On ‘False, Collusive Objectification’: Becoming Attuned to Self-Censorship, Performance and Interviewer Biases in Qualitative Interviewing

Abstract

In this paper, we discuss a problem in qualitative interviewing labeled by Bourdieu as ‘false, collusive objectification’. As described by Bourdieu, interviews where this occurs appear authentic because they often echo social science concepts and terminology and therefore may please the interviewer; however, they are actually unusable. We evaluate Bourdieu’s claim for the existence of ‘false’ interviews in light of the predominant postmodern position in qualitative research, offer examples from our own research on people diagnosed with mental illness and raise the issue of whether, when and how qualitative researchers should concern themselves with the shortcomings of interviews. We conclude with suggestions derived from Bourdieu’s view on how to address the problem he described.

Introduction

What does it mean when we sense—alerted by some discordant note during the encounter or mounting unease upon reading the transcript—that an interview is too good to be true? What standard, grounded in what metric of authenticity or truth, is implicitly at work? What tools have we used (consciously or not) in detecting the potential fraud, and how good are they? Looking back at the interview itself, can an easily established rapport work against you if it is revealed to have glossed over what seemed inconsequential white lies at the time—small misrepresentations of habit or attitude—that turn out to be central material to the purposes of the interview? In the interest of narrative drive, can respondent and interviewer unwittingly devise traps that work against ‘an honest story, honestly told’ (Geertz, 1990, p. 9)?

These and other quandaries of the talking trade are commonly enough alluded to when practitioners get together. A number of them (Atkinson & Coffey, 2002; Birch & Miller, 2000; Charmaz, 1995; Dean & Whyte, 1958; McCleod, 2003; Nunkoosing, 2005; Power, 2004; Rosenblatt, 2002; Wiersma, 1988) have examined these issues with some rigor and imagination. But nowhere, in our judgment, are these addressed with more demanding a standard of excellence than in Pierre Bourdieu’s odd, unsettling methodological essay, ‘Understanding’ (Bourdieu, 1999). In his own crabbed fashion, Bourdieu called this trap ‘false, collusive objectification’: false, because it is unfaithful to the particularities of the life under scrutiny; collusive, because the interviewer unwittingly goes along with the artifice; and objectifying because a readymade account has been substituted for the lived, unfinished angularities of a reflective subject. The net result is that respondents effectively elude the interview’s incentives, opportunities and instructions, all of which have been formally designed to secure some modicum of truth peculiar to the singularities of this respondent. Instead, following some unstated convention, respondent and interviewer both default to a mode of self-presentation of pre-emptive closure, one that sidesteps uncomfortable issues and rides roughshod over inconsistencies. The sidestepping is accomplished in a manner that is deceptively pleasing to many interviewers (and is therefore not noticed or addressed by them)—because the language that is used echoes social science concepts and terminology, because the account is packaged persuasively or because it fits in neatly with theories that the interviewer is familiar with. Bourdieu contrasted false and collusive interviewing with an ideal interview, itself a kind of ‘spiritual exercise,’ in which the researcher works with the respondents in a ‘simultaneously painful and gratifying effort’ to explore issues ‘which they may find it most difficult openly to declare and assume’ (Bourdieu, 1999, p. 616).

Bourdieu’s view that some types of data can be judged to be of poor and even unusable quality stands in contrast with the currently popular ‘postmodern’ position in qualitative research, which holds that it is misguided to pursue authenticity (Atkinson & Coffey, 2002; Power, 2004). In ‘ Understanding,’ however, Bourdieu takes the position, which others (Charmaz, 1995; McCleod, 2003; Rosenblatt, 2002) have echoed, that even if it is not possible to find objective truth, there are important consequences to abandoning the quest for objectivity. Our purpose in this article is both to unpack some of Bourdieu’s densities and complexities and to explore possible correctives, both pre-emptive and remedial, to suspected falsehoods in interviewing. Through our discussion, we hope to increase awareness of the issue of whether, when and how qualitative researchers should concern themselves with identifiable shortcomings of interviews.

How Does Authenticity Become Compromised?

Bourdieu did not discuss particular conditions under which ‘false, collusive objectification’ is more likely to occur. However, we suspect that authenticity is most likely to be compromised when interviewing about the topics that reflect powerfully on respondents’ identities. Supporting this hypothesis, we have noted a few articles that identified a tendency for respondents to use the interview in a manner consistent with Bourdieu’s warning. For example, Wiersma (1988), in a study of women who had transitioned from homemakers to careers outside the home, found that her respondents initially presented stereotyped, irrelevant and implausible accounts that she termed ‘press releases’. These accounts sounded like promotional advertisements for career change but offered very little explanation of how the transition to a career utside the home actually came about, or the difficulties that it caused. Snow and Anderson (1987) found that roughly a third of a sample of homeless persons (and particularly those who had been living on the street for between two and four years) engaged in ‘fictive storytelling’ about themselves during interviews. The authors interpreted these stories partially as efforts to present oneself as ‘different’ from other homeless people and as protection against identification as ‘homeless’. These stories about personal history or future plans were determined to be ‘fictive’ (and therefore inauthentic) because they stood in contrast to the observed behavior of the person or contradicted other statements that he or she had made in other contexts.

In our own research on issues related to the recovery and the social integration of people diagnosed with severe mental illness (and frequently co-occurring substance abuse problems), we have found that there may also be reasons for participants to present accounts of what they do with their time, or how they were able to effect changes in their lives, that may be ‘too good to be true’. In part, this may have to do with perceived pressures to conform to a type of ‘model’ story that is consistent with what is typically seen in widely distributed autobiographical ‘success’ stories, or to conform to what is heard in group settings where stories following a standard script are routinely told, such as 12-step meetings. Other persons may have picked up on the types of stories that mental health professionals encourage them to tell—for example, ones that emphasize ‘insight,’ overcoming ‘denial’ and the importance of finding caring professionals. At other times, an individual’s actual behavior may involve practices that do not fit in with what is recommended by professionals (e.g. the participant may continue to use substances rather than abstaining altogether, or may use psychiatric medication in a manner other than as prescribed). The end result can be that the interviewer feels good that an uplifting version of the story of ‘recovery’ has been confirmed, but that little objective information is actually obtained about the person’s experience.

For an example of what we have in mind, consider the following transcript excerpt from an interview with an individual discussing the factors that have influenced her recovery from mental illness:¹

I’m kind of proud of myself. And I’ve thought of doing things I couldn’t do which I found out I can do. At the time I didn’t know. At the time I thought I couldn’t do anything. I was wondering—why am I here?

Interviewer: When did you think that you couldn’t do anything?

Participant: Back then. That is how I was coming up. Then I didn’t know I had a mental illness until I got in my mid-thirties. My mid-thirties, that is when I had a mental illness and I got upset. I said I can’t do anything or else I start crying. I can’t do anything right. What did I do with my life? And then, until I started, started getting back into computers … Find out all the information I could find on it … my motivation started coming … slowly, it was a process. And, I really got motivated when I came to this programme …. I learned a lot of coping skills … you know I should be using. Um, it gives me a lot of energy. … No stress, there is no stress. I am stress free. Makes me feel good and helps me a lot with my mental illness and I always know about it, because I can research on it. … At first, when I found out, I couldn’t deal with it. I kept saying, I ain’t crazy. But now I’m dealing with it … I did research on mine. … I do a lot of kinds of research that … by me doing that … all the research, it gives me motivation. It says, well I do have a mental illness but look what you achieve. Look what you achieve as having a mental illness. So, I say, I tell other people because you have a mental illness don’t think you can’t do anything. You can do a whole lot.

In this transcript the participant offers an account that stresses how she was able to overcome ‘denial’ through education, and acquire ‘insight,’ ‘motivation,’ ‘coping skills’ and ‘hope’. As such, it neatly fits in with theories of how psychosocial rehabilitation can influence the process of recovery. Initially, the relatively inexperienced interviewer did not find these responses to be potentially problematic, but, upon closer examination, these came off as too neat and a bit canned. In particular, statements such as ‘I am stress free’ and the use of such terms as ‘coping skills’ and ‘motivation’ adopt professional terms that gloss over the complexities of managing severe symptoms and problems in everyday life. In retrospect, the interview does not provide much useful information about the process of coping with severe mental illness.

Another example illustrates how discrepant information can sometimes be gleaned from observations or statements outside the interview context, which can illuminate false statements made during the interview. In an interview with a respondent with a long history of mental illness and substance-abuse problems, the respondent spoke excitedly about his participation in substance-abuse treatment, but when asked about issues of psychiatric problems and treatment, the participant was undemonstrative, almost dismissive. When asked about his use of psychiatric medication he stated:²

so far the Zyprexa …, I don’t know what it’s doing for me, you know. But uh, it’s like methadone, right, after the first six weeks, you know, it’s like drinkin’ water. You know. Like you don’t get high anymore, you know. And it’s uh, it blocks the opiates, that’s all it does. (Interviewer: And you take the Zyprexa and the Prozac on a sort of regular basis?) Yeah. (Interviewer: The way it’s prescribed, in the same doses?) Mm.

After the third interview, however, when the tape-recorder had been turned off, the participant off-handedly revealed something altogether different: that he does not take his psychiatric medications on a regular basis at all, but only takes them at times when he is concerned about doing something that might lead to others noticing him as different or disordered. Here is the fieldnote recording that revelation:

Fieldnote: Casual conversation after tape goes off … talking about a recent test he had to take …: ‘no meds prior to taking the exam—no meds at all! —except once in a while “if I feel like I’m going off the deep end” and “in order not to call attention to myself” when under stress (e.g. spat with neighbors). Been throwing the prescriptions away since leaving jails [several years ago].

Here, when the ‘staging apparatus’ of the recorded interview had been removed, this brief revelation revealed something more accurate and genuine than anything said previously about the participant’s use of psychiatric medications—that he used them sparingly, and only when he thought he needed help in holding things together, to avoid making a spectacle. This throwaway remark offered not only a glimpse into his real world of coping, but also a commentary as well on the unreal one of interviews. What had been a white lie tossed off in the initial feeling-each-other-out interview and then later reaffirmed had morphed into an unwanted fiction maintained in the interest of presenting the responsible self he found himself wanting to be taken for—a self that both the lie and the uneven medication practice were at odds with.

What to Do About Interview Problems: The Postmodern Response

Postmodern qualitative researchers have not ignored the issue of what to do with self-censorship and performance in interviewing, but they have in a sense neutralized concern about these issues by ruling the question moot. Atkinson and Silverman (1997) criticized a naive faith among social science researchers in the authenticity of interview data, which they characterized as being rendered inherently suspect by the ubiquitous nature of interviews in contemporary media (cf. Denzin, 2002); however, they did not address or acknowledge the possibility that there can be degrees of objectivity in interview data. Rather, their critique suggested that they consider all interview data to be suspect. More recently, Atkinson and Coffey (2002) summarized the postmodern perspective on how concerns with interview objectivity can be addressed. The role of the interview is not to assess the accuracy of what is being stated, they argue, but to generate data that are ‘examined for their properties as accounts’ (p. 808). From this perspective ‘contrasts between what people do and what they say become irrelevant,’ and instead ‘attention is paid to the coherence and plausibility of accounts, to their performative qualities, the repertoires of accounts and moral types that they contain, and so on’ (p. 808). Lack of authenticity is apparently of little concern when one is focused purely on the interview as a vehicle of identity construction.

Though Bourdieu did not directly engage postmodernism in his essay, we think it clear that he would argue otherwise. True, if all an interviewer is interested in is how this highly artificial exchange might shed light on identity construction, then the performative position might suffice. But, what if the researcher is interested in something that reflects on people’s activities outside of the staged confines of the interview context—perhaps an issue of considerable public health or social policy importance? For example, what if we really want to know how to help other people with severe mental illness negotiate workable modes of social integration? Or, how to design interventions for improving people’s ability to collaborate with their physicians to arrive at workable medication regimens—interventions that took into account functional illiteracy, for example? In such cases, it is important that the information we obtain be as objective and ‘nuanced’ as possible, if what we learn is to be of use to others. Here, the postmodern perspective offers little guidance. Clearly, then, we need ways to increase interview objectivity, enhance authenticity and generate alternative means of doing so when the original approach falters. At the least, we should be able to identify when severely compromised objectivity/authenticity is present with some reliability.

Alternatives to Postmodernism: Implications for Qualitative Interviewing

Rejecting the postmodernist position and accepting that ‘false, collusive objectification’ is a pressing concern for researchers, merely underscores the pragmatic question of what to do instead. A close reading of Bourdieu’s ‘ Understanding’ provides some guidance on how interviews can avoid the mutual bad faith (connivance in a staged production) that can subtly sabotage the encounter. To avoid well-rehearsed (and initially appealing) performances, to move both interviewer and interviewee to uncharted ground where both tread with anxiety and uncertainty, Bourdieu suggests a few key practices described as ollows.

Active and Methodical Listening

This posture ‘combines the display of total attention to the person questioned, submission to the singularity of her own life history—which may lead, by a kind of more or less controlled imitation, to adopting her language and espousing her views, feelings and thoughts—with methodical construction’ (p. 609). Methodical construction is ‘based on a sociological “feel” or “eye” [that] enables one to perceive and monitor on the spot as the interview is actually being carried out, the effects of the social structure within which it is taking place’ (p. 608), and to correct for or control those effects. Thus, interviewers need to develop the ability to have a complete attunement to their respondents’ worlds in order to avoid pitfalls such as thinking that they have ‘heard it all before,’ while simultaneously drawing upon their own scientific knowledge to inform their understanding of what participants describe.

There are some technical similarities between what Bourdieu refers to as methodical listening and what phenomenological researchers have referred to as the ‘empathic bridge,’ in which the interviewer plumes personal experiences in order to place himself or herself in the respondent’s position, thereby obtaining a deeper understanding of the factors that motivated the respondent in a given situation (Davidson, 2003). A key difference, however, is that Bourdieu would assert that the interviewer also needs to go beyond experiential knowledge, bringing a sociologically informed understanding of structural aspects of the ‘habitus’ that have influenced the respondent, and of which the respondent is not likely to be consciously aware. Presumably, both experiential and sociological knowledge will be able to lead the interviewer to the conclusion that something that is being described is not likely to be authentic.

In-Depth Knowledge of the Circumstances Relevant to the Respondent

Bourdieu insists that the interviewer needs to have ‘knowledge of objective conditions common to an entire social category’ (p. 613) in order to carry out good interviews. This can be acquired in a number of ways, including drawing upon in-depth sociological knowledge of the participants’ conditions (as discussed above) and minimizing social distance between interviewer and respondent (i.e. matching interviewers to respondents with similar socioeconomic and cultural backgrounds whenever possible). Minimizing social distance may provide the interviewer with greater access to the types of experiential knowledge discussed above.

An additional technique that can be added to accumulate background knowledge about respondents is to follow Ortner’s (2003) suggestion for incorporating elements of participant observation into primarily interview-based studies whenever possible. This could include making arrangements to observe behavior as well as ask about it, which can help to illuminate inconsistencies, and documenting casual conversations occurring outside of the formal interview context (e.g. when the tape is not rolling). Certainly, Snow and Anderson’s (1987) research supports that casual observations can sometimes illuminate inconsistencies, as did our experience in the second example from our own research described previously.

Repeated Interviews

Deeper knowledge of the respondent can also be earned by conducting repeated interviews. With time to test out, refine and revise accounts as necessary, and the slow luxury of real rapport, both interviewer and interviewee are more likely to find the wherewithal to locate and survey previously unexplored ground. Images that participants present of themselves are also less likely to be contrived upon the second or third interview. Wiersma (1988) observed that the women in her study were more likely to let go of their ‘press releases’ during second or third interviews.

Attending to ‘Clichés Behind Which Each of Us Lives,’ Including the Interviewer’s Own Biases

This axiom of Bourdieu’s reflexive sociology is one of the most challenging aspects of the process, because working clichés can feed the ‘false, collusive objectification’ that converts artifice into insight. Although Bourdieu was vague on how to detect this process, his counsel to reflexivity would seem to demand that researchers tune themselves to detecting contradictions in interviews and cultivate an alertness to accounts that seem ‘canned,’ rehearsed or a repetition of an existing cliché. Statements that sound particularly rehearsed or similar to ‘model’ stories (e.g. stories from pop culture, the AA model) can be a signal that something is amiss and that the authenticity of statements made in a particular area should be taken with a grain of salt. Alternately (and paradoxically), they may demarcate the areas of genuine inarticulateness that will require much more work if the material faithful to the distinctive experience of that respondent is to be plumbed (Swidler, 2000). When encountering such situations, it would seem prudent to make attempts to revisit the issue with the interviewee. Obviously, study protocols too will need to build in flexibility to accommodate revisiting such issues and provide for repeated interviews as needed.

Willingness to Classify Interviews as Uninformative

In a footnote within ‘Understanding,’ Bourdieu indicated that interviews demonstrating ‘false, collusive objectification’ were considered ‘botched’ and were ‘dropped from analysis’ (Bourdieu, 1999). This approach is likely to be a radical one for many researchers. (To begin with, for most funded and institutionally approved research, budgets and protocols have explicitly stipulated that X number of interviews will be conducted.) How can a researcher judge data to be unusable? Is it even ethical to discard data? With appropriate practice and foresight, researchers may need to stipulate criteria and procedures for bracketing certain interviews as ‘unusable’; others might be re-broached by a different interviewer or approach and still others could be discontinued after the initial efforts prove unproductive. Obviously, guidelines are needed to ensure that such correctives do not inadvertently introduce fresh biases of their own.

The above recommendations are not ones that can be easily put into practice. By invoking the metaphor of ‘spiritual exercise,’ Bourdieu wanted not simply to set a high bar for quality. He meant as well to remind us that, like the religious adept’s goal of perfection (approachable but never reached), the good interview will require rigorous preparation and disciplined practice—along with a keen eye for illusory insights and counterfeit visions. Ignatius termed this latter gift ‘discernment,’ and it is the cultivation of that faculty that perhaps best distinguishes the competent interviewer.