Introduction - NARRATOLOGY

Introduction - NARRATOLOGY

Preliminaries

Definition of narratology

Narratology is, etymologically, the science of narrative. The term was popularized, however, by such structuralist critics as Grard Genette, Mieke Bal, Gerald Prince and others in the 1970s. ¹ As a result, the definition of narratology has usually been restricted to structural, or more specifically structuralist, analysis of narrative.

The post-structuralist reaction of the 1980s and 1990s against the scientific and taxonomic pretensions of structuralist narratology has resulted in a comparative neglect of the early structuralist approaches. One positive effect of this, however, has been to open up new lines of development for narratology in gender studies, psychoanalysis, reader-response criticism and ideological critique. Narratology now appears to be reverting to its etymological sense, a multi-disciplinary study of narrative which negotiates and incorporates the insights of many other critical discourses that involve narrative forms of representation. Consequently, while our selection of texts in this reader gives ample representation to the original structuralist core of the discipline, it also includes samples of approaches which are narratological in the wider if not in the strict formalist sense of the term.

Wider and narrower definitions

Is Aristotle Poetics the first narratological treatise? ² Although Aristotle's work focuses on one specific genre, tragedy, it offers extraordinary narratological insights which are applicable to all genres that use plots. For Aristotle, plot (mythos) is the central element in a literary work, a narrative structure which is common to dramatic and narrative genres proper. The conceptual ambivalence of 'narrative' is present, then, from the very beginning of the discipline. A wider Aristotelian definition of 'narrative' might be 'a work with a plot' (e.g. epic poetry, tragedy, comedy); a narrow one would be 'a work with a narrator' (epic poetry, but not, in principle, drama or film). Today, narratology studies the narrative aspects of many literary and non-literary genres and discourses which need not be defined as strictly narrative, such as lyrical poems, film, drama, history, advertisements.

Narrative as mediated enunciation

The difference between narrative in the wider sense (as 'a work with a plot) and narrative proper has a parallel in the distance between the poles of dramatic narrative and mediated narrative ('showing' and 'telling'). Linguistic narrative (history, the novel, short stories) would be narrative in what we have called the narrow sense of the term, that is, narrative mediated in this case by the discursive activity of a narrator. Still, there may be other forms of mediacy: in film, the camera is a mediating device, albeit non-verbal (the use of the word 'enunciation' in this case is therefore a new, analogical development). Drama itself, of course, is a mediated presentation which uses a variety of linguistic and non-linguistic strategies, some of which may be more 'narrative' than others (e.g. the chorus in classical Greek tragedy, the device of the messenger who reports offstage events or, at the extreme, Brecht's 'epic' theatre).

The specificity, of each narrative medium

Each medium and each genre allows for a specific presentation of the fabula, different point-of-view strategies, various degrees of narratorial intrusiveness and different handlings of time. Consequently, each narrative medium requires a specific analytical approach to narrative structures and levels (see, for instance, chapter 14).

The novelistic tradition is one of continuous redefinition of the narrative voice, from fake memoirs and letters, to intrusive and omniscient narrators, to modernist experiments with objective presentation or a limited point of view. Some critics have argued that innovative novel-writing is inherently reflexive: the discourse of the novel is simultaneously a reflection on past and present ways of telling a story. This reflexivity leads both to a defamiliarizing foregrounding of technique and to experimental variations on the specific areas of narrative structure exploited by a given novelist (e.g. temporal structure, perspectival control). It should be stressed, however, that defamiliarization is not necessarily a characteristic of narrative (or the novel, or literature, for that matter): there are also

eminently ritualistic genres in which narrative pleasure derives from strict adherence to generic conventions (e.g. jokes, neoclassical tragedy, or noh drama in Japan).

One of the main differential traits of the novel is its ability to mix reflection and narrative (thence, in part, its reflexivity). The classic novel became the genre which revealed the inner life of characters, showing not just their behaviour but the relationship between action and character. By contrast, drama, in general, is much more directly focused on action. It is easy to see that the foregrounding of action in drama and of character in narrative leads to a whole array of consequences for the treatment of such aspects as time or psychological representation. Drama generally tends to focus on a significant and dearly defined action, with a strong plot based on cause and effect. The 'unity of time' of neoclassical drama, which of course was not meant by neoclassicist critics to apply to written narrative, was a recognition of this difference. It was also a recognition, albeit an exaggerated one, of the specificity of dramatic illusion.

The theatrical element of drama involves a further difference: the written, verbal text of a play is only the basis for the actual performance, which is a different and constantly changeable interpretation of that text and in fact on every occasion a new text for the audience. In film, unlike drama, the visual/aural presentation becomes a fixed text, which nevertheless remains subject to the viewer's perceptual activity. Drama and film, like narrative paintings, comic books, and so on, have a visual element in common. Images may be used to narrate just like words; it is here perhaps that narratology shows most clearly that it is not just a subsection of literary theory, but rather of a general semiotic theory.

Definition of narrative

A narrative is the semiotic representation of a series of events meaningfully connected in a temporal and causal way. Films, plays, comic strips, novels, newsreels, diaries, chronicles and treatises of geological history are all narratives in this wider sense. Narratives can therefore be constructed using an ample variety of semiotic media: written or spoken language, visual images, gestures and acting, as well as a combination of these. Any semiotic construct, anything made of signs, can be said to be a text. Therefore, we can speak of many kinds of narrative texts: linguistic, theatrical, pictorial, filmic. Any representation involves a point of view, a selection, a perspective on the represented object, criteria of relevance, and, arguably, an implicit theory of reality. Narrative structuring may become most elaborate in literary texts, but narrativization is one of the commonest ways of applying an order and a perspective to experience.

The term 'narrative' is, then, potentially ambiguous. As we have seen, it has at least two main senses: the broad one, which we have just defined, and the narrow one, according to which narrative is an exclusively linguistic phenomenon, a speech act, defined by the presence of a narrator or teller and a verbal text. This definition would restrict the area of analysis to oral or written narrative, and in the case of literary studies, to such literary genres as the novel, the short story, epic poetry, ballads, jokes. Here we shall be concerned mainly with the narrower sense, with the study of verbal narrative, and specifically with fictional/artistic narrative. The contributions we have selected deal in the main with the narrative element in the literary genres, although a few items deal with non-verbal narrative media such as film.

Narratological analysis concentrates on those aspects of textual production, structure and reception which are specific to narrative: for instance, the study of plot, or the relationship between action and character portraiture. Narrative may of course be approached in other ways: historically, thematically, stylistically, archetypally, deconstructively. In fact, most of the structures studied by narratologists do not exist exclusively in narrative works, but in narrative they are central and noticeably distinct. This is the case with regard to point of view or enunciation, for example, which may be found in a meditative sonnet as well as in a novel.

The analysis of narrative structure

Narratology in the strict sense of the word is usually associated with structuralism. Thus, in selecting relevant contributions for this reader we have assumed that structuralist approaches constitute the core of the discipline. The work of Saussure and the Russian formalists early in this century prepared the ground for structuralist thought. The formalists argue that words in poetry do not function only as signifiers: they are also signifieds. Literature is defined as a functional system, as a set of devices whose value is determined by other devices which are played off against them (those of other genres, past styles, and so on). A work presupposes conventions, other works, styles, genres, structures of meaning which go beyond the work itself. Literature is for these early structuralists a kind of langue of which each specific work is an instance of parole. French structuralists carried these linguistic analogies further during the 1960s and 1970s.

Structuralists often hesitate, however, when it comes to deciding the level at which the analogy should work: is it literature as a whole that works as a language, or is it the individual work that does so? Each work may be argued to constitute to some extent a langue of its own, may be seen as a self-regulating structure, since it creates, up to a point, the conditions for its own meaning and helps define the language in which it is interpreted. When the structuralists opt for this second alternative and seek to analyse the functioning of individual works, they are closer to the New Critical analysis of works as 'organic wholes'. If, on the other hand, they choose the first, that is, the analysis of general literary mechanisms, the individual work comes dose to disappearing. It becomes a mere crossroads of different codes, the codes being the real object of analysis. A kind of turning-point away from this tendency is represented by Barthes in S/Z ( 1970), which rejects the idea that a work can be reduced to the codes that enable its existence. ³ S/Z is often considered as the opening statement of the post-structuralist analysis of narrative, which tends to emphasize the reader's active manipulation of semiosis.

The initial question a narratologist would try to answer is: in what sense can we analyse the structure of narrative? How can we begin? The very definition of narrative we propose -- 'the representation of a series of events' -- assumes that narratives are composite entities in a number of senses, that a narrative can be analysed into the events that compose it, and that these events can be studied according to their position with respect to each other. In a series of events some are at the beginning, some in the middle, some at the end. A narrative therefore consists of a number of successive parts: it has a longitudinal structure of time and actions. This 'horizontal' approach to narrative description is analogous to syntactic analysis in linguistic studies. We shall call it the syntagmatic axis in analysis.

A narrative, then, is in one sense a succession of elements. But it is a compound in other senses, too, and can be analysed in more ways than one. In our definition, it should be noted, a narrative is not 'a series of events', but 'the representation of a series of events'. Here the composite nature of narrative appears not as a number of successive parts, in length or horizontally, but, as it were, vertically, in depth: the narrative is not what it seems to be; it is a sign which represents a state of affairs. This 'vertical' direction in analysis leads us from the sign to its signification. The basic activity in this sense is interpretation, and therefore we shall call this the hermeneutic direction in narrative analysis. What we get in a narrative text are not events as such, but signs, the representations of events. Here an infinite complexity may arise. In what way are the events represented? In what way is the narrative similar to or different from the events it represents? Narratological theories will largely consist in the formulation of possible answers to these questions.

We see, then, that the very definition of narrative leads us to the beginning of analysis, and in several directions at once. We shall examine different theories which analyse narratives either horizontally, or vertically, or both. As far as horizontal analysis is concerned, we have spoken so far of beginning, middle and end. Other concepts will complicate this simple account of parts. As far as vertical analysis is concerned, we may speak of levels of analysis. Our definition distinguishes at least two basic levels. If narrative is a semiotic representation of a series of events, one level of analysis will examine the events represented. Another level of analysis will examine the structure of the representation. We shall find that narratological theorists often differ when it comes to defining these levels of analysis: some distinguish two, while others speak of three or four. Mieke Bal tells us that there are three basic levels of analysis of narrative: fabula, story and text; Tomashevski only speaks of two, fabula and siuzhet. ⁴ En fact, this problem arises in all areas of literary study. Theories which appear to be similar often turn out to originate in entirely different critical projects. For the purpose of this discussion, we assume a framework of three levels of 'vertical' or hermeneutic analysis of the narrative text: text, story and fabula (see e.g. Bal, Narratology). Thus, if we take a work such as Robinson Crusoe, we will say that the text is the linguistic artifact that we can buy and read, written de facto by Defoe and supposedly by Robinson. The fabula is whatever happened to Robinson in his travels and on his island. The story is the precise way in which that action is conveyed, the way the fabula is arranged into a specific cognitive structure of information.

Bal has defined these concepts as follows:

1. A TEXT is a finite and structured set of linguistic signs.

1.1. A narrative text is a text in which an agent relates a story

2. A story is the signified of a narrative text. A story signifies in its turn a fabula. ⁵

We may represent these levels of signification by means of the following diagram: ⁶

The fabula is, according to Bal, a bare scheme of narrative events which does not take into account any specific traits that individualize agents or actions into characters and concrete events. A description of the fabula (or action) would also omit any temporal or perspectival distortions: there are no flashbacks or variations in point of view at this level of analysis. In other words, in Bal's conception the fabula is actually an action-scheme: it is a synthetic abstraction, not the concrete, full-blown action that we construct when reading or watching a narrative. It may be confusing that other theorists ( Ingarden, Martnez Bonati, Ruthrof) use the other concept of deep structure (concrete action and world, not abstractive fabula). We shall retain both concepts, since both are analytically significant: the full-blown or concretized action can be meaningfully opposed to a more abstract and reduced fabula or action-scheme.

It is also possible to draw up schemes of the story and the text (the 'reduced' version of the text being best called a summary). By this method we obtain the following critical tools:

The term 'plot' as used in everyday language often designates a storyscheme or an action-scheme, or a structure in between, mixing traits of both. Here we shall use it to refer to a scheme consisting of the structures of action and perception which shape the story. However, 'plot' is a tricky word because of its rich meaning, and several theories take into account other phenomena implicit in the everyday use of he word. ⁷

The concept of story needs further elucidation. A story is a fabula which has been given a presentational shape: a specific point of view and temporal scheme have been introduced. We could say that a story is a fabula as it is presented in a text -- not the fabula as such. The text is not the story, either: 'story' is still a synthetic abstraction we produce from the text, taking into account only its narrative aspects, considering it only in so far as it represents an action. We may recall that for Aristotlemythos was merely one of several 'aspects' of a literary work. A text is a linguistic construct, while a story is a cognitive scheme of events. The same story can give rise to a number of texts: for instance, when Kafka wrote The Castle in the first person and then rewrote it in the third person, the story remained essentially the same, but the text became a different one. The same story could in principle be told by means of different texts: a film, a comic book or a novel. But film adaptations of novels usually tell very different stories, even if the basic elements of the action are preserved. The story, then, can be looked on as a further structuring of the action. It may be ideally defined as the result of a series of modifications to which action is subjected. These modifications can be relative to time or to informational selection and distribution ( Genette 'mood', Narrative Discourse). Telling a story from a single character's point of view is one of many possible modalizations of the action.

A narrative text is also an instance of discourse, of linguistic action. Discourse is the use of language for communicative purposes in specific contextual and generic situations, called discourse situations. These can be described at different levels of specificity: there is written discourse in general, but also specific fictional written discourse. Increasing specificity would take into account historical or generic considerations. Of the infinite number of discourse situations which could be defined in this way, we may draw attention to a few: the writing of narrative discourse which is intended to be read as literature, the 'naive' reading of the same, the critical (academic) discourse on literature, writing factual narratives such as reports, or more personal narratives such as diaries and memoirs.

A text cannot be reduced to its linguistic codification, especially if by 'linguistic' we mean 'relative to the abstract system of language'. From the standpoint of linguistic pragmatics there are many cultural codes, apart from the Saussurean langue, structuring discourse (including narrative discourse). For instance, the social interaction rituals which allow speakers to position and identify themselves in conversation remain active when social interaction is represented in a text. An author uses a multiplicity of codes in shaping a narrative, with various degrees of deliberateness or consciousness. Many of these codes are to be recognized and used by a receiver in interpreting the text, if the interpretation is to be shared and accepted by other speakers. This retrieval of the author's meaning can happen with various degrees of awareness. There are no doubt many codes organizing the meaning of texts which have not yet been identified by theoreticians, although readers will use them intuitively. Not all the codes used by the author need to be identified or retrieved, even in this intuitive way. A portion of the meaning of the text is usually enough for the purposes of most readers and critics, who may, moreover, interpret the text according to codes which were not used by the author, and in this way construct new meanings. The legitimacy and value of these or any other meanings are defined in a specific discourse situation -- they cannot be determined a priori.

The intrinsic context of a work is a communicative context. It can be conceived as a virtual communicative situation, in which a textual author communicates with a textual reader. The concepts of 'textual author' and 'textual reader' derive from Russian and German formalism, ⁸ and more immediately from Walker Gibson's 'mock reader' ⁹ and Wayne Booth's 'implied author' and 'mock reader'. ¹⁰

The textual author is a virtual image of the author's attitudes, as presented by the text. The textual reader is a virtual receiver created by the author in full view of the actual audience he or she presumes for his or her work. The textual reader need not coincide with the author's conception of the audience: this reader-figure may be a rhetorical strategy, a role which the author wishes the audience to assume (or even to reject). Likewise, the reader's textual author and the author's textual author need not coincide any more than the meaning of the work for author and reader. But if communication is to occur these figures must have elements in common.

The levels of analysis just mentioned can be conceived as a series of semiotic strata, in which each level is the result of the application of a set of transformational rules to the previous level. A reader will consider the (verbal) text as a given, and will use it to construct the story. In its turn, the fabula is constructed on the basis of the story, by 'undoing' the transformations which gave rise to the latter.

The complexity of this enunciative structure can be exploited aesthetically in literature. A variety of displacements are possible. The subject required by the narrative act, the narrator, need not coincide with the subject of the fictional statement, the textual author. The narrator may be an entirely fictional figure, as in most first-person novels, or may coincide formally though not ideologically with the textual author -- an unreliable third-person narrator ( Booth, Rhetoric).

Many combinations are possible, and the differences between the various textual subjects may be clear-cut or extremely shady. This is to be expected, since the textual author, like the narrator, is not a substance but a discursive role. Discursive selves, permanent or provisional, proliferate in literature as much as in other modes of discourse. When we speak in our professional capacity, for instance, we use a set of discursive conventions to fashion an official persona, a provisional self designed for use in a given sphere of action. Irony is another way a speaker may modulate the presentation of self. Through the use of irony, a textual author creates a provisional, evanescent enunciator (subject) which does not coincide with the author's overall discursive self. More sustained irony will produce something like a hypothetical character, and by pushing this a bit further we can create a fictional narrator consistently differentiated from the author by means of ironic distance. Such narrators may use first- or third-person narrative strategies; they may write a (supposedly) factual narrative -- letters, memoirs, a diary -- or an explicit fiction.

The narrator's utterance is addressed to a hearer (reader) located at the same structural level: the narratee. ¹¹ Just as the various fictional narrators merge gradually into the textual author, the various narratees shade into the textual reader. This means that in some tales the differences between the narratee and the implied reader are crucial and dear-cut, while in others they are only latent.

So far we have identified a variety of textual figures, roles or subject-positions. They each perform an activity which has a direct object (if it is transitive) and an addressee: ¹²

We can also represent the structure of fictional narrative diagrammatically:

Figure 2

From the surface level of the linguistic text the reader constructs two kinds of interpretive scheme: the discursive schemata (the fictional speech situation, the textual senders and receivers) and the narrative deep structures (the narrative and the story). An interpreter will usually end by constructing some kind of literary statement or interpretation, which becomes for him or her the meaning or significance of the story.

Figure 3 represents this process of construction in a schematic way. The vertical and slanting double arrows indicate that the process of interpretation is not linear; it does not proceed neatly from one level of the textual structure to the next. Instead, there is a constant feedback between interpretation of the action, of the narrative structure and of the textual subjects. Differences in construction of the implied authorial attitude therefore often result in different constructions of the action. We have used one-way arrows in the first step of the process in Figure 3 to emphasize that the first substantial contact of the text rests on verbal denotation; but it is clear that even the denoted meaning of the text may be subject to revision once construction of the other narrative levels is under way. Nothing in the work is fully given from the start: everything is subject to revision and interpretation.

An historical overview

History of narrative and history of narratology

It is not clear yet what a history of narrative as such might be like. There are of course histories of the novel, of film, even of historical study itself; but a more general history of narrative forms would have to be an interdisciplinary achievement. Most work in narratology consists of synchronic formal analysis, and the discipline still needs to develop a comparative and historical perspective on narrative genres and structures. The history of the discipline of narratology itself is also largely unwritten. What follows sketches its development through the early prescriptive poetics of specific genres, through formal and structural analysis, to recent trends which stress the relationship between representation and specific ideological or cultural forms. A study in depth of the historical development of narrative genres would result in a cultural narratology, the study of narrative forms in their relationship to the culture which generates them. Cultural studies, derived from neo-Hegelian, Marxist, feminist or Foucauldian sources, open up a new speculative area for narratology that lies beyond the classification of formal devices and the purpose of this book. Narrative theory before 1950

Classical and post-classical

In Plato Republic we find the groundwork of genre theory and of the analysis of literary enunciation. ¹³ After expounding a theory of art as mimesis, Plato's spokesman Socrates discusses the style of poetic compositions: 'All mythology is a narration of events, either past, present or to come. . . . And narration may be either simple narration, or imitation, or a union of the two' (p. 27 ). That is, the poet may speak in his own voice (simple narration) or may speak through the voice of a character (imitation, mimesis). Tragedy and comedy are wholly imitative, while in dithyramb, lyric poetry and similar genres the poet is the sole speaker, 'and the combination of both is found in epic, and in several other styles of poetry' (p. 28 ). From a narratological viewpoint, this is the first theoretical approach to the problem of narrative voice, preceding a pragmatic approach to the text as utterance.

Aristotle Poetics further develops the formal approach to literature. Aristotle apparently takes for granted that 'serious' literature (of which tragedy is the highest form) is narrative -- in the wider sense of 'telling a story'. He roughly preserves the Platonic classification of genres on the basis of enunciation, distinguishing the wholly imitative form of dramatic performance from the mixed presentation of epic narrative. Within the genres based on incident and event (narrative proper and drama), the mythos, 'plot', or 'structure of the incidents' is for him the basic infrastructure; it is foremost among the aspects of tragedy (plot, characters, diction, thought, spectacle, song):

The most important of these is the arrangement of the incidents, for tragedy is not a representation of men but of a piece of action, of life, of happiness and unhappiness, which come under the head of action, and the end aimed at is the representation not of qualities of character but of some action. . . . And furthermore, two of the most important elements in the emotional effect of tragedy, 'reversals' and 'discoveries', are parts of the plot. . . . The plot is then the first principle and as it were the soul of tragedy: character comes second.

( Poetics VI, pp. 25-7)

The mythos, the plot of the tragedy, is defined as 'the arrangement of the incidents', so, strictly speaking, 'it is the plot which represents the action' ( Poetics VI, p. 25). We therefore have two possible ways of looking at a tragedy, two possible levels of analysis of the story which is being represented. On the one hand, it is an action (praxis), just as our daily activities may be described as actions. On the one hand, it is a plot (mythos), an artistic structure which the poet builds out of the action. That is, on the one hand we find mere incidents, on the other, the disposition of incidents. The poet is the maker not of verses or of incidents but of this important intermediate structure, the plot. We may note that Aristotle did not include action as a separate constituent of tragedy; he probably felt that the presence of plot in that list of parts accounted for both of them.Critical traditions in any cultures usually begin with foundational texts which, as Miner has noted, normally take one specific literary genre, most frequently 'lyric' poetry, as the model or epitome of all literature. ¹⁴ In spite of the centrality of narrative to all cultures, we find that only one major critical tradition (the Japanese one, deriving from Murasaki Shikibu Tale of Genji) takes narrative as the central literary genre. Significantly, Aristotle Poetics is based on drama, specifically tragedy. Many aspects of the analysis of narrative form sketched in the Poetics remained largely undeveloped until the twentieth century, when they were taken up by the formalist and structuralist critical schools.The classical discipline of rhetoric was not primarily concerned with narrative. Nevertheless, it provided many insights into the mechanism of style and composition which were gradually incorporated into the analysis of narrative. The most influential treatises on rhetoric follow a standard development of the discipline, concentrating on:

The Renaissance and Enlightenment periods developed their own literary systematics often through a re-reading and expansion of classical treatises. The Aristotelian theories of Robortello, Scaliger and Castelvetro constitute a gigantic step forward in terms of the detailed discussion of formal issues. ¹⁷ Countless other treatises discuss the poetics of tragedy, epic poetry, romance, and so forth. Although we may disparage the neoclassical obsession with generic laws and rules, as well as the prescriptive nature of these treatises, we should not overlook their increasing analytical power. For example, debates on the famous 'three unities' of drama helped to refine such basic concepts as the opposition between represented and representational time, narrative ellipsis and compression, the use of physical space to signify fictional space, or the relationship between dramatic illusion and convention. G. E. Lessing Laokoon, for instance, remains within the neoclassical and prescriptive episteme, but its analytical subtlety and its wealth of conceptual abstraction foreshadow later developments in aesthetics and semiotics. Lessing defines literature as an art intrinsically conditioned by the temporal sequentiality of linguistic signs (while the plastic arts use spatial signs). This abstract definition is the starting-point for practical analysis of such issues as the immediacy effect, dramatization and the use of point of view. ¹⁸

The theory of the novel was neglected during the emergence of the genre in the seventeenth and eighteenth centuries. Novels or 'romances' are not to be found in the classifications of Boileau or in the criticism of Dryden. In the eighteenth and early nineteenth centuries such critical statements as do exist are usually far behind the criticism of poetry in theoretical development. Nevertheless, the appearance of the novel leads to a qualitative step forward in analysis. From the point of view of theory, the new form invites a new paradigm, asking new questions and demanding new answers. Novelists are among the first to make significant statements about their craft. Innovative works themselves were such statements: the novel is an intrinsically parodic genre, and the best novels are very often a parody of, or at least a commentary on and interpretation of, previous modes of fiction writing. Don Quixote is often defined as a satire on romances, and this implied critique can be read everywhere in fiction, more explicitly than in drama or poetry. Great novels have always been to some extent metafictions, or anti-novels.

Henry Fielding, the great heir of Cervantes in the British scene, integrates commentary and fictional writing, most obviously in Tom Jones, where each book is headed by one introductory chapter of commentary. Fielding calls himself 'the founder of a new province of writing' with independent laws, ¹⁹ the 'comic epic poem in prose' ²⁰ -- a deliberately paradoxical formulation, stressing both that the novel is born out of the convergence of diverse genres and that it is essentially parodic in nature: a way of setting previous conventions of writing against one another. More specifically, the novel is the parodic genre which results from setting the conventions of epic and romance against prosaic reality.

Additional valuable insights are provided by Samuel Richardson and Laurence Sterne. ²¹ Drawing on the classical distinction between pure, imitative and mixed narrative, Richardson distinguishes between first-person narration (in which the writer tells of his own adventures), epic narration (controlled by what we would call an authorial narrator) and a technique which is more dramatic, introducing dialogue and direct speech. Richardson especially values this dramatic mode, clearly with his own epistolary technique in mind. Sterne pushes narrative experiment to a limit: Tristram Shandy can be read as an entertaining and often teasing commentary on the way narrative expectations are created and frustrated.

German critics, such as Friedrich Schlegel and G. W. F. Hegel, provided some of the earliest theoretical approaches to the novel. In his Brief ber den Roman, Schlegel voices in an explicit way the notion of the novel as a medley or convergence of all previous literary genres. For Hegel, the novel is the modern version of the epic: both bourgeois and subjective -- a result of the turn of romantic literature towards subjectivism and reflexivity. Hegel's theories were developed in the twentieth century by the young Lukcs, one of the main theorists of narrative realism, who sees the novel as the product of bourgeois demythologization of aristocratic ideals. ²²

The aesthetics of realism

The early theory of the novel was formulated for the most part under the realist aesthetics of the nineteenth century. While in this century the theory of the lyric is expressive, the theory of the novel remains largely mimetic. That is, the lyric is defined as an expression of the poet's feelings, and its representative element is subordinated to this expressive function (hence the 'pathetic fallacy', the projection of subjective passions on to the landscape). Victorian theorists of fiction often drew an opposition between the genres of the romance and the novel. Romance was light entertainment, making free use of fantasy and stirring adventure. The novel, on the other hand, was on the way to becoming 'serious' narrative, through its aesthetics of verisimilitude. This theory often originated with the novelists themselves, such as Balzac in France. In England we find an eloquent defence of realism in the essays of George Eliot and George Henry Lewes, who attack popular fiction and contend that realism is a novelist's moral responsibility. ²³

Among the basic terms of analysis in the realist aesthetic are plot, character, setting, theme, moral aim and verisimilitude. For realist critics, a novel should have a good construction, starting with a coherent plot. The distinguishing trait of the novel, however, is not the plot but its mimetic aim in the depiction and characters and setting. Many realist critics see deliberate plotting as a somewhat extraneous element which may distort the spontaneous revelation of character. In the Victorian age, the term 'novel of character' is often used as a hallmark of narrative quality: the novel of character is privileged over the simple-minded 'novel of action' or romance. Sometimes it is the regional specificity of the characters and setting that is emphasized, as in the late nineteenth-century school of 'local colour'. The realist novel, in any case, should be a psycho-social study, one that reveals new truths about human feelings and relationships. Such a novel has a theme and is linked to a well-defined moral intention, an authorial stance towards that theme, which is easily identified, whether it is conveyed by direct or by indirect means. It is this moral intention that makes realism something more than an attempt at copying nature.

Bulwer Lytton essay 'On Art in Fiction' is a typical nineteenthcentury approach to the narrative specificity of the novel. ²⁴ Lytton gives great weight to the author's intended effect and deliberate manipulation of the materials. It is essential to have a plan in a novel: an artistic shape, although this shape does not coincide with that of drama. For instance, the novelist uses description, not used by the playwright, which as Lytton notes is integrated with plot and character. He also argues that since the plot of a novel is less tight and less guided by cause and effect than that of a play, the new genre should be understood to have its own autonomous poetics. The novelist has interests other than those of the playwright: in character study and range of intimate emotions.

One of the most influential nineteenth-century approaches to narrative construction is found in Edgar Allan Poe's theory of the short story. Poe identifies the unity of a work not so much in the structure of the work itself as in its effect on the reader: the unity of the reading experience is essential for the unity of the work. He sees the novel as a genre devoid of a true unity of impression, while the short story is for him the most artistic prose genre, because the unity of effect on the reader can be calculated and preserved. ²⁵ Poe advocates a hyper-conscious theory of writing: everything is controlled by authorial intention. The end of the work must be complete in the writer's mind before actual composition begins: there is to be no improvization or change of plan during the writing. Thus, Poe situates himself in opposition to novelists like Dickens or Trollope, who often worked without a pre-established plan and improvised their plots in the making. Poe's conception of narrative is centred on its closure: 'It is only with the dnouemmt constantly in view that we can give a plot its indispensable air of consequence, or causation, by making the incidents, and especially the tone at all points, tend to the development of the intention.' ²⁶

We have seen that early theorists like Lytton placed some emphasis on defining the conventions of the novel as such, and on distinguishing the narrativity of fiction from that of drama. However, many novelists have praised narrative techniques which approach the effects of drama. Certain comments on narrative technique by Richardson, Stendhal or Dickens are interesting forerunners of Henry James's theories of fictional form because of the value they set on the dramatic elements of the novel: writers should not tell the whole of the story in their own person, but should show it -- make their characters tell it by means of dialogue and action. Stendhal proudly notes that all other novelists tell the story, while only he shows it to the reader.

Studies of narrative technique developed piecemeal, in Switzerland with Edmond Scherer and in Germany with Friedrich Spielhagen. ²⁷ Spielhagen adumbrated a fully-fledged theory of 'dramatic' narrative technique especially suited to psychological realism.

With Henry James, likewise, we witness the evolution of realism towards subjectivism and perspectivism, in part because of James's psychological bent. According to James, the novel (unlike drama) can reveal to us the inner life of characters, and this is the essence of the genre, which otherwise must follow, in his opinion, a dramatic ideal of concentration. ²⁸ But the novel is a free form, he says. It has no grammar which can be defined, no rules that can be taught, because it is 'a personal, a direct impression of life' (p. 664). Execution and intensity of impression are the grounds of its value, and they cannot be defined. They stem directly from the personal way each novelist sees life. In James's essays and prefaces to his own novels we find some of the clearest and most influential statements of the period on point of view and narrative voice, as well as on action and character.

James makes a distinction between voice and point of view in his novelistic practice as well as in his rhetorical statements. This distinction arises from his concern with the novel's ability to depict experience and psychological life. First-person narrative is not adequate for his purposes, because he is not looking for a conscious revelation of character, or for a novel based on recollection of past experience, which is what first-person narratives are most suited to reveal. His novels are usually written in the third person, which is less 'intrusive', more 'dramatic'. The story should in any case unfold in a transparent way without the writer stepping in to make his own comments. Rather than being simply 'told', we are 'shown' action and character as they develop through significant scenes. And there is an ideal way of 'showing' in third-person narration which is at once dramatic and psychologically immediate. This is what James calls narration through 'centers of consciousness' (Preface to The Portrait of a Lady), 'vessels of sensibility' or 'reflectors' (Preface to The Wings of the Dove), which many narratologists now call focalizers. The scenes usually act on a perceiving character, a reflector or focalizer, whose psychological reaction, the development of his or her understanding of the action, contributes to the organic unity of the plot. This is the role of Strether in The Ambassadors, or of Maisie in What Maisie Knew. James does not consider it necessary, as do some of his followers, to avoid changes of perspective during the narrative, but he does seek to cut the story into perspectival blocks that are internally coherent. For instance, in The Wings of the Dove, the story of Milly Theale is seen mainly through the eyes of two characters, Merton Densher and Kate Croy, as well as through Milly's own eyes. Every change of point of view, James says, has its aesthetic justification, its dramatic coherence, but it is essential for a novel to establish a 'register', a set of perspectival rules to which it consistently adheres.

Just as Aristotle argued that an action or praxis had to be treated artistically before it became the plot or mythos, James distinguishes between the 'subject' and the 'wrought material' or novel, thus prefiguring the Russian formalists' opposition between fabula and siuzhet. The 'register' defines the relationship between the material and the finished novel. Form and psychology converge: the dramatic form allows the reader a new insight into the characters' perception and interiority. James conceives of the rules governing point of view as organic and internal, springing from the very nature of the psychological material of the novel. The influence of James's ideas is readily apparent in most important twentieth-century writers on fictional technique and point of view: Percy Lubbock ( The Craft of Fiction, 1921), Cleanth Brooks and Robert Penn Warren ( Understanding Fiction, 1943), Jean Pouillon ( Temps et roman, 1947), R. K. Stanzel ( Typische Erzhlsituationen, 1954), Norman Friedman ( ''Point of View in Fiction'', 1955); Wayne C. Booth ( The Rhetoric of Fiction, 1961), Grard Genette ( Figures III, 1972), Mieke Bal ( Narratologie, 1977). ²⁹

Since the early phase of modernism ( James Conrad), theoretical reflection on narrative has consistently been opposed to experimental, avant-garde or at least 'highbrow' narratives. The various genres of popular fiction continue to rely heavily on the classical narrative devices of plotting and stereotyped characters. Still, critics have often valued the pleasures provided by narratives of action (as opposed to narratives of character, of point of view or of linguistic experimentation). R. L. Stevenson's defence of the romance, an answer to James's defence of psychological fiction, is a case in point. ³⁰

Early modernism

As we have seen, Henry James helped theorize the transition from Victorian realism to modernism. Critics like Joseph Warren Beach and Percy Lubbock were to systematize and popularize these ideas. ³¹

Beach coined the phrase 'exit author' to describe the new dramatic autonomy of the novel, whose action was to unfold directly under the eyes of the reader, without the mediating value judgements of the narrator. Percy Lubbock book The Craft of Fiction was something of an unofficial textbook of the modernist aesthetics of indirection. Lubbock draws an opposition between two methods, 'showing' and 'telling': 'The art of fiction does not begin until the novelist thinks of his story as a matter to be shown, to be so exhibited that it will tell itself.' ³² The aim of the novelist is to create a whole and full impression, to produce a controlled effect on the reader through the careful arrangement of form and subject matter. The aim is still to tell a story which is morally or metaphysically relevant, but the point now is that the reader must perceive and feel the story together with the character -- as an experiential process, not as a finished product seen from the outside. Related to this aesthetic position is Dos Passos' and Hemingway's concept of impersonal fiction, as well as Lawrence's injunction to 'trust the tale' instead of the teller or Joyce's image of the author standing apart from his creation, 'paring his fingernails'.

According to Lubbock, what makes the story be shown rather than told is a matter of composition, of the adequate treatment of the story material through the use of subjective point of view and scenic presentation. The novelist must use a coherent style: a consistent narrative mode. If the subject matter requires transitions between different modes they must be made smoothly: the seams must be invisible. Nothing must remind us of the novelist's presence. Everything that is told in a story must be motivated; that is, it must be there on account of some character's experience. Of course, in calling for a limited point of view Lubbock is also assuming a subject matter that is psychological in nature: some kind of personal drama, instead of the vast social frescoes of the Victorians. His model is Henry James The Ambassadors. Lubbock calls for conscious craftsmanship, an attention to composition, and the development of an adequate critical vocabulary to describe it. Similar ideas, the stress on character, point of view and composition rather than on plot, are also found in Ortega y Gasset's theory of the novel, and in various other modernist critics. ³³

Other valuable aesthetic approaches to the genre in the Englishspeaking world are E. M. Forster ever-popular Aspects of the Novel and Edwin Muir The Structure of the Novel. ³⁴ Forster studies the basic narrative elements of realist fiction, under such headings as 'story', 'plot', 'people', 'pattern and rhythm'. His emphasis, following what is perhaps the mainstream British tradition, falls on the depiction of 'people'; that is, of character. His classification of characters into 'flat' (based on one trait and therefore predictable) and 'round' (complex and lifelike) became universally accepted (incidentally, this conception was far from new -- the essentials can be found in neoclassical critics, such as Dryden). Edwin Muir's work, less popular with later readers, tries to establish different types of novel on the basis of their experiential treatment of time, action and point of view: the character novel, the dramatic novel based on conflict, or the chronicle, the wide-ranging multiplot novel of social panorama.

High modernism and New Criticism

Henry James's ideas (as well as those of Lubbock, Forster and Muir) were representative of the transition between the classical realist novel, with its emphasis on story, setting and character, and the modernist novel with its stress on writing and composition. In the 1920s and 1930s there was a widespread critical revolution against the aesthetics of late Romanticism. In literature this revolution is called modernism and is identified with a self-conscious avant-garde; in critical theory it was identified as New Criticism or as formalism. The modernist/formalist revolution had deep

consequences for the writing and criticism of all literary genres. The New Critics moved further away from mimetic considerations. They dismissed the Romantics and favoured lyric poetry that was complex, ironic and intellectualized. They criticized literature in terms of its structural complexity, not in terms of its immediate fidelity to life. That is, the aesthetic judgements of the New Critics tend to be intrinsic rather than extrinsic. A work is above all a pattern of words, a self-sufficient entity which constructs and manipulates emotions and thoughts that have only an analogical relationship to reality. It is a self-enclosed structure, meaning that any element has to be judged within the pattern, taking its function into account, rather than being identified in an immediate way with its equivalents in the historical world. Originally the New Critics did not pay much attention to fiction, although later we find readings of fiction in terms of tone, pattern, irony and balance. ³⁵ With the 'close reading' developed by William Empson and F. R. Leavis, the novel suddenly became a 'dramatic poem' -- its language became significant in terms of tension and image, like the language of poetry. ³⁶ Virginia Woolf claimed that modern fiction would assume the quality of a poem, and opposed fiction modelled on fact or report (like that of the naturalists). For her, as for other modernists, fiction must work through poetic suggestiveness rather than through narrativity. ³⁷

Plot and character as critical terms seemed to fade into the background, all emphasis falling on language and imagery, and on the overall pattern woven by all these elements. This intrinsic turn in critical thought eventually favoured the development of a reflexive theory of fiction, although it took some time for this to be explicitly formulated in the Anglo-Saxon world. For the time being, mimetic concerns still occupied the foreground, but mimesis had become internalized ( Erich Kahler spoke in this respect of an 'inward turn' of the novel). ³⁸ Critics from the 1930s to the 1950s paid particular attention to the modes of representation of inner life developed by the modernist novel, by Joyce, Woolf or Faulkner. Terms such as 'free indirect style', 'interior monologue', 'camera eye' narrative or 'stream of consciousness' occupy the centre of the critical stage. ³⁹ We shall not dwell long on this phase; our main concern here is with the next stage of theoretical development: the theory of the novel as it stood in the 1960s and 1970s. It was with the second wave of formalism, in other words with structuralism, that narratology underwent a wholesale expansion. But first we need to consider additional formalist approaches which prepare the ground for this development. Formalisms

Continental criticism witnessed an increase in formalist analysis of fiction, with the advanced work of German and Polish critics and the work of the Russian formalists in the 1920s. It is true that aesthetic studies simultaneously appeared in the English-speaking world, ⁴⁰ but in general theoretical speculation was more common in continental Europe. The tradition of theoretical reflection descending from Schlegel to Spielhagen and Walzel includes such interesting practitioners as Kte Friedemann. Her book Die Rolle des Erzhlers in der Epik is a fully-fledged narratological treatise written many years before the appearance of mainstream narratology. ⁴¹ She analyses the technique of early modernist 'dramatic' narration as formulated by Spielhagen (whom we could usefully compare to James in the Anglo-Saxon sphere) and contends that other types of narrative voice which allow for the narrator's intrusions or commentary are equally 'artistic' and useful for the novelist.

While German aesthetics is one of the main influences on the Russian formalist school, the systematic and functional approach to form developed by Shklovski, Propp or Tomashevski is considered by many as the inaugural statement of narratology proper. ⁴² The formalists react against both impressionist and historicist approaches to literature, and supplement their aesthetic background by rethinking Aristotelian insights, which they enrich with concepts borrowed from the new developments in theoretical linguistics ( Baudouin de Courtenay, Saussure, Jakobson). Key narratological concepts inherited from the Russian formalists are the opposition fabula/siuzhet (the source of the opposition fabula/story/ text mentioned above), or the concept of pseudo-oral narrative voice, skaz. Such concepts are not to be applied in a mechanical way: the aim of the formalists is to account for the organic effect of the work and the interaction of all its elements. For instance, in their discussion of 'realistic motivation' they argue that a technique such as epistolary narrative has a specific informative function vis--vis the reader, while it is simultaneously 'justified' or motivated by the story, therefore acting as a sign of realism. The work of Vladimir Propp on the Russian folktale, influential in both anthropology and literary theory, introduced such key analytical tools as the concept of 'narrative functions' and their organization into 'sequences'. Propp's work is a grammar of narrative which identifies the basic 'deep' structure underlying any number of 'surface' manifestations.

The most important contribution of the formalists, then, is one of general method: they aim at devising a general science of literature (narrative) capable of describing the systematics of literary forms and also of literary evolution. Form and function are intrinsically related: literary forms, for instance, may become worn and give rise to new forms (parody) while their previous social function is taken up by originally minor forms that evolve and come to the fore.

The systematic study of literary forms is characteristic of twentiethcentury criticism. The systematics may derive from linguistics and aesthetics, but also from philosophy as well as from comparative studies of anthropology, religion and myth.

Although there are a variety of philosophical approaches relevant for the study of narrative, we may single out phenomenology as the most akin to narratology stricto sensu. Phenomenology is a systematic study of experience. It approaches reality as a formal system of relationships, so leading naturally to the problem of additional subsystems such as literary works and fictional objects, and converging with semiotics in the study of sign systems and representations. Roman Ingarden The Literary Work of Art, which applies Husserl's phenomenology to the description of literary works, is a foundational text in this lane, ⁴³ usefully complementing structuralist descriptions of narrative form. In his later work Ingarden antedates many aspects of the reader-response approaches of the 1960s and 1970s. ⁴⁴ Critics such as Wolfgang Iser have further developed the phenomenological study of reception and reading in closer convergence with the structuralists. ⁴⁵ There is, in addition, a rather well-defined phenomenological approach to narrative which is associated with existentialist philosophy, represented by the work of Jean-Paul Sartre or Jean Pouillon.

Like literary phenomenology, 'myth-and-ritual' studies of narrative have a formal (narratological) dimension. Many literary critics have drawn inspiration from Sir James Frazer The Golden Bough, Carl Jung's studies on collective psychology or Joseph Campbell The Hero with a Thousand Faces. ⁴⁶ Frazer's book is an attempt at finding a common narrative structure beneath a wide range of myths and rituals. Campbell's book identifies the basic stages of archetypal narratives similar to those studied by Propp. A comparison of these books reveals striking similarities not only in the object of study but also in the abstractive nature of their approaches, although each is grounded in a totally different discipline and intellectual tradition.

Northrop Frye Anatomy of Criticism is yet another major treatise on archetypal plot structures. ⁴⁷ Indeed, Frye constructs a complex interpretation of all literature in which the different genres and modes are organized as phases in a narrative structure associated with the cycle of life. The creativity, learning and subtlety evinced in this work have been rewarded by its immense influence. Like many other

works not strictly narratological, Frye Anatomy is essential reading for any student of narrative.

Myth criticism is often combined with early psychoanalytic approaches. Sigmund Freud himself devoted some attention to the psychoanalytical interpretation of narrative literature as well as to the narrative dimension of psychoanalysis. ⁴⁸ Early analyses based on Freud's work lay more emphasis on the former, that is, on mechanisms of identification in reading, the writer's fantasies of sexuality and power, or the 'pathological' origin of plot structures and patterns of images or motifs. By contrast, present-day psychoanalytic criticism often privileges the second perspective, that is, the fantastic element in the analyst's (or the critic's) interpretive activities, the formal and institutional constraints of the diagnosis, and so on.

Contemporary narratology

Comparative narratology

Interdisciplinary studies of narrative afford a promising area for future development. A 'comparative narratology' -- in the sense of 'comparative literature' -- addresses such matters as the structural differences of given narrative genres or sub-genres, the phenomenological difference between narrative and other literary and artistic phenomena, and the comparative poetics of different cultures and traditions. Interdisciplinary narratological studies also try to strengthen the ties between narratology and other critical endeavours, such as the theory of interpretation or reception, women's and gender studies, deconstruction. The interdisciplinary direction offers the most interesting avenues for development, although it often leads beyond narratology proper. ⁴⁹

From a strictly narratological perspective, we may classify theories of narrative according to their main object of study within the narratologically defined structure of the text. A theory is always a limited model which isolates or gives preference to certain features of the object of study. Most theories of narrative, therefore, privilege either narrative as process or narrative as product, the level either of the fabula, of the story, or of textual representation.

Theories of authorship

Authorship and literary production have been a traditional area of literary research. In the nineteenth century, historical scholarship developed alongside the bourgeois conception of the writer as a detached observer of society. The historicist orientation, together with an individualist conception of writing, are still largely dominant today. Roughly speaking, we can distinguish two tendencies. One is towards impressionistic criticism, often biographical in tone and interested in the personality of the authors, their lives and at best the quality of their imagination and style, the comparison of different works and the study of influence. This kind of criticism had its academic heyday in the work of Walter Raleigh or David Cecil, ⁵⁰ and is perhaps the tone most often associated nowadays with journalistic reviews and interviews. The other is more concerned with literary history: the demarcation of literary periods and their defining characteristics, the study of schools and movements, the interplay between literature and other cultural phenomena (from Saintsbury, Ker and Lanson to contemporary theorists of postmodernism.) ⁵¹ French, German and often American critics are more prone than the British to grounding literary scholarship on contemporary theories of history, whether positivist ( Taine), evolutionist ( Brunetire), classicist ( T. S. Eliot), Hegelian (the early Lukcs), Marxist (the later Lukcs, Weimann), Weberian ( Watt) or structuralist ( White). ⁵² The specifically narratological aspect of these theories often lies in the links they establish between the narrative process and the 'master narrative' of history they use as their basic framework.

Psychoanalysis is another major source for theories of authorship. We should keep in mind here that Freudian theories, too, provide a master narrative of the development of the self. The critic's aim is usually to relate stylistic or thematic patterns in the work to the structure of an author's personality, evaluating the interplay between conscious and unconscious influences. The influence of psychoanalysis on studies of narrative is too pervasive to attempt even a limited overview. There is, besides, the closely related issue of psychoanalytic treatment considered as a narrative process whose patterns and symbolism can be approached much like those of a literary text.

Theories of enunciation

Theories which foreground the study of enunciation and its protocols are related to twentieth-century formalist criticism (stylistics, Russian formalism, New Criticism, the Chicago school, structuralism and literary pragmatics). They try to identify different strands or levels in the voice of the text: it is no longer, or not only, the voice of the author which is analysed, but also a number of fictional masks of personae which mediate between the authorial voice and the characters. Such theories differentiate the historical author from the narrator and the implied author; Wayne Booth The Rhetoric of Fiction is a seminal text in this respect. Linguistic theories studying the articulation of subjectivity in language ( Benveniste, Austin, Bakhtin) are often applied to the analysis of narrative enunciation. ⁵³ Narrative pragmatics may also focus on the enunciative pecularities of fictional and non-fictional narrative. The specific rhetorical devices of each of the textual voices help define the style of the text.

From the point of view of literary pragmatics, fictional narrative is a second-degree discourse activity (or complex speech act), whose understanding presupposes the understanding of more primitive or literal discourse situations from which it derives. In the early classifications of speech acts there was no place for fictional narrative. This is not surprising, since these classifications were not really concerned with actual speech acts, but with idealized or normative speech-act types. That is why Austin or Searle could afford to posit a sentence-grammar as the basis of their studies of speech activity. The study of real discourse, however, must perforce be based on a textual grammar, and is bound to yield somewhat less clear-cut results. Writing a novel is obviously a kind of speech act, but its specificity has to be captured by a theory of discourse which would take into account the real circumstances and contexts in which novels are written. Writing a novel, or writing fiction, is not a 'statement', though it is a derived act of a kind which has statements as its remote ancestor in a structuralist/genetic conception of speech activity. For practical purposes of analysis, writing a novel is that kind of speech act called 'writing a novel': linguistics at this point shades off into the literary theory of genres. Between the linguistic speech act called 'statement' and the literary speech act called 'novelwriting' several conceptual steps could be distinguished, among them a study of the fictional statement as derived from the literal statement and a study of the narrative as an extended statement derived from the simple sentence. ⁵⁴

Theories of action or fabula

The theory of action is a philosophical discipline, with work ranging from traditional ethical treatises ( Aristotle Nicomachean Ethics) to modern approaches in the fields of formal logic (Von Wright), hermeneutics ( Ricur) or cognitive psychology ( Goffman). ⁵⁵ Such theories focus on the study of events, action sequences and schemes, functions, actants, characters, settings and the internal laws of narrated worlds. In the realm of literary narratology we find some influential developments in the Russian formalist school. Vladimir Propp develops a model for the functional classification of events and the description of action sequences at fabula level. He draws a clear distinction between, on the one hand, the text itself, which is the manifest level and, on the other, the abstract level of function sequences and spheres of action of the characters (similar to Greimas' 'actants'). ⁵⁶ Drawing on Veselovski's definition of plot or narrative (siuzhet) as a succession of basic unanalysable units, the 'motifs' (a subject plus a standardized action), Tomashevski distinguishes two means of classification. ⁵⁷ First, he differentiates linked motifs from free motifs. Linked motifs are indispensable for the identity of the fabula; free motifs are not important and can be altered without any significant change in the fabula. Secondly, he opposes static to dynamic motifs. Descriptions, for instance, or unimportant actions, are static, while significant actions are dynamic. Barthes and other structuralists further refined these concepts. While the essay by Barthes included in our selection of texts is interesting in many other respects, it has been especially influential as a model for the analysis of narrative actions. Several formalized models, drawn by analogy with linguistics or formal logic, were devised by structuralist critics such as Bremond, Greimas, Todorov and Prince. We include selections from the work of Bremond and Greimas. Many additional formalist theories approach the analysis of action structures, their generation, variations, and possible classifications. ⁵⁸ Among recent developments of action theories, the most fruitful contributions are those which bridge the gap between literary narratology and cognitive psychology (e.g. in Goffman, Bordwell, Branigan). ⁵⁹ The study of action patterns in terms of schemata, scripts or frames allows for the formulations of a common model for all narrative actions (both 'inside' the fabula and 'outside', at the levels of story construction and the process of reading). It also permits links to be established between narratology and artificial intelligence. ⁶⁰

Theories of story and narration

Many theorists deal with, and many place their main emphasis on, the intermediary structures of story construction and narration. Such theories constitute the traditional narrow core of narratology. They devise modes of analysis of the time structure of the story (order of events, temporal distortions such as flashbacks or flashforwards, duration and selection of scenes, narrative rhythm, etc). The study of point of view (focalization, dramatic irony, suspense, omniscience) and of presentational mode (showing/telling) also fall within the scope of theories of story structure, as does the analysis of characters' discourse (free indirect style, dialogue presentation) and narrative voice. The most influential approaches to story construction and narrative voice were developed by formalist critics (Russian formalists, students of stylistics, New Critics, structuralists). Although they are too numerous for a fair selection, we may refer the reader to key concepts coined or developed by Henry James (reflectors, restricted point of view), Lubbock (scenic or panoramic presentation, showing/telling), Bally (free indirect style), Booth (reliable and unreliable narrators, moral or intellectual distance), Genette (focalization, narrative levels, anachronies, homodiegetic and heterodiegetic narrators, the narratee), Bal (the focalizer, levels of focalization), Bakhtin (dialogism, poliphony, heteroglossia), Dujardin, Humphrey or Cohn (stream of consciousness), Fowler (mind-style). ⁶¹ As these studies make up the main body of our selection, we refer the reader to the texts in question, where many of the central narratological concepts are analysed in detail. We have included extracts from works by Culler, Sternberg, Bal and Ricur which analyse the relationship between the represented sequence of events and the representational structures of the story from the point of view of causality, temporality and perspectival presentation. The two extracts in our brief section on film, by Branigan and Deleyto, also focus on such representational structures. The selections grouped under 'Text', on the other hand, focus on the linguistic surface of narrative and the communicative activity of the narrative subjects (author, reader, narrator, characters). In this section we include key contributions from Booth, Gibson, Stanzel, Genette, Prince and Hutcheon.

Theories of reception

Narrative is, among other things, a communicative speech act, a message transacted between a sender and a receiver. Poststructuralist critical schools have developed the analysis of the reader's role in literary communication, stressing the active and creative nature of reading and, generally speaking, of understanding. Early narratological theories usually took the reader's role for granted, but successive elaborations have approached the reader's role from different perspectives, and reader-figures have proliferated. Walker Gibson speaks of the 'mock reader', a readerimage which is inscribed in the text by means of presupposition and ideological assumptions. ⁶² Nowadays the term 'implied reader' is more commonly used. The text may also feature a fictionalized version of the reader, a 'narratee'. ⁶³ Narratees may range from fictional characters to less obtrusive fictional addressees to the 'mock' or implied reader.

A different line of inquiry consists in analysing the text from the reader's viewpoint. The static structure of meanings becomes suddenly animated: form becomes a sequential process of construction, a series of provisional hypotheses in the mind of the reader. Total form is not the final 'figure in the carpet' the reader makes up once the book is finished, but rather the progressive formulation and discarding of structural hypotheses during the reading process. ⁶⁴

The creative aspect of reading becomes more and more prominent as we move from theories that deal with the construction of action and story structures to theories which try to analyse the ideology or the significance of the text considered as a literary statement. Contemporary theories of reading (deconstruction, hermeneutics, reader-response criticism, reception theory) have systematically questioned the possibility of an objective, formalized analysis of this ideological level. Their contention is that the ideology or significance of a text is not a structure, but a process, the result of the interaction of text and reader. Brooks's text in our selection relates the reader's construction of the plot to the play of desire and basic psychical drives as defined by Freudian psychoanalysis. Some of these enquiries lead towards social, political and cultural studies. Marxist and feminist post-structuralism tries to show how reading is a political act involving the acceptance or questioning of preexisting representations, ideologically charged models of interpretation and presuppositions. De Lauretis' text in our selection is a re-reading of some previous formalist models in this light, as well as a good instance of the convergence of psychoanalysis, narratology and feminist ideological criticism. Other schools lay more emphasis on the phenomenology of understanding. They analyse the significance of texts as it relates to the figural nature of language (deconstruction), to the experience of otherness and tradition (hermeneutics) or to contemporary developments in psychology, artificial intelligence, linguistics and cognitive science. ⁶⁵

Theories of self-referentiality and intertextuality

Theories of literary reflexivity can be traced back to the German romantics, ⁶⁶ and have become increasingly important in the second half of the twentieth century. Much structuralist and post-structuralist criticism is concerned with experimental fiction, the analysis of reflexive fiction or metafiction perhaps being the most closely related to narratological concerns. ⁶⁷ By metafiction we mean fiction which experiments with its own form as a way of creating meaning. We usually associate the novel with realism, just as we usually associate art with representation. But realism, the mimetic pull, is always counterbalanced by the fictionality of the novel, establishing a characteristic tension between the strategies of fiction and the drive towards realism. The tension between reality and mimetic representation is perhaps the best starting-point for a discussion of metafiction. Instead of taking reality or realism for granted, the reflexive novel explores the epistemological foundations of both, lays them bare, opening the way either for a more self-conscious realism ⁶⁸ or for something else. Metafiction, then, can be defined as a way of writing, or more precisely as a way of consciously manipulating fictional structures, of playing games with fiction. Metafiction as writing would constitute a specific sub-genre in which the reflexive element is the dominant one. Robert Alter sees a difference between self-conscious novels and novels which contain self-conscious moments. ⁶⁹ The self-conscious or reflexive novel must be informed by a consistent effort: self-consciousness must be central to its structure and purpose. The term 'reflexive' calls our attention both to mirror structures (doublings, analogies, frames, mise en abyme) ⁷⁰ and to thought, consciousness, reflection, awareness accompanying action. Indeed, metafiction is reflexive fiction in the sense not only that mirror images are found in it, but also that these mirrorings and reflexive structures are used as a meditation on the nature of fiction. Alter defines the self-conscious novel as 'a novel which systematically flaunts its own condition of artifice and by so doing probes into the problematic relationship between real-seeming artifice and reality'. ⁷¹

But this definition of metafiction is a text-centred one: it leaves out many reflexive phenomena in the literary process. Metafiction may also be defined as a way of reading. The 'reflexive reading' of texts is a critical paradigm which provides a new way of exploring meaning in literary works, unearthing structures of meaning which may be automatic, not deliberately planned by the writer but rather reached by the reader through the spontaneous play of meaning or the interplay between writing and reading.

Metafiction challenges a common critical assumption: that the work is silent about itself and waits for the critic to interpret it. A metafictional work will often be outspoken enough about its aims and technique. It criticizes previous literary conventions (to some extent all literature does this), but also its own conventions. All modern art shares this self-conscious attitude to the past, this anxiety deriving from its belatedness. Josipovici argues that 'There is no better way of defining the achievement of Picasso, Stravinski or Eliot than to say that it is an exploration both of the medium in which they are working and of the traditional exploration of that medium.' ⁷² The artist's exploration of the possibilities of art is perhaps the main characteristic of the twentieth century: it is not surprising that these explorers transgressed many borders and reached many dead ends. Hutcheon's typology of reflexive devices, included in this reader, is a good instance of the structuralist approach to reflexivity.The structuralist concern with reflexive texts is both questioned and developed by the deconstructive school. Generally speaking, a deconstructive reading of a text (any text) will trace its staging of its own metaphysics only to undermine this by showing how it leads to contradiction, paradox and aporia -- that is, how the text effects its 'self-deconstruction'. Hillis Miller's contention in 'Line' (pp. 286 -95 below) is somewhat more ambitious, as it attempts to trace the aporia underlying any use of narratological terminology.Metafiction often amounts to an act of criticism of previous traditions, because of its links with parody and self-consciousness. A discussion of reflexivity therefore cannot be isolated from intertextuality, the theory of which asserts that no text exists as an autonomous and self-sufficient whole: the writer's and the reader's experience of other texts conditions its form and interpretation. Classical accounts of narrative ( Aristotle, Horace) usually assume that the poet as a plot-maker will use traditional stories, rather than invent the action of the work. Plot construction was traditionally discussed within the context of intertextual transformations, so that intertextual studies are anything but new. The term itself, however, was coined by Julia Kristeva, who derived it from Mikhail Bakhtin's work on social linguistics. ⁷³ Bakhtin defined the self and society as being in a dialogical relationship. Individual and social languages define each other: any individual discourse is traversed by social discourses (a state of affairs Bakhtin calls 'heteroglossia'). Some modes of discourse, such as the novel, are especially open to this play of competing ideological voices, hence the 'polyphonic' nature of the novel for Bakhtin. Intertextual analysis leads, therefore, both to the analysis of genre conventions and to ideological and cultural studies.In sum, we can distinguish several types of intertextual relationship, according to the reader's perception of the relations existing between a given text and:

The critical commentary of the text (which frequently incorporates types 1, 2 and 3). The most sophisticated analyses of this type of intertextuality have been carried out by contemporary theorists of interpretation (deconstructivists, reader-response theorists, etc), who have systematically problematized the relationship between a text and its subsequent readings.

Applied narratology

Under this heading we cover theories that formulate relations between narrative, society, history and ideology. While narratology has usually been considered an 'intrinsic' approach to literature, it is possible, in a wider perspective, to see a relationship between the study of narrative structure and extrinsic approaches to literature which assess its relations to history, social structures, institutions and cultural phenomena in general. There are many different schools of cultural criticism: Weberian, Marxist, Foucauldian, gender studies, multicultural studies. All of them in one way or another use the concept of ideology as a critical tool. Ideology is a problematic term, since its definition depends in part on the ideology of the person defining it. The original Marxist definition understood 'ideology' as false consciousness -- a set of superstructural representations of society and culture which tried to perpetuate the status quo. Ideology was therefore a concept opposed to the scientific social description provided by Marxism. Later approaches (starting within Marxism itself in the work of Bakhtin) conceptualize ideology as a necessary network of semiotic representations: we cannot have access to reality in an ideology-free way. A corollary of this view is that ideology is not a substance of some kind we find in a text, but rather a relational function between the text and the reader.

In The Rise of the Novel ( 1957) Ian Watt explains the appearance of the novel on the basis of ideological changes brought about by bourgeois hegemony in the eighteenth century. He relates social change to such structural elements as increasing accuracy in the depiction of individual character, and concrete social settings and plot circumstances which build up realism as a mode of representation.

Marxist critics have studied the ideological function of narrative genres in the self-representation of social classes and as an ideological manifestation of social conflicts. We might single out Fredric Jameson's work on narrative. ⁷⁶ Jameson incorporates and reinterprets structuralist narratology in order to study narrative as a 'socially symbolic act', as an imaginative (and ideologically motivated) response to class struggle. For him, narrative is the main structuring principle in the articulation of desire as well as of political and cultural representations.

The question of narrative and desire is also central to psychoanalytic approaches. Freud himself inaugurated the psychoanalytic study of narrative functions. In 'Creative Writers and Daydreaming' Freud analyses the function of the hero and of point of view as fantasies of power and desire. He also analysed individual works. ⁷⁷ But perhaps his most fundamental contribution to the analysis of narrative is an indirect one: he conceives the whole process of the development of the self, as well as the process of psychoanalytic therapy, as narratively structured. Innumerable critics have built on Freud's insights in ways which are relevant to narratological analysis. Peter Brooks Reading for the Plot ( 1984), an extract from which is reprinted here, is an example of the present-day convergence of post-structuralist reader-response criticism and psychoanalysis.

Another extrinsic approach to narrative is gender studies. Among these, feminist criticism has undoubtedly been the most influential. There is an immense variety of feminist approaches to literature, and most of them have made significant contributions to the analysis of narrative. Feminist analysis of the constructedness of gender has had far-reaching effects on the study of authorship, character representation and plot structures. Some key works in this context include Kate Millett's Sexual Politics ( 1969), Ellen Moers' Literary Women ( 1976) and Sandra M. Gilbert and Susan Gubar The Madwoman in the Attic ( 1979). ⁷⁸ Feminist critics have also pointed out that narrative verisimilitude and coherence are gender-conditioned categories, ⁷⁹ and have denounced the patriarchal ideology inherent in plot patterns, implied authorial attitudes and implied readers, calling for a feminist 'resisting reader' (Fetterley). Teresa de Lauretis' Alice Doesn't ( 1984) develops many of these tenets from the perspective of poststructuralist psychoanalysis and applies them to the study of filmic narrative. An excerpt from this work has been included in our reader.

There are countless other significant extensions or applications of narratology to a number of disciplines. A philosophical variety of narratology is to be found in the work of Paul Ricceur. ⁸⁰ A writer in the tradition of philosophical hermeneutics inaugurated by Heidegger Sein und Zeit, Ricur develops his thought as a critique of modern rationalist approaches such as phenomenology and

structuralism. He insists that myth is an essentially temporal phenomenon, and cannot be reduced to a semantic matrix as LviStrauss contends. He considers narrative ordering as a fundamental human experience, a way of structuring human existence in time and of opening up the possibility of meaningful action. Ricur's work is a good example of narratology applied to philosophical and religious (Christian) hermeneutics.

The theory of history has always had to account for issues of representation, with the question of narrative as a central problem. The realization that modernist history leaves out of its account many facts that remain ungraspable or untranslatable informs the postmodernist concept of history, which thus becomes 'the problem of the past as unrepresentable burden'. ⁸¹ Consistently with this, the contemporary theory of history sets out to question the cognoscibility of the past and the ideological function of history writing in the selection of past events erased or recorded. According to the modernist interpretation, history is not a text at all but rather an idea which becomes accessible in textual form. By contrast, poststructuralist theorists of history such as Hayden White or Dominick LaCapra have questioned the validity of the modernist approach. ⁸² From their postmodernist perspective, historical meaning is the result of the inextricable unity of idea and textual envelope, of content and form, so that the study of the real can only be the analysis of the textual form reality takes when apprehended by subjects. In Metahistory ( 1973), for example, Hayden White studies the techniques of historical writing as versions of literary or mythical plots. His essay in our selection is also a good instance of the application of structural analysis of narrative to historical writing.

Conclusion

Much might be said about the possible developments of a science of narrative, because, as we have tried to show, narrative is a complex phenomenon whose analysis allows infinite perspectives. Many critics would argue that narratology should be understood as exclusively referring to formalist and structuralist analysis. Other critics envision narratology as an umbrella term, the meeting-place of multiple approaches to narrative, from the standpoint of a variety of disciplines: history, anthropology, psychology, cognitive science, hermeneutic philosophy, ideological criticism, and so on. In this introduction we have attempted to sketch an ample theoretical and historical conception of narrative theory. In our selection of texts, however, we have deliberately followed more restrictive criteria. First, we have only included texts from 1950 onwards. Secondly, the bulk of our selection is based on structuralist narratology, although we have also included a few examples of the most influential alternative theoretical approaches to narratology in the wider sense of the word. We are aware, however, that the differing narratives on narratology which we propose in the introduction and contents of this reader are just two of the countless possible plots in a garden of forking paths.

Notes

EARL MINER, 'Narrative', in Comparative Poetics: An Intercultural Essay on Theories of Literature ( Princeton: Princeton University Press, 1990, pp. 135-212.

For instance, by JOHN DRYDEN, 'An Account of the Ensuing Poem [Annus Mirabilis] in a Letter to the Honourable Sir Robert Howard' ( 1667), in John Dryden: Selected Criticism, ed. James Kinsley and George Parfitt ( Oxford: Oxford University Press, 1970), pp. 7-15.

GEOFFREY OF VINSAUF, Poetria Nova (c. 1200), trans. Margaret F. Nims ( Toronto: Pontifical Institute of Medieval Studies, 1967).

F. ROBORTELLO, In librum Aristotelis De arte poetica explicationes/Paraphrasis in Librum Horatii, Qui Vulgo De Arte Poetica Ad Pisones Inscribitur ( 1548); LUDOVICO CASTELVETRO , Poetica d'Aristotele vulgarizzata e sposta ( 1576) in Castelvetro on the Art of Poetry: An Abridged Translation of Ludovico Castelvetro's Poetica d'Aristotele vulgarizzata e sposta, ed. A. Bongiorno ( Binghampton, NY: Medieval and Renaissance Texts and Studies, 1984); JULIUS CAESAR SCALIGER, Poetices libri septem ( 1561) in Select translations from Scaliger's Poetics, ed. and trans, F. M. Padelford ( New York, 1905).

G. E. LESSING, Laocoon: An Essay on the Limits of Painting and Poetry ( 1766) trans. Edward A. McCormick ( Baltimore: Johns Hopkins University Press, 1984).

HENRY FIELDING, The History of Tom Jones, a Foundling (Ware: Wordsworth, 1992), vol. 1, p. 38.

FIELDING, preface to The History of Joseph Andrews ( London: Hutchinson, 1904), p. 8.

SAMUEL RICHARDSON, excerpt from Novelists on the Novel, ed. Miriam Allott ( London: Routledge and Kegan Paul, 1959), p. 258.

FRIEDRICH SCHLEGEL, Schriften und Fragmente ( Stuttgart: Krner, 1956); G. W. F. HEGEL , Esthetics ( 1835) trans. T. M. Knox ( Oxford: Clarendon, 1975); GEORG LUKCS , The Theory of the Novel ( 1916) trans. Anna Bostock ( Cambridge, Mass.: MIT Press, 1971).

GEORGE ELIOT, 'Silly Novels by Lady Novelists' ( 1856), rpt. in Victorian Criticism of the Novel, ed. Edwin M. Eigner and George J. Worth ( Cambridge: Cambridge University Press, 1985), pp. 159-80; GEORGE HENRY LEWES, 'Criticism in Relation to Novels' ( 1865), rpt. in ibid., pp. 181-92.

EDWARD BULWER LYTTON, 'On Art in Fiction (II)' ( 1838), rpt. in ibid., pp. 22-38.

EDGAR ALLAN POE, 'Nathaniel Hawthorne', in Poems and Essays ( London: Dent, 1927), pp. 177-94.

POE, 'The Philosophy of Composition' ( 1846), in ibid., p. 164.

EDMOND SCHERER, Zur Technik der modernen Erzhlung ( 1879); FRIEDRICH SPIELHAGEN , Beitrge zur Theorie und Technik des Romans ( Leipzig: Staackmann, 1883).

HENRY JAMES, 'The Art of Fiction' ( 1884, rev. edn 1888), in Adams (ed.), Critical Theory since Plato, pp. 660-70.

JAMES essays and prefaces are collected in The Art of Criticism: Henry James on the Theory and the Practice of Fiction, ed. William Veeder and Susan M. Griffin ( Chicago: University of Chicago Press, 1986).

R. L. STEVENSON, 'A Humble Remonstrance' ( 1884), in Victorian Criticism of the Novel, ed. Eigner and Worth, pp. 213-22.

PERCY LUBBOCK, The Craft of Fiction ( London: Cape, 1921; 1926 edn); JOSEPHWARREN BEAC

WARREN BEACH, The Method of Henry James ( 1918; rev. ed., Philadelphia: Albert Sciler, 1954.)

LUBBOCK, Craft of Fiction, p. 62.

JOS ORTEGA Y GASSET, The Dehumanization of Art and Other Essays on Art, Culture, and Literature ( 1925, etc.) trans. Helene Weyl ( Princeton: Princeton University Press, 1948).

E. M. FORSTER, Aspects of the Novel ( London: Arnold, 1927); EDWIN MUIR, The Structure of the Novel ( London: Hogarth Press, 1928).

CLEANTH BROOKS and ROBERT PENN WARREN, Understanding Fiction ( 1943; Englewood Cliffs, NJ: Prentice Hall, 1959).

F. R. LEAVIS, ''The Novel as Dramatic Poem: Hard Times'', Scrutiny 14 ( 1946-7): 185-204; rev. version in The Great Tradition ( London: Chatto, 1948).

VIRGINIA WOOLF, 'Modern Fiction' ( 1919), in The Common Reader [1st series] 1925 ( London: Hogarth, 1929), pp. 184-95.

ERICH KAHLER, The Inward Turn of Narrative ( 1970) trans. Richard Winston and Clara Winston ( Princeton: Princeton University Press, 1973).

CHARLES BALLY, 'Le style indirect libre en franais moderne', GermanischRomanische Monatschrift 4 ( 1912): 549-56, 597-606; MARGUERITE LIPS, Le style indirect libre ( Paris: Payot, 1926); L. E. BOWLING, 'What Is the Stream of Consciousness Technique?' PMLA 65 ( 1950): 333-45; MELVIN J. FRIEDMAN, Stream of Consciousness: A Study of Literary Method ( New Haven: Yale University Press, 1955); NORMAN FRIEDMAN, ''Point-of-View in Fiction: The Development of a Critical Concept'', PMLA 70 ( 1955): 1160-84.

For instance: VERNON LEE, The Handling of Words and Other Essays in Literary Psychology ( 1923; Lincoln: University of Nebraska Press, 1968); BLISS PERRY, A Study of Prose Fiction ( 1902; Boston: Houghton Mifflin, 1920); JOSEPH WARREN BEACH , The Twentieth-Century Novel: Studies in Technique ( New York: Appleton, 1932).

See the selections from essays by VIKTOR SHKLOVSKI and BORIS TOMASHEVSKI in Lemon and Reis (eds and trans.), Russian Formalist Criticism, and VLADIMIR PROPP's Morphology of the Folktale ( 1928) trans. Laurence Scott. 2nd edn, rev. and ed. Louis A. Wagner ( Austin: University of Texas Press, 1968).

ROMAN INGARDEN, The Literary Work of Art: An Investigation on the Borderlines of Ontology, Logic, and Theory of Literature ( 1931; 3rd edn 1965; Evanston, Ill.: Northwestern University Press, 1973).

INGARDEN, The Cognition of the Literary Work of Art ( 1968; Evanston, Ill.: Northwestern University Press, 1973).

WOLFGANG ISER, The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett ( 1972; Baltimore: Johns Hopkins University Press, 1974).

JAMES G. FRAZER, The Golden Bough (abridged edn 1922: Macmillan, 1956); JOSEPH CAMPBELL, The Hero with a Thousand Faces ( 1949; Princeton: Princeton University Press, 1968); CARL G. JUNG, Collected Works ( 19 vols), ed. Herbert Read , Michael Fordham and Gerard Adler ( London: Routledge, 1981).

NORTHROP FRYE, Anatomy of Criticism: Four Essays ( Princeton: Princeton University Press, 1957).

See, for instance, 'Der Dichter und Phantasieren' ( 1907), or 'Das Unheimliche' ( 1919). The Standard Edition of the Complete Psychological Works of Sigmund Freud ( 24 vols), trans. James Strachey ( London: Hogarth Press/Institute of Psychology, 1953-66).

See for instance CHRISTOPHER NASH (ed.), Narrative in Culture: The Use of Storytelling in the Sciences, Philosophy, and Literature ( London: Routledge, 1989).

DAVID MASSON, British Novelists and Their Styles ( London: Macmillan, 1859); Sir WALTER RALEIGH, The English Novel ( 1894); Lord DAVID CECIL, Early Victorian Novelists ( London: Constable, 1934).

GEORGE SAINTSBURY, The Flourishing of Romance and the Rise of Allegory ( Edinburgh: Blackwood, 1897); W. P. KER, Epic and Romance ( 1896); GUSTAVE LANSON , Histoire de la littrature ftanise ( 1894), rev. edn Paul Tuffrau ( Paris: Hachette, 1951); E. A. BAKER, History of the English Novel ( 9 vols; London: Witherby, 1924-38).

HIPPOLYTE TAINE, History of English Literature ( 1863-4) trans. H. Van Laun ( New York: Ungar, 1965); FERDINAND BRUNETTIRE, L'volution des genres dans l'histoire de la littrature ( 1890; Paris: Hachette, 1980); T. S. ELIOT, 'Tradition and the Individual Talent' ( 1919), in Selected Essays ( London: Faber and Faber, 1951); LUKCS, Theory of the Novel and The Historical Novel ( 1937) trans. Hannah Mitchell and Stanley Mitchell ( Lincoln: University of Nebraska Press, 1983); ROBERT WEIMANN, 'Erzhlerstandpunkt und Point of View. Zur Geschichte und Aesthetik der Perspektive im englischen Roman', Zeitschrift fr Anglistik und Amerikanistik 10 ( 1962): 369-416; IAN WATT, The Rise of the Novel ( 1957; Harmondsworth: Penguin, 1983); HAYDEN WHITE, Metahistory: The Historical Imagination in Nineteenth-Century Europe ( Baltimore: Johns Hopkins University Press, 1973).

MILE BENVENISTE, Problems in General Linguistics ( 1966) trans. Mary Elizabeth Meck ( Coral Gables, Fla.: University of Miami Press, 1977); J. L. AUSTIN, How to Do Things with Words ( 1962; Oxford: Oxford University Press, 1980); M. M. BAKHTIN , The Dialogic Imagination ( 1938, pub. 1970) trans. Caryl Emerson and Michael Holquist ( Austin: University of Texas Press, 1981).

Some relevant works in literary pragmatics: TEUN A. VAN DIJK (ed.), Discourse and Literature: New Approaches to the Analysis of Literary Genre ( Amsterdam: Benjamins, 1985); MARY LOUISE PRATT, Toward a Speech Act Theory of Literary Discourse ( Bloomington: Indiana University Press, 1977; JOHN R. SEARLE, Expression and Meaning: Studies in the Theory of Speech Acts ( Cambridge: Cambridge University Press, 1979); ROGER D. SELL. (ed.), Literary Pragmatics ( London: Routledge, 1990).

GEORGE HENRIK VON WRIGHT, An Essay in Deontic Logic and the General Theory of Action ( Amsterdam: North Holland, 1968); PAUL RICUR, Du texte l'action ( Paris: Seuil, 1986); ERVING GOFFMAN, Frame Analysis: An Essay on the Organization of Experience ( Cambridge, Mass.: Harvard University Press, 1974).

PROPP, Folktale; A.-J. GREIMAS, Structural Semantics: An Attempt at a Method ( 1966) trans. Daniele McDowell, Ronald Schleifer and Alan Velie, introd. Ronald Schleifer ( Lincoln: University of Nebraska Press, 1983).

A. N. VESELOVSKI, Poetika siuzhetov, in Sobranie Sochinenii, 2/1 ( Petersburg, 1913), pp. 1-133; TOMASHEVSKI, 'Thematics' (select. and trans. of Teorija literatury, 1925) in Lemon and Reis (eds and trans.), Russian Formalist Criticism, pp. 61-98.

As a sample of three possible directions chosen among innumerable works: KENNETH BURKE, A Grammar of Motives and A Rhetoric of Motives ( Cleveland:

World, 1962); R. S. CRANE, ''The Concept of Plot and the Plot of Tom Jones'', in Critics and Criticism, ed. R. S. Crane ( 1952; abridged edn Chicago: University of Chicago Press, 1957), pp. 62-93; WILLIAM O. HENDRICKS, Essays on Semiolinguistics and Verbal Art ( The Hague: Mouton, 1973).

GOFFMAN, Frame Analysis; DAVID BORDWELL, Narration in the Fiction Film ( Madison: University of Wisconsin Press, 1985); EDWARD BRANIGAN (see pp. 234-48, below).

For instance, in MARVIN MINSKI, ''A Framework for Representing Knowledge'', in The Psychology of Computer Vision, ed. R. H. Winston ( New York: McGrawHill, 1975); DANIEL G. BOBROW, R. KAPLAN et al., ''GUS: A Frame-Driven Dialog System'', Artificial Intelligence 8 ( 1977: 155-73; JAMES MEEHAN, ''Tale-Spin'', in Inside Computer Understanding: Five Programs Plus Miniatures, ed. Roger C. Schank and Christopher K. Riesbeck ( Hillsdale, NJ: Erlbaum, 1981), pp. 197-226; GREGORY G. COLOMB and MARK TURNER, ''Computers, Literary Theory, and Theory of Meaning'', in The Future of Literary Theory, ed. Ralph Cohen ( New York: Routlege, 1989), pp. 386-410.

JAMES, 'Art of Fiction'; LUBBOCK, Craft of Fiction; CHARLES BALLY, ''Le style indirect libre en franais moderne'', Germanisch-Romanische Monatschrift 4 ( 1912): 549-56, 597-606; BOOTH, Rhetoric of Fiction; GENETTE, Narrative Discourse and Narrative Discourse Revisited ( 1983) trans. Jane E. Lewin ( Ithaca, NY. Cornell University Press, 1988); BAL, Narratology; BAKHTIN, Dialogic Imagination; DOUARD DUJARDIN, Le Monologue intrieur ( Paris: Messein, 1931); ROBERT HUMPHREY , Stream of Consciousness in the Modern Novel ( Berkeley: University of California Press, 1954); DORRIT COHN, Transparent Minds: Narrative Modes for Presenting Consciousness in Fiction (Princeton: Princeton University Press, 1978); ROGER FOWLER, Linguistics and the Novel ( 1977; London: Methuen, 1985).

ROBERT ALTER, Partial Magic: The Novel as a Self-Conscious Genre ( Berkeley: University of California Press, 1975).

See LUCIEN DALLENBACH, The Mirror in the Text ( 1977) trans. Jeremy Whiteley and Enuna Hughes ( Oxford: Polity Press, 1989).

ALTER, Partial Magic, p. x.

GABRIEL JOSIPOVICI, The World and the Book: A Study of Modern Fiction ( Stanford: Stanford University Press, 1971), p. xiv.

BAKHTIN, Dialogic Imagination; JULIA KRISTEVA, ''Bakhtin, le mot, le dialogue et le roman'', Critique 239 ( 1967): 438-65.

GENETTE, Introduction l'architexte ( Paris: Editions du Seuil, 1979).

FRYE, Anatomy of Criticism.

FREDRIC JAMESON, The Political Unconscious: Narrative as a Socially Symbolic Act ( Ithaca, NY. Cornell University Press, 1981).

FREUD, 'Creative Writers and Daydreaming' ( 1908) in Adams (ed.), Critical Theory since Plato, pp. 749-53, and Der Wahn und die Trume in W. Jensens Gradiva ( Leipzig: Hellen, 1907).

KATE MILLETT, Sexual Politics ( 1969; London: Virago, 1977); ELLEN MOERS, Literary Women: The Great Writers ( Garden City, NY: Doubleday, 1976); SANDRA M. GILBERT and SUSAN GUBAR, The Madwoman in the Attic: The Woman Writer and the Nineteenth-Century Literary Imagination ( New Haven: Yale University Press, 1979).