Lafayette’s play with the historical record now starts to assume its full significance: the counterfactuality of her heroine has made it impossible to read her novel as the characters in that novel “read” each other. While those characters give themselves over to gossipy reading that always refers narrative to the real world, Lafayette’s real readers are discouraged from such a mode of consumption. They are not warned off by a disclaimer, since in Lafayette’s day saying that there are no keys is the best way to ensure that readers will look for them. They are discouraged, rather, by the author’s bypassing the normal rules of poetic invention. Had Lafayette supplied a historical cast of characters with plausible or attested adventures, as she did in La Princesse de Montpensier and as the historical novella generally did, she would have been guilty of involvement in the very trafficking her tale denounced.27 Her guilt would clearly be no less great had she set the narrative in the present and hinted at keys, as many other stories of the day did. But La Princesse de Clèves is not a true story, and there is no key. Only an impossible heroine, set safely in the past, could keep Lafayette’s readers from being implicated in the culture of gossip that brings misfortune and death down on her characters.28 Quite simply, the Princess permitted Lafayette to write about intrigue without participating in intrigue.29
Implausible Identifications
Such obliquity represented, then, a gain for Lafayette: it enabled a noncomplicit critique of a culture’s narrative promiscuity, that is to say, its unceasing circulation of sexual secrets. Yet an impossible heroine implied disadvantages as well. After all, historical truth underwrote not only Aristotelian poetic production but also theories of consumption; without it, what were readers supposed to do with this novel? Placing a nonexistent heroine at the center of the novel’s web disables the pretense of teaching about the illustrious events and figures of the past; and being impossible, the Princess cannot help us entertain hypothetical “secret” causes of those events. Nor can one read for edifying examples of good and bad behavior, since the ethics of exemplarity likewise presupposed the assertion of historical truth.30 And veracity enabled even the dominant explanations of emotional response to the artwork. For theorists of the Italian Renaissance as well as of Classical France, the emotions audiences took away from books and plays were predicated on the historical reality of heroes, whose emotion was in a sense channeled by the poet before flowing outward to the actor or character and finally to the spectators or readers.31 So what was a nonentity good for?
Of course, even leaving aside comedy, readers of the period did have experience with characters lacking in real-world attestation. And the most obvious thing to do with them at the time was to look for the hidden historical referents—to look for keys. The success of Le Grand Alcandre, the disguised account of Henri IV’s love life to which I have already alluded, clearly indicated such interest in veiled accounts of scandal; originally destined for private circulation, Bussy-Rabutin’s Histoire amoureuse des Gaules on publication in 1665 cloaked its referents in disguised names (and earned its author a stint in the Bastille); Subligny’s La Fausse Clélie (1670), the subject of the following chapter, was meant to be interpreted as an account of current events. Keys were used moreover in works, like Scudéry’s Clélie, whose characters maintained a nominal historicity. Curiously, however, no one appears to have taken a key to La Princesse de Clèves.32 The fact that keyed readings are not attested in contemporary documents is admittedly surprising, given that part of the novel’s publicity campaign would seem to incite them: in the January 1678 issue of Le Mercure galant, Donneau de Visé published a story, La Vertu malheureuse, having the same basic plot outline as La Princesse de Clèves but recast as the purportedly genuine story of a woman who has just retired from court life. The lack of historical antecedent for the Princess, coupled with the framing of La Vertu malheureuse—the real Princess is out there right now, in a convent somewhere in France—could have made people read Lafayette’s work as a “fake” historical novella that was in fact about a present-day adulterous love. Yet it didn’t. Putting a nonhistorical character into an otherwise historical setting was evidently not interpreted as an invitation to look for a real person.33
There was, however, a second mode of reading that could accommodate Lafayette’s invented princess. This reading was ethical: the Princess, and specifically her avowal, was taken as posing a general problem of conduct to be parsed and debated. Antecedents for this are manifold. In Marguerite de Navarre’s Heptaméron, the commentary of the tale-tellers on their true stories can be viewed as a colloquial version of humanist exegesis.34 The shepherds of d’Urfé’s L’Astrée, meanwhile, embody certain transparent bents, passions, or ways of loving. Hence, to L’Astrée’s Celadon, a virtual allegory of steadfast faithfulness, was opposed his mirror image Hylas, eloquent spokesman for the virtues of change; numerous other characters illustrated additional approaches to amorous service. Such an ethical concept of character manifestly has nothing in common with the creation of a “deep” individual; but nor do these characters function as types in the comic sense. Rather, they are more like hypothetical instances through which readers can work through problems of conduct. By the time of Scudéry’s great romances of the 1640s and 1650s, the use of characters to illustrate amorous conundrums—what were known as “questions of gallantry” (questions galantes)—was typical. And Lafayette knew the practice well: her own Zayde is effectively built around one such conundrum, borrowed in fact from Scudéry: is knowledge of a possible partner an impediment or a stimulus to love? The cast of Zayde (1670–71) illustrates the possible permutations of the question. Hence, one man will insist on knowing everything about his love object, another will demand that his lover not know anything about him, while the hero, Consalve, who knows absolutely nothing about the eponymous heroine, will alone find his love rewarded.
Such a mode of reading was immediately pressed into service by Donneau de Visé, who in a successful bid to solicit the involvement of readers of Le Mercure galant used the Princess’s dilemma to pose a general problem, nowhere articulated in Lafayette’s novel itself: should a virtuous woman who respects her husband tell him of her love for another man so that her husband will allow her to leave court, and thereby to distance herself from the source of her temptation? Or does the calculus of happiness and pain dictate that she suffer in silence? Donneau de Visé was able to construct an elaborate chart in which possible opinions about the Princess’s avowal—its positives and negatives—are organized. And readers wrote to him, explaining why this or that reasoning seemed to them the strongest. Even lacking in historical credentials, the Princess could, then, be treated like any other romance character, that is, as an occasion for posing a general problem.
At least one reader, however, found this ethical mode of reading inadequate to a novel in which for the most part characters do not act in accordance with some governing temperament or philosophy. This was Valincour, and once again he is critical of the novel: La Princesse de Clèves has no proper heroes, with the Princess herself being, unsurprisingly, the worst offender. In situation after situation, she cannot respond to challenges in a manner the reader expects of a serious protagonist, which, in the court culture of wit and repartee, largely comes down to verbal mastery.