Accueil Revues Revue Numéro Article


2006/3 (Vol. 9)

  • Pages : 24
  • DOI : 10.3917/mana.093.0095
  • Éditeur : AIMS


Votre alerte a bien été prise en compte.

Vous recevrez un email à chaque nouvelle parution d'un numéro de cette revue.


Article précédent Pages 95 - 115 Article suivant

This paper seeks to open up for discussion the emotional world of researchers in a manner that encourages and supports reflective practice. Drawing on the work of Clifford Geertz (1968) we focus on the irony inherent to research —elaborated via the concept of covertness– whereby ethnographic researchers construct mutual fictions in their relationships with respondents, which obscure the authenticity and sincerity of the emotional exchange between researcher and researched. Specifically we discuss examples of interpersonal dynamics which generate uncomfortable emotions and identity work on the part of researchers. Ultimately, we advance understanding of how emotions and identity work influence the collection and interpretation of data. The methodological implications for conducting ethnographic research are discussed.



Our thinking for this article began with a throwaway comment by Richard Sennett and Jonathan Cobb in The Hidden Injuries of Class. In discussing the power relationship between teacher and pupil, they write about « Max, an obnoxious fifth grade bully » and sometime writer of doggerel poems, and his teacher, who was less than impressed after hearing him read one of his poems (Sennett and Cobb, 1972: 86- 87). This casual emotive descriptor of Max is quite an unusually frank and subjective remark in published research. Most of the emotively prejudiced and everyday gut reactions to meeting and talking to people in the field don?t make it to the published work, even if they do reach the field notes. Hubbard, Backett-Milburn, and Kemmer (2001: 125) acknowledge this reality and suggest that we are compelled to hide our occasional loathing. And, as Gary Fine has commented « hated individuals are found within our ethnographic world [but] we crop them from the picture » (1993: 273).


These reactions and their management via identity work —such as adopting a mask of objectivity— form part of the emotional labor of the researcher. In our view, researchers often pay insufficient attention to this labor and the emergent, compromised, and continuously negotiated nature of the research process, producing instead post-hoc explanations that appear over-planned and over-rationalistic (Geertz, 1995; Czarniawska, 1998). Our analyses of emotions show how intellectually difficult, emotionally stressful, and politically strenuous real world research actually is (Kleinman and Copp, 1993). Our paper explores the emotional dissonances —fear, disgust, loathing, discomfort, embarrassment, etc.— produced via our field experiences, some identity work strategies we used for « dissipating » them (Geertz, 1968: 154), and the implications of these experiences and practices for our research. In other words, we explore how, despite our efforts to obscure the emotional dissonances of ethnographic research, we often can?t help but call Max obnoxious.


In the empirical sections of the paper, based on a large six-year study of culture change at a steel works in Australia (Badham, Garrety, Morrigan, Zanko, and Dawson, 2003; Badham and Garrety, 2004; Garrety, Badham, Morrigan, Rifkin, and Zanko, 2003; McLoughlin, Badham, and Palmer, 2005), three examples of emotional dissonance and the identity work used in its dissipation are discussed: 1/reactions to displays of emotion; 2/reactions to specific respondents; 3/reactions to generalized others. The subsequent discussion section explores the methodological implications of this analysis for research practice in general.



In order to meet the predominantly methodological aims of this paper we will need to give short shrift to some rather large and contested areas of social and scientific thought: emotions, identity and their intersections. Our general methodological purpose in this paper stems from a growing interest in the intersections of emotions and identity in our specific research field of organizational culture change (Carr, 1999; Turnbull, 2002; Keifer and Müller, 2003). The most famous work in this area is Hochschild?s (1983) research into the emotional labor of air stewardesses. In this and other studies that address cultural change in organizations, specifically the move to more normative (Barley and Kunda, 1992) or emotional (Kunda and Van Maanen, 1999) forms of control, there has been a relative lack of in-depth exploration of the effect of the research process on the emotions and identity of the researcher, despite the emphasis it has received in the methodological literature (Kleinman and Copp, 1993). If we use Hochschild?s (1983) terms for example, the questions —what are the messy and contested feeling rules inherent in research situations and how do researchers wrestle with these?— often remain unasked and unanswered.


This relative lack can, for instance, also be seen in Kunda?s (1992) influential study of cultural engineering in a high technology company. Despite the exemplary nature of the work, he writes only vaguely of « pain » and « discomfort » regarding his fieldwork (1992: 238 and 239). This is not a criticism of such analyses; we are all too aware of the need to avoid ethnographic research becoming « a black hole of introspection » (Van Maanen, 1988: 92). However, the gaps in his account do merit more generalized investigation. Did Kunda make friends? Did he dislike some of the people he met? What did he feel about his experience? How did he present himself and his work to his subjects? Our purpose in asking these questions is not to prescribe research conduct (though this may be one implication), but to uncover sources of emotional dissonance produced in the field, and the effects of researchers? attempts to dissipate them through identity work. Thus, we will suggest later that who we are as individuals and as academics, cannot but influence the conclusions we are able to draw.


While a systematic overview is beyond the scope of this article, some consideration of the broader concepts of emotion and identity is necessary for our discussion here. The nature of human emotion is, of course, much disputed both within and between different scientific fields. However, two categories seem to encompass most of what we recognize as emotion. Firstly, there are those that seem universal to all cultures and can be described as an affect program of « reflex [emotions that combine] behavioral responses, physiological changes, facial expressions and (somewhat later) thoughts ». Secondly, there are those « complex cognitive emotions that have to do with morality and social relationship[s] » (Jasper, 2002: 146), which differ depending on the cultural context. Arguments about emotions often pivot on the distinction between these two categories.


In this article we are primarily interested in the social and relational aspects of emotions —the meanings ascribed to them: specifically the meanings which emerge from those emotional dissonances created in the process of research. The so-called self-conscious or social emotions, such as shame, guilt, embarrassment and pride, are only generated in response to the perceived (real and/or imagined) evaluations of specific or generalized others (Lewis, 1993; Tangney, 1999). They play an important role in social control, as individuals will try to conform to prevailing social expectations in order to avoid embarrassment and shame (Scheff, 1988). Emotions generated in the course of interpersonal relationships frequently express « a judgment of the moral quality of some other person?s action » (Harré and Gillett, 1994: 146-147). In the coming examples of our emotional reactions to displays of emotion, specific and generalized others, we will see the dissonances produced via our attempts to judge or avoid judging others, and the identity work required to dissipate these dissonances.


Identity is also a contested concept. Nevertheless within the social sciences, at least, a consensus that we become selves via processes of narrative construction is orthodox (Giddens, 1991; Somers, 1994; Jenkins, 1996; Down, 2006). However, despite the sophistication of recent debates which have identified the manner in which individuals manage identity work (Kärreman and Alvesson, 2001), particular difficulties remain in accounting for the intersections of emotion and identity (Bosma and Kunnen, 2001) —for example, how, and in what way, the coherent selves that we present or adhere to are altered in response to the emotions generated by social interaction (Scheff, 1988).


In the specific case of the identity of researchers, the narrative role category of the researcher is a widely understood and legitimate character (MacIntyre, 1981). The researcher is allowed, and often expected to be, aloof, analytical and emotionally disengaged (Emerson and Pollner, 2001). Nevertheless, he or she is also susceptible to the full range of human emotions. Researchers do react emotionally to others, and want to be accepted (Kleinman and Copp, 1993). They —we— want to produce useful work. These tensions are exacerbated by the fact that researchers, especially in ethnographic fieldwork, are placed in the difficult position of standing between two social worlds (Geertz, 1968): their own and that of the researched. In order to do his or her work, the ethnographer needs to walk a fine line between doing closeness and doing distance. This line can be disrupted by subjects who seek to exclude the researcher, or who draw him/her into their worlds in ways that make the researcher feel uncomfortable (Emerson and Pollner, 2001). The identity work of the researcher in managing the closeness/distance dynamic and in bringing coherence to the intersection of the two social worlds through the creation of viable narratives of self and others (the researched) is interesting because of the implications it has for what academic knowledge claims to be able to achieve. The inherent moral ambiguity of the researcher (friend, emancipator or judge? Emotionally engaged or detached?) places him or her in a contradictory position. Though the claim that researchers bring both identity and emotion work into their analyses may be obvious, showing how our identity work is linked to emotional dissonances produced in the field, and the methodological implications of the practices we employ to dissipate them, is a novel contribution.



Traditionally of course, emotionality is considered taboo in research, and is —even if recognized as unavoidable in the qualitative field— squeezed out in most research accounts (Blee, 1998: 383; Fine, 1993). Whilst the desire to appear detached and objective in the face of experience may not police our accounts quite so carefully today, many researchers still feel that their emotional responses are irrelevant. After all, it?s them we?re researching, not us. However, increasingly, and especially in fields outside management and organizational studies, scholars are recognizing that analyses can benefit from attention to researcher emotionality (Kleinman and Copp, 1993; Blee, 1998; Hubbard et al., 2001; Lerum, 2001; Garot, 2004). Thus, Hubbard et al.?s (2001) account of their fieldwork discusses how we need to avoid impoverishing our understanding of the social world by omitting reference to the emotionality of researchers. And Lerum (2001) has argued that highlighting the emotional attachments of ethnographic research by dropping the protective armor of detachment can lend objectivity to the research because the researcher no longer has to protect his or her expert position. As part of her study of racist groups, Blee (1998) showed how fear and threat influenced research interactions and the knowledge that emerged out of them. She also suggested that the emotions evoked in the researcher can provide useful data regarding the social dynamics engaged in by the researched in a range of situations (Blee, 1998: 382), not just when subjects are difficult and threatening (1998: 395).


These accounts add greatly to our understanding of emotions and research practice. Where our analysis advances understanding is by stressing the moral ambiguity of the research act via investigation of how emotional dissonances, and the identity work which dissipates them, influence the collection and interpretation of data. Specifically our discussion is guided by Geertz?s notion of anthropological irony, a phenomenon that emerges out of two ethical dimensions of fieldwork: « the imbalance between the ability to uncover problems and the power to solve them, and the inherent moral tension between the investigator and his subject » (Geertz, 1968: 155). His reflections and analysis of his research in Java point to the ironic position of the fieldworker in standing between social worlds. In his case, in regard to the first point, it was the social and material distance between the Western anthropologist and the Javanese informant. The distance at the steel works between us and them is clearly not as great, but as our empirical analysis of the identity work we do to dissipate emotional dissonance will show, it is nonetheless powerfully incumbent on our experiences.


Secondly, Geertz was concerned to show that the fieldworker has very little to offer respondents, such that the researcher is left « ethically disarmed » (Geertz, 1968: 151). He suggests that the only thing we have to give, after bribery and highfaluting idealism is discounted, « is oneself » (1968: 151). That is, we tend to build up personal and emotionally engaged relationships in order, in part, to maintain a degree of self-respect. But these relationships are partial fictions and produce an ironic and « inherent moral asymmetry » (1968: 151). The partial fictions that both researcher and respondent ascribe to the social interaction are ironic in that although the researcher and his/her subjects go along with the fiction, they are « never completely convincing for any of the participants » (1968: 151).


For Geertz, as for us, there are no easy ways to escape the ambiguities and tensions inherent in ethnographic fieldwork. The only way forward is to recognize the situation and reflect on it, and somehow dissipate the tensions so that work can proceed: « To recognise the moral tension, the ethical ambiguity, implicit in the encounter of anthropologist and informant, and to still be able to dissipate it through one?s actions and one?s attitudes, is what encounter demands of both parties if it is be authentic, if it is to actually happen. And to discover that is to discover also something very complicated and not altogether clear about the nature of sincerity and insincerity, genuineness and hypocrisy, honesty and self-deception » (Geertz, 1968: 154- 155).


This dissipation requires emotional labor and identity work. In dealing with emotional dissonance and constructing identities, we create the fictions that allow the research to proceed. But for Geertz (1968: 154) these « fictions » were not the same as « falsehoods ». While the latter are clearly concocted to deceive, the former can accommodate irony and different, sometimes contradictory, perspectives, which can add to the research, not undermine it, a point to which we return later, when we look at the methodological implications of our analyses.


For us, the chief irony of our position at the steel works was that we would on occasion justify our presence by claiming that our research would contribute to better change management in the future, whilst being unable to do much in the here and now. To at least this extent then —recognizing that « all research is secret in some ways » (Fine, 1993: 277, emphasis in original)— all field research is imbued with a sense of covertness on the part of researchers, where ironic reflection could exercise and inform proper conduct.


Of course others have discussed the emotionality provoked by their research. However, discussion of the covertness —the disjunction between the substantive aims of researchers and their need to establish emotionally meaningful relationships with respondents— inherent to field work is more limited. Before we explore our own examples, it is worth discussing Lerum, Blee and others? research, who highlight different sorts of tensions, and suggest different avenues through which they can be dissipated —even though they don?t really address the inherent ambiguities embodied by the notions of anthropological irony and covertness.


Lerum conducted research on sex work. She called her task « politically and emotionally tricky » (Lerum, 2001: 468) because (among other things) she was sometimes emotionally « hooked » by her informants (2001: 469). Traditionally, researchers in this field have hidden their own sexual feelings, standards and preferences behind a wall of detached objectivity. According to Lerum, this inhibits the development of critical knowledge because it ignores the subjective dimensions of the research topic. For her, dissipation of these tensions involves dropping the pretence of academic armor that researchers use to maintain control and a sense of superiority over their informants (Lerum, 2001: 473, cf. Geertz, 1968: 157). This armor is manifested through appearances and beliefs —the way we talk, dress, and possibly believe ourselves to be intellectually superior. Lerum claims that we can drop this armor by talking and dressing differently, and by allowing ourselves to become emotionally engaged with our research subjects. This facilitates the collection of « truly subjective, emotionally engaged, embodied data » (Lerum, 2001: 481) that, combined with data obtained through traditional positivist methods, leads to the creation of knowledge that is more « objective » than that collected from a purely detached position (2001: 479-80).


We agree with Lerum?s claim that surfacing, rather than suppressing, emotional responses in the field can contribute to the production of more realistic ethnographic accounts. However she oversimplifies the issues at stake. Firstly, as we show below, personal and academic armor is not so easily discarded. The identities we bring into and construct in the field remain an intractable feature of the research encounter, and the emotional dissonances produced as a result of the interactions we make are not so easily dissipated via dropping our occupational identities: our armor. Secondly, there are reasons other than a desire to appear intellectually superior which can justify a wish to remain emotionally detached. Detachment may also be a way of dealing with undesirable emotions, such as aversion to a particular research subject or situation. Detachment is not necessarily less sincere and authentic than overt emotional display and involvement, as Geertz has observed. Thirdly, we are both persons and researchers and it is legitimate to protect our own sense of who we are. In other words, while immersing ourselves in subjective experience may help dissipate the emotional dissonances produced in the field in some situations, it does not necessarily do so.


The complexities of the issues at stake are further highlighted when we juxtapose Lerum?s article with Blee?s observations on conducting research on racist groups. While Lerum?s (2001: 473-74) account was suffused with empathetic expectations that emotional experiences in the field would be positive, Blee?s account of doing research on racist groups gave a very different picture and for her « fear was ever-present » (Blee, 1998: 388). Blee dissipated these tensions by actively distancing herself from her informants and by using the emotional dynamics of her encounters in the field as data. In making this latter suggestion, she was agreeing with Lerum. However, because of the nature of the research and her emotions, she announced to informants that she did not share their views, and that they should not view her as a sympathizer or potential convert. She actively eschewed any attempt to establish rapport with her informants (Blee, 1998: 388). Nevertheless, in order to gain their involvement, she promised « to try to present an accurate depiction of women racist activists » without judging them « in advance [as] crazy [or] personally pathological » (1998: 385). In achieving her research, Blee thus performed some very complex emotion and identity work.


Neither however, is especially explicit about the mutual fictions or covertness of their research. Lerum (2001) is perhaps somewhat naïve about her ability to drop her armor: to believe that there are no ironies that will require dissipating identity work on her part. Blee (1998) places more emphasis on the ambiguity of her position, as she considers the possibility of being used as a platform of racist views and perceives herself, on occasions, as being lured by racists making friendly overtures. Both, however, seem to infer that taking a position vis-à-vis research subjects is a one-off affair that can address or resolve the tensions of the situation in which they find themselves. This is not to criticize these and others? work (Hubbard et al., 2001; Garot, 2004). It is rare that emotions are considered at all. Rather, it is our intention to further extend this conversation, and in a direction that takes into account research settings that are much more mundane than the examples referred to above.



We have all been members of a larger research team examining multiple aspects of change in Cokemaking Inc., part of a large steel works in Australia. Since the 1980s, the plant has undergone a series of culture change initiatives designed to improve performance. The fieldwork that provides the focus for this paper was conducted over a period of six years (1999-2005), and though the various projects have not all been ethnographic, the ethos guiding the research team has generally applied this sensibility.


Specifically regarding this paper, Simon spent an average of a day a week for eleven months in a section called Utilities. This is a section of Cokemaking that maintains and repairs the doors, ovens and other aspects of the batteries (Photograph 1). These are hundreds of tall narrow ovens, in which ground coal is cooked to produce coke, which is then sent off to the nearby steel-making blast furnaces. There are fourteen Specialists (operator level), six Technicians (projects, super-visory), and a manager in the section. Simon?s involvement in Utilities started with the manager, and most time was spent with the Technicians. It was in the second half of the research period that Simon spent more time with the Specialists. Karin?s research was conducted in several different parts of the steelworks as a whole. In the cokemaking division, she was a non-participant observer of meetings at shop floor, middle and senior management levels. She also carried out interviews in a project investigating leadership training across the steelworks. Because of multiple involvements, she often switched between projects, and did not form a long-term attachment to any one group within the plant. Richard was the principal investigator of the various projects and was involved throughout. He is well known in the plant and his relationship with the cokemaking plant manager, Garry, was a continuous and emotionally important thread linking several of the projects.

Photograph 1 - Specialists adjusting battery doors


Partly because of these different involvements and partly because of who we are, each of us reacted to being in the field in different ways and have recorded our thoughts on meeting and interacting with the people we have met. It is these observations from contemporaneous and retrospective field notes, plus data collected more generally from meeting notes, that provide the material for our analysis. In baring our emotive souls we are not wishing to claim methodological righteousness. Rather, there is an interesting general point that all researchers need to recognize in producing their analyses: the inherent covertness of the research act produces emotional dissonances (of disgust, embarrassment, disquiet, anger, etc.) and the identity work which dissipates them structure subsequent analyses and theoretical engagement.



Each of the following examples explores different aspects of the dissonances produced by the inherent covertness of research in the field. We split them into three types of emotional dissonance: those generated by displays of emotion, those provoked by specific others, and those provoked by reactions to generalized others.



We take the first examples from Karin?s fieldwork experiences. In reflecting on her experiences of observing a series of work redesign meetings she wrote: « Although I tried to be friendly, I found it difficult (impossible!) to do the matey, blokey stuff. I thought it would be too patently false to even try. The major emotion was discomfort, feeling out of place, like a spy or a parasite ». Karin did however « enjoy it, as it was a window into another world which was quite interesting ». In another set of off-site all-day personal development meetings with more senior men aimed at « progressing » their people skills and attitudes to change, she noted that, « they were sometimes quite emotional. I felt even more uncomfortable at these [meetings], because the group was smaller (making it even more difficult to be inconspicuous), and because, due to the occasional emotional intensity, being there felt voyeuristic ». In both situations, we see how mutual evaluations between researcher and researched generate emotional reactions (in this case discomfort) in the researcher.


This dissonance of discomfort became all the more apparent when we consider Karin?s response to the emotions displayed by others in the field, including her co-researcher Richard. Richard also attended the senior management personal development meetings, in which the managers were encouraged to « talk candidly about themselves —their strengths and weaknesses, etc. ». An exchange at one of these meetings is illuminating. Karin wrote, in a post-research reflection, that the plant manager, Garry, « was talking about his own strengths and weaknesses and Richard made a comment and became emotional. In my notes I have it recorded as: “Richard: Can I put in some strengths [of Garry?s] without appearing to grovel? Commitment to [the] underdog. Having confidence in people to do things” ». Karin didn?t contemporaneously note the emotionality of this interjection, which does appear in Richard?s own original notes: [Richard talking about his feelings about the plant manager] « some strengths, without appearing to be grovelling, it is the defence of the underdog, and the confidence (and then emotionally chokes up) ». Karin reflects on the contrast between her own and Richard?s notes: « I remember thinking “He?s more emotional than I thought he was”. I didn?t include “emotionally chokes up” or any similar comment in my notes for fear of embarrassing him. Likewise, fear of embarrassing our research subjects and a general delicacy/discomfort regarding displays of emotions (blame the Germanic upbringing) meant that I generally left these things out of my notes. I have to say, though, that I liked him for being emotional. (Being very confessional here) I?m quite emotional myself, but have been heavily socialized into not showing it; especially with people I don?t know well. This means that I appreciate and understand emotions in others, but I?m supposed to keep mine hidden […] and somehow it seems unseemly to take notes about them. »


Here we see discomfort at the disclosure of her own and others? emotions which leads in the second case to a retrospective acknowledgement of selective recording of data. The consequences of the first example are reasonably clear: Karin maintains the distance between herself and the respondents and offers little, certainly not herself. In the second example, the dissonance of discomfort is similar, but upon reflection Karin explains herself and dissipates the emotional dissonance via a process of identity work: she rationalizes the omission of emotionality in her account as a product of emotional repression in her upbringing. With respect to the covertness and irony of our position as researchers, are we to criticize Karin?s management of these emotional dissonances? No, but we need to recognize what can be learnt from the differences between researchers in this regard. We return to this later when we discuss the methodological implications of our analysis.



A different form of dissonance is at the root of the next example. It might also be broadly described as discomfort, and includes embarrassment, but as with Sennett and Cobb?s (1972) comment about Max, is directed toward specific research respondents and focuses on disgust, dislike, irritation, and so on.


Working at the plant is a rough, dirty and foul-mouthed affair (Photograph 2). It?s a man?s job, and some of the men lead lives very different from academics. One individual, Jack, who worked as a Specialist, presented particular challenges to Simon. He notes of his initial meeting: « What stuck out for me was the negative vibes from Jack », and talking to another worker he noted « how Jack is to me sullen, moody and not saying hello, and that he wouldn?t be my first choice for interviewing ». Here Simon is making decisions about the conduct of the research based on emotional reactions to individuals, and though numerous group conversations were had with Jack involved, he was not asked to do a one-to-one interview.


Jack is a thin, small and wiry individual with the unhealthy pallor that seemed only to confirm his oft-mentioned insomnia. Though swearing was de rigueur in the plant, Jack would use the F and C words regularly and venomously. Jack was perceived by Simon as being far more obnoxious than Max ever was. What was particularly disturbing was the way Jack seemed to take swaggering pride in his moonlighting activities of paying for sex and being paid as a driver for prostitutes. The veracity of this activity was confirmed to the extent that he talked openly about it and his fellow work mates said he wasn?t « bullshitting ». In the Simon?s field notes of a conversation with Jack and his teammates, Pauly and Dimitri, this was noted: « He said he earns $300 per night. Pauly asked if he got to “sample the goods”. Jack smirked (yes that was it, smirked) and said of course. He seemed a bit proud, in that I-dabble-in-things-you-wouldn?t-understand type of way. I?m sure he?s right. Jack is a very odious little man who swears a great deal and has many a grudge with the world. He has a Neanderthal appearance ».

Photograph 2 - “Hard Yakka”: specialists at work


On another occasion up on the Batteries talking to Jack, Pauly and Dimitri, the conversation turned to their aspirations in life and broader issues of class. Jack was fairly quiet and explained that he wasn?t really interested in aspiring to anything. Simon?s notes show that Jack explained that « with him “you get what you see” —it might not be polically correct to say it, but in his case I think this is right! » In the same conversation Jack also explained that: « he?d “fucked some woman” recently, and not paid for it. “She had been sore!” he said. No real embarrassment from Pauly and Dimitri [at this], but I sensed that they were having a bit of sport with him ». Immediately following these notes Simon made the following comments referring to a meatpacking job early in his working life: « I remember from the butchery that sexual conquest was much discussed and disputed currency. Shaun [a butcher] used to brag a lot and get the piss taken for it ». Simon?s embarrassment at this talk provoked memories of analogous work situations: identity work in action.


For Simon, in the case of Jack, dealing with and dissipating emotional dissonances of disgust, embarrassment, and fear is achieved via reference to his narrative identity. But this identity work is not sufficient to bridge the two worlds of Jack and Simon or the others. The partial fiction of mutuality is exploded, and what remains is faint unease over the lack of authenticity —the covertness— in the exchange.



As well as reacting to specific people and situations in the plant, we also reacted emotionally to generalized others. Generalized others are internalised representations of collective attitudes, norms and expectations within which, and against which, we place ourselves socially and monitor and modify our own behaviour (Mead, 1934: 154-155). As we noted earlier, ethnographers must straddle diverse social worlds. The notion of generalized others enables us to capture the dissonances generated by our membership in social collectives —class, gender, occupational groupings— that were quite different to those of our research subjects.


For Simon, a key source of emotional dissonance was his ambiguous feelings of anxiety regarding his own class origins. He spent much of his twenties struggling to run a creative enterprise and working in menial jobs, including for the London Underground which has many similarities with the steel works. Simon noted three quarters of the way through the research period, in relation to his disappointment about not getting as close to the Specialists as he would have liked: « I guess part of my reluctance stems from the difficulty I have always faced at Cokemaking, namely the similarity to the people and feel to my own work and class experiences at London Underground. […] The feeling of not wanting to touch and engage with the working blokes has pervaded this research. I was always outside of that group of people at the Underground and still feel separate. […] I?ve done my best, but in terms of the research I could have been more entrepreneurial with the workers than I have. It?s an issue of identity I guess. I don?t want to go too close because I?ve really been there myself and for real. I know the feeling I had when working in a shit job. You don?t like it. […] You don?t feel particularly dignified.


Simon?s attitude also had its consequences. Reflecting on the workers? sometimes unpredictable and childish behavior he noted that: « I definitely feel that I have avoided dealing with these guys and am aware that I feel threatened by some of their behaviors. For instance in the sports science place whilst [listening to a talk] about the sports bra [on a visit to the University Sport Science Centre], Andy made some comment about nipples which was, typical, predictable and a bit embarrassing. Whilst it didn?t bother me, I think that there is a fear there for me ».


Despite his unease about these manifestations of working class culture, Simon would use his familiarity with the experience of manual work to establish rapport. He would find himself thinking about how he sounded, being conscious of emphasizing his London working class accent. He would also swear more than normal: that would often spill over into work meetings and the home. He was conscious of trading on the past identity of his working background, and was perceived by many on the batteries as being a decent and acceptable bloke (Photograph 3). An example of this can be seen when Dimitri, a Specialist, mentioned his view of « University types » and middle class people in general following the visit to the University: « He talked about beards and glasses [seen in photographs of science academics in University corridors] and said “you just knew what they?d be like: cocksuckers”. I challenged him on this and said that “this is what I?m like, living in Kiama and so forth” [small middle class town]. But he said that I?m “different I guess and you can relate to us guys”. The fact that I?d worked in the Underground went in my favor ».

Photograph 3 - Simon (left) playing cards with technicians


Karin also selected and publicized aspects of her identity in an attempt to decrease the gulf between herself as a « university type » and the men she was researching. In retrospective notes she wrote: « Unlike Simon and Richard, I grew up in the local area during the time that the steelworks was a major employer. I was always aware of lots of people working at the steelworks. When I was a university student (early 70s) many of the male students, including my then boyfriend (now husband), worked at the steelworks during university holidays. I had some steelworks connections to talk to the workers about —that is, I could try and present myself as a local “girl”. Class was not the dominant issue. Being a daughter of migrants is fairly typical —the differences between me and the blokes at the steelworks were not about class but about level of education and, of course, gender ».


For researchers, drawing on the « generalized other » to try and be like them can only be a conceit, because the researcher is if nothing else a « walking display case, of the sort of life-chances » (Geertz, 1968: 149) they will not achieve. In the steel works, the disparities are ambiguously defined, for in terms of salaries, for instance, the researchers would typically earn less than the Specialists. Thus for many the presence of a working class or migrant success story will not necessarily inspire, though academic work and/or education is nevertheless held in high regard. For example, one Specialist commented to Simon, after a brief exchange of career stories, that he thought the researcher had done well since leaving London Underground. The moral ironies of these engagements are apparent. Because different interpretations of similarity, difference, and success are possible, researchers are left with feelings of covertness and unease in their constructions of self.


With our stereotypical but in all three cases non-existent « beards and glasses », there were many times when we, sometimes unconsciously, worked on our emotions in ways that reinforced particular readings of ourselves as academics. This is interesting for a number of reasons, particularly with regards to how the identity work of researchers affects analyses. Thus Karin notes retrospectively: « Overall, I think my approach to the ethnographic research was in keeping with a traditional approach —detached and objective, at least as far as the notes and my degree of social involvement are concerned. Though I felt a lot of emotion, I didn?t write about it or show or talk about it (also traditional). This was not a conscious choice, but more an outcome of the context and my personality. As I result, I didn?t get as much rich data as you two, who were more socially involved and emotionally expressive (at least it seems so to me). »


Here the construction of an academic identity leads to specific and tangible impacts on the nature of the data collected. The research involved Simon dealing with feelings and working through issues of working class/middle class identity in constructing his academic self. In simplistic terms the consequence of this has been proportionally less data collected from the workers (though this is not unusual in organizational research). For Karin, potential nuances in the emotional dynamics of the fieldwork may not have been emphasized, because of the construction of an academic identity of scholarly detachment.


Gender was a significant issue for Karin. It affected the data that were available for collection, as the men were seemingly more self-conscious when being observed by a female researcher. She noted retrospectively that: « I think my presence inhibited [the managers]. They were polite and kind, but there were times when they wanted to be crude, and even though I tried to convey that I?m not prudish, they still held back. This was a general thing throughout my whole time at Cokemaking. Many of the guys feel uncomfortable swearing in front of women no matter how much you say you don?t mind. (I swear at home and with people I know well and who I know wouldn?t be offended, but I would never swear at the steelworks) ».


Note here the different strategies in relation to the generalized other of the workforce. Simon swore more than usual ( « a blokey thing ») while Karin felt compelled to abide by traditional gender stereotypes and refrained from swearing at all. These gendered influences on the conduct of the research also had mixed effects, as noted by Karin: « Being female had its occasional advantages, and these [management] guys were probably not as uncomfortable with an educated female as I suspect some of the shop floor guys were. [One manager] in particular liked to lay on the charm and likes a female audience. I appreciated the attention and used him as source of information. The overall effect of all this is that I didn?t make friends with any of the employees, like Simon and Richard. But then again, I?m not particularly sociable at the best of times. Trying to make new friends for the purposes of fitting in or gathering information would have felt false, uncomfortable and unethical, and I would have been crap at it anyway ».


Here Karin?s structurally gendered identity allied to her aforementioned traditionally distant sense of herself as an academic, had implications for the conduct of the research and the extent of the emotional bond she developed with respondents. Did Richard or Simon feel any more false for making friends in order to gather information? These and other questions underlie the general point we are making here about the inherent irony and covertness of research caused by reactions to emotional dissonance, and the variable identity work used to dissipate it. Individual researchers cannot avoid facing the « not altogether clear (…) nature of sincerity and insincerity, genuineness and hypocrisy, honesty and self-deception » (Geertz, 1968: 155).


Thus there is also a structural dimension to these examples of class, gender and work identity. As researchers, not all the presentations of self described above can be conceptualized « as “acting” or as issues of “self” or “identity” » (Garot, 2004: 738), but can be seen as part of the responses to structural aspects of socially ascribed public narratives about roles, such as woman in a man?s world and/or academic researcher and the particular emotional stances evoked by these roles (Warren, 2001). The dissonances generated by doing research in different social worlds would vary according to who the researched are. Had we been conducting investigations in a university, or some other locale in which the researched were highly educated, the dissonances would have been different. Thus, the inherent moral ambiguities of the researcher are, to some extent, emotionally neutralized by the structural expectations of these roles. We are let off the hook, and substitute or merge our emotional issues with traditional gender stereotypes and/or the inherent power, authority and distance of the researcher role.



We have shown how the inherent irony and covertness of the research situation creates emotional dissonances that we try to dissipate through different types of identity work, some aimed at doing distance, and others aimed at doing closeness (Emerson and Pollner, 2001). We are left however with a question that each and every research paper has to answer: so what?


At this point it would be rather foolish of us to adopt a po-faced position of sage-like superiority and start preaching an un-ironic position by stipulating a list of how-to do or avoid this or that, or to risk provoking the response of “yeah, right, who are you kidding?”, to unrealistic and over-idealized notions of research practice which often accompany methodological papers. We will however attempt to conclude by pointing out the utility of our reflections on emotional dissonance and identity work in the field.


The first point is that although honesty may be honored as the best policy, we have seen that it?s not always easy to know exactly if we are being honest and sincere in the face of covertness. We operate in two worlds and have conflicting interests. As researchers and individuals, we have varying dissonance dissipating strategies and tolerances. We should not ignore this. In this regard, and as we suggested earlier, Geertz?s (1968) imputation of the moral ironies of fieldwork is useful in explaining and furthering analyses. Our role as academics is predicated on us being able to produce useful knowledge. As a research group our project has had more or less explicit emancipatory as well as normative economic aims. That is, we want to improve the lot of people in organizations in some way. As with Geertz in Java, however, the distance between our creating useful knowledge and being able to change organizations for the better of all (as opposed to managers in particular) is inherently limited. This is the partial fiction we use —at times explicitly in justifying the need for the research to respondents— to construct a robust sense of ourselves as academics. Because the partial fictions that emerge in the field are not falsehoods and therefore are not attempts at deception, the identity work that researchers need to do in order to dissipate the dissonances can potentially generate difficult questions for the research, and force the researcher to accommodate irony and different, sometimes contradictory, perspectives. This can benefit critical reflection on the research process and on substantive issues.


For us recognition of these ironies via our discussions and particularly in preparing this paper has undoubtedly had a positive effect on the quality of our forthcoming research analysis elsewhere. Identifying the ironies of our own engagement has helped in understanding and theorizing the ironic engagement of other actors in the plant (Badham, 2006). One implication of our work here would be to exhort ethnographers to engage in research group discussions (see also Hubbard et al., 2001: 132-136) addressing these issues. Or if you work in the field on your own, discuss your work with other colleagues. On the one hand, reflections on your own emotions and engagements in the field may assist in the analysis of the feelings, actions and identities of those others you are studying. Also, in addition to improving your analysis by incorporating the emotional context of the research setting (Lerum, 2001; Blee, 1998), such discussions may also help to dissipate the effects of emotional dissonance. The ironies will not be resolved, but the identity work you need to do might be made more self-conscious, less privately stressful, and support greater collaborative understanding and bonding.


Moreover, when an earlier version of this paper was discussed at the 2006 European Group for Organizational Studies Colloquium (at a sub stream which was focussed on ethnography) junior researchers found the presentation encouraging because they were relieved to hear that they were not alone in experiencing these negative and morally messy feelings. And, for the more senior ethnographers in the session, there was a more knowing recognition (but all the same normally implicit and relegated to craft-knowledge) of the realities of real world research. Thus, one practical implication of this paper would be for those who advise younger researchers (either face to face, or in published work) to be more open and reflective about the moral ambiguities of research.


The second methodological point follows the theme of ambiguity, and warns against the reassuring notion that baring all will dispose of the conundrums it raises. We grant that reflexive practices —in which we ask « how does who I am, who I have been, who I think I am, and how I feel affect data collection and analysis? » (Pillow, 2003: 176)— have become increasingly common in qualitative social inquiry. However, researchers often practice what Pillow (2003: 187) has called « comfortable reflexivity ». That is, we assume that by confessing our identities and emotions we make our knowledge claims more transparent and reliable (Lerum, 2001), and that by allowing informants to speak for themselves we help address asymmetries in power. However, there are dangers, as Pillow (2003: 186) notes: « Self-reflexivity can perform a modernist seduction —promising release from your tension, voyeurism, ethnocentrism— a release from your discomfort with representation through a transcendent clarity ».


While, as Geertz (1968) noted, some release from tension is necessary if work is to proceed at all, Pillow warns us not to seek solace too readily in the illusion that confessing emotions and revealing identity work will somehow allow us to produce truly emancipatory knowledge. She advocates instead the practice of uncomfortable reflexivity —a continuous grappling with the difficult questions, an « often uncomfortable task of leaving what is unfamiliar, unfamiliar » (Pillow, 2003: 177). In our case, we are unable to conceive of ethnographic methods that would remove the moral ambiguities of the research relationship. The researchers? position, and their reflections, remain an ironic one —dissipation occurs, at least in part, as a result of a higher road to the acceptance (Gray, 1960) of emotional dissonance rather than its removal or denial.


A third, broader point relates to the variability of the ethnographic experience. Lerum (2001) and Blee (1998) have clearly reflected on their emotional entanglements from two very different research contexts — one that provoked positive feelings and the other which provoked negative feelings in the researcher. By comparing their accounts, and the research advice that has emerged out of them, we can see that there is no one-size-fits-all way of responding to, and dealing with, the tensions associated with ethnographic fieldwork. While emotional engagement may be beneficial in some situations, distance may feel more appropriate in others.


Even our own humdrum research site, though ostensibly not as emotionally charged as the situations encountered by Lerum and Blee, was full of emotional dissonance. Emotions are part of everyday life and should be part of everyday ethnography. There was even considerable variability within the micro-situations through which the research work was conducted. The experience of Simon, by virtue of his more concentrated involvement with a well-defined sector of the plant, and the discourses of identity that were available to him, was quite different to that of Karin. The tensions they experienced were different as were the strategies they used to dissipate them. While Simon took extensive contemporaneous notes on, and reflected at length on his emotional responses —to the extent of initiating and taking the lead in writing this paper— Karin tended to excise emotions from her note-taking, preferring (at least initially) to suppress or ignore them. While neither method is necessarily superior to the other, they produce different sorts of accounts. Each has advantages and disadvantages, which would vary according to situations. By focusing on his or her own emotionality, a researcher may generate valuable data about workplace relationships and power dynamics. However, s/he may also become self-indulgent and self-absorbed, thereby missing alternative readings of situations. On the other hand, researchers who resist acknowledging emotions, in themselves and others, may miss valuable emotion data and risk producing less than honest accounts. On the plus side, concentrating on the objective aspects of encounters may prevent researchers being overwhelmed by their own feelings and those of others, and may contribute to harmonious relationships in organizations where emotional outbursts (and researchers? accounts of them) may have negative consequences. We aim to continue exploring these differences, and hope you do too.


The final point picks up on the moral and political consequences of acknowledging feelings about individuals and situations. We may not like or appreciate all our respondents: it is alright to find Jack, or Max, or anyone else obnoxious. When reading ethnographic work one is often left feeling that the respondent is to be positively accepted for who he or she is, and that the ethnographer is emotionally hollow, bland or absent. This narrative strategy in published work denies the emotional realities of life as much as it ignores the covertness we have discussed here. We are aware today that all research positions, including objectivity, are social constructions, and as such, have moral and political consequences. As the journalist Hunter S. Thompson —the inspiration behind our title– once remarked, « Objective journalism is one of the main reasons American politics has been allowed to be so corrupt for so long. […] You can?t be objective about Nixon » (Washington Post, 2005). Perhaps, like Thompson, we should be more prepared to be open and reflexive about our fears and loathings as part of direct and acknowledged engagement with forms of gonzo (emotional/subjective) research. While eschewing the seductive idea that acknowledging our feelings and identity work allows us to claim some kind of moral high ground (Pillow, 2003), we can still hope that doing so (or perhaps sometimes deciding not to) will help us to enhance our understanding of ourselves, those in the field, and ways in which we can work (together or apart) on improving the worlds that we live in.


  • Badham, R. 2006 Mind (ing) the Gap: The Irony of Practice and the Practice of Irony, Second Organization Studies Summer Workshop, 15-16 June, Mykonos, Greece.
  • Badham, R., K. Garrety, V. Morrigan, M. Zanko, and P. Dawson 2003 Designer Deviance: Enterprise and Deviance in Culture Change Programmes, Organization, 10: 4, 707- 730.
  • Badham, R. and K. Garrety 2004 Living in the Blender of Change: The Carnival of Control in a Culture of Culture, Tamara, 2: 4, 22-38.
  • Barley, S., and G. Kunda 1992 Design and Devotion: Surges of Rational and Normative Ideologies of Control in Managerial Discourse, Administrative Science Quarterly, 37: 3, 363- 399.
  • Blee, K. M. 1998 White-Knuckle Research: Emotional Dynamics in Fieldwork with Racist Activists, Qualitative Sociology, 21: 4, 381-399.
  • Bosma, H. A., and E. S. Kunnen (Eds.) 2001 Identity and Emotion: Development through Self-organization, New York: Cambridge University Press.
  • Carr, A. 1999 The Psychodynamics of Organization Change: Identity and the ?Reading? of Emotion and Emotionality in a Process of Change, Journal of Managerial Psychology, 14: 7/8, 573-585.
  • Czarniawska, B. 1998 A Narrative Approach to Organization Studies, London: Sage.
  • Down, S. 2006 Narratives of Enterprise: Crafting Entrepreneurial Self-identity in a Small Firm, Cheltenham: Elgar.
  • Emerson, R. M, and M. Pollner 2001 Constructing Participant/Observation Relations, in R. M. Emerson (Ed.), Contemporary Field Research: Perspectives and Formulations, 2nd ed., Prospect Heights: Waveland Press, 239-259.
  • Fine, G. A. 1993 Ten Lies of Ethnography: Moral Dilemmas of Field Research, Journal of Contemporary Ethnography, 22: 3, 267-294.
  • Garot, R. 2004 “You?re Not a Stone”: Emotional Sensitivity in a Bureaucratic Setting, Journal of Contemporary Ethnography, 33: 6, 735-766.
  • Garrety, K., R. Badham, V. Morrigan, W. D. Rifkin, and Zanko, M. 2003 The Use of Personality Typing in Organizational Change: Discourse, Emotions and the Reflexive Subject, Human Relations, 56: 2, 211-235.
  • Geertz, C. 1968 Thinking as a Moral Act: Ethical Dimensions of Anthropological Fieldwork in the New States, Antioch Review, 28: 2, 139-158.
  • Geertz, C. 1995 After the Fact: Two Countries, Four Decades, One Anthropologist, Cambridge, MA: Harvard University Press.
  • Giddens, A. 1991 Modernity and Self-Identity: Self and Society in the Late Modern Age, Cambridge: Polity Press.
  • Gray, J. C., 1960 Irony: A Practical Definition, College English, 21: 4, 220-222.
  • Harré, R., and G. Gillett 1994 The Discursive Mind, London: Sage.
  • Hochschild, A. R. 1983 The Managed Heart: Commercialisation of Human Feeling, Berkeley, CA: University of California Press.
  • Hubbard, G., K. Backett-Milburn, and D. Kemmer 2001 Working with Emotions: Issues for the Researcher in Fieldwork and Teamwork, International Journal of Social Research Methodology, 4: 2, 119-137.
  • Jasper, J. M. 2002 What the Devil Are Emotions?, Qualitative Sociology, 25: 1, 145-148.
  • Jenkins, R. 1996 Social Identity, London: Routledge.
  • Kärreman, D., and M. Alvesson 2001 Making Newsmakers: Conversational Identity at Work, Organization Studies, 22: 1, 59-89.
  • Keifer, T., and W. R. Müller 2003 Understanding Emotions in Organizational Change: the Role of Identity in Ongoing Change Processes, Academy of Management Conference, Seattle, August 1-6.
  • Kleinman, S., and M. A. Copp 1993 Emotions and Fieldwork, Newbury Park, CA: Sage.
  • Kunda, G. 1992 Engineering Culture: Control and Commitment in a High-Tech Corporation, Philadelphia, PA: Temple University Press.
  • Kunda, G., and J. Van Maanen 1999 Changing Scripts at Work: Managers and Professionals, The Annals of The American Academy of Political and Social Science, 561: 1, 64-80.
  • Lerum, K. 2001 Subjects of Desire: Academic Armor, Intimate Ethnography, and the Production of Critical Knowledge, Qualitative Inquiry, 7: 4, 466-483.
  • Lewis, M. 1993 Self-Conscious Emotions: Embarrassment, Pride, Shame and Guilt, in M. Lewis and J. M. Haviland (Eds.), Handbook of Emotions, New York: Guilford Press: 563-73.
  • MacIntyre, A. 1981 After Virtue: A Study in Moral Theory, London: Duckworth.
  • McLoughlin, I. P., R. J. Badham, and G. Palmer 2005 Cultures of Ambiguity: Design, Emergence and Ambivalence in the Introduction of Normative Control, Work Employment and Society, 19: 1, 67-89.
  • Mead, G. H. 1934 Mind, Self & Society from the Standpoint of a Social Behaviorist. Chicago, IL: University of Chicago Press.
  • Pillow, W. S. 2003 Confession, Catharsis, or Cure? Rethinking the Uses of Reflexivity as Methodological Power in Qualitative Research, International Journal of Qualitative Studies in Education, 16: 2, 175-196.
  • Scheff, T. J. 1988 Shame and Conformity: The Deference-Emotion System, American Sociological Review, 53: 3, 395-406.
  • Sennett, R., and J. Cobb 1972 The Hidden Injuries of Class, New York: Knopf.
  • Somers, M. R. 1994 The Narrative Constitution of Identity: A Relational and Network Approach, Theory and Society, 23: 5, 605-649.
  • Tangney, J. P. 1999 The Self-Conscious Emotions: Shame, Guilt, Embarrassment and Pride, in T. Dalgleish and M. J. Power (Eds.), Handbook of Cognition and Emotion, Chichester: Wiley, 541-568.
  • Turnbull, S. 2002 The Planned and Unintended Emotions Generated by a Corporate Change Program, Advances in Developing Human Resources, 4: 1, 22-38.
  • Van Maanen, J. 1988 Tales of the Field: On Writing Ethnography, London: University of Chicago Press.
  • Warren, C. A. B 2001 Gender and Fieldwork Relations, in R. M. Emerson (Ed.), Contemporary Field Research: Perspectives and Formulations, Prospect Heights: Waveland Press, 203-223.
  • Washington Post 2005 Hunter S. Thompson Dies at 67: ?Fear and Loathing? Writer Apparently Committed Suicide, Monday, February 21: A04.

Plan de l'article


Pour citer cet article

Down Simon, Garrety Karin, Badham Richard, « Fear and Loathing in the Field: Emotional Dissonance and Identity Work in Ethnographic Research », M@n@gement 3/2006 (Vol. 9) , p. 95-115
DOI : 10.3917/mana.093.0095.

Article précédent Pages 95 - 115 Article suivant
© 2010-2014
back to top