But British chauvinism isn’t the only thing challenged by Lafayette’s novel. Brief reflection is enough to bring into focus the oddity of an ancestor who is separated from her descendants by so many generations. After all, it is not as if the French novel in the century between Lafayette and Balzac looked anything like La Princesse de Clèves, either. On the contrary, with a few exceptions it looked very much like its counterpart across the Channel: it was dominated by the pseudofactual forms of the memoir novel and the letter novel. For its brevity, its psychology, its modern values, La Princesse de Clèves no doubt makes a convincing “first novel.” But is it really, in spite of appearances, fiction? A positive answer entails explaining why so much time would go by before novelists imitated Lafayette’s example by inventing protagonists without asserting their existence. A negative answer frees us from that tricky task, but imposes another, that of facing up to the possibility that at least some great books are not harbingers but hapaxes. This is, I will argue, the case with La Princesse de Clèves. It is an occupational hazard of literary historians to regard the history they study as an evolution driven by innovative masterworks or, at the very least, exemplified by those masterworks. Lafayette’s novel was a lasting success at home and abroad; its patently invented heroine spurred readers to envision their relations with characters in a totally new light—to identify with her, as we will see. And yet this novel did not change the shape of the novel: third-person narrations of the doings of openly fictional protagonists set within a firmly drawn historical and geographical frame were not a thing of the immediate future. Lafayette invented her princess, but she did not invent modern fiction, because La Princesse de Clèves becomes “fiction” only in the rearview mirror of literary history. Instead, the novel was an isolated manipulation of longstanding conventions and local practices that changed precisely nothing.
Those conventions and practices belonged to the Aristotelian tradition, and they were adhered to by writers of both long historical romances and the short historical novellas proliferating before and after the appearance of La Princesse de Clèves.2 Had Lafayette pushed a plausible figure into the foreground of her tale—a woman who might have existed—she would have certainly been stretching the bounds of normal practice, centered, as we will see, on real heroes of the past. Probably such a stretch would have gone unnoticed as just another example of Classical vraisemblance, of invention that is grafted so seamlessly onto history that we cannot tell where one leaves off and the other begins. But Lafayette was doing something much more radical still, and someone did take note—the otherwise unknown critic Jean-Baptiste de Valincour, who vigorously objected to the fact that Lafayette had created a protagonist who was impossible. This chapter begins, then, with Valincour’s indictment of Lafayette’s princess for being in direct contradiction with the historical record. I go on to point out how startlingly different this practice is with respect to an earlier historical novella of Lafayette’s, La Princesse de Montpensier (1662), whose heroine was, to the contrary, real to the point of indiscretion; La Princesse de Clèves’ impossible protagonist, I argue, allows Lafayette to break from a traffic in “true stories” and critique a court culture of gossip. And the invention of the Princess has another effect as well, as an appreciative Valincour points out. Far from being merely a retrograde churl, the critic readily admits the peculiar emotional bond he feels with Lafayette’s characters, a bond he goes on to praise and describe at length, and that is only possible, I suggest, because the character of the Princess disables normal reading protocol. The chapter concludes with some thoughts on the implications of the type of poetic contextualization I offer. For if viewing La Princesse de Clèves through the lens of the Aristotelian tradition robs Lafayette of her stature as the inventor of modern fiction, it has the compensating effect of making the subsequent history of the novel much more comprehensible even as it overturns many common assumptions about the evolution of literary forms. (An appendix to the chapter, intended for specialist readers, describes the types of invention practiced by Lafayette’s competitors in the genre of the historical novella.)
Inventing a Princess
Immediately following the anonymous publication of La Princesse de Clèves in 1678 appeared two book-length appraisals of the novel, also anonymous: first, the fairly critical Lettres à Madame la Marquise *** sur le sujet de la Princesse de Clèves, attributed to Jean-Baptiste de Valincour, and then, in the novel’s defense, Jean-Antoine de Charnes’s Conversations sur la critique de la Princesse de Clèves. Valincour is far from uniformly hostile, as we will see, yet a number of things about Lafayette’s heroine bother him—her behavior, as many literary critics have underlined; but also her existence itself, or rather the fact that she never existed. This aspect of Valincour’s criticism has received little attention, and no wonder: as readers whose codes have been formed by, say, Fabrice del Dongo’s experience at Waterloo, we are apt to take the mixture of historical precision and fabricated characters as intuitively novelistic. Indeed, on this point it is difficult not to side with Charnes, who mocks his adversary for staging confrontations between novels and history books. Such has been the position of Charnes’s modern commentators, who credit him with recognizing “the autonomy of novelistic fiction with respect to history”3; the implication, of course, is that Lafayette’s Princess is an ancestor of Stendhal’s Fabrice, and that Valincour may be right to say that Lafayette is breaking with tradition, but, turned to the past, he is blind to the novel’s future. In fact, Charnes is not clairvoyant; he mounts a serviceable defense of Lafayette’s practice, but ultimately Valincour’s incomprehension is a better guide to Lafayette’s idiosyncratic use of history and the challenge it posed to longstanding Aristotelian accounts of poetic invention.
Steeped in Renaissance interpretations of Aristotle’s Poetics, Valincour cannot abide what he sees as Lafayette’s violation of a rule of literary composition: poets and novelists (faiseurs de romans; V 63), he writes in his Lettres, choose historical episodes of sufficient notoriety to pique the reader’s curiosity but not so familiar in the details as to limit the writer’s room for maneuver.4 The alloy of historical fact (the basic episodes and personages) and poetic invention (the maneuvering) was not of Valincour’s manufacture: it was the one found commonly in Racine and Corneille’s discussions of tragedy; Gomberville and the Scudérys say much the same thing about romance, a genre whose increasing recourse to history helped cement its prestige.5 And such seventeenth-century French use had deeper sanction, as makes clear Valincour’s recourse to the authoritative Aristotelian commentary of Renaissance scholar Castelvetro. For according to the latter’s Poetica d’Aristotele vulgarizzata et sposta (1570), cited toward the close of Valincour’s first letter, inventing new characters was an impermissible infraction of Aristotelian laws. That Aristotle himself had said no such thing is of little concern; Castelvetro’s position, doubtless a bit strict, followed from an interpretive tradition that consistently relegated invented characters to the comic. Serious writers, by contrast, take famous people and events from history—fame being the very basis of readerly interest—and then invent the meetings, dialogues, motivations, and very frequently love stories that had been left unrecorded. The goal, Valincour notes, is to give readers the impression of discovering “what historians forgot to write” (V 70). Authorial latitude is huge, “as long as [invention] [is] not directly opposed to historical truth” (V 71).6 By contrast, Lafayette does not respect this accepted practice. Setting her novel in the well-documented court of Henri II, she makes the false step of contradicting the documentation, marrying a woman who never existed to a man who did, and a man who, moreover, was never married in the first place. Lafayette’s heroine is—Valincour repeats the expression twice—“visibly false” (V 69).
Modern critical attention to the subject of verisimilitude in and around Lafayette’s novel has focused almost exclusively on what has been called cultural plausibility: what were the ideological presuppositions that made readers at the time discount episodes or actions as implausible?7 This interest is far from anachronistic, since cultural plausibility is thematized in the novel itself, and present both in Valincour’s comments and in the testimony of individuals such as Bussy-Rabutin and Fontenelle.8 Valincour, however, objects to the “visibly false” heroine because the princesse de Clèves, whatever her