THE NARRATOR

Fictional narratives have two dimensions: 1) the story, that is, the chain of actions and events
performed or experienced by characters in a setting, and 2) the discourse, the means by which the
story is told or otherwise transmitted. In rough terms, the story is the "what," and the discourse is
the "how." Although story (sometimes called the "narrated") and discourse (the "narrating" or
"narration") are useful terms devised by narratologists to explain the structure of the novel, each
novel, of course, actually comes to a reader as a string of words. That string---the published text---is
an actualization or instance of the narrative structure. The reader reads the words and interprets
them as events, characters, settings, and the like.

The main agent of the discourse is the narrator. Some narratologists believe that every narrative has
a narrator; others believe that narratives are "non-narrated" if the reader cannot discern an audible
voice. In the simplest kind of narrative, the oral personal anecdote, the narrator is the speaker; he or
she tells what happened, usually after the fact, to another person, the listener. On the analogy of
"narrator," this audience has been named (by Gerald Prince) the narratee.

A narrative is one of several types of text---others are descriptions, arguments, expositions, lists,
and so on. Unlike the other text types, narrative necessarily communicates a chain of events or
happenings. Recall the main events of Charles Dickens' Great Expectations (1861). Upon the
invitation of an anonymous benefactor, Pip, a poor country boy, leaves his humble home for
London to become a gentleman. He believes that his benefactor is the lady of the manor, Miss
Havisham, but later learns that it was really Magwitch, a convict whom he had helped escape and
who has made a fortune abroad. When Magwitch returns to England, sacrificing everything for a
chance to see "his boy," Pip learns the truth and is ashamed of the snobbishness he has developed
toward the poorer classes, including Magwitch and his brother-in-law, Joe. Pip goes to the Orient to
seek his fortune, returns to his village, makes amends, and finally goes off to a glorious future with
Estella, Miss Havisham's ward.

Time-sequence is essential to the text type we call narrative: every story rests on a chronology of
events. That chronology may be called story-time. But there is another kind of narrative time,
namely the time (whether stated by the narrator or reconstructed by the reader) that it takes the
narrator to tell the story; this kind of time may be called discourse-time. In story-time, events can
only occur in one sequence, following the natural order of life. Jack has to climb the hill before he
can fall down it; he has to fall down before his crown is broken. In the nursery rhyme "Jack and
Jill," the discourse-time follows the same order as the story-time. But the discourse-time could have
been arranged in a different way: "Jack lay on the ground moaning. He had just broken his crown
after falling down the hill." Here the last event is told first, and the earlier events are told later. This
discourse arrangement is usually called flashback (or analepsis). It appears as early as Homer's
Iliad.

Narrative, then, unlike the other text types, has a double time arrangement. Compare the double
time-scheme of narrative with the single time of a different kind of text, argument. An argument
turns not on events but on propositions and proofs. Of course, a story may be used by an arguer to
support the argument. In that case, however, the story is not told for its own sake, but as evidence.
A defendant's attorney may tell a story of how his client was bowling when the murder occurred, so
a chronology is invoked. Yet the larger structure of the lawyer's argument does not depend on time-
sequence, but on the support furnished by the reasons to the assertion in question (in this case, "He
is not guilty"). Of course, the argument is a text that unfoldsin time, the time it takes to read or hear
it. But that is external text-time: it is not intrinsic to the structure of the argument.

The story the defense attorney tells about his client's bowling at the time of the murder is an
instance of a narrative subserving an argument. The reverse also happens: in Henry Fielding's Tom
Jones (1749), the narrator often interrupts the story by advancing a moral argument. The text type
of description even more typically subserves narratives, for instance, in setting the scene.
(Narratives may subserve descriptions, also, as when a travel book recounts the history of a castle as
part of its description of the building.)

Just as we distinguish between story-time and discourse-time, we may distinguish between story-
space and discourse-space. Story-space is the setting, the place(s) where the story events occur.
Story-space in Great Expectations includes Joe's forge, where Pip lives as a boy, Miss Havisham's
manor, and London. Discourse-space, on the other hand, is the place occupied by the narrator when
he tells the story. In Great Expectations, the discourse-space occupied by the narrator is
indeterminate; but in some novels it is clearly located. In Joseph Conrad's "Heart of Darkness"
(1902) the discourse-space is the yacht Nelly, lying becalmed in the London estuary, whereas the
space of Marlow's story extends from Belgium to Africa, up the Congo River. The discourse-space
of Emily Brontë's Wuthering Heights (1847) is Thrushcross Grange, the house Lockwood has
rented from Heathcliff; it is there that Lockwood, the initial narrator, becomes narratee to Mrs.
Dean's story of Heathcliff's tempestuous love for Catherine. The story-space of Mrs. Dean's tale, of
course, is the estate of Wuthering Heights.

In the personal oral anecdote, the narrator is also the author. If I tell you about something that
happened to me yesterday, I am both the author of the story and its narrator or transmitter. If I tell
you a joke that I heard, I am only the narrator; the person who originally made the joke up is the
author. The situation in novels and short stories is still more complex. When an author decides to
write a novel, she chooses not only the story elements---events,characters, and settings---but also
the discourse elements, including the best kind of narrator. Mark Twain decided to have
Huckleberry Finn narrate the novel The Adventures of Huckleberry Finn (1884) because he wanted
a naive spokesman, the better to ironize life on the Mississippi River. Jane Austen decided on an
anonymous narrator to narrate Pride and Prejudice (1813), the better to convey the conflict between
Elizabeth's prejudice and Darcy's pride. It is important not to confuse the real author with the
narrator, even if the narrator refers to herself as "your author" or the like. The author is or was a real
historical person; the narrator, on the other hand, is a personage in the novel, no less fictional than
any character, who comes to life each time the novel is read.

Some narrators are also characters in the story; some only tell the story and do not participate in it.
The first kind may be called character-narrators; the second external narrators. Pip, in addition to
being the narrator of Great Expectations, is also its hero or protagonist. The character-narrator can
tell what happened in the story even up to the present moment, the "now" of the discourse. He was a
character "back then," during storytime, and is now a narrator in present discourse-time. A more
technical term for the character-narrator is homodiegetic narrator. ("Diegetic" refers to story-
telling, and "homo-" means "the same.") Thus, the narrator-Pip is the same person as the
protagonist-Pip. However, it is important to recognize that "Pip" refers to two different narrative
agents, even though Pip is a single person. Sometimes the character-narrator is not the protagonist
but a lesser character who tells what happened to the protagonist. For example, in F. Scott
Fitzgerald's novel The Great Gatsby (1925), Gatsby is the protagonist, but the narrator is his
neighbor, Nick Carraway. The narrator of Fedor Dostoevskii's The Devils (1872) is an even more
minor character, an unnamed and undistinguished inhabitant of a provincial Russian town. A
character-narrator who is only an observer is called a witness narrator.
Character-narrators often are referred to as "first-person-" or "I-narrators," but those terms can be
misleading since an external narrator, one who does not inhabit the story-world but only the
discourse-world, could just as well refer to herself as "I." A more technical term for the external
narrator is heterodiegetic narrator. Classic examples are the narrators of Henry James' The
Ambassadors (1903) and his short story "The Beast in the Jungle" (1903). The narrator of the
former, for example, says on the first page "The principle I have just mentioned ..." and the latter
alludes to himself in sentences beginning "Our point is ..." or "What we are especially concerned
with ..." But neither narrator is a character in the story. In a few cases the firstperson plural "we" is
used to refer to the external narrator, either in the royal sense (the monarch is so grand a figure as to
encompass a plurality, not an individuality, speaking as she does for the entire kingdom) or,
contrarily, as a mark of modesty, to avoid the suggestion of egotism. But some narrators are plural
in the ordinary sense: in William Faulkner's short story "A Rose for Emily" (1931), the "we"
represents a village community who witness the bizarre aftermath of Miss Emily's jilting by her
fiancé. The collective "we" -narrator is rare, even in modern fiction, and often ambiguous. It can
refer to such different entities as an actual community of speakers speaking as one (like the chorus
in a Greek tragedy), to one speaker presuming to speak for others (who may or may not be willing
to let her be spokesperson), to a speaker who includes the addressee in the group (that is, makes the
narratee, by definition a "you," into a member of his party), and so on.

The expression "third-person narrator" is even more confusing than "first-person narrator," since the
third person in question is generally not a speaker or an addressee but some other person.
Grammatically, the first person is the one who speaks, the second person is the one spoken to, and
the third person the one spoken about. That means that the latter is by definition a non-participant in
the immediate communication of the discourse, although her consciousness may "filter" the story-
world. To call the narrator a third person is to confuse story and discourse. (An exception is the
narrative in which, for some extranarratorial reason, the first-person speaker refers to himself by his
own name, as does Julius Caesar in The Gallic Wars.) However, one may reasonably speak of third-
person narratives, remembering that the focus is now on the character, not on the narrator.

Is there such a thing as a second-person or "you"-narrator? Again we can speak of "second-person

narratives," but in this unusual form the "you" again seems always to refer to a character (or to the
character aspect of a character-narrator). In second-person narration, the narrator may be referring
to him-/herself in past (or future) time, but also may be the narratee, or a third person who, for some
reason, is spoken to as well as about. Some examples are Rex Stout's How Like a God (1929),
Michel Butor's La Modification (1957; Change of Heart), a short story by Mary McCarthy called
"The Genial Host," in The Company She Keeps (1942); Jay McInerney's Bright Lights, Big City
(1984), and Lorrie Moore's Self-Help (1985). How are we to understand this rather unnatural way of
telling a story? In some cases we seem to be dealing with an external narrator telling the events to
some "you" who actually has experienced them but is either unwilling or unable to articulate the
story (as a psychiatrist might tell an amnesiac patient his own story to help him recover his memory,
or a prosecuting attorney might tell the accused his version of a crime in the hopes that the accused
will break down and admit that it is true). In other cases, the narrator may be speaking to himself
about himself, as if from the outside. Or the passage may be in interior monologue, the narrator
simply recording what the character who refers to himself as "you" is thinking. In some cases, as in
parables, the "you" may be generic, indicating that what is happening to the character may easily
happen to anyone. Or "you" may simply be a vivifying form of "I" (as in some detective novels).
Whatever the case, the "you" form seems intended to make the reader identify closely with the
addressed character.

Despite the fact that they inhabit only the discourse, some external narrators emerge as full-blown
personages in their own right, expressing opinions, attitudes, and prejudices, even describing their
own personality, appearance, and so on. Others are simply voices, giving us few clues, either
explicit or implicit, about themselves. The former are sometimes called overt, the latter covert.

Narratees are even more likely to be blanks, simply ears, listeners living in the discourse-world, or
"dear readers," who make no response to the narrator's tale. But sometimes they are fleshed out, as
when the text is "nested" and the discourse becomesitself a little narrative. The outer or containing
narrative may be called the frame, and the inner, contained narrative the framed. For instance, in
Conrad's "Heart of Darkness," theframing narrative recounts a conversation among four men on a
yacht. An unnamed member of the party narrates the story of the men's conversation; mentioning no
narratee, he is presumably telling the story of the conversation to "us readers." At a certain point,
another member of the party, Marlow, embarks on the framed story, about his trip to Africa and
meeting with Kurtz. Marlow is the narrator of the inner, framed story, and his narratees are the other
three men, including the anonymous narrator of the outer, framing story. (A similar structure
informs Wuthering Heights.)

We can also distinguish among narrators according to the kind and amount of information they
provide. At one end of the scale is the narrator who has the power to tell the narratee explicitly, and
at any moment he likes, everything one needs to know to understand the story. Traditionally, such a
narrator has been called the omniscient narrator, but we might do better to call her the unlimited
narrator. "Omniscient"---"all-knowing" (from Latin "omni-" "all" and "sciens" "knowing")---is not
felicitous because the narrator's function is not to know something, but simply to tell the story. To
do that she must already know the story. It makes even less sense to say that the author is
omniscient, since in a fiction the author is the one who has invented everything. Ordinarily, the only
one for whom knowing is an issue is the reader, who learns everything in the course of reading the
novel.

There are, however, some exceptions, that is, narrators who do not know the whole story from the
outset. For example, a correspondent in an epistolary novel, like Samuel Richardson's Pamela
(1740), or the diarist of a diary novel, like Jean-Paul Sartre's La Nausée (1938; Nausea), do not
know in mid-novel how things will ultimately turn out for them. It could be argued, however, that
these are not the main narrators of their novels, that the narrator, rather, is the one who compiled
and/or edited the letters or diary. Such a narrator is like the passive narrator of a story told
completely through the quoted dialogue of one or more of the characters (for example, Dorothy
Parker's short story "Lady with a Lamp").

Another narrator who may not know the whole story, or who knows it only inaccurately, is called
an unreliable narrator . An unreliably narrated story is one whose "true version," we infer, differs
strikingly from what the narrator actually tells us. We piece together the "real" story because we
find reasons in the narrator's character, age, intelligence, the way he expresses himself, or whatever,
for doubting that he understands the import of the events, or that he is being truthful about them. In
the former case, when the narrator does not undersand, we may speak of naive unreliability, which
we find in Huckleberry Finn and Ford Madox Ford's The Good Soldier (1915), for instance. When
the narrator is all too knowing, we may speak of lying unreliability (as, for instance, in Jason
Compson's section of William Faulkner's The Sound and the Fury [1929]).

The unlimited narrator can, at any moment of his choosing, provide a great deal of information
directly to the narratee; he can summarize events and situations, interpret, judge, and comment upon
what characters do, think, and feel (or what they don't do, think, or feel), introduce metaphoric
language, philosophize about the real world (as opposed to the fictional world of the story), make
comments about the narrating process itself, and so on. The narrator of Tom Jones offers copious
explanations of his narrative procedures, explaining, for example, why he has withheld the
information that Mr. Square as well as Tom Jones was Molly's lover (he saw no reason for doing so,
since Mr. Square hadn't announced it either). Such a narrator is sometimes called self-conscious.
The (implied) author puts no limits on what the unlimited narrator can say.

On the other hand, the narrator may be limited. Usually the limitation takes the form of restricting
what can be told at any given story-moment to what one character experiences. A classic example is
Henry James' What Maisie Knew (1897): the adult external narrator narrates the erotic carryings-on
of a small girl's estranged parents, but only insofar as she perceives and understands---or fails to
understand---them. Another example is Gustave Flaubert's Madame Bovary (1857), most of which
is screened through Emma Bovary's flawed consciousness. As for character-narrators, like Marlow
in "Heart of Darkness," most are limited, but some may enter the minds of other characters without
explaining how they could do so (the minor character-narrator of The Devils recounts the secret
thoughts of the protagonist, which neither he nor anyone else possibly could have known).

The limited narrator tends not to give overt clues about his own personality, but we can often catch
hints of it. The stories in James Joyce's Dubliners (1914), often cited as classic examples of limited
narration, imply some attitudes of the narrator. For example, in "A Painful Case," a sentence
summarizing the protagonist's general attitude toward life---"Mr Duffy abhorred anything which
betokened physical or mental disorder"---is followed by "A mediaeval doctor would have called
him saturnine. His face, which carried the entire tale of his years, was of the brown tint of Dublin
streets." Here the narrator stands back from the character's consciousness and gives an independent
view of the character's appearance, as well as hints of his own attitude toward Dublin and its
inhabitants (not to speak of his knowledge of medical history).

There are extreme cases of limitation in which the narrator has the power to present only the
contents of a character's mind. This is the effect called interior monologue or stream of
consciousness. The most celebrated example is the last section of Joyce's Ulysses (1922), which
presents nothing but the rambling thoughts of the heroine, Molly Bloom. It is important to
understand that the character whose flow of thoughts is recorded is not the narrator. She is not
telling anybody a story but simply living her life in the story. The narrator here is a kind of mind-
reading recorder faithfully reproducing her every thought. He is even more restricted than the
narrators of Dubliners.

Finally, there is a kind of narrator who may not enter any character's mind. This sort of objective
narrator functions somewhat like a camera---hence such colorful synonyms as camera-eye or fly-on-
the-wall narrator. Objective narratives are sometimes said to be "shown" rather than "told," but the
terminology is only metaphoric: you can literally "show" a story only in a visual medium like a
movie or comic strip. The objective narrator, common in hard-boiled detective novels that insist on
hard evidence, usually recounts only what could be seen by someone standing in the vicinity of the
events narrated. A classic example is Ernest Hemingway's "The Killers." The objective narrator
"follows" the protagonist, Nick Adams, but tells us practically nothing of what he is thinking,
leaving us to guess Nick's state of mind as the events transpire before him. Not that the objective
narrator need be the equivalent of a photographic device. He---or better "it"---may "overhear," like
an audiotape recorder or stenographer who does nothing more than transcribe the dialogue of the
characters. Another Hemingway story, "Hills Like White Elephants" (1927), consists almost
entirely of the speeches, wrapped in quotation marks, of the two characters. Or the objective
narrator may present only written documents, like letters. This kind of narrator functions like a fax
machine or manuscript copyist.

These categories---unlimited, limited, and objective---only roughly explain the characteristics of

narrators of novels and short stories. Often it is difficult to place individual narrators. Some seem to
belong in more than one category or blur the norms of a given category. Some novels switch
repeatedly among different kinds of narrators (James Joyce's Ulysses is a preeminent example).
Some tales are narrated multiply by narrators whose versions conflict, leaving what "really"
happened open to question. Robert Browning's narrative poem The Ring and the Book (1868-69)
and the Japanese short story and film Rashomon are good examples of this technique. Instead of
forcing narrators into one or another of the categories listed above, we may do better to characterize
them in terms of the individual powers assigned them. Take, for example, Marcel Proust's À la
recherche du temps perdu (1913-27; In Search of Lost Time; also translated as Remembrance of
Things Past). Although narrated by the character-narrator Marcel, and in many ways limited to what
Marcel-as-character could know, Marcel-as-narrator also has the power to delve into the intimate
thoughts of other characters. The narrator of John Barth's "Lost in the Funhouse" (1968) has access
only to the consciousness of the protagonist Ambrose but at the same time feels free to comment
and generalize, not only about events and characters in the story, but also about the art and
mechanics of writing fiction.

Whatever the powers of the narrator, he or she inevitably presents the story from a certain point of
view or perspective. This point of view, which we can call the narrator's slant, may be stated quite
explicitly by the narrator. In the first chapter of Tom Jones, the narrator argues that authors should
provide plenty of information---in introductions, chapter headings, and the like---about what the
reader may expect to find in the text. His slant takes the ostensible form of honest business dealings:
like a good restauranteur, he offers the reader a menu by which to decide whether to partake of the
text. The narrator of Joyce's "A Painful Case," comparing Duffy's saturnine face with the "brown
tint" of Dublin's streets, conveys a rather negative slant on life in Ireland. In the opening sentence of
Pride and Prejudice, the narrator reveals her slant about a social class whose sole occupation is to
marry off its daughters to affluent men: "It is a truth universally acknowledged, that a single man in
possession of good fortune must be in want of a wife." In "Heart of Darkness," Marlow's slant is
against the brutal treatment of indigenous peoples by entrepreneurs, although not necessarily
against colonialism as an institution.

Unlike the external narrator, the character-narrator has two perspectives---her present, narrator's
slant toward the story and her earlier, character's perspective. We may call the character's point of
view "filter." The character-narrator can report the filtered attitude that she had back when she was
a character (what I did made sense at the time) but then modify it with her current slant (but I see
now that it was quite stupid). The narrator's slant may reveal psychological, moral, political,
ideological, and economic attitudes---indeed, any mental predisposition whatsoever.

Slant entails the notion of distance. A narrator may be said to be more or less distant from the
characters, the action, the setting, or all three. This distance may be literally spatial, as in "Heart of
Darkness," in which the narrator recounts events that happened thousands of miles away.
Sometimes, distance is used to refer to a broad perspective. Tom Jones begins as if high above
England: "In the western half of this kingdom ...". Critics speak of the bird's-eye or God's-eye view,
as opposed to the microscopic view. It is common to use terms like "long shot" and "close-up," but
we must remember that these apply only metaphorically to verbal fiction.

Another kind of distance is temporal; that is, the time of the telling of the story may begin only
moments after the last storyevent occurred, as in Richardson's Pamela, or long after, as in Sir
Walter Scott's historical novels or John Fowles' The French Lieutenant's Woman (1969), whose
narrator lives in the 1960s but narrates events that took place 100 years before.

Most novels are told by the narrator after the fact, that is, retrospectively. But a few novels suggest
that the discourse and the story are simultaneous. There are even a few novels cast in the future
tense, which give the effect of prophecy, or in the conditional mood, that is, envisaging what might
have happened.

In English, the standard verb form for retrospective accounts is the ordinary past or preterite tense.
Conrad's The Secret Agent (1907) begins, "Mr. Verloc, going out in the morning, left his shop
nominally in charge of his brother-in-law." Sometimes,however, the retrospective discourse is
couched in the present tense, whether progressive (the verb "to be" followed by the present
participle---"John is singing") or simple ("John sings"). The present tense is used in ordinary
conversational narrative to impart a kind of vivid immediacy to the tale: the person next to you in a
bar says "So I'm standing there and he comes up to me and makes this remark ...". Some novels do
the same: Ann Beattie's Chilly Scenes of Winter (1976) begins, "It is raining,and Sam's hair streams
down his face." Quite often in fiction, when the narrator tells the story in the simple present tense,
we understand that the account is retrospective, that the present tense refers to past activity. This is
a use of the "historical present" tense to vivify the action.

Even when the narrator tells the story in the preterite, the story may be vivified by the use of present
time adverbs (technically called deictics) to convey the sense of immediacy felt by a character.In
Edgar Allan Poe's "The Tell-Tale Heart" (1843), for example, the narrator refers to his feelings
upon being interrogated by the police: "No doubt I now grew very pale" (rather than "No doubt I
then grew very pale").

Finally, there is a kind of distance that we may call personal, referring to the degree of sympathy or
empathy that the narrator exhibits toward the character, whether emotionally, morally,
sociologically, or philosophically. The narrators of Virginia Woolf's Mrs. Dalloway (1925) and
Katherine Mansfield's "The Garden Party" (1922), for instance, are mostly in harmony with the
protagonists. But the narrator may be very distant, speaking of the protagonist with detachment or
even distaste, as is the case in Conrad's The Secret Agent. Even a character-narrator may be very
distant from and unsympathetic toward her former self. Sympathy toward the protagonist has been
termed consonance; indifference or antipathy has been termed dissonance (see Cohn, 1978).

The properties of narrators described above are not rules but norms, that is, conventional
characteristics of narrators as manifested throughout the history of the novel. But art is dynamic,
and authors press the limits of these norms, seeking ever new ways to tell stories. This is
particularly true of 20th-century novelists, both modernist and postmodernist. To many an author, a
narratorial norm is like a red cape to a bull, not something to be revered but something to be
subverted or transformed into new realms of expression.