SOURCE: Smith, Roch C. “Gaston Bachelard and Critical Discourse: The Philosopher of Science as Reader.” Stanford French Review 5, no. 2 (fall 1981): 217-28.

[In the following essay, Smith provides an assessment of Bachelard's contribution to critical discourse.]

Usually original, often provocative, Bachelard's quarter century of work on the literary imagination has itself caught the imagination of numerous critics and imitators who would apply his approach to other texts. This is particularly true of his works on the imagination of elements, or simply the “Elements,” as I shall call them here.¹ Some of the early adaptations of Bachelard's Elements may seem somewhat rigid or naive to us now.² But despite numerous attempts to determine Bachelard's influence on critical discourse, including Vincent Therrien's heroic endeavor to document Bachelard's “revolution in literary criticism,” by identifying eight critical methods in his work,³ the specific nature of Bachelard's contribution remains frustratingly elusive. For some, Bachelard is primarily a psychological critic.⁴ Others consider his so-called phenomenology of the literary image his most original legacy.⁵ Still others continue to see in his theory of the four elements the fundamental principle of his poetics and his main contribution to literary research.⁶ Even Therrien, whose meticulous analysis of Bachelard as a literary critic attempts to establish the plurality of his approach to literature, must finally admit that the distinctions between the eight methods are necessarily blurred (344).

Perhaps what is needed, in order to reexamine Bachelard's contribution to critical discourse, is to consider his work in light of his own frame of reference, namely that of reader and, ultimately, of philosopher. For Bachelard proposes a personalized reading of the text. His several reminders, throughout his work, that he is a mere reader are echoed in his final and most intimate book, La Flamme d'une chandelle: “Jamais deux fois nous n'avons fait la même lecture … Quel mauvais professeur de littérature nous eussions fait!”⁷ And reading for Bachelard has always meant to respond to a work's images, to allow them to reverberate within him. At the same time, even when dealing with the poetic image, Bachelard never fully abandons his concern with the twin philosophical questions of epistemology and ontology. Central to his philosophy of science, they continue to serve as implicit and, on occasion, quite explicit norms for his oneiric work.

Bachelard's attitude as a reader of literature is essentially to be open to the literary image, to allow it to exist on its own terms rather than to seek its causes or to attempt to find its place within the totality of the work. Such a posture is justified, in Bachelard's view, primarily because he sees the literary image as autonomous and therefore irreducible. For Bachelard, an original literary image is not dependent upon observation, it is not secondary. Radically new, it precedes both perception and thought because it is created by an imagining consciousness directly from language.⁸ In what could serve as a concise example of Derrida's “logocentrisme,”⁹ Bachelard advances the general thesis that “tout ce qui est specifiquement humain dans l'homme est logos” (PE [La Poétique de l'espace], 7). Language is thus the source of all human production, including the literary image. As he explains in La Poétique de l'espace: “la nouveauté essentielle de l'image poétique pose le problème de la creativité de l'être parlant. Par cette créativité, la conscience imaginante se trouve être, très simplement mais très purement, une origine” (PE, 8). It is to the independent product of this original consciousness that Bachelard seeks to respond through reading. The reader follows the lead of the image, he experiences its reverberation or “retentissement” as he himself becomes poetically creative: “Il s'agit en effet, par le retentissement d'une seule image poétique, de déterminer un véritable réveil de la création poétique jusque dans l'âme du lecteur” (PE, 7). The poetic image elicits wonder rather than criticism from the true reader who obeys its “dynamique immédiate” (PE, 3). And, while the image is singular and unexpected, it is nevertheless communicable from poet to reader because of the latter's astonished attraction to this image which reverberates within him as if it were his own creation. Indeed, for Bachelard, the image cannot be communicated in any other way. Receptive reading is not only desirable, it is necessary.

Concomitantly, reading demands that more narrow critical attitudes be put aside, for they would interfere with its processes. Beginning with his earliest works on the imagination, which for Bachelard means the literary imagination,¹⁰ he consciously rejects conceptual, positivistic methods of analysis. Initially his concern is with overly rational approaches which would threaten the unconscious source of literary imagination. “Il ne faut pas trop vite s'adresser aux constructions de la raison pour comprendre un génie littéraire original. L'inconscient, lui aussi, est un facteur d'originalité,” he cautions in La Psychanalyse du feu.¹¹ This warning is echoed in L'Eau et les rêves, where he reminds us that “la raison n'explique pas les rêves.”¹² And, while he will begin his Poetics¹³ with a condemnation of psychoanalytical explanations because “le psychanalyste intellectualise l'image” (PE, 7), his fundamental posture has not really changed: the image must be understood on its own terms, which is to say as an original creation. And that, for Bachelard, necessarily excludes any sort of conceptual constructs: “Le monde est beau avant d'être vrai. Le monde est admiré avant d'être vérifié” he had pointed out in L'Air et les songes.¹⁴

While it is true that Bachelard begins the Elements with an attempt to discover the objective conditions of literary images by exploring their reference to the imagination of fire, water, air, or earth, he also recognizes that imaginative processes are basic to creative activity and cannot, ultimately be determined. As a consequence, he admonishes, we must decompose the forces of the poem and break with “l'idéal naïf, l'idéal egoïste, de l'unité de composition” (PF [La Psychanalyse du feu], 181). Such mistrust for conceptual critical discourse is manifested throughout the Elements. Indeed, it is mildly surprising to see him talk about the “prudence scientifique” (PE, 3) of the Elements when he declares in his introduction to La Poétique de l'espace that he now intends to respond more personally to the image. The change between the Elements and the Poetics is by no means a shift away from a series of scientifically objective works. His attack, in La Poétique de l'espace, on what he sardonically calls the critic's “superiority simplex” could just as easily have been made earlier in the Elements. It might be more accurate to speak merely of the taxonomic rather than the scientific prudence of these earlier works. For conceptual scientific methodology is consistently rejected in these works, as we have seen, but Bachelard's classification of images according to the four elements of antiquity as well as his distinctions between the formal, material, and the dynamic imaginations do allow him to speculate on the conditions for the existence of the poetic image without concerning himself overmuch with how the reader comes to know that image. His categories offer an explanation of the nature of the literary image: their success as a method of literary interpretation is limited, as Suzanne Hélein-Koss has demonstrated.¹⁵ Bachelard's classifications are comfortably familiar tools which allow this rationalistic sorcerer's apprentice to explore the mysteriously fascinating world of the poetic image under the initial illusion that he will be left unchanged.

Bachelard's works on scientific epistemology specifically reject mere classification of substance. One of the most frequently occurring observations of these early works is that modern science differs from the pre-scientific attitude in abandoning completely the notion of a priori substance. In La Formation de l'esprit scientifique, he speaks of “l'explication monotone des propriétés par la substance”¹⁶ which has blocked experimental knowledge, thus constituting an epistemological obstacle. “Toute cette énorme surcharge du rêve,” he adds, “toute cette valorisation des substances par le temps passé à les préparer, il faudra en débarrasser la pensée scientifique” (FES [La Formation de l'esprit scientifique,], 123). Imagination, on the other hand, encourages a return to substance, as Jacques Gagey has pointed out.¹⁷ This nostaliga for substance permeates much of Bachelard's work on the imagination. Where science has had to avoid the illusions of value-ridden substance, imagination delights in such values, especially when they enrich literary expression. Literary imagination, for Bachelard, asserts the worth of substance naively, without recourse to the rigorous reality tests of science. He sees creative imagination as a “fonction de l'iréel qui est psychiquement aussi utile que la fonction du réel” (TRV [La Terre et les rêveries de la volonté], 3). Using the term phenomenology to mean merely the study of the phenomena of literary images in La Terre et les rêveries de la volonté, Bachelard suggests that “Pour l'imagination matérielle, toute phénoménologie révèle une ontologie, tout phénomène [which is to say each literary image] a sa substance” (TRV, 236).

Bachelard sees in the communicability of an original literary image evidence that such an image grows out of a shared valorization of substance between poet and reader. Because it is communicated, this new image does not remain a subjective illusion of the poet. Having taken linguistic form, it is now able to awaken a similar subjective response in the reader. Not an object but an “événement du logos” (PE, 7), it has, nonetheless, acquired a kind of ontological status.

Indeed, Bachelard's preoccupation throughout the Elements is primarily ontological as he seeks to determine the nature of the literary image. And while it would be beyond the scope of this essay to fully explore Bachelard's ontology of the image, it should be noted that there are some variations in the way Bachelard relates the image to being. For, in addition to seeing the image as a communicable linguistic event, Bachelard also claims that language reveals a substratum of being through the literary image. Speaking of the material imagination in Sartre, for example, Bachelard asserts that “la matière est révélatrice d'être … [elle] nous donne le sens d'une profondeur cachée, elle nous enjoint de démasquer l'être superficiel” (TRV, 115). In addition, as I have indicated elsewhere,¹⁸ his view of an open, dynamic, and immediate poetic image betrays a conception of being which shares with science a rejection of the topological nature of everyday language while viewing the real as a particular case of the possible. Where the scientist gives ontological value to the possible through mathematics, the poet uses language to create a literary image which has not existed before. But, whether considering the literary image as an event of language, as a revealer of hidden depth, or as an instance of possibility, Bachelard's consistent concern, throughout the Elements, is with its ontological status.

Having begun to develop, in the Elements, what Gilbert Durand would call an “ontologie symbolique,”¹⁹ Bachelard is then able to abandon the taxonomic guide that allowed him to speculate on the nature of the image in the first place. Freed from the classificatory apparatus of the Elements, he is now able to consider the grounds for knowing this new reality. Thus, as a complement to an evolving symbolic ontology, Bachelard might be said to develop an “epistemology of symbol” in the Poetics. After summarizing the ontological status of the literary image in the first few pages of La Poétique de l'espace, he rejects, as we have seen, both objective criticism “[qui] étouffe le ‘retentissement’” (PE, 7) and, more importantly, the psychoanalytical approach “[qui] explique la fleur par l'engrais” (PE, 12). Bachelard seems to have discovered, not only that the methodology borrowed from psychology and psychoanalysis is inadequate to understanding the special nature of the literary image, but that it has kept him from even considering the question of how he can best know that image. As Vincent Therrien so aptly expressed it, “il s'agit … d'améliorer l'analyse de l'image poétique en en proposant un mode d'approche qui soit plus conforme à la nature même, au fond ontologique de la poésie” (324). Since, as Bachelard explains in La Poétique de l'espace, “on demande au lecteur de poèmes de ne pas prendre une image comme un objet, encore moins comme un substitut d'objet, mais d'en saisir la realité spécifique” (PE, 3), traditional objective means of knowing are no longer appropriate. Speaking, not to the constructed thought of science, where concepts are logically related, but to an admiring and willfully naive consciousness, the image, itself a product of a creating consciousness, elicits a subjective response. While scientific knowledge “augmente à chaque conquête de l'abstraction constructive,” Bachelard explains in La Poétique de la rêverie, “C'est un non-sens que de prétendre étudier objectivement l'imagination, puisqu'on ne reçoit vraiment l'image que si on l'admire.”²⁰ The specific nature of the image is such that it requires a nonobjective method in order to be “known” in the broadest sense of that term. Consequently, Bachelard is perfectly consistent when he claims that, “Il faut en venir, pour éclairer philosophiquement le problème de l'image poétique, à une phénoménologie de l'imagination. Entendons par là une étude du phénomène de l'image poétique quand l'image émerge dans la conscience comme un produit direct du coeur, de l'âme, de l'être de l'homme saisi dans son actualité” (PE, 2). Bachelard's phenomenology stems from epistemological concerns; it is, in fact, a means of knowing the poetic image. Moreover, in light of the specific reality he assigns to the literary image, he views phenomenology as the only valid means of initially knowing that image.

If we take our distance for a moment from Bachelard's Poetics and consider the totality of his work, including his scientific epistemology, it becomes clear, as Jean-Pierre Roy has pointed out, that there exists not only a duality of object in Bachelard as he deals with science and with literary imagination, but also one of method.²¹ Bachelard's approach to science, Roy reminds us, is based on “concept,” while his more subjective approach to literature is grounded in “image.” In Roy's view, such methodological duality is a fundamental flaw in Bachelard's work because it removes literature from the scope of scientific inquiry. It is a “refus de la science” (7), which, in Roy's opinion, allows literary discourse to salvage a humanistic ideology that Bachelard's scientific epistemology had attempted to discredit. Rather than the syntagmatic discourse of science, in which what is known is constructed conceptually, Bachelard's paradigmatic emphasis on depth in his oneiric work results, according to Roy, in a prejudicial preference for reading instead of criticism (175). While Roy is undoubtedly correct in maintaining that “la lecture [est] … l'activité bachelardienne par excellence” (173), his contention that Bachelard's predilection for reading blinds him to the value of criticism is questionable. Not only does Bachelard specifically recognize the role of such criticism, but his epistemology indicates that he has, in fact, the required critical competence for the kind of conceptual criticism Roy has in mind.

While Bachelard is clearly a practitioner of reading in responding to the literary image, and while, as we have indicated, he specifically rejects reductive critical approaches in order to read more faithfully, he nevertheless acknowledges the importance of conceptual criticism as a continuation of his own activity. This is especially so in the Poetics, where he insists on the need to study the image as a function of both the soul and the mind. “A notre avis, âme et esprit sont indispensables pour étudier les phénomènes de l'image poétique … pour suivre surtout l'évolution des images poétiques depuis la rêverie jusqu'à l'exécution … Pour faire un poème complet, bien structuré, il faudra que l'esprit le préfigure en des projets” (PE, 5-6). Returning to this idea in La Poétique de la rêverie, he associates it to the Jungian distinction between animus, with its concepts and projects, and anima, with its revery: “En somme, il faut bien avouer qu'il y a deux lectures: la lecture en animus et la lecture en anima” (PR [La Poétique de la rêverie], 55-56). Thus Bachelard's practice as a reader in anima does not preclude a conceptual, in animus consideration of the poem. The fact that Bachelard himself generally does not go beyond in anima reading does not mean that he fails to recognize the possibility or even the necessity of conceptual activity in dealing with literature. On the contrary, he quite deliberately leaves aside the conceptual problem of the work's totality, not because he considers in anima reading the only legitimate approach, but because, in his view, the poetic image must first be understood on its own terms before the work as a whole can be validly approached. To do otherwise, in Bachelard's estimation, would easily result in never responding adequately to either the image or the poem. As he explains in his introduction to La Poétique de l'espace:

En limitant de cette manière notre enquète à l'image poétique dans son origine à partir de l'imagination pure, nous laissons de côté le problème de la composition du poème comme groupement des images multiples. Dans cette composition du poème interviennent des éléments psychologiquement complexes qui associent la culture plus ou moins lointaine et l'idéal littéraire d'un temps, autant de composantes qu'une phénoménologie complète devrait sans doute envisager. Mais un programme si vaste pourrait nuire à la pureté des observations phénoménologiques, décidément élémentaires, que nous voulons présenter … Il y aurait pour nous immodestie à assumer personnellement une puissance de lecture qui retrouverait et revivrait la puissance de création organisée et complète touchant l'ensemble d'un poème.

(PE, 8)

Clearly, for Bachelard, conceptual criticism applies to the totality of the work, to that aspect of literature he neglects for epistemological reasons. And, just as clearly, he recognizes the worth of an activity which is able to confront the work in its entirety.

Well before his works on the imagination, Bachelard had insisted on the role of concept in determining totality. This is precisely the function he recognizes in mathematics, which he describes as “une aspiration au complet” in Le Nouvel Esprit scientifique.²² Since it is a question of “la primauté [de] la relation sur l'être” (NES [Le Nouvel Esprit scientifique], 67) in science, Bachelard is epistemologically consistent in dealing with the totality and composition of reality in science where, as he puts it, “Le mathématisme est non plus descriptif mais formateur” (FES, 5). When that reality changes to the poetic image, it remains consistent for him to neglect totality and composition. But, beyond this, when Bachelard recognizes that, in order to be considered in its entirety, a work demands methods of analysis earlier associated with science, he is, once again, adjusting his epistemology to the reality at hand. As part of his phenomenology of the literary image, Bachelard may be said to “bracket” his analysis of the poem. Yet such bracketing does not limit either in animus or in anima reading; it is designed to expand both.

Nor can it be inferred from Bachelard's focus on in anima reading that he was incapable of a more conceptual criticism. His works on scientific epistemology amply demonstrate his familiarity with questions of composed or structured reality: “On arrive ainsi à penser une sorte de structure préalable, de construction en projets, de réel en plans, de moule rationnel pour la technique expérimentale” (NES, 158-59). Bachelard, no novice in the structural methods of science, makes a thoroughly informed judgment in deciding that, “En poésie, le non-savoir est une condition première” (PE, 15). Eschewing a criticism that might validly be called “pre-structural,” Bachelard maintains, to use Murray Krieger's phrase, that poetry requires “distinctive treatment.”²³

This brings us to a fundamental paradox in Bachelard's work. As we have seen, he is aware that the nonobjective ontological status of the literary image requires a subjective response in order not to destroy the specific reality into which he is inquiring. Yet, because subjective responses necessarily vary from reader to reader as the image emerges into consciousness, it is inescapable that the specific reality of the image will vary according to the interpreter. Or, to put it differently, while Bachelard adapted his epistemology to ontological concerns, the specific ontological status of the image is, to some degree at least, determined by Bachelard's special epistemological adaptation. What is known depends in part on how it is known.

Yet the paradoxical nature of this notion cannot merely be attributed to Bachelard's avoidance of conceptual criticism when dealing with the image. It cannot be dismissed as the inevitable consequence of a subjective approach to the image. For in his scientific epistemology Bachelard also recognizes that the object to be known depends on the method of knowing: “L'objet mesuré n'est guère plus qu'un degré particulier de l'approximation de la méthode de mesure. Le savant croit au réalisme de la mesure plus qu'à la réalité de l'objet. L'objet peut alors changer de nature quand on change le degré d'approximation” (FES, 213). Thus Bachelard reminds us that our knowledge of physical reality is dependent on modulations of the quantitative means of knowing: “dis-moi comment l'on te transforme, je te dirai qui tu es” (NES, 28), he quips. In science a variational method reflects the scientist's attempt to perfect his approximation to reality, although such variation is ultimately incorporated into a coherent conceptual system.

Similarly, in the case of the literary image, the method and the specific reality to be known are subject to modification. But here Bachelard insists on an important difference: the transformations in the image, as distinct from the poem as a whole, cannot be incorporated into a larger conceptual system. “L'image poétique est en effet essentiellement variationnelle” (PE, 3). It cannot merely be reduced to a constituent part of a larger whole and it cannot be known merely as a function of that totality. “Elle n'est pas, comme le concept, constitutive” (PE, 3), Bachelard reminds us. For the image is unique and so is the reader's response, since in responding to a particular image the reader assigns it value and makes it personal.

In recognizing the necessary interplay between reader and text, Bachelard goes beyond mere interpretation. The reader, for Bachelard, is actively involved in determining the specific nature of the text at any given moment. While Bachelard does not claim that all meaning is assigned to the text by the reader, neither does he assert that the complete textual meaning, at least as expressed through the literary image, pre-exists the act of reading. Literature, for Bachelard, is a process before it can be anything else, and that process is transsubjective, that is, it grows out of a link of consciousness between author and reader. The image takes on its full meaning by being communicated while remaining unique.

In assessing Bachelard's perception of and contribution to critical discourse, it becomes clear that one must distinguish between a conceptual approach to the literary object and a more subjective approach to the literary process. As a practitioner of the latter, intent on responding fully to the literary image, Bachelard may be considered a forerunner of “Response Criticism.” Such critics, it has been suggested, “treat the literary work as autonomous in that they believe it has no significant relation with a causality, but they deny it objectivity inasmuch as objectivity implies that it contains one or several more or less immutable meanings that the critic-reader must find or discover. Instead, they say that its meaning is acquired by it in the act of its being read by the reader in whom meet many historical currents.”²⁴ While Bachelard is not especially concerned with identifying historical currents that inform the reader's interpretation, his particular approach to the text otherwise corresponds remarkably well to traits attributed to Response Criticism.

This kind of transsubjective response, which disciples of Bachelard such as Georges Poulet and Jean-Pierre Richard have practiced, is, of course, hermeneutic, as Gérard Genette reminds us.²⁵ Yet, beyond his own practice, Bachelard recognizes that a complete accounting of the work requires a more objective criticism, one which would apply principles of his own scientific epistemology with its emphasis on structured relations. Such a conceptual discourse, Bachelard repeatedly admits, would counterbalance his own subjective attention to imagistic detail and his neglect of the work's totality which so frequently results in what Vincent Therrien calls accusations of “pointillisme” (85). Such a counteraction is a potential function of structuralism which Gérard Genette recognizes when he writes that, “Le structuralisme serait alors, pour toute critique immanente, un recours contre le danger d'effritement” (157). To the extent that any criticism growing out of Bachelard's scientific epistemology and his several statements about objective criticism may be considered structural, or pre-structural,²⁶ Bachelard's work does not point to an opposition of hermeneutics and structuralism, as Ricardou implies,²⁷ but rather to the possibility of a complementary relationship between the two. One cannot, of course, overlook fundamental philosophical differences between these two approaches and, yet, Genette is surprisingly close to Bachelard when he recognizes the possibility that, “la relation qui unit structuralisme et herméneutique pourrait être non pas de séparation mécanique et d'exclusion, mais de complémentarité: à propos d'une même oeuvre, la critique herméneutique parlerait le langage de la reprise du sens et de la recréation intérieure, et la critique structurale celui de la parole distante et de la reconstruction intelligible. Elles dégageraient ainsi des significations complémentaires, et leur dialogue n'en serait que plus fécond, à cette réserve qu'on ne pourrait jamais parler ces deux langages à la fois” (161). Indeed, Bachelard refuses, not science, as Roy would have it, but the pitfall of speaking “both languages at once.” And he does this, it must be emphasized, not out of ignorance of one of the “languages,” but out of a deliberate choice dictated by the object of his inquiry. In recognizing the possibility of a complementary approach, Bachelard argues for the kind of openness and flexibility in literary criticism he had found in science. This flexible, complementary critical discourse transcends any individual method, including Bachelard's own activity, which we have characterized as an early form of Response Criticism. Insisting on some attention to the conditions of the critic's enterprise and warning against a premature quantification of the work which would distort its meaning, it echoes Bachelard's earlier reminder to developing scientists that, “Il faut réfléchir pour mesurer et non pas mesurer pour réfléchir” (FES, 213).

Notes

1. La Psychanalyse du feu, 1938; L'Eau et les rêves, 1942; L'Air et les songes, 1943; La Terre et les rêveries de la volonté, 1948; La Terre et les rêveries du repos, 1948.

2. In retrospect, Robert Champigny's “L'Expression élémentaire dans L'Etre et le néant,” PMLA, 68 (1953), 56-64, may seem a bit too literal, while Gilbert Durand's “Psychanalyse de la neige,” Mercure de France, 318 (1953), 615-39, displays what now appears to be a certain ingenuousness.

3. Vincent Therrien, La Révolution de Gaston Bachelard en critique littéraire (Klincksieck, 1970). (Subsequent references given parenthetically.)

4. These include, François M. Dagognet, “M. Gaston Bachelard, philosophe de l'imagination,” Revue Internationale de Philosophie, 14 (1960), 32-42; C.G. Christofides, “Gaston Bachelard's Phenomenology of the Imagination,” Romanic Review, 52 (1961), 36-47; Bernard Elevitch, “Gaston Bachelard: The Philosopher as Dreamer,” Dialogue, 7 (1968-1969), 430-48; and Robert Champigny, “Gaston Bachelard,” in Modern French Criticism: From Proust and Valéry to Structuralism, ed. John K. Simon (Chicago: University of Chicago Press, 1972), pp. 175-91.

5. A view shared by Jean-Claude Pariente, “Présences des images,” Critique, 20 (1964), 3-27; Laurent Le Sage, “Gaston Bachelard,” in his The French New Criticism (University Park: Pennsylvania State University Press, 1967), pp. 27-35; and Gilbert Durand, “Exploration de l'imaginaire,” in Circé, I (Lettres Modernes, 1969), 15-45.

6. See, for example, Rolf Vige, “Gaston Bachelard—elementloven og drømmelivets poetikk,” Vinduet, 20 (1966), 198-215; and Michel Mansuy, Gaston Bachelard et les éléments (Corti, 1967).

7. Gaston Bachelard, La Flamme d'une chandelle (Presses Universitaires de France, 1961), p. 105.

8. Gaston Bachelard, La Poétique de l'espace (Presses Universitaires de France, 1972), p. 8. (All references are to this edition and are given parenthetically in the text, preceded by the abbreviation PE.)

9. Jacques Dérrida, De la grammatologie (Minuit, 1967), pp. 11-12, et passim.

10. See, for example, Gaston Bachelard, La Terre et les rêveries de la volonté (Corti, 1948), pp. 5-6. (Subsequent references are given parenthetically in the text, preceded by the abbreviation TRV.)

11. Gaston Bachelard, La Psychanalyse du feu (Gallimard, 1949), p. 144. (All references are to this edition and are given parenthetically in the text, preceded by the abbreviation PF.)

12. Gaston Bachelard, L'Eau et les rêves (Corti, 1942), p. 97.

13. I.e., La Poétique de l'espace, 1957, and La Poétique de la rêverie, 1960.

14. Gaston Bachelard, L'Air et les songes (Corti, 1943), p. 192.

15. Suzanne Hélein-Koss, “Gaston Bachelard: Vers une nouvelle méthodologie de l'image littéraire,” French Review, 45 (1971), 353-64.

16. Gaston Bachelard, La Formation de l'esprit scientifique (Vrin, 1972), p. 21. (All references are to this edition and are given parenthetically in the text, preceded by the abbreviation FES. Original edition published in 1938.)

17. Jacques Gagey, Gaston Bachelard ou la conversion à l'imaginaire (Marcel Rivière, 1969), pp. 207-08.

18. See my “Gaston Bachelard and the Power of Poetic Being,” French Literature Series, 4 (1977), 234-38.

19. Gilbert Durand, “Science objective et conscience symbolique dans l'oeuvre de Gaston Bachelard,” Cahiers du Sud, 4 (1964), 48.

20. Gaston Bachelard, La Poétique de la rêverie (Presses Universitaires de France, 1974), p. 46. (All references are to this edition and are given parenthetically in the text, preceded by the abbreviation PR.)

21. Jean-Pierre Roy, Bachelard ou le concept contre l'image (Montréal: Presses de l'Université de Montréal, 1977), p. 17. (Subsequent references are given parenthetically.)

22. Gaston Bachelard, Le Nouvel Esprit scientifique (Presses Universitaires de France, 1968), pp. 32-33. (All references are to this edition and are given parenthetically in the text, preceded by the abbreviation NES. Original edition published in 1934.)

23. Murray Krieger, Directions for Criticism: Structuralism and Its Alternatives (Madison: University of Wisconsin Press, 1977), p. 10.

24. Carol Sherman, “Response Criticism: ‘Do Readers Make Meaning?’” Romance Notes, 18 (1977), 289.

25. Gérard Genette, Figures (Seuil, 1966), p. 158. (Subsequent references are given parenthetically.)

26. Jean-Pierre Roy points out that the conceptual methodology of Bachelard's scientific epistemology is essentially structural (74), but he claims that Bachelard neglected such methodology entirely when dealing with the image (89 and 187). I have already indicated that Bachelard did indeed recognize the value of such methodology when applied to literature and that he was quite capable of applying it himself. Left essentially unexplored by Roy are the reasons for Bachelard's methodological choice, given his obvious ability and his demonstrated understanding of the need for conceptual criticism.

27. Jean Ricardou, “Le Caractère singulier de cette eau,” Critique, 23 (1967), 718-33. See also Tom and Verena Conley's comment on Ricardou's observations: “Ideological Warfare: Ricardou's Purge of Bachelard,” Sub-Stance, 1 (1971), 71-78.