Good Prose: The Art of Nonfiction

by Tracy Kidder

  Some general truths apply. For instance, one sure way to lose the reader is trying to get down everything you know about a person. What the imaginative reader wants is telling details. Characters can emerge in long descriptive passages, as in Tolstoy, but brevity can also work. Graham Greene rarely gives us more than a detail or two—a face “charred with a three days’ beard” or a pair of “bald pink knees”—and Jane Austen often gives us less than that, and yet the people those writers create have come alive for generations of readers.

  Whether it is brief or lengthy, mere description won’t vivify a statue. What we want are essences, woven into a story in moments large and small. A character has a wart. You could describe it in detail, but the reader would probably see it more clearly if you described not the wart but how the character covers it when he’s nervous.

  Or take the small matter of a character’s age. It’s a bad idea, categorically bad, to offer it via a tired pretext: “She looked a decade younger than her seventy years.” Nor do you want to tack her age onto her name, as in a news story. Instead, think of a character’s age not as an obligatory but as a potentially significant fact. Wait for the moment when we need to know her age in order to understand an event in the story. Or, if her age has no narrative importance, slip us the number quietly at the moment when, if you were reading, you would need it for the picture of the woman that is forming in your mind.

  There are multiple ways of presenting characters. Some people will go a long way toward defining themselves through their own speech. Others may talk volubly but without revealing much. In all cases, diligent reporting produces more quoted material than a book or article can hold. Selection is necessary, and, a more delicate matter, authorial management. Take for instance a character who talks so colorfully that the writer has to acknowledge the fact, as in The Earl of Louisiana, A. J. Liebling’s extended profile of the flamboyant governor Earl Long. At one point, Liebling recounts a campaign speech in which the governor tells a crowd that a political opponent wears expensive suits, then says of himself: “A four-hundred-dollar suit on old Uncle Earl would look like socks on a rooster.” Liebling breaks in:

  It is difficult to report a speech by Uncle Earl chronologically, listing the thoughts in order of appearance. They chased one another on and off the stage like characters in a Shakespearean battle scene, full of alarums and sorties.

  An author can also rely on others to comment for him. Liebling spends most of the first four chapters of The Earl relating others’ stories and opinions of the governor. Point of view is a powerful tool for making people materialize in our imaginations. Characters who seem lifeless at the end of a draft can sometimes be revived by letting the reader see them through another set of eyes, or several sets of eyes, as in Liebling. Minor characters encountered in the flow of a story can function as lenses on the main characters. Minor characters can also serve as foils, the human background against which we see the main characters more clearly.

  A minor character can also become a liability. Sometimes in a draft, a minor character grows dimensions and threatens to wreck the narrative scheme. It is assumed that something like this happened to Shakespeare. In Will in the World, his book about Shakespeare’s life and times, Stephen Greenblatt writes: “… in an anecdote that circulated in the seventeenth century, [Shakespeare] is said to have remarked of Romeo and Juliet that he had in the third act to kill Mercutio—the wildly anarchic mocker of romantic love—before Mercutio killed him.” Probably every serious writer, in nonfiction and fiction, has had the experience of seeing a minor character grow so vivid as to seem more important or at least more interesting than the principals of the story. It’s a better kind of problem than many. Such characters are always hard to give up, but they are often diagnostic: maybe you haven’t worked hard enough on the main characters, or maybe that minor character ought to be major, and if so, maybe you should ask yourself if there is something wrong with the point of view or the structure of your story.

  Some minor characters who remain minor are, as E. M. Forster puts it, “flat,” known in one dimension only. The type is abundant in Shakespeare’s plays and in any work that includes a wide cast. Flat characters are part of the glory of Charles Dickens. The dinner guest Twemlow in Our Mutual Friend is an extreme example of flatness, known mainly by his resemblance to a table:

  There was an innocent piece of dinner-furniture that went upon easy castors and was kept over a livery stable-yard in Duke Street, Saint James’s, when not in use … The name of this article was Twemlow. Being first cousin to Lord Snigsworth, he was in frequent requisition, and at many houses might be said to represent the dining-table in its normal state. Mr. and Mrs. Veneering, for example, arranging a dinner, habitually started with Twemlow, and then put leaves in him, or added guests to him.

  Whether flat or multidimensional, minor or central, characters need settings in order to live. A setting can be an actual place, but it is always more than that. Above all, a setting tells what is at issue—what a character is trying to do, what a character fears or is trying to hide, hopes to gain or stands to lose, what a character is up against. Depicting such circumstances and feelings is one way you can get the reader imaginatively involved; something matters to this person on the page, and we can imagine its mattering to us.

  Sometimes what’s at stake, as in a narrative of revelation, is the author’s own quest to discover a character’s identity. The challenge implicit in all biographical writing is made explicit in these cases. In Joe Gould’s Secret, for example, we accompany Joseph Mitchell in getting to know an outlandish Harvard-educated bum and in ultimately discovering the poignant lie by which the man has lived. A. J. Liebling’s essay “Quest for Mollie,” less well known but no less beguiling, begins with Liebling in the role of war correspondent reporting from North Africa. He comes upon the body of a young American enlisted man and hears stories about him from other soldiers: the daring, wildly eccentric “Mollie” supposedly captured an entire regiment, six hundred Italian soldiers, single-handedly. We return to New York with Liebling and accompany him on his further search for Mollie’s identity, or identities.

  •

  Many well-wrought characters live in nonfiction, but the library of fictional characters is much larger. Nonfiction writers spend useful time there, trying to figure out what they can borrow and, equally important, what they can’t.

  When we read fictional and factual narratives, we conjure up characters through their deeds: characteristic actions and contradictory actions, behavior in moments of stress, of mastery, of weakness. Suggestions of a character’s motives may be implicit in the deeds, but many readers want more. We want to imagine that we know why characters do what they do and feel as they do. We want to understand characters in a story better than we understand ourselves. This, of course, is an illusion available only in fiction. The writer of factual stories is constrained by what the subject is willing and able to reveal.

  Fiction writers tell us that sometimes their imagined characters seem to take over, to grow and change on the page, acquiring unexpected qualities and doing unexpected things, just as if they were alive. Fiction writers can invent ways out of whatever problems this creates; if necessary, they can always kill off the character in the midst of the story, as Shakespeare did Mercutio. Nonfiction storytellers don’t have that option.

  In nonfiction, events and characters stand in paradoxical relation to each other. There is a fundamental difference between writing about a man who gets into an accident and writing about the accident. The event is the event. It happened. It’s a fact. As for the man, no one knows for sure who he really is, what skein of motives and desires led him to this event. And yet he, not the accident, is your fixed star. Once you have selected a person to write about, that person has become the central mystery you want to solve, knowing that you never will solve it completely.

  In your research, you spend all the time and perspicacity you can muster on that man. Eventually, you grasp a
vision of him in the round, a vision that honors his variousness. Then you try to re-create that vision in writing, presenting him through things you’ve seen him do, things you’ve heard him say, things you’ve heard others say about him. You choose those things from among many candidates in your notes, selecting them partly because they seem interesting or colorful or funny in themselves, but mainly because they express your vision of the character. It may be that when you met him, the man thought of his accident as the most important thing in his life. But suppose nothing about the accident enhances or complicates or expresses your sense of who he is. You can’t invent another, more revealing accident for him to be in, but you can choose not to write about the real one. The honest nonfiction storyteller is a restrained illusionist.

  STRUCTURE

  Things happen in time, and time is crucial in storytelling. Kidder and I once established a rule: you can mess with chronology, but you have to have a good reason to do so. This is one of those empty propositions that can have a good effect, this one as a check against the deliberate scrambling of narrative time, a gambit very appealing to some writers, but frustrating to readers when it is not employed in service of the story.

  Our rule was put to the test with Kidder’s Strength in What Remains. The book recounts the life from boyhood to middle years of a man named Deogratias, a native of the east central African country Burundi. The basic story contains many powerful and, at least superficially, improbable events: Deo’s escape on foot, first from civil war and then from the Rwandan genocide; the painful reconstruction of his life and psyche in New York City (Deo slept for some months in Central Park); the resumption of his education (at Columbia University); and ultimately his return to Burundi as an American citizen, intent on helping with the reconstruction of his native country.

  Although the events were all in the past, the story was nonetheless heavily reported. Kidder revisited the scenes of Deo’s travail, spending time with him in New York and Boston, interviewing people who had helped him, and traveling with him to Burundi and Rwanda.

  All this research in hand, how to tell Deo’s story? This became a problem of both point of view and time. Our solution had two parts.

  Kidder’s first draft roughly honored the chronological rule. It had wonderful moments, but as a whole it was unsatisfying. The chronology asked a great deal of the reader. Specifically, it asked American readers to sit still for an account of a painful childhood in a place most had certainly never heard of. (I did not have to imagine this ignorance—Deo was my introduction to the very existence of Burundi.) Another option was to start with the most dramatic moment, the harrowing story of Deo’s escape, but this would have committed the old theatrical crime of starting too high. Where would we go from there? Moreover, the reader wouldn’t have much reason to care about Deo, not having met him in less extreme and more familiar circumstances. Yet another possibility might have been to hew strictly to the order in which Kidder reported the story. This is often tempting and almost always a mistake. At crucial moments the presence of the narrator would have served only to mute the drama. How the narrator would ultimately appear was a problem to be solved, but this was clearly not the narrator’s story.

  With the comfort of a rough draft to remind us that we had a book, that it was just a matter of getting it right, we set about giving the story a new structure. We determined that it would be a good idea to get Deo to New York as soon as possible. But then how would we get him back to Burundi, and where would the narrator be?

  It was a summer day on the coast of Maine, the sort of afternoon when weather and season invite you not to be urgent about anything, but we were feeling urgent about this. We sat in silence for some time on the porch of Kidder’s cottage. I found myself sketching out a plan on a yellow pad. After a while Kidder said, “Well, this is all I’ve come up with,” and held up his own yellow pad. I smiled and held up mine. The drawings were identical—four arrows coming in from the left-hand margin, interleaved with four arrows coming in from the right. Kidder said, “We’ve been doing this too long.”

  But we were pleased, once we realized that we were indeed thinking of the same thing. In the first part of the book, Kidder would create two alternating chronologies, each moving straightforwardly in time. One chronology would begin with Deo’s departure from Burundi and take us through his time in New York. The second would recount his youth in Burundi, culminating in his narrow escape from war and genocide. This second chronology would end where the first began, and, as it turned out, it could end on a line that Deo remembered clearly, a line that would by that time, with Deo’s sufferings in New York already told, be nicely ironic: with the consular official in the US embassy in Burundi handing Deo a visa and saying, “Good luck in New York.” Some other refinements suggested themselves as we talked. The New York sections would get progressively shorter, the Burundian sections longer. And it was plain that the narrator had no useful presence: the story would be told in the third person through Deo’s perceptions—that is, it would be told in Kidder’s words mostly, but they would describe Deo’s memories.

  I suppose the book could have ended there. But much was left over from Kidder’s reporting trips. The original purpose of these had been to understand and verify Deo’s story, to gather detail firsthand. But the results went far beyond sights and sounds and smells: portraits of the people who had come to Deo’s aid in New York, and Kidder’s observations while traveling in Burundi. In seeing Deo revisit places of trauma, Kidder also witnessed some of its effects. He had the sometimes heartbreaking privilege of watching a man deal with ungovernable, tormenting memories. We wanted to find a form for these experiences, too, for Kidder’s first-person observations. We decided that a prologue would be the best way to establish the fact of a first-person narrator. But many pages told from Deo’s point of view would intervene. How to bring the narrator back? After a while it seemed there was no advantage in doing this other than boldly.

  In effect we decided that the story should be told twice. After 145 pages, the reader comes upon Part Two, and the disarming phrase with which the author steps onstage for the first time since the prologue: “I first met Deo in Boston, about a decade after he had fled Rwanda and Burundi.” Part One would be told through Deo’s eyes, Part Two through the author’s. This yielded many benefits. Much that got in the way in the rough draft—the observations Kidder had made during reporting, observations which did not come from Deo—now became useful. And Part Two, in allowing the reader to see how the author reported the story, worked to dispel any doubts a reader might have about the story’s credibility. Most important, retelling the story in the first person helped the reader see (and in candor it helped author and editor realize) what the book was really about. It was not just about extraordinary resilience and courage, but also about memory and how the mind can work to repair itself after devastating experience. This kind of thing has happened often in our years of working together. More than most writers, I think, Kidder discovers his stories by writing and rewriting them. In this case, finding the story’s structure was the means to finding its theme.

  So, a twice-told tale that starts near the end, interleaves two chronologies, and then the second time around is told through a different set of eyes, with a new chronology, which reenacts the author’s research. Not the simplest structure imaginable (and far from the one with which the book began). For all of that, the rule for us continues to hold. Don’t mess with chronology unless you have a good reason.

  —RT

  The fundamental elements of a story’s structure are proportion and order. Managing proportion is the art of making some things big and other things little: of creating foreground and background; of making readers feel the relative importance of characters, events, ideas. Often this means upsetting normal expectations by finding a superficially trivial detail or moment that, on closer examination, resonates with meaning.

  As for order, its fundament is time. Writers profit from knowing when events
actually occurred. It is always a good idea to construct a detailed time line and, for some, to write rough drafts chronologically. If you know the actual sequence of your story, all the details dated in relation to one another, you avoid muddle and misunderstanding, and can write with a feeling of authority, which tends to insinuate itself into prose.

  It is especially important to know a story’s chronology if you are tempted to alter it. For the writer, an important tension arises between chronology and deliberate alterations of chronology. As you’re writing, you ought to feel that pull. It challenges you. Do you want to tell parts of your story out of sequence in order to be arty, just to show off? Or do you have reasons that arise from your exploration of the story itself?

  Serious narratives offer us good reasons for caring what will happen to the characters. Why those things happen, especially the characters’ motives, is a higher order of question than what will happen next, but most stories lack propulsion if they lack sequence. Telling stories in chronological order has a distinguished lineage in Western literature, which includes, among others, all the classical narratives and most novels well into the twentieth century. One might say that the straight-ahead treatment of time reflects a moral rather than a psychological understanding of the world. In Jane Austen, characters define themselves and are judged by what they do, from present moment to present moment, event by event. We aren’t told that Mr. Darcy is courting Elizabeth because of what happened to him when he was eight years old. Rather, we are asked to believe that he has an essential nature, and that we should be more interested in what it is than in where it came from. This is, after all, the frame of mind in which most human beings spend most of their waking hours.

  The straightforward structure is not obsolete, even in our psychological age. Ron Suskind’s A Hope in the Unseen dramatizes the progress of a young African American named Cedric Jennings. We meet him as a student in an inner-city high school in Washington, DC. Jennings has a gift for math, and he has a code of behavior formed by his mother’s faith in him, his own faith in God, and a fierce pride. He is determined to escape the ghetto, and he does, improbably, to Brown University. In high school he is a pariah, scorned for his industry and ambition and what some see as arrogance. Things aren’t always easier for him at Brown, where he is an anomaly of another sort, genuinely poor in a subsociety of mostly middle-class African Americans, passionate in an academic world that values reason and nuance, and religious among secularists. Nonetheless, he ultimately finds his place, and—the psychological engine of the book—he finds a way to adjust and also to preserve the values that got him there. Significantly for students of structure, this book, though abundant in event and detail, could not be simpler in form: a straightforward chronology, with only the smallest detours to account for Cedric’s family history.