AS EMOTIONAL LABOUR
From Journal of Contemporary Ethnography, vol. 27, no. 3, (October 1998) ps. 346-347.
Emotional labour is what workers do with their feelings to comply with organizational role requirements. This article explores the concept in Professional organisations, examining the psychotherapeutic discourse of objectivity, neutrality, and care as feeling rules. Based on a study in a residential psychiatric facility in Israel, the authors found that counselors laboured to display aspired professional feelings despite the absence of memos, protocols, or training sessions. Who told them to do so? How did they know what to feel? The authors claim that therapeutic discourse constitutes professional feelings through the use of specific concepts and techniques. However, the term professional feelings disguises a complicated process of negotiation between different ideologies. The difference between two groups of counselors indicates that both scientific and intersubjective knowledge represent modes of emotional control. The authors claim, thus, that emotional labour in professional service organisations is the product of contested professional discourse.
"YOU'RE A FAT WHORE:" Ruth, a resident in a shelter for the rehabilitation of chronic mental patients, stated, bursting with laughter. Ann (all names have been changed), a third-year psychology student who worked there stood looking bewildered but calm and restrained. Later, she would tell us, "I was burning with anger. All I wanted to do was to hit Ruth. You know I have a problem with my weight" Yet, she turned very quietly to Ruth and asked her to leave. "I couldn't let her see my anger," she explained. Ann rationalised her calm reaction:
"I had to remind myself that Ruth is schizophrenic, that she cannot always control what she says. Do you remember Dr. Berger's analysis in one of our group sessions? He said that Ruth's curses are introjects of shame toward her own body and sexuality, and that defences of psychotic patients are often very weak. Recalling his words, I could remove myself from the situation. I even felt pity."
Ann was telling us about her emotional work as labour. She transmutes a negative feeling, such as anger, into pity through a process of "deep acting" (Hochschild 1983) by raising what she saw as professional claims of reason. Why did Ann choose to control her anger? Who told her to do so? How did she know what to feel? We shall argue that Ann's emotional conduct was controlled by her professional inclination and by her anticipation of becoming a therapist. Moreover, we shall claim that emotional work in professional service organisations, such as the shelter at which Ann works, is managed and operated by professional (in this case, psychotherapeutic) discourse. In our analysis of this discourse, we suggest that some workers submit to the authoritative norm of emotional control, while others produce counter narratives to the dominant standpoint of objectivity and neutrality. In other words, we aim to reveal the struggles and ideologies that structure the emotional management of professional workers.
Emotional labour (Hochschild 1983) is a practice often engaged in by people in the domain of work. When feelings are underplayed, overplayed, neutralised, or changed according to specific organizational feeling rules and in order to advance organizational goals, workers perform emotional labour. Thus, the concept of emotional labour refers to what workers do with their feelings in compliance with organizational role requirements. At Delta Airlines, specific feeling rules (Hochschild 1977) were directly connected to work standards. Anger, aggression, and resentment, even toward annoying passengers, were prohibited by strict rules and controlled by the selection of employees, training, and supervision. Workers went through periodic sessions to learn how to change private anger into public empathy, kindness, and smiling. Hochschild views such manipulation of feelings, aimed at giving the company a competitive edge, as a form of exploitation contributing to worker alienation and loss of emotional authenticity in the sense that the company, rather than the worker, provides the "right" interpretations to incidents and events during work. Managerial rather than private interpretations construct or constitute workers' feelings. Service workers such as flight attendants, tax collectors, or cashiers who use empathy, kindness, and smiling to encourage customer commitment and loyalty to the company lose touch with their private feelings, unless they resist the company's "natural" right to "possess" their emotions.
In a similar vein, Kunda's (1992) ethnography in a high-tech corporation opens a window to the design, development, and maintenance of a high-tech "emotional culture" engineered to achieve the "right mind-set" and the "appropriate gut reactions" (p. 93). In detailed descriptions, Kunda demonstrates how internalised commitment is constructed and how thoughts and feelings are enlisted to serve the company's interests. Specific rules (do what is "right" in each situation) and rituals (presentations, messages, manuals, and speeches) constitute not only the definitions of membership, but also of the employee's self - "that ineffable source of subjective experience" (p. 11). Corporate culture, Van Maanen and Kunda (1989) argue, claims our hearts and souls as well as our minds and bodies.
Emotional work in organisations such as Delta (Hochschild 1983), the high-tech corporation (Kunda 1992), Mary Kay (Ash 1984), and Disneyland (Van Maanen and Kunda 1989) produce feelings of warmth, affection, enthusiasm, and commitment. These emotions are programmed by managers and practised by employees in order to achieve organizational objectives. Workers who successfully internalise their membership roles, conform to the company's meaning of good work, invest themselves in corporate culture, and use their personalities as a vocation (Hunter 1993) are rewarded with money and status.
What happens, then, in professional service organisations that are oriented toward social and human welfare, such as hospitals, psychiatric institutions, correctional agencies, social agencies, or nursing homes? These helping organisations are dominated by professional associations that "serve as the ultimate authorities on those personal, social, economic, cultural and political affairs which their body of knowledge and skill address" (Freidson 1994, 33; see also Abbott 1988). Freidson (1994) argues that, in professional service organisations, knowledge is dominated by professional jurisdiction that monopolises the definition of work; unlike industrial organisations, professional service organisations 'remove or withhold from the hands of management authority to create and direct the substance, the performance and even the goals of the work itself' (p. 62). In Bourdieu and Wacquant's (1992) terms, individual performers in these organisations (the practitioners) are the agents of a specific field of knowledge. Therapists, for example, represent in their practice the dispositions of the field of psychotherapeutic discourse.
Studies of medical and psychiatric institutions (Abbott 1988; Freidson 1988; Smith 1992; Smith and Kleinman 1989) show that, like Hochschild's flight attendants, employees in the helping professions manage their feelings as part of the work requirements. Confronted on a daily basis with strong emotions of pain, suffering, aggression, bewilderment, and neediness, professionals (doctors, nurses, social workers, psychiatrists, psychologists) control emotions of anger, anxiety, and dislike and express empathy, calmness, optimism, and kindness. Are they following specific feeling rules? Do they subjugate themselves to those rules, or do they negotiate with them? How does such a struggle affect their professional selves? Can we conceive their emotion management as labour? Do they use their emotions as service? In other words, do professionals, such as therapists, use professional feelings as their own?
To answer these questions, we studied the processes through which psychology students managed their emotions in a psychiatric facility in Israel. We focused on the ways in which students dealt with their own emotionality in relation to the professional ideology of emotional self-control. Many studies take for granted that professionals, and therapists in particular, "manage their feelings toward clients" (Hardesty 1987, 247; see also Meissner 1996). Such studies, however, rarely specify how these processes of self-control are achieved. What are the struggles and conflicts involved in these processes? How do people with power over the emotions of others (Thoits 1996) use tacit methods to gain control over themselves? Notwithstanding the importance of calmness, courtesy, and empathy, we believe that emotion management as an issue of professional identity deserves further distinction in order to demystify the processes by which therapists reclaim emotional control. A need to critically examine the power and limits of the professional discourse of emotional control motivated this study at the shelter for psychiatric residents.
The shelter is a residential psychiatric facility in Israel. Its three-story building is surrounded by a small stone yard opening onto the gardens of the county mental hospital on one side and a large parking lot on the other. There is nothing particularly pleasing about the building or its locale, but the atmosphere inside holds the promise that new and innovative programs often have. The shelter is affiliated with the county mental hospital but is a financially autonomous unit supported by the Department of Health.
In the course of one year (thirty hours each week), we conducted a study, based on participant observation and in-depth interviews, exploring patterns of emotional interaction between counselors and residents. Particularly, we were interested in the reactions and emotional expressions of the counselors. These were third-year students of psychology who volunteered to work with the residents as rehabilitation role models or, in the jargon of the institution, as social guides. At the time of our fieldwork, the shelter housed about twenty psychiatric residents, one senior psychologist, two social workers, five nurses, a service manager, a secretary, and sixteen students of psychology employed as counselors. Turnover among the counselors was high. Therefore, we decided to interview only twelve students who had worked in the shelter for at least six months.
We focused our study on this group because the counselors were the largest group of employees, they carried the burden of service and care, and they were responsible for overseeing the residents' daily routines. Most important, the decision to concentrate on their emotional work was based on the contention that psychology students, who are highly motivated to become therapists but are still novices, experience the emotional demands of their profession more intensely and consciously. At the early stages of socialisation, processes of emotion management are more salient and transparent, whereas expert therapists often manage their emotions automatically, to the extent that emotional labour becomes "natural" and is forgotten and the boundaries between private and professional feelings blur.
The rehabilitation program at the shelter relied on a milieu therapeutic structure aimed specifically at the "normalisation' of the residents. The overall goal for each participant was maximum socialisation with minimal medication. The program espoused functional independence limited only by the residents' ability to respond appropriately. The rehabilitation program strove to reintegrate participants into the community, despite their symptoms, by teaching them social skills and self care. They were expected to clean their rooms and various public areas in and around the shelter, get their own food from the kitchen, wash their laundry, and perform minor repairs. In addition, the residents worked a few hours in the community and were paid by the hour. They participated in various learning and recreational activities, such as news analysis, money management, hygiene, cosmetics, sports, creative writing, and other expressive activities in order to enhance their interpersonal skills.
These activities were guided, mediated, and supervised by the counselors (the psychology students), who conveyed a youthful spirit and earnestness. According to their job description, the counselors were to accompany residents to their workplaces, teach them interpersonal and social skills by taking them on trips to the local mall, run workshops, supervise the residents' personal hygiene and eating habits, and mediate between residents and other professional workers at the shelter. Unofficially and against explicit policy, some students initiated therapeutic - or what they believed to be therapeutic - interactions.
The students in the shelter were employed primarily as social guides and not as therapists. Nonetheless, they worked with mental patients, and often, the students took this opportunity to "play the therapist." For them, it was an opportunity to "be close to the profession," to "get experience," and to "get credit" for their future training. Obviously, during their work as social guides, they encountered strong emotional upheavals of anger, love, fear, guilt, empathy, or helplessness, as they were working for the first time in a mental institution and with mental patients. Strong emotions and awareness of emotional reactions characterised their work. Yet, never during our study did the students participate in training sessions or workshops organised to discuss and exercise feeling-management practices.
Similarly, we could not find any written memos or protocols that laid out rules of emotional display and expression. Students frequently compared performance and emotions among themselves to help each other and to socially validate their own feelings, but they were not guided in how to suppress outbursts or how to show empathy. From time to time, they received short notices from the management reminding them to attend staff meetings and to be prompt. However, during these meetings, only brief reports on specific residents were discussed, and some procedural business was carried out.
While the reception of a new resident was carefully planned in detail, both in written protocols and in practice (for three weeks the new participant was closely accompanied to teach him or her the habits, behaviors, schedules, and expectations of the place), new counselors were superficially briefed about routines and tasks. Nowhere in the shelter could we find a complete or even partial description of what counselors ought to do and how they should act or feel.
These observations correspond with Smith and Kleinman's (1989) conclusions that stressful emotions related to the treatment of the human body, which are held by medical students, were never recognised institutionally or acknowledged openly among the students. There were no managerial rules or pressures to control strong emotions. Nonetheless, medical students laboured individually to manage affective neutrality by using various strategies of control.
In light of the heavy emotional demands in the shelter, on one hand, and the lack of organizational rules, on the other, we were not surprised to find the students complaining about the lack of definitions regarding the "right" emotional display during work. Jacob, one of the counselors, recalled, 'When I came to the shelter, no one told me what I should be, they just told me to be. Even our name was problematic. Who are we? Therapists? Therapeutic guides?" Lilly, another counsellor, added,
"At first I was very confused, I didn't know what to do in various situations. With time, I stopped being troubled by my display of feelings. I learned that. But inside me, I didn't. I don't know if over-identification or anger are feelings which people in such roles usually experience. When you have no one to consult, you just do your best."
Lilly, like Jacob and most other students, believed that the therapeutic interaction dictates a specific emotional state and a display of emotions that are endemic to the therapeutic role. She believed in a "professional way of feeling." This belief partly explained the distress and confusion some students expressed as a result of receiving no training or instructions about "the proper way to feel," 'how to control emotions," "what to do with anger, disgust and love,' or "how to show empathy even when you don't feel it."
At early stages of professional socialisation, the need to control uncertainties (Light 1980) generates a need for regulation. In Lilly's words,
"Each of us saw things from his or her point of view. We are not professionals, and we did not relate to each other as such. So, I had my opinion, someone else had his, and no one had the truth. If only there were someone professional, from above, with more experience ... things would have been different. Consultation with someone professional would have helped."
Lilly was upset about having to rely on intuition alone:
"I felt rage, and then doubts as to whether I had chosen the right occupation. Do I fit in? Why do I need all this? What have I done? At first, I used to tell myself: "OK, I am not going to be overly sensitive." I wouldn't let things shake me. Then I became angry about not having someone to consult. When I understood that things are not going to change for the better, I felt helpless. Work became meaningless."
Like others, she felt that professional feedback in the form of supervision would have given her access to the truth - to a unified knowledge without which work was meaningless. Lilly's anxiety changed to anger because she was not receiving what she hoped to get out of her work at the shelter: 'real" knowledge. This did not mean school material and theories, which Lilly could read in books; for her, as for others, it meant objectivity gained by training and mastery of her feelings.' Lilly's anger is understood if we recognise her motivation: it reflects her need to be trained as a therapist in a place that defined her role as a social guide.
Emotional labour requires training and practice. The students in the shelter believed that supervision would unlock the secrets of the profession and cause the practice of professional feelings to become habitual. These ideas were rooted in their vague knowledge of the discipline and were based on what they had learned, heard, and imagined. After three years of studying psychology, they hoped for "real" training in the form of supervision, and they were extremely disappointed and annoyed by its absence.
Although an essential goal of supervision is to secure and monitor clients' care, one of the consequences of supervision is that trainees are socialised into a profession and develop a sense of professional identity (Bernard and Goodyear 1992, 7). The role of the supervisor is to "help the supervises to examine aspects of his or her behavior, thoughts and feelings that are stimulated by a client, particularly when these may act as barriers to the work with the client" (p. 5). In other words, supervision creates a competent practitioner who learns to react to clients in line with the rules and methods known to the field. To put it more strongly, supervision is the place where discourse is realised as practice and where professional authority is exercised. Experts who represent the logic of therapeutic discourse initiate beginners into the field not only by shaping their knowledge, skills, and attitudes, but also by reconstructing their emotions. This perspective of emotional "taming" is hardly discussed in the literature; nor was it recognised by the students as problematic. They wanted answers to troubling questions: what to feel - what is right and what is wrong, what to show during work with patients, and how to build the right emotions. For them, supervision was the ideal solution to unsettling feelings. In fact, it was the only answer they could conceive of to solve the problem of uncertainty and confusion.
Doing one's best, as Lilly put it, signifies a rhetoric of control. Students of psychology adopt a rhetoric of emotional control early in their schooling. They understand that members of the therapeutic profession are not people with 'regular' feelings. They are professionals, and as such they ought to display professional feelings. Too much affection or too much anger, the counselors claimed, is inappropriate. Control was a key word among them.
This in itself is not unusual. Various groups talk about control as a way to assert and exercise power (Lutz and Abu-Lughod 1990). However, therapists, we claim, identify the concept of control with professional identity. The director of the shelter, Dr. Berger, believed that the good clinician is a professional. His typical advice to students was to treat residents from "a professional rather than a personal point of view." The distinction between the personal and the professional represent the essence of the therapeutic role. The personal, he claimed, is to experience raw feelings, whereas the professional way is to perceive the resident's provocative behavior (e.g., cursing) as representing his or her psychopathology:
"If you cannot see a resident through the eyes of the clinician, you cannot help him, because then your reaction resembles the reaction of a non-professional who responds with raw feelings. Raw feelings impair professional intervention. Raw feelings are the most dangerous enemy to the good clinician. They keep him or her from using all the knowledge, training and skill which we received."
This belief was shared by the students in the shelter, who often voiced the logic that anger, hate, or disgust belong to the realm of the private, or the wild, and are therefore illegitimate emotions. This distinction between professional knowledge and "raw feelings" suggests adoption of the standard medical model (Smith and Kleinman 1989). The view that raw feelings are against the rules of good therapy derives from the wide consensus to keep the therapist's feelings under control or outside of the room by objectifying techniques that draw the line between the clinician and the non-professional. The definition of professionalism as the skill to react emotionally to patients through recognition of their pathology, or in a particular way (Greenberg 1996), rather than with spontaneity, suggests that emotional labour is a necessary part of the work that therapists do.
Notwithstanding this distinction between the professional and the personal, Dr. Berger seemed to be unaware of the students' need to learn the process by which affective neutrality (Smith and Kleinman 1989) is acquired. In fact, he insisted on not teaching them, because the good clinician, he believed, internalises the codes and secrets of emotional control through experience. In Dr. Berger's view, emotional competence is gained through a long process of learning and practice, and emotional knowledge is part of becoming a professional. It was therefore his belief that each practitioner would eventually find his or her own techniques for controlling personal feelings:
"Experience helps to internalise a sense of competence and emotional security which derives from knowledge and practice. Professional therapists learn, with experience, to cope with their own anxiety. They can anticipate patient's reactions and therefore are usually not surprised or frightened. The students are still young in a professional sense; they are not yet secure in their reactions or their competence to do what is necessary."
Some of the counselors, particularly those who strongly envisioned psychotherapy as their future career, did attempt to control their emotions in what they viewed as a professional manner. Jacob explained, "It is, in my view, illegitimate to feel anger or hatred. Residents are like foreigners in our world. It's hard on them, their dynamic processes are less controlled, their behavior is less intentional, they are less rational." When he felt angry he occasionally talked to himself about "what is right and what is not." He labelled his mode of thinking "intellectualisation," but what he was doing was a version of what Hochschild cable deep acting. By perceiving the residents as 'ill,' 'not responsible,' or 'not logical,' he was able to turn what he believed to be professionally illegitimate emotions into therapeutic feeling such as empathy and, by such cognitive manipulation of th emotions, assert his control. This point becomes clearer when Jacob fails to perceive a particular resident as mentally ill:
"I used to scream at Joel. I couldn't see him as a psychiatric patient. I used to treat him as a regular person, despite his manipulations. I didn't inhibit my feelings toward him. With him I gave myself all the freedom in the world to scream, to be angry, to fight with him, because as far as I was concerned, he was reacting like any normal person."
Jacob turned Joel into a person like himself. With this change of judgement, his field of emotional dispositions changed. He was then himself a regular person, not in the role of social guide or counselor. He could be out of control, 'natural.' Somewhat differently phrased, but still presenting a similar theme, Ann, another counselor, claimed,
"I can be angry at a nagging resident if I don't believe him or her. There are some residents who are less ill in my view. They are more aware of the things happening to them. When they nag and I know they can control themselves, I get very angry. But when they are really helpless, even when they nag, I feel pain for them, I can understand them."
Both Jacob and Ann controlled their emotions with an inner talk about illness. The thought of how seriously ill certain residents are constituted their emotional tendencies and mediated the management of display.
COMMITMENT TO THE PROFESSIONAL IDENTITY
Most students who worked as counselors were strongly motivated to becoming therapists and hoped that the work in the shelter would help them get into graduate school in clinical psychology because it demonstrated their commitment to and serious intentions regarding the field of therapy. Some volunteered in order to try out the field. Thus, even though the shelter was not a therapy-oriented rehabilitation center, and intervention was mostly based on social training, for the students the work there was an opportunity to "play therapist' and they resisted such limitations. In the words of one participant, "One day a notice was circulated prohibiting students from therapeutic relations with patients. That's good for the books, but unreal for us. If I cannot talk with a resident about his suffering, what am I doing here?"
We can also learn about the inclination and motivation of the counselors from the daily reports they wrote as part of their job. The reports were written in a special notebook and used by the students to instruct each other. There were comments on routine and unusual events that occurred during their shifts ("Rebecca was absent the whole afternoon," "Please note that Tova forgot her medication," "Zvi needs clean clothes") and some references to clinical interpretations and diagnoses ("Levi feels very tense and nervous," "Shlomo showed paranoid delusions; all morning he isolated himself.") In addition, students regularly added personal comments, drawings, jokes, and questions. These personal missives reflected their moods, frustrations, and struggles, revealing a strong need for coherence and professional guidance. They were also a common vehicle to mediate disagreements and conflicts.
Commitment to professional identity was not equally strong among all the students. Some clearly anticipated becoming therapists in the future; others were unsure. Students who strongly identified with the therapeutic role were more troubled by issues of emotional control and articulated more strongly the wish to be supervised by a professional psychologist.
Noah, who was recently accepted into a graduate program in clinical psychology, gave us a good clue as to how students with strong professional inclinations manage their feelings. From the moment he approached the shelter, he said, a transformation of feelings began. As he entered the door, he was already a different person:
"When I get to work, I immediately enter the shoes of a very defined role. Noah at home is not Noah at work. When I am at work, I don't allow myself to express emotions. I can show some empathy, but even this is not necessarily true; I act.
In addition, Noah tried ways to "universalise [his] role as a psychologist beyond the specific place of work." By reading literature on abnormal and developmental psychology, he came to the conclusion that "pathological reactions are disconnected from specific therapist-patient interaction." This conclusion helped him 'treat symptoms in an objective way." By being "objective," which he uses interchangeably with being rational, Noah meant that he can absorb insults and attacks from residents, without projecting back anger or anxiety, and maintain a "poker face," which he believed was the image of a professional psychologist. Just like the Shetland Islanders who watched "the visitor drop whatever expression he was manifesting and replace it with a sociable one just before reaching the door" (Goffman 1959, 8), Noah replaced whatever he was feeling with a balanced and detached emotional demeanour. Unlike the Shetlanders, however, Noah was practising not cultural habits, but a specific professional self, or what he thought to be so.
The commitment to professional behavior and emotions was perhaps most salient in Jacob's rhetoric. Jacob experimented dangerously with clinical situations, in clear defiance of accepted shelter policy. He was also the most vocal in his complaints of not receiving enough training, guidance, or supervision and of not having enough talks, discussions, or workshops. He said, "in order to protect your feelings, you must be connected to some social standard in order to have a reference point. You must comply with a standard and react accordingly." Not knowing what to do, Jacob sought his "social standard" in the books. Referring to a specific case, he explained,
"I read in order to ease my feelings. I went to the library and read about borderline cases: what does it mean, how do you reach a borderliner, what do they feel and how do I react to that? I wanted to understand the case objectively. It gave me meaning in the emotional sense."
Jacob talked continuously about his need to be rational and to control his feelings. 'Who said you should control your feelings?' we asked. "No one in the shelter," he answered. "I think it's part of the myth of what it means to be a psychologist. Part of the symbols of what therapeutic work is in general." From the position of playing psychologist, he related the following story:
"Al, a borderline patient, defecated in his pants. I escorted him to his room and stayed with him while he was changing.... I was appalled by the smell, it was disgusting. It's not easy... this case was clearly good for both of us. I was exposed to his most sensitive place ... his homosexuality, his penis."
Jacob was aware of the fact that a clinical psychologist may not have accompanied AI to his room nor stayed there while he was changing. This incident, however, gave Jacob an opportunity to act out his fantasy, to play therapist. He continued,
"We talked about ah ... about the fact that AI goes at night to find sexual partners in public gardens. I asked him what kind of sensation he has when he feels the urge to go out. After three days AI came to me with an answer. He said he feels arousal in his anus. OK, I said, let's buy you a bicycle. Each time you feel the sensation, go for a ride on the bicycle ... that was just an experiment, I didn't know exactly ... well it was pretty clear. The meaning of it was that every time he rides the bicycle he is having intercourse with me ... that the bicycles are me ... that was based on the idea that the analogy between the bicycles and me is something that will connect him to transference and me to countertransference in the sense that I will acknowledge the sexual tension between us ... the fact that he courts me. This was very difficult ... very difficult ... in the emotional sense ... if I just had some training.... In supervision I could have worked on it, but I couldn't by myself."
Jacob told this story to demonstrate the way in which he controlled his feelings of pity and disgust. This incident also revealed his attitude against just "being yourself' and reacting spontaneously to Al's embarrassing episode. He wanted an interaction of a therapist and patient. His need for training and supervision reflected his belief that there is essentially a right therapeutic response that he ought to learn and follow and that his emotional responses could and should be planned in advance, follow a certain logic, and entail specific techniques.
Jacob's story points out the unique nature of emotional management in the mental health setting. Neutrality was not a sufficient stance. Unlike medical students who worked hard to establish emotional distance and uninvolvement (see Smith and Kleinman 1989), Jacob aspired to transform pity and disgust into empathy and care. Indeed, the idea of care and its place in the psychotherapeutic discourse reverberates in Jacob's story. Therapists, as much as lay people, conceive the capacity to care as a natural gift and a personality skill. However, the idea of care is a "thought collective" (Douglas 1986) of psychotherapists. Moreover, the normative belief that therapists must be empathy directed and fundamentally interested in people and their problems implicitly influences the selection, training, and evaluation of 'the good" therapist (Light 1980). Spurling and Dryden (1989), who interviewed several prominent psychotherapists on the process of becoming a therapist, found that understanding, empathy, and the drive to repair appeared as a dominant calling in the lives of those who talked about their occupation in a distinct "tribal' language, creating a therapeutic world like the worlds of science, art, or music.
We were not surprised, then, to find students like Jacob who adopted the language of care as part of their professional self identities. Natan, another counselor, talked, for example, about love: "it is the kind of love that you feel toward people who you must help. It is also the kind of love that comes with the role. The definition of our role is that you must be empathic and caring." Natan's sense of "must" reflects the common belief that role taking emotions (Shott 1979), such as care and empathy, are necessary dispositions of therapists.
Another story demonstrates the importance that students at the shelter ascribed to their control of anger and other "unwarranted" feelings. This story focuses on the failure of empathy, the shame, and the deep sense of inadequacy that followed it. Dan, who saw himself as a novice in the field of psychology, but who had great ambitions "to be good," related the following sad incident:
"I arrived at my night shift.... We have a patient whom, I believe, should have not been in the program. Then from 1 1 P.m. until 6 in the morning she drove me crazy in every possible way: she started by asking for a piece of bread, which we customarily don't give so as not to fixate such habits. I explained to her reasonably that I cannot. But she came back, not so much for the bread, just to annoy me. So, I gave her a piece of bread. From that moment the whole situation began to deteriorate. Every other minute she came back, saying: "OK, I am going to sleep, I am going to sleep." If I could, I would have given her a slap in the face. I could hardly control myself. The situation just slipped from my hands. Now-the anger, you just seek to throw it on someone, because you can't be angry. I wanted to understand, but there was nothing I could do. The problem is that she is diagnosed as obsessive-compulsive. I can't be angry at her. Yet, I am so angry. At whom should I be angry? At the hospital? Myself? I tried every possible way to deal with her. Finally, I wrote three pages of case report. I tried to look at the case objectively, and I let out my anger through writing.... I was swept away by the situation.... I became part of the situation, which I know I am not supposed to do."
Dan's presentation of the case was not the whole story. As his written report indicated, during that night he also yelled at the patient and threatened her, finally removing all the tokens patients receive for good behavior from her door. As a result he felt, in his words, 'sick' and 'like an idiot.' His relief came only by 'transferring all this to writing.' Dan knew that he was "not supposed to feel and express anger," but he could not control himself. In Dr. Berger's words, he was "too young in a professional sense" to know how to do that. As a consequence, he experienced low self-esteem and shame. The incidents we heard from Dan and Jacob, although different emotionally, suggest that both had a clear image of the "good therapist," and that included emotional work in a way that was yet hidden from them.
At the same time that they worked to control their anger against patients, the "psychologists" appeared to use their anger at the lack of supervision and training to promote their professional identity. As Clark (1990) notes, emotions often generate a micro hierarchy of status claims (p. 316). Through anger, the students claimed a place and identity for themselves. Moreover, they could blame Dr. Berger, the place, and the program, overlooking the fact that some of their problems derived from playing therapist. Dr. Berger claimed that his door is always open and that he is available to help the students whenever they feel anxious or cannot solve an emotional problem, but few students took advantage of his open-door policy. We suspect that the students were not making use of this open-door policy in order to keep their anger to themselves. Holding on to the anger gave them a sense of power and difference.
Ora looked on her role as counselor from a different perspective than Jacob and Dan did. She believed that her task was "to be as human and natural as possible." She saw clearly the tension between these two perspectives-being professional and being human:
"Look, I relate to residents as human beings. People, as you know, are sometimes nice and you love them, and sometimes are irritating and you feel like yelling at them, as I do. if our goal is to rehabilitate people, to enable them to function outside the hospital, then they should be prepared. In the outside world people are not always soft and tolerant. When they are annoyed, they yell. After all, we are human beings. It is true we are also the guides, and this is our role, but we work with people, and when people provoke you, you tend to react."
Ora was resistant to students who played, in her words, the role of the "perfect therapist" and who 'supposedly knew what a good therapist was." "What do you mean by playing therapist?" we asked. "Playing therapist," she explained, "means not to laugh at funny things because it is not therapeutic or ethically correct, to be always the one who understands, accepts, explains and never gets angry." As she got more aroused, her anger became more blatant:
"Psychology students came with the pose of the "psychologist'. .. then you don't hate, but you don't love either... no, you don't show emotions. I am sure that they had a lot of feelings inside them, they had residents that they loved and others that they hated, but the norm of the "psychologist" would not let them display it or talk about it. It is a shame to touch life from the pose of the "psychologist.""
Ora's criticism raised a distinction, common among the students in the shelter, between those students who controlled their emotions - here, called the psychologists-and those who freely engaged in emotional episodes of love and anger - the guides. Her attitude reflected the lay belief that there are two kinds of emotional ethics: professional and human. Her role as a guide, she believed, was 'to be real.' She said: 'Our role is to be guides, not therapists. As guides, our role is to construct real-life situations in the shelter. In real life, you sometimes love, sometimes are angry, sometimes laugh, sometimes are sad.' Like Ora, Avi saw himself as a guide and not as a therapist. To sit with residents in their rooms, drink, laugh, and take walks with them was, he believed, part of being human. In his interview, he defined the borders between himself and the residents as illusory. 'Only one step separates me from the residents,' he said. 'I can hide my paranoid thoughts, while they cannot.' To our question about displaying emotions, Avi answered,
"I show them everything. I don't get out of control, but I tell them what I feel whether it's good or bad. I do that intentionally, in order not to represent a role model which is either always good and affectionate, or bad and tough. They should see me as a real person who sometimes gets angry and sometimes smiles."
Other students fell somewhere between Jacob the therapist and Ora the guide. Natan, for example, admitted to being an emotional person with extreme motivation to help and express love:
Interviewer: Don't you feel like shouting at residents sometimes?
Natan: Yes, sometimes.
Interviewer: Do you?
Natan: Yes, if it is constructive. If I believe that it might be therapeutic.
Interviewer: Don't you yell at residents without therapeutic intentions, just because you feel like it?
Natan: No. I control myself.
Interviewer: Is it forbidden to yell?
Natan: No, I don't think so.
Interviewer: Is it allowed or forbidden?
Interviewer: Even if it's not a therapeutic act?
Natan: Even so. After all you are a person, and if someone torments you, you are permitted to yell.
Interviewer: Then why aren't you yelling?
Natan: It's a personal hang-up. But I also believe that the role obligates you to be in emotional control and to avoid strong emotional reactions which are not constructive from a therapeutic point of view.
In his position, Natan was aware of the schism between the personal and the professional. He was also aware of the gap between his attitude (yelling is permitted) and his behavior (yelling is only allowed if therapeutic). Natan perceived himself as a guide but practised emotional control as a therapist, perhaps because he had not yet internalised a clear identity role. It is clear that his spontaneity was modified by what he saw as the therapeutic point of view. Natan's reaction is a good example of the tension between spontaneity and institutional (or discursive) control of feelings (Gordon 1990, 168). By believing in spontaneity, he made a claim to the place he desires (Clark 1990). His behavior, however, marked him as a psychologist.
THE ECONOMY OF PROFESSIONAL FEELINGS
Students who clearly saw their work in the shelter as part of their professional socialisation and took it as an opportunity to participate in the 'tribal act" of therapy (Light 1980) spoke of emotional self-control as an obligation. They incorporated the personal ("being real") and the professional ("being a therapist') into a single moral identity role (Sarbin 1995), overcoming the tension and inconsistency between these two identities by their strong motivation to be professionals. The guides, on the other hand, adopted the personal. They laughed when something was funny and got angry when insulted or maddened. Although the psychologists did not confront this attitude with direct animosity or open conflict, in private talks among themselves they often criticised the tendency of the guides to disregard professional emotional boundaries. The psychologists created, through various social and rhetorical activities, an aura of importance and exclusivity. They often talked about their uniqueness as a group, in both the shelter and the university. Clearly, the psychologists perceived themselves as more adept, serious, and responsible than the guides were.
Surprisingly, they were also treated in this light by the administration, even though they were the ones who challenged the policy not to play therapist. When an administrative decision was made one day to divide the residents and staff members into three groups, each located in a different area of the building, Dr. Berger chose those students who we considered the leading figures among the psychologists to serve as group supervisors of the counselors. In this way, the administration gave tacit support, legitimacy, and power to the psychologists and their attitude. This incident illustrates the vulnerability of groups outside the dominant discourse (i.e., the guides). At the same time, it also suggests that the counter narrative of emotional self-disclosure and interpersonal meanings poses a threat to the professional order.
The distinction between the psychologists and the guides reflects two possible competing approaches of emotional work in organizational environments: scientific management and work feelings (Putnam and Mumby 1993). Putnam and Mumby use scientific management to refer to the scientific language of objectivity, rationality, detachment, and control that workers use to talk about emotional interactions at work. Work feelings, in contrast, allude to emergent feelings through the negotiation of interpersonal meanings.
"Work feelings are those emotions that emerge from human interaction rather than being imposed by instrumental goals and bureaucratic rationality .... That is, work feelings aid in negotiating meanings about roles and relationships rather than in conforming to predetermined display rules or to prescribed norms." (pp. 49-50)
Although the students who took on the role of guides could not negotiate meanings on an equal basis with the residents, as Putnam and Mumby's (1993) concept of work feelings would require, they nonetheless used personal life experiences and real-life responses as their criteria for regulating emotional display, showing anger or annoyance when the situation warranted. In contrast, the students who identified themselves with psychologists relied on their knowledge from classes, books, and discussions with professional therapists. Their motivation to mask, distance, or change feelings of anger and aggression reflected their occupational aspirations and their professional anticipation. They wanted to act as psychologists, not as lay people. Although their knowledge of psychology in general, and psychotherapy in particular, was still uncritical and idealised, they believed wholeheartedly that their attempt to hide and control negative feelings would draw them closer to real therapeutic work and place them higher in the hierarchy of professional practice (Clark 1990). The point we wish to make is that students who played psychologist were not as free as the guides to exercise their emotions as individuals. They "knew" from the little they had learned, that therapists must discipline their emotional display and that one day they would be paid to do so (Fineman 1993). Their notions regarding the nature of exchange between patients and therapist coincide with Fineman's (1993) conclusion that professional workers-doctors, nurses, or social workers-believe that "they are to look serious, understanding, controlled, cool, empathic, and so forth with their clients or patients" and that they ought to protect themselves from 'private feelings of pain, despair, fear, attraction, revulsion, or love; feelings which would otherwise interfere with the professional relationship' (p. 19). They also match Putnam and Mumby's (1993) concept of scientific management of feelings: the students believed that rules, techniques, and strategies rationally regulate the therapist's feelings in order to reach objectivity in treatment.
TWO FORMS OF EMOTIONAL LABOUR
The differences between the psychologists and the guides, their distinct beliefs, approach, and identity roles, indicate that the concept of emotional labour cannot be uniformly theorised even within a particular coherent discursive field. Every dominant discourse confronts the challenge of subversive voices. As often is the case, the differences between the two groups are not incidental and in some way reflect the debate within the field of psychotherapy and psychoanalysis over the meaning of objectivity. The debate is particularly acute within theoretical bounds: how far must therapists control their emotions and in what ways?
In his lecture on "wild" psychoanalysis, Freud ([l910] 1964) located psychoanalysis among the sciences, establishing a field of inquiry intolerant of popular and superficial apprehensions. Freud's concept of psychoanalytic expertise was based not only on proficiency of knowledge, but also on a mastery of techniques. His well-known preference to sit behind the patient clearly demonstrates that techniques are often ideological strategies. It is not surprising that this well-thought-out methodological preference "naturally" generated, as Gabbard (1995) notes, the theories of emotional distancing and neutrality that have dominated the mental health field.
While many therapists still maintain that loss of boundaries is unprofessional and detrimental to therapy, this stance has recently been challenged by a new psychoanalytic approach known as two-person psychology (Modell 1984). Within this developing field of therapy (Mitchell 1988; Stolorow, Brandchaft, and Atwood 1987), the interaction between therapist and patient approximates personal relations between two individuals who negotiate their associations, ideas, and feelings. Understanding the patient's inner world is viewed as an intersubjective endeavour that is contingent on the therapist's willingness to use his or her inner world (Bollas 1992), sometimes to the point of self-disclosure (Ehrenberg 1982).
Even proponents of two-person psychology do not suggest that method, technique, and professionalism be discarded. Questioning the rigidity of boundaries, treating therapeutic data as a shared creation, and allocating space for the therapist's subjectivity are all used in the service of cure. It is, therefore, incumbent upon the therapist, as an agent of change, to practice these techniques. The two-person therapist is not exempt from the proper stance or from obeying the rules of professional feeling.
Going back to the psychologists and guides, it is clear that the former wilfully adopted the dominant rhetoric of emotional control in the common meaning of distancing and neutrality. Most of all, they wanted to own those objective properties that would change them from wild analysts to professionals. The ideological position of the guides was more ambiguous and complicated. They believed in "real feelings" and in "being a real person." Their insistence on authenticity, however, did not free them from emotional labour. They, too, as we observed, were struggling to find "the right way" to deal with their feelings. In their own way, they were also producing a discourse of emotional control.
In this article, we discussed the concept of professional feelings. We claimed that not only workers, patients, or clients are subjects of emotional manipulation and control. Mental health professionals, who have the power and the authority to manage the feelings of others by the definition of their expertise, are likewise subjected to emotional dispositions and feeling rules. It is indeed common knowledge that therapists manage their feelings toward clients in order to display neutrality (Hardesty 1987, 247). However, it is not obvious how neutrality, emotional distance, or empathy are in fact negotiated and practised. Do therapists identify or struggle with the ideology of emotional control and, if so, in what ways? Emotional labour in professional organisations is not easily identified or recognised, mainly because rules of regulation and disciplinary practices are disguised as ethical codes, professional techniques, and specialised knowledge. Our project was, then, to show that emotional labour in professional fields is not simply the management of neutrality. In the psychotherapeutic field, for example, emotional labour is a self-regulated process by which the right feelings, whether anger or empathy, are constituted through discourse. It is a process by which propriety, not only neutrality, is established, managed, and displayed. Even hatred can serve a point or be helpful when it is well managed and controlled (Winnicott 1975). This is why, in the process of socialisation, students yearn for guidelines: when to show or hide anger, how to reduce emotional uncertainties, and how to enhance emotional competence beyond strategies of distancing. Their sentiment is particularly revealing considering that professional institutions, such as the shelter or medical school (Smith and Kleinman 1989), emphasise emotional control but deny students full membership through training. As we noticed, students are not interested in who controls their feelings and what it means to stay calm when feeling anger. For most of them, mastery of the "right" feelings signifies professionalism and a professional identity. Their ability to detach, displace, transform, and substitute feelings reflects their belief in a legitimate difference between trained and untrained people, as well as the legitimate contrast between common sentiments, and the science of emotions.
Learning to manage professional feelings is, however, connected to ideology. The guides rejected the strategy of neutralising feelings, believing instead that emotional control actually meant being personal and responsive without losing sight of their role as mediators. The psychologists firmly believed in neutrality and objectivity and, as a result, often focused on the "problem" rather than the individual resident. By contrast, the guides responded with emotionality, treating the residents as historical beings and as their equals.
The difference between the two groups reflects distinct meanings of emotional labour and the connection between different discourses and the display of feelings. It is easy to see how a rhetoric of objectivity acts as a mechanism of control, but it is less obvious when the practice of control is represented by claims of spontaneity and humanness. The subversive challenge of such a position is revealed in the harshness with which it was opposed by the psychologists (Foucault 1994, 41) and the rivalry between the two approaches to feeling management. We claim, thus, that emotional labour in professional service organisations is the product of contested professional discourse.
Abbott, A. 1988. The system of professions.' An essay on the division of expert labour. Chicago: University of Chicago Press.
Ash, M. K. 1984. Mary Kay on people management New York: Warner.
Bernard, M. J., and K. R. Goodyear. 1992. Fundamentals of clinical supervision. Boston: Allyn & Bacon.
Bollas, C. 1992. Being a character: Psychoanalysis and self experience. London: Routledge.
Bourdieu, P., and L.J.D. Wacquant. 1992. An invitation to reflexive sociology. Chicago: University of Chicago Press.
Clerk, C. 1990. Emotions and micro politics in everyday life: Some patterns and paradoxes of 'place.' In Research agendas in the sociology of emotions, edited by T. Kemper, 305-33. Albany: State University of New York Press.
Douglas, M. 1986. How institutions think. Syracuse, NY: Syracuse University Press.
Ehrenberg, D. B. 1982. Psychoanalytic engagement: The transaction as primary data.
Contemporary Psychoanalysis 18:535-55.
Fineman, S., ed. 1993. Emotion in organisations. London: Sage.
Foucault, M. 1994. Genealogy and social criticism. In The post modem turn: New perspectives on social theory, edited by S. Seidman, 39-45. Cambridge, UK. Cambridge University Press.
Freidson, E. 1988. Profession of medicine: A study in the sociology of applied knowledge. Chicago: University of Chicago Press.
Freidson, E. 1994. Professionalism reborn. Chicago: University of Chicago Press.
Freud, S.  1964. Concerning 'wild' psychoanalysis. Standard Edition, vol. 11, 221-27. London: Hogarth.
Gabbard, O. G. 1995. When the patent is a therapist: Special challenges in the psychoanalysis of mental health professionals. Psychoanalytic Review 82:709-25.
Goffman, E. 1959. The presentation of self in everyday life. New York: Doubleday.
Gordon, S. L. 1990. Social structural effects on emotions. In Research agendas in the sociology of emotions, edited by T. Kemper, 145-79. Albany: State University of New York Press.
Greenberg, J. 1996. Psychoanalytic interaction. Psychoanalytic Inquiry 16:25-38'
Hardesty, M. J. 1987. The social control of emotions in the development of therapy relations. Sociological Quarterly 28:247-64.
Hochschild, A. R. 1975. The sociology of feeling and emotion: Selected possibilities. In Another voice, edited by M. Millman and R. M. Kanter, 280-307. New York: Anchor.
Hochschild, A. R. 1977. Emotion work, feeling rules and social structure. American Journal of Sociology 85:551-75.
Hochschild, A. R. 1983. The managed heart. Berkeley: University of California Press.
Hunter, 1. 1993. Personality as a vocation: The political rationality of the humanities. in Foucault's new domains, edited by M. Gane and T. Johnson, l53-92. London: Routledge.
Kleinman, S. 1996. Opposing ambitions. Chicago: University of Chicago Press
Kunda, G. 1992. Engineering culture. Philadelphia: Temple University Press.
Light, D. 1980. Becoming psychiatrists: The professional transformation of self. New York: Norton.
Lutz, C., and L. Abu-Lughod. 1990. Language and the politics of emotions. Cambridge, UK: Cambridge University Press.
Meissner, W. W. 1996. Empathy in the therapeutic alliance. Psychoanalytic Inquiry 16:38-53.
Mitchell, S. 1988. Relational concepts in psychoanalysis: An integration. Cambridge, MA: Harvard University Press.
Modell, A. H. 1984. Psychoanalysis in a new context. New York: International Universities Press.
Putnam, L., and D. Mumby. 1993. Organisations, emotion and the myth of rationality. In
Emotion in organisations, edited by S. Fineman, 36-57. London: Sage.
Sarbin, T. R. 1995. Emotional life, rhetoric, and roles. Journal of Narrative and life History 5: 213-20.
Shott, S. 1979. Emotion and social life: A symbolic interactionist perspective. American
Journal of Sociology 84: 1317-34.
Smith, P. 1992. The emotional labour of nursing. London: Macmillan.
Smith, A. C., and S. Kleinman. 1989. Managing emotions in medical school: Students' contacts with the living and the dead. Social Psychology Quarterly 52:56-69.
Spurling, L., and W. Dryden. 1989. The self and the therapeutic domain. In On becoming a psychotherapist, edited by W. Dryden and L. Spurling, 191-214. London: Tavistock/Routledge.
Stolorow, R. D., B. Brandchaft, and G. E. Atwood. 1987. Psychoanalytic treatment., An intersubjective approach. Hillsdale, NJ: Analytic Press.
Thoits, P.A. 1989. The sociology of emotions. Annual Review of Sociology, 15:317-42.
Thoits, P.A.. 1996. Managing the emotions of others. Symbolic Interaction 19: 85-109.
Van Maanen, J., and G. Kunda. 1989. 'Real feelings': Emotional expression and organizational culture. In Research in organizational behaviour, edited by L. L. Cumming and B. M. Stew, 43-103. Greenwich, CT: JAI.
Winnicott, D. W. 1975. Hate in the countertransference. In Through paediatrics to psychoanalysis, 194-203. New York: Basic Books.
NIZA YANAY is a senior lecturer at the department of behavioural sciences at Ben Gudon University in Israel. Her current research focuses on the relations between national discourses and emotional identities. She has written several articles on the social construction of hostility and national hatred.
GOLAN SHAHAR is a graduate student in clinical psychology at Ben Gudon University in Israel. He is currently writing his Ph.D. dissertation on vulnerability to depression.
 Deep acting is a form of emotional labour entailing a change of the inner world. One prompts oneself to feel or not feet certain emotions or to change the interpretation of the situation by either raising emotional memories or manipulating 'as-if' images. In contrast, surface acting is restricted to facial and body language. It is important to note that surface acting was much more dominant in the shelter than deep acting was. A possible explanation is the duration of work with the residents. The longer one works with people, the harder it is to maintain as-if images.
 Here, Merton's concept of anticipatory socialisation (Merton and Kitt 1950) comes to mind. In the current study, psychology students adopted empathy, neutrality, and objectivity as 'emotion norms' (Thoits 1989) to affiliate themselves with the group of professional therapists.
 Hochschild (1975) defines feeling rules as norms and standards that reconstruct inner experiences in cultural, social, or organizational settings. Feeling rules, she writes, 'define what we should feel in various circumstances' (p. 289). Such rules differ from culture to culture, indicating what is appropriate and desirable.
 Smith (1992) reported similar findings in her fieldwork on nurses' emotional labour. Nurses at public hospitals expressed more self-confidence and satisfaction when they received training in expressing feelings with patients and their families, disguising pain, and radiating optimism. Smith observed that emotional uncertainty led to frustration, distress, and serious doubts about adequacy.
 We were inspired by Kleinman's (1996) notion of legitimate alternative, which indicates that professionals are concerned with standards, conventions, and boundaries even within alternative organisations. Moreover, her work demonstrates that emotions play an important role in creating work standards and legitimacy.