Teaching an undergraduate course on the novel reminds us of how little effect a century of “literary theory” has had on how we read. Lacking our dubious skill in dealing with novels as “texts,” undergraduates remind us of an essential epistemological point: one can’t follow a story without constructing its actors in our mind in the same way that we construct real people. We speak of these textual entities as though they are real because that is indeed the only way to talk about a novel’s content. But this very fact reflects the historic specificity of novelistic form.

A novel, however short, must be long enough to allow us to experience the passage of time in the lives of the characters. The novel, whether classical, “realist,” or avant-garde, alternates between interactive scenes and extra-scenic passages characteristically narrated at too fast a pace to provide material for the reader’s vicarious experience: “telling,” not “showing.” The characters leave the scene (I say “the scene” because scenic confrontations all derive from the same model), return to it, and leave it once more. Although these acts of leaving and returning are often literal, their narrative implementation permits of alternatives: interior monologues, flashbacks, repetitions. We make the passages of summarizing narration part of our reading experience only by integrating them within our scenic knowledge of the character: the fact that X has studied here, worked there, traveled somewhere else is relevant only to the X whom we know now, in this scenic moment of his life. Novelistic scenes are islands of meaning in a non-scenic sea.

What the apparently merely quantitative fact of length makes distinctive about the novel is the tension between the scene and its narrative environment. What holds it together is an overarching sense of the unity of life, usually of a single character, but extensible to the family and even the community in the collective sagas or romans-fleuves of the plethoric first decades of the twentieth century. The novel models the texture of life in its systolic-diastolic opposition between scenic and non-scenic moments. This texture, by making us aware of the fragility of the scene, creates a phenomenon that I shall call novelistic felicity: the narration of a revelatory experience analogous to that of the sacred because, at that one precious moment, wholly enclosed within the scene. Love is the source of most such experiences, but there are felicitous scenes of battle, of murder, even of whaling. What we take for granted in classical theater (and, in an ironic mode, in modern theater as well) has in the novel the status of a rare epiphany–a status that self-conscious modern novelists (think of Proust’s petite madeleine) increasingly thematize, but whose privilege is already clear in France’s first great novel, La princesse de Clèves.

Let us consider in this light the association of mediated desire with the novelistic context in which René Girard discovered it in Mensonge romantique et vérité romanesque (1961). Although Girard has since found mimetic desire equally present in the theater, notably in Shakespeare, his original insight should not be lost. The ahistorical tenor of Girard’s writing that make some see the triangle of desire as a “structure” should not blind us to the historical specificity of his analyses. All desire is mimetic and all culture a means of deferring it, but the novel defers mimetic desire in a new and radical way.

The mimetic theory of desire is a theory of the “double bind”: the mediator both incites our desire and discourages it, incites it because he discourages it. (Georges Bataille’s model of (erotic) desire as transgression respects the indissociability of desire’s double bond.) Thus it is not always so easy to distinguish between the positive and negative moments of triangular desire. “Mimetic theory,” sad to say, has never gone beyond Girard’s basic scenario: (1) the mediator, knowingly or not, designates the object to the subject-disciple (imitate me!); (2) the disciple displays desire for the object; (3) the mediator forbids the object to the disciple (don’t imitate me!). If the mediator is external to the world of the disciple, he is unaware of the latter and therefore presumably cannot carry out the negative moment of interdiction. This appears to imply that only internally mediated desire can produce the triangular double-bind effect. But it is clear from Girard’s own association of external mediation with the sacred that attraction and interdiction both inhere in any mediation process and therefore in any desire.

Considered in this light, moments (1) and (3) of Girard’s triangular scenario are merely two descriptions of the same operation. To “designate” an object is both to display its availability as a focus of desire (which is why harems are not open to the public) and to display its unavailability to everyone but its designator (which is why rich people drive expensive cars). The positive (imitate me!) and negative (don’t imitate me!) moments of mimetic desire both inhere in the same act.

This analysis permits us to add a degree of freedom to the triangular model in order to account for the different ways in which mimetic desire may be made the basis of a literary (or, more generally, “cultural”) plot. Instead of limiting the model’s variability to the distance–internal or external–of the mediator from the disciple, we consider as well the degree to which the positive and negative moments of mediation predominate at the origin of desire. On this point, I propose the following hypothesis: whereas the classical, typically theatrical, plot turns on negatively mediated desire, the modern, typically novelistic, plot turns on positively mediated desire.

This explains, for starters, why Girard first discovered mimetic desire in the novel and only later thought to apply his schema to tragedy. Oedipus is no Julien Sorel seeking in Napoleon’s memoirs the course to follow. Always already the object of Laius’ aggression, he chooses a path always away from his paternal mediator. What drives the plot of Sophocles’ play is Oedipus’ quest not for Jocasta but for his father’s rival, that is, for himself as disciple of the negative moment of mediation. What Oedipus learns from mimetic desire is above all the impossibility of the subject’s coexistence with his mediator. Sandor Goodhart points out in “Oedipus and Laius’ Many Murderers” (Diacritics , 1978: 55-71) that we never really learn whether Oedipus killed Laius, because once the shepherd brought in to testify about the murder recognizes Oedipus as the child Jocasta gave into his care, Oedipus accepts his guilt as his mother’s husband and forgets about finding his father’s murderer. This switch, which Goodhart attributes to Sophocles’ awareness of the arbitrary necessity of Oedipus’ guilt, also demonstrates the priority of the negative moment of mimetic desire over the positive. The positive moment (incest with the mother) provides Oedipus and the audience with an imaginary object of desire, and guilt, that dissimulates for a moment the more banal negative moment (murder of the father), but it is the latter that drives the story, as it does in all tragedy. The tragic agon re-presents mimetic crisis not as a common desire for a central object but, on the contrary, as the chaotic aggression occasioned by a central object’s loss of desirability. The well-known fact that tragedy typically deals with plots known in advance is but another way of stating this fact; the “already-known” is the negative moment of desire. The name, and the history, of the novel make the contrast clear.

Girard’s title sets up as a foil to the novel (romanesque) the “romantic” (romantique) that dissimulates mediation behind the apparent spontaneity of desire. Without accepting the invidious nature of Girard’s distinction, we may understand romance in its courtly origins as the first stage of modernity’s shift from interdiction of the sacred world to valorization of the profane world. The donna m’apparve that inaugurates courtly love marks the descent of the sacred to the earth. Whether or not Dante loved Beatrice because some nameless rival designated her to him, the crucial mediation that marks his love is the consecration of the profane world by the spirit of Christianity.

The novel is no longer concerned with the descent of the sacred but with the disenchanted world of profane desire–the world of the marketplace. Even if “falling in love” still follows the Dante-Beatrice model these many centuries later, sacred mediation is not otherwise an effective way to chart a course through such a world. To construe the novel as a condemnation of mediated desire requires that we take its sacrificial structure as a moral, as in a fable. (Tobin Siebers’ Morals and Stories  [Columbia UP, 1992] insightfully examines the paradoxical relationship between narrative and the lesson we are expected/asked/dared to draw from it.) In its contribution to the practical world of its readers, the novel is much rather an encouragement to mediated desire; its cautionary moral is no less necessary in form and artificial in content in the novels that Girard accepts into his Pantheon than in Les liaisons dangereuses (see Chronicle 235). If the novel reveals the “demonic” nature of mediation, it also reveals that the lives of those who read novels are constructed on a demonic basis.

Girard concludes his history of the novel with Alyosha Karamazov’s renewal of the Gospel message of love transcending worldly desire, a Christian equivalent of apocalyptic socialism. Once the critique of mediated desire truly becomes the moral of the novel, as it might arguably be said to do with Dostoevsky, from whose world the evolving consumer society so visible from Balzac through Flaubert and Zola is virtually absent, the novel becomes a prelude to spiritual-political change–such as the Bolshevik Revolution was supposed to bring about. But however much Girard wants his story to end with Dostoevsky, it ends in reality with Proust, whose esthetic transcendence of desire is at the same time the recycling of desire in a bourgeois world. To understand Proust’s artistic praxis as a tribute to the punctual revelation of the vanity of mimetic desire is, in his case above all, to succumb to the myth of the novel rather than to elucidate it.

It is not surprising that the heroic age of the novel coincided with the birth, evolution, and implementation of the illusion that future society would do away with the mediations of human exchange. The novel concretely recounts a life of mediation while pointing abstractly to a set of meanings beyond mediation. Its survival through postmodernity depends on our willingness to identify with lives that can no longer be felt to incarnate any such meanings. Impatient with the exemplarity of these unexemplary biographies, I, like many others, have come to prefer the more concrete–and temporally limited–scenicity of cinema. The novel requires that we follow a protagonist from scene to scene in the faith that the time we have invested in reading will be capitalized in a thrill of narrative felicity. For me, like Marcel’s mother’s kiss, this moment, so eagerly awaited, cannot compensate by anticipation the loss of meaning that is its inevitable aftermath.