|
A
Starting Place
This article has emerged from concerns with comments made by fellow
students that suggest avoiding ethnographic research methods because
of their colonial roots and historical valuing of academic knowledges
at the expense of those of research subjects. Underlying such comments
is often a wish, which I share, to construct more democratizing research
purposes and practices. However, understanding the specifics of how
certain ideas acquire status through ethnographic research methods provides
a necessary and useful starting point for constructing ethical research
relationships. My concern is to provide a detailed analysis of interview
practices as they pertain to the production of knowledge. I seek to
contribute to the discussion of ethical and authorial issues involved
in interviewing by exploring the following: How is it that knowledge
associated with the interviewer might be authorized as more credible
than the local knowledge of those interviewed? How might this be challenged?
To Interview?
Inter draws attention to the in-between, the relations, the
positioning among people. Embedded within this short prefix is the complex
relationship between me and the person with whom I will speak. Both
of us are situated within social and historical relations, dynamically
repositioning, and continually rethinking the experiences and social
relations we talk about in light of new insights. How will I approach
this process in ways that reflect these complexities?
View. The Concise Oxford Dictionary (1976) defines "view" as
"a manner of considering a subject" or "inspection of eye or mind" (p.
1297). This leads me to question the potential for two people to engage
not only together, but also in relation to a focus of inquiry. Who establishes
the purpose for this inquiry? Whose knowledge counts in this process?
Whose interests are served by this interaction? How are previously subjugated
knowledges, histories, and interests brought to bear on more dominant
ideas through this process? The relevance and responses to these questions
depend on the specific purpose and circumstances for each interview.
Related debates have been ongoing (Fontana and Frey 2000, Roman 1993,
Smith 1999, Tedlock 2000). The way in which I enter this debate is to
consider how Said's (1994) analysis of orientalism provides a framework
for considering social relationships in the production of knowledge.
His analysis lacks consideration of the interplay of social dimensions
such as gender and class. But it does bring to light the potential for
the academic information-seeker to inspect a subject, to take knowledge
generated through observation or interview, to reconstruct it in her
own terms to serve her own interests regardless, or to the detriment
of the information-provider. I argue that the questions raised can be
brought to bear on interviewing principles and practices to understand
how certain ideas are privileged during the research process.
My intent with this paper is to argue that some interviewing principles
and practices advocated in ethnographic literature perpetuate orientalizing
practices, while others attempt to acknowledge and challenge them. I
begin by explaining Said's (1994) concept of "modern orientalism" and
outline questions that provide a framework for analyzing discourses
of interviewing. Second, I illustrate how Hammersley and Atkinson's
(1997) interviewing principles and practices privilege the meaning made
by the interviewer. While I acknowledge that power inequalities are
never completely eliminated in the research process, I argue that researchers
ought to understand how certain ideas acquire status through the research
process. Finally, I highlight alternative practices of critical feminist
ethnographers that reflect and extend Said's analysis by challenging
orientalist practices and by encouraging the development of theories
through a dialectic between the interviewee's and the interviewer's
knowledge.
What is orientalism?
In Orientalism, Said (1994) analyzes how eighteenth century
French, British, and American literary writers constructed representations
of the Orient. In his analysis, orientalism has three meanings. First,
it refers to an academic study of "Oriental" languages, societies, and
peoples. Second, it is defined as a "style of thought that assumes epistemological
and ontological distinctions between the East and West as a basic division
in the world" (p. 1). The third meaning of orientalism, and that which
is central to Said's argument and to my consideration of interviewing,
is as a mode of discourse to represent the Orient as being other than
the Occident: mysterious, unchanging, and ultimately inferior (p. 2).
Said argues that starting in the late eighteenth century, there was
a "corporate institution for dealing with the Orient - dealing with
it by making statements about it, authorizing views of it, describing
it, teaching it, settling it, ruling over it: in short, Orientalism
as a Western style for dominating, restructuring, and having authority
over the Orient" (p. 3).
Rooted in this concept of orientalism is the premise that knowledge
is political, not neutral nor absolute. Considering how ideas "acquire
authority, normality, and even the status of natural truth" (p. 325),
Said argues that practices that pursue knowledge are deeply connected
to historical, cultural, and institutional forces, to a dominant ideology,
to political imperatives of the society in question, and to colonial
practices (Moore-Gilbert, 1997, p. 36). Drawing on Foucault's concept
of discourse, Said analyzes how the West's power operated through channels
of colonial management and representation to speak on behalf of, or
for, others, and to determine what counted as official knowledge and
official history. Said argues that "ideas, cultures, and histories cannot
seriously be studied without their force, or more precisely their configurations
of power, also being studied" (p. 5). While Said's analysis focuses
on textual representations and on the binaries of West and East, his
analysis of orientalism provides a general framework for my analysis
of the interview process:
-
How do these researchers challenge or maintain the role of the interviewer
as expert and interpreter, who reconstructs the interviewee in light
of knowledge rooted in scholarly research, and who is detached from
the circumstances of life and from involvement with a class, a set
of beliefs, and a social position (Said, 1994, p. 10)?
-
How do these researchers disrupt or perpetuate the role of the interviewee
as a "static and essentialized" (p. 283) object, who does not speak
for herself (p. 6), and is incapable of defining herself?
-
How do they disrupt or maintain the notion of the interview as a
predominantly apolitical process, where the interviewer, positioned
as superior in relation to the interviewee, authorizes views about
the interviewee and designates her own interpretations as true (p.
19)?
INTERVIEWER - interviewee
Hammersley and Atkinson's 1997 "updated guide to ethnographic practices
and principles" provides an example of the ways in which a disregard
for social relations and the production of knowledge serves to privilege
the knowledge of the interviewer. Judged, according to their own book
cover, as "a must text for theory and methods courses," I argue that
their characterization of the interview, interviewer, and interviewee
delegitimizes the knowledge of the interviewee and perpetuates colonial
research practices among prospective ethnographers. How does this happen?
Analyzing Hammersley and Atkinson's interviewing discourse requires
some consideration of their orientation to ethnography. They define
ethnography as a social research method whereby the ethnographer
participates, overtly or covertly, in people's daily lives
for an extended period of time, watching what happens, listening to
what is said, asking questions - in fact, collecting whatever data are
available to throw light on the issues that are at the focus of research.
(1997, p. 1)
They argue that watching, listening, asking and collecting are NOT
neutral, value-free acts separate from the influence of bias of those
involved in research. However, analysis of their recommendations for
interviewing highlights contradictions in their argument. In the following
paragraphs, I outline these contradictions and explain how Hammersley
and Atkinson's characterization of the interviewer, the interviewee
and the process itself promotes an orientalizing discourse.
First, Hammersley and Atkinson (1997) refer to the interview as a
research instrument that can be used by ethnographers to extract information
about events and as "evidence [revealing] perspectives, and discursive
practices of [people] who produce them" (pp. 124 -125). They claim that
the interviewer and the interviewee are both participant observers in
this process, "acquiring knowledge about the social world in the course
of participating in it" (p. 124), and the interview, like any other
form of social interaction, is structured by both the researcher and
the "informant" (pp. 151-152). Contradictions emerge, however, with
their assumption that the interviewer, and specifically an interviewer
without an explicit motivation may participate in social interaction
and minimize her own subjectivities and beliefs.
Implicit within their chapter entitled, "Insider Accounts: Listening
and Asking Questions," are assumptions that informants and explicitly
politically motivated researchers risk being more biased than researchers
whose key goal is the production of knowledge. To them, the purpose
of research is to produce knowledge and not to challenge the status
quo or to advocate changes to policy or practice (p. 21). They argue
that "when [people] are engaged in political or practical action, the
truth of what [they] say is often not [their] principal concern," revealing
their assumption that knowledge may somehow be separated from political
influence. While they argue that the interview process is not free of
bias, they do not address how to acknowledge or deal with the deeply
held assumptions and interests of all researchers that may significantly
shape the interpretations of the interview.
Said (1994) illustrated how such assumptions of researcher objectivism
were central to orientalist practices where the researcher's ideas acquired
the status of truth over other ideas. He revealed how information generated
by the West about the Orient and claiming to be impartial reflected
more of the attitudes and beliefs of the West than of some real Orient.
Thus, he argued that "no production of knowledge in the human sciences
can ever ignore or disclaim its author's involvement as a human subject
in her/his own circumstances (p. 10). Drawing also on Harding's (1991)
critique of "weak objectivism," I argue that Hammersley and Atkinson's
assumption that knowledge production is separate from political influence
fails to recognize how "broad, historical social desires, interests,
and values …have shaped the agendas, contents, and results" of their
own motivations for research (1991, p. 143).
What is at issue is a lack of researcher reflexivity. Hammersley and
Atkinson neglect to consider how the interviewer's establishment of
the purpose and nature of inquiry, and factors such as education, class,
gender, and racial relations relate to knowledge produced through the
interview process. While Hammersley and Atkinson acknowledge that "ethnographers
must remain aware of the likely effects of their questions on what is,
and is not, said by informants" (p. 155), they do not consider how the
interviewers' actions are closely connected to, and possibly responsible
for, "negative" reactions to questions. Instead, they blame poor results
on individual interviewees. Hammersley and Atkinson draw on one such
example in an attempt to illustrate how the ineffectiveness of an interview
is often attributable to the interviewee herself:
The nudity and perceived seductiveness of the stripper, and
the general permissiveness of the setting, had interfered with our
role as researchers. The respondent, not we, had been in control of
the interaction; we had been induced to play her game her way even
to the point that she made the decision when to end the interview.
(Skipper and McCaghy, as cited in Hammersley and Atkinson, p. 150, italics
added)
This exemplifies orientalizing discourse for two reasons. First, it
casts the interviewee as deviant and as responsible for derailing the
interview. Second, Hammersley and Atkinson imply through their subsequent
analysis that the interviewer must always be in control of the content,
length, and direction of the interview. I would agree that the interviewer
has a responsibility to guide the interview by focusing on the purpose
of inquiry. However, without considering their assumptions of the purpose
or approach to controlling the interview, they do not address the potential
for the interviewer to impose his interpretations on the interviewee.
Hammersley and Atkinson's (1997) maintenance of the authority of the
interviewer is further reinforced by their characterization of the interviewee.
In several accounts "respondents" are judged as giving "appropriate"
answers, being "deliberately inconsistent," providing "evasive," "incorrect"
answers, or "defensive reactions" (p. 127-128). The overarching message
to me as a researcher is to listen with suspicion, and to concentrate
on the evaluation of the trustworthiness of all interviewees' responses.
The effect of this message is to portray interviewees as typically untrustworthy,
and in so doing, to further authorize the researcher's interpretation
of what is 'true.' This is echoed with Hammersley and Atkinson's only
explicit reference to power relations when they urge interviewers to
control the agenda of research, the time and location with an interviewee
deemed more powerful than the interviewer. Their failure to explicitly
deal with interviewees of less power than the interviewer suggests a
disregard for the effects of power on the knowledge production of the
interview process.
Assumed within the discourse of this chapter is the notion that the
interviewer is the chief actor who "asks questions," "reads for what
the interview tells about the phenomena," "negotiates a role," "[cultivates]
and [trains] people as informants," and in one case "persuades informants
to spill the beans." Hammersley and Atkinson's advice to the interviewer
to "establish trust," "monitor the effects of the audience," establish
the "territory," and "capitalize on the more dominant role" of the researcher
in the interview process, sets up a discourse reminiscent of Said's
(1994) report of Napoleon's calculated maneuvers to invade Egypt in
1797. Determined to know about Egypt for the purpose of conquest, Napoleon
involved seven academics "to gain their trust of " and "convince the
Egyptians" that he was fighting for Islam by speaking to them in Arabic
(p. 92). This recommended control of the process by the interviewer,
coupled with the patronizing characterization of the interviewee, assumes
that the emerging production of knowledge will favor and even reify
the interpretations of the interviewer, thus perpetuating an orientalizing
discourse.
Finally, it is ironic that Hammersley and Atkinson argue for the interviewer
to establish trust, while the central message in their chapter is not
to trust the interviewees' responses. While they claim that bias is
inevitable and impossible to shed entirely, they reiterate that "the
[respondents'] answers might be deceptive and such questions will be
necessary if one suspects that informants have been lying" (p. 154).
This obsession with trustworthiness of the interviewees' responses disregards
Clifford's (1986, 1988) argument that ethnography can work to shake
off its colonial legacies through the recognition of truth as partial,
and of ethnography as involved in the invention, not the representation
of cultures. Clifford concurs with Said (1994) that one way to disrupt
the position of authoritative researcher making claims about an inferior
other, is to acknowledge fully the limitations of the ethnographic process
and product. Instead, Hammersley and Atkinson keep the authority of
the interviewer rigidly in place by failing to scrutinize the influences
of the interviewer's own bias, while scrutinizing that of the interviewee.
I have argued that Hammersley and Atkinson promote orientalizing discourse
through their recommendations for interviewing. This occurs through
the prioritization of interviewer questioning over listening, an absence
of interviewer reflexivity, and a predisposition to judge interviewer
insights as "true" and the interviewee's ideas as untrustworthy. What
is missing is analysis of how emerging constructions of knowledge privilege
existing knowledge frameworks of the interviewer over those of the interviewee.
Wishing to disrupt the orientalizing practices of Hammersley and Atkinson
and to find alternatives, I turn to selected feminist and critical ethnographers
who have engaged directly with issues of social relations as they relate
to knowledge produced in the interviewing process. The sources I draw
on do not reflect a unified approach to research, nor do they resolve
such complex issues. Rather, they represent divergent notions of how
to promote a more collaborative approach to knowledge production.
INTERVIEWEE/INTERVIEWER
Critical and feminist ethnographers such as Oakley (1981), Stacey
(1988), Young and Tardif (1992), Roman (1993) and Smith (1987, 1999)
have grappled with the disjunctions between self and other, affective
and cognitive, political and personal, and local and objectified knowledges
in attempts to develop the interview as a critical tool for developing
new frameworks and theories based on people's lives (Fontana and Frey
2000). Such researchers have commonly approached the interview as a
negotiated, joint creation of knowledge. However, they differ in the
ways they talk about and seek to address issues of how knowledge is
authorized in the interview process.
For example, Oakley (1981) argues that conditions must be created
for a two-way exchange of knowledge, such as discussing and documenting
interviewees' questions and reactions to the interview. She and Young
and Tardif (1992) independently focus on the interpersonal processes
and the affective dimensions of the interview process to enhance the
quality of information generated in the interview. Young and Tardif
(1992) propose an explicit mapping-out of the power shifts between the
interviewer and interviewee over the course of an interview relationship.
By articulating the shifts of control over the interview according to
the differences in socio-cultural and educational backgrounds, information
shared, interviewing skills and the transition from interviewing to
authored product, they seek to identify ways to better negotiate the
interviewing relationship. They recommend that interviewers disclose
more information about themselves and their theoretical orientation
to the interviewee, and present a written interpretation back to the
interviewee for feedback.
While such considerations challenge the division of feelings and thought,
and silences of the interviewee so characteristic of orientalizing representations,
enhancing the interpersonal relationships does not adequately address
issues of authorship. Stacey (1988) rightly questions "whether the appearance
of greater respect for and equality with research subjects in the ethnographic
approach masks a deeper more dangerous form of exploitation" (p. 113).
Arguing that the "delusion of alliance" can be as problematic as "the
delusion of separateness," Stacey warns that the interviewer has many
possibilities of betraying the research subjects as a result of her:
1) relative freedom to leave the research site; 2) institutional position;
3) ultimate role to interpret, to judge, to evaluate the data; and 4)
role as sole author (with rare exceptions) of the final product (p.
117). So, how might we further enhance the potential for co-construction
of knowledge through the interview process?
One proposal for disrupting the authority of the researcher is informed
by Anderson and Jack's (1991) re-conceptualization of the interview
process. As they write,
the interview demands a shift in methodology from information
gathering where the focus is on the right questions, to interaction,
where the focus is on process, on the dynamic unfolding of the subjects'
viewpoint. It is the interactive nature of the interview that allows
us to ask for clarification, to notice what questions the subject formulates
about her own life, to go behind the conventional expected answers to
the [subject's] personal construction of her own experience. (p. 12)
They argue that to "[hear] [people's] perspectives accurately, [researchers]
have to learn to listen in stereo, receiving both the dominant
and muted channels clearly and tuning into them carefully to understand
the relationship between them" (p. 11, italics added). To listen acutely
requires focus on the interviewee's formulations of her/his experiences
and questions, and clarification of how his interpretations of those
experiences go beyond the expected answers and concepts or values of
the researcher (p. 23).
More specifically, Roman (1999) advances a framework for acute listening
to extend theory building beyond the assumptions and beliefs of the
interviewer and/or of the academy. Collaboration comes, Roman argues,
from listening to the interviewee's experiences and his interpretations
of these experiences. The interviewer's role is to build theories out
of the interviewee's insights by probing with follow-up questions, listening
for silences, and testing any hypothesis by reflecting it back to the
interviewee, paraphrasing what he said, and asking for examples and
counter examples of terms that the interviewee has used.
This does not mean that the interviewee alone guides the process.
Rather, such recommendations emphasize the need for collaborative authorship
to disrupt orientalizing practices. As Anderson and Jack (1991), Roman
(1993), and Smith (2000) argue, the interviewer's education and time
allotted to research contributes essential insights for knowledge generation.
A central role of the interviewer is to acknowledge her theoretical
orientation and purpose of research, and intervene to help to frame
the interview in such a way that legitimizes and activates the knowledge
of the interviewee. Further, the interviewer must avoid the tendency
to steer interviewees solely to the her concerns while tuning into her
own feelings and responses to the interview, and taking note of her
own areas of confusion or of too great certainty (p. 24).
To begin to grapple with such issues in relation to knowledge generation
requires dealing openly and continually with the problems, contradictions,
and failures of the research process (Roman, 1999). Roman argues for
positioning the interview and its participants in relation to historical
and political contexts and analyzing "connections between the structures
and processes that give rise to [power] inequalities or to people's
experiences and knowledge of them" (p. 280). To challenge orientalizing
tendencies, she suggests creating a dialectic between "practical ethical
dilemmas and conflicts" of research and "emergent theory." She urges
researchers to "expose how their prior beliefs and structural (class,
gender, and racial) interests partially constituted the empirical evidence
for or against their descriptions and analyses of the research subjects"
(p. 281). This argument for strong reflexivity concurs with Harding's
(1991) call to critically consider the subject or agent of knowledge
along with the object of her inquiry. By promoting analysis of the purpose
and nature of inquiry as well as the positioning of the research participants,
Roman and Harding challenge assumptions of the interviewing process
as apolitical, of the interviewer as free from influence, and of the
interviewee as an unchanging, passive respondent.
Smith (1999) brings another important perspective to this conversation.
Rather than focus on the immediate relations between the interviewee
and the interviewer, she directs attention to the institutionalized
discursive practices that frame what gets recognized as worth knowing.
Arguing that researchers do not always have control within the universities
in which they work nor within the interview relationship, she finds
it more useful to focus on how social research often produces an "alienated
mode of knowing society expressing the ruling of one class, gender or
race," excluding researchers, peoples, and subjugated knowledges from
the discourses (1999, p. 74). She illustrates that despite the intentions
of researchers to create knowledge that is useful for people, their
tendency to begin with questions and concepts determined by academic
discourses serves to subsume the actual lives and knowledges of people
and produce people as objects. The challenge in her view is to frame
the research (and interviews) in ways that the respondents' own knowledge
provides the basis from which subsequent stages of research emerge.
Smith extends Said's analysis of how it is that certain ideas acquire
authority to include consideration of how people (we as researchers)
participate in the authorizing of those ideas (1999, p. 73).
In Closing
This paper only scratches the surface of complex discussions about
ethical concerns related to interviewing. If the paper has been successful
it convinces the reader to consider her/his assumptions about the role
of the interviewer, the interviewee, and the ways in which knowledge
is constructed through the interview process. By critiquing Hammersley
and Atkinson's interviewing practices and principles, I attempt to expose
issues related to a disregard for the social relations as they relate
to knowledge production. Some strategies for challenging orientalizing
practices are identified. These include: rigorous self-reflexivity on
the part of the interviewer, shared exploration of dilemmas and power
shifts involved in the research process, and acute listening to encourage
a dialectic among knowledges of the interviewee and interviewer. A significant
difference between the reporting of Hammersley and Atkinson (1997),
and that of Anderson and Jack (1991), Roman (1999), and Smith (1999)
is the commitment of the latter to explore the ongoing dilemmas and
negotiations in the interview process. Of all the factors contributing
to collaborative knowledge production, this strikes me as powerful starting
place.
|
PRINTABLE VERSION
