Types of Narration

Types of Narration

WAYNE C. BOOTH

Wayne C. Booth devotes chapter 6 of The Rhetoric of Fiction to an analysis of the different types of narration theoretically available. As he explains, earlier classifications of point of view, such as Norman Friedman ''Point of View'' ( 1955), are simplistic in that they are exclusively based on the notions of first/third-person narration and the degree of omniscence of the narrator. Although these notions are important, he contends, further refinements should be made. He proposes a differentiation between real author, implied author, narrator, characters and readers. The implied author is the real author's literary version of him/herself. The narrator is the mediating instance between author and reader, the one who tells. Figuratively placed between implied author and characters in the narrative chain, the narrator may be closer to one or to the others. Impersonal or 'undramatized' narrators, who try to efface themselves from their narration, are often difficult to distinguish from the implied author. 'Dramatized' narrators, that is, narrators with a well-developed personality, are more easily perceptible in their own right. These may choose to participate in the action as characters or 'narrator-agents', or to stand apart as mere 'observers'. Narrators can participate in the action in different ways according to the moral, physical and/or temporal distance separating them from the other characters and/or from the author and reader. Thus, narrator-agents can be further classified as 'reliable' or 'unreliable' -- if their opinions and values coincide or clash with those of the others -- and they can also be 'isolated' or 'supported' by other narrators in the story. Finally, all kinds of narrators can choose to be omniscient -- including free access to the minds of the characters, which is the most interesting kind of omniscience, according to Booth -- or to limit their knowledge to what could be learned by natural means or inference, thus producing a realism-enhancing effect.

Booth's typology constituted a landmark in the analysis of the narrative instance. Although some of his categories are not clearly defined and his terminology is often tentative or misleading (for example, he often forgets his own distinction between real author, implied author and narrator), he nevertheless coined key concepts, like 'implied author', 'unreliable narrator' and 'distance', that would provide the basis for the more systematic typologies of narratologists like Genette, Bal and Stanzel.

We have seen that the author cannot choose to avoid rhetoric; he can choose only the kind of rhetoric he will employ. He cannot choose whether or not to affect his readers' evaluations by his choice of narrative manner; he can only choose whether to do it well or poorly. As dramatists have always known, even the purest of dramas is not purely dramatic in the sense of being entirely presented, entirely shown as taking place in the moment. There are always what Dryden called 'relations' to be taken care of, and try as the author may to ignore the troublesome fact, 'some parts of the action are more fit to be represented, some to be related.' ¹ But related by whom? The dramatist must decide, and the novelist's case is different only in that the choices open to him are more numerous.

If we think through the many narrative devices in the fiction we know, we soon come to a sense of the embarrassing inadequacy of our traditional classification of 'point of view' into three or four kinds, variables only of the 'person' and the degree of omniscience. If we name over three or four of the great narrators -- say Cervantes' Cide Hamete Benengeli, Tristram Shandy, the 'I' of Middlemarch, and Strether, through whose vision most of The Ambassadors comes to us, we realize that to describe any of them with terms like 'first-person' and 'omniscient' tells us nothing about how they differ from each other, or why they succeed while others described in the same terms fail. It should be worth our while, then, to attempt a richer tabulation of the forms the author's voice can take.

Person

Perhaps the most overworked distinction is that of person. To say that a story is told in the first or the third person ² will tell us nothing of importance unless we become more precise and describe how the particular qualities of the narrators relate to specific effects. It is true that choice of the first person is sometimes unduly limiting; if the 'I' has inadequate access to necessary information, the author may be led into improbabilities. And there are other effects that may dictate a choice in some cases. But we can hardly expect to find useful criteria in a distinction that throws all fiction into two, or at most three, heaps. In this pile we see Henry Esmond, ''A Cask of Amontillado,'' Gulliver's Travels, and Tristram Shandy. In that, we have Vanity Fair, Tom Jones, The Ambassadors, and Brave New World. But in Vanity Fair and Tom Jones the commentary is in the first person, often resembling more the intimate effect of Tristram Shandy than that of many third-person works. And again, the effect of The Ambassadors is much closer to that of the great first-person novels, since Strether in large part 'narrates' his own story, even though he is always referred to in the third person.

Further evidence that this distinction is less important than has often been claimed is seen in the fact that all of the following functional distinctions apply to both first- and third-person narration alike.

Dramatized and undramatized narrators

Perhaps the most important differences in narrative effect depend on whether the narrator is dramatized in his own right and on whether his beliefs and characteristics are shared by the author.

The implied author (the author's 'second self'). -- Even the novel in which no narrator is dramatized creates an implicit picture of an author who stands behind the scenes, whether as stage manager, as puppeteer, or as an indifferent God, silently paring his fingernails. This implied author is always distinct from the 'real man' -- whatever we may take him to be -- who creates a superior version of himself, a 'second self', as he creates his work.

In so far as a novel does not refer directly to this author, there will be no distinction between him and the implied, undramatized narrator; in Hemingway 'The Killers,' for example, there is no narrator other than the implicit second self that Hemingway creates as he writes.

Undramatized narrators. -- Stories are usually not so rigorously impersonal as 'The Killers'; most tales are presented as passing through the consciousness of a teller, whether an 'I' or a 'he.' Even in drama much of what we are given is narrated by someone, and we are often as much interested in the effect on the narrator's own mind and heart as we are in learning what else the author has to tell us. When Horatio tells of his first encounter with the ghost in Hamlet, his own character, though never mentioned, is important to us as we listen. In fiction, as soon as we encounter an 'I,' we are conscious of an experiencing mind whose views of the experience will come between us and the event. When there is no such 'I,' as in 'The Killers,' the inexperienced reader may make the mistake of thinking that the story comes to him urumediated. But no such mistake can be made from the moment that the author explicitly places a narrator into the tale, even if he is given no personal characteristics whatever.

Dramatized narrators. -- In a sense even the most reticent narrator has been dramatized as soon as he refers to himself as 'I,' or, like Flaubert, tells us that 'we' were in the classroom when Charles Bovary entered. But many novels dramatize their narrators with great fulness, making them into characters who are as vivid as those they tell us about ( Tristram Shandy, Remembrance of Things Past, Heart of Darkness, Dr. Faustus). In such works the narrator is often radically different from the implied author who creates him. The range of human types that have been dramatized as narrators is almost as great as the range of other fictional characters -- one must say 'almost' because there are some characters who are not fully qualified to narrate or 'reflect' a story ( Faulkner can use the idiot for part of his novel only because the other three parts exist to set off and clarify the idiot's jumble). ^*

We should remind ourselves that many dramatized narrators are never explicitly labeled as narrators at all. In a sense, every speech, every gesture, narrates; most works contain disguised narrators who are used to tell the audience what it needs to know, while seeming merely to act out their roles.

Though disguised narrators of this kind are seldom labeled so explicitly as God in Job, they often speak with an authority as sure as God's. Messengers returning to tell what the oracle said, wives trying to convince their husbands that the business deal is unethical, old family retainers expostulating with wayward scions -- these often have more effect on us than on their official auditors; the king goes ahead with his obstinate search, the husband carries out his deal, the hell-hound youth goes on toward hell as if nothing had been said, but we know what we know -- and as surely as if the author himself of his official narrator had told us. [. . . ]

The most important unacknowledged narrators in modern fiction are the third-person 'centers of consciousness' through whom authors have filtered their narratives. Whether such 'reflectors,' as James sometimes called them, are highly polished mirrors reflecting complex mental experience, or the rather turbid, sense-bound 'camera eyes' of much fiction since James, they fill precisely the function of avowed narrators -- though they can add intensities of their own. []

Observers and narrator-agents

Among dramatized narrators there are mere observers (the 'I' of Tom Jones, The Egoist, Troilus and Criseyde), and there are narrator-agents, who produce some measurable effect on the course of events (ranging from the minor involvement of Nick in The Great Gatsby, through the extensive give-and-take of Marlow in Heart of Darkness, to the central role of Tristram Shandy, Moll Flanders, Huckleberry Finn, and -- in the third person -- Paul Morel in Sons and Lovers). Clearly, any rules we might discover about observers may not apply to narratoragents, yet the distinction is seldom made in talk about point of view.

Scene and summary

All narrators and observers, whether first or third person, can relay their tales to us primarily as scene ( ''The Killers,'' The Awkward Age, the works of Ivy Compton-Burnett and Henry Green), primarily as summary or what Lubbock called 'picture' ( Addison's almost completely non-scenic tales in The Spectator), or, most commonly, as a combination of the two.

Like Aristotle's distinction between dramatic and narrative manners, the somewhat different modern distinction between showing and telling does cover the ground. But the trouble is that it pays for broad coverage with gross imprecision. Narrators of all shapes and shades must either report dialogue alone or support it with 'stage directions' and description of setting. But when we think of the radically different effect of a scene reported by Huck Finn and a scene reported by Poe's Montresor, we see that the quality of being 'scenic' suggests very little about literary effect. And compare the delightful summary of twelve years given in two pages of Tom Jones (Book III, chap. i) with the tedious showing of even ten minutes of uncurtailed conversation in the hands of a Sartre when he allows his passion for 'durational realism' to dictate a scene when summary is called for.[. . .] The contrast between scene and summary, between showing and telling, is likely to be of little use until we specify the kind of narrator who is providing the scene or the summary.

Commentary

Narrators who allow themselves to tell as well as show vary greatly depending on the amount and kind of commentary allowed in addition to a direct relating of events in scene and summary. Such commentary can, of course, range over any aspect of human experience, and it can be related to the main business in innumerable ways and degrees. To treat it as a single device is to ignore important differences between commentary that is merely ornamental, commentary that serves a rhetorical purpose but is not part of the dramatic structure, and commentary that is integral to the dramatic structure, as in Tristram Shandy.

Self-conscious narrators

Cutting across the distinction between observers and narrator-agents of all these kinds is the distinction between self-conscious narrators aware of themselves as writers ( Tom Jones, Tristram Shandy, Barchester Towers, The Catcher in the Rye, Remembrance of Things Past, Dr. Faustus), and narrators or observers who rarely if ever discuss their writing chores ( Huckleberry Finn) or who seem unaware that they are writing, thinking, speaking, or 'reflecting' a literary work ( Camus The Stranger, Lardner 'Haircut,' Bellow The Victim).

Variations of distance

Whether or not they are involved in the action as agents or as sufferers, narrators and third-person reflectors differ markedly according to the degree and kind of distance that separates them from the author, the reader, and the other characters of the story. In any reading experience there is an implied dialogue among author, narrator, the characters, and the reader. Each of the four can range, in relation to each of the others, from identification to complete opposition, on any axis of value, moral, intellectual, aesthetic, and even physical. (Does the reader who stammers react to the stammering of H. C. Earwicker as I do? Surely not.) [. . .]

narrator and Pyle the American in Greene The Quiet American, both departing radically from the author's norms but in different directions); morally and emotionally ( Maupassant 'The Necklace,' and Huxley 'Nuns at Luncheon,' in which the narrators affect less emotional involvement than Maupassant and Huxley clearly expect from the reader); and thus on through every possible trait.

The narrator may be more or less distant from the reader's own norms; for example, physically and emotionally ( Kafka The Metamorphosis); morally and emotionally (Pinkie in Brighton Rock, the miser in Mauriac Knot of Vipers, and the many other moral degenerates that modern fiction has managed to make into convincing human beings).

With the repudiation of omniscient narration, and in the face of inherent limitations in dramatized reliable narrators, it is hardly surprising that modern authors have experimented with unreliable narrators whose characteristics change in the course of the works they narrate. [. . .] The mature Pip, in Great Expectations, is presented as a generous man whose heart is where the reader's is supposed to be; he watches his young self move away from the reader, as it were, and then back again. But the third-person reflector can be shown, technically in the past tense but in effect present before our eyes, moving toward or away from values that the reader holds dear. Authors in the twentieth century have proceeded almost as if determined to work out all of the possible plot forms based on such shifts: start far and end near; start near, move far, and end near; start far and move farther; and so on. [. . .]

The implied author may be more or less distant from the reader. The distance may be intellectual (the implied author of Tristram Shandy, not of course to be identified with Tristram, more interested in and knowing more about recondite classical lore than any of his readers), moral (the works of Sade), or aesthetic. From the author's viewpoint, a successful reading of his book must eliminate all distance between the essential norms of his implied author and the norms of the postulated reader. [. . .]

The implied author (carrying the reader with him) may be more or less distant from other characters. Again, the distance can be on any axis of value. [. . .]

What we call 'involvement' or 'sympathy' or 'identification,' is usually made up of many reactions to author, narrators, observers, and other characters. And narrators may differ from their authors or readers in various kinds of involvement or detachment, ranging from deep personal concern (Nick in The Great Gatsby, MacKellar in The Master of Ballantrae, Zeitblom in Dr. Faustus) to a bland or mildly amused or merely curious detachment ( Waugh Decline and Fall).

For practical criticism probably the most important of these kinds of distance is that between the fallible or unreliable narrator and the implied author who carries the reader with him in judging the narrator. If the reason for discussing point of view is to find how it relates to literary effects, then surely the moral and intellectual qualities of the narrator are more important to our judgment than whether he is referred to as 'I' or 'he,' or whether he is privileged or limited. If he is discovered to be untrustworthy, then the total effect of the work he relays to us is transformed.

Our terminology for this kind of distance in narrators is almost hopelessly inadequate. For lack of better terms, I have called a narrator reliable when he speaks for or acts in accordance with the norms of the work (which is to say, the implied author's norms), unreliable when he does not. It is true that most of the great reliable narrators indulge in large amounts of incidental irony, and they are thus 'unreliable' in the sense of being potentially deceptive. But difficult irony is not sufficient to make a narrator unreliable. Nor is unreliability ordinarily a matter of lying, although deliberately deceptive narrators have been a major resource of some modern novelists ( Camus' The Fall, Calder Willingham Natural Child, etc.). It is most often a matter of what James calls inconscience; the narrator is mistaken, or he believes himself to have qualities which the author denies him. Or, as in Huckleberry Finn, the narrator claims to be naturally wicked while the author silently praises his virtues behind his back.

Unreliable narrators thus differ markedly depending on how far and in what direction they depart from their author's norms; the older term 'tone,' like the currently fashionable terms 'irony' and 'distance,' covers many effects that we should distinguish. Some narrators, like Barry Lyndon, are placed as far 'away' from author and reader as possible, in respect to every virtue except a kind of interesting vitality. Some, like Fleda Vetch, the reflector in James The Spoils of Poynton, come close to representing the author's ideal of taste, judgment, and moral sense. All of them make stronger demands on the reader's powers of inference than do reliable narrators.

Variations in support or correction

Both reliable and unreliable narrators can be unsupported or uncorrected by other narrators ( Gully Jimson in The Horse's Mouth, Henderson in Bellow Henderson the Rain King) or supported or corrected ( The Master of Ballantrae, The Sound and the Fury). Sometimes it is almost impossible to infer whether or to what degree a narrator is fallible; sometimes explicit corroborating or conflicting testimony makes the inference easy. Support or correction differs radically, it should be noted, depending on whether it is provided from within the action, so that the narrator-agent might benefit from it in sticking to the right line or in changing his own views ( Faulkner Intruder in the Dust), or is simply provided externally, to help the reader correct or reinforce his own views as against the narrator's ( Graham Greene The Power and the Glory). Obviously, the effects of isolation will be extremely different in the two cases.

Privilege

Observers and narrator-agents, whether self-conscious or not, reliable or not, commenting or silent, isolated or supported, can be either privileged to know what could not be learned by strictly natural means or limited to realistic vision and inference. Complete privilege is what we usually call omniscience. But there are many kinds of privilege, and very few 'omniscient' narrators are allowed to know or show as much as their authors know. [. . .]

The most important single privilege is that of obtaining an inside view of another character, because of the rhetorical power that such a privilege conveys upon a narrator. There is a curious ambiguity in the term 'omniscience.' Many modern works that we usually classify as narrated dramatically, with everything relayed to us through the limited views of the characters, postulate fully as much omniscience in the silent author as Fielding claims for himself. Our roving visitation into the minds of sixteen characters in Faulkner As I Lay Dying, seeing nothing but what those minds contain, may seem in one sense not to depend on an omniscient author. But this method is omniscience with teeth in it: the implied author demands our absolute faith in his powers of divination. We must never for a moment doubt that he knows everything about each of these sixteen minds or that he has chosen correctly how much to show of each. In short, impersonal narration is really no escape from omniscience -the true author is as 'unnaturally' all-knowing as he ever was. [. . .]

Inside views

Finally, narrators who provide inside views differ in the depth and the axis of their plunge. Boccaccio can give inside views, but they are extremely shallow. Jane Austen goes relatively deep morally, but scarcely skims the surface psychologically. All authors of stream-ofconsciousness narration presumably attempt to go deep psychologically, but some of them deliberately remain shallow in the moral dimension. We should remind ourselves that any sustained inside view, of whatever depth, temporarily turns the character whose mind is shown into a narrator; inside views are thus subject to variations in all of the qualities we have described above, and most importantly in the degree of unreliability. Generally speaking, the deeper our plunge, the more unreliability we will accept without loss of sympathy.

Narration is an art, not a science, but this does not mean that we are necessarily doomed to fail when we attempt to formulate principles about it. There are systematic elements in every art, and criticism of fiction can never avoid the responsibility of trying to explain technical successes and failures by reference to general principles. But we must always ask where the general principles are to be found. [. . .]

In dealing with the types of narration, the critic must always limp behind, referring constantly to the varied practice which alone can correct his temptations to overgeneralize. [. . .]

Notes