A Kite in the Wind: Fiction Writers on Their Craft

INTRODUCTION

What do you think of that for a kite? he said.

I answered that it was a beautiful one. I should think it must have been as much as seven feet high.

I made it. We’ll go and fly it, you and I, said Mr. Dick. Do you see this?

He showed me that it was covered with manuscript, very closely and laboriously written; but so plainly, that as I looked along the lines, I thought I saw some allusion to King Charles the First’s head again, in one or two places.

There’s plenty of string, said Mr. Dick, and when it flies high, it takes the facts a long way. That’s my manner of diffusing ’em. I don’t know where they may come down. It’s according to circumstances, and the wind, and so forth; but I take my chance of that.

CHARLES DICKENS, David Copperfield

Writing fiction is, like most writing, a solitary pursuit. Finding the words, the images, the characters, the scenes, the voice, and the structure that bring a story or novel to life, and that come close to fulfilling the writer’s ambition for the work—or, in the happiest cases, exceed that ambition—sometimes seems as improbable as life itself. In Climbing Mount Improbable, evolutionary biologist Richard Dawkins wrote,

However many ways there may be of being alive, it is certain that there are vastly more ways of being dead.... If you think of all possible ways of arranging the bits of an animal, almost all of them would turn out to be dead; more accurately they’d mostly never be born. Each species of animal and plant is an island of workability set in a vast sea of conceivable arrangements most of which would, if they ever came into existence, die. The ocean of all possible animals includes animals with their eyes in the soles of their feet, animals with lenses in their ears instead of their eyes, animals with one left wing and one right fin, animals with skulls around their stomachs and nothing around their brains. (99)

Bringing a piece of fiction to life, especially one with all the parts in the best possible place, requires preparation, patience, persistence, practice, and some combination of inspiration, intuition, and conscious manipulation. It makes perfect sense that at some point, during those long hours at the desk, or with the laptop warming one’s thighs, the writer might wonder, Isn’t there an easier way to do this?

It’s tempting to say that the answer is no. Assuming the writer in question is interested in discovery, as opposed to mere production, the blank page, or screen, always presents new challenges. Writing one good story, even a marvelous story, is no guarantee that the author can write another as good. Having written one novel does not, as everyone in the situation quickly realizes, guarantee any kind of success with the next one.

But to some significant extent the answer is yes. That answer comes in many forms: in books and essays, in classes and private workshops, even in podcasts—ways in which writers tell each other, Here is something that might help. Those things might be examples, explanations, or bits of advice. Most tempting—and most dangerous—are rules. Rules for writing fiction are sometimes presented boldly, sometimes disguised, but boil down to the reassuring This is how you do it.

While such firm instruction can be useful to the beginner, following it dutifully ultimately leads to the sort of writing Flannery O’Connor criticized when she wrote, in Mystery and Manners, So many people can now write competent stories that the short story as a medium is in danger of dying of competence. We want competence, but competence by itself is deadly. What is needed is the vision to go with it, and you do not get this from a writing class (86). It’s good to be a competent driver, good to be a competent cook; but the term competent artist is an oxymoron. Part of the artist’s challenge—the fiction writer’s challenge—is to transcend the familiar, and simply following rules will not lead to transcendence.

What this book—like its precursor, Bringing the Devil to His Knees: The Craft of Fiction and the Writing Life—offers, instead, is good company for that solitary pursuit, in the form of considerations of craft by writers actively involved in exploring possibilities. While there are plenty of insights, and while there is no shortage of strongly urged advice, none of these essays began as an attempt to tell anyone how fiction should be written. They began instead as personal investigations: attempts to understand why a decision in a particular story or novel seemed successful; to define a quality or problem or subcategory that seemed either unrecognized or unsatisfactorily defined; to pass on understanding gained by experience; to understand what, despite years of experience, resisted comprehension; and to pursue haunting, perhaps even unanswerable questions. Eventually, these musings took the form of oral presentations (more about that in a moment); all were revised for print.

The contributors’ common denominator is that they have all taught—some a few times, some over decades—in the Warren Wilson College MFA Program for Writers, the first low-residency program in creative writing. Founded by poet Ellen Bryant Voigt at Goddard College in 1976, the program—not only its design but also its students and faculty—moved to Warren Wilson College, in Asheville, North Carolina, in 1981. From the outset, the expectation at the residencies was that, as Ellen has put it, everyone—poets and fiction writers, faculty and students—would go to everything. (Sadly, at most colleges and universities, if one faculty member sits in on another’s class, the purpose is usually not to learn anything, or to contribute to the discussion, but rather to fill out an evaluation.) Faculty who lectured at Warren Wilson quickly adapted to this challenging audience—a combination of wonderfully talented students and a diverse group of peers with high expectations—and to the context. Instead of giving lectures they’d given a dozen times wherever else they taught, instead of passing along familiar guidelines, they created a tradition of reporting on their own investigations into aspects of the art that for one reason or another captured their interest.

Since everyone at a given lecture then goes on to a class, and then to a reading, and the next day to another lecture, more or less as a group, any one presentation is part of the larger, ten-day-long discussion; and since many of the same people return six months or a year later, the Warren Wilson lectures have become an ongoing conversation about poetry and fiction involving literally hundreds of writers. Representative essays have been published not only in Bringing the Devil to His Knees but also in Poets Teaching Poets: Self and the World and Poet’s Work, Poet’s Play: Essays on the Practice and the Art. Fiction writers on the faculty also created The Story Behind the Story: 26 Stories by Contemporary Writers and How They Work.

Like any good conversation, there is no strict logic or progression to the one that follows, but we’ve grouped the essays into sections for easy reference. The first concerns narration, with particular attention to various kinds of narrators and several perspectives on unreliability; the second, the strategic creation and presentation of character; the third, some of the roles of the visual, beginning with establishing setting; and the fourth, structural and organizational issues, from movement through time to the manipulation of information to create mystery and suspense.

But even those broad categories are too narrowly defined. These essays are true excursions and contain any manner of side trips: personal anecdotes, jokes, references to music and film. Anyone looking to be told what to do in a story or novel’s first sentence, how to write the second, and so on, is bound to be disappointed. Writers and readers interested, however, in contemplations of various aspects of the fiction writer’s craft will, we think, find this collection surprising, provocative, and even useful. Toward that goal of usefulness, the contributors and the editors have donated 100 percent of the book’s royalties to Friends of Writers, Inc., to provide scholarships for developing writers. Beyond that, the book is dedicated to everyone who has a voice in this rich, ongoing discussion of the writer’s craft at Warren Wilson—faculty and students, poets and fiction writers, participants in the past, present, and future. Rather than wanting to have the last word, the writers here hope to inspire the next one.

Peter Turchi

Andrea Barrett

part 1

• Narrative Distance and Narrative Voice

FREDERICK REIKEN

The Author-Narrator-Character Merge: Why Many First-Time Novelists Wind Up with Flat, Uninteresting Protagonists

During my years in an MFA fiction writing program, I wrote a novel that was never published. It was not for lack of trying, as I radically rewrote it three or four times during the course of those three years, made one more attempt the year after I finished the program, and showed it to half a dozen literary agents before finally putting it in a drawer and moving on. I now consider it the novel I had to write in order to teach myself how to write a novel, and soon after I let it go it became obvious that the novel shouldn’t be published, though it has taken me many years to determine why.

The material for the novel was derived from my year working as a field biologist in Israel’s Negev Desert, studying Persian onagers, a species of wild ass that had recently been introduced into the Israeli desert wilderness. I faced many interesting conflicts, including drug smugglers and terrorists routinely passing through the desert crater that formed the study area, as well as onagers that sometimes crossed into Jordan, where they were usually shot by Jordanian border patrols. Some of the animals wore radio collars, and one night the Jordanian TV news ran a story claiming that soldiers had shot a trained Israeli donkey carrying a bomb. After a while it dawned on me that what I was living through seemed a lot like fiction.

I started an MFA program at the University of California at Irvine about a month after I returned from Israel, and my adviser there was instantly in love with my idea for a novel that featured a young American biologist studying wild asses in Israel. I was twenty-three and had written a dozen or so short stories, perhaps three of which had any substance whatsoever. But I plunged in with the standard mixture of idealism and egomania, believing my adviser’s promise that if I wrote this book I’d soon be the next twenty-five-year-old literary star. My writing models were limited, but I immersed myself in a study of classic literature, one result of which was that my writing tended to emulate whoever I’d been reading. In the first draft of that novel—with working titles ranging from The Land of Solomon to Wild Ass Mirage—sections read alternately like Virginia Woolf, Gabriel García Márquez, William Faulkner, and James Joyce. The second draft was so derivative of Joyce that one of my friends joked that I should change the title of the novel to U-asses. In hindsight, emulating these authors was a valuable step in the formation of my own voice, but only two years later did I become conscious of what I was doing. I recall deciding that if I wrote a second novel, I would have to invent a voice that would, paradoxically, sound like me.

More significant than the pastiche of style, however, was the fact that there was no distinct story, which in turn was linked to the fact that all three point-of-view characters in this third-person narrative were quite passive, serving as observers of the action around them but rarely becoming dramatic focal points themselves. I’d written a kaleidoscopic portrait of an idealized small desert town where biologists studied wild asses and female soldiers fell in love with biologists, against the backdrop of the Israeli-Palestinian conflict, which had started while I was there. I tried and tried to make the characters more interesting and alive, and though they were somewhat improved in successive drafts, the whole thing never came together. Many readers, agents included, had similar comments—which I’ll sum up this way: "The writing is really great, some of the scenes in the desert very beautiful, and I love all the descriptions of wild asses. But I was bored and uninterested in the characters. It’s almost like you’re not seeing them."

When I wrote my first published novel, The Odd Sea, I was relieved to have solved this problem. At the time, I chalked it up to learning the art of restraint through working as a journalist for several years. Then in 1999 I began teaching at Emerson College, and in my first semester I taught a graduate workshop in novel writing. The student novels ranged in topic from a woman grappling with a rare body disorder to a young man whose MIA father returns from Vietnam in the 1980s, to a woman searching for family secrets in Northern Ireland. In my mind, all of these story conceptions had tremendous potential, but in each case my main reactions were: 1) I was bored and uninterested in the characters, and 2) it seemed as if the author wasn’t seeing them. I have now taught this novel course every fall since then, and at least half of the novels I wind up seeing share this problem. My first reaction was that most novelists needed to go through this, to do what I did—spend years writing a failed novel that would teach them how to write a successful one—and I would still say this is true. Most would agree that Stephen Hero, Joyce’s first version of what would later become A Portrait of the Artist as a Young Man, is almost unreadable, and that The Voyage Out was similarly a novel Woolf had to write in order to teach herself what she didn’t want to write.

During my second year at Emerson I had my first glimpses of understanding about why so many first novels by MFA students, particularly those that are autobiographical in some way, are, in a word, flat. When I say it feels as if the author isn’t really seeing the characters, I am talking about the particular flatness that comes from what I call the author-narrator-character, or ANC, merge.

At its simplest, the merge may be thought of as the narrative structure that occurs when an author, for reasons ranging from naïveté to authorial narcissism (which often go hand-in-hand), fails to invent and/or reinvent—in the case of autobiographical novels—the main character, both visually and in relation to some objective external context. The author has not separated him- or herself imaginatively from the character—has not conceived the fictional construct as an other—and is in effect stuck inside the character, usually right behind the character’s two eyes. What typically results is a narrative in which there is virtually no distance between the story’s narrator and character, and a sense that the main character is nothing more than a narrating device—and hence not much of a character at all. Most aspiring novelists are aware of this phenomenon—the first-person narrator or third-person point-of-view character as passive observer—and it’s my sense that a problematically passive protagonist is often, though not always, a merged one. With a merged narrator and character, both are neutralized. Not only is there a flat character; there is also no narrating apparatus to cue us on the logic of the story flow—that is, no apparatus for translation of what’s going on dramatically—and hence no perspective, texture, or depth, and frequently a complete lack of narrative voice.

Separating Author, Narrator, and Character

Before this can make much sense, we must be able to delineate instinctively between the three domains of author, narrator, and character, which is not as simple as it sounds. For one thing, we often mistake the narrator or even the character for the author. Strictly speaking, an author is a human being who exists outside the novel’s textual universe. (Under this strict definition, the same would hold true even in the case of nonfiction memoir.) The author is, of course, the book’s prime mover, but as a biological creature the author can never physically enter the textual dimension and hence is limited to a presence that depends on the level of symbolic correspondence between the author and his or her textual components—the narrator(s) and character(s). Some theorists have even claimed that a novel’s power rests with the paradox that an author asserts his or her presence through a form of absence.

Meanwhile the narrator, strictly speaking, is a construct of language, invented for the purpose of both presenting and translating the novel’s action such that a reader can stay oriented with the narrative’s sentence-by-sentence flow. Every thought, gesture, or action—indeed, every absence of thought, gesture, or action—falls under the narrator’s jurisdiction, as do the mechanics of the writing, right down to choices of punctuation. Obviously the author makes these decisions, but they are presented via the narrator, who governs the textual universe and shapes our relationship with the characters, who are, of course, the focus of a conventional, storyoriented literary novel. The ANC structure can be represented as a series of so-called Chinese boxes, in which the narrator’s domain subsumes that of character and the author’s domain subsumes those of both narrator and character:

Let us first consider the case of a first-person narrative. One might immediately suggest that when the narrator is also the book’s protagonist, there is no distinction to be made between narrator and character. To which I would say: Wrong. I’d go so far as to argue that the key to a successfully executed first-person novel usually lies precisely in the relationship between that first person as narrator and as character. Consider Holden Caulfield in the opening passage of J. D. Salinger’s The Catcher in the Rye.

If you really want to hear about it, the first thing you’ll want to know is where I was born, and what my lousy childhood was like, and how my parents were occupied and all before they had me, and all that David Copperfield kind of crap, but I don’t feel like going into it, if you want to know the truth. In the first place, that stuff bores me, and in the second place, my parents would have about two hemorrhages apiece if I told anything pretty personal about them. They’re quite touchy about anything like that, especially my father. They’re nice and all—I’m not saying that—but they’re also touchy as hell. Besides, I’m not going to tell you my whole autobiography or anything. I’ll just tell you about this madman stuff that happened to me around last Christmas just before I got pretty run-down and had to come out here and take it easy. (1)

As we learn, the book is structured as a story in which Holden—who is now out here, in a place that lies beyond the time line of the story’s dramatic action—is looking back at all this madman stuff that occurred during a period of days in the previous year. In other words, there is an implied separation between where Holden sits as narrator and the time frame he occupies as a character. In this opening we are with him in the expository domain of Holden as narrator, but within a page or so, as is the case with most novels, we enter the domain in which Holden begins to look at himself as a dramatically oriented character. In this case the transition occurs in the novel’s third paragraph: Anyway, it was the Saturday of the big football game with Saxon Hall. The game with Saxon Hall was supposed to be a very big deal around Pencey. It was the last game of the year, and you were supposed to commit suicide or something if old Pencey didn’t win. I remember around three o’clock that afternoon I was standing way the hell up on top of Thomsen Hill, right next to this crazy cannon that was in the Revolutionary War and all (2).

Here Holden establishes the real-time setting and the ground situation of the story and then places himself, as a character, on top of Thomsen Hill. As the passage continues he moves back and forth between the character domain and the narrator domain, which orients the story. By page 4, he reaches a point where the dramatic or character mode becomes predominant. Note that Holden’s strong narrative voice greatly assists in this movement. It is a perfect example of how useful the rhetorical effect of a strong first-person voice is in allowing the narrator to move effortlessly between the character domain, in which the story is rendered, and the narrator domain, in which the story’s action is translated, commented upon, and placed in the larger context of the character’s life trajectory. The voice also helps to characterize Holden, even when he is speaking within the domain of narrator. Part of the power of strong voice-driven fiction comes out of this synergistic effect.

Many first-person narratives I have read by beginning novelists never visualize or establish the first-person narrator as a physically placed character—that is, envisioned in a sequence of temporally contextualized, dramatically evolving scenes. The narrator spends an inordinate amount of time reflecting internally—that is, navelgazing—and the action that does occur is presented in relation to other characters or is simply summarized rather than dramatized in relation to the protagonist/narrator. Within the author’s imagination, I believe, the author, narrator, and character are perceived as synonymous rather than being structured as separate domains. The result can be an almost essaylike compression of the three. This happens in intentionally autobiographical material as well as completely invented material, although in such cases there are usually similarities between author and protagonist or at least subtle autobiographical associations.

With Holden, we can also consider how the author-narrator-character relationship sets up the effect created by a limited or unreliable first-person perspective. Holden, of course, is absolutely clear on the veracity, legitimacy, and logic of his every thought or action, and this is made clear in the relationship between Holden-as-narrator and Holden-as-character. The relationship, however, of author Salinger to narrator/character Holden is quite different from that of Holden-as-narrator to Holden-as-character, and the result is a dialectic in which these two relationships interact and in which we, as readers, are situated next to the author:

Salinger’s perception of Holden objectively renders him as lost, confused, and disoriented, even as Holden is telling us otherwise, and the pathos we feel is largely tied to the rift between what we and Salinger see objectively and what narrator Holden, in his desperation, perceives with regard to his own odyssey. If we think about certain exquisitely rendered unreliable first-person narratives— Eudora Welty’s story Why I Live at the P.O. or Kazuo Ishiguro’s novel The Remains of the Day, for example—we can instantly see this effect. And if we think about the ANC relationship, we come to realize that in addition to whatever dramatic tension is occurring, we are equally riveted by a different kind of narrative tension that arises from the rift between the author’s objective rendering of a character and that character’s limited perception of what’s going on. It’s also worth noting that the power of well-executed metafiction is tied to a similar strain of narrative tension, derived from a manipulation of the ANC relationship such that a character takes on the awareness of a narrator or author, or in which the author pretends to be a character.

Though I will not dwell on it here, I want to point out that even in the special case of a first-person present-tense narrative—a point-of-view choice that, because there is no implicit temporal gap between narrator and character, relies on high artifice—there is still a clear separation between narrator and character in successful story-oriented narratives. Usually this is because the artificiality of the tense actually does allow for a temporal gap, though it may be a gap of seconds rather than years and it is usually combined with a subtly metafictional narrative objectivity (that is, the character can shift between the domains of narrator and character much like Holden Caulfield, even though this degree of instant reflective clarity would be impossible in real life). Often a great deal of narrative space is devoted to flashbacks that get presented in the past tense. Without a successful separation, a first-person present-tense narrative tends to become a stream of impressionistic minutiae that may engage on the basis of tone for a short while but that will soon lapse into annoying stylistic pretension. Or, as in the case of the following anonymous student example, it may result in a narrator and character being merged to the extent that we can barely tell what’s happening:

The warmth from his hand passes through my eyelids and into the tired sockets encasing two spheres with ocean-blue enigmas, as he had called them. I attempt to place my hand upon his, desperate to feel his touch again, only to find nothing. Frantically tossing the comforter from my shirtless torso, I open my eyes to scan the room. The bureau we found on the street stands regally in the corner leaving enough space to hide behind. This could all be a cruel trick.

Moving on to third-person narratives, the distinction between narrator and character is, on the one hand, obvious, as the narrator is not embodied in the storyline. However, I have found ANC merges to be more common in third-person than in first-person narratives, and in several cases students have solved the merge problem by switching to a first-person past-tense point of view, where a natural split occurs between the future place from which the narrator is reflecting and the timeline of the character moving through the story. With a third-person point of view, the sense of where the narrator is in relation to the story can get a lot fuzzier, since the narrator is not situated in time. In order to consider the position of a third-person narrator, a useful distinction can be made with regard to what John Gardner, in The Art of Fiction, refers to as the psychic distance between narrator and character. He offers these five examples, in order of progressively diminishing psychic distance:

1. It was winter of the year 1853. A large man stepped out of the doorway.

2. Henry J. Warbutton had never much cared for snowstorms.

3. Henry hated snowstorms.

4. God how he hated these damn snowstorms.

5. Snow. Under your collar, down inside your shoes, freezing and plugging up your miserable soul. (111)

In example 1, we have the third-person narrator looking at an unnamed man with great psychic distance. This is the type of longrange psychic distance you might expect at the opening of a novel. In example 2, we are getting situated in the character’s point of view, but we still hear the oral quality of the narrator, who is orienting what seems more likely to be the beginning of a short story. In example 3, this voice-over quality of the narrator drops off, and as a result we no longer see the character; rather, we are situated inside the character’s head, privy to his perceptions but no longer looking at him from the outside. In example 4, Gardner makes use of a technique known as free indirect discourse, through which the character’s thoughts are subsumed by the narrator. In example 5, we get stream-of-consciousness interior monologue, which is technically a variety of free indirect discourse.

Most third-person narratives proceed with constant modulation of the psychic distance, moving like a camera eye between longrange establishing shots and very limited, close-range character point of view, and then back out to longer-range shots again. But in a case in which the author has not fully imagined the point-of-view character—often because the author has not yet truly conceived the character as a bona fide other—the ANC relationship gets structured so that there is little or no psychic distance between narrator and character, no way for us to see the character moving through a setting or situation, and hence, though unintentionally, what I am calling a merged effect. With regard to Gardner’s examples, we might say that in the case of a problematic merge, the psychic distance never becomes greater than and usually stays continuously at what we see in example 3. That is, the most we get is a sense of being inside the character’s head, but we can never actually see him.

This effect is apparent in the following anonymous student example (in which the distance remains at roughly that of Gardner’s example 3):

Joey stepped over the snow and a pang of jealousy passed through his body. Being that it was so cold, he wondered how his girlfriend Mia had been so comfortable that afternoon in just a cotton sweater. She is so beautiful, he thought, again and again, and wished he had never met her. She is with Dave, he thought, remembering what his father always told him about young, attractive girls. He heard the snow crunching under his feet and saw again the spot where she had left him. The footprints they had left in the snow were gone.

I want to note that the above point-of-view system is not egregiously problematic. We do see the girlfriend Mia, in her cotton sweater, and theoretically we might come to see Joey in the next paragraph. For a story about Joey, this point-of-view system would be fine if the author pulled back periodically and allowed us to see him with more distance. Even if we don’t ever get enough psychic distance to see him, it might be possible to get away with this merged point of view in a short story, as long as what he’s looking at proves interesting and specific. But in a novel focused primarily on Joey, a merged point of view is likely to undermine the narrative, and the writer will probably lose the reader by page 30. As Gardner notes in The Art of Fiction, Fiction does its work by creating a dream in the reader’s mind (31). If we can’t see the protagonist, we usually can’t dream him, and as readers we will quickly tire of trying.

Be aware that an intentional (that is, conscious) merge of narrator and character may be a useful stylistic effect and, unlike the accidental merging I’m describing as problematic, may aesthetically enhance a story rather than detract from it. Consider this short excerpt from the opening of Joyce’s A Portrait of the Artist as a Young Man: When you wet the bed first it is warm then it gets cold. His mother put on the oilsheet. That had the queer smell (7).

Here we have an obvious merge—the narrator has taken on the voice of young Stephen Dedalus—but because it is done consciously narrator and character remain discernable within the merge. Note that the technique is more or less the same as in Gardner’s example 4. That is, Joyce moves the psychic lens to the point that the narrator subsumes the character’s voice and/or thought patterns. But I would be remiss not to acknowledge that Portrait eventually does become a problematically merged narrative. Once Stephen Dedalus gets beyond the age of ten or so, the playful free indirect discourse drops off and is replaced by a more serious and weighty narrative tone in which, as Vladimir Nabokov once noted, Stephen is a rather abstract young man . . . never quite clearly visualized by the reader, a projection of the author’s mind rather than a warm new being created by an artist’s imagination (286). It’s not surprising that Stephen is thought to be an autobiographical portrait of the young Joyce himself, whereas Joyce’s most successfully envisioned character, Leopold Bloom (an ad canvasser of Hungarian-Jewish origin who stars as the wandering protagonist of Ulysses), was truly a creation of Joyce’s artistic imagination. While Bloom seems to live and breathe even during extensive sections of stream-of-consciousness interior monologue, Stephen, in both Ulysses and the second half of Portrait, tends to feel like a mouthpiece for the author. We tolerate Stephen because Joyce happened to be an intellectual genius, capable of endowing even a deeply merged narrative with brilliance.

Other examples of an effective conscious merging of narrator and character may be seen in the use of free indirect discourse by authors ranging from Virginia Woolf to Alice Munro, each of whom has had her share of merged narratives but succeeds in creating warm new beings most of the time, often through the use of free indirect discourse. The Beggar Maid—a collection of ten interrelated stories about a single protagonist, Rose, whose life strongly resembles that of Munro—is a marvel of characterization despite its autobiographical quality and heavily expository style. The secret seems to be Munro’s strong-voiced, ironic, and playful third-person narrator, who constantly appropriates protagonist Rose’s words, voice, thoughts, and perceptions, such that the narrative itself becomes a mirror of Rose’s personality. For example, this expository passage follows a scene in which Rose’s ex-husband, Patrick, makes a hateful face at her during a chance run-in at an airport many years after their divorce:

Sometimes when Rose was talking to someone in front of the television cameras she would sense the desire in them to make a face.... They were longing to sabotage themselves, to make a face or say a dirty word. Was this the face they all wanted to make? To show somebody, to show everybody? They wouldn’t do it, though; they wouldn’t get the chance. Special circumstances were required. A lurid unreal place, the middle of the night, a staggering unhinging weariness, the sudden, hallucinatory appearance of your true enemy. (100)

We can also notice the manner in which Rose’s thoughts become synonymous with the narrative. Yet because they are such specific thoughts, the character of Rose is discernible, and preserved. A similar effect can be observed in the following example of free indirect discourse in Woolf’s Mrs. Dalloway. You can hear the character’s voice running clearly through the narrative, and the effect is a simultaneous telling and showing and a constant modulation of psychic distance, even though the narrative is driven primarily by Clarissa’s thoughts:

That is all, she thought, looking at the fishmonger’s. That is all, she repeated, pausing for a moment at the window of a glove shop where, before the War, you could buy almost perfect gloves. And her old Uncle William used to say a lady is known by her shoes and her gloves. He had turned on his bed one morning in the middle of the war and said, I have had enough. Gloves and shoes; she had a passion for gloves but her own daughter, Elizabeth, cared not a straw for either of them.

Not a straw, she thought, going on up Bond Street to a shop where they kept flowers for her when she gave a party. Elizabeth really cared for her dog most of all. The whole house this morning smelt of tar. Still, better poor Grizzle than Miss Kilman; better distemper and tar and all the rest of it than sitting mewed in a stuffy bedroom with a prayer book! Better anything, she was inclined to say. But it might only be a phase, as Richard said, such as all girls go through. (15)

The personality of this third-person narrative voice—which mirrors that of Clarissa Dalloway—achieves the same kind of synergistic effect we saw earlier in the first-person Holden Caulfield example. The free indirect discourse animates the narrator’s voice and in doing so helps to characterize Clarissa Dalloway, while at the same time reinforcing the implicit textuality of that narrative voice—that is, as separate from a living, breathing author, Virginia Woolf.

By contrast, what you often get with an unintentional merging of narrator and character is the so-called fallacy of imitative form, or imitative fallacy. Because there is never any distance between narrator and character, there is no