2
BUILDING WORKING RELATIONSHIPS
1 introduced myself. I was writing about (English| football supporters. Did he mind if I asked him some questions?
He stared at me. Then he said. "All Americans are ssankers." And paused. "All journalists," he added, showing, perhaps, that his mind did not work along stricdy nationalist lines, "jre cunts." We had established a rapport.
Dili Uuford(l<»3.26)
Journalist Dill Buford may haw been writing tongue-in-cheek, but this brief vignette still makes an important point. An interviewee may not like, respect, or even wr Iconic the prying attention of a journalist or researcher, and the researcher, in turn, may not like the interviewee, but this does not prevent the two from developing some way to work together.
The Limits of Rapport
The usual advice in textbooks on interviewing is to build good rapport with interviewees. A quick search of the key word "rapport" in the Handbook o/htier-vicw Research (Gubrium and Holstein 2<X>lb), for example, brings up nineteen out of forty-four chapters where authors reference or assume the importance of establishing rapport with interviewees. But what exactly is rapport? Oxford-dictionaries.com defines it as "a close and harmonious relationship in which the people or groups concerned understand each others feelings or ideas and communicate well."' By that definition, what passed between Buford and the English football fan was anything but rapport. Yet. despite this. Buford still managed to spend years with a group of such fans going to restaurants, pubs, and matches throughout England and Europe. If it was not rapport that enabled
Building Working Relationships 11
Uuford to get close to a group of men whom the popular press was calling "thugs." then what was it?
I argue it was the working relationship that Uuford was able to establish with this group, despite, not because of. their divergent proclivities toward violence, mayhem, and excessive drink. A working relationship allowed the "thugs" to talk to Uuford frankly, and tor Uuford to talk back in equally blunt terms. What a good working relationship requires is that the researcher treat participants as etuis in themselves, rather than as means to an end (the "end" being the book, article, or dissertation the rcseanrhcr is writing). Adopting this type of "deontologii.il" approach to research ethics is conducive to building a working relationship because it requires the researcher to treat all participants with the same dignity and respect, regardless of what the researcher thinks of their actions or beliefs, and regardless of how forthcoming or recalcitrant the interviewee turns out to be (Murphy and Dingwall 2007).
All too often, researchers treat rapport as an end in and of itself, as if achieving rapport obviates other problems or dilemmas. Yet. rapport in the sense of "closeness" or "harmony" is not necessary or even sufficient for a working relationship to develop, as the example of Uuford indicates. The assumption that rapport is critical to interview's may be based on the belief that only interviewees who trust the researcher will talk openly.This assumption does not always hold, however. In research settings where insecurity is rife and mistrust the norm, rapport might be impossible to achieve (Uelousov cl al. 2007; Pickering 2007, 195). Even in more peaceable environments, feelings between interviewer and interviewee may not ever become warm or friendly. And friendliness itself may not be an indicator of meaningful rapport, but rather a superficial relationship or "ethnographic seduction" whereby the outward "civility" and "affability" of an interviewee actually leads researchers away from their original focus, not closer to it (Robben 1995,83).
f.ven when rapport is established, it does not guarantee that the other person will agree to be interviewed or that the interview will go in a desired direction. I once spent several hours chatting with an elderly Mack woman in her kitchen in my research site in Maryland. The methods literature might say that I was building rapport with her. I explained that my project was on the lynching that had occurred in her hometown (where she still lived). I asked if I could interview her ami handed her a copy of my written consent form, which included a description of my project and what participating in it entailed. She glanced at the document and asked if we could do the interview in a couple of weeks. I said that would be fine. She did not seem to want me to leave so I stayed and \sv continued to chat. The conversation was friendly and relaxed.
Two weeks later, I called the woman to make sure it was still okay for me to come by that day to interview her. I arrived at her home at the agreed-upon hour. Uefore starting the interview. I explained that I would not share the data with anyone or use any real names or identifiers in any published work. I also told her she
14 Chapter 2
could stop the interview at any lime or refuse to answer any question. She agreed to be interviewed, at which point I took out my notebook and pen. I began by asking when and where she was born and how far back her family went in her hometown. After these questions, she began talking about the lynching. As she spoke, I began scribbling, trying to capture what she was saying in her actual voice. I quickly turned the page to write down more when she said in a stern tone. "You've got enough now." Incredulous, I looked up at her and asked plaintively, "You want me to stop?" She replied,"Yes. you've got enough." I had only finished writing two notebook pages in very Urge and messy scrawl. The interview had stopped just as quickly as it had begun.
I did not feel a lack of rapport with this woman. In fact, my sense was just the opposite. I was very comfortable sitting and talking with her in her tidy kitchen, with the air conditioner humming in the background. She was in her early nineties and had a quick mind and sharp sense of humor. She also seemed to welcome the company. Hut when it came time for the actual interview—that is, the formal conversation that began with my asking for her consent and then taking out pen and paper to record what she said—she became reticent. Perhaps it was the topic of the lynching that made her uneasy. Maybe the whole format of the interview felt too invasive. Perhaps my furious note-taking drew more attention to her than she wanted, especially with such a difficult topic. The lynching, after all, was not a "data point" to her; it was a highly charged life experience, which she recalled with great clarity and emotion. Perhaps revisiting tliat day elicited emotions she did not want to experience or to share with me.a stranger.Stopping the interview may have been her way of keeping herself—and our interaction—within certain bounds. This experience taught me that rapport has no magical properties when it comes to ensuring the depth, breadth, or duration of an interview. Rapport may be nice to have, but it is not sufficient for purposes of relational interviewing.
In other instances, rapport may not be a realistic goal. A participant may not like the researcher, as Hill Huford's football fan made clear to him in this chapter's epigraph. Another possibility is that the researcher does not like the interviewee or may disapprove of what interviewees have done in the past or how they spend their time in the present. Kathleen Bice (1998). for example, interviewed Ku Klux Klan members, whose white supremacist beliefs were not in line with hers, but this did not prevent Blec from seeing her interviewees as fully fleshed-out individuals. As Blee (1993,600) recalls,"Oral histories of Klan women reveal that many held complicated attitudes toward gender, race, economics, and nationalism, attitudes that did not tit traditional political categories, such as reactionary or progressive." To arrive at such a nuanced understanding, Blee had to find a way to work with her intcrviewrcs—and thev with her.
Similarly, studies of the Rwandan genocide and Rwandan prisons have relied heavily on interviews that scholars conducted with self-identified £ah\ul\m>, that is, people who took part in the 1994 genocide that resulted in the slaughter of over half a million people (Fujii 2009; McDoom 2013; Mironko 2004;
Building Working Relationships IS
Straus 2006; Tcrtsakian 2008).These studies would not have been possible without interviewers and interviewees arriving at some way of working together.
What leads people to talk openly, moreover, is not always feelings ofclosencss" or "harmony." In some contexts, it might be distance and lack of trust that allow people to talk in frank terms. Italian-born historian Alcssandro Portclli (2011), for example, conducted oral histories with residents of a mining community in Harlan County, Kentucky. His interviewees included one black woman named Mrs. Cowans, who told him outright that she did not trust white people, a category that included Portclli in that context. Portclli (2011.8) reflected on what he learned from this exchange.
Fieldwork handbooks always tell you to gain the trust of the interviewee; but it was the distance and difference that made the interview meaningful, while Mrs. Cowans s boldness in speaking across the line and explaining its meaning made it possible. There were lines of age. class, gender, education, religion, language, color, and nationality between myself and most of the women and men who spoke to me in Harlan County. The mutual effort to speak across those lines taught me to think of the interview as an experiment in equality, where trust is achieved not by pretending that we are all the same but by laying the difference and the inequality on the table and making it, as Mrs. Cowans taught me, the implicit subject of the conversation.
Portclli did not assume that bridging the divides that separated him from the people of Harlan County was a prerequisite to meaningful interaction. Rather, he learned that acknowledging those divides from the start was the way to build a working relationship with participants.
The examples of Portclli and Blee point to the importance of developing a particular kind of relationship with interviewees, one that is borne out of a certain type of work interviewers and interviewees do together. The concept of a working relationship does not foreclose the possibility of establishing rapport with interviewees, but at the same time, it does not assume that rapport is necessary. A working relationship simply means that interviewer and interviewee arrive, explicitly or implicitly, at mutually agreeable terms for interacting, conversing, listening, and talking with one another.
The Promise and Pitfalls of Positionality
Huildinga working relationship starts with becoming aware of how the researcher and interviewee "see" one another—what assumptions they are making about who the other person is. what he or she is after, or what that other person might know. It also involves becoming aware of differences in power, social status, and privilege between the researcher and interviewee. What enables researchers to
16 Chapter 2
develop such awareness is reflexivity. The kind of reflexivity I have in mind is context dependent, belying the one-size-(its-all language of codified rules and regulations. It entails a "keen awareness of. and theorizing about, the role of the self in all pluses of the research process" (Schwartz-Shea 2014, 133). The ausons for such awareness are twofold. First, issues of positionality relate to the ethical dimensions of the researchers practices and strategies. Second, issues of power and privilege bear directly on the knowledge claims the researcher can advance.
Structural Power and Privilege
Not all researchers go to the field with the same resources or talents, but many enjoy a wide range of privileges while they .ire there. Privilege is the ability to choose—where to live, what to eat, how to travel, and with whom to spend time. Whenever the researcher exercises this ability to choose, he or she is exercising privilege. The most fundamental privilege that all researchers enjoy is gaining entrée into people's worlds. Choosing what to do with that privilege is an issue no researcher should take lightly. Some scholars will take the time to ponder such thorny questions as whose priorities—the researchers or the participants— should take precedence in a given moment or over time. Some will not tike such time and not doing so is itself an exercise of privilege that can have deleterious effects on the research process.
My inability to recognize my own power and privilege may explain one relationship that involved a prisoner in Rwanda whom my interpreter and I interviewed twice in 2004. To the second interview, the nun brought a small black notebook, which he wrote in throughout the session. I did not ask him what he was writing, but it appeared he was "recording" our discussion or parts of it. My fear was that he was reporting back to the other prisoners what kinds of questions I had asked. Only later did it occur to me that any of the prisoners could have done the same, whether they were writing down the questions or not.
Looking back, I recall my reaction as he began writing. "Showing" me and my interpreter that he was "recording" our interview felt antagonizing. At the time. I did not reflect much on the power differential that marked our interactions. Only through writing this book did I begin to realize that this man may have been trying to reclaim some measure of dignity and control in a very unequal situation, by turning the tables on my interpreter and me through notating what we were saying. The power differences urn' undeniable. My interpreter and I were in a more powerful position: we were the ones asking questions. We had the prison directors permission to conduct interviews there, and we could come and go as we pleased. Gender effects may have also contributed to his hostility, gisxn the fact that those in power were two women. He may have also resented the fact that I would be benefiting from our exchange, while he would get little out of it that was meaningful. For my part. I resented his obvious lack of cooperation, an indication of my expectations for how these interviews "should" go and how
Building Working Relationships 17
interviewees "should" behave during these exchanges. Because I did not consider the kind of power I exercised in these encounters. I tailed to comprehend why a man would resist the attempts of a privileged outsider to engage him. Had I been more attuned to the power dynamics in this relationship. I could have tried to find ways to interact that were more palatable to him (if any), rather than deciding not to interview him again during that trip, which is what I decided to do after that encounter.
Individual Attributes and Background
In addition to sources of privilege and power, the researcher needs to contemplate issues of positionality. Positionality takes many forms. It includes the array of personal traits, such as skin color, accent, age. and sexual orientation, that signify "who" or "what" a researcher is in the local context. Which combination of traits matters depends on what people in the research site find salient when identifying others. For a project involving interviews with immigrants, for instance, the most salient aspects of the researcher s positionality might be native language abilities, accent.and the researchers own immigrant status (Berger 2015;(»awlewicz 2016). For projects that take place in male-dominated or masculinized settings, by contrast, the most important aspects may be gender, race, and profession (Cohn 2006; Jentsch 1998).
Because positionality depends on how people in the research context view the researcher, there are no traits that are predictably moa* important or more salient than others. Which characteristics matter will vary not only across different sues, but also across different researchers working in the same site, even those with similar backgrounds.The case of two graduate students doing interviews in Indonesia illustrates this point.Jessica is a young-looking, Indonesian-Canadian woman who speaks Bahasa Indonesia, the country's official language, with a North American accent. Aarie is a tall, white. Canadian man similar in age to Jessica but he does not speak Bahasa. For Jessica, gender norms dictate that she be accompanied by another person during most meetings and interviews, despite her fluency in the local language. For Aarie. no such norms apply and because he is conducting interviews with English-speaking officials, he can meet participants on his own, n ithout the extra step of arranging for someone to accompany him.
The disparity in how local people "read" Jessica and Aarie translated into very different treatment. During his time in the field. Aarie noticed that local people often smiled and said hello to him, even, at times, giving him a mock salute as if to acknowledge his high social "rank." As one local explained to him. people behaved that way because they were "in awe" of a tall, white, presumably wealthy foreigner. Jessica received no such "awe-filled" treatment, despite coming from the same university and country and possessing the same credentials as Aarie.*
As the example of Jessica and Aarie underscores, people in the research site will use categories and labels that have meaning to them when they arc deciding who
18 Chapter 2
or "wliat" the researcher is. These categories and labels might bear no relation to the ways that the researchers themselves self-identity. In Rwanda, for example, local people often assumed that one oř my parents must be Rwandan, even though I haw no family connections to die country. In one of my sites, a rumor even circulated that I was the long-lost daughter of a local woman who had had a baby with a muzunou (European or white foreigner) man.This rumor helped to explain why I had chosen to do research in that particular hill rather than any of the other thousand in the country (Fujii 2009.2014).J
Social and Theoretical Vantage Points
In addition to personal features of the researcher, positionality can also be a matter of the physical location or vantage point that the researcher adopts. Researchers often make careful choices about where to "situate" themselves within the site, whether a firm, an organization, or a community, because different locations afford different mow of power relations, work habits, cultures, and norms. Lorraine bayard de Volo (2009. 220-1). for example, worked as a cocktail waitress in three casinos in Reno. Nevada to uncover how her fellow servers constructed the nature of their work. As a waitress. Bayard dc Volo was able to glean a very different view of the casino floor than had she worked as a pit boss or manager. Similarly, Timothy I'achirat (2009, 145) worked on the "kill floor" of a slaughterhouse in the United States in order to see and experience "everyday, hidden and violent labor" (emphasis in the original) at the level of those who engage in such work. Had he taken a job in the management offices upstairs, his view of power relations within the slaughterhouse would have been quite different. For his research on nightlife. May (2<>I4) frequented local nightclubs as a regular club-goer, which put him on equal footing with the other patrons, many of whom were students at the local university where he was a professor. In these examples, where and how the researchers situated themselves shaped what they could see and whom they could access.
Positionality can also be a matter of the theoretical orientation and disciplinary training that the researcher brings to the field. When Susan Abbott (1983. 175) was conducting research in Kenya, her training in positivist epistemology made her blind to the effects of her presence in the community: "I was passing out research instalments, collecting systematic behavior observations, and largely ignoring my role as actor in the local community. I was busy doing science." Her methodological training and socialization prompted her to see and do certain things and overlook others.
In a very different setting, theoretical training and orientation also led Cyrus Ernesto Zirakzadeh (2009) to look for certain kinds of evidence and not others. When he headed to Spain to do fteldwork on Basque nationalists. Zirakzadeh went "armed" with modernization theories that impelled him to look for evidence that these nationalists were simply "disoriented burghers
Building Working Relationships 19
and small farmers" resistant to "modern social trends" (Zirakzadeh 2(X)9, 1(X>). Instead of finding support for this theory, he discovered a much more complex picture. To pursue these new discoveries, he relocated himself both figuratively and literally, severing all ties to right-wing parties (which he had made by way of a letter of introduction from a friend of his mother). He rented a room near several bars that were popular among antigovernment activists and began to frequent one in the hopes of meeting some of them (Zirakzadeh 2009, 105-6). Zirakzadeh's shift in theoretical stance led him to change his physical and social vantage point.
All these elements of positionality matter to research because they shape the research process—namely, which individuals, communities, or groups will agree to be interviewed, what they will say in those interviews, and how they will interact with the researcher (Symposium 2009; Fujii 2009; Meadow 2013;Turner 2011), If local people believe the researcher is a harmless, "cute" girl, they might readily agree to an interview (Jcntsch 1998). If they view her as a "feminazi." they might refuse outright (Cohn 2006). If certain individuals in the research site regard the researcher as a rival, they may try to block the latter s access (Abbott 1983). Conversely, if they view the researcher as a potential benefactor, they might be all too eager to participate (Goduka 1990; Pickering 2007, 194).
Insiders v. Outsiders
Positionality in the field is relational and context dependent, not fixed or absolute. How people in the field site view the researcher depends on the context and can shift over time. This means that no researcher is a "true" insider or outsider from beginning to end. Many will occupy both categories at various points in time or shift from one to the other. Elizabeth Chacko (2004.54-6), for example, set out to study healthcare among women in an Indian village in West Bengal. Chacko grew up in Calcutta ami was fluent in Bengali, which made her an insider in one way. but she was not conversant in the dialect spoken in her research site, which rendered her an outsider in another. Her degree of education also led people to identify her as an upper-caste (and upper-class) woman, even though she nude it clear that she was not Hindu. Over time, however, and as a result of living with village families for two months,Chacko found that local women began seeing lur differently, first as a "guest" and eventually as an "insider" who cared about them. Living among the women provided Chacko a much more intimate view of daily existence than she had had before. As she explains.
It was at the local well, where women congregated to wash clothes, draw water or bathe at dusk, and at informal social gatherings that I was able to obtain a depth of understanding that would have eluded me if I had restricted myself to formal data gathering procedures.
(CIukL-o 2004. 56)
20 Chapter 2
Whether participants sec the researcher as more of an insider or an outsider can present advantages and disadvantages. Commonality between researcher and participants may facilitate access. Koni Bergers (2<)15. 223) status, for instance, .is .in immigrant studying other immigrants to the United States helped to put interviewees at ease: "The moment they heard my accent, one could hear the sigh of relief and feel the atmosphere relax." Bergers (2015. 223) insider status also enabled her "to hear the unsaid, probe more efficiently, and ferret out hints that others might miss." A potential downside to being considered an insider is that interviewees might not bother to articulate what they assume the researcher "already knows" (Bcrgcr 2015, 224). For example, when Anna Gawlcwicz (2016, 35), a Polish immigrant to England, was interviewing other Polish immigrants to England, participants sometimes gave her a "knowing look" instead of putting their answer into words.
Being an outsider can also have advantages. When Portclli (2011) was collecting oral histories in Harlan County, he worried that people might resent the intrusion of an outsider. Yet, the kind of outsider he was made a difference.That Portclli came to learn and not teach made people receptive to him.That he came from Italy and not Chicago, Lexington, or other centers of power that can intrude on life in Marian County also made a difference. As one local explained: "You're not trying to influence people or anything. All you're doing is trying to gather a little knowledge or get people to tell you stories, and they don't resent that" (Portclli 2011.7). Over time. Portelli became more of an insider with local "kin-folk" who made phone calls whenever he was on the road to make sure he arrived safely at his destination.
Because "insider" and1 "outsider" are fluid rather than stable categories, and because they present potential upsides as well as downsides. Herod (1.999* 326) calls on scholars to think of them not as a binary pair, but rather along a "sliding scale of intimacy" whereby a researcher might experience different degrees of "insidcriK'ss"' or "outsiderness" at various points in time. Portclli and Chacko, for example, started out toward the outsider end of the spectrum, but over time, both moved closer to the insider end. Herod's recommendation is a valuable intervention into debates over whether it is better to be one or the other, or, indeed, both at the same time.
Elites v. Non-Elites
As with insider/outsider positioning, social status is also relationally defined. This makes the terms "elite" or "elite interviewing'" problematic since there is no set of social relations in which status and power are fixed prior to the interview (Smith 2006, 651). Sutus is also situationally contingent (Neal and McLaughlin 2009, 696); and even "elite" status can be fragile and transitory (Harvey 2011,433; Neal and McLaughlin 2009, 703*. Smith 2006, 645). Because status is relational, contingent, and transitory, there is no reason to assume that "elites." however defined.
Building Working Relationships 21
will respond to researchers in a particular way. Elite responses toward researchers w ill vary in the same ways as responses from any other type of interviewee.
Like non-elites, elites diverge in how the)* respond to a request for an interview. Some will see it as a burden (Odcndahl and Shaw 2002), while others will welcome the opportunity (Smith 2006.651). In a study of labor unions, for example, Herod (1999) found a clear difference in how high-level union officials in the United States and Eastern Europe viewed academics. In the former. Herod (1999, 316-17) felt that "I was really impinging upon interviewees' tune." whereas in the Czech and Slovak Republics, union presidents welcomed him as a valued collaborator on research of mutual interest. I his warm welcome made a big difference in the type of working relationship Herod svas able to build with participants. An he observed."Of all the research I haw done this was probably the most two-way" (Herod 1999,317).
I litis also vary in their openness or forthrightness with researchers. Some might stick to answers that conform to what public relations offices train them to say. Others will talk in frank terms. Cohn (2006, 101) experienced both types of responses when conducting interviews with active-duty military officers in the 1980s, while the Cold War was still at its peak:
A very high percentage of people seemed extremely open and forthcoming, often revealing things that clearly would cause difficulties for them if exposed, or clearly deviating from "the official line." Others were guarded, but in only one case did I have the clear sense that an informant had decided he was just going to stonewall straight through the interview.
Cohn experienced wide variation across interviewees despite the fact that the research setting and larger political context remained constant.
In addition, elites vary in their desire to exert power and authority over the researcher during interviews.4 During an interview that William Harvey conducted with a pharmaceutical CEO, for example, the man kept criticizing Harvey's questions and gave only wry short answers (Harvey 2011. 437). Linda McDowell had an easier time interviewing bank employees in London, but most were still "vers* keen to establish both how important their work or they themselves were and how busy they were, barely able to spare a minute or two" (McDowell 1998. 213S). Katherine Smith had yet a different experience interviewing policymakers and senior academics. These interviewees surprised her with the "level of self-reflection, uncertainty and nervousness" they displayed (Smith 2006.646-7).Those who did try to assert authority over her were more junior than the other interviewees.
As these examples indicate, status and power differences are more fluid and shifting than a reified notion of "elite" implies. Smith (2006. 696) even suggests, "Perhaps we need to make room to consider the possibility of'vulnerable elites.*"' Sarah Neal and Eugene McLaughlins (2009) experience interviewing authors of
22 Chapter 2
the /WWi RejK>rtt a study of multiculturalism in Britain, exemplifies this point. The report's authors were members of The Future of Multi-Ethnic Britain Commission. Neal and McLaughlin (2009.690) refer to these Commissioners as "elite policy-makers" because of their extensive experience engaging with issues of race, ethnicity, and immigration. The group met for over two years. When the report was finally published, however, the public backlash was swift and vicious.Various media sources accused the authors of being "unpatriotic." "idiots," "fools." and "anti-British." Such charges challenged the "elite" status of the Commissioners, .ill of whom had enjoyed distinguished careers in government, media, policy-making, and academia (Neal and Mcl aughlin 2009, 696).
Neal and McLaughlin interviewed the Commissioners two years after publication of the report. Rather than finding "elites" who exercised power and authority during interviews, they found interviewees who were still reeling from the public lambasting they had received. A few found it so difficult to explain their feelings, the "interview would have to stop until the individuals felt able to continue" (Neal and McLaughlin 2009, 697). The emotional intensity .md vulnerability that permeated their self-reflections unsettled the power dynamics between the so-called elite interviewees and the researchers.The public lashing the interviewees recalled made them less like people "on high" and more like discredited intellectuals and scholars. As Allan Cochrane (I99X, 2128) observes, "Elites change over time, with reputations not always reflecting continuing power."
The Importance of Paying Attention and Respect
Thinking through matters of power, privilege, and positionality requires reflex iv-ity. Reflcxivity. in turn, can help to ensure greater accountability in terms of the project s "ethical merit" as well as its knowledge claims (Benatar 2002). Given the array of responses that participants may have during an interview, the researcher should treat everyone with the same dignity and respect, regardless of how forthcoming, reticent, or hostile an interviewee turns out to be (Bclousov el al. 2007; Czarniawska 2014; Herod 1999,316; Odendahl and Shaw 2002).*
Minimizing Harm through Meaningful Protections
Treating people with respect and dignity starts with the researcher becoming attuned to sources of risk or harm to participants. These sources may lie in existing structural inequalities that place the researcher and the interviewee in radically different economic spheres, as is often the case in wry poor countries or neigh-borhoods.Thcy might also lie in current social or political conditions, such as an ongoing civil war or neighbors prone to spreading malicious rumors. Regardless of the source, researchers need to reflect on these potential risks and harms so they can take steps to guard against them.
Building Working Relationships 23
Harmful practices can manifest in mundane ways. Meeting an interviewee at a local coffee shop in town, for example, might be convenient for the researcher, but may force die other person to incur travel costs he or she can ill afford. To most Americans, paying twenty-five cents for bus fare is nothing, but to many Rwandam, twenty-five cents may constitute a substantial portion of their dairy earnings. The average daily income of Susan Thomsons (2010.22) participants in Rwanda in 2006 was the equivalent of eleven American cents. Needless to say. promising to pay the person back is immaterial if that individual does not have the money to expend in the first place.
Potent lal harms can also arise in and through the interview itself. Talking about certain topics might re-traumatize interviewves. rather than provide catharsis (Bell 2001 ).Visiiingan interviewee at her home might make rumors fly and ncighborsjealous. In situations marked by significant economic or educational inequalities between the interviewer and interviewee, asking for consent can become a "loaded" act. shaped by various incentives and pressures for the participant to say "yes." None of these potential harms is a reason to avoid certain communities or research sites entirely, for that would constitute another kiinl of harm—that of systematic neglect by researchers. Neither docs it mean avoiding certain topics altogether. What treating people with dignity and respect does require is for the researcher to become familiar enough with local conditions to anticipate why some participants might say yes when they would prefer to say no, or what kinds of social or psychological harms might befall interviewves who agree to be interviewed.
Related to the researcher's responsibility to avoid harmful practices is the equally important obligation to seek protections that are meaningful in the local context. This is especially crucial in settings such as war zones or authoritarian states, where talking to a researcher can present risks to participants (Nordstrom and Robben 1995; Sriram el ai 2009; Straus and Waldorf 2011), Such conditions might warrant making adjustments to the consent process so that participants can exercise greater control over how the researcher can use the data. In research on popular participation in the insurgency in El Salvador in the 1980s, Elisabeth Jean Wood (2006,180) created a "tiered" menu of consent levels. Interviewees could agree to share information that she could hear but not write down, information she could write down but not use in any publication, and information that she could write down and use in publications (with person and place names de-identified). In addition. Wood (2006. 382) took an extra precautionary step: site decided not to use some of the information that participants had told her she could use in her book, because she feared that Salvadoran military officers would read the book or reviews of it. Wood s fears were well-founded. She discovered later that a review of her book appeared in a publication of the US Army's John F. Kennedy Special Warfare Center, which Salvadoran military officers may have read (Wood 2006.382. fn 14).
Taking steps to ensure the protection, safety, and dignity of interviewees does not guarantee that good working relationships \s-ill develop, but a lack of
24 Chapter 2
ethical treatment can undo close relationships, even after publication. Carolyn Ellis's (1995) ethnography of a small fishing village on the Chesapeake Bay, in the mid-Atlantic region of the United States, pros-ides an example. Thinking that people in her research site would never read her book (because literacy rates were so low in the area), she did little to hide participants' identities. Although she used pseudonyms, they could be easily matched to the actual individuals she spoke to during her research. When Ellis returned to the village five years later tor a visit—as a friend, not a researcher—she was met with anger and outrage on the part of several participants. Another researcher, a sociologist who had originally introduced her to the villagers, had been reading portions of the book out loud to people, focusing on the sections that revealed the most private parts of their lives. Several individuals who had been close to Ellis told her how angry and betrayed they felt by what she wrote. As this example underscores, researchers need to maintain ethical awareness during all stages of research, including publication.
Becoming Mindful of Local Norms and Hierarchies
In addition to sharpening their ethical awareness, relational researchers should consider what it means to treat people with respect in everyday terms. These meanings will depend on the context. In some research settings, showing respect might start with observing local mums concerning idf-prcsentation It might mean adopting modest forms of dress (covering arms or heads), or greeting people using forms of address that acknowledge their status or role in die local context (e.g.."Colonel" or "Madame"). It might also entail first introducing oneself to the person (or persons) with the highest social rank in the research site. Oluwashola Olaniyan. who worked as an assistant for British researcher. Margaret Pasquini. attributes their success at interviewing farmers in Nigeria to following proper protocol.
That we held our first meeting with the head farmer was also wry important. Position (in Hausa. "maifayT) is very important in Nigeria. Nobody wants to be taken for granted ami each wants to be recognised for the position the\' hold. So the head farmer was gratified and pleased that the "hattiriya" (foreigner] and her assistant had come to pay him proper "homage." Had we simply gone ahead to address a meeting of all the farmers at once, he might haw felt slighted.
(Pactum ami Olaniyan 2004, 28)
Paying respect by following local norms shows that the researcher (or assistant) has taken the time to learn what matters to local people.
Conversely, not taking the time to do so can haw deleterious effects on the project, as the experience of Abbott (1983) indicates. After months of fieldwork in a Kenyan village, she sought to do one-on-one interviews with local women.
Building Working Relationships 25
Upon learning of this plan, a sixty-year-old woman with some influence it) the community tried to block Abbotts access by telling the other women not to cooperate with her. Abbott was shocked by this behavior, especially because she considered the woman a friend. Only later did she learn that one reason for the woman's resentment was that Abbott and her assistants had not come to her first before starting their interviews (Abbott 1963, 170). As a result of this oversight, the woman felt she had been disrespected.
Respecting Participants' Knowledge and Boundaries
Treating people with dignity and respect can involve smaller but no less meaningful gestures, such as arriving on time or on the agreed-upon day, rather than showing up unexpected. It might mean, wherever possible, calling, texting. or sending a message if the researcher is running late or cannot make the meeting because of traffic, illness, or unforeseen circumstances. It might mean cutting an interview short to accommodate the participants personal schedule or needs.
Showing respect might also involve larger gestures, such as recognizing participants'expertise. All interviewees have something to teach.otherwise there would be no point in interviewing them. Acknowledging that expertise might mean approaching people with humility rather than showing off one's own knowledge. As I'ortelli (2011, 7) observed about his experience gathering oral histories in Harlan County: "The most important things I had to offer were my ignorance and my desire to learn." In other research contexts, the opposite may be true. Acknowledging the participants' expertise might involve showing them that the researcher has come prepared to talk in an informed, rather than naive, way.
In yet other situations, treating people with dignity and respect might involve observing boundaries that interviewees draw around certain subjects. In my own experience, whenever an interviewee made certain topics offlimits. I took it as an invitation to show him or her that I would respect these limitations. Doing so led to some startling insights. Two very different experiences, one in Maryland and the other in Rwanda, illustrate this point.
In Mai viand. I interviewed a man who had a reputation for being one of the biggest racists in town. Several interlocutors had told me I should talk to him precisely for that reason. I think they assumed that as one of the "old timers"—the small group of older, white, male elites in town—he must know quite a bit about the lynching that occurred there in 1933. He would have been a teenager at the tune, hence, the right age to have Liken part, had he lived in the area at the time.
My research assistant. Linda Duyer/' and 1 went to his favorite gathering spot for the interview. He turned out to be nothing like what I expected. He was irascible but did not try to make us leave.To the contrary, he seemed to enjoy the attention. At the same time, he did not hesitate to push back on any question he-did not like. In fact, one of the first things he said was that he would not talk about anything related to World War II. I told him I would not ask him anything about
26 Chapter 2
his wartime experiences. He then proceeded to provide brief fragments from this period of his lite, as if to show us that he would be the one to decide what we would and would not discuss.
Linda and I talked to the Old Tinier on four different occasions. Each time, we hoped to learn more about a man whose life stories turned out to be more itunplex than his reputation would imply. The initial surprise came early on. During our first interview. I was amazed to learn that he had not grown up in the Maryland town where he had lived most of his adult life, but in a completely different part of the country. He had moved to Maryland after his discharge from the army.This meant that by local standards, he was not a "true" insider. He even joked about being a "foreigner," the term that local people use to refer to anyone who was not born on the Eastern Shore, as the region is called (Stewart 1990).
The other surprise came later. At each interview, he had made it clear that he would not tell us anything about what he knew or had heard about the lynching. For the most part. I did not pursue the topic, believing that respecting his boundaries was the best way to build a good working relationship. What became evident over time was his reason for not sharing what he knew. His refusal did not stem from a desire to protect the reputation of certain individuals or the tow 11 itself, which is what I initially thought, but from his own personal ethics. To this man, there was dishonor in the telling itself, especially if it meant passing on anything that was salacious or damning about those who gave him the information. He did. however, let us know what he thought of the lynching. During one interview, he said otT-handcdly. "It's a good thing I wasn't there." I was unsure how to interpret this remark so I asked why. "Because I don't approve of that kind of stuff," he replied. I was dumbfounded by this statement. I had never expected him to condemn the lynching, especially since he was part of a well-entrenched town elite, members of which may well have been involved in the lynching or at least supported it.
He surprised me yet again at the end of our last interview. I asked him where the lumberyard was where authorities had dumped the body of the lynching victim. Without hesitating, he pointed in its direction and explained some of its history. Linda and I were startled to learn these facts, since there' were several lumberyards in the vicinity and neither of us had known exactly which one had been the site. The Old Timer may haw given us this information because we had not pressed him earlier. By not pursuing topics he had stated were oft" limits, we were able to respect his boundaries and allow him to tell us what he wanted when he wanted. In this way. he was able to control the direction and content of much of our conversations.
The second example of an interviewee who set clear limits on what we could talk about involved a woman I interviewed many times in Kwanda. I originally decided to talk to Theresc7 because she was the wife of the genocide leader in the community. Since I could not interview her husband—he had been killed in 1997—I decided to talk with her about the genocide as a second-best option.
Building Working Relationships 27
Based on scholarly accounts that highlighted the active role that women played in the genocide Forges 1WJ), I was certain that she was aware of her husband's activities and may have participated in them. I just did not know to what extent or in what ways.
The very first time my research assistant and I visited her. Therese agreed to be interviewed but said almost immediately that she would not speak about tin-war or genocide. I interpreted her reluctance as confirmation of her complicity She must have been deeply implicated. I told myself, or why would she refuse to talk about those events? Over time, however. I completely changed how I saw her.
Like the Old Timer in Maryland.Therese was quite willing to talk about other topics, despite making the one most germane to my research oft* limits. As in Maryland. I proceeded to bring up other subjects as a way to respect her boundaries. During our very last interview,just before I was to leave Rwanda to return to the United Stales, she accompanied my interpreter and me down the long, dirt road to our car. as was the local custom. As we walked, she kept talking to my interpreter, who did not usually continue translating once the formal interview was over. After we viid our goodbyes at the car. the interpreter told me that Therese had finally spoken about the genocide—how her husband had assumed leadership tor the community and then became the local leader of the genocide. She revealed no information that 1 had not already gleaned from other sources, but the fact that she said anything about a topic that had been oft' limits for so long was revealing in another way. I interpreted this gesture as an indication that she had come to trust as, and part of that trust. I believe, was borne rrom my not asking questions about topics she did not want to discuss.
I had come to interview Iherese and the OKI Inner with assumptions about who these people were, what they knew.and what they had done during the periods of violence in their respective communities. But after interviewing both several times, I realized that most of my assumptions had been wrong. I had assumed Therese must have supported her husband in his genocidal activities, but I begin to shift my sense of her as the interviews progressed. By the end of my research. I came to believe that she had not participated at all in her husbands genocidal activities, but instead, had done her utmost ttol to know what was happening. Therese s coping strategy did not make her a hem; but neither did it make her an active accomplice to mass murder, as I had originally thought.
Similarly, the Old Timer was not "simply" a racist. He was, no doubt, a product of his time, a white nun in a world where being while always carried power and privilege over anyone Mack. I never doubted that he had used that perch of power to demean and. perhaps, even humiliate others, as I had heard from several interlocutors.Yct. his position in and relationship to the community where he had lived most of his life were more complex than the mere fact that he was white and male and had a reputation for being a racist. Being a racist did not necessarily make him a supporter of the lynching. He may have said he opposed the lynching because he was talking to two people (Linda and me) whom he rightly typed as
28 Chapter 2
"liberals." but I do not think this was the case. He was fiercely proud of thinking and acting independently, without regard to others' expectations. I did not view his opposition to the lynching naively, however. Had he been living in the area at the time, he may not have taken part, but given the context, I doubt that he would have tried to stop it or publicly condemned anyone who participated in it. Nevertheless, his life story challenged my own facile assumptions about what it meant to be "the biggest racist in town."
These revelations emerged through engaging with two people who. for separate reasons, refused to talk about the topic that was the focus of my research. By paying attention to what they were willing to discuss and not inquiring alter subjects they had each placed off limits, I was able to build a good working relationship with both.The bounds of the relationships were negotiates! between us. through repeated interactions, though not necessarily through explicit discussion. We felt our way as we went along. In this way, both parties had a hand in steering the direction and content of the interviews.
Learning from Relationships That Do Not Work
Building working relationships is an ongoing process that takes time and effort. Some relationships will work better than others. Some may not work at all. Such was the case with the prisoner in Rwanda whom I call "the man with the black notebook." He brought a small black notebook to our second interview and proceeded to write in it during the entire session.This action was not just distracting, it felt antagonizing. It turned the interview into a contest of wills between us and I was not able to figure out how to shift to something more productive. Perhaps I could have asked him what he was writing, but my sense, confirmed years later (see Excerpt 1 in the Appendix), was that he would not have told me anything. Suffice it to say that once the interview became more of a battle, it no longer mattered what questions I asked since his goal seemed to be to show me that he too could take control of the interaction, and control it he did. After that second encounter. I did not interview the man again during that research trip.
My decision not to interview him further made me wonder whether I was only re-interviewing people I liked or those who did not frustrate me. "People I liked" did not seem like a defensible selection criterion, so I nude sua* to return to some interviewees who also elicited the same defensive response in me. Another prisoner, whom I call Olivier in Killing .W/^/i/hw. was one such man. He used to answer every question by grinning to himself, then giving a one-word reply. As with the nun with the black notebook, this may have been a tactic to equalize an asymmetric relationship.Whatever his reasons, I stuck with him over many months, determined not to be deterred by his demeanor, despite how much it frustrated me. I lie strategy paid off. Over time, the relationship changed. By the end of field-work, he had become one of my most valuable participants in terms of helping me understand the kinds of group dynamics that operated during the genocide.
Building Working Relationships 29
By our list interview, he answered all my questions in a straightforward manner. The experience taught me that I did not have to like my interviewees—or they me—in order to build good working relationships with them.
Working with Research Assistants and Interpreters
Learning how to build and maintain good working relationships also extends to research assistants and interpreters. For this discussion."research assistant" refers to any person from the research site (or region or country where the site is located) whom the researcher hires to help with interviews and other research-related tasks, such as nuking appointments or arranging logistics. With some projects, this person might also serve as an interpreter during interviews, interpreting questions, answers, and cultural references from one language to another and back again. Indeed, it is not uncommon for an assistant to wear "both hats" (Turner 20IO,207).N
Not every project calls tor a research assistant. Working with a local assistant can be beneficial, but it also places additional costs and responsibilities on the researcher, since an assistant and especially one who serves as .m interpreter will play an active role in the knowledge production process. As Diane Wolf (1996, 23) cautions,"These relationships often require considerable negotiation." At the very least, the researcher will have to budget money to pay the assistant and may need to discuss with him or her what constitutes a fair wage and what the nature of the work will be. Additionally, the researcher will have to train the assistant and the two will have to learn how to work together effectively. If the avsistant is also interpreting during interviews, the researcher will have to budget extra time since interviewing through an interpreter generally goes at a slower pace than interviewing without an interpreter. Depending on the kinds of responsibilities the assistant will be undertaking, he or she might also have to undergo online training mandated by the researcher's university ethics board. All these tasks take time (Fujii 2013).
The benefits of working with a research assistant are myriad. With my own projects, a local assistant was immensely helpful. My assistants in Bosnia. Rwanda, and Maryland were local women close in age to me (Lite thirties to fifties at the time of research). They came from or near the rural regions where my research sues were located, were fluent in the local language and conversant in local place names, family names, and landmarks, and wen.- familiar with the culture and social norms. All three were crucial in helping me find and contact potential participants and arrange meeting places and times. They also helped me to nuke sense of what people said during interviews.
These benefits notwithstanding, working with an assistant multiplies and complicates issues of power, privilege, and positionality because there is now an additional person whose presence will shape interactions with interviewees in several ways. First, interviewees might find an assistant more legible than the researcher.
JO Chapter 2
and as a result they may focus all their attention on the assistant, leaving the researcher feeling "invisible" or at risk of losing control of her own interviews, as Margaret Pasquini felt when she began interviewing Hausa farmers in Nigeria (Pasquini and Olaniyan 2(K)4, 25). I. too. felt invisible at times in Rwanda and Bosnia, but made the same calculation as Pasquini. 1 decided early on tli.it if being invisible helped the conversation to flow, then I was happy to remain in the background. Looking back on my experiences, however. I realize that even if I felt invisible in a given moment, this did not mean I was. In settings where there is a large asymmetry in wealth and status between the researcher and the interviewee, the former's presence and privilege are always in play.
Second, research assistants come with their own ties and affiliations which can shape access to participants in important ways. As Thomas Molony and Daniel Hammett (2007.294) point out. going through a "national research agency" might expedite the task of rinding a research assistant, but the "presence of an officially sanctioned assistant can Ik constraining." Nowhere is this truer than in countries such as Rwanda where ordinary people often assume tli.it anyone with status or wealth, including foreign researchers, must have close ties to the government. For this reason, I tried to steer clear of anyone who had tics to a government body or high-level government official when 1 was interviewing candidates tor the job of research assistant/interpreter. Such ties would have undermined my ability to pro-te< t participants' identities and confidentiality. I was also careful to avoid hiring an assistant who came from the large diaspora that returned to Rwanda after the war and genocide ended in 1994. since Rwandans who never left the country tended to view these returnees as aligned with the government. There were many other factors I considered because I believed that my assistant's positionality would be crucial to building good working relationships with participants.'
Third, assistants have their own prejudices, preferences, and worldviews, which may color how they treat interviewees and interpret what they say. How assistants and interpreters treat interviewees (and how interviewees feel they are treated by them) will a fleet the kind of working relationship that the researcher is able to establish. Interviewees, lor example, may not trust the interpretations of a local interpreter who conies from a different political party or social group. Such fears might be heightened by the political climate. In post-war Bosnia. Paula Pickering (2<K»7, 193) talked with a woman who did not believe that local interpreters working for international organizations "were conveying all of her concerns." In other words, the woman did not fully trust the interpretations of those who belonged to the dominant "ethnic" category.
Two other examples illustrate how the interpreter's positionality can shape working relationships.The first conies from Sarah Turners (2010) interviews with two young women who worked as local research assistants and interpreters for two different doctoral students studying ethnic minority populations. One pair was doing research in Vietnam and the other in China.The disparate views of the two assistants toward the study populations are striking. The Vietnamese assistant
Building Working Relationships 11
had very little previous contact with the minority groups that were the focus of the project. Her impressions of these groups had come mainly from television and a single trip she took to the region before the research began. As this woman explained to Turner (2010.211):
At that time I had no idea about those people. I had no intent to talk with them, or to find out if they are interesting or not... Of course I still feel that thc\r are dirty, even after I meet them and know them very well, I still think that they arc dirty |laughs).
The assistant who worked in China had a very different reaction to the groups the researcher was studying. This woman was an ethnic minority herself. When she trawled to the region where the Dong minority lived, she was impressed by the beauty and size of the Dong village and described the local people as "pure hearted and friendly" (Turner 2010.211).
As Turners interviews make clear, both assistants approached the research with opposing views toward minority groups, and these views seemed to endure even after the projects were completed. Though Turner docs not specify how their stances affected the research itself, it is quite possible they shaped what the assistants heard and how they chose to interpret what people said during interviews. Interpretation is not a neutral act, after all. As Bogusia Temple and Kosalind Edwards (2002. 5) point out. "Language can define difference and commonality, exclude or include others." Interpreters make quick decisions about what to interpret, and because many words and concepts do not have equivalent terms in the target language, they often choose from a "dazzling array of possible word combinations" (Temple and Hdwards 2<H>2.2). As a result, they actively shape what the researcher hears and how the working relationship develops.
In the other example, Catrina MacKcnzic (2016,176) also discovered in what ways the interpreters positionality shapes how he or she chooses to interpret participants' words. MacKenzie hired two different interpreters for a project on protected zones near National Parks in Uganda. Both interpreted during the same focus groups. By comparing the respective results, she was able to sec how their interpretations differed from one another and found that each "filtered" his interpretations through his own background knowledge, experience, and education. One of the men, for example, had trained to be a nurse and had prior experience working on studies of the spavad of primate diseases. This assistant tetided to bring clinical terms into his interpretations when the topic was about disease transmission, even though these terms were not necessarily part of the participants* vocabulary. The other interpreter manifested different tendencies. He chose, at times, to paraphrase both the questions the focus group leader posed and the answers that people gave, which resulted in interpretations that "sounded more like an explanation of the issue in more educated terms." MacKenzie (2016. 176—7) speculates that this mans practice of paraphrasing may have been due
32 Chapter 2
to his background as a teacher and/or to his personal views on land use near national parks.
Fourth, the joint positionality of researcher and assistant docs not always operate in predictable ways. When I was preparing to conduct interviews in two prisons in Rwanda in 2004. I started with naive assumptions about what kind of assistant I needed for these interviews. My assumptions were based on television images of American prisons and crude ideas about how gender works in highly gendered settings. I assumed I needed to hire a male interpreter because male prisoners would not take me seriously if I were to work with a fcnule interpreter. These assumptions turned out to be wrong. Rwandan prisons were nothing like my imagined American prisons. I worked exclusively in an outdoor courtyard, where prisoners (mostly men) engaged in daily chores ami other activities while a handful of male and female guards looked on.Wh.it turned out to be far more important than gender was the trustworthiness of my interpreter. One prisoner even said as much during an interview. He stated that the only reason the prisoners were willing to talk to me was because of their trust in my interpreter. "Did it matter what you and the other prisoners thought of me?" I asked him. "Everything depends on the interpreter," he replied.
In a different setting, however, a mixed-gender researcher and assistant may bring definite advantages. For example, Pasquini was a young, white. British woman studying Hausa farming practices in Nigeria. Her assistant was a male Nigerian of similar age. He believed that the pairs mixed-gender status worked to their advantage, speculating that "two young women might have been regarded |by the local farmers) as a 'joke'and the farmers might not have taken them very seriously" (Pasquini and Olaniyan 2<MM. 29). The reality may have been more complicated. Part of what nude this team successful was the role that each fulfilled. Pasquini. the female, was the researcher, and therefore retained some measure of status while working with a male assistant in a context in which men tend to have higher status than women of the same age. Reversing the roles may have had a different effect on the research, stripping the female assistant of any status she might claim on her own and even making her vulnerable to mistreatment by others.
Such was the case with Molony. a white, European male researcher, who hired a "well-qualified" female assistant for a project in Tanzania. When the woman began working with Molony. she became subject to taunts and jeers from local young men in Dar es Salaam, the biggest city in the country. Her harassers accused her of working as a "prostitute" for the rich, white foreigner, whom they called her "sugar daddy."These taunts made her uncomfortable walking around town with Molony (Molony and Hammett 2007. 297). In this instance, the mixed-gender status of Molony and his assistant made the latter vulnerable to abuse by local men. a situation that Pasquini s male research assistant presumably never faced.
Fifth, power inequities between researcher and assistant have ethical implications. The onus is on the researcher to reflect on whether the relationship might
Building Working Relationships 33
be putting the assistant in harms way or exploiting that persons skills for the researchers benefit. As Molotiy and Hammett (2007, 294) argue, exploitative relationships can easily arise when researchers from the global North conduct projects in the global South, because they provide job opportunities for educated individuals that might be scarce in the local context. In environments where jobs ate few. lot. al assistants might "resent the feeling that the benefits arising from |the research process| greatly favour the researcher in the long-term" (Molony and Hammett 2007, 298). Molony s female assistant in Tanzania, for example, gained employment and experience by working with him, but she also paid a high price. The reputational effects may lwve even lingered after her employment ended.
Turner also calls on researchers who work with local assistants to give more thought to the sacrifices and costs the assistant might be incurring. As she points out:
How often do we carefully reflect upon the feelings of our employees? ...They too might be in a very foreign, unfamiliar situation and physical location for the first time, away from family and friends, and coping with a number of uncertainties, anxieties and mixed emotions.
(Timirr 2010,211)
These issues relate not only to the ethics of the research but also to the integrity of the project itself. If research assistants or interpreters feel overburdened or resentful toward the researcher, those feelings can affect the work they do for and with the researcher and the working relationships they are able to build with participants.
*****
Imilding good working relationships is a process. For the researcher, it is part of the privilege of doing research. With this privilege comes the responsibility to treat all people with dignity and respect, regardless of how forthcoming they are or whether the researcher agrees with their views or approves of their behavior. This responsibility docs not start or end with institutional approval by university bodies such as IRBs or RF.ISs. It is ongoing.
Treating people with respect ami dignity requires the researcher to develop and hone an ethical sensibility that can alert him or her to ways in which his or her own power and privilege might be contributing to practices that demean, violate, or overlook the needs of others—participants as well as research assistants ami interpreters—or that exploit inequalities, hierarchies, or norms that benefit the researcher at the expense of others. To understand these sources of power and privilege, the researcher must become attuned to how people in the field are categorizing or perceiving her. What status, influence, or power docs the researcher enjoy because of her nationality, putative race or ethnicity, age, gender, educational
34 Chapter 2
level, local alliances, or vantage points? Reflecting on issues of positionality across various contexts can help the researcher find meaningful ways to treat people with respect and dignity. This might include showing up only when expected, heeding interviewees' boundaries about what is open for discussion and what is not. and taking the time to notice when participants are pushing back against the interview format or inequalities in the relationship. Respecting rights and boundaries can have positive outcomes. People who initially refuse to speak about certain subjects might change their minds bter. or talk about other nutters in ways that are quite revealing. This will only happen, however, if they trust in the working relationship they and the researcher haw built together.
Notes
1 See \v\v\v.oxforddictionarks.eom/us7dennition/an)crican_atg|ish/rappt>rt (accessed II January 2016).
2 Personal communication with Aaric Glas and Jessica Soedirgo. 29 September 2014.1 thank both for letting me use their stories.
3 Rwanda's nickname is the "land of a thousand hills."
4 These attempts can extend to publications. In one extreme example of interviewee* controlling the finished product. Man Uradshasv (2001, cited in Smith 2006.649) had to agree to embargo his dissertation for fifteen years after publication in order to gain access to the corporate maiuger* he interviewed.
5 Questions about how to form good working relationships take on a different meaning when a researcher uses deception to gain access. For a discussion of'the trade-oris between transparency and deception, see Kachirat (2011) and Goodc (1996). For competing views, see the exchange between Kai Erikson (1993,1996) and Richard A. Leo (1996).
6 I am using Lindas real name with her permission. I do not identify the location svhere ssxr met with the Old Timer to protect his identity.
7 This is the pseudonym I uve for her in Killing Xeighlw*. All personal names from the book are pseudonyms to protect identities.
8 For this discussion, 1 also assume that research assistants who double a* interpreters are not trained as professional interpreters. For reflections on their experience* working with professional interpreters, tee Herod (1999) andjentsch (1998.280).
9 For a detailed discussion of the criteria I used to hire my research assistant in Rwanda, see ch. 2 of Killing Sdglibors.