3

3.3. Temporal structure

Definition

Selection

Order

Duration

Definition

Literature is, like music or cinema, a sequential form of art; a literary work, just like any other linguistic product, unfolds itself in time. Language exists in time, but its mission is to represent, to signify. And among things which can be represented we find time itself. A text, unfolding itself in time, can at the same time represent time. Texts do not have to be narratives in order to do this. The tense system of any language is designed for the representation of time in any kind of text, narrative or other. We might argue that from the moment we find a represented time in a text we are identifying if not an actual narrative at least a narrative trait or structure. But we come closer to the spirit of narrative temporality proper when we define it as the use of the time of the representation as an icon of the represented time. By virtue of this mapping of one time into the other, the textual time becomes a representational time, and the action time becomes a represented time. Things might stop there, and the two temporal sequences would overlap in a perfectly homogeneous way. But we must remember that action time is also represented semantically by means of the linguistic signs of the text, not merely iconically through the linguistic chain. The mapping of action time into textual time is governed by the semantics of the text, to the extent that, far from one-to-one homogeneity being the rule, no two texts present the same temporal formula. The interplay of iconicity and semantized time ensures that the represented temporality is distorted in a manifold of ways and degrees. What we experience in a narrative text is therefore not the action time as such, but a represented action time, what we call the story time.

The temporality of a literary work can be studied from a variety of viewpoints, since literature, like any human activity, is temporal through and through. Now we shall focus on just one specific way in which time is present in narrative: some kinds of relationships between the temporality of the action and its representation as a story. There are other aspects of narrative temporality, such as the temporal-causal relationships we have studied under the heading of action, or the temporality of the narrative act. Narrative temporality is the result of the interaction of these different types of narrative time.

We are going to concentrate on story-level temporal structures, which are the result of mapping the time of the action onto the different temporal sequence of the discourse. In doing so, therefore, we shall presuppose those structures which are peculiar to action and discourse time, and which underpin the structure of story time. The story time is the result of the interplay of action time with another temporal sequence, the textual time of the linguistic chain. It is very frequent to meet descriptions of narrative time which assume only two temporal threads.[1] Either the action time as a necessary referent or the duration of the enunciative act get lost somewhere in the description. We must assume on the contrary that each of the levels of analysis we distinguish can generate a temporality of its own: for instance, it makes no sense to speak of the "temporality of enunciation" as if enunciation were a simple, univocal phenomenon. The complexity of the temporality of enunciation will mirror the complexity of the enunciation itself, the interplay of the author's and the narrator's voices. If the basic scheme of action, story and narrative is complicated in any way, if, for instance, any of the levels duplicates itself, the temporal structure of the narrative will become proportionally more complex.

The study of narrative tense is the study of three kinds of relationships between the succession of events in the action and the succesion of signs which represent them in the discourse. There are relationships of selection, of order, and of duration.[2]

Selection

The most famous classical reference to the problem of narrative selection is Horace's differentiation between narration in medias res and narration ab ovo. These terms are introduced in a paragraph discussing the problem of how to give a well-known subject a brilliant treatment, or, more to the point, of how to turn a traditional action into an artistic story. Horace, following Aristotelian suggestions, sets Homer's practice as an example:

He does not begin a "Return of Diomede" from the death of Meleager, nor the war of Troy from the twin eggs [gemino ab ovo ]. He ever hastens to the issue, and hurries his hearers into the midst of the story [in medias res ], just as if they knew it before; and what he thinks his touch will not turn to gold, that he lets alone. (Art of Poetry 70, lines 146-50)

Preliminaries can be dispensed with, and writing manuals often advise the prospective novelist to begin with a significant happening rather than with a lengthy exposition.

Selection will define the beginning and the ending of the story, which need not be those of the action.

It also helps determine the kind of exposition which will bew offered the reader: whether it will rely on the unfolding of the events themselves or whether it will consist in reminders, deductions, etc. based on later happenings and conversations of the characters.

Traditional criticism has always defined the well-written story as an organic unity, a system where each of the elements should contribute to the overall construction and meaning. "Unity of plot" is an Aristotelian requirement for artistic construction. "Unity of plot," Aristotle says, "does not, as some persons think, consist in the unity of the hero" (Poetics VIII). The actions of a man do not necessarily build up a single pattern, a unified action which makes a coherent whole with a sense. And the action must be a whole:

A whole is that which has a beginning, a middle, and an end. A beginning is that which does not itself follow anything by causal necessity, but after which something naturally is or comes to be. An end, on the contrary, is that which itself naturally follows some other thing, either by necessity, or as a rule, but has nothing following it. A middle is that which follows something as some other thing follows it. A well constructed plot, therefore, must neither begin nor end at haphazard, but conform to these principles. (Poetics VII)

This definition of beginning, middle and end derives from Plato's Phaedrus. It is not wholly truistic: "the acceptance of the statement that a story must have a beginning would seem to be that the story must start more or less with where its antecedents may be taken for granted, that is, where they are generic rather than specifically relevant" (Wimsatt and Brooks 30). The distinction is especially relevant in Greek tragedy, which relied for its plots on stories which were well known to the public. Aristotle draws this conclusion from the requirement of unity:

As therefore, in the other imitative arts, the imitation is one when the object imitated is one, so the plot, being an imitation of an action, must imitate one action and that a whole, the structural union of the parts being such that, if any one of them is displaced or removed, the whole will be disjointed and disturbed. For a thing whose presence or absence makes no visible difference, is not an organic part of the whole. (Poetics VIII)

Aristotle sees the whole as more than the sum of its parts, if only in that it includes the relations among the parts. Defined in this way, the unity of action has a much more general and comprehensive nature than will be allowed by later interpretations. We may notice that it is not too much to say that this is a structural definition of unity: its abstraction allows us to account for any kind of unity we may find in a plot, and in this respect it is difficult to go beyond it.

Aristotle compares the unity of the plot to the unity of a living being. It is only a comparison, but it has its importance. The comparison of the unity of a literary work with that of a living organism refutes the charge that Aristotle is describing a formal, dead, mechanical kind of unity.[3] This is "unity" in a sense similar to that used in modern structuralist poetics. However, it has an obvious shortcoming : being a structural definition of the plot, which is only one of the constituent elements of a tragedy, it fails to account for the whole of the tragedy. This we have already seen; Aristotle's theory of tragedy and indeed the whole of his Poetics is plot-centered, and so it fails to account for many literary phenomena. It is the work that we might wish to define as a whole and as a structure, and not merely the plot.

This more comprehensive concept of organicism is found in nineteenth-century theorists of poetry, like Coleridge, and later on with reference to the novel, e. g. in a Victorian critic like Walter Besant:

In Fiction the power of selection requires a large share of the dramatic sense. Those who already possess this faculty will not go wrong if they bear in mind the simple rule that nothing should be admitted which does not advance the story, illustrate the characters, bring into stronger relief the hidden forces which act upon them, their emotions, their passions, and their intentions. All descriptions which hinder instead of helping the action, all episodes of whatever kind, all conversation which does not either advance the story or illustrate the characters, ought to be rigidly suppressed. (Besant 1885: 15, 24)

This conception, Henry James will complain, is plot-centered and smacks too much of a ready-made recipe; but still it introduces an organic standard which James himself will use.

We shall define ellipsis as a relationship between action and story in which the story suppresses action elements. Ellipsis is the opposite of representation.

Ellipsis may occur not only at the beginning and the end of a story, but at any point within the story. It may involve not just secondary material (space or characters not described, minor actions or catalysts omitted) but also crucial information (action kernels). Ellipsis of this kind would seem to be a problematic concept, since the action only exists through its rendering in the story, and ellipsis is defined as that part of the action which is omitted in the story. Nevertheless, action materials which have not been included in the story may still be active in a variety of ways. The definition of ellipsis needs, therefore, a different kind of knowledge to justify its presence, and this is usually the result of the literary competence of the reader and the predictability of action patterns, apart from the more general scenarios which govern our everyday behaviour. Our concept of ellipsis should take these phenomena into account. Story schemata both organize the pattern of presence and ellipsis and guide the reader in determining which ellipses are significant and which are caused by redundant information. Ellipsis is therefore connected to the driving principle of the narrative, the narrative design which shapes relevance and interest, the plot. So many ellipses are mere time gaps which are used to give shape to this plot, to emphasize causal over temporal connection by modelling the temporal sequence on the model of the causal one. A relevant ellipsis, on the other hand, is a narrative gap which is flaunted in some way as significant, through explicit comment or through the break it causes in action sequence which we interpret through a well-known intertextual model. Usually the gap will be filled later on through the use of anachrony.

Ellipses can be divided from a formal point of view into explicit, implicit, and hypothetical, and according to a temporal criterion into definite and indefinite.[4] Usually the definiteness of the ellipsis is not explicit, and is left to the inference of the reader.

[1] E.g. Tomashevski (1982, 54); Barthes (1981, 12); Kristeva (1974, 250), or Genette himself (1972, 77).

[2] In what follows we adapt Gérard Genette's vocabulary and concepts to our own. Genette divides the temporal relationships between action and story into three types:

we will study relations between the time of the story [action] and the (pseudo-) time of the narrative [story] according to what seem to me to be three essential determinations: connections between the temporal order of succession of the events in the story and the pseudo-temporal order of their arrangement in the narrative . . . ; connections between the variable duration of these events or story sections and the pseudo-duration (in fact, length of text) of their telling in the narrative--connections, thus, of speed . . . ; finally, connections of frequency, that is . . . relations between the repetitive capacities of the story and those of the narrative. (1980, 35)

Genette's "frequency" is for us one more among several kinds of narrative aspect, not tense.