl wearing red roses” who temptingly brushes up against Marcel at a reception. Subsequently he cannot find either her identity or her person, and she becomes an obsession. One can easily begin to remember the girl wearing red roses as a character in the novel. But Proust did not include her. The ordinary reader has enough to keep track of without adding this confusion not essential to the novel. I would welcome her in a separate volume of the Pléiade devoted to Proust’s drafts and revisions, or more appropriately in a scholarly publication with a larger format.

It is difficult to attribute the hypertrophied Pléiade Proust edition to commercial motives. Widely recognized but not a best seller, Proust has no cult status like Elvis Presley’s in which anything he touched achieves market value. What prompted the publishers of the Pléiade to approve for Proust an order of magnitude far exceeding his own plans or expectations was the hubris of literary scholars, armed with a theory. The genetic critics, particularly when led by so disciplined and informed a figure as Jean-Yves Tadié, have tried to do something the deconstructionists never succeeded in accomplishing. They have tried to unmake a work of literature. Intending to carry In Search of Lost Time to its final apotheosis in their sumptuous, four-volume, 7,000-plus-page edition, Tadié and his associates have in effect buried Proust’s novel in trappings and distractions and commentary.

The volumes honor scholars’ decisions about what to include more than they honor Proust’s decisions about what to exclude. The miraculous construction of a coherent roman fleuve out of an ocean of drafts and sketches is reversed by plunging the work back into that amorphous solution. The project is undertaken in the name of finding the genesis of the work, a genesis not recognized in Proust’s choices for inclusion and exclusion but tracked back into his whole extended life as a writer. We have no common word for this operation, the reverse of cutting or editing. Hypertrophize? Whatever we call it, we should not confuse this intended tribute in the new Pléiade with an acceptable way of honoring an author’s accomplishment. It shrouds and demeans the author’s work.

Here in the United States, we stand forewarned against such an attempt. In the 1960s, the Modern Language Association’s Center for the Editions of American Authors began publishing Emerson, Melville, Hawthorne, and Howells in elaborate volumes edited with full apparatus and manuscript variants by committees of five to eight scholars. Each major author was assigned around forty such volumes, each called “An Approved Text.” In a stentorian polemic published in this journal, “The Fruits of the MLA,”2 Edmund Wilson denounced the new editions as top heavy with “rejected garbage,” as offering unreadable texts, and as representing a bare-faced academic “boondoggle.” Wilson pointed to the first 1954 Pléiade edition of Proust as a model solution to editorial problems for such an undertaking. One volley was enough. Thanks in great part to Wilson’s intervention, we have today not the MLA hypertrophized editions but the generally competent, well-presented titles in the Library of America.

Proust himself warned us about the present juncture. In the spring of 1922, only a few months before his death, he considered selling the manuscript and the corrected proofs of Sodome et Gomorrhe to the great couturier and collector Jacques Doucet for 7,000 francs. In July, Proust wrote to his English friend Sydney Schiff (who published fiction under the pseudonym Stephen Hudson) that he did not regret that the transaction had fallen through. For he had learned that Doucet’s collection would become public.

Now the thought is not very agree-able to me that anyone (if my books continue to interest people) will be able to paw through my manuscripts, to compare them to the definitive text, to form suppositions, which will always be wrong, on my work methods, on the evolution of my thought, etc. All that bothers me a little and I wonder if I wouldn’t do better to cut back 5 or 7 thousand francs in my expenses rather than to expose myself to this posthumous indiscretion.3

In his biography, Tadié quotes the first of these two sentences and concludes tersely, “Proust had foreseen everything.”

In addressing the 1998 meeting of the French Literature Conference at the University of South Carolina, I gave a more comprehensive account of genetic criticism in relation to Proust. In front of that audience of scholars of French literature familiar with the circumstances, I felt an obligation to draw conclusions that I renew here.

I propose that we boycott the overblown, misconceived, and overpriced new Pléiade edition. It saps and traduces Proust’s lifelong devotion to a single work. It includes materials of concern primarily to scholars, four thousand pages best published in a separate scholarly edition keyed to a reader’s edition. Every time we cite the new Pléiade edition, we are endorsing a scholarly enterprise that seeks not to identify and present the text of a major literary work but to blur and smear it back into a congeries of manuscripts. The new four-volume Pléiade conveys an image of drowning in one’s own excessive creation, which is exactly wrong for Proust’s resolute molding of a free-standing narrative out of the available mass of materials. Let us not yield to the temptation to accept unthinkingly the prestige of the Pléiade collection.4


In the face of the resolute march toward expansion and elephantiasis in Proust studies, it is reassuring to encounter a short biography of Proust by a seasoned writer inaugurating a new series. Edmund White has written a number of novels and a long, somewhat adulatory biography of Jean Genet. Now, in a mere 165 pages, he offers a contemporary version of the traditional “life and works” treatment. The first chapter informs us indirectly that Proust resisted identification as either a Jewish or a homosexual writer, and also that he could display symptoms of anti-Semitism and of what is now called homophobia. Furnishing the basic information without excessive detail, White writes pertinently on Proust’s addiction to and withdrawal from Ruskin’s influence and provides in Chapter Eight a succinct account of the unexpected origin of Proust’s novel during a period of doubt and indecision. We learn about his odd work routines and his equally odd social life. White acknowledges and follows Tadié’s biography in giving constant attention to the way many events of Proust’s untypical life were transmuted into fictional episodes.

At the beginning, White’s concern with his subject remains perfunctory, as if he is writing on assignment. Later chapters display vigor and conviction. On a few questions, his exposition wavers. In Chapter Seven, he takes a sound view of Proust as a philosophical novelist. “…If he was a philosopher, at the same time he had more faith in the senses and in memory than in the intellect to experience ultimate truth.” Sixty pages later, however, he allows himself to speak too easily of “the immortal ideas” which can be “codified in great works of literary art.” And on the next-to-last page, White refers to “Proust’s rejection of rustling, wounded life in favor of frozen, immobile art.”

Let me assure all potential readers of Proust that White is right the first time and then goes astray. Proust never abandons rustling, wounded life even during the profound reflections on art and literature in the next-to-last section of his novel. Following those pages, the narrative turns unmistakably back to the comedy and conflict of social life among the aging cast of characters assembled at the final reception, and closes with the narrator engaged in a process of writing so down-to-earth and lifelike that it is compared to Françoise’s cooking and dressmaking. “Frozen, immobile art” is an inappropriate metaphor for the sustained, propulsive flow of Proust’s river novel. It forms pools; it never comes to a standstill.

Despite this complaint, I find White’s little book a serviceable introduction to a sometimes intimidating author. White has obviously lived in close sympathy with Proust’s universe and can express his absorption in it. At the end, White refers offhandedly to “the homosexual bias of my little book.” He does pay attention both to established facts and to hearsay about Proust’s sexual tastes, and he makes a strong argument, following Tadié, that the transposition of Proust’s chauffeur Agostinelli into the major female character Albertine elevates the emotional and narrative level of the book rather than distending it. I believe, however, that White is both naive and unperceptive when, almost a century after the fact, he calls “absurd” Proust’s determination to avoid the label “homosexual.” In order to maintain a dispassionate view of the whole range of human society, Proust could not afford to identify himself as engaged in special pleading for a single narrow segment of it. The derogatory expression “the closet” that White uses to describe Proust’s insistent discretion about his private life does not do justice to the integrity of Proust’s goals of detachment, meticulous description and analysis, and frequent satire in capturing the textures of life.

A more probing examination of the role of homosexuality, sadism, and evil in Proust’s novel is contained in the central essay, “This shuddering of a heart being hurt,” of Antoine Compagnon’s Proust Between Two Centuries.5 Compagnon never berates Proust about staying in the closet and deals sensitively with the disgust and pain that form an important part of Proust’s description of erotic feelings. Both White and Compagnon consider the thirty pages of Sodom and Gomorrah I describing the first encounter between the Baron de Charlus and Jupien as the crucial episode admitting male homosexuality into the novel as a major motif. The botanical analogies and the comic detachment Proust deploys to render the encounter alternately neutralize the “vice” of homosexual behavior and caricature it to grotesque extremes. These pages make a great literary set piece, like Swift’s brilliantly ambivalent description of the Yahoos.

But neither White nor Compagnon does full justice to these searing pages on homosexuality and humanity. They contain, like a steep mountain range blocking the way, the longest sentence in Proust’s novel—close to two and a half pages.6 It enumerates nine nearly disabling burdens under which this “accursed race” must live and compares its predicament to that of the Jews. Having surmounted the barrier of that mammoth sentence, Proust interjects the story of a solitary homosexual, abandoned by his only lover “like a sterile jellyfish stranded on the beach.” The last page of this intense section offers us a succinct summary of the narrator’s and, I believe, Proust’s attitude toward homosexuals.

In all countries surely they form a colony of exotic, cultivated, musical, sharp-tongued members, who have charming qualities and unbearable faults. They will appear in higher relief in later pages of this book. But here it is appropriate to warn provisionally against the fatal error of creating, as people have encouraged the creation of a Zionist movement, a sodomite movement and the rebuilding of Sodom.

As it is possible to include too much extraneous material in an edition of Proust’s novel, it is also possible even for astute critics to overlook passages crucial to understanding Proust’s thought. We must never stop working to raise the level of our reading to the scale—immense, yet still human—of his writing.

This Issue

March 18, 1999