No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 02 December 2020
A critical cliché often heard today is that Proust was fundamentally a poet rather than a novelist. The historians of literature and the critics do not put it quite as crudely as that, but their remarks frequently permit such an assumption on the part of the reader. Thus the Castex and Surer manual, in its twentieth-century volume, finds in “toute l'œuvre [de Proust] un climat d'intense Poesie” (p. 82). And Georges Cattaui, in his recent survey of the present status of Proust, though he does not in so many words call Proust a poet or his novel a poem, does say that Proust is above all the heir “de Nerval, de Baudelaire, de Mallarmé,—de ces poètes qui lui ont enseigné l'art de transfigurer les choses, l'art de délivrer la beauté prisonnière … ” Now all this is true if it is merely taken as a vivifying figure of speech, if it merely means that Proust was not a realistic novelist, and that he shows the influence of the great French poets of the late nineteenth century, or that, to use a convenient term, he was a symbolist, like his contemporaries, Claudel, Gide, and Valéry. But it has so often been said in our time that the twentieth century has seen the breaking down of the distinctions between the novel and poetry, that it seems to me useful to demonstrate, by studying two treatments of the same subject, one that of a novelist, Proust, the other that of a poet, Valéry, that there remains a fundamental and profound difference between the intent and the method of prose fiction and of poetry, at least the type that is today called “pure” poetry.
1 “Proust après trente ans. H,” Critique, No. 90 (novembre 1954), p. 918.
2 La Dormeuse appeared in the volume Charmes, of which tie achevé d'imprimer is of June 1922; “La regarder dormir” was publisbed in the Nouvelle Revue française of Nov. 1922 (pp. 513–519), just a few days before Proust's death. Aside from minor variants, it differs in three important details from the text published over a year later in the first volume of La Prisonnière (a corrected version of which is to be found in the recent Pléiade edition, Paris, 1954): tie transition which links the passage to that which precedes is omitted (Proust must have decided that a transitional passage was a poor beginning for an isolated fragment, forgetting that the figure in tie sentence was a key figure for tie whole passage, as I explain further on); the paragraph concerning letters in Albertine's kimono is advanced to about a page from the beginning of the passage (a change which seems to me an improvement). A more startling, although superficial, change is that of the name of the sleeping woman from Albertine to Gisèle! This was an attempt to disguise the fact that the fragment came from La Prisonnière, which Proust was then planning to have printed in Les Œuvres Libres. (See Marcel Proust-Jaques Rivière Correspondance, Paris, 1955, p. 287.) To avoid confusing my readers I shall call the sleeping woman Albertine. Although the version appearing in the review, prepared for publication by Proust before his death, is, as far as I can determine, the most nearly finished version of this passage, I shall refer to and quote from the more readily available Pléiade text.
3 La Dormeuse is given complete a few pages farther on. The length of “La regarder dormir” made it impossible to reproduce it here.
4 It is not the province of this paper to consider whether it is mere coincidence that these two works appeared in the same year, or whether one could have influenced the other. Merely in passing, I may remark that it is highly unlikely that Valéry could have been influenced here by Proust. (He stated in “Hommage à Proust”—NRF, Jan. 1923, p. 117— that he knew Proust only slightly and that he had read only one volume of his werk.) On the other hand, it is possible that Proust read Valéry's sonnet shortly after its appearance, and, since he was working on La Prisonnière in the summer of 1922, some influence of the sonnet on the fragment published in the NRF is not impossible.
5 Between 1914 and Proust's death (18 Nov. 1922) it had become customary for the NRF to print important fragments from the successive volumes of A la recherche du temps perdu before they were published. A letter of Proust to Gaston Gallimard of Sept. 1922, reveals that he considered “La regarder dormir' one of the ”moins mauvais morceaux,“ as he modestly put it, of La Prisonnière (Proust, Lettres à la NRF, Paris, 1932, p. 255).
6 Exemplified in n. 2 above.
7 Pléiade, m, 69. As pointed out in n. 2, this transitional sentence and most of the one following it as well are omitted in the passage published in the NRF. This and the change of the name from Albertine to Gisèle weaken the novelistic structure. But I am examining the passage as a portion of the novel, which it was certainly originally conceived to be, rather than as a unit in itself.
8 The reading “contre elle” is found only in the NRF, Nov. 1922, text. All other editions give “comme elle,” which seems to me to make less sense, and is probably the result of a typist's error.
9 The repugnance Valéry felt toward writing fiction is well known. Breton, in the Manifeste du Surréalisme (Paris, 1925), p. 16, says that Valéry assured him that “il se refuserait toujours à écrire: La marquise sortit à cinq heures.”
10 A discussion of these theories, with numerous quotations from Valéry, will be found in Jean Hytier's La Poétique de Valéry (Paris, 1953), notably in Chapter iv, “Obscurité et Poesie absolue,” and especially on pp. 115–116.
11 Valéry, Variété (Paris, 1924), pp. 98–99 (“Avant-Propos” à La Connaissance de la Déesse de Lucien Fabre).
12 Valéry, “Je disais quelquefois à Stéphane Mallarmé,” Variété III (Paris, 1936), p. 14. In Degas, Danse, Dessin (Paris, 1938), Valéry developed, at some length, his theory of the bad effects of the growth of description in literature.
13 Valéry, “La création artistique,” Bulletin de la Société française de Philosophie (1928), p. 11, quoted from Hytier, p. 172.
14 Valéry, in a conversation with Frédéric Lefèvre, expressed the opinion that a perfect, pure poem could hardly be more than one verse long (F. Lefèvre, Entretiens avec Paul Valéry, Paris, 1926, p. 66; quoted from Hytier, p. 115).
15 This was one metaphor that Valéry could hardly have given directly; he could not, in so many words, call the woman a calorifère or a chaudière.
16 A la recherche du temps perdu, éd. Pléiade, in, 69.