Abstract
Using examples from qualitative health research and from my childhood experience of reading a poem about a boy devoured by a lion (Belloc, 1907), I expand on a framework for reflexivity developed in Bischoping and Gazso (2016). This framework is unique in first synthesizing works from multidisciplinary narrative analysis research in order to arrive at common criteria for a “good” story: reportability, liveability, coherence, and fidelity. Next, each of these criteria is used to generate questions that can prompt reflexivity among qualitative researchers, regardless of whether they use narrative data or other narrative analysis strategies. These questions pertain to a broad span of issues, including appropriation, censorship, and the power to represent, using discomfort to guide insight, addressing vicarious traumatization, accommodating diverse participant populations, decolonizing ontology, and incorporating power and the social into analyses overly focused on individual meaning-making. Finally, I reflect on the affinities between narrative – in its imaginatively constructed, expressive, and open-ended qualities – and the reflexive impulse.
As qualitative researchers, we tend to eschew positivist claims to objectivity, and instead understand ourselves to be entwined with the research questions that we pose, to infuse the data that we collect and to inform both the analyses that we conduct and the forms in which we articulate them (e.g., Denzin, 1994; Mauthner & Doucet, 2003; Stanley & Wise, 1983, 1993). We aspire to be reflexive about how our selves imbue our research processes. Yet, in the documenting of the details, uncertainties set in. Old habits of objective science may dog us, we may not know where to begin, and what we end up with may simply be a laundry list of facts about ourselves. I am white, a woman, the child of immigrants, a settler, living in a Western country, 54 today, and sometimes of uneasy mind…
Without any ado behind the scenes, the three of us addressing the recent Qualitative Health Research Conference each spoke to the question of how to better be reflexive. Johnny Saldana (2018) answered in terms of how the values we share as qualitative researchers permeate our work. Elizabeth Hordge-Freeman (2018) replied in terms of how embodied emotions, ever at the heart of our methods, can be brought to analytic pride of place. My response comes from a first love of my childhood: stories. Accordingly, here I will flesh out a framework for reflexivity that Amber Gazso and I developed (Bischoping & Gazso, 2016), one prompted by concepts from the field of narrative analysis.
To readers thinking, “But I’m not using narrative data”: stay with me. Although narrative analysis certainly can be about stories—about the telling of experiences over time and the meaning-making that this involves—many narrative analysis strategies are equally relevant to any process of narration, to any time one person tells something to another, any answer given in an interview, any memo, movie, posting of a meme, crafting of a memorial quilt, and so on. And, to readers thinking, “But I already have an analytic strategy and it’s not narrative analysis,” not to worry. Narrative analysis is an umbrella term for approaches from an eclectic span of disciplines. On these pages alone, you will find concepts and works from anthropology, linguistics, literature, health studies, history, Mad studies, nursing, philosophy, various stripes of psychology, social work, sociology, and more. At least from my disciplinary vantage point in sociology, narrative analytic strategies are neither especially contentious nor weighed down by dogma, so much as they are generous mixers that we can draw on selectively, lightly or heavily, to our taste.
In particular, I will address how, in order to be more capably and fully reflexive, we can be guided by what narrative analysts have learned about how people go about considering whether a story (or an interview answer, a memo, etc.) strikes them as a “good” one (Bischoping & Gazso, 2016, pp. 54–62). Each of four criteria for a “good” story will be taken up in turn and connected to key reflexive questions. Illustrations will come from my own and others’ qualitative health research and—to lighten up the complexities—by revisiting a story dear to me as a child. It is a cautionary tale of the demise of a boy named Jim (Belloc, 1907), told through the verse of Hilaire Belloc (United Kingdom, 1870–1953) and sketches of Basil T. Blackwood (United Kingdom, 1870–1917). Jim’s story splits its readers into those who consider it a lively parody of moral tales for children and those who would shudder at giving it to a child. This makes it rich ground for thinking through how who we are enters into how we read and write the stories of our data and of the social world.
Reflexive Questions About Reportability
Jim’s tale opens with the lines: There was a Boy whose name was Jim; His Friends were very good to him. They gave him Tea, and Cakes, and Jam, And slices of Delicious Ham, And Chocolate with pink inside And little Tricycles to ride. (Belloc, 1907, p. 9)
…and thus far in, none of Jim’s circumstances engrossed my childhood self. In those days, I was more enchanted by the lifeworlds of children whose circumstances differed substantially from my own, who foraged for shellfish in the Channel Islands (O’Dell, 1960), or held midnight feasts in Swiss boarding school dormitories named after Alpine flowers (Brent-Dyer, 1958). But what made Jim’s story promise to be a “good” one nonetheless was its title: “Jim, Who ran away from his Nurse, and was eaten by a Lion.” (Belloc, 1907, p. 9)
My childish deliberations about stories’ goodness employed a criterion that narrative analysts refer to as reportability, tellability, or newsworthiness (Labov, 1972; Sacks 1995): Is the story reporting something new and worthy? That reportability is not inherent in a story, so much as in the interaction between a story (or story teller) and story recipient is where reflexivity comes in. My father’s steel plant shiftwork income, and the food security and dietary standards of 1960s Canada, where I first came across Jim, made the markers of his life’s goodness unremarkably different from mine; a reader in another context might well be transfixed by the novelties of Jim’s cake-and-tricycle lifestyle. Thus, a first reflexive question that narrative analysis can prompt us to ask is: How do we and our research participants differ in our understandings of what is reportable? This question often lies near the heart of qualitative research projects using focus group, photo-voice, participatory action, and other methods that aim to be attuned to participants and their lifeworlds rather than to researchers’ expectations. For example, an education project in this vein found that although policy discourse depicts persons with disabilities as continually struggling, they themselves regard their happiness, joys, and aspirations—and not only their struggles—as reportable (Sunderland, Catalano, & Kendall, 2009).
To introduce the second reflexive question, I will draw from a research project I have underway that illustrates a challenge of using qualitative data absent of direct, personal interaction with any interview or fieldwork participant. The data come from binge-viewing Taiwanese romance series, and my analysis is informed by my standpoint as the viewer of many a North American television drama, accustomed to considering serious illness or injury to be reportable. Sickbed moments often serve as the motors of North American plots, leading characters to sudden epiphanies and urgent disclosures of the heart. In the United States–produced show Grey’s Anatomy, for instance, the grand passion of Dr. Izzy Stevens and her heart transplant patient, Denny Duquette, was memorably enacted in his sickroom and ultimately on his deathbed (Rhimes & Tinker, 2006). But when I began watching Taiwanese romances, I noticed that a different standard for the reportability of a health condition was being applied. Attention was lavished, not only on major conditions but also on the great quantity of minor burns, blisters, cuts, scrapes, and bruises that characters seemed to so unluckily sustain. In the show Miss Rose, for instance, even being served a bowl of soup or caressing the petals of a rose on the tabletop could lead to a finger being pricked or a hand being burned (Lai, Lin, Bi, & Deng, 2012a, 2012b). It turned out that these trifling injuries routinely advanced heterosexual romance plots. Thus, when Miss Rose’s leading female character, Luo Siyi, refuses to bandage the pricked finger of her ex-boyfriend, the audience learns that their romance is nowhere near being rekindled. When Luo Siyi’s ex-boyfriend and her new boss each visit her at home to treat her burned hand, the audience becomes witness to their rivalry and to her family’s reactions to the rivals (Lai et al., 2012b). If a systematic investigation does find North American-produced shows to differ from Taiwanese shows with regard to what injuries are reportable, the next reflexive question I could ask is: How do we understand differences in reportability in a responsible fashion?
By responsible, I mean, in a way that refrains from exoticising the apparent difference between ourselves and our research participants. This means setting aside my childish longing to learn about mysterious “Others” who inhabit unfamiliar landscapes such as the Channel Islands or the Alps, who live hand-to-mouth or feast extravagantly; it means engaging critically with the politics of difference. For instance, a ready explanation for the frequency of bandage scenes in the Taiwanese romances that I’ve watched is a cultural one, based in Confucian notions about maintaining proper distance between men and women. According to these notions, minor injuries might move plots along by permitting men and women to make virtuous, compassionate bodily contact. But, it would be unwise to understand the reportability of minor injuries simply in Confucian terms. Such an analysis would default to a stereotype of Chinese culture, and indeed Confucianism, as static and unchanging in its essence (Cao, 2007; Li, 2000), with the upshot of allowing Western audiences to smugly see their ways as more progressive and free. My earlier reference to Grey’s Anatomy was meant to undercut this reading by suggesting that the Taiwanese shows are close cousins to North American deathbed dramas, rather than altogether “Other”; as Zhang (1988) points out, it is worthwhile to keep an eye out for a “fusion of horizons’” (p. 131). A more responsible analysis would also have to account for how certain characters on these Taiwanese shows are having unproblematic, none-too-Confucian premarital sex (e.g., Chien, Lin, & Yu, 2011). Last, as Taiwanese television storylines can be responsive to the authoritarian censorship standards of the large mainland Chinese market (Che, 2008), a responsible analytic story will address shifting terrains of politics and economics, and not simply static notions of culture.
This example has shown a single aspect of how reportability interlocks with the power to represent and the politics of being represented (Clifford & Marcus, 1986). With power and politics in mind, the reflexive question of reportability becomes: Who has the right or entitlement to report what story to what audience? As folklorist Amy Shuman (2015) explains, issues of story ownership stretch from censorship to appropriation. For example, within health research, we see concerns about the power of pharmaceutical companies in sponsoring medical trials and censoring reports of unfavorable findings (Schafer, 2004). Accordingly, even in qualitative journals, the disclosure of funding is increasingly considered part of ethical practice. Further, governmental influence makes itself felt in certain lines of qualitative inquiry and reporting (Bell & Elliott, 2014; Grimm & Saliba, 2017; Ho, 2014; Human Rights Watch, 2017; Price, 2004). In Canada, for instance, when oral historian Alanis Obomsawin was seeking National Film Board funding for a documentary on a Mi’kmaq fishing rights dispute (Obomsawin & Thomson, 2006), the Board directed her not to interview white people (Alioff & Levine, 2016). After Obomsawin did anyway, [W]e had to go back to the program committee for finishing money, and the same member said, “Well, Alanis, I thought we told you not to interview the whites.” And then I jumped on this person. “Now I’m going to tell you how I feel,” I said. “I know you told me that, but you or nobody is going to tell me who I’m going to interview, or not interview. And if you feel that I, as a Native person, cannot interview white people, we’ll go through everything the Film Board has done with Native people, and see who interviewed them.” (Alioff & Levine, 2016)
Qualitative scholars concerned about the politics of representation have been adopting practices such as consulting the communities that they are studying to obtain feedback on draft research papers. Educational psychologist Jodi Kaufmann’s (2010) draft analysis of an interview on gender identity, conducted with Jessie, a male-to-female transsexual, offers a cautionary case in point. Kaufmann, concerned about the rigidity with which sex–gender binaries are medicalized and societally regulated, was eager to highlight the moments in Jessie’s narrative when she troubled these binaries. To Kaufmann, the possibility of fluid, queer understandings of sex and of gender was the story worth reporting. She was shocked to find that, having read the draft analysis, Jessie was reduced to tears. To her, the binaries were realities, and not constructs. Having experienced her younger self as a woman trapped in a male body, she read the story that Kaufmann wanted to report as an erasure of her journey. As anthropologist Brettell (1996) discusses in When They Read What We Write, such disagreements are complex to resolve. Perhaps at best we can struggle, as Kaufmann does, to do better in our rewrites, particularly by addressing interpretive tensions and identifying their stakes.
The points raised so far involve being reflexive about story content, about how who we are enters into what stories we deem reportable. Yet when we assess the reportability of a story, we are simultaneously assessing the process of the story’s telling. In my family, the process of recounting Jim’s story was enhanced when the storyteller used toe-to-knee tickling to punctuate Jim and the Lion’s fateful meeting: Now just imagine how it feels When first your toes and then your heels, And then by gradual degrees, Your shins and ankles, calves and knees, Are slowly eaten, bit by bit. (Belloc, 1907, p. 12)
But what if the person you’re tickling isn’t ticklish? What if, as listeners, we feel uninterested in the stories that we’re hearing or in the way these stories are being told? It’s been shown that narrators then begin to flail, even to doubt their own story’s worth (Bavelas, Coates, & Johnson, 2000). Our reflections might therefore encompass the question: How do our standards for telling a story in ways that make it reportable enter into how we listen and thus co-construct our data? An illustration comes from a photo-voice project that sociologist Marisol Clark-Ibáñez (2004) carried out with inner-city Los Angeles schoolchildren, asking them to document their everyday lives. She writes, “One of my first interviews was with Janice, who took 38 photos of her new kitten (see Figure 3)” (Clark-Ibáñez, 2004, p. 1513). The kitten in her Figure 3 is small and blurry, hard to make out against its background, pale in comparison to the Lion that consumed poor Jim. “I admit,” she continues, “I dreaded this interview. What would we discuss besides her gatito?” (Clark-Ibáñez, 2004, p. 1513). It turned out that Janice’s kitten was the spark for many a topic relevant to Clark-Ibáñez’s study: the child’s loneliness, her family’s capacity to afford the kitten, and her migration experience, which included having to give up an earlier pet. But this example is that rare, invaluable one in which a researcher, by confessing her fear of boredom, alerts us to how our data and analysis may be swayed by our reactions to research participants who have different views from us about what’s reportable or newsworthy, or different practices from us for telling stories engagingly.
Reflexive Questions About Liveability
Everyone loves a good story. But does everyone live a good story? (Schechtman, 2007, p. 155)
What makes a good story good? The reportability criterion had focused us on how engrossing or significant a story is taken to be. But in what linguist Neal Norrick calls “the dark side of tellability” (2005, p. 323), aspects of an interesting, even gripping, narration may be too disturbing, too discomposing to be voiced or listened to, characterizing a world or situation that we can’t bear to live in. Thus, a “good story” is also liveable. (There are alternative terms here too. Oral historian Alistair Thomson (1994) speaks of whether a narrative can be told with composure while social psychologist Wendy Hollway and criminologist Tony Jefferson (1997) together introduce the concept of the defended subject to characterize a narrator struggling to tell a disturbing story liveably.)
Like a story’s reportability, a story’s liveability is no objective matter; it lies in the ears of its teller and listener. The tale of Jim, for instance, would likely be unliveable to many a potential reader. Jim’s sugar-, fat-, and caffeine-laced diet of tea and jam, ham and chocolates, so delightful in 1907, might appall today’s parents. Then, too, there is the matter of the Lion images (Figure 1). Perhaps because my parents had lived through a war and thought of gallows humor as a sign of resilience, as a child, I relished these images. It was only a few years ago, after I showed the lion to a wide-eyed young friend—and then spent half a frantic hour photoshopping the lion into a cage for her, erasing its fangs, and turning it a peaceable blue—that I had any notion that the image might be unliveable to others.
Illustrations of the Lion by Basil T. Blackwood (Belloc, 1907, pp. 9, 11, 12).
That one’s standpoint can thus inform one’s take on what a story’s liveability suggests reflexive questions parallel to those posed about reportability. We can begin by asking: How is the knowledge our research produces influenced by what our research participants are able to liveably narrate? From social work comes a case study of Sinding and Aronson’s (2003) efforts to map how health policy flaws and the Canadian welfare state’s decline make having good end-of-life care a struggle. But interviewing family members and friends about their loved ones’ deaths, Christina Sinding found participants uniformly narrating stories of “good deaths.” “She didn’t suffer,” “it was how she would have wanted it,” and “they did everything they could” were examples of what Sinding interpreted as the participants’ “consoling refrains” (Sinding & Aronson, 2003, pp. 100–103). Although participants voiced glimmers of concerns, as when mentioning an oncologist who might not have done quite all he could, in the embodied experience of conducting interviews Sinding realized that probing further would risk exposing what the participants understood as their own failures, rather than as failures of the state. Because these unliveable stories proved impossible to collect, this constrained the knowledge that Sinding could produce about the flaws of end-of-life care.
Ethics principles and policies, such as Sinding and Aronson observed, direct researchers to be attuned to what participants may find unliveable to narrate. Yet we may nonetheless go badly wrong. For instance, criminologist Claudia Cojocaru (2015), who for several years had experienced being trafficked for sex, writes scathingly of how academic audiences had pressed her for the intimate minutiae of her experiences, of learning that what they sought as reportable was “the powerlessness, the shocking details, the humiliation, the horror and the sexual domination, all common themes in the survivor narratives. They wanted a Freak Show” (p. 187). Reading this, we may hope that we are doing much better than Cojocaru’s listeners. But, Mad studies scholars Costa et al. (2012) would dash this hope. Their work shows how a sympathetic elicitation of experiences can still be understood to go hand in hand on with appropriation—a reportability issue. To Costa et al., it is empathetic, justice-oriented qualitative researchers who ironically “may pose the most threat; the ones who, by their very self-reflexivity have discovered how to be really effective at stealing stories for their own academic gain” (2012, p. 91). Criticizing the appropriation of reportable psychiatric patients’ stories as “‘disability tourism’ and ‘patient porn’” (Costa et al., 2012, p. 85), these scholars have organized to alert psychiatric survivors to ask themselves: Who profits from you telling your story? What purpose does personal story sharing serve? How do large organizations use stories to make material change? Storytelling is an exercise of labour/work. Do you get paid? The internet lasts forever. Because of the technology available today, your interview or story will likely be accessible to the public for a very long time. That includes future employers and landlords (Costa et al., 2012, p. 91).
These questions direct us to contemplate how researchers and research participants’ perspectives on what is liveable may differ, not only in the short term of the fieldwork, but also long thereafter. Accordingly, reflexivity can mean asking: In what ways is the knowledge we produce imbued with our limited grasp of what’s liveable to research participants in the long term?
As the counterpart to these participant-focused questions, we may also ask: How does what is unliveable to us as researchers, whether in the short term of fieldwork interactions or the longer term of a research program, inflect our knowledge? The short-term example I will give comes from an enraging fieldwork moment that I experienced during qualitative interviews for a project with sociologist Elizabeth Quinlan on the health consequences of precarious theatre work (Bischoping & Quinlan, 2013). Conducting the job history part of the interview, I asked an actor, whose pseudonym is Charles, how he had gotten a particular job. The question was a bit foolish: both Charles and I knew full well that, based on our earlier theatre work together, it was I who had once again hired him. Charles replied, “Nine hundred ladies’ room walls can’t be wrong. ‘For a good time call—’” I took this answer, which imaginatively cast me into a women’s washroom, reading graffiti that gives Charles’s phone number and testifies to his sexiness, as unliveable and sniped back in anger. Although afterward I would gladly have swept the episode under the rug, feminist researchers such as Kleinman, Copp, and Henderson (1997) alert us to how qualitative researchers can employ their passionate, discomfiting emotions as sites of knowledge, rather than as threats to objectivity. By focusing on the unliveable moment, Amber Gazso and I (Gazso & Bischoping, 2018) began to unearth how a host of discourses that Charles and I instantiated were relevant to the stresses of cultural work. For instance, that I am a woman and much older than Charles means that he can deploy a “cougar” discourse (Montemurro & Siefken, 2014) to castigate me; that he is using humor puts me in a double bind, because if I criticize him, I can be called humorless (Quinn, 2000).
In studying traumatic experiences, such as terminal diagnoses, pregnancy loss, sexual assault, experiences of torture—or, anything unliveable to a particular researcher to contemplate—reflexivity about the long-term consequences of research work is especially important. Taking up a dementia project with sociologist Riley Chisholm (Chisholm & Bischoping, 2018), I initially felt shielded from my anxieties about the topic. After all, it was Riley, and not I, who was in the field, conducting interviews with the family members of “Alexander,” a man diagnosed with dementia. The emotional labors of the fieldwork, as discussed by Ezzy (2010) and Fitzpatrick and Olson (2015), fell entirely to her. But even working at arm’s length, reading the literature, as well as participating in data analysis and writing, left me miserable. I began to have nightmares in which I had dementia, to feel guilty because the literature plainly told me that fear-ridden views contributed to the condition’s indignities (Beard & Fox, 2008), and to feel that my misery was preposterous compared to the stresses I imagined Alexander and his family to be facing. A reflexive question this provokes is: How does regarding one’s research topic as liveable or unliveable color an analysis? Riley and I noticed in writing our joint paper that she initially tended to emphasize the family’s hopefulness as a fact, whereas I tended to treat it as tenuous accomplishment (Chisholm & Bischoping, 2018, p. 7). Where she saw a glass half full, I saw one continually being drained.
As I dragged my feet through the revisions, I asked, Is this work of knowledge production sustainable for me? The answer was a resounding “no.” Nothing I could imagine saying further about dementia would be worth the joylessness of this effort. However, exit is not always desired or possible. For researchers devoted to a topic that brings them to despair, but who are embedded in the embattled community that they are studying or who perceive an “easy exit” from the field as contrary to feminist fieldwork principles (Murphy & Dingwall, 2001), the follow-up question would be: What way of conceptualizing this experience fits it most closely and leads to the most helpful interventions? Concentrating on the psyche, the key conceptualizations from social and clinical psychology are burnout (Maslach, 1982), compassion fatigue (Figley, 1995), and vicarious traumatization (Pearlman & Saakvitne, 1995). These vary as to (1) how specifically they pertain to trauma, (2) whether they regard the person experiencing them as having a pathology, (3) whether they point to the positives of such experiences as well as the negatives, and (4) whether they recommend individual or community-based strategies (Bischoping, 2004; Maček, 2014). Even broader frameworks have emerged for examining psychic harms alongside the physical risks of dangerous field settings (e.g., Jessee, 2017).
Reflexive Questions About Coherence
As researchers or as participants talking about something unliveable, we quite often become incoherent: we can’t find words, stumble, make slips of the tongue, use double negatives, contradict ourselves, lose our place in time, present cause-and-effect chaotically, and the like. These phenomena intrigue narrative analysts because the third common answer to “What makes a good story good?” is that it be coherent, that it hangs together and is clear at the level of syllables, words, sentences, and so on, up to its overall themes and structure. A coherent story is “good” not only for listeners, it has also been shown to be good for storytellers. The capacity to tell one’s life story coherently is associated with greater well-being and social support (e.g., Baerger & McAdams, 1999; Burnell, Coleman, & Hunt, 2010), as well as the capacity to negotiate one’s group memberships (Linde, 1993).
Coherence or incoherence may seem to inhere in a story or storyteller. But, narrative analysis shows that listeners and their social milieux, once again, to be implicated in it, often without knowing. Returning to the story of Jim, if I had a quibble as a child, it was with the passage, quoted earlier, in which the Lion begins to consume Jim: Now just imagine how it feels When first your toes and then your heels And then by gradual degrees, Your shins and ankles, calves and knees, Are slowly eaten, bit by bit. (Belloc, 1907, p. 12)
Incoherence can direct us toward reflexive questions of at least four kinds. When a sensitive or painful topic is being discussed, and a narrator or a certain part of their narrative seems incoherent, we may reflexively ask: What discourses 1 condition what research participants are able to narrate coherently? How am I, as the researcher–listener, positioned in relation to these discourses? McKenzie-Mohr and Lafrance (2011) provide an apt exemplar. They discuss how Suzanne McKenzie-Mohr, a social worker, had conducted interviews with women who had been raped so as to learn how they went about living well in the aftermath. In the interviews, participants often struggled to express themselves, as in this statement by “Margaret”: “I don’t feel helpless I guess is what I am trying to say. I didn’t feel helpless over the situation. It is probably the best way to put it. I don’t feel responsible but I don’t feel helpless either” (McKenzie-Mohr & Lafrance, 2011, p. 63, italics mine).
Reflecting on this passage and others like it, McKenzie-Mohr and Lafrance interpret their participants’ struggle for coherence using sociologist Marjorie DeVault’s (1990) concept of linguistic incongruence, in which the available language doesn’t capture their experiences. Margaret understands herself to have agency and autonomy, and to not feel helpless, but a powerful, seemingly sympathetic discourse adjures us that victims are not to be blamed. Many other participants likewise could not easily find words surmount this rigid binary. Further, McKenzie-Mohr discovered that her decision to recruit participants by using the word “rape” rather than the legalistic, clinical “sexual assault” inadvertently aligned her with a dominant discourse characterizing rapists as force-wielding strangers (McKenzie-Mohr & Lafrance, 2011, pp. 54, 55). It transpired that several of her participants had long struggled to understand themselves as having been raped when their experiences did not conform to this discourse.
Second and more broadly, when researching populations whose marginalization can be exacerbated by seeming incoherence, we can ask: Do our methods incorporate the best practices for countering this marginalization? Fieldwork conducted through language barriers is a prime example. When people are required to speak in a language in which they are not fluent, their contributions may become less coherent, nuanced, and quotable than other participants’. Those who cannot speak the researcher’s language may be deemed ineligible to participate; phrasing this reflexively, we would say that they are excluded because the researcher is unable to speak their language. Further, if untrained interpreters become involved, data may become substantially transformed and even distorted. Nursing researchers Almalik, Kiger, and Tucker’s (2010) study of the use of interpretation when studying how migrant women in Scotland experienced postnatal care provides some alarming illustrations. For instance, in response to “So was the midwife telling her and teaching her how to deal with the baby?” a lay volunteer interpreted a participant to have said, “Yes, they show her how to hold the baby; she thinks that it was good” (Almalik, Kiger, & Tucker, 2010, p. 261). A professional interpreter, however, found the participant to have replied quite differently: Yes, she showed her, she showed her how to hold her but she doesn’t think it is good because in her opinion her midwife told her to hold the baby like a doll and she thinks that the baby is so weak and so small that she couldn’t hold her how she showed it to her. (Almalik et al., 2010, p. 262)
Last, although this section has been characterizing being incoherent as symptomatic of a deeper problem, and being seen as incoherent as stigmatizing, narrative analyses also offer a sense of incoherence as ordinary and natural. Literary scholar Mikhail Bakhtin (1935/1981) tells us that the genre of the novel appeals to us because it uses characters as vehicles for setting diverse perspectives into dialogue. Our psyches, too, are dialogic in nature: We ordinarily ponder what’s “on the one hand” and “on the other hand,” we speak through diverse discourses and naturally contradict ourselves, break off, and start anew in new directions, our new realizations emerging as we speak (Sools, 2012). We are prompted, therefore, to ask reflexively: How can we become attuned to the diversity within each participant’s perspectives, recognizing that our own selves are implicated in how we selectively hear? Kaufmann’s (2010) study of trans-representation, discussed earlier, reflects on her initial preference for what she took to be her participant’s most reportable voice; in Riley Chisholm and my dementia study (Chisholm & Bischoping, 2018), meanwhile, our different projections about whether the participants’ situation was liveable arguably led us to focus on the transcripts’ more hopeful or more despairing passages. An elegant data analysis method that can aid us is the I-poem (Gilligan, Spencer, Weinberg, & Bertsch, 2006), which isolates moments when participants speak of themselves, and directs us to listen specifically for such diversity.
Reflexive Questions About Fidelity
I believe, Woman, said she, thou tellest me a Story. (Richardson, 1740, p. 272)
What makes a “good story” good? Although an informal sense of story is that it is a lie, the fourth, commonly held criterion for the goodness of a story is that it be faithful to the truth, correspond to reality, convey accurate knowledge. Of course, “truth” has been notoriously difficult for philosophers to define (Tarski, 1944); “reality” is no less murky. But sociolinguists Stivers, Mondada, and Steensig (2011) help us out by pointing to how, in our day-to-day life practices, we act on the assumption that quite a lot is knowable. At the microlevel of interaction, in exchanges as brief as, “Where’s my toothbrush?” “How should I know?” we continually pay heed to who is entitled or obligated to know how much about what and with what certainty. Moreover, say Stivers, Mondada, and Steensig, we hold each other accountable to fulfill obligations and not overstep entitlements. A (morally) good narrative therefore is faithful to what is knowable and to the system by the transmission of knowledge is organized.
In keeping with this chapter’s refrain, although fidelity may seem intrinsic to a story, it can be produced by both a storyteller and listener, from how they negotiate entitlements and obligations to knowledge and from their standpoints and positionality. The resultant reflexive questions will depend on the ontology and epistemology that an inquiry involves. Broadly speaking, realist questions aim to ascertain the “actual facts” of a matter, while constructionist ones take interpretations and interpretive processes as their object of analysis. It should be noted that a research project may incorporate elements of both. A case in point is sociologist Deborah Davidson’s (2007) study of how Canadian hospitals came to change their protocols for addressing perinatal loss: She asks both realist questions about when and by whom the changes were instigated and constructionist ones about how caregivers came to new understandings of mothers’ perinatal bereavement.
When we do realist research, seeking facts and faced with data containing contradictory possibilities, common strategies are to deliberate on various data points’ relation to other known information, and on participants’ trustworthiness, competence at understanding the matter at hand, and extent of firsthand experience (Howell & Prevenier, 2001). A reflexive question to then pose is: How does your standpoint play into your assessment of a statement’s plausibility or a narrator’s trustworthiness and competence? The reason is that, when reading fiction and judging whether a narrative is realistic, we use our experiences—and our experiences with stories—as a baseline. Media researchers Busselle and Bilandzic (2008) explain how, when a narrative comfortably aligns with our expectations, we refrain from analyzing and judging; lulled, we immerse ourselves in the story world. But when an element in the story strikes us as unexpected, we are jarred out of the story world and begin criticizing the narration as unrealistic or implausible. An example comes from anthropologist and midwife Janneli Miller’s (1996) interview with a Guatemalan Maya woman who narrated an experience of having been enspelled and subsequently expelling a toad from her body, in what “was like having a baby” (Miller, 1996, p. 111). When Miller recounted this, many Anglo listeners took the story as told to be implausible and deemed it to instead be a veiled story of an abortion; only an American Indian woman who had heard of such an enspellment before took the story to be plausible and the narrator to be trustworthy and competent.
When I presented the earlier version of this article, an attendee conducting research on Indigenous knowledge posed a related question: What if I believe my participants but the audience for my research is skeptical about their knowledge? Literary theorist Gayatri Spivak (1988) famously showed exclusion from power structures to go hand in hand with being treated as lacking knowledge worth learning about. If your audience is immediately dismissive, you face exactly this problem. An option is to preface your work with a commentary on how claims about objectivity are permeated with power relations. When audience members default to skepticism or to the assumption that their way of knowing is automatically superior, they may be carrying forward a colonial legacy in which the research gaze has served the purposes of empire, devaluing and appropriating knowledge from the “Other” (Battiste, 2002; Denzin, Lincoln, & Smith, 2008; Hall, Dei, & Rosenberg, 2000; Kovach, 2010; Smith, 2013). To signal that now-independent states have been decolonized in formal terms only, and that the work of dismantling the colonial legacy remains underway (Ngũgĩ, 1986), decolonizing methodology is the name that Māori education scholar Linda Tuhiwai Smith (2013) has given this line of work.
Let me now turn to the reflexive questions arising when we do research as constructionists. In this circumstance, we usually agonize less over whether what a research participant narrates is—strictly speaking—“factual.” We know that the everyday in which stories are told is no court of law, we know that a story may be embroidered to become more reportable, have parts lopped out to become more liveable, or have unwieldy facts glossed over so as to gain coherence. We may treat a story, whether understood as fictional or “factual enough,” to be authentic in voicing a “deeper truth.” For instance, in Jim, Belloc (1907) gave me a protagonist who spoke what I took to be a deeper truth: that it was possible to run toward rather than away from danger, and hang the consequences. Although orderly and timid, I viewed the wilder boys in my class with the same fascination as I did Jim.
Constructionists can encounter certain pitfalls, some of them related to not being quite constructionist enough. Ontologically, constructionist analyses may mistakenly proceed as though research participants’ psyches and bodies contain unsocialized, unmediated “raw” experience (Scott, 1991), and as though participants’ “deeper truths” about the meanings of their experiences emerge from an individual, quest-like process that is somehow solitary, and presocial. Epistemologically, constructionist analyses may hold to the naive notion that a well-meaning interviewer can straightforwardly extract what’s in a research participant’s mind (Roulston, 2010). (Previous sections of this article, however, have emphasized myriad ways in which story listeners and tellers co-construct what is told.) Simultaneously, constructionist analyses may fail to recognize that constructions often matter because their consequences transcend the individual storyteller. From discourse analysts come the insights that language is a form of action rippling out beyond the individual storyteller (Fairclough & Wodak, 1997), and intimately linking knowledge to power (Foucault, 1978). The deeper truth I perceive in Jim’s story, for example, reproduces and celebrates widely circulating discourses about liberal individualism and a “good-girls-like-bad-boys” line of hegemonic masculinity.
The reflexive question arising for constructionists prone to these pitfalls, then, is: How would the knowledge we produce be altered by recognizing that interpreting one’s experience is not simply an individual process, but a socially informed one with social consequences? This can be a thorny question in qualitative health research. We may be reluctant to shift our emphasis away from participants’ individual interpretive agency and creativity toward how their interpretations recirculate popular notions, or to seem to undermine the heartfelt words that ill people are saying, especially if we understand those words to be bringing them hope, encouragement, and equanimity.
But would we necessarily be undermining our participants? In a provocative study of how breast cancer patients spoke of their coping strategies, psychologists Wilkinson and Kitzinger (2000) suggest not. By closely observing how their participants spoke of thinking positively, they found that it was frequently as a moral exhortation (“You’ve got to think positively” (p. 206)), often coming from others. Instead of serving these participants, thinking positively seemed to serve their interlocutors, alleviating their burden of hearing about misery, and allowing them to define what a “good cancer story” sounds like. Further, participants sometimes vehemently countered these moral injunctions, as “Pauline” did when she had just been advised by an acquaintance: “[B]ecause you’re so positive you will fight and you will conquer this.” And I was quite aggressive at this point, and I actually said (pause) “If anything happens to me,” I said, “I don’t want the likes of you turning round and saying, P’rhaps she wasn’t as positive as she made out she was.” (Wilkinson & Kitzinger, 2000, p. 808)
Last, let me return to Riley Chisholm and my dementia study (Chisholm & Bischoping, 2018), in which “Alexander’s” family members often spoke positively. The following quote from “Holly,” his daughter-in-law, encapsulates the family’s narrative: He [Alexander’s] never been a big talker and I guess you could say that’s the Scottish way about him. […] So, he doesn’t need to speak for us to know that he’s all around us and more than anybody he is this land. (Chisholm & Bischoping, 2018, p. 13) Scottishness is totally about land. They came over and they farmed, you know, and that’s how they lived was working the land. If you are Scottish and live here, […] it’s likely your father was some kind of farmer. Dad is pretty much a purebred. If you have grown up out in the County, who you are (pause)—well, it’s all about the land. I mean we’ve been farming for six generations so the land, yeah, it’s everything. (Chisholm & Bischoping, 2018, p. 14)
Conclusion
In closing, I wish to briefly reflect on what the writing of this article has shown me about the fit between narrative analysis and the reflexive impulse, a fit that first appealed to me in a more intuitive than explicit way. First, that the notion of story contains a fluid, shifting mélange of fact and fantasy, myth and dream, and truth stranger than fiction, enlivens our analytic imaginations. The notion seems suited to inviting qualitative researchers of all ontological stripes to ponder the constructed nature of knowledge. Second, that stories so often hold elements of joy, entertaining us, arousing our curiosity, giving first voice to inchoate dreams, pressing forth toward visions of justice, or—even at their direst—conveying a teller’s personality means that a narrative analysis-based reflexivity is in touch with an uplifting, expressive force. (In other words, it’s often fun.) Third, that the meanings of stories are frequently open-ended and multiple in their possibilities is consistent with how a narrative analysis-based reflexivity can pose open questions that flexibly encompass a sprawl of directions and scales. We can leap from censorship, appropriation, inclusion, trauma, and epistemology to minute stumbles or pronoun choices; we can freely join our reflexive questions to innumerable analytic strategies. Finally, a narrative analysis-based reflexivity is sensitive to the duality of our roles. We are both co-constructing listeners to our research participants’ stories and tellers, to co-constructing research audiences, of what we hope will be “good” stories of our own.
Footnotes
Acknowledgments
Amber Gazso’s presence in this text exceeds what citation can indicate. We not only developed the framework together but read and wrote together in ways that shaped the examples chosen here and the voice in which I write about them. She commented on this paper’s earliest draft; her daughter, Macartney, alerted me to the perils of the Lion. Rawan Abdelbaki contributed a literature review on translation; Victoria Palmer introduced me to the “I-poem” method; Riley Chisholm, Zhipeng Gao, and Elizabeth Quinlan, also coauthors, ongoingly prompted me to think about reflexivity in our joint projects; and Zhipeng Gao brought generosity and a probing curiosity to the final draft.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research for this article was supported by a York University Research Development Fellowship and the York University Sabbatical Leave Fund; a Social Sciences and Humanities Research Council of Canada (SSHRC) Grant in Aid of Travel and a SSHRC Conference Opportunity Grant; and a Faculty of Liberal Arts and Professional Studies, York University, International Conference Travel Grant.