[In the following essay, Farrant discusses Balzac's shorter works—stories, articles, and fragments—and their relationship to and development into his longer works, exploring in the course of the discussion the idea of brevity and the role of the fragmented in Balzac's fiction.]

Balzac, on paper and in person, has a reputation for being long. Here is one sleepless victim of the author:

La nuit qui suivit cette soirée un peu tendue, se passa dans une chambre à deux lits, offerte aux deux visiteurs; au lieu d'écrire quelques notes, Balzac se plut à exprimer à son compagnon avec une verve et un brio intarissables, les impressions qu'il venait de recevoir; il eût fallu sténographier!; M. Loiseau répétait souvent qu'il n'avait jamais moins dormi, ni plus admiré son cher compatriote, que pendant cette conférence si spontanée et si prolongée.¹

Not all Balzac's audiences have been as delighted. The duchesse de Dino, who had been treated to him earlier that evening, found him ‘sans verve ni facilité dans la conversation. Il y est même très lourd … Il vise à l'extraordinaire et raconte de lui-même mille choses auxquelles je ne crois nullement’.² ‘C'est un bavard’, declared Delacroix.³

Balzac is generally thought to be long without good reason; this study hopes to turn that notion on its head. It explores Balzac's short stories in the light of their genesis, as individual fictional entities, in relation to others, and in the context of his work's overall development. Short stories make up over half La Comédie humaine, in addition to the thirty published Contes drolatiques, and scores of other narratives and newspaper articles.⁴ Balzac's writing career began with short fiction—the first trace of narrative in his work is an anecdote⁵—and ended with it, to all intents and purposes, in what are vastly expanded stories, Le Cousin Pons and La Cousine Bette. ‘Tout s'agrandit sous ma plume’:⁶ virtually everything in La Comédie humaine either is short fiction, or grew from it. Beyond his short stories proper, brief narrative is ubiquitous: in the anecdotes of his parodic guides to human behaviour, the Code des gens honnêtes and the Physiologie du mariage; in stories framed within others (Une conversation entre onze heures et minuit, Autre étude de femme), or in works like Le Médecin de campagne or Albert Savarus; in article-series such as Les Français peints par eux-mêmes, Petites Misères de la vie conjugale, or Le Diable à Paris; in articles concatenated into stories (Les Comédiens sans le savoir), stories concatenated into novels (Même histoire, now La Femme de trente ans), and novels linked into the vast frescoes which are Illusions perdues and Splendeurs et misères des courtisanes.

Such narratives are hardly brief; but Balzac's view of brevity is very large. Untroubled by the anachronistic anglophone term ‘short story’,⁷ with its regrettable intimations of inferiority, and undesirable invitation to start counting words,⁸ he took the brevity of contes or nouvelles as read. Yet Balzac the jobbing journalist and sometime printer did calculate numbers of pages, even of typographical characters. In this respect, textual length was a practical matter, of quantity of copy and of payment. At this basic level, contes or nouvelles were units of currency—ones which, like everything he touched, were subject to inflation: David Séchard, for example, proposed as a nouvelle, eventually formed the last third of Illusions perdues.⁹ A working definition for Balzac's stories at the height of his activity as conteur (1829-32, when about half those in the Comédie humaine were written and published) might be set at two-and-a-half feuilles, or forty octavo pages,¹⁰ his limit with the Revue de Paris (Corr. ii. 174-6), which published most of them. But, as Balzac, and the press, evolved, the boundaries shifted: his later projections of the number of typographical characters (Corr. iii. 533, 536; iv. 33) are significant less as quantitative absolutes, than as revealing reflections of the crucial junctures at which he made them. In 1839 and 1840, the newspaper serialized novel was just beginning to take off, and to extend beyond the five- or six-episode nouvelles which had hitherto marked the outer limit of most feuilletons.¹¹ And it was precisely at this time that Balzac was working on one such work, Une fille d'Eve (a nouvelle which stretches to thirteen instalments in Le Siècle), and that he first envisages a collection of oeuvres complètes called ‘la Comédie humaine’ (Corr. iv. 35). Une fille d'Eve is a cas-limite, one of the longer works considered here in detail. But it is chosen less for its length, than as symptomatic of larger tensions between story and novel, brevity and extension, wholes and parts, macrocosm and microcosm, explored also in other texts such as La Peau de chagrin or Albert Savarus.

Such tensions form some of this book's central concerns. These are less with theoretical definition (which is ultimately subjective)¹² than with exploring Balzac's conception and practice of short fiction—both his individual stories, and shorter fiction as grille d'écriture and de lecture for his opus. Its title phrase, ‘shorter fictions’, points to these tensions, and to the aporia inherent in any study of genre: discussing genres implies that they can be absolutely distinguished, yet they are invariably mixed; conversely, to speak of mixing implies the reverse.¹³ It is impossible to solve this paradox; but it is the motor which drives this study forward. What matters is not the boundary-line, but the relative status of the realms concerned. Genre, fundamentally pre-reflective,¹⁴ is a question of conception—the writer's, and the reader's: the expression of an intention, and a set of pointers towards reception.¹⁵ Hence the linking of genesis and genre.¹⁶ My subject is not only Balzac's ‘short stories’,¹⁷ but also other brief forms, articles, and fragments; their relationship to, and development into, even the longest novels; and ultimately, the philosophical question of brevity. And my aim differs from that of much contemporary genetic criticism in being less to question the finality of the printed text, than to follow some of the steps which led to it, and to explore the role of the provisional, the fragmentary, and the virtual in the final work.¹⁸

Balzac gives only two substantive theoretical discussions of short fiction: the 1831 Théorie du conte (which is as much an autobiographical self-examination as a theory), and his remarks in the 1840 Lettres sur la littérature. Yet they leave his regard for the genre in no doubt. The conte, ‘la plus haute expression de la littérature’, is a traditional genre, the embodiment of authenticity and truth:

Ceux qui ont conté sont rares, bien conté, on les compte, et ce sont des hommes de génie—Lucien—Pétrone—Les Fabliaux (autores incertos)—Rabelais—Verville—Boccace—L'Arioste—La Fontaine—Voltaire—Walter-Scott—Et la reine de Navarre! … Hamilton—Sterne—Cervantès—et Le Sage donc?¹⁹

Le Conte, cette magnifique, cette puissante forme de la pensée humaine et qui va si loin, témoin Peau d'âne, Barbe bleue, La Courtisane amoureuse, Roméo et Juliette, porte avec lui quelque secret quand il a conquis la vie refusée à tant d'œuvres.²⁰

La Comédie humaine is scattered with vestigial references to this corpus, which the Contes drolatiques seek to emulate, and works such as Eugénie Grandet or La Vieille Fille exploit.²¹ For Balzac, the conte is a vital form containing the object lessons of human experience. Ceci n'est pas un conte is ‘un des grands morceaux de l'histoire du coeur humain’; Manon Lescaut, Adolphe, Werther, and René ‘donnent la clé de presque toutes les situations du coeur humain en amour’. Its weakness is a potential for triviality: ‘un livre comme Volupté a plus de chances de vie littéraire que les bijoux dont je vous parle [Musset's Frédéric et Bernerette] … Conter pour conter est l'arabesque littéraire; mais l'arabesque n'est un chef-d'œuvre que sous le pinceau de Raphaël’. Stories must have an exemplary quality, an illustrative function, present in all great examples of the genre:

En terminant ces six nouvelles, on se demande comme le mathématicien: ‘Qu'est-ce que cela prouve? A-t-on voulu prouver quelque chose: Y a-t-il quelque grand et vaste symbole comme dans Adolphe, comme dans Paul et Virginie, comme dans telle page qui devient un monument au milieu des ruines d'une littérature?²²

Demonstrated by Balzac's own practice (in, for example, the 1830 Scènes de la vie privée), this illustrative function is linked to his concern with examples, types, and systems. To be successful, Musset should have ‘élevé chacune de ces narrations à la hauteur où elles deviennent typiques’ and

présenté l'un de ces sens généraux auxquels s'attachent invinciblement les coeurs … Ceux qui sont doués de talents rares, comme l'est M. de Musset, doivent étudier les causes de ces phénomènes de l'esprit humain, afin d'en augmenter la divine nomenclature.

For Balzac, the conte is the Urform of literature.

Few today would think of Scott, Sterne, or Lesage as conteurs, or of Manon Lescaut or Adolphe (which Constant calls a ‘petit roman’) as nouvelles.²³ But Balzac saw the conte, and conter, as transcendent, capable of absorbing diverse discourses and of near-infinite variety, via contes philosophiques, bruns, or drolatiques—as something, indeed, which rather resembles the Bakhtinian novel. In this, he celebrates a polyvalence which contemporaries, wearied by the craze for stories known as the folie du conte,²⁴ implicitly acknowledged, but sought to deny. ‘Jamais siècle n'a été aussi engloutisseur que le nôtre’ noted La Charge;²⁵ for Le Conteur, ‘le conte … n'est en somme que le squelette d'un roman’; for the Journal des débats the nouvelle was ‘un roman au petit pied’, an ‘improvisation de roman’.²⁶ ‘Ah! s'il était possible de l'allonger, de l'amincir, de l'étendre à l'infini, comme une feuille d'or sous le marteau du batteur, il n'y aurait pas de contes; on les laisserait à Voltaire; il n'y aurait que des romans’, exclaimed Nisard; ‘mais le conte contemporain n'est pas une feuille d'or’.²⁷ Yet Balzac was a goldsmith: the voice of the conteur dominates in even the most fully realized novels of La Comédie humaine²⁸; it is through him that all else is relayed.

However, the conte is not the same as the novel, even if made of similar stuff. As well as embodying archetypal wisdom,²⁹ the various types of short fiction share an ubiquitous but under-acknowledged concern with form which determines even his longest works. ‘Quel homme eût été Balzac’, wrote Flaubert, ‘s'il eût su écrire! Mais il ne lui a manqué que cela. Un artiste, après tout, n'aurait pas tant fait, n'aurait pas eu cette ampleur.’³⁰ Janin declared Illusions perdues to be written ‘au hasard, sans but, sans plan’; ‘un écrivain n'est pas un chiffonier, un livre ne se remplit pas comme une hotte’.³¹ Yet for Balzac, form is not as a mere envelope for the material, or the superficial effect of a cause, but the internal and shaping dynamic which of necessity brings together cause and effect.³² In Illusions perdues, Vernou ‘porte des articles, fera toujours des articles, et rien que des articles … Félicien est incapable de concevoir une oeuvre, d'en disposer les masses, d'en réunir harmonieusement les personnages dans un plan qui commence, se noue et marche vers un fait capital.’³³ D'Arthez's comments are just that: a thèse d'art. Balzac sees form as integral to conception: his concern with unity of form finds its due expression in his compositional practice. His habits of meticulous redrafting and recasting manifest a will to unity and a purposiveness engendered by his engagement with, and practice of, short fiction, and which shapes even his longest novels.

Such stress on form is, then, a salient feature of short fiction, and of Balzac's fiction tout court. Concentration, which he repeatedly advocates in the first half of his career,³⁴ becomes, in the second, a demand for unity. On the model of Ceci n'est pas un conte, the conte must be ‘simple, vrai, complet’(CHH xxviii. 134); Ourliac's Suzanne, derived from Diderot's work, uses too much extraneous material: ‘Diderot, en grand artiste, n'a pas donné les antécédents de Gardeil et de Mademoiselle Delachaux … Le génie de l'artiste consiste à choisir les circonstances naturelles qui deviennent les éléments du Vrai littéraire, et, s'il ne les soude pas bien, si ces métaux ne font pas une statue d'un beau ton, d'un seul jet, eh! bien, l'œuvre est manquée’ (CHH xxviii. 136). Unity, together with justifying causation and drama, are the criteria which produce the ideal composition:

La nature n'a pas besoin de livre, le fait est expliqué par cela seul qu'il est. Pour le faire passer de l'action vivante à l'action probable d'un livre, l'écrivain doit nous en montrer toutes les racines.

(CHH xxviii. 136-7)

Balzac's criteria seem contradictory: on the one hand, praise for an absence of explanation in Diderot; on the other, a demand for it elsewhere.

This contradiction is only partly explained by the fact that Balzac is differentiating between developments possible in novels, but not, perhaps, in stories: it is, in fact, inherent in the ambiguous generic status of the conte and the nouvelle, and innate in the emergence of the Renaissance (or Tuscan) nouvelle from the conte. Where the conte shapes the world as we would like it to be (or in such a way as to warn us that it is not thus), the nouvelle, a more modern form, starts from a real, or supposedly real, incident to exemplify a human truth.³⁵ The world of the conte is self-contained, circular, and on occasion fantastic,³⁶ bearing an archetypal truth, often stated in a separate moral, and, as its name suggests, echoing or replicating the oral transmission of wisdom; the nouvelle, in contrast, exemplifies the real world, stresses less narration than event, is linear, and directed towards its ending; its moral is inferred.³⁷ The conte obeys an internal logic in which the very absence of the extraordinary might seem incredible; the nouvelle provides what is a proto-realistic exploration of the world.³⁸ Most of the stories of the founding modern collection, The Decameron, can be seen as contes, but some are more morally problematic, and offer no unambiguous interpretation.³⁹ The apparent ambiguity of Balzac's comments parallels that of short fiction itself; and the conte's idealization of its subject matter is embodied in the philosophical concerns of the Études philosophiques, conceived mostly in the earlier part of his career, whilst the nouvelle's protorealistic concern with realia, developed in canonical psychological explorations such as La Princesse de Clèves, gives rise to the stories of the 1830 Scènes de la vie privée, and ultimately to the Études de mœurs.

If the burden of Balzac's work thus gradually shifts from the older form of the conte to the more modern nouvelle, striking realities, dramatically presented, characterize many of the short fictions of La Comédie humaine. Balzac praises such qualities in his review of Musset: ‘L'histoire, car de telles compositions arrivent à la valeur de l'histoire, est d'un dramatique horrible, d'une épouvantable vérité, d'un sens cruel, et, par-dessus tout, amusante’ (CHH xxviii. 143-4). A deftly exploited historical context, a focus on moments of terror, are two prominent elements identified by Bourget in Balzac's stories,⁴⁰ although they certainly do not account for all of them; and Balzac's stress on unity anticipates Poe's suggestively loose definition of the story as something which could be read at one sitting⁴¹ (even if Balzac could sit longer and more busily than most, breaking his chair on more than one occasion: LH i. 99). Many of his stories contain a central symbol, a Heyserian falcon,⁴² a Tieckian turning-point⁴³ (often a revelation of identity—the discovery that Maximilien, a shop boy, is really an aristocrat in Le Bal de Sceaux, Chabert's exposure of his wife's identity in Le Colonel Chabert); or exemplify Goethe's demand for ‘eine sich ereignete unerhörte Begebenheit’,⁴⁴ Storm's description of the Novelle as ‘die Schwester des Dramas’, or Chekhov's comparison of it to the stage.⁴⁵ There is a fundamental unity of effect in Le Colonel Chabert or Le Curé de Tours; a central symbol or imagic nexus co-ordinates, for example, La Paix du ménage, La Maison du chat-qui-pelote, or La Fille aux yeux d'or—devices carried over in the determinant symbols of longer works (La Peau de chagrin, Le Lys dans la vallée, La Rabouilleuse). Mystery, drama, often of the most lurid kind, pervade the early stories especially—El Verdugo, Un épisode sous la Terreur, L'Auberge rouge, and La Grande Bretèche.

Yet there is drama and there are dramas; cause, and causes; effect, and effects. The intense impact of El Verdugo is very different from the theatrical dialogue of a later work like La Fausse Maîtresse, or the spectacular realm of Gaudissart II. After Le Père Goriot and the middle of Balzac's career, the systematic deployment of recurring characters represents a change in the teleology of his creation. Unity itself becomes problematic—a setting of the unity of the individual story against that of the work as a whole. The early stories of the Études philosophiques, most of them written between 1830 and 1832, are testbench examinations of philosophical causes, the ‘sens caché’ (x. 1210) behind events: Adieu, L'Élixir de longue vie, Le Chef-d'œuvre inconnu. The real is a means of questing the absolute, more than a subject of enquiry in itself. From 1833, with the Études de mœurs, the tables turn: the ‘effets’, ‘plus considérables que les causes’ (ibid.), become the primary study of his opus.

This has two results. First, it leads to an extensiveness which takes the form of longer fictions assembled from shorter ones (Même histoire in 1834; from 1836, Illusions perdues) and of a continuing production of stories (nearly all, with the significant exceptions of La Confidence des Ruggieri, Gambara, and Massimilla Doni, outside the Études philosophiques) which present the wider personnel of La Comédie humaine: La Messe de l'athée, Z. Marcas, Un prince de la Bohème, La Fausse Maîtresse. Secondly, it creates a confusion of cause and effect: L'Interdiction explores the effects of rapacity in an unhappy marriage; yet the episode it relates is also a cause of continuing rancour, evoked as late as Splendeurs et misères des courtisanes (vi. 514, 720); Les Secrets de la princesse de Cadignan is both an effect of the creation of its protagonists in Le Cabinet des Antiques and Illusions perdues, but also, in the new illumination it brings to them, a cause.

Such circumstances make the unity of the individual story rather more questionable than it might be in the case of other contemporary writers of freestanding narratives, such as Mérimée, or Gautier. Indeed, one of Balzac's characteristics is to diversify the forms of brevity, to subvert categories in the very act of their creation. Little given to ‘la Théorie dont la parole est toujours au futur’ (v. 720), he had little time for its hierarchies:

Descartes établit des divisions, des nomenclatures; j'aurais mieux aimé le voir s'occuper de choses plus intéressantes … rien n'est si facile que de classer les fleurs, mais celui qui cherche à deviner la cause de leurs couleurs, de leur naissance, leur principe de vie, est le véritable botaniste …

Je ne prétends pas faire le procès de Descartes: il faut des divisions, mais il y a fagots et fagots, dirait Sganarelle.

(OD i. 576 7)

In his discussion of Musset quoted above, he slips seamlessly from ‘ces six nouvelles’ to the ex-cathedra statement ‘Conter pour conter est l'arabesque littéraire’. Balzac's flexible terminology, his use without explanation of conte, nouvelle, histoire, and the like, reflects the strikingly imprecise, often deliberately aberrant usage of his age, with its challenge to Classical hierarchies of genre. ‘Les règles sont en littérature ce que sont les lois en morale: elles ne peuvent tout prévoir’.⁴⁶ In this area, as in the matter of length, the desire for precision can lead to confusion,⁴⁷ so I attempt no more than to follow these general pointers in my use of terms like anecdote, conte, nouvelle, and roman.⁴⁸ The anecdote and conte derive from oral narrative; the anecdote, generally the shortest, is often told to illustrate another discourse; the conte is generally longer and exemplary, whilst the anecdote's stress on event and ending⁴⁹ is shared with the nouvelle. Baudelaire's oft-quoted remark on the pre-eminence of form in the nouvelle (‘Le roman … est un genre bâtard dont le domaine est vraiment sans limites … La nouvelle, plus resserrée, plus condensée, jouit des bénéfices éternels de la contrainte’)⁵⁰ is still a useful characterization of differences between it and the novel, but its Poe-inspired emphasis on unity as a primarily aesthetic consideration (‘rien ne se perd de la totalité de l'effet’) appears somewhat anachronistic, given Balzac's conception of unity as an altogether more substantive affair of character, event, and philosophy. Simpler, but closer to the mark, is Eikenbaum's view of the nouvelle as ‘une forme fondamentale élémentaire’ and of the novel as ‘une forme syncrétique’;⁵¹ but it is still rather too reductive to be applied to Balzac.

Beyond such broad statements, then, there can be no absolute or universally applicable distinction between genres. To relate systematically the formal features of Balzac's work to the generic or structural typologies of writers such as Jolles or Propp⁵² would be to produce a typology tied to an extraneous hierarchy, rather than to the internal imperatives of Balzac's work, to that evolving relationship between genesis and genre which is the study of ‘le véritable botaniste’. More seriously, it would overlook the central fact that such contradictions result from attempts to isolate as absolutes generic features which operate contextually and synergically, which ask to be seen in conjunction with each other⁵³—a particularly grave error in relation to Balzac, whose work is both a cathedral of categories, yet also their carnival,⁵⁴ and in which genesis and evolution are indissociable from genre.⁵⁵ The bafflement of a Godenne at such diversity (‘La nouvelle française, c'est tout à la fois une histoire de quelques pages … une histoire aux dimensions plus importantes … un récit sérieux, grave, dramatique, et un récit plaisant, comique, un récit de caractère oral ou non, une histoire vraie, une histoire fantastique’)⁵⁶, is merely a failure to recognize a critical vice as a creative virtue.⁵⁷ Balzac's ambition is to ‘représenter toutes les formes littéraires’ (xi. 474; cf. vii. 54); I include as many kinds of shorter fiction as possible, to acknowledge the mobility of Balzac's novels and stories, written at a time when notions of genre were notably in flux. Newspaper articles like La Dernière Revue de Napoléon could be expanded into a story and concatenated with others to become what, faute de mieux, we must call a novel, Même histoire; the story Les Deux Musiciens turned into Le Cousin Pons.

The interanimation of Balzac's creation, of genesis and genre, makes it as pertinent to say what Balzac's fictions do, as what they are. Apart from the conte philosophique Corsino (1819), and the tête de roman or proto-nouvelle Une heure de ma vie (1821-2), the few Balzac short fictions written prior to 1829 form brief illustrative anecdotes or exempla in longer works (such as the Discours sur l'immortalité de l'âme, the Code des gens honnêtes, or the Physiologie du mariage), embodying a fundamentally exemplary, demonstrative, even didactic role which his fiction was never entirely to abandon. But the 1829 publishing crisis made bringing out longer works difficult, and was soon to lead to the folie du conte. Between 1829 and 1832, Balzac wrote about half the total number of stories in La Comédie humaine, nearly all of which were kept in print and ultimately included in it. Thereafter, the conte, especially the Renaissance conte, genuine or reproduction, fundamentally or vestigially oral in nature, continued to have for him the totemic status and proselytizing function as the vessel of authenticity and repository of archetypal national wisdom which is embodied in the uncompleted project for ten dixains of Contes drolatiques, begun in 1831, effectively abandoned six years later, but still being discussed at the end of his career. Most of the other contes in La Comédie humaine are included within framed narratives (La Grande Bretèche, Le Médecin de campagne), often fragmented or problematized (L'Auberge rouge, La Grande Bretèche), but none is explicitly called a tale: Balzac's frame stories or first-person narratives, free-standing or extracted from novels, were invariably entitled histoires (Histoire du Capitaine Beauvoir, Histoire de Napoléon racontée dans une grange, Histoire d'une clarinette, Histoire intellectuelle de Louis Lambert), connoting both objectivity and authentic first-person or first-hand experience. It is extraordinary to note that a storyteller as prolific as Balzac rarely published in isolation stories labelled simply conte (coming closest to doing so in contes fantastiques such as Zéro, Tout, and Le Chef-d'œuvre inconnu). The term conte was reserved for cycles of stories (Contes bruns, Contes artistes, Romans et contes philosophiques) denoting some thematic and human community of experience, and often characterized by other features, unity of action, imagery, and symbolism, developed in Balzac's nouvelles and novels.

If the conte forms the first broad category of Balzac's fictions, the nouvelle might be a second, emerging in 1829 from the condensing of the historical novel and its conflation with other genres—the conte moral, visual art, and satirical dialogue scènes.⁵⁸ The first series of Scènes de la vie privée embody the story as exemplary, socially engaged form, ‘composé en haine des sots livres [i.e. contes moraux] que des esprits mesquins ont présentés aux femmes jusqu'à ce jour’(Préface, I. 1173).⁵⁹ But Balzac's nouvelles are always more complex than Eikenbaum's ‘forme fondamentale élémentaire’: these scènes adopt a belt-and-braces approach, having both the explicit didacticism of the conte, expressed in an appended moral, and the inwardness of the nouvelle, embodied in the action. They tell by recounting and warning of danger, in the way of the conte moral (‘marquer d'une branche de saule les passages dangereux de la vie’), yet also tell by showing ‘le tableau vrai des moeurs que les familles ensevelissent aujourd'hui dans l'ombre’ (i. 1173). In 1832, Les Célibataires, now Le Curé de Tours, develops Balzac's use of the nouvelle in a penetrating psychological exploration of the experience of the individual. From it, via notably La Femme abandonnée and Louis Lambert, would emerge Balzac's first major novels.

The distinction between conte and nouvelle thus becomes largely redundant, other than as an extraneous category used in private communications with publishers, but (with the signal exception of the Contes drolatiques) hardly in public labelling of his work. For the whole drive of Balzac's enterprise is to transcend the conventional generic boundaries which limit narrative fiction to women and children, or, as Stendhal put it more disparagingly, ‘femmes de chambre’,⁶⁰ to make the ‘ouvrages improprement appelés Romans’ with which Balzac's Scènes are associated (I 1175) bear the weight of philosophy and history. From early on, in his project for an Histoire de la France pittoresque, a cycle of novels running from the Middle Ages to the present, via Le Dernier Chouan, he aimed to subsume narrative, histoire, within History—the ambition still voiced in the 1842 Avant-propos to La Comédie humaine (i. 11). Balzac's scènes, a genre brought to prominence by the Scènes de la vie privée in 1829, and the principal generic denominator of his creation, are much more than simply longer, or more objective, contes. Whilst contes narrate archetypal events, Balzac's scènes are primarily concerned with the characters' emotions and experience; by its veracity and exemplary validity in the context of contemporary society. Even the stories which do rely on the sensational place it in an emotional, usually family, context: La Vendetta, Adieu, El Verdugo, L'Élixir de longue vie, Le Réquisitionnaire, La Grande Bretèche. It is this increasing understanding of, and weight given to, characters' feelings which enables Balzac to develop La Transaction from a Gothic tale of burial alive into the quest for emotional wholeness which is Le Colonel Chabert, or to move from the merely sympathetic treatment of the protagonists of La Femme abandonnée and La Grenadière to the full-blown empathetic development of the protagonists' subjectivity in texts like Eugénie Grandet or La Duchesse de Langeais, to an identification with the characters which is the prerogative of the novel, to the engagement of substantive individuals with each other. Balzac's generic boundaries are indeterminate, but suggestive. If he prefers the implicit objectivity of scène or étude to the fictional (and, for other writers, often negative) connotations of conte, nouvelle, or roman, story and novel are held in tension, as are fiction and reality in La Comédie humaine.

These tensions, together with Balzac's consistently genetic, evolutionary presentation of his work in the Introductions to the Études de mœurs and the Études philosophiques, and above all in the Avant-propos, invite us to eschew the sometimes arcane typologies of previous studies,⁶¹ and to approach his stories and novels as symbiotically as they grew. This is why, seeking to work with the material, I have chosen a broadly diachronic approach, covering every completed self-contained story in La Comédie humaine, some unfinished works, and touching on pivotal novels such as Le Dernier Chouan and La Peau de chagrin. This means as much looking at the way genres interact within individual texts, as sketching their contemporary background; for this interaction is the name of the game.⁶² Florence Goyet has distinguished late nineteeth-century nouvelles from novels on the basis of monology: the first are monologic, the latter are not. This argument may hold for other writers, but probably not for Balzac.⁶³ Balzac's writing begins with a dialectical ambition, in his Discours sur l'immortalité de l'âme; but it almost immediately becomes dialogic, in his various ‘asides’ and ‘voices off’ (like the remark on Descartes quoted above), and in the very first anecdote in the Discours (OD i. 555), explored in Chapter I; Balzac entertains diverse viewpoints, and in his fictions tends to transcend the narratorial distance which Goyet (and implicitly others, like Du Bos and Fernandez)⁶⁴ find characteristic of the récit or the nouvelle. Balzac's shorter fiction emerges from this proto-dialogic hesitation, in this Discours and the Physiologie, and carries it throughout his career;⁶⁵ it is an inevitable effect of his plurivocal œuvre. Yet his later absolutist conception of a transcendent unity in his work and in the world, of a whole composed of parts, ‘un et varié’ (Massimilla Doni, x. 593), knowable in its totality solely by God (or the author), visible only partially, as fragments, to fallen creatures (or readers) via the illusory relativism of human viewpoints, ultimately makes his dialogism as problematic as his synthesis is virtual.⁶⁶

This ‘novelistic’, ‘dialogic’ quality is perhaps implied by A. J. George's unappetizing description of Balzac's stories as ‘dehydrated’ novels;⁶⁷ but in terms of chronological development, the truth is rather the reverse. There is little evidence that the techniques of Balzac's novels had much bearing on his short narrative before 1829, when the failure of Le Dernier Chouan, the first novel he published under his own name, produced a cure d'amaigrissement which resulted in his first breakthrough as a story-writer, his first series of stories, the 1830 Scènes de la vie privée.⁶⁸La Peau de chagrin, his first success as a novelist, is a development of the story, as are his canonical nouvelles Le Colonel Chabert and Le Curé de Tours, and the first great novels of his maturity, Eugénie Grandet and Le Père Goriot, created by a ‘lateral’ expansion, via subplots and description, of the fundamentally ‘linear’ diachronic narrative of the story. Eugénie Grandet and Le Père Goriot exemplify a major aspect of this symbiosis, emerging in tandem with the stories L'Illustre Gaudissart and Un drame au bord de la mer, like Madame Firmiani and Le Colonel Chabert before them, or, subsequently, César Birotteau and La Maison Nucingen, Une fille d'Eve and Béatrix, and La Cousine Bette and Le Cousin Pons.

From the purely illustrative function of the philosophical exemplum in the Discours, Balzac's early fiction evolves via attempts at short fiction, Corsino and Une heure de ma vie, towards a more mimetically credible treatment of freestanding reality in his early novels. But this is only brought into focus in the combination of doxa and exemplary narrative in the 1825 Code des gens honnêtes. Chapter 2 shows how this treatment develops through the anecdotes of the second, 1829 version of the Physiologie du mariage, an elaboration and undermining of the prescriptive discourse of the first version of 1826. This undermining of the exemplary was to be at the centre of Balzac's work: Chapters 3-5 cover the high noon of his career as a writer of short stories, during the 1829-31 crisis in novel-publishing and the subsequent folie du conte. Chapter 3 shows how, in the Scènes de la vie privée, Balzac moves from a narratorially centred historical account via other genres, the conte moral, the contemporary satirical scène, and the historical novel, to a more rounded and more ‘embodied’ treatment of reality. In this period, Balzac's fiction develops via his early journalism—in his development of narrative from pen-portraits in, for example, Gobseck and Étude de femme, and in small-readership magazines and satirical sheets (Chapter 4). Stories like Sarrasine or Le Chef-d'œuvre inconnu, which have since become canonical, owe much to the publishing environment in which they emerged, and in which I have attempted to resituate them (Chapter 5). Seen in this context, they, along with others, such as the Contes drolatiques or Maître Cornélius, can resume their ideological purpose as would-be renewers of the Renaissance, or Hoffmannesque contes artistes promoting the artist as an absolute romantic, and by implication, absolute anti-bourgeois. La Peau de chagrin's synthesis of these genres is a major advance, the creation of an identity of microcosm and macrocosm, part and whole, which was to underpin Balzac's aesthetic, but one whose potential was not fully to be achieved until later (Chapter 6); for the time being, oral narrative and the frame story embody a threatened social cohesiveness of which the Études de mœurs, and ultimately La Comédie humaine, were to be the analogue. Via the myriad contemporary species of the conte—conte brun, conte drolatique, conte artiste, conte fantastique—Balzac, dubbed ‘roi de la nouvelle’ in January 1832, moves to more considered psychological treatment of its subjects in Le Colonel Chabert and Le Curé de Tours (Chapter 7). If these mark the pinnacle of his first period as a storyteller, they also introduce the hesitant self-examinings of the first-person narratives of 1833, Louis Lambert and Le Médecin de campagne, and the more thoroughgoing psychological investigations of La Grenadière, La Femme abandonnée, Eugénie Grandet, and La Duchesse de Langeais, explored in Chapter 8. With Le Père Goriot, Balzac completes an evolution from predicative to psychological,⁶⁹ from objective to empathetic narration, from conte to novel.

The story, of course, is not that simple. Evolution is too facile a model to account for the complexity of genre in Balzac's fiction. As well as being diachronic, it is also, perhaps even more, synchronic, and it is from 1834, the year of Le Père Goriot, that this becomes most pointedly apparent. The systematic reappearance of characters initiated by this novel renders redundant the cycle of stories hitherto at the centre of Balzac's creation, the Contes drolatiques, where recurring characters originated,⁷⁰ the consequence of the creation of the Études de mœurs au XIXème siècle and the Études philosophiques (1833 and 1834), successors to the first Scènes de la vie privée and Romans et contes philosophiques (1830 and 1831). If the core of these new series is initially formed by earlier fictions, mainly stories, they are soon joined by others, affected by their dichotomous, paradoxical relationship to narratives elsewhere in the creation. Whilst Balzac's novels continue to expand short fictions, building narrative around a single major protagonist (La Vieille Fille, César Birotteau), single-protagonist stories, explored in Chapter 9, La Messe de l'athée, Facino Cane, are characterized by a kind of ‘tangentiality’, a ‘centrelessness’ which makes clear their dependence on tales or events elsewhere. It is a decentredness which will become more common as Balzac develops the system of recurring characters, in stories like La Maison Nucingen (Chapter 10), Les Secrets de la princesse de Cadigman, which recount secrets de polichinelle (Chapter 11), Z. Marcas, which alludes to a political doctrine never revealed, or Un prince de la Bohème or La Fausse Maîtresse (Chapter 12), which are about love affairs which never really occur. Driven by Balzac's burgeoning treatment of the contemporary, it is a development which spells the effective end of the Contes drolatiques, whose 1837 third dixain has little of the abandon of its two predecessors.

In this synchronic symbiosis, Balzac's 1839 preface to Une fille d'Eve (Chapter 11) marks, after Le Père Goriot, an important second stage. Its emphasis on his creation as a mosaic implies the interrelationship of its parts—characters and narratives—to each other, and the often disordered and anachronistic way in which this may emerge. We may encounter the middle or the end of a story before its beginning, or the established Rastignac before meeting the student. This is partly a response to circumstance: Balzac's frenzied and simultaneous composition of different fictions, the uncertainties of book and newspaper publications, and the pressure of writing for the ill-fated enterprises he engaged in, in order to escape them: his inordinate contributions to his newspapers (the Chronique de Paris) and single-handed writing of the Revue parisienne (1840) gave rise to short fictions like La Messe de l'athée and Z. Marcas. But there are also aesthetic reasons for the development. As Balzac's creation grew, his short fictions in particular assume unprecedented complexity (Un prince de la Bohème).

The 1842 Avant-propos (Chapter 13) confirms the tendencies announced in 1839, mixing perspectives and generic frames, at once handing narratorial control to an embodied creation and voicing the author's own views more powerfully than anywhere else. As the Comédie humaine emerges, the independence of its fictional world becomes ever more apparent: Balzac's most important new short fictions are told by recurring characters (Honorine, Autre étude de femme), a factor which, in Un homme d'affaires, creates a complexity verging on incomprehensibility. Yet the whole project of publishing a Comédie humaine, rather than just Œuvres complètes, emerges from short fiction, via the publisher Hetzel and a series of articles, Scènes de la vie privée et publique des animaux; and the fragmentation implied by the mosaic evoked in Une fille d'Eve's preface is created both by the disordered appearance of the Comédie in instalments, and by the intermittent publication of Balzac's articles and pen-portraits in newspapers and collective works such as Le Diable à Paris. His last stories, Un homme d'affaires and Les Comédiens sans le savoir, are largely concatenations of such articles, embodying in their form the modern metropolitan fragmentation they describe. Their piecemeal, alternatory techniques are reflected in the quick-fire sequences of contemporary serial novels, La Cousine Bette and Splendeurs et misères.

But older forms of Balzac's fiction also re-emerge and are reinvented towards the end of his career. The early 1840s see a series of ‘supernovellas’, Un début dans la vie, La Muse du département, Modeste Mignon, developed from short stories, under the influence of his sister Laure and Madame Hanska, and with an implicit moral aim (chapter 14). It is this augmented nouvelle form to which Balzac turns in his final diptych, Les Parents pauvres (La Cousine Bette and Le Cousin Pons), which combine the cogency of short fiction with the linearity, fragmentation, and eclecticism of the serialized novel. ‘Tout est bilatéral dans le domaine de la pensée’ declares Illusions perdues (v. 457); ‘Tout est double, même la vertu’, the preface to Les Parents pauvres (vii. 54). Balzac's genres are oxymorons,⁷¹ holding contrary elements in tension. The subjective is implicit in the objective, the fantastic in the real, the literal in the literary—and vice versa; the particular in the general, unity in variety, the part in the whole; immensity, import, in the exiguous; brevity, the story, in the novel. Read with focus, Balzac's fictions, even the longest, may seem shorter than we thought.

Notes

1. Loiseau d'Entraigues, letter to Spoelberch de Lovenjoul, May 1891: Imprudence et bonheur, Lov. A 109, fo. 72^r. The letter erroneously places the visit in 1833; it in fact occurred on 26 Nov. 1836 (P, 298).

2. Duchesse de Dino: chronique de 1831 à 1862 (1909), ii. 108-9: P 298.

3. Delacroix, Correspondance générale, 22 June 1842, ii. 112, cited by R. Pierrot, Corr. iv. 460.

4. On Balzac's unfinished work and journalism, cf. T. Takayama, Les Ouvres romanesques avortées de Balzac (1829-1842) (Tokyo and Paris, 1966); R. Chollet, Balzac journaliste (1983) and OD i and ii, passim; and F. Schuerewegen, ‘Avortements’, in S. Vachon (ed.), Balzac: une poétique du roman (Montreal and Saint Denis, 1996), 307-21.

5. Discours sur l'immortalité de l'âme, OD i. 555. Cf. Ch. I.

6. LH i. 304, 16 Aug. 1846.

7. Established by Brander Matthews's article ‘The Philosophy of the Short Story’, The London Saturday Review (5 July 1884); Lippincott's Magazine (July 1885) (cf. F. Garcier, ‘Du nom au genre: le cas de la short story’, La Licorne, ‘Dynamique des genres’ (1992), esp. 19-20; repr. in C. E. May, The New Short Story Theories (Athens, Oh., 1994), 73-80).

8. Forster classes novels as not less than 50,000 words (Aspects of the Novel (Harmondsworth, 1966), 13; do nouvelles stop at 49,999?), Engstrom the nouvelle as a vaguer ‘two hundred and fifty average thirty-line pages’; whilst McCormick and Des Loges restrict themselves to 50-60 and a maximum of 100 pages respectively (A. G. Engstrom, ‘The Formal Short story in France and its Development before 1850’, Studies in Philology, 42 (1945), 630; D. F. McCormick, Les Nouvelles de Balzac (1973), 14; S. Des Loges, L'Art structural de la narration dans la nouvelle de Balzac (Wroclaw, Warsaw, Cracow, 1967)). Such definitions are largely irrelevant to matter apprehended not arithmetically, but formally and linguistically.

9. In 1839, Le Siècle had exchanged Mémoires d'une jeune mariée [sic] for ‘six nouvelles inédites, dont trois d'environ soixante colonnes de quarante lignes environ chacune et les trois autres de quarante colonnes environ chacune’ (Corr. iii. 605-6); and, on 6 Oct. 1841, Balzac told Le Musée des familles that if the nouvelle David Séchard exceeded 3,000 lines, ‘je la remplacerais par une autre intitulée les Parisiens en province’ (Corr. iv. 321), a title eventually employed in La Comédie humaine for l'Illustre Gaudissart and La Muse du département.

10. A feuille is a printed sheet, the equivalent of sixteen octavo pages; a feuillet, that of one printed octavo page.

11. Anticipating the length of Véronique, now part of Le Curé de village, Balzac wondered ‘si vous avez à la Presse assez de caractère pour composer toute ma nouvelle, qui sera d'une étendue de dix feuilletons’ (May 1839, Corr. iii. 605), whilst the following year its administrator, Dujarier, calculated that ‘les 50 feuilletons que vous devez publier … formeront au moins huit nouvelles, soit environ six feuilletons pour chacune d'elles’ (Corr. iv. 42). Indeed, at this time, overlength was more likely to be a problem: Nettement writes that the editor of La Gazette de France ‘a peur d'un roman en trente feuilletons’ (24 May 1840, Corr. iv. 122). But, by 1843, length has become a virtue: ‘vos quatre cent mille lettres ont enlevé la question’ wrote J.-A. David, accepting Esther ou les amours d'un vieux banquier, which actually occupied 39 instalments, for Le Parisien (16 May 1843, Corr. iv. 602), and even articles had grown: Ce qui plaît aux parisiennes (Philosophie de la vie conjugale à Paris) was 3 folios, 48 octavo pages, paid 600 frs—considerably longer, and more, than Balzac's first articles for Le Diable à Paris (12 Dec. 1843, Corr. iv. 632). Cf. R. Guise, ‘Balzac et le roman-feuilleton’, AB (1964), esp. 298-9; ‘Le roman-feuilleton (1830-1848): la Naissance d'un genre’, Doctorat d'État thesis, Nancy, 1975, repr. Lille, 1985, pt. I, chs. 4-6.

12. See R. Godenne: ‘Il serait vain de réduire la nouvelle, ce que l'on fait trop souvent, à un type de récit unique’ (Études sur la nouvelle française (Geneva, 1985), 9). Cf. also H. Steinhauer, ‘Towards a Definition of the Novella’, Seminar, 6 (1970), 154-74; V. Shaw, The Short Story: A Critical Introduction (London, 1983), vi, 8; M. L. Pratt, ‘The Short Story. The Long and the Short of It’, Poetics, 10: 2-3 (1981), 175-94, reprinted in May, The New Short Story Theories 91-113; A.-M. Baron (ed.), Balzac et la nouvelle, L'École des lettres, 13 (1998-9), 3-17.

13. J. Derrida, Parages (1986), 253, cit. D. Combe, Les Genres littéraires (1992), 148.

14. Combe, Les Genres littéraires, 13 et seq.

15. J.-M. Schaeffer, Qu'est-ce qu'un genre littéraire? (1989), esp. 136-55.

16. For an overview of Balzac's manuscript corpus, but with a more globally synchronic approach than is taken here, see S. Vachon, ‘Les Enseignements des manuscrits d'Honoré de Balzac’, Genesis (1997).

17. Cf. A. W. Raitt (ed.), Balzac: Short Stories (Oxford, 1964 and reprints).

18. Cf. e.g. M. Contat and D. Ferrer (eds.), Pourquoi la critique génétique (1998), 9. If the conventional historical approach thus falls into a ‘piège téléologique’ (A. Grésillon, Eléments de critique génétique (1994), 18), at least it has some hope of being well anchored.

19. Les Cent Contes: Théorie du conte (hereafter: Théorie du conte), OD i. 518; Pensées, sujets, fragments, Lov. A 182, fo.11^r, PSF fo. 17, CHH xxviii. 670, cit. OD i. 1155; Balzac had deleted ‘Marmontel pour mémoire’ after Scott. The list is repeated virtually verbatim, but with the addition of Aesop and ‘les Arabes inconnus des Mille et Une Nuits’ in Petites Misères de la vie conjugale, xii. 107. Balzac wrote an eulogious Notice sur la vie de La Fontaine in 1825 (OD ii. 141-6).

20. Lettres sur la littérature, le théâtre et les arts, Revue parisienne, 3 (25 Sept. 1840), CHH xxviii. 145-6; Balzac has in mind Bandello's novella (Novelle, ii. 9), rather than Shakespeare's tragedy. Though he cites Shakespeare's play more frequently, he has high esteem for its source: vii. 53. Two of these three Lettres, along with other critical texts, have been anthologized and annotated by S. Vachon: Écrits sur le roman (2000).

21. There are some 39 references to La Fontaine in La Comédie humaine, 30 to Marguerite de Navarre, 14 to Boccaccio (xii. 1716; 1584, 1741-2, 1621); but a complete study of the role in Balzac's work of this traditional literature of the conte remains to be made. Cf. R. Chollet, R. Guise, and N. Mozet, OD i. 1148-58; J. Gurkin, ‘Romance elements in Eugénie Grandet’, L'Esprit créateur, 7 (Spring 1967); N. Mozet, iv. 806.

22. Lettres sur la littérature, Revue parisienne (25 Sept. 1840), CHH xxviii. 134, 135, 144-5, 145-6, 144.

23. Constant, Adolphe, ed. J.-H. Bornecque, 2nd edn. (1968), preface, 3, 5; Balzac calls it variously a ‘petit roman’, a ‘roman’, and a ‘nouvelle’ (ii. 699, iv. 775, 765), but there are reasons for seeing it as a récit: cf. C. Du Bos, cit. Ch. 10(v) below.

24. The folie du conte reached its zenith in mid-1832, beginning to decline after 1833, though there are still complaints about the conte's ubiquity in 1835. Cf. R. Guise, ‘Le Roman-feuilleton’, pt. I, ch. 3; P. Berthier, La Presse littéraire et dramatique au début de la monarchie de juillet (1830-1836) (Villeneuve-d'Ascq, 1997), pp. 1119-33, esp. 1129; F. P. Bowman, ‘La Nouvelle en 1832’, CAIEF 12 (1975) esp. 191-2, and ‘The Splendors and Miseries of Reading beyond the Canon’, in W. Paulson, Les Genres de l'hénaurme siècle (Ann Arbor, 1989), 3-13.

25. La Charge ou les folies contemporaines (28 Apr. 1833): Guise, ‘Le Roman-feuilleton’, pt. I, ch. 3, 142-3.

26. Le Conteur, recueil de contes de tous les temps et de tous les pays [1832], i. 9, 13; Journal des débats (29 June 1833): Guise, ‘Le Roman-feuilleton’, pt. I, ch. 3, 185.

27. D. Nisard, ‘D'un commencement de réaction contre la littérature facile’, RP 57 (Dec. 1833), 211-18, at 217: Guise, ‘Le Roman-feuilleton’, 186-7. The deprecation of Nisard's final phrase relates to the moral more than the literary qualities of the conte.

28. Cf. R. Fernandez, cit. Du Bos, Ch. 10(v); and, on the Bakhtinian interpretation and the determinant role of the narrator, E. Bordas, Balzac, discours et détours (Toulouse, 1997), esp. 154-5 et seq. But Bakhtin seems to be anticipated by Blondet: ‘Notre jeune littérature procède par tableaux où se concentrent tous les genres, la comédie et le drame, les descriptions, les caractères, le dialogue sertis par les noeuds brillants d'une intrigue intéressante. Le roman, qui veut le sentiment, le style et l'image, est la création moderne la plus immense’ (Illusions perdues, v. 459).

29. A notion shared by Grimm and Arnim (A. Jolles, Formes simples (1972), 175-9.). Cf., inter alia, B. Bettelheim, Psychanalyse des contes de fées (1976), 58-67 and 119-22; S. Loiseau, Les Pouvoirs du conte (1992); C. Velay-Vallantin, L'Histoire des contes (1992); and N. Belmont, Poétique du conte (1999), ch. 6.

30. Letter to Louis Colet, 16 Dec. 1852, in Flaubert, Correspondance, ed. J. Bruneau and B. Masson (1998), 214.

31. J. Janin, ‘Un Grand homme de province à Paris’, RP 7 (July 1839), 145-78, cit. E. Zola (‘Jules Janin et Balzac’, Le Roman expérimental, ed. A. Guedj (1971), 316, 318). Zola elsewhere declares: ‘Il a pourtant, dans les Contes drolatiques, donné des pages qui sont des bijoux de ciselure; je ne sais rien de plus joliment inventé comme forme, ni de plus finement exécuté … D'ailleurs, il a beau s'embarquer dans des phrases fâcheuses, son style est toujours à lui. Il le pétrit, le refond, le refait entièrement à chacun de ses romans. Sans cesse il cherche une forme’ (‘Du Le Roman expérimental, 223).

32. Cf. M. Andréoli: ‘La forme constitue le centre, l'espace privilégié de la fusion nécessaire entre la cause et l'effet’; it is a ‘manifestation sensible de l'idée’ (Le Système balzacien (Paris and Lille, 1984), 678). In pointing to the dominance of the shaping narratorial dynamic, Andréoli and Bordas, from their different viewpoints, are saying similar things; cf. n. 28 above.

33. v. 427; cf. Le Cabinet des Antiques, preface, iv. 963-4.

34. Cf. Notice sur la vie de La Fontaine, OD ii. 146; Le Gars, Avertissement, viii. 1675; La Peau de chagrin, x. 51; Les Cent Contes drolatiques, OD i. 150, 251, 315.

35. Jolles, Formes simples, 180 et seq.

36. R. Godenne, ‘A propos de quelques textes critiques du XXème siècle sur la nouvelle’, CAIEF 27 (1975), 249.

37. Balzac's view seems to anticipate Flaubert's wish that the novel should ‘faire rêver’: the ‘arabesque’ of the conte ‘fait songer, comme la fumée du cigare qu'on brûle’: CHH xxviii. 145.

38. Jolles, Formes simples, 180.

39. Cf. e.g. Decameron, i. iii and ii. vii: J.-P. Aubrit, Le Conte et la nouvelle (1997), 108-9. The ambiguity inherent in the pointe is exemplified by nouvelles 5 and 6 of the Heptaméron, which invite at least two contradictory interpretations, and found in exempla as early as Seneca: cf. P. de Lajarte, ‘La Nouvelle aux frontières du commentaire et du dialogue dans L'Heptaméron de Marguerite de Navarre’, and A. Cullière, ‘De la controverse à la nouvelle. Alexandre Van den Bussche, lecteur de Sénèque’, in V. Engel and M. Guissard (eds.), La Nouvelle de langue française aux frontières des autres genres, du Moyen Age à nos jours (Ottignies, 1997), i. 106 et seq. and 46.

40. Études et portraits (1889-1906), iii. 250-7. Bourget is discussing El Verdugo, Un épisode sous la Terreur, Le Réquisitionnaire, and La Grande Bretèche.

41. The Philosophy of Composition (1846), which even suggests that a long work cannot be unified (E. A. Poe, Selected Writings, ed. D. Galloway (Harmondsworth, 1984), 482); cf. Twice-Told Tales, ibid. 445-6.

42. For Heyse, the Novelle requires ‘eine starke, deutliche Silhouette … deren Umriss, in wenigen Worten vorgetragen, schon einen charakteristischen Eindruck marke’, finding a perfect example in the Decameron, v. ix (Jugenderinnerungen und Bekenntnisse (Berlin, 1900), 348).

43. For Tieck, the Novelle's ‘sonderbaren, auffallenden Wendepunkt’ distinguished it from all other kinds of tale: Schriften (Berlin, 1829), xi, p. lxxxvii.

44. ‘Was ist eine Novelle anders als eine sich ereignete unerhörte Begebenheit?’: J.-P. Eckermann, Gespräche mit Goethe in den letzten Jahren seines Lebens (25 Jan. 1827), cit. R. Paulin, The Brief Compass (Oxford, 1985), 93.

45. Storm, Novelle, ed. J. Kunz (Darmstadt, 1973), 72: Paulin, Brief Compass, 7; Chekhov, letter to Pleshcheyev, 30 Sept. 1889: May, New Short Story Theories, 195.

46. V. Hugo, Odes et Ballades, ed. P. Albony (1969), 31n. (1824 Preface); S. Nash, ‘Victor Hugo's Odes et ballades and the Romantic Lyric’, in Paulson (ed.), Les Genres de l'hénaurme siècle, 73 95, and Bowman, ‘La Nouvelle en 1832’, 193, 206. Balzac lampoons such usage in Les Litanies romantiques (OD ii. 826), and acknowledges it in Illusions perdues: ‘J'ai lu L'Archer de Charles IX’ says Petit-Claud, ‘c'est plus qu'un ouvrage, c'est un livre!’(v. 661).

47. A. Engstrom's system of classification causes him to conclude that Balzac wrote only nine contes (two of which, L'Idée et le fait and Cambremer's tale, can only by saved by wresting them from their narrative frames in L'Auberge rouge and Un drame au bord de la mer), and Nodier only one (‘The Formal Short Story’, 630); Desloges's attempt to organize Balzac's narratives by quantity, quality, structure, type, and plot creates so many categories that her entire study becomes one of classification. R. Godenne (‘A Propos de quelques textes critiques sur la nouvelle’), produces a similarly inconclusive list.

48. This approach, whilst open to the danger of anachronism, is intended to be synchronic, seeking both to historicize generic concepts and to be accessible to present-day readers.

49. Cf. Jolles, ‘Le Cas’, Formes simples, 137 57.

50. Baudelaire, ‘Théophile Gautier [IV]’, Curiosités esthétiques: l'art romantique, ed. H. Lemaître (1962, rev. 1986), 677-8.

51. B. Eikenbaum, ‘Sur la théorie de la prose’ in T. Todorov and R. Jakobson (eds.), Théorie de la littérature (1965), 202.

52. V. Propp, Morphologie du conte (1970). J. Gurkin, ‘Romance elements in Eugénie Grandet’, suggests the potential of such readings.

53. Cf. Schaeffer, Qu'est-ce qu'un genre littéraire?, 74-8.

54. Cf. Andréoli, Le Système balzacien, 25 6, 33 5. For a robust appraisal of the readings of Balzac ‘contre Balzac’ instituted by L. Dällenbach, ‘La Comédie humaine et l'opération de lecture’, ‘Du Fragment au cosmos’, and ‘Le Tout en morceaux’, Poétique (1979 and 1980), and marked notably by N. Mozet, Balzac au pluriel (1990), 287 et seq., and F. Schuerewegen, Balzac contre Balzac (Paris and Toronto, 1992), see M. Andréoli, Lectures et mythes (1999), 26.

55. Schaeffer's scepticism about the application of evolutionary models to questions of genre Qu'est-ce qu'un genre littéraire?, 58 9, 71 2) is doubtless justified—texts, unlike biological species, do not reproduce themselves. But Balzac explicitly invites an evolutionary approach (Avant-propos, i. 7), and his texts are, to a degree, self-generating, determined by a creative imperative shared with other genres (cf. e.g. Chs. 2 and 3 below on the shared origins of the Physiologie du mariage and the Scènes de la vie privée, or Ch. 8 on Eugénie Grandet and L'Illustre Gaudissart). Balzac's work is a matter of competition and progression between different genres (pace S. Vachon, ‘Balzac nouvellier’ in Baron (ed.), Balzac et la nouvelle, 27).

56. R. Godenne, Études sur la nouvelle française (Geneva, 1985), 9.

57. Cf. Schaeffer, Qu'est-ce qu'un genre littéraire?, 79.

58. I omit the stories of uncertain authorship Le Pacte (1822) and Le Droit d'aînesse. Cf. Ch. 1(iii) and Mozet, Balzac au pluriel, 243-6.

59. Balzac was still targeting the genre as late as 1848, in La Femme auteur, written partly as an attack on writers such as Albertine Necker and Sophie Ulliac de Trémadeure, whose L'Éducation progressive and Le Petit Bossu et la famille du sabotier had in 1828 and 1834 won the Montyon Prize for virtue which Le Médecin de campagne was denied: M. Regard, xii. 598-9.

60. ‘Projet d'un article sur Le Rouge et le noir’, in Le Rouge et le noir, ed. P.-G. Castex (1973), 719.

61. Notably Des Loges, L'Art structural de la narration dans la nouvelle de Balzac, and McCormick, Les Nouvelles de Balzac, and M.-C. Amblard, L'Œuvre fantastique de Balzac (1972).

62. S. Swahn, Balzac et le merveilleux (Lund, 1991), takes a step in this direction, in the light of the fairy-tale.

63. F. Goyet, La Nouvelle, 1870-1925 (1993), esp. chs. 3 and 4. Goyet's view contrasts with A. Vial's conception of the nouvelle as a hybrid of conte and novel (Guy de Maupassant et l'art du roman (1954), 435-506); cf. also Bakhtin's ambivalence on this matter in relation to Balzac: Problems of Dostoevsky's Poetics, trans. R. W. Rotsel (Ann Arbor, c.1973), 29.

64. Cf. Ch. 10(v) n. 78 below.

65. Bahktin's account of the parodic-travestying literature of ancient Rome as ‘unstable, compositionally still unshaped, lacking a firm or definite generic skeleton … something like an immense novel, multi-generic, multi-styled, mercilessly critical, soberly mocking, reflecting in all its fullness the heteroglossia and multiple voices of a given culture, people and epoch’, along with the importance of the mosaic, seems like a description of Balzac's enterprise avant la lettre, as characterized in Complaintes satiriques sur les moeurs du temps présent and the preface to Une fille d'Eve. (M. Bakhtin, ‘From the Prehistory of Novelistic Discourse’, in The Dialogic Imagination, trans. M. Holquist (London, 1981), 59-60, 69; cf. below, Chs. 6 and 10(iii).)

66. Cf. Andréoli, Le Système balzacien, ch. 2, esp. 109 et seq. The debate about the unity or diversity of Balzac's work, and the associated question of the relation between monology and pluralism, constitute two of the main currents in contemporary Balzac criticism. Cf. particularly Andréoli, Le Système balzacien, and A. Vanoncini, Figures de la modernité (1984), esp. 76-7, 138-40, 177, 186, 195-9; views crystallized in their articles ‘La Représentation de l'utopie dans Le Médecin de campagne, and ‘Le Médecin de campagne: idéologie et narration’, AB (1988 and 1989), and to some extent synthesized in Andréoli's Lectures et mythes: see esp. 343-61.

67. A. J. George, Short Fiction in France, 1800-1850 (Syracuse, NY, 1964), 101; echoed by A.-M. Baron's view of them as ‘modèles réduits de l'ensemble’ (Balzac et la nouvelle, 17).

68. This is why I give only very brief consideration to the romans de jeunesse, on which cf. M. Bardèche, Balzac romancier (1940), and P. Barbéris, Aux sources de Balzac (1965) and Balzac et le mal du siècle (1970), and A. Lorant (ed.), Premiers Romans (1999).

69. Cf. T. Todorov, ‘La Grammaire du récit’, in Poétique de la prose (1978), 47 57, at 49 et seq.

70. Argow, pirate-protagonist of Le Vicaire des Ardennes (1822) had reappeared in Annette et le criminel (1824), but Balzac did not develop the device at that stage.

71. Cf. J.-L. Diaz, ‘Balzac-oxymore: logiques balzaciennes de la contradiction’, RSH 44 (July-Sept. 1979).

References

References to La Comédie humaine designate the ‘new’ Pléiade edition, edited by P.-G. Castex et al. (1976-81) and reprints, supplemented, in the same series, by Œuvres diverses, ed. R. Chollet, R. Guise, and N. Mozet (1990, 1996); they take the form of volume and page numbers bracketed without identifying title, e.g. (i. 63), except in the case of Œuvres diverses, where the volume number is preceded by the abbreviation OD. Where a textual variant is concerned, it is designated by a lower-case letter after the volume and page number. Thus (iv. 44a) refers to vol. iv, variant a to p. 244; variants and notes appear at the end of each Pléiade volume.

The place of publication of all works cited is Paris, unless otherwise stated.

Other editions are designated as follows:

In addition, the following abbreviations have been adopted: