I have defined a plot as the relationship among the incidents in a work of literature and devoted two chapters to characterizing the five kinds of relationships and the two worlds in which they can exist. This chapter will discuss an even more basic question: what, if anything, is an incident?
One can argue, of course, that an incident is any part of a text that a reader decides to call an incident. In this extreme position, all characterizations of incident in principle are equally valid and in practice seek their justification not in any epistemic dependence on the text but in the elegance, coherence, ideological usefulness, psychological gratification, or richness in resonance of the remarks the text stimulates the reader to make. Roland Barthes seems at first to be making this argument in S/Z, but stops short of complete arbitrariness:
The proairetic sequence is never more than the result of the artifice of reading; whoever reads the text amasses certain data under some generic title for actions (stroll, murder, rendezvous), and this title embodies the sequence; the sequence exists when and because it can be given a name, it unfolds as the process of naming takes place, as a title is sought or confirmed.
(19)
In this passage, Barthes assumes that the text will talk back: a sequence needs a name if it is to be perceived, or even to exist, but to be confirmed, a name needs something at least arguably present and concrete, a sequence in the text. Barthes’s restrained epistemic nihilism lacks the philosophical impregnability of the extreme position. The word “confirmed” deprives him of the privilege of ignoring the text, but it also avoids the question we tend to ask when Freudians, Marxists, postmodernists, or sophomores tell us portentously that the human mind can never react with complete objectivity to any matter under consideration: so what else is new?
Barthes’s chief vulnerability lies not in the question of how the reader projects his categories upon the text but in the study of migratory plots that exists in many different pieces of folklore or other objects of explication. In the critical traditions where modern plot studies began, it takes strong devotion to theory to treat as an artifact of reading an incident that has retained its identity through many different texts, often for millennia. The devoted nihilist could always answer, “The identity of the migratory plot is in itself an artifact of the reader’s perception,” preserving ideological consistency at the expense of a body of experience to which Aarne, Thompson, and Jean Girandoux, who wrote the play Amphitryon 38, have devoted enormous attention. An ingenious nihilist could challenge the exact boundaries of any incident since incidents change, taking on the coloration of the culture or the genre where they appear, but the generations of philologists who have explored such changes could attack the nihilists’ adversarial consistency by claiming that if incidents could talk, they might say, “I change; therefore I am.” This appeal to experience is no more sophisticated than Samuel Johnson’s kicking of a rock to prove the world was there, but this book rests modestly on just such a rock.
The Russian Formalist critics of the 1920s were trained in Saussurian structural linguistics and inherited Alexander Veselovsky’s positivistic dream of a clear and unambiguous science of literature that would account for everything by using simple terms and consistent rules. Tomashevsky, Propp, and Shklovsky all begin discussions of plot with a deferential acceptance of Veselovsky’s definition of a plot (siuzhet) as “a complex of motifs” and a deferential rejection of his characterization of a motif. Veselovsky drew from the folklorists the perception of a motif as a persistent entity, like the killing of a dragon or the abduction of a princess, that could move from work to work across the barriers of time, space, language, class, and culture, even penetrating communities which could not understand the original meaning of a particular motif. In the tradition of Max Mueller, he explains that a motif is a
formula which in early society answered questions which nature everywhere posed to man or which reinforced those especially vivid impressions of reality which seemed important or were recurrent. The mark of a motif is its figural single-termed schematicism; such are the irreducible elements of primitive mythology and folktale: someone steals the sun (an eclipse); a bird bears lightning-fire from the sky….
The simplest type of motif can be expressed by the formula a + b: the wicked hag dislikes the beauty—and assigns her a life-threatening task. Each part of the formula is subject to transformations; b especially undergoes expansion; the tasks may be two, three (the favorite folk number), or more…. Thus a motif would grow into a plot.
(Istoricheskaia poetika, 494–95)
In this passage, Veselovsky is establishing the bases for an ambitious effort to adapt rules already developed for folk plots to more extensive and intricate narrations, a task that would later attract investigators as different as Shklovsky and Claude Bremond. Within the two pages which I have abridged here, Veselovsky shifts from the atomic perception of motifs as the many irreducible units out of which a plot may be assembled to the alchemical perception of a motif as a single unit, which may grow into a plot through repetition and whose forms and uses are restricted by culturally inherited rules. None of the Formalists objected to this double way of looking at a motif or to the coexistence in Veselovsky’s text of two different ways of perceiving and presenting the world, but most of their work leaned more toward the structural than the algorithmic.
Tomashevsky saw no conflict between the two approaches when he called a plot (tema) “a certain more or less integral system of events, one ensuing from another, one linked with the other,” and goes on to describe these events as follows:
Before organizing a plot, it must be divided into parts, then be hung on the narrational core. The concept of plot is a summarizing concept, uniting the verbal material of a work. A whole work may have its plot, and at the same time each part of a work possesses its own plot. Such a separation out from a work of parts each unified by a particular unity of plot is called the analysis of the work….
Through such analysis of a work into units of plot, we finally arrive at irreducible parts, at the most minute fragments of plot material: “Night fell,” “Raskolnikov killed the old woman,”…etc. The plot of an irreducible part of a work is called a motif. Basically, every sentence possesses its own motif.
(Teoriia literatury, 136–37)
In this strongly positivistic passage, Tomashevsky writes like a mathematician describing sets whose members are sets whose members are sets. He would not have known about fractals, which only became popular with mathematicians in the past few decades. A fractal curve has kinks, and kinks on the kinks, and smaller kinks on those kinks, and so on indefinitely, so that fractal patterns repeat themselves on a smaller and smaller scale and an enlargement of any part of a fractal curve will look exactly like an unenlarged portion of the same curve.
Tomashevsky asserts that plots can be analyzed into units that are also plots, so that no generic difference distinguishes “Raskolnikov killed the old woman” from “Night fell,” although one summarizes a hundred pages of text and the other recapitulates two words of text. He understands that the irreducible plotlet and the scheme for a whole novel are merely the extreme ends of a spectrum of plot sizes.
Propp carries Tomashevsky’s attack on Veselovsky to its logical conclusion:
The motifs which [Veselovsky] cites as examples do decompose. If a motif is something logically whole, then each sentence of a tale gives a motif. (A father has three sons: a motif; a stepdaughter leaves home: a motif; Ivan fights with a dragon: a motif; and so on.)…. But let us take the motif “a dragon kidnaps the tsar’s daughter.”…This motif decomposes into four elements, each of which, in its own right, can vary. The dragon can be replaced by Koshchej, a whirlwind, a falcon, or a sorcerer. Abduction can be replaced by vampirism or various other acts by which disappearance is effected [sic] in tales…. The final divisible unit, as such, does not represent a logical whole.
(Morphology of the Folktale, 12–13)
Propp’s apparent quarrel with Veselovsky centers upon the word “irreducible,” which Veselovsky himself did not intend to preclude internal structure, as his formula “a + b” indicates. But Propp really was objecting to the vast and messy collection of variables Veselovsky, like a lexicographer, would have to deal with, unlike Propp, a morphologist who pursued a limited and ordered collection of variables. Propp finds that Veselovsky’s motifs relate to particular plots but cannot be generalized. To ascertain the rules for a folklore plot, Propp must isolate himself from particular folktales and deal with “functions” that can be reduced not to sentences like “Raskolnikov killed the old woman,” but to abstractions like “interdiction, interrogation, flight,” whose very bloodlessness makes them applicable to many tales. He defines a function as “an act of a character, defined from the point of view of its significance for the course of the action” (21), and adds that the functions of characters serve as stable, constant elements in a tale, independent of how and by whom they are fulfilled. They constitute the fundamental components of a tale, and the number of functions known to the fairy tale is limited.