Look around you. Don’t we all have one foot in the air? We all look as though we are traveling. No one has a definite sphere of existence; no one has proper habits; there are no rules for anything; there is no home base.

(Pyotr Chaadaev, Letter on the Philosophy of History)

NARRATIVE VOICE: THE TROUBLE WITH REPRESENTATION

The relationship between art and life examined in Madame Bovary becomes an obsessive concern of the nouveau roman. While Flaubert's heroine reads her books all too literally, suffering from what Valéry would have diagnosed as “literary superstition,”¹ the confusing world of the nouveau roman can be attributed to a linguistic inquiry into the crisis in representation through language. La Modification is a text that is as much about finding a narrative voice as it is the story of Léon Delmont's train trip from Paris to Rome. This search for a voice demands the involvement of the reader not only because the latter is implicated in the vous (“you”) narration, but also because the je (“I”) that speaks in the final chapter does so as the writer of a future book.

Butor critics, as will be demonstrated, tend to view the writing of Léon's book as an emergence of self, one that definitively puts an end to the narrative crisis reflected in the vous narration and the mise en abyme. I shall argue, however, that it is precisely in the role of writer of a future book that Léon will be forever deprived of an authentic narrative voice. My reading of Léon's intricate description of the promised book takes into account Robert Weimann's statement that “what, more than anything else, representational art presupposes and what it thrives on is indeed the loss, the undoing of the plenitude of that property in which the self and the social are mutually engaged and in which their engagement is unquestioningly given and taken for granted” (96). The desire to represent becomes so acute in La Modification that the narration disengages from the story, producing a textual subject that, unlike the traveler-protagonist, experiences what I shall call “homelessness” in discourse. Before locating this in the text, it would first be useful to briefly discuss contrasting theories of representation.

Jean Ricardou's work on the novel approaches the relationship between language and representation from an antireferential perspective. Fiction takes a stand against realism in contrast to Ian Watt's earlier account of the genre of the novel as a representation of reality. Ricardou writes that any narrative description, for example, the wedding cake in Madame Bovary, has two opposing tendencies, one referential in which the object is suggested as the sum of its parts and one literal in which it can be experienced only as the linear succession of its various components (34). Watt holds the opposite view that the novel, even if it looks very different in the forms given it by Sterne, Balzac, and Proust, is inherently referential: “The basic continuity of the tradition of the novel is made clearer if we remember that these differences in narrative method are differences of emphasis rather than of kind, and that they exist within a common allegiance to the formal or presentational realism which is typical of the novel genre as a whole” (294).

An engaging inquiry into the relationship of realism to Bakhtin's theory of dialogism is carried out by Ann Jefferson, who demonstrates that neither Watt's nor Ricardou's view actually helps us deal with the problem of reference. “Despite their many differences,” she explains, “both the realist and the reflexive theories of the novel imply a view of language as unitary and self-consistent—intrinsically ahistorical and fundamentally asocial” (171). Instead, she argues, Bakhtin's polyphonic conception of language allows one to sidestep the polarized issue of reference because the utterance is always socially constituted. Reference is still relevant, however, but only for its place “within the artistic organization of the ‘diversity of social speech types …, and individual voices’ and not as their pre-condition.”² Reference has a place in dialogism and an important one at that because “objects are themselves already inhabited by other discourses” so that it is this “very impulse toward the referential object that engenders dialogism” (177). Even in the nouveau roman, therefore, it becomes essential to pay close attention not to the rejection of the referent, but rather to the ways its supposed banishment from the text informs the enunciative model on which dialogism depends.

Representation in the nouveau roman, approached from Jefferson's perspective, becomes all the more problematic when one considers the proliferation and high visibility of the mise en abyme technique. Considered “one of the major modes of textual narcissism” (Hutcheon 4), the mise en abyme functions by means of “regress” through the embedding of one narrative level in another (Bal 146). It should be noted in passing, as Lucien Dällenbach points out, that this is not unique to the nouveau roman (152). Literature has always demonstrated a concern for the process of textual production, a preoccupation that develops in the nineteenth century into an exploration of representation that would seem to depersonalize a text. What the nouveau roman does, of course, is to push this to an extreme with an insistence on the mise en abyme technique so great that reflexivity itself fuses both story and narration.

In La Modification, simultaneity between diegesis and narrative act reflects a crisis in narrating. Enonciation takes on an urgency that produces dislocation: Léon Delmont tries to be in Paris via Rome and in Rome via Paris so that enunciative situations are superimposed on one another. As a result of this, from the opening of the novel, discourse is decentered as the protagonist searches for a voice that will enable him to be in several places at the same time. His use of the pronoun vous, a much-discussed technique that will be a focus of this essay, illustrates both an inability to speak in his own name and an enactment of discourse as dialogic address rather than monologic self-expression.

In the concluding chapters of the novel, when Léon expresses his desire to write a book, the frequent shift from vous to je has often been viewed as an emergence of self on the narrator's part. This general claim, I believe, is the result of allying oneself either with Watt or with Ricardou, whose

conceptions of language as a single and unified entity may be regarded from a Bakhtinian viewpoint as instances of the centripetal force of monoglossia since neither the referential language of realism, nor the self-generating signifiers of the nouveau roman acknowledge the existence of ideologies or of registers alien to their own.

(Jefferson 171–72)

It is here that I would like to begin my debate with existing scholarship on La Modification by bringing Jefferson's views, themselves an echo of Bakhtin's, into the conceptual framework of my analysis. I will argue that the narrator, in contrast to the protagonist, is as “homeless” at the end of Léon's trip as at the moment of his departure.

VOUS NARRATION: DISCOURSE AS HOMELESS

One of Léon's tasks in the story is to reconcile the myth of Rome with modern Paris. The desire to be a foreigner in his own city becomes an obsession so great that he sees Cécile as his vehicle for feeling Roman in his native Paris:

ce salut, Cécile, cette gorgée d'air, ce surcroît de forces, cette main secourable qui se tend vers vous messagère des régions heureuses et claires, depuis cette lourde ombre tracassière dont vous allez pouvoir enfin vous séparer de fait, cette magicienne qui par la grâce d'un seul de ses regards vous délivre de toute cette horrible caricature d'existence, vous rend à vous-même dans un bienfaisant oubli de ces meubles, de ces repas, de ce corps, tôt fané, de cette famille harassante. …

(40)

that refuge, Cécile, that breath of fresh air, that source of new strength, that rescuing hand reaching out to you from the realm of happiness and light, away from that oppressive, tiresome ghost whom you're about to shake off for good toward that enchantress who, with a single glance, can deliver you from this whole horrible caricature of an existence and make you yourself again, in blessed oblivion of all that furniture, of all those meals, of that prematurely faded body and that exasperating family. …

(29–30)

Cécile functions as the mediating force that he believes can restore him to himself (“vous rend à vous-même”). But this results from viewing the text only as an énoncé.³ For the mediating position in discourse is different from that in the story (Cécile in this instance, Rome in others) when both the énoncé and énonciation are examined for their point of intersection; it is at this juncture that what François Flahault calls la parole intermédiaire comes into view, a move revealing that “les individus ne sont pas maîtres d'opérer leur mise en place, puisque c'est au contraire cette mise en place qui établit leur identite” (Flahault 52).⁴ If the subject must submit to the mise en place (“setting into place”), then Léon's vous establishes just this power. It guarantees that he is perpetually in dialogue with himself and thus simultaneously both the addresser and addressee of his “own” discourse.

The unusual dynamic here between self and self as other produces the narrator as a split subject. This discursive fissure, however, should not be understood as analogous to the diegetic opposition between Rome and Paris on the one hand and between Cécile and Henriette on the other: in the role of protagonist Léon seeks to establish a distance between his Parisian and Roman “selves” while the divided self of the narrator is in no way sought out. This is significant because it indicates a lack of correlation between reality and thought, between a double life and a divided self.

Thus the narration is thrown into doubt because discourse is not mimetically faithful to the thought processes being portrayed. This would mean that the narration is homeless and not the traveler. For example, when Léon tries to capture the feeling of Rome in Paris, he goes to a small bathhouse of dubious cleanliness to remind him of the quaintly Italian Albergo Diurno, after which we learn: “vous avez profité du temps qui vous restait pour flâner un peu, tel un touriste romain à Paris, comme si c'était Rome votre habitation régulière et que vous ne vinssiez à Paris que de temps en temps” (61) (“you took advantage of the time you had left to stroll about like a Roman tourist in Paris, as if Rome were where you usually lived and you came to Paris only from time to time, every two months or once a month at most, on business” [48]). Léon's desire to replace the Parisian “self” with a Roman one is never realized; instead, this desire deviates, so that a fragmentation of the subject occurs discursively. Léon's voice continues to occupy simultaneously the positions of addresser and addressee, but the positions themselves become more problematic as each marks a desired distinction between a Roman and Parisian “self.” Thus the addresser sees himself with a dual identity and so too does the addressee. Positions have now multiplied and suggest the beginning of a crisis that will intensify in the duration of the text. The words “comme si” (“as if”) demonstrate the narrator's awareness of the lie he himself has fabricated but cannot control. When, back in Paris, Léon dines Italian style “pour prolonger cette impression de ne pas être tout à fait rentré” (62) (“in order to prolong that feeling that you hadn't really come home yet” [49]), discourse is indeed away from home because of the multiple positions occupied by Léon-narrator as the latter portrays Léon-character carrying out the desire to be a foreigner in his native city.

The entire novel becomes a study in the fulfillment of this desire to be displaced. The story fails to achieve such hoped-for fragmentation: at the end the protagonist does not reach his goal and so the narrator changes his scheme to the only one that can be realized: that is, the writing of a book that by all appearances is the one we hold in our hands, “ce livre futur et nécessaire dont vous tenez la forme dans votre main” (286). The narration metonymically fulfills this displacement where the story does not. The subject's language unconsciously produces meaning elsewhere than it intended. Lacan's theory of desire helps explain this unwilled turn of events: the inaccessibility of the signified, forever thwarted by the signifier from coming into view, means that the metonymic desire for something else is textual as well because of its driving insistence on the production of meaning (277). What happens in La Modification that is essential to a reading of the novel is that the displacement desired in plot occurs instead in discourse. It is here that the vous narration results in the splitting of the subject who is at once addresser and addressee. Indeed, this is the only place in the novel where he succeeds in being a foreigner in native territory, despite the fact that it is not at the level of discourse that he had resolved to perform this exile.

FURTHER DIVISION OF THE TEXTUAL SUBJECT: SEEKING THE MYTH OF ANCIENT ROME

The dynamic of this displacement of homelessness from plot to discourse requires a look at the nature of reality as imagined by the narrator: his conception of reality as something that he can manipulate backfires⁵ and proves a complicated illusion that effects a further decentering of discourse. The split subject of the narrator manifests itself in still other ways than that of a simultaneous addresser and addressee. The continual breaking down of the subject's voice is directly linked to the mise en abyme's regressive power to prevent the narrator from ever attaining a goal. The technique produces, on his part, an ongoing failure to control reality. For this reason the place of the myth of ancient Rome deserves special examination in this context.

As this myth unfolds in the course of the narrative, it aggravates the instability of the narrative voice. The numerous cultural forces at work in the novel are derived from intertextual references to Imperial Roman and Christian Renaissance painting, architecture, sculpture, and myth. When the two worlds dialogically collide to share the same physical space, such as when Léon strolls through galleries in the Louvre that contain art from both eras, the split subject of the narrator into addresser and addressee is divided further so as to place the reader in a similarly split position. Just as the museum's space is shared between two epochs, so too is Léon's vous occupied by both himself and the reader.

In his admiration for the paintings of Pannini, Léon is especially struck by the fact that his canvasses are representations of representations, depicting

deux collections imaginaires exposées dans de très hautes salles largement ouvertes où des personnages de qualité, ecclésiastiques ou gentilshommes, se promènent parmi les sculptures entre les murs couverts de paysages, en faisant des gestes d'admiration, d'intérêt, de surprise, de perplexité, comme les visiteurs dans la Sixtine, avec ceci de remarquable qu'il n'y a aucune différence de matière sensible entre les objects représentés comme réels et ceux représentés comme peints.

(64)

two imaginary art collections displayed in high, well-lighted rooms in which people of fashion, nobles and clergy were walking about among the sculptures, between the landscape-hung walls, making gestures of admiration, of interest, of surprise and bewilderment, like visitors to the Sistine, with this peculiarity, that there was no perceptible difference between things supposed to be real and things supposed to be painted.

(51)

On one level, the “personnages” fascinated by their surroundings function as an obvious mise en abyme of the narrator-protagonist of La Modification who is examining the paintings with great care. But on a second level, the reader is also represented observing the narrator watching his “personnages” watching, thereby implicating the reader's presence at the scene. If the narrator is at once spectator and spectacle, then the reader, invited into this text from its opening words, also occupies this same space alongside him. The reader thus becomes an essential part of the text, recalling Barthes's words: “Ce lecteur, il faut que je le cherche, (que je le ‘drague’), sans savoir où il est. Un espace de la jouissance est alors créé. Ce n'est pas la ‘personne’ de l'autre qui m'est nécessaire, c'est l'espace” (Plasir 11) (“I must seek out this reader [must ‘cruise’ him] without knowing where he is. A site of bliss is then created. It is not the reader's person that is necessary to me, it is this site” [Pleasure 4]). The reader has something the text wants, even if it means, as it does in this case, a further movement away from a subject-centered voice.

The desire for the reader indicated in Barthes's words is closely linked here to a need for mimesis. For if we accept that in the world of the text the reader is “real” and the narrator merely “represented,” then the latter's preference for Pannini's vision of ancient Rome over Michelangelo's Renaissance masterpiece in the Sistine Chapel because “there was no perceptible difference between things supposed to be real and things supposed to be painted” suggests that the narrator is searching for a convergence, a pictorial one between the real and the painted, and a textual one by a simultaneity between himself and the reader, the latter being forced to occupy the same position as the narrator because of the vous narration. This textual search for a convergence between narrator and reader can be read as the desire for collaboration between discourse and the real, so that the narrative voice should mimetically represent the diegesis. Although the previous section illustrated that this does not occur because the narration is in fact thrown into doubt where the story is not, the desire for a harmonious relationship between discourse and the real reveals the importance of the enunciative moment in the text.

The Pannini paintings become a mise en abyme of the utterance in two ways. On the one hand, as demonstrated above, they produce a textual mirroring of the enunciative situation between narrator and reader, two positions fused into one. On the other hand, the figures in the paintings stroll through ancient Rome, a scene that reflects Léon-narrator describing Léon-protagonist walking selectively around the Louvre to study only art representing the Roman Empire. This act parallels that of the reader observing the narrator's attempt to appropriate the myth of Rome to his Parisian self. These two ways in which Pannini's art produces a mise en abyme of the utterance place the reader both inside and outside the text, inside because the vous forces the reader to participate in the story and outside because the reader occupies the final position from which to view the narrator observing the protagonist observing Pannini's figures.

The vous that narrates this story is thus occupied by a variety of voices all at once. The one that Léon seeks out as his very own is not to be found, so that he is doomed to a polyphonic definition of self. The attempt to rid himself of the dispersion of the subject through a number of voices becomes the discursive equivalent of the protagonist's desire to revive the myth of ancient Rome. If Rome is “pour vous le lieu de l'authenticité” (146) (“the place where you felt genuine” [123]), it is not simply because it represents the origins of civilization as Léon sees it, but also because it holds the promise of an original, authentic voice, one that can truly be called one's own.

But the properties of dialogism render such a claim impossible. For as voices mutually condition one another, they are unable to function independently in any enunciative situation. In one of the few instances of direct discourse between the two lovers, Léon complains to Cécile: “Alors, si toi aussi tu te mets à compliquer les choses, comment pourrais-je avoir l'air naturel?” (185) (“So if you too are going to start making things complicated, how can I behave naturally?” [158]). John Frow's theory that the subject is not “the origin of utterance but its effect” (61) would necessarily eliminate any concept of the natural from the enunciative moment, making the constitution of the subject dependent on the particular situation. There is no place for the natural in la parole intermédiaire (“the intermediary world”) since the positioning that occurs therein is what constitutes the subject. Interestingly enough, Flahault explains that this concept is better understood by those raised outside of a bourgeois intellectual climate. As an example, he discusses Pierre Jakez-Hélias's Le Cheval d'orgueil in which Jakez-Hélias describes the ritual of Breton peasants visiting one another:

Que quelqu'un pénètre chez quelqu'un d'autre n'est pas dépourvu de sens et ne doit pas se dérouler “n'importe comment.” Le rituel permet à chacune des parties (visiteur et visité) de s'approcher en douceur, de savoir où elle en est par rapport à l'autre et de préserver la distance souhaitée par l'une sans blesser l'autre. Ce n'est pas que la bourgeoisie soit dépourvue de pareils rituels d'approche et de mise en place, mais elle les efface volontiers de sa conscience (notamment dans les milieux dits intellectuels) afin de pouvoir mieux développer l'idée d'une relation “naturelle.”

(52n)

The act of entering someone else's home is not without meaning and should not be taken lightly. The ritual permits each party (visitor and visited) to carefully approach one another, to know where one stands with regard to the other and to preserve the distance desired by one without hurting the other. It is not that the bourgeoisie are without such rituals of approach and placement, but they are fond of erasing them from their consciousness (particularly in so-called intellectual circles) in order to be able to develop better the idea of a “natural” relationship.

Léon Delmont, although purely fictitious, has all the bourgeois trappings one could imagine; indeed, one reason the reader is so immediately drawn into the text to share his vous is that the referents it designates (Parisian architecture and Roman ruins) are ones that so many readers of this novel are undoubtedly familiar with.⁶ If the idea of the natural is an intellectual invention, then Léon's “chasse au naturel” (“hunt for the natural”) is as much our problem as it is his. The difficulty lies not so much in the fact that there is no such thing as the natural, but, more significantly, in the fact that the reader, participating in the story alongside Léon, also experiences confusion at having to share a voice with another: for the reader and Léon-narrator, it is the decentering of the narrative voice that produces disorder, while for Léon-character it is the mission to transpose Rome to Paris, to make natural what is clearly unnatural.

The cultural world represented is one that is recognized by the reader at the same time that it cannot actually exist in the text itself. The significance of intertextual references to, for example, various Parisian and Roman monuments depends on the reader's own awareness of numerous cultural forces at work outside of the text. Claude Duchet writes that the exclusion of the referent from the textual world can be explained by “l'impossibilité du signe à être la chose même, mais il n'en peut devenir la représentation vraisemblable qu'en se fondant sur notre expérience et usage du monde” (450) (“the impossibility of the sign to be the thing itself, but it can become a good representation of that thing only in building on our experience and use of the world”). It could be said then that Léon depends on us to bring his dream to fruition. The vous in the text, clearly occupied by both Léon and the reader, needs this intimate coexistence. It becomes the bearer of textual dialogism because (1) it enacts the character interacting dialogically with himself as addresser and addressee and (2) it puts the reader into a similarly split position. Decentering of the subject is so aggravated by this intimacy that the vous ends up acquiring an alter ego in the narrator. Discursive homelessness is so complete that there are not only two split positions in the text, but an alter ego as well. The myth of Rome has set discourse into a spin. As it achieves momentum, decentering will occur in another form, that of the vous expressing itself as je.

JE NARRATION: DISCURSIVE HOMELESSNESS AND THE PROMISE OF THE BOOK

Towards the end of the novel, as Léon is recovering from nightmares about being cross-examined first by the Great Huntsman and second by various Roman ecclesiastics, the accelerated pace of the narration slows down as we observe him catch his breath. A significant change takes place in the aftermath of his judgment day: vous frequently shifts to je as the narrator abandons his plan to bring Cécile (and, consequently, Rome) to Paris and decides instead to write a book.

This new project has none of the uncertainties that characterized his original one. On the contrary, he discusses the protagonist (a close resemblance to himself) and storyline (the relationship between Paris and Rome) with confidence and resolution:

Il me faut écrire un livre.

(273)

Je ne puis espérer me sauver seul. Tout le sang, tout le sable de mes jours s'épuiserait en vain dans cet effort pour me consolider.

Donc préparer, permettre, par exemple au moyen d'un livre, à cette liberté future hors de notre portée, lui permettre, dans une mesure si infime soit-elle, de se constituer, de s'établir, c'est la seule possibilité pour moi de jouir au moins de son reflet tellement admirable et poignant.

(emphasis mine; 276)

I ought to write a book.

(238)

I cannot hope to save myself by myself. All the blood of my being, all the sand of my days would run out in vain in my effort to achieve integration.

So then, the only possible way for me to enjoy at least the reflected gleam—itself so wonderful, so thrilling—of that future liberty which is out of our reach would be to prepare the way for that liberty, to enable it in however minute a measure to take shape and substance by means of a book for instance.

(240)

Unable to integrate (“consolider”) his disparate voices without the help of the book, Léon gains the self-awareness that he is weak, the recognition of which is represented through repeated use of the first person. Similarly, his reference to the future takes the reader into account with the first-person plural, thus articulating with certainty the reader's presence in the text that the vous narration had only previously implied. What are the implications of such resolve for closure in this novel?

It has been argued that the fragmentation that occurs in the narration throughout the text ends when Léon decides to write his book. Helling has likened Léon to Aeneas so that the two heroes “changent et arrivent finalement à une certaine connaissance de soi, une connaissance qui décidera du reste de leurs vies” (74) (“change and finally attain a certain self-knowledge, a knowledge that will determine the rest of their lives”). This “connaissance de soi” suggests that La Modification promises a notion of self as unique and verifiable, characteristics that are eliminated, as demonstrated above, by the very properties of la parole intermédiaire.

In an analysis by Patricia Struebig, Léon has also been shown to reach a destination. She understands the act of his continually being displaced (or “modified” as the title directs us) as an indication that Léon is also simultaneously arriving somewhere. Again, homelessness is no longer an issue (54). Of greater interest is Bernard Pingaud's theory that the device of the vous, in addition to dividing itself into various voices, also functions as a way of joining the subjectivity of first-person narration with the objectivity of the third person. A compromise is effected because in this fusion of persons each is dissolved so that the vous can bridge the gap between them (94). Thus the tendency towards displacement of the subject is counteracted by the urge of je and il (“he”) to cohere as a vous.

Achieving a balance between these centripetal and centrifugal forces is obviously one way to define the relationship between self and other. Another version of this can be seen in Marianne Hirsch's view that self and other are both “decentralized” and “reconstituted” in Butor's work, that the city of Rome offers the model of a center that has lost its power so that all that remains is a “balance of points” around which self and other orient themselves.⁷ Thus, even with a “decentralized vision,” there appears to be movement towards a center, one that Hirsch conceptualizes as a “balance of points.”

Françoise Van Rossum-Guyon, too, in her comprehensive reading of La Modification, locates the constitution of the subject in the use of the first-person pronoun in the final chapter and the promise of the book. She notes that “this I is a vanquished one” because it has yet to, among other things, “master the images that assail it” (279–80). But her argument that the je at the end of the novel is not a triumphant one is weakened when she explains that this character discovers the one mechanism that can free him from this attack (281). If the search for a subject-centered voice is laid bare in La Modification, Rossum-Guyon, curiously enough, understands this activity to be a liberating force for the narrator (and the reader as well by implication) since his next move is to write a book. Although all of these discussions disagree about the location of closure in La Modification, there is general consensus that the je suggests an end to tensions between the various voices occupying the text so that a stable subject can emerge. Critical attention has been directed at the movement towards unification of the subject implied by Léon's future book. I argue that the je that speaks as a writer-to-be is in fact the most disembodied voice in the text.

The novel is permeated with obstacles to finding a voice and, despite the je that precipitates a decision to write, is no nearer its goal at the end of the text that at the beginning. The discussions referred to above demonstrate that the hero has reached his destination in the decision to become an author. But with regard to the narrator, he is still far from any goal. Indeed, the end of the novel can be read as throwing the narration into doubt once and for all, thus eliminating any sense of closure whatsoever.

The je that speaks at the end of the novel is one that has consciously decided to become the author of a story resembling the one just told us, with one important difference: this new version will admit the illusory nature of the “croyance secrète à un retour à la pax romana” (279) (“secret belief in a return to the Pax Romana” [243]) that lay beneath the present book. Acknowledging the power of history means accepting that the narrator's voice is not really his own, that it is technically a result of conditions beyond his control. In other words, the recognition of the fact that history will play a role in the future book presupposes that the decision to write the text is already an acceptance of the failure to find a voice.

Having been condemned by history for having ignored it,⁸ Léon is left with a “fissure béante en ma personne” (276) (“yawning fissure in my being” [240]). Looking to the future book as a way to close the fissure, he makes himself a character in his own fiction, thereby asserting his power as author. The idea of the legitimizing authority of writing to which critics owe their reading of La Modification as ending with the emergence of a subject suggests that the writer is an originator of meaning and invention, an idea poststructuralist thought has severely criticized. Foucault, for example, claims that “In writing, the point is not to manifest or exalt the act of writing, nor is it to pin a subject within language; it is, rather, a question of creating a space into which the writing subject constantly disappears” (142). And Derrida's definition of écriture has addressed the irresponsible, because infinitely repeatable, dimension of the sign in written language. An endless postponement of the present results in writing as absence. But, argues Foucault, to view writing as absence actually supports the notion of the writer's authority because it is “a simple repetition, in transcendental terms, of both the religious principle of inalterable and yet never fulfilled tradition, and the aesthetic principle of the work's survival, its perpetuation beyond the author's death, and its enigmatic excess in relation to him” (145). To keep from falling into this trap, as readers, he continues, we would do better to “locate the space left empty by the author's disappearance, follow the distribution of gaps and breaches, and watch for the openings that this disappearance uncovers” (145). In La Modification, I find this space precisely where the narration is not faithful to the thought processes it portrays. The text, obsessed with representing reality, comes undone when the je tries to take over. It is at this point that the narration, unlike Léon the traveler, remains homeless. Two passages should be examined in order to show this decisive move away from a center.

In the first, Léon wonders how he will depict the two cities in his life:

Il faudrait montrer dans ce livre le rôle que peut jouer Rome dans la vie d'un homme à Paris; on pourrait imaginer ces deux villes superposées l'une à l'autre, l'une souterraine par rapport à l'autre, avec des trappes de communication … le trajet d'un point à un autre serait modifié selon la connaissance, la familiarité que l'on aurait de cette autre ville, de telle sorte que toute localisation serait double, l'espace romain déformant plus ou moins pour chacun l'espace parisien, autorisant rencontres ou induisant en pièges.

(280)

This book should show the part Rome can play in the life of a man in Paris; the two cities might be imagined one above the other, one of them lying underground below the other, with communicating trip doors … the distance from one point to another, the way from one point to another would vary according to one's knowledge, the degree of one's familiarity with that other city, so that every man's consciousness of place would be twofold, and Rome would distort Paris to a greater or lesser degree for each individual, suggesting authentic or misleading parallels.

(243–44)

Rome and Paris, the two “centers” (he uses the word several times throughout the novel) in his life that have proven unreconcilable, are brought together here for the first time. However, only by imagining them superimposed in his book can they function simultaneously, although even then it is not a peaceful coexistence because one “deforms” the other, forcing the Parisian sphere into submission. To arrange centers so that “every man's consciousness of place would be twofold” removes from each respective space any ability to remain a center. The two spaces now exist as one “off-center” so that the logocentric desire for a center is built into the proposed book. The narrator of the present book thus only envisions himself as narrator of the next book in terms of the perpetual quest to make the text right itself. In Peter Brooks's words describing the course of desire as “totalizing in intent” and “tending toward combination in new unities,” this is “metonymy in search of metaphor” (106). The present narrator condemns himself to an eternal state of metonymic homelessness that offers no way out. Although Léon later rejects the idea of superimposing Paris and Rome, what he envisions next for his future book will actually aggravate, rather than alleviate, this troubled state.

He changes his mind in the last lines of the novel when he decides that

le mieux, sans doute, serait de conserver à ces deux villes leurs relations réelles et de tenter de faire revivre sur le mode de la lecture cet épisode crucial de votre aventure, le mouvement qui s'est produit dans votre esprit accompagnant le déplacement de votre corps d'une gare à l'autre à travers tous les paysages intermédiaires, vers ce livre futur et nécessaire dont vous tenez la forme dans votre main.

(285–86)

The best thing, surely, would be to reserve the actual geographical relationship between these two cities and to try to bring to life, in the form of literature, this crucial episode in your experience, the movement that went on in your mind while your body was being transferred from one station to another through all the intermediate landscapes, toward this book, this future necessary book of which you're holding in your hand the outward form.

(249)

This passage has been viewed as a positive step towards Léon's recovery. In what could be considered a Freudian analysis, Armine Kotin Mortimer finds that Rome represents the unconscious and Paris the conscious. An assertion of the distance between the two is a result of Léon's having enlisted the reader to help him carry out an analysis that brought the unconscious to the surface (183). Mortimer's study accounts for the addressee's presence and relationship to the narrator, but it cannot be forgotten that when the latter recovers from his crisis, he can only do so as a character in fiction in both the present text and the promised one, thus ensuring that he will continue to be subsumed by language. To complicate this “recovery,” the mise en abyme of the book as an object implicates an addressee whose presence is accentuated as the narrative develops. Plurivocal by nature, the mise en abyme decenters discourse for good in Léon's final lines.

To return to Foucault's concept that the death of the author should be viewed in terms of the “space into which the writing subject constantly disappears,” we next need to examine the space uncovered by this absence. He offers a clue to the composition of this space:

In a novel narrated in the first person, neither the first-person pronoun or the present indicative refers exactly either to the writer or to the moment in which he writes, but rather to an alter ego whose distance from the author varies, often changing in the course of the work. It would be just as wrong to equate the author with the real writer as to equate him with the fictitious speaker; the author function is carried out and operates in the scission itself, in this division and this distance.

(152)

The idea of an alter ego adds yet another voice to this text. If the je, that successfully speaks at the end of the novel creates space for an alter ego rather than a unification of self, then the death of the author gives birth not only to the reader,⁹ but also to this ego as alter.

The relationship between the author's effacement and the consequent positing of an alter ego illustrates how Léon's je does not possess the ability to represent an authentic voice. In the frenzied dream sequences, his continual interpellation of Janus, the double-faced god of beginnings, is evidence of his desire to reach an origin (as is the myth of ancient Rome), the ultimate place of authentication. But the position of author that he cultivates is actually his greatest obstacle to achieving a centered voice because, as Foucault explains, beginning in the seventeenth or eighteenth centuries the author-function in literature, unlike that in science, represented a multivoiced text that could not be reduced to any one particular voice (149–53). Historically speaking, then, the concept of the polyphonic emerges at the same time as that of the author. Because Léon ignores history, it follows that he would mistake the voice of an author as one that could lead to a place of authentication. Not accepting the implications of history for discourse excites the decentered voice rather than calms it.

Barthes accounts for the lack of a discursive center in terms of a war inherent in language. The writer becomes the site for battle, yet he is denied any fixed meaning therein. Barthes explains:

Comme créature de langage, l'écrivain est toujours pris dans la guerre des fictions (des parlers), mais il n'y est jamais qu'un jouet, puisque le langage qui le constitute (l'écriture) est toujours hors-lieu (atopique); par le simple effet de la polysémie (stade rudimentaire de l'écriture), l'engagement guerrier d'une parole littéraire est douteux dès son origine. L'écrivain … est nécessaire au sens (au combat), mais privé lui-même de sens fixe.

(Plaisir 57)

As a creature of language, the writer is always caught up in the war of fictions (jargon), but he is never anything but a plaything in it, since the language that constitutes him (writing) is always outside-of-place (atopic); by the simple effect of polysemy (rudimentary stage of writing), the warrior commitment of a literary dialect is dubious from its origin. The writer is … necessary to the meaning (the battle), but himself deprived of fixed meaning.

(34–35)

The recourse to the imagery of battle is a useful method for determining how different levels of the mise en abyme function in this text. Léon-narrator has made an enemy of himself by deciding to become a writer, a role that can be characterized by submission and, eventually, death. Furthermore, “Butor's” own position as a writer is now thrown into doubt so that even the putative starting point of the mise en abyme resists location. Léon, in opting for salvation through a book, actually maintains the unstable voice that permeates the present text. Far from mending the division of the self, the announcement of the future work guarantees the immutability of this fragmentation. The mise en abyme at work here in the dynamics between writer and character suggests that the process of gaining access only to fictional versions of himself is an interminable one. The impossibility of finding a subject-centered voice is thus inscribed in the text.

Continual displacement constitutes the self as a construction of textual relationships that has no beginning or end; writing is essentially a never-ending process in which an act of closure is forever subsumed by the promise of a text that inheres in Léon's je. Recalling Frow's words that the subject is not an origin, but an effect, I contend that this ungrounded configuration of the signifying chain produces anything but the emergence of a definitive self that so many critics have argued for.

The fact that the distance between ego and alter is in a continual state of flux recalls the conditions inherent in la parole intermédiaire. If positions are always shifting vis-à-vis one another, then the enunciative situation, even with the author laid to rest, is made up of voices that must be dynamic by nature. If La Modification does not see the emergence of a voice, it does affirm the enunciative model on which all discourse is founded. Discursive decentering may create permanent obstacles for Léon-narrator, but it ensures contact between addresser and addressee at the very least, thus securing communication. In a narrative that is left without a center, discursive homelessness seems the best place to settle in.

CONCLUSION

To read La Modification only as an énoncé involves the risk of being blind to the discursive decentering so central to the novel. Léon-narrator ends his search for a unified voice by emphasizing the role of the reader, thereby guaranteeing homelessness ad infinitum for the subject. The paradox lies in the fact that “in a world whose meanings are multiple and heterogeneous, complex and contradictory—in a world that, therefore, demands to be charted—the act of reading remains the only legitimate individual activity.” (Hirsch, “Decentralized” 348), even if the reader is “without history, biography, psychology” (Barthes, “Death” 148). This relentless insistence on the reader occurs at the expense of the author as a logocentrically defined concept.

By fashioning the promised book into an extension of himself, a way of prolonging the self, Léon can be heard metonymically. But as a whole, the subject cannot be located because the mise en abyme forces the stratification of meaning so that a subject-centered voice is beyond hope of being recuperated. Discursive decentering is metonymic rather than metaphoric because the text is all promise rather than revelation, processual rather than totalizing. Léon's voice remains disembodied because, to recall Lacan's theory of desire as metonymy (277), it is doomed to perpetual wandering out of desire that can never be fulfilled. Mieke Bal envisions the mise en abyme thus: “What is put into the perspective of infinite regress is not the totality of an image, but only a part of the text, or a certain aspect” (146; emphasis hers). Indeed, the “totality of an image” is not available to Léon, who may return home to Paris, but with no voice intact.

Notes

1. Valéry defined this term as “toutes croyances qui ont de commun l'oubli de la condition verbale de la littérature” (569) (“all beliefs that overlook the verbal condition of literature”).

2. Jefferson 172; emphasis mine. Jefferson quotes from Bakhtin here.

3. William Helling's comprehensive study of the intertextual correspondence between La Modification and The Aeneid falls into this trap. Had Helling studied the énonciation as well, I believe he would not have concluded that Léon is indeed worthy of being called a Virgilian hero.

4. “[I]ndividuals are not masters of performing their own placement in space, since, on the contrary, it is this setting into place that establishes their identity” (my translation). Flahault's study of discourse keeps language from being reduced to a mere object of communication. Emphasis is placed on the intersubjective moment in which the speaking subject is constituted according to the positions it attributes to an other.

5. For example, he believes his failed lunch Italian style at a Parisian café can be compensated for by lighting the last of his Italian cigarettes, only to have it quickly put out by the rain.

6. Gerald Prince notes that the narratee, even if seemingly impersonal (“the degree zero narratee”) “connaît la langue, le(s) langage(s) de celui qui raconte. Dans son cas, connaître une langue, c'est connaître les dénotations—les signifiés en tant que tels et, s'il y a lieu, les référents—de tous les signes qui la constituent” (180) (“knows the universal language, the language[s] of he who is doing the telling. For him, to know a language is to know the denotations—the signifieds as such and, if need be, the referents—of all the signs that constitute the language”).

7. Hirsch, “Decentralized” 326. The concept of a “balance of points” is borrowed from an interview that Hirsch held with Butor entitled “An Interview with Michel Butor” (272).

8. At the end of his dream, Léon, suspended in the air, listens with terror as the King of the Last Judgment identifies those accusing him: “Au simple son de mes paroles, tes membres commencent à se convulser, comme déjà dévorés de vers. Ce n'est pas moi qui te condamne, ce sont tous ceux qui m'accompagnent et leurs ancêtres, ce sont tous ceux qui t'accompagnent et leurs enfants” (263) (“At the mere sound of my words your limbs are beginning to twitch convulsively, as if worms were devouring them already. It is not I who pronounce judgment upon you but all those who are with me and their ancestors, all those who are with you and their children” [227]). The history of Christianity now weighs on him as it delivers its sentence, which, in the waking world, will result in the book.

9. The death of the author allows the reader not only access to a text, but free access because the reader can gain entry while disregarding the signified. Barthes explains: “Thus is revealed the total existence of writing: a text is made of multiple writings, drawn from many cultures and entering into mutual relations of dialogue, parody, contestation, but there is one place where this multiplicity is focused and that place is the reader, not, as was hitherto said, the author” (Barthes, “Death” 148).

I would like to thank Ross Chambers for his comments on an earlier version of this essay.