The Construction Zone

Snow White, Revolutions, the American Dream, and Other Fairy Tales

The first illustrative narrative is one of our own (Coulter, 2003). The reason we use it here is that, while the narrative was constructed, every choice was considered and reconsidered. For example, Coulter experimented with different tones, voices, and the like to see how each squared with the evidence and represented the experience of the participants with fidelity and respect. As we reflected on this decision making and the effects of each choice, we formulated general hypotheses about narrative construction. Thus we had more material to draw on to describe the use and narrative effects of literary elements. In Coulter’s study of the experiences of high school students in a program for English language learners, she functions first as the teacher of those students, then as the researcher of those experiences, and finally as the author of the students’ stories. Eight people, including Darek, Vicki, and David, function first as students, then as research participants, and finally as the characters in the nonfiction novel that Coulter wrote. (All names in the study are pseudonyms.) Darek divides his time between a Northwestern city and his Cu’pik village. His mother is Cu’pik, and his father, now deceased, was African American. As Darek navigates life at “Northwest High School,” he is faced with questions of access and identity. Vicki is a second-generation immigrant whose father is from Mexico and whose mother is from El Salvador. Her story reflects the choices she had to make. David lives with his single mother, who is first-generation Filipino. His story portrays insight into low expectations and tracking in schools.

This Boy’s Life

For our other three story selections we had less detail. We could not reproduce the acts of constructing the narratives except by drawing on the texts themselves. This Boy’s Life (Wolff, 1989) is a memoir that encompasses the author’s preadolescent and adolescent years and leaves off as he drops out of college to join the army. He begins the narrative as he and his mother drive west to Utah, escaping the life of wealth and Protestantism of his father and older brother in Connecticut, looking for transformation and independence. He changes his name from Tobias (Toby) to Jack and hopes to become a different boy. His mother hopes to strike it rich by mining uranium in Utah. “But we were late—months too late,” he writes. “Moab and the other mining towns had been overrun. All the motels were full. . . . There were no jobs, and people were getting ornery” (p. 6).

His mother finally settles for a job and a poor place to live. He goes to Catholic school. An abusive man from her past tracks them down. They escape by driving farther west. Later, she is courted and then marries a man named Dwight, and the narrative intensifies to reveal the depth of each character and the conflicts among them.

Why did we choose this text? It does not match the conventional education research format of preserving observation notes and interview transcripts contemporary with the acts recorded. It is a retrospective account, similar to historical research. It potentially suffers from a self-interested reconstruction, but in that respect it is the same as all memoirs. Its subject matter is not focused on formal schooling, but nevertheless the author’s school life constitutes much of the experience that he deems worth remembering and writing about. We chose this text because it is iconic in courses and programs in writing for its vivid detail and consistent, strong voice, and because of Wolff’s ability to construct scenes—all of which is instructive for narrative researchers.

Down by the River

The third narrative, Down by the River (Bowden, 2004), we include because of its complex structure and interrelationships of author and characters over time. The author investigates a murder he reads about in the newspaper. The victim is a man named Bruno who lives in El Paso. A young man from across the river in Juarez is arrested for the murder. The circumstances of the arrest, conviction, and overturning of the conviction leave open the question of whether larger forces were involved. Bowden finds that the incident is connected to international drug cartels and the governments of both the United States and Mexico. Everywhere he turns in his own study, he finds that the connective tissue that ties the various elements together makes the story almost impossible to prove by the standards of research or journalism. The literary forms he uses mimic the complexity and ambiguity of the content and provoke complex and ambiguous responses in the reader.

Why did we choose this text? Among the four, only this one typifies the postmodern turn in literary criticism (Barone, 2000). A noted investigative journalist, nonfiction writer, and teacher of literary nonfiction at the University of Arizona, Bowden writes with variations in tone, voice, distance, and style and thus requires readers to make their own sense of his stories. By avoiding pat answers, he forces readers to rethink their basic views of the world and of the phenomena he describes. If there are conclusions or themes, they are of the reader’s making, not the author’s. We chose this particular text because of these qualities and because we could not find anything approaching its creative and skillful use of literary elements in the education research literature.

Sneaky Kid and Its Aftermath

The fourth narrative we use to show literary elements is Sneaky Kid and Its Aftermath (Wolcott, 2002). The initial section of this book is a redescription of Wolcott’s ethnographic study of schools and how they fail some students. He uses as his case a young man, named Brad, who has taken up residence on Wolcott’s property. Brad provides extensive material, and the resulting ethnography is published in a scholarly journal. The narrative then is transformed into a story of the personal relationship between the researcher and Brad. After the relationship ends, news comes that Brad has had a psychotic break. He blames Wolcott and comes back to attack him and burn down his house. The next segment of the narrative is the story of Brad’s trial and the censure directed at Wolcott from his scholarly community and from the society in general. We chose this particular nonfiction narrative because of the rich material the author has made available in his other works and his prominence in teaching about the writing of research.

In the preceding subsections, we have attempted to justify our use of these four texts. They do not represent any universe of texts, but they do represent authentic examples of the ideas we try to describe in this article.4

We now turn to our discussion of literary elements as research tools in narrative research. We first introduce each literary element and describe how it is used in writing in the general sense. We then discuss how the element can be used in research, drawing from our four narratives for examples.

Point of View

Point of view is the lens through which a narrative is told. The reader sees and hears through the sensibilities and emotions of a focalized character. In narrative research, point of view consists of how the story presents the relationships between researcher and participants, and the relationships of the participants to each other. The researchers are the writers, and the characters are participants, people who have played a role in the lives of the participants, or researchers.

In both narrative theory and narrative research methodology, the relationships among writers, characters, and readers have become contentious issues. The writer controls the telling. When the writer decides which character has agency to “speak” the story or carry the action forward, she assumes a position of privilege. In narrative research, the writer who is also the researcher—as she transforms research participants into characters and narrators in her story—faces the same issues and more, often juxtaposing her position of societal privilege against the marginalized position of her participants. Barone and others argue that the researcher must attempt to counteract the inherent privilege of this relationship from the very beginning, making literary decisions that debunk it (Barone, 2000; Emihovich, 1995; Goodson, 1995; Nespor & Barber, 1995). Relevant to point of view, there are literary decisions that researchers can make to mitigate the researcher–participant relationship. These are discussed next, in connection with the other literary elements.

Person

Addressing the complex issue of point of view involves deciding on person. Who tells the story? In some cases the author is the narrator, but the author can also construct another, a character, as narrator. The narrator must stand somewhere in relation to the actions taken by the characters in the story. What knowledge does the narrator have available at the time of the narrating? The author can choose from first, second, or third person and can choose whether a character, a narrator as character, or the author herself tells the story. A story told in third person uses the third-person pronouns—he, she, it, or they—and tells the story from a narrator’s perspective, whether or not that narrator also plays a part in the action. A story told in first person uses the first-person pronoun “I.” If that story is told by a character, it is filtered through her unique perceptions. We later present excerpts of narrative constructions in the third and first persons.

The following excerpt represents a passage Coulter constructed to illustrate third-person storytelling:

I.

Darek stood up and started gathering dishes from the dinner table. He clanked the pots a little too loudly, and then tried to calm himself. It wasn’t Danny’s fault that he had failed two classes. Danny looked over at Darek. He knew that he had made Darek angry. He hadn’t meant to. He just wanted his mother to know how well he did in school. The mother looked at her two sons. Good-natured Danny, who tried so hard in school, and strong Darek, who looked after Danny. Darek must not have done so well, she thought. It had been such a hard year for him. She hoped he wouldn’t be too hard on himself.

Given the concern about the inherent privileging in the researcher–participant relationship (Barone, 2000; Emihovich, 1995; Goodson, 1995; Nespor & Barber, 1995), both first- and third-person renderings can be problematic. How can a researcher get into the head of a participant and tell her story from her perspective, using her voice, her words, her thoughts, as in first person? But then how can a researcher presume to get into everyone’s head, as in the third-person point of view? Richardson (1990) described the dilemma in “occupying a godlike position” in narrative research texts:

Do researchers have the right to speak for others, distancing themselves from the text, acting as if their own subjectivity were not being inscribed in the text? And, conversely, how can the credibility of the writer’s claims to knowledge be decided?

Omniscience

The term omniscience is used to refer to the amount of knowledge the narrator (and particularly the author as narrator) assumes in the narrative. The omniscient narrator is an objective “all-knower,” possessing all of the facts about characters and events and having access to the heart and mind of every character, such as in Excerpt I, above, in which the narrator knew what Darek, Danny, and their mother were thinking. Omniscience functions to provide general descriptions and interpretations and thus has a place, for example, at the beginning of the story, where the author establishes the initial state of affairs, and between scenes, where the author indicates how much time has elapsed and how things have changed. Writing the scenes themselves calls for use of rich and thick description of the concrete particulars necessary to establish the characters’ actions and intentions. Distanced telling backs up and fills in between the focused showing (thick description). Writing a passage such as “Seven years went by between the time of Chansy’s death and Darek’s reflection on it” may convey the researcher’s omniscience. But it also functions to structure the plot line. Researchers cannot represent every single occasion that happened in the participants’ lives but must choose which events to concentrate on and which to highlight; otherwise, the “map” of representation would be as extensive as the “territory” of the life it represents (cf. Becker, 2007). Choices are made based on the significance of the included event in the analysis of the evidence as a whole. This is how Geertz (1973) distinguished thick description from simply detailed or rich description.

The omniscient narrator has access to and can describe the inner thoughts and actions of every character. Thus the omniscient third person serves the purpose of economy even as it gives the impression that the narrator knows equally and completely the thoughts of each character. This impression is problematic. Researchers hope to construct adequate understanding of the lives of their research participants, but most acknowledge that their understanding, even after years of patient and thorough work, falls short.5

Another aspect of point of view is limited omniscience, in which the narrator has access to the thoughts and sensibilities of one character but not other characters. This perspective lends the reader a sense of closeness to the story because it is most similar to what we experience in life. In the following example, Coulter (2005) used the limited omniscient point of view from the perspective of 12-year-old Sahar:

III.

At lunch Taylor Maloney and her groupies had thrown grapes at her again. She had been walking to her usual spot at lunch recess: the picnic table in the far corner, right under the mesquite tree. Suddenly she felt something wet hit her cheek. She looked down and saw a squished grape on the ground. She didn’t need to look up to see where it had come from. Taylor and her friends always sat in the center of the square, on a short cement wall that had been built around the biggest tree on campus. There they engaged in various forms of social torture. The latest was “graping” the least popular passers-by.

“What happened Sa-hair?” Taylor called out. Sahar kept on walking. Taylor persisted. “Oh my God! There’s a grape in my sa-hair!!” she said and everyone laughed. Sahar kept her eyes on the pavement beneath her feet. Taylor had a short attention span. Out of sight, out of mind. She finally made it to her place under the mesquite tree. Wiping the rest of the grape off her cheek, Sahar watched Taylor from afar. She was on to her next victim, poor little José Cardenes. He was new this year and didn’t speak much English. He ignored Taylor just like she had. Sahar wondered if it ever occurred to anyone that it was mostly minority students that were ostracized. Not that anyone would care.

To answer that question, we turn now to the next aspect of point of view: narrator reliability.

Narrator Reliability

If the researcher chooses to write the account in the first person, through the eyes of a character, or from a limited omniscient third-person perspective, he can draw upon the element of narrator reliability. Narrator reliability refers to the degree to which the reader can trust the narrator. The author can build in signals to the reader that the narrator’s perspective may or may not be entirely trusted. The reader is presented with the problem of assessing how much information the narrator has at a given point in the story and what the interests and intentions of the narrator may be. The reader is thus led to see the narrator’s account as partial or uncertain rather than definitive. In this way, narratives lead readers to questions rather than to single answers. Varying the degree of narrator reliability has the potential to introduce counternarratives and debunk the inherent tendency toward dominating functions in literary elements when they are applied to narrative research.

In the following excerpt from Coulter’s study, David describes how he feels other students at his high school “made themselves victims of racism.” Notice the detachment created by placing several years between the narrated event and the occasion of narration:

IV.

Some people will tell you exaggerated stories. Stories to show how racist Northwest was. To me, you make your own fate. The kids that got into trouble got themselves into trouble. Like Eddie. I mean, they thought Eddie was in a gang, but Eddie really was. And he made the choice to kill a guy. He made the choice to go to jail and sit there the rest of his life. Sure, I know there is racism out there. But people make themselves victims of racism. When people stereotyped me, I manipulated it to my own benefit. If other people are stereotyped and then they become what people think they are, really become it, that’s the choice they make. It’s not the racism that causes it. It’s a choice they make themselves. The kids that got into trouble caused it themselves, and they just want to blame racism for it.

After writing a draft of her findings, Coulter presented relevant sections to each person she had rendered as a character and requested suggestions for revisions (all of which were honored). David was aware of the content and interpretation and representation that Coulter had made and was comfortable with the portrayal. He stood firm in his point of view and expressed that Coulter’s portrayal, given his revisions, was accurate. Researchers of various qualitative studies would no doubt point out the complexities and inconsistencies in the varying accounts of David’s story. In this instance the researcher made no attempt to stamp a single definitive meaning on David’s experience, instead employing the literary technique of unreliable narrator to show the inherent complexities in the situation. The ethics of manipulating reliability in participants’ stories requires a deeper discussion than is possible here.6

Other aspects of narrator reliability warrant discussion, such as how a researcher as narrator can establish trustworthiness. Coulter varies person, omniscience, and reliability, but the narratives of Wolff and Wolcott maintain the first-person point of view throughout. Both Wolff and Wolcott function as narrators. When other characters speak, the reader knows that their words are channeled through the narrator from a distance of some years. The reader allows for ordinary lapses in human memory and understands that the characters’ words at the time of the event may not have been rendered exactly or completely. Nevertheless, the reconstructions are plausible, so the reader regards the author as narrator as trustworthy.

However, Wolff and Wolcott establish trustworthiness in quite different ways. Wolcott draws on three sources. The first is his status and authority as a professional anthropologist. The second is his device of interrupting the narrative to introduce discursive arguments from the literature; such introjections are meant to help readers make the meanings that the narrator wants them to make. The third is the research methods he used for the first part of the narrative to study Brad, the “sneaky kid.” He conducted and tape-recorded interviews and took notes from his extensive number of observations of Brad’s activities over a 2-year period. The authoritativeness of his data generation, analysis, and reporting was reinforced by the publication of his paper in a refereed work. By the standards of professional anthropology, and even those of journalists and historians, such methods lend trustworthiness to a narrative. In contrast, Wolcott’s description of events subsequent to the publication of his study was not based on research in the traditional anthropological manner. In those sections of the book, he was a full participant in the events and recounted them as life story. His narrator is therefore less trustworthy in those sections; this effect is purposeful on his part. Because these narrated events sound less scientific and less distanced, the narrative becomes more personal and thus evokes the readers’ emotions to a greater extent.

Wolff’s trustworthiness emerges from a different source: his harsh and unrelenting portrayal of the foibles of his younger self. One expects self-presentation bias to be positive. But not so with Wolff. He describes Jack’s exploits: stealing money, stealing cars, starting fights, cheating, destroying property, torturing animals, and lying incessantly—the lies meant not only to avoid punishment but to preserve a good image of himself in others’ eyes and his own. Wolff writes this story without angst or justification, in cool, detached, detailed prose. In this example, Jack and his friends have been acting up in the high school lavatory, and Jack scratches “Fuck You” on the wall:

V.

During the first period after lunch the vice-principal visited every classroom and demanded the names of those responsible for the obscenity that had been written in the boys’ lavatory. . . . Well, he was going to get them if he had to keep every single one of us here all night long. . . . The vice-principal was new and hard-nosed; he meant what he said. I knew he wouldn’t let this drop, and he would keep at it until he caught me. I got scared. Even more than his anger, his righteousness scared me to the point where my stomach cramped up. As the afternoon went on the cramp got worse and I had to go to the nurse’s office. That was where the vice-principal finally came for me. . . . He kicked the cot where I lay doubled up and sweating. “Get up,” he said. . . . I sat up part-way, still miming incomprehension.

Jack describes more of his high school experience:

I brought home good grades at first. . . . They were a fraud. I copied other kids’ homework on the bus. . . . The report cards were made out, incredibly enough, in pencil, and I owned some pencils myself.

To accomplish that, Jack steals school stationary, transcript forms, and envelopes and forges letters of support from counselors and community members. In this spirit,

Now the words came as easily as if someone were breathing them into my ear. I felt full of things that had to be said, full of stifled truth. It was the truth known only to me, but I believed in it more than I believed the facts arrayed against it. . . . These were ideas about myself that I had held for dear life.

Bowden’s Down by the River contrasts with the other three narratives in point of view. The author switches back and forth between first and third person, sometimes putting words and thoughts in the mouths and minds of characters; he changes narrator distance from near to far, formal to informal. Here is an example of projecting into the thoughts of Patricia, Bruno’s fiancée, on the evening of his murder:

VI.

Where is he? . . . [A]s the engine idles, Patricia can feel her irritation rising. . . . She makes an effort for him. And she is tired of being late because of him. . . . Bruno can’t seem to ever be on time.

Still later in the narrative, Bowden uses the second person to describe the life of the boy who was tried for Bruno’s murder:

VII.

You are a natural entrepreneur so you leave school to get a part of the action. You realize there is a greater market for your products abroad than at home so you leave your nation and your culture and become a pioneer of free trade and open markets. . . . You leap out from the curbside at intersections and busily begin washing the windshields of motorists. You allow the customers to set their own fees. . . . Your entertainment venture requires more preparation. You apply makeup to your face until you have the classic visage of a clown. Then when the light turns red you leap out and begin to juggle.

A prevailing theme throughout Bowden’s book is that all sources of information contain distortions because of underlying individual and institutional interests. At best, the sources should be considered part of the truth: “Facts crumble at a touch” (Bowden, 2004, p. 160). Here, Bowden addresses this theme and demonstrates his stance toward the material:

IX.

I am in a black hole. . . . Here is the problem: once you enter this black hole and truly live in it and taste it, then you understand. . . . And this understanding doesn’t matter at all. It becomes a curse and the curse never lifts.

Narrative Voice

As research studies are transposed into narrative texts, the researcher can choose to vary the point of view from section to section, portraying multiple voices. Bakhtin (1981) calls this approach heteroglossic. The term suggests that different characters view the action differently or see different parts of the action portrayed. Such an approach suggests greater complexity of interpretation: A polyvocal account—one in which several characters narrate their stories—varies the voice from section to section or chapter to chapter. Narratives and counternarratives are constructed from the data and analyses such that the reader is free to interpret the construction as rendered. In fiction, the writer has no obligation to characters to present them and their role in the story with a sense of fairness or truth. The researcher, however, has multiple interests and obligations.

One of the ways that narrative researchers have tried to remove themselves from the omniscient point of view is through the use of multiple narrators and perspectives. Coulter (2003) decided that the eight participants in her study should narrate their own stories in first person from their own perspectives. She intended that the sum of the separate stories would represent the school as a whole, creating a polyvocal, layered account. Choosing the first-person point of view also gave Darek the agency to tell this story from his point of view. After writing Darek’s story, Coulter confirmed with him that her representation of his story was consistent with his view at the time. All of the people behind the characters had a chance to review and alter their stories.

Narrative researchers approximate a heteroglossic account in a number of ways. One way is the form known as pastiche (Ely, 2007), which allows participants to express their points of view in different forms, for example, in poems, songs, or the like, or in separate chapters or sections allotted to them. A second way is dialogue, a necessary part of all narrative texts. Through dialogue, the author presents different points of view as expressed by different participants. Dialogue also presents opportunities to show how various actors have made meaning from confrontations with others (Becker, 2007). A third way is through the actions of participants; for example, Darek’s friend Chansy laughs at the security guard and Darek notes Chansy’s laughter. Darek and the reader can interpret Chansy as mocking, as an agent in defying authority. This approach makes possible additional layers of interpretation on the part of both participants and readers.

Where does the writer place herself in the complex network of readers and participants in a narrative? A person transforms herself to researcher, then to writer. By using narrative rather than a conventional research report, the researcher can become a character in the story. This provides many options for problematizing the researcher’s telling. For example, the researcher can take on two perspectives, that of the writer and that of a character within the construction. When these two perspectives oppose each other, there is room for readers to come to their own interpretations.

Authorial Distance

Authorial distance refers to the degree of immediacy between the events portrayed and the time and place of the writing. Authors choose less narrative distance when they want the reader to experience the story almost as if he were living it along with the narrator and perhaps identify with one or more characters. The writer can choose temporal distance (as in presenting events as past or even future) or spatial distance (how far from events the writer is “standing” as she tells the story). These kinds of decisions can suggest affiliations and relationships among author, narrator, and characters. Concrete detail, scene, and interior monologues by characters can be used to achieve a feeling of closeness. Abstract nouns, summaries, and apparent objectivity can make the reader feel detached from a character or event (Burroway, 2003). Both effects can be achieved simultaneously by using a combination of techniques. For example, Wolff speaks coolly and from a distance in his generalizations between scenes. But the scenes themselves, particularly those with dialogue, make the reader feel as if he hears and sees what the characters are doing and perceiving.

Because of the conventions of education research, researchers tend to write as if they were viewing events from a great distance. For example, they use interview segments of disembodied voices, out of context and depersonalized. In the past, researchers often used passive third person, as if to convey to the reader a sense that no human being had a hand in the study. Choosing a distant position suggests an intention to portray oneself as objective, rational, disinterested, and thus scientific. But this is only one of several choices that researchers can make. Choosing less authorial distance brings readers directly to the scene, as they “hear the characters’ heavy breathing and smell their emotional anguish. Readers feel so close to the characters that, for those magical moments, they become those characters” (Burroway, 2003, p. 289).

Heavy breathing is evident in this excerpt from Wolff’s (1989) narrative. In advance of his mother’s marriage, Jack was sent ahead to live with Dwight. Dwight was attentive, charming, and jovial toward him while wooing Jack’s mother. But there is an undercurrent that Jack alone detects:

X.

Dwight made a study of me. He thought about me during the day while he grunted over the engines of trucks and generators, and in the evening while he watched me eat, and late at night while he sat heavy-lidded at the kitchen table with a pint of Old Crow and a package of Camels to support him in his deliberations. He shared his findings as they came to him. The trouble with me was, I thought I was smarter than everybody else. The trouble with me was, I thought other people couldn’t tell what I was thinking. The trouble with me was, I didn’t think. . . . The trouble with me was that I had too much free time. Dwight fixed that.

Authorial distance can be used in narrative research to accomplish various effects. Like other literary elements, it should be used with purpose.

Tone

Narrative researchers use tone to “match, emphasize, alter or contradict the meaning of words. . . . As author, you manipulate intensity and value in your choice of language, sometimes matching meaning, sometimes contradicting, sometimes overstating, sometimes understating, to indicate your attitude to the reader” (Burroway, 2003, pp. 292–293). In the following excerpt, Vicki uses ironic humor to describe the location of the ESL classes—in the darkest corners of the school—and recognizes the significance of that placement:

XI.

So this is me my first day of school. I’m carrying my schedule, looking for room G-13, thinking, where the hell is this place? I’m walking down H, walking up and down Upper G, not finding it, and then I notice the stairs on the far corner. I can’t help it. I start to laugh. I mean could they have put ESL any further away from the office? So then I go down the stairs and I still can’t find it. So I finally ask a security guard, and he’s like, “G-13? That’s the mechanic shop!” And he points to another set of stairs, and a door, about midway down and on the left. I laugh again. You know they put ESL in the dreariest, darkest place they could find, hiding them like a pair of dirty socks. They say, “That’s your classroom. Walk past the transmission, then turn left at the axle. DON’T TOUCH ANYTHING!” They put them where the white kids go to trade school, you know? And I’m thinking, great, can their place at Northwest be made any clearer?

Richardson (1985) uses tone to offset her choice of the omniscient third person in her writing of The New Other Woman:

Tone is revealed in many different ways, such as choice of metaphors, organization of material, how a quotation of a person’s experience is framed and treated by the narrator, what and how much the narrator lets who say and so on. Tone, consequently, becomes a way for the omniscient writers of qualitative research to accomplish two interlocking tasks: reduction of their authority over writing for others, and amplification of their credibility as writers of interpretive social science.

Metaphor, Figurative Language, and Theme

Use of a unifying metaphor and figurative language increases the chance of provoking intellectual and emotional responses on the part of readers. In Vicki’s account just presented, a metaphor comes out of her mouth and Coulter needs only to select it as part of the account. “DON’T TOUCH ANYTHING” is more than just a statement about the equipment in the shop. It points to the larger meaning that these students may not intrude on the precious culture of Northwest High. Bowden uses figurative language to bring the reader into the complexities and ambiguity of the story:

XII.

Listen . . . and you can hear them whispering a fable. Look and you can glimpse the shadows darting down the midnight streets. . . . In the summer of 2002, the new reform government of Mexico opened the secret files, accumulated by the state between 1952 and 1985. . . . The secret files for this period number eighty million. They hold the screams of interrogations, the taps on phones, the details of executions and hidden graves. . . . People line up in the streets of Mexico City to glimpse these secret files. They are searching for their own lives and the lives of husbands, wives, sons, daughters, and lovers who have vanished as if they had never existed. Of course there are limits to what people can find out. . . . And then there is an additional barrier, the millions of files accumulated by the U.S. government on the killings and tortures, files that remain sealed. . . . files that chronicle the intimate connection between the United States government and the slaughter. . . . Grab a shovel, the work begins. Come, down by the river, the dead speak.

Constructing a narrative almost always involves invoking some theme or moral. Themes that emerge from data can be alluded to—never named explicitly or asserted directly. The following excerpt from Darek’s story introduces an allusive theme:

XIII.

We were crossing the parking lot to the gym when Jason drove up with his F 250 full of green and gold teammates. We had to step back as he rolled his window down. . . .

“Get in, Stephens!” he said. I couldn’t move. Jason and Dirk got out and walked over to me. They made a big show of shoving me into the back of the truck, with Dirk pulling and Jason pushing. I let myself be coerced like it wasn’t my own decision, holding my ball with both hands. . . . Thai Boy just stood there and watched, his mouth open again. I threw the basketball to him and shrugged as we drove away. He caught the ball, hitched it on his hip and watched us go. . . .

. . . “That’s it, Dirk! He needs a jacket!” Jason said. He stood up and showed me his jacket. He had a big grin on his face. “Put it on, Darek.”

“It won’t fit,” I said. I was the biggest guy on the team.

“Let’s check the tag, see what size it is,” Jason said. He felt around for the tag, found it and showed it to me. It was a handmade tag that said, “Welcome to the team, Darek!” in bubble letters. It looked like a girl’s handwriting. . . .

. . . I drove in the cab on the way back, watching the clock. I couldn’t afford another tardy. Jason noticed me looking at the clock.

“Don’t worry about it Darek. Like I said, you’re one of us now. Just wear your jacket to class.”

“Yeah, it’s a magic jacket,” said Dirk, waving his hands in the air. “You magically walk in on time even if you’re late.”

“But you gotta walk right, Darek. Walk in like you own the place.”

Fiction writers consider it a mistake to let a theme determine other elements of the writing, an approach that often results in a text that is two-dimensional and predictable. Instead, the writing itself should suggest themes to the reader. This poses a dilemma for the narrative researcher who must address two communities at once (Becker, 2007; Richardson, 1990). The scholarly community expects the report to include discursive arguments that yield a result—a truth claim. In committing to a narrative rendering, the researcher makes arguments implicit. We believe that unifying themes should emerge through the researcher’s data analysis and through the writing process. Using metaphors is a means to this end. And because the narrative also serves a research purpose, the author carefully ties symbolic language to the concrete details that make up everyday life in the setting of action.

Placing Action Within Scenes

Essential to story construction is to portray action performed by actors within scenes. Storytelling requires the researcher to describe the particulars of setting, actors, and action (Connelly & Clandinin, 1990). In the details a text becomes compelling and provides a vicarious experience for the reader. Scenes require rich detail; general exposition or transitions span the time between scenes. Setting is a vital aspect of the construction because it helps create a close, believable telling. If the researcher was a participant in the events, he can rely on his recollection of his experiences (acknowledging that his own memories are imperfect). If the researcher was not himself present as the events unfolded, he faces difficulties in contextualizing and must compensate by relying on multiple data sources. The researcher can retrospectively visit the place where events took place or are taking place. In addition, he can include interview questions designed to evoke a sense of place from the perspective of the participants. The methods of interviewing proposed by Spradley (1979), for example, encourage participants to respond at such a high level of detail that they enable the researcher to “witness” events secondhand and thus generate data from which scenes can be reconstructed with verisimilitude.

Describing the setting is only the beginning. Actions must likewise be described at a fine level of detail and preserve the sense of sequence and social interaction. The scenes we have presented so far reflect both actions within scenes and character inventories, as participants or characters throw basketballs, dip French fries in ketchup, drink Old Crow, and smoke Camels. The description of significant details allows readers to imagine themselves into those scenes, which then has the effect of creating empathy and resonance.

Time

Without representing the passage of time, an author has no narrative. The narrative starts at one point and ends at another, but the arrangement of scenes need not follow chronology. The author can use foreshadowing techniques, flashbacks, and flash-forwards to elicit various effects, for example, to keep the reader interested and expectant. What does this mean for the researcher? First, her findings have to reflect some kind of arc on which participants move through time. She must identify key participants and settings that can become focal points for carrying the action forward. She must look for trajectories, plot points, and hinges that empirically and logically reflect the lived experiences of those she has studied. She must be able to recognize the intentions and goals of key individuals. So many decisions follow: Where does the story begin? How much time elapses between the beginning and the end? How many scenes should be portrayed? How much general description and exposition should be included at the beginning and end and between scenes for ideal rhythm and pace? The arrangement of scenes has an effect on the reader’s sense of causality, contingency, and explanation (Becker, 2007). Readers orient to earlier passages as probable explanations for later events; thus researchers need to establish continuity, contingency, and simple causality from their data before they make sequencing decisions in developing their narratives. All these decisions and more are needed to structure the report. The endings need not conform to conventions of happy or sad endings or even to the convention of tying up all the loose ends.

Coulter’s narrative begins with her first day at Northwest High School, describing the way the school looks to her as she drives toward it. She engages in an internal monologue that establishes some of the significant events that led her here. The second scene takes place in the principal’s office, where both she and the ESL program are placed in a diminished role in the school’s culture. From there, the narrative proceeds generally toward the end of her tenure at the school. However, the scenes in between are presented from the characters’ points of view. A key event in most stories is the death of their friend Chansy. Each scene has its own arc and sense of time. The overall structure of the narrative is novelistic, with the methodology and review of literature consigned to appendixes.

The structures of Wolff’s and Wolcott’s narratives follow a fairly straightforward path, with the sequence of scenes reflecting the chronology of events in that part of their lives. Wolff begins at the point when he and his mother broke away from their former circumstances. At that point he had intended to become a different person and signaled that by renaming himself. The starting point Wolff has chosen implies that his purpose is not to present his whole life story but to encompass a period in his life when his sense of self was most at risk and might have gone in a very different direction. The story ends with his enlistment in the army. He dispenses with his college years in a few pages and his enlistment in a single line of text. His tone then changes to one of brief and simple exposition. However, this alteration in pace and tone still leaves room for multiple interpretations on the readers’ part.

Wolcott divides his narrative into segments. The first begins with his discovery of Brad on his property. He then considers his decision to conduct an ethnographic study with Brad as his subject. He describes his research methods and then presents segments of data organized in themes, such as “getting paid for dropping out” and “I’m not going to get caught.”

After the preamble quoted earlier, in Excerpt XIII, Bowden presents another beginning, and then another, before he tells the story of the unsolved murder and all its connections to the drug war. He also fails to end the narrative in a way that we might expect, thus showing that there is no real ending and no final understanding of the events he has narrated.

Whatever the researcher’s approach to structuring time in a narrative construction, the structure serves as a cohesive element throughout the report, tying together elements and perspectives and providing temporal context for the audience.

The literary elements described in this section interact with each other in complex ways, posing countless combinations of decisions for the researcher regarding how to manipulate the text. The aim is to achieve various responses in readers without changing the underlying story,7 which itself is the result of evidence and analysis.

Narrative and Qualitative Research: Evidence and Analysis

Unlike those who draw sharp boundaries around narrative research, differentiating narrative research from qualitative research, we argue for continuities. We see commonalities in how evidence is generated—through interviews with participants, detailed observation of the participants’ social worlds, searches of archives—by researchers immersing themselves over the long term in those social worlds. All this research is done in the service of revealing as many aspects of those worlds as possible, from as many perspectives as possible, recognizing that any one perspective will be inadequate. Narrative researchers and qualitative researchers share an orientation to data analysis as a complex transaction between researcher and evidence, the ends of which are provisional and fallible. Both probe the themes, hypotheses, categories, and assertions that emerge from analysis to see how they stand up to the weight of evidence and counterclaims. Both judge their work on characteristics of verisimilitude, fidelity, coherence, plausibility, usefulness, and evidentiary warrant. They seek to test their accounts with participants and peers.

At the moment of preparing representations of their work, however, the paths of narrative and qualitative researchers are likely to diverge. Narrative researchers use literary devices such as we have described to develop time- and process-oriented accounts. They are less likely than qualitative researchers to use discursive logic to frame their descriptions of the social worlds of their participants. They do not usually state explicitly the themes they have discovered, or reveal the methods they have used, preferring to let the narratives stand on their own literary merits. Narrative researchers allow multiple interpretations to emerge, yet hope to persuade readers by their artistic visions (Eisner, 1981, 1998). In contrast, qualitative researchers are more likely to mingle discursive arguments with descriptions, case studies, and vignettes.

We also see commonalities in the ways that qualitative and narrative researchers analyze evidence and choose what parts of the evidence will go into their representations. To support our view, we first describe in cryptic terms how an interpretivist qualitative researcher goes about the process. For Erickson (1986), a good interpretivist researcher analyzes evidence as follows: He reads and rereads the body of evidence as a whole; generates preliminary assertions (specific statements of what he believes to be true) by inductive means; warrants each assertion by first assembling all segments of data that confirm the assertion, assembling all the data segments that seem to disconfirm the assertion; examines extreme cases for how they shed light on patterns; weighs the evidence, discarding or redefining assertions that do not stand up to the warranting process; looks for an organization system that links assertions to one another (e.g., hierarchies or processes); for each surviving assertion, constructs a vignette, something like a short story with actors, settings, and an arc that demonstrates the truth of the assertion in narrative form; and frames assertions and vignettes in interpretive commentary, with general and particular data.

Erickson recognizes that the generating and warranting of assertions does not follow a formula or rule. Rather, he promotes the idea of bounded rationality: The researcher makes decisions and choices about the course of data generation and analysis with the best thinking at his disposal, critiquing these decisions, examining their consequences, and making the process fully transparent to readers. Similarly, in choosing what evidence to draw on in writing vignettes, the interpretivist researcher has no algorithm to follow. In practice, the event that is transposed into a vignette is often representative of a larger set of confirming data instances. It is representative not in the statistical sense but in being typical, and perhaps a vivid example, of the larger set of events that warrant the assertion. Researchers in this model (Graue & Walsh, 1998) discuss the practice of writing a vignette as a composite of material from more than a single observed event, or with composite characters; they also reveal the crafting decisions in what Erickson (1986) termed “the life history of the study.”8

Compare Erickson’s model of analytic induction with Polkinghorne’s method of narrative explanation.9 Again in cryptic form, the narrative researcher begins by developing a story from the database and configuring it into an emplotted account (Polkinghorne, 1995, 2007). The researcher tests the evolving story with the database. When the data conflict with the evolving story, or when there are contradictions, the researcher changes the configuration of the story. Recursive movements are made between the evolving story and the database until the plot begins to take form. As this happens, the researcher is able to decide what events and data elements need to be included in the final account. The evolving story embodies the data and corresponding analysis.

Clearly, there are many continuities and areas of convergence among narrative research and qualitative research studies.

Narrative and Fiction

Education research generalists worry about whether the word narrative implies fiction. They argue that readers of a text called research have legitimate demands for texts that are factual and, by ordinary standards, true. If an author or researcher departs from the literal facts as known, readers rightfully expect some declaration of how she went about such departures and where those appear in the text. Such declarations make it possible for the reader to decide how to come to grips with the text and find it credible. In postmodern novels, however, such declarations are omitted. The author uses literary devices as cues that something nonliteral is going on. In O’Brien’s (1999) novel Going After Cacciato, the narrative begins with realist trappings. Suddenly the reader becomes aware that the author is playing tricks with time, place, and causality, creating a realm of magical realism. As we described earlier, Bowden varies his approach from section to section, veering from realism to dream states to something like poetry.

Although narrative research and fiction may share some literary techniques, narrative research accounts are renderings of the results of research (data collected in multiple forms, analyzed, and reported) in which the researcher balances the interests of storytelling against the inclusion of every literal detail uncovered in her research. For example, she renders two research participants as a single character or reworks two equivalent incidents into a single scene. Although such constructions are sometimes called fictionalizing, we believe such a label furthers the confusion; we prefer the terms reworking, rendering, or crafting. We argue that the purposes of research are not antithetical to the purposes of narrative, which include keeping the reader reading to the last page, and that the use of literary elements helps the process. Skeptics of crafting must consider how even strong and formulaic quantitative studies exhibit additions and subtractions and seldom tell the whole story: Interventions are rarely described in sufficient detail for replication; few sampling designs are executed as originally intended; the personal stake and ideology of researchers are rarely noted. Researchers transform their materials according to conventions at every stage of the research process (Becker, 2007), corresponding to the purposes of their studies. Moreover, the reports of conventional research—much as researchers in that tradition would like to persuade readers otherwise—are themselves narratives (Barone, 2001c; Richardson, 1990).

Geertz (1973) summed up this view: “Anthropological writings are themselves interpretations, and . . . they are, thus, fictions; fictions, in the sense that they are ‘something made,’ ‘something fashioned’ . . . not that they are false, unfactual, or merely ‘as if’ thought experiments” (p. 15).

Truth and Narrative Research

Most of the literature on narrative research in education, which is sampled in our first section, is focused on drawing a sharp distinction between narrative research and other forms of social science, both quantitative and qualitative.10 This literature focuses on positivism as the canonical difference between narrative research and methodologies in similar terrain. Positivism assumes that a real world exists separate from perceptions and interpretations of it. Positivism also assumes that a single truth can be found about the social world and that science demands the pursuit of this truth through objective methods. In contrast, the dominant stream of narrative research theory in education denies the existence of a single, external reality and declares that the establishment of objective methods and single, definitive knowledge claims is therefore impossible. Instead, narrative researchers emphasize constructing accounts of participants’ subjective experience and pursue particular rather than general understandings of phenomena. To understand participants’ subjective experience, researchers cannot rely on rule- and technology-governed procedures and methods. Knowledge is constructed through transactions among researchers, participants, evidence, and the social context.

The emphasis on this canonical division of positivist-informed traditional research versus narrative research has consequences. The first is that advocates for narrative research tend to downplay and even decry11 projects that search for factual knowledge and understanding of the social world. Epistemologies of social and historical realism, to name two alternative theories of research, assume that a real world exists but deny the ability of human researchers to capture a singular truth about it. They deny the possibility that any single method or perspective might capture the full complexity of human and social life, and they include as necessary the subjective experiences of participants’ lives. They treat participants’ actions as the result of their definitions of the situation, their motives, and their intentions. In these epistemologies, the orientation to knowledge claims is that such claims are flawed, provisional, and partial, although they do, at least in part, reflect some aspect of the social world before them, much in the way that maps model territory, despite inadequacies in technical methods of measurement and background knowledge.12 Except for this last feature, the assumptions of these epistemologies are not unlike those embraced by narrative research theories in education.13

A further consequence is to mark off as legitimate the work of author–researchers such as Wolcott (2002), Rose (1995), and Kozol (1991), who intend to write the warranted but still imperfect truth about their research topics, using narrative forms.

The third consequence of canonically divided literature is a downplaying of what a narrative is. Current publications in narrative research devote only a few pages to defining narrative in any way familiar to narrative theory in literary criticism (see, e.g., Bal, 1997). Finally, the postmodern denial of reality, truth, and truth-seeking abdicates the responsibility of researchers to act as witnesses to the conditions of everyday life for people who are silenced or abused. Susan Sontag (2003) wrote, “The reports of the death of reality—like the death of reason . . . the death of serious literature,” when considered in light of the effects of wars and atrocities on individual lives, “[suggest], perversely, unseriously, that there is no real suffering in the world” (p. 110). In a world that punishes (writ small) English Language Learners with anomie and marginality and (writ large) others with atrocities, disappearances, and poverty, there is a need for witnesses, however imperfect.