04-078918-Toerien.qxd http://soc.sagepub.com Sociology DOI: 10.1177/0038038507078918 2007; 41; 645 Sociology Merran Toerien and Celia Kitzinger Beauty Salon Emotional Labour in Action: Navigating Multiple Involvements in the http://soc.sagepub.com/cgi/content/abstract/41/4/645 The online version of this article can be found at: Published by: http://www.sagepublications.com On behalf of: British Sociological Association can be found at:Sociology Additional services and information for http://soc.sagepub.com/cgi/alerts Email Alerts: http://soc.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.navReprints: http://www.sagepub.co.uk/journalsPermissions.navPermissions: http://soc.sagepub.com/cgi/content/refs/41/4/645 Citations by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://www.britsoc.co.uk http://soc.sagepub.com/cgi/alerts http://soc.sagepub.com/subscriptions http://www.sagepub.com/journalsReprints.nav http://www.sagepub.co.uk/journalsPermissions.nav http://soc.sagepub.com/cgi/content/refs/41/4/645 http://soc.sagepub.com Emotional Labour in Action: Navigating Multiple Involvements in the Beauty Salon ■■ Merran Toerien University of York ■■ Celia Kitzinger University of York A B S T R AC T Building on Hochschild’s path-breaking analysis of service providers’ ‘emotional labour’, this article demonstrates some of the interactional skills required for emo- tional labour to be performed. Using conversation analysis (CA), we examine a sin- gle case from a database of recorded beauty salon interactions. The episode was chosen because it makes visible the mechanics of how a beauty therapist manages conflict between her ‘multiple involvements’ in the salon: between her simultane- ous engagement in topic talk and hair removal. We show first how she navigates this conflict and then how her actions may be understood as an example of emo- tional labour. The article addresses, then, both the feminist concern with making visible the skills of emotional labour and the conversation analytic concern with how participants manage multiple involvements in a socially meaningful way. K E Y WO R D S beauty therapy / conversation analysis / emotional labour / feminist sociology Introduction Emotional labor emphasizes the relational rather than the task-based aspect of work found primarily but not exclusively in the service economy. It is labor-intensive work; it is skilled, effort-intensive, and productive labor. It creates value, affects productiv- ity, and generates profit. It is why frontline service workers and paraprofessionals have been referred to … as the ‘emotional proletariat’. (Steinberg and Figart, 1999: 9) 645 S o c i o l o g y Copyright © 2007 BSA Publications Ltd® Volume 41(4): 645–662 DOI: 10.1177/0038038507078918 SAGE Publications Los Angeles, London, New Delhi and Singapore by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com I n her classic analysis of flight attendants’ work, Hochschild (1983) first used the term ‘emotional labour’ to refer to self and other emotion management in the paid workplace. She distinguished this from ‘emotion work’: emotion management in the non-paid sphere. Although this distinction has been cor- rectly criticized as simplistic (Bolton and Boyd, 2003), the act of naming ‘the often invisible dimensions of the relational work which people do as part of car- ing for their families or performing their paid jobs’ (Mirchandani, 2003: 721) was path breaking. Hochschild’s thinking has since been subject to extensive development (see Steinberg and Figart, 1999) and applied to numerous work- place settings: from hospice nursing (James, 1989) to sex work (Sanders, 2004); from call-taking (Korczynski, 2003) to small-business management (Mirchandani, 2003). We focus on emotional labour in the workplace because our analytic interest is in professional beauty therapy. As the opening quote suggests, emotional labour takes skill, may be effort intensive, and has the aim of generating profit, improving workplace function- ing and/or adding value to an organization’s service. Emotional labour includes the idea that many employees are ‘paid to “look nice”, smile, be caring, be polite’ (Fineman, 1996: 546), and may entail an array of relational tasks, such as, ‘soothing tempers, boosting confidence, fuelling pride, preventing frictions, and mending ego wounds’ (Calhoun, 1992: 118). Emotional labour may involve suppressing or inducing one’s own emotions. Crucially, the employee must effect whatever emotional impact is required by the organization; e.g. flight attendants must relax nervous passengers while debt collectors must intimidate (Hochschild, 1983). Emotional labour researchers have pursued diverse agendas, including, as Steinberg and Figart (1999) show, assessment of: (i) the effect on employees of performing emotional labour; (ii) the relationship between emotional labour and organizational effectiveness; and (iii) the extent to which emotional labour is recognized and remunerated. Feminist researchers have been particularly interested in the third of these, finding emotional labour/emotion work to be useful concepts because they name ‘as work behaviours which, whether performed in the private or in the public sphere, are typically invisible and unrecognised as such’ (Frith and Kitzinger, 1998: 302, emphasis in original). Thus, although often a job requirement, emotional labour tends to be poorly rewarded (Steinberg and Figart, 1999). This is a feminist concern because, as a group, women are worse affected than men. This partly reflects women’s overrepresentation in the ser- vice industry, where emotional labour is a job requirement and pay is notori- ously poor. Crucially, however, feminists have underscored the socially constructed nature of skill, showing how the tasks traditionally performed by women within the reproductive sphere are: … defined as requiring minimal skill – even though the very same tasks carried out in the productive sector, particularly if associated with male workers, are considered to require extensive training and qualifications … the ‘socially constructed’ nature of skill includes the assumption that women are born with certain ‘natural’ skills which require neither talent nor training, and which are merely part of their ‘natu- ral’, ‘feminine’ behaviour. (Tancred, 1995: 17) 646 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Consequently, such skills are rendered invisible, and remain so even when utilized by women in their paid jobs. Thus, the tendency for women to be hired by ‘organi- zations which require the constant display of friendliness … on the assumption that women are better at displays of warmth’ (Mirchandani, 2003: 723) does not trans- late into adequate remuneration for the skills performed. An important feminist pro- ject has thus been to ‘expand the definition of skill to include emotional labour’ (Steinberg and Figart, 1999: 14) and to make visible the skills involved in perform- ing emotional labour effectively. It is to this project that we aim to contribute here. Previous researchers have also identified emotional labour as a job require- ment for beauty therapists. Gimlin (1996), for instance, asserts that ‘beauticians deal with their customers by attempting to create a personal relationship with them, listening to and remembering the intimate details of their lives, and claim- ing emotional attachment to them’ (p. 514). Similarly, Sharma and Black (2001) comment that beauty therapists are expected to give the client a ‘morale boosting “treat” … delivered in the context of attentiveness to that client’s indi- vidual needs and circumstances’ (p. 919). Black (2004) highlights the mismatch between job requirements and pay very clearly. She shows that emotional labour is needed in the salon to fulfil both short- and long-term organizational requirements. In the short-term, salons need ‘orderly and compliant’ clients, not clients ‘who will scream and thrash around as the pain of the electrolysis nee- dle kicks in, disturbing the woman in the next cubicle’ (Sharma and Black, 2001: 925). The long-term goal is to establish a loyal client base in a competi- tive market. Both goals require skills beyond the practical ones taught at beauty college, yet even senior therapists are paid very poorly. Black reports that the pay for UK beauty therapists actually decreased slightly from 2000 to 2002: In 2000 the average hourly pay for a therapist was £4.93, for a senior therapist £6.31 and for a salon manager £8.63. In 2002 these respective figures stood at £4.88, £6.18, and £6.86. Nail technicians received £3.89, a surprising figure, since this is actually below the legal minimum wage. (2004: 104–5) Women’s emotional labour in general, then, is a feminist issue; the poorly paid, highly gendered world of the beauty salon offers a particularly stark example. Although concerned with actual workplace practices, much emotional labour research has relied on self-report data (Frith and Kitzinger, 1998). An alternative approach is to analyse what actually takes place in workplace inter- action. Observational studies have gone some way towards doing this (e.g. Korczynski, 2003; Sanders, 2004). However, almost no emotional labour research has analysed recorded interactions (for an exception, see Wingard, forthcoming, who uses conversation analysis to analyse instances of emotion work in video-recordings of families interacting at home). In the absence of such recordings, the researcher is reliant on summaries or idealizations of what occurred; the complex details of talk-in-interaction are irrecoverably lost. Yet an analysis of these details is crucial, we argue, to the feminist project of mak- ing visible the skills involved in performing emotional labour. As part of a broader study of hair removal as paid work (see Toerien and Kitzinger, 2007), this article uses conversation analysis (CA) to examine emotional labour in 647Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com action. Analysing a recorded instance of salon interaction, we ask: what does it take to do that which sociologists have theorized as emotional labour? Part of the answer, we show, lies in employees’ capacity to be ‘multiply involved’ (LeBaron and Jones, 2002): to engage in more than one activity simul- taneously. How participants negotiate multiple involvements in a socially mean- ingful way has recently come under scrutiny within CA. Goodwin and Goodwin (1992), for example, address this complexity in a family’s mealtime activity. They show how the younger daughter negotiates competing claims for her alignment: by her sister, who is telling an extended story, and by her boyfriend who is seek- ing to get her ‘on side’ with his heckling of the story-teller. In addition, they show how the story-teller manages – through a gesture – the potential disruption of her story by an offer of more food. They deal also, then, with the coordination of talk and non-vocal interaction (see also Stivers and Sidnell, 2005). Erickson (1992), similarly, analyses the coordination of turns at talk and eating during another family dinner. While multiple activities may be managed smoothly, such studies demonstrate that this is an accomplishment, requiring interactional skills; some- times, conflict between activities may have to be navigated. For example, LeBaron and Jones (2002), who analyse a chance reunion between two women at a hairdresser’s, show how it conflicts with the physical activity of hairdressing. In this article, we also focus on conflicting multiple involvements during a ser- vice encounter. We examine a single case that caught our attention precisely because it makes visible a moment of conflict between a beauty therapist’s engage- ment in the physical practices of shaping a client’s eyebrows and the relational practices of ‘chatting’ with the client. We begin by showing how the therapist nav- igates this conflict. We then ask what might be accomplished interactionally by her actions. It is a key argument of this article that both the ‘chat’, and the way the therapist navigates the conflict between ‘chat’ and hair removal, are empirical examples of what sociologists theorize as emotional labour. Methodology and Data We employ the tools of CA, which allow for a fine-grained analysis of recorded talk and non-vocal interaction. Founded in the late 1960s and early 1970s by Sacks et al. (1974; and see Sacks, 1995), CA conceptualizes talk as a fundamental form of social action; as the means by which much of ‘the work of the constitutive institutions of societies gets done – institutions such as the economy, the polity, the family, socialization, etc.’ (Schegloff, 1996: 4). From this perspective, CA is not just a method for studying talk; it is ‘a distinctive way of doing sociology’ (ten Have, 1999: 7). CA’s basic goal is the ‘description and explication of the competences that ordinary speakers use … in participating in intelligible, socially organized interac- tion’ (Heritage and Atkinson, 1984: 1). Classically, the focus has been on ‘ordinary’ interaction. However, there is now a substantial body of work concerned with institutional interaction, which involves at least one participant representing an insti- tution and an orientation to an institutionally defined set of tasks (Drew and Heritage, 1992). Examples include counselling (Silverman, 1997), calls for emergency assistance (Whalen et al., 1988), presidential press conferences (Clayman 648 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com and Heritage, 2002), courtroom proceedings (Drew, 1990), and doctor–patient con- sultations (Heritage and Maynard, 2006). In keeping with CA’s ethnomethodolog- ical origins, a key concern is with how participants themselves constitute the encounter and their social roles ‘as having some distinctively institutional character’ (Drew and Heritage, 1992: 21); with how ‘institutional realities are evoked, manip- ulated and even transformed in interaction’ (Heritage, 1997: 162). There is not space here to review CA’s methods and findings. For excellent introductions to CA gener- ally, and the application of CA to institutional interaction, see Heritage and Atkinson (1984) and Drew and Heritage (1992), respectively. We used video and audio-tapes to record our data. As Heath and Hindmarsh (2002) emphasize, video-recordings are particularly important because they cap- ture the non-vocal actions that characterize most interaction. Our data extract is taken from a salon session of about 40 minutes long, recorded in February 2002. The agreed goal is the removal of hair from the client’s eyebrows, legs, ‘bikini line’ and underarms. We transcribed all talk using Jeffersonian transcription notation1 and include descriptions of key physical actions using curly brackets to show where these begin and end. Stills from the video are shown as Figures 1–6, and the clip is available on the first author’s website: http://www.york.ac.uk/depts/soci/s_toer.htm Provision of the clip is important so that readers do not have to rely on (unavoid- ably incomplete) transcription. However, this is only ethically acceptable where participants have given fully informed consent to the data being made widely available. In our study, participants – who were given detailed written and verbal information about our work – signed a consent form specifically allowing the data to be displayed. The beauty therapist requested that we include her and her salon’s name, which we are glad to in acknowledgement of her involvement. The extract (around one minute long) occurs 1min 18secs into the record- ing. The therapist is ‘threading’ the client’s right eyebrow; this is an Ancient Eastern technique, involving plucking the hairs with a loop of cotton. The client has experienced this procedure before with this therapist. In the extract, the therapist is standing at the top right-hand corner of a table, on which the client lies. The therapist holds a spool of thread in her right hand, with a loop of thread in her left and one end in her mouth (she is able to speak). The client’s eyes are closed throughout the procedure as protection against the light. (Please note that the client’s name is a pseudonym.) 01 Cli I still find it hard to imagine how you 02 actually do it? 03 ( . ) 04 BTh Do [you: :. Real]ly. 05 Cli [ °hem hm°] 06 (2 . 4) 07 Cli >I think it m[ust< take yea : : rs and year] 08 BTh [V e r y popular darli :ng,] 09 Cli ↑YEA: : [ : H] 10 BTh [Very] very popular. Because: u : : : m 11 ( 0 . 6 ) I g e t p e o p l e t r a v e l f r o m f a r f a r 649Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com 12 away. 13 Cli Do you teach people¿ 14 BTh .hhh I do: : : . If they want to. That’s what 15 other thing I’m thinking of.=Like teaching you 16 kno: {w, = {Bth stops threading 17 Cli = { ‘ C a u s e I k n o w l i k e i n a l l t h e {BTh’s R hand held briefly alongside R side of Cli’s face 18 magaz { i n e s n o w t h e [ y ’ r e ] {BTh extends R index finger 19 BTh [ Y ]eah 20 Cli { they’ re saying you know [threading’s] {BTh runs R index finger twice along Cli’s R eyebrow 21 BTh [Threading] 22 Cli d e f i n i t e l y t h e b e s t { w a : [ y , {BTh runs R index finger along Cli’s R eyebrow 23 BTh {[ B e s t w a y . T h a t ’ s t r u e : : . {1st extension of BTh’s R hand 24 Cli But I mean so {few {people know how to do it, {1st retraction of BTh’s R hand {2nd extension of BTh’s R hand {BTh’s gaze shifts 25 Cli { [s o ] { 2nd retraction of BTh’s R hand 26 BTh [That]’s {true{[: . {BTh’s gaze shifts { 3rd extension of BTh’s R hand 27 [You could prob’ly do very 28 { w e l l w i t h t e a c h i n [ g . ] {BTh’s R hand extends towards Cli’s R shoulder 29 BTh [Mm] hm 30 BTh {Suza: :nne? {BTh’s gaze shifts {BTh turns right hand so facing palm-down over Cli’s body 31 Cli Yea: [ : h? 32 BTh [With {this one darling, {BTh touches Cli’s R shoulder 33 Cli Mhm, 34 (.) 35 BTh Pull your eye:l[id, ((22 lines omitted during which the therapist issues further directives, with which the client complies, and checks the client is okay)) 58 { (4.4) {After about 2 seconds, the BTh begins threading 59 BTh °Alright darlin’° 60 (2 secs) 61 Cli What I also like about the threading is it 62 doesn’t take a l(h)ong ti(h)me. ((Talk about threading continues)) 650 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Analytic Section 1: Navigating Multiple Involvements (i) Simultaneous Threading and Topic Talk Our extract begins with the threading underway. The client initiates talk, topi- calizing the threading (lines 1–2), and the therapist responds, while maintain- ing a (predominantly) rhythmic series of plucks. Although much could be said about this slice of interaction, our main point is that the therapist is engaged in both activities simultaneously; they are not in competition. (ii) Interruption of Threading; Continuation of Topic Talk From lines 17–29, the threading stops while the topic talk continues. During this time, the therapist’s hand movements are crucial to our analysis. We focus first on lines 17–22. Having stopped threading, the therapist’s right hand hov- ers momentarily along the right side of the client’s face (line 17). She then extends her right index finger (line 18) towards the client’s right shoulder. She also runs her finger twice along the client’s right eyebrow (lines 20 & 22), after which she retracts the finger, ending physical contact with the client’s face. These actions are not visible to the client, whose eyes are closed. However, to 651Emotional labour in action Toerien & Kitzinger Note: Figures 1–6 printed with kind permission from Mina Rajdev (owner of Chic Beauty Clinic, London) and client Figure 1: First hand extension Figure 2: First hand retraction Figure 3: Second hand extension Figure 4: Second hand retraction Figure 5: Third hand extension Figure 6: shoulder-touch by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com an onlooker they contrast distinctively with the almost continuous threading action occurring prior to line 17 – there is a sense of the threading being inter- rupted. While this may be only a ‘sense’ at this point, there follows some strong evidence that the therapist is deliberately delaying the threading with reference to the ongoing topic talk. This is made visible by the therapist’s hand move- ments over five seconds of the interaction, which we describe next. (iii) Evidence of Conflict:Two Aborted Attempts to Reinitiate Threading Jumping ahead to line 32, notice that the therapist uses touch as part of her directive to the client to pull her eyelid. The shoulder-touch displays to the client – whose eyes are closed – which hand the therapist means by ‘with this one’ (line 32). The directive serves to elicit the client’s assistance, thereby reini- tiating the threading. In what follows, we show that the therapist makes two prior attempts to initiate this shoulder-touch, which she aborts in favour of the topic talk. The transcript and figures show where the therapist twice extends her hand towards the client’s shoulder only to retract it. We first describe where these hand movements occur before explicating their significance: • First aborted shoulder touch: As the therapist launches her agreement at line 23, she extends her right hand away from the client’s face, toward the client’s shoulder (Figure 1). A few words into the client’s next turn (line 24), she retracts her hand (Figure 2). • Second aborted shoulder touch: Almost immediately after the first retrac- tion, the therapist extends her hand again (line 24; Figure 3). She also glances away from the client’s face, along the right side of her body. Towards the end of the client’s turn (line 25; Figure 4), the therapist again retracts her hand. Just before completing her responsive turn at line 26, the therapist again glances along the right side of the client’s body. • Successful shoulder touch: As the client produces another assessment (lines 27–28), the therapist again extends her right hand, this time as far as the client’s shoulder, with her palm facing inwards towards the client’s body (Figure 5). As the therapist produces her summons to the client (line 30), she turns her hand so that the palm is facing down over the client’s shoul- der. Again, she glances down the right side of the client’s body. As she begins to issue her first directive to the client, she touches the client’s shoul- der (line 32, Figure 6). The therapist’s hand extensions are unavailable to the client, whose eyes are closed. They should not, then, be understood as recipient-designed gestures, which the client deliberately ignores or accidentally misses. Rather, what makes them analytically interesting is that they make visible to us the therapist’s inten- tion to elicit the client’s assistance some time before she does so. In what fol- lows, we examine the relationship between the therapist’s hand movements and the talk in order to unpack this claim. 652 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com (iv) Relationship Between the Hand Movements and Topic Talk When the therapist first extends her hand, she is producing an assessment (line 23) that agrees with the prior turn and adds nothing substantive. No further talk is conditionally relevant. When beginning her first hand extension, the therapist could, then, project topic closure and hence an upcoming place at which to initiate task-directed talk non-interruptively. However, the client con- tinues talking. The therapist’s retraction of her hand, which occurs partway through the client’s turn, may be understood as responsive to the resumed talk; as an abortion of the incipient shoulder-touch to avoid interrupting the client by initiating a directive sequence. The second retraction of the therapist’s hand, likewise, appears finely tuned to the ongoing talk: the hand is retracted as the client says ‘so’ (line 25), by which point it is apparent that her turn offers an assessment, making relevant a second assessment from the therapist (Pomerantz, 1984). Thus, the therapist’s retracted hand may be understood to display an analysis of the client’s turn as making relevant a response, which she produces (line 26) instead of initiating task-directed talk. The shift in the therapist’s gaze (line 24) provides further evi- dence that this retraction is an aborted version of the shoulder-touch. Not only does the therapist glance in the direction of the shoulder, but this gaze shift occurs again (line 26) just before the onset of the final hand extension and dur- ing the summons (line 30), suggesting that the gaze on all three occasions is directed at the goal of the incipient touch (the shoulder). The onset of the third extension of the therapist’s hand occurs, almost exactly like the first, as the therapist is coming to the end of a topic closing implicative turn: an agreement with a prior assessment (line 26). Again, therapist and client are aligned and the therapist has added nothing new. Again, they could be done – a good place for the therapist to shift the talk back to the task, non- interruptively. Again, however, the client resumes the topic talk (lines 27–28). This time, the therapist does not abort the hand extension. Instead, she deals with the conflict between the topic talk and her attempt to elicit the client’s assistance in a different way: she closely coordinates the extension of her hand with the talk, extending it but not touching the client’s shoulder until she has the floor. Notice how she gains the floor: not only does she produce a minimal agreement (line 29) as soon as the client’s turn could be possibly complete, but she then issues a sum- mons (line 30). The summons produces a clear shift between the prior talk and what is to come next, requesting the client to focus her attention on something new. Moreover, it makes relevant something different to topic talk: a response to the summons (Schegloff, 1968). The client responds (line 31), giving the therapist the go-ahead. She now has the right to produce her directive – and complete her shoulder-touch – and does so, beginning at line 32. One feature of the hand movements, which supports our analysis of the first two as aborted attempts at the shoulder-touch, is their ‘progressivity’. By this we mean that each extension gets physically closer to the shoulder (com- pare Figures 1, 3 and 5, which show the maximum extension for each prior to 653Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com the actual touch) and each retraction takes the hand less far away (compare Figures 2 and 4, which show the maximum retraction for each). Similarly, each of the therapist’s responsive turns is more minimal, adding less to the talk. The therapist is not mandated to abort the shoulder-touch by the client’s talk or the relevance of a responsive turn; this is a choice from alternatives (e.g. initiating her directive interruptively). She is actively managing the physical tasks of threading in relation to the task-irrelevant topic talk. She navigates the conflict between these multiple involvements by delaying the threading in favour of the topic talk. Notice that from line 61 the topic talk, reinitiated by the client, is resumed in tandem with the threading. In sum, the beauty therapist’s work includes managing the dual demands of topic talk and depilation; this requires, at times, navigating conflict between them. Analytic Section 2: Emotional Labour in Action We now consider the interactional significance of the therapist’s actions. Linking our microanalysis with the sociological concept of emotional labour, we argue that our extract provides evidence of two ways that a beauty therapist may per- form emotional labour: (i) by being ‘multiply involved’ and (ii) by navigating between her involvements in a way that maintains a ‘smooth’ interaction. (i) Engaging in Topic Talk:Treating the Client as an Individual As we have shown, the therapist engages in topic talk while performing the threading. At first glance we might dismiss this ‘chat’ as an example of emo- tional labour. It is not of an overtly emotional nature and is probably not what beauty therapists mean when they report a relational aspect to their work. Sharma and Black (2001) found that beauty therapists reported taking on a counsellor-like role when, for example, ‘clients confide about their family or marital problems in the course of a “treatment”’ (p. 926). We suggest, however, that the relational side of salon work is far more pervasive than such instances imply. Emotional labour includes, crucially, the successful personalizing of an encounter that is utterly routine for the therapist. As Sharma and Black put it: ‘at all costs the client must not feel that she is on a “production line” served by a robot’ (p. 928). In our extract we see evidence of a practice for personalizing the session: engaging in topic talk while performing the depilation. This talk serves to build a relationship between client and therapist and demands that the therapist be responsive to the client ‘without resort to formulaic or standard- ized (“regimental”) responses’ (Sharma and Black, 2001: 928). The therapist’s ability to be ‘multiply involved’ – to perform the threading while simultane- ously ‘chatting’ with the client – is one of the many competences that enable her to treat the client as an individual. It should be understood, then, as an exam- ple of emotional labour in action. 654 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com (ii) Navigating Between Multiple Involvements: ‘Smoothing’ the Interaction We have also shown how, when navigating conflict between her multiple involvements, the therapist delays the hair removal. She gives precedence, not to the physical tasks for which she is officially paid, but to the relational tasks of not coming off as rude, overly hurried, or not listening properly. In the extract we see something of what it takes to ‘smooth’ an interaction, to prevent interactional ‘frictions’ (Calhoun, 1992: 118), to ‘be polite’ (Fineman, 1996: 546): the therapist must not only know how to be appropriately ‘multiply involved’ but also how to navigate smoothly between her involvements. She must perform the interactionally complex task of reorienting the client’s atten- tion to the job at hand, without interrupting the topic talk, which the client repeatedly sustains. If managed seamlessly, this navigation between multiple involvements may be invisible. That is the point: to ‘smooth’ an interaction suc- cessfully, the process cannot appear effortful. The extract is a clear example of this aspect of emotional labour not because the therapist gets it just right: rather, it is informative because the therapist’s hand extensions make visible to us her efforts to avoid generating interactional ‘friction’. While the moment of conflict is kept below the surface of the interaction – as it must be if the thera- pist is to come off as attentive and polite – her hand extensions allow us to glimpse the underlying mechanics of so doing. The interactional competences we analyse here are not ‘spectacular’ in the way that dealing with an enraged or grieving client might be. They are, how- ever, the sorts of mundane (i.e. everyday, taken-for-granted) relational skills we expect from our service providers. They are the sorts of skills that, if practised well, leave us feeling ‘looked after’, and if not, leave us thinking we may go else- where in future. Discussion and Conclusions Hochschild’s (1983) coining of the terms ‘emotional labour’ and ‘emotion work’ has generated a multi-faceted body of research. Almost none of it, how- ever, has examined recordings of naturally occurring interaction to understand what emotional labour/emotion work might look like in practice. As Wingard (forthcoming) argues: Without actually viewing interactions in which emotion work occurs, researchers can- not tap into how emotion work is interactionally accomplished, what kinds of conse- quences it has for participants in that interaction, or even register an actual range of the different kinds of emotion work that might exist in particular interactions. Showing how wide a ‘variety of disparate kinds of interactions and strategies’ have been termed emotional labour/emotion work, Wingard proposes that ‘an initial typology of the kinds of verbal and nonverbal strategies that are instrumental in performing emotion work’ would be useful. The present article 655Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com contributes to this fledgling project by explicating a recorded empirical example of emotional labour in the beauty salon. Focusing on a moment of conflict between a beauty therapist’s simultaneous involvement in depilation and ‘chat’, we have demonstrated two kinds of strategy for performing emotional labour: (i) being ‘multiply involved’ and (ii) navigating appropriately between these involve- ments. In this case, the therapist’s delay of the hair removal in favour of the ‘chat’ gives precedence to her relational tasks; it is in this sense that her navigation of her involvements may be understood as an example of emotional labour. The idea that beauty therapy entails emotional labour is not new; neither is the political point that emotional labour is poorly recognized and rewarded. Yet the tendency to rely on interview or observational data has rendered researchers unable to demonstrate, in action, the interactional skills required to perform emotional labour. This is ironic given the feminist commitment to making women’s skills visible. Tancred (1995), for instance, argues that emo- tional labour is an excellent example of work involving invisible skills and con- cludes that feminist analysis is crucial to help extend the definition of skill. Interestingly, in light of our analysis, she suggests that, ‘if women’s work were to be taken as the norm, the ability to deal with several tasks simultaneously might well head the list in evaluating the difficulty of the job’ (p. 17). However, without a video recording, it is hard to imagine having recognized the signifi- cance of the beauty therapist’s hand movements with respect to her manage- ment of simultaneous tasks. More generally, while beauty therapists would likely recognize the importance of ‘chat’ for making the client feel good, they could not be expected to articulate the interactional skills required to navigate between ‘chat’ and depilation.2 Conversation analysis, with its tools for expli- cating how participants do interaction, offers an ideal method for making rela- tional skills visible. Our key contribution to the emotional labour literature, then, stems from our application of CA to the study of beauty therapy. With respect to CA, our main contribution is to the understanding of how participants manage their multiple involvements. In this respect, our analysis is closely aligned with research by LeBaron and Jones (2002), who also address a moment of conflict between topic talk and work on a client’s body (her hair). However, while they analyse conflict generated by an additional interaction, we show how multiple involvements may conflict within dyadic talk. Moreover, the nature of the conflicts is different. In the former, it is overt and enacted between hairdresser and client. The hairdresser resists the client’s attempts to interrupt the hairdressing: Jane [the hairdresser] repeatedly turns Katie’s body … and even sprays Katie’s face … to preserve the spatial integrity of the hairdressing activity. Thus, Katie must work to initiate a reunion, which is bodily at odds with other involvements. (LeBaron and Jones, 2002: 554–5) By contrast, the conflict in our extract is never allowed to rise to the surface of the interaction, and remains a conflict – for the therapist alone – between com- peting activities. Like Katie, she needs to close down one activity in favour of 656 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com another. To do this she must accomplish a mundane interactional task: eliciting the client’s attention. What makes our extract unusual is that the client’s eyes are closed. The therapist cannot, then, make use of the non-vocal attention-getting devices (e.g. body movements) that Heath (1984) has shown speakers using to encourage a display of recipiency. The therapist’s hand movements are not recipient designed. Rather, they are technically interesting because they afford a glimpse of the mechanics of how the therapist manages her multiple involvements. Obtaining the client’s attention is of interactional delicacy because – unaware of the therapist’s intentions – she repeatedly reinvigorates the topic talk. Non- vocal attention-getting devices deal with such delicacy by allowing ‘the partici- pants to maintain involvement in the business at hand, without having to address the problem of involvement as a topic in its own right’ (Heath, 1984: 332). Unable to subtly get the client on side, the therapist’s difficulty is how to obtain her attention non-interruptively. Ultimately she uses a vocal summons. However, as we have shown, this is delayed precisely in order to deal with the interactional delicacy. Our analysis points to some of the difficulties of manag- ing multiple involvements when participants have differential access to non- vocal actions. Further similar research has the potential to produce findings of practical applicability (e.g. for interactions between sighted and blind partici- pants). An important implication of our work for conversation analysts is that topic talk should not be assumed to be independent of the ‘official business’ of the institution. Like McHoul and Rapley (2000), who analyse ‘chat’ on a com- puter helpline, we would argue that: While the term ‘chat’ may, conventionally, gloss some kind of disengagement from businesslike conversation, while it may seem to be ‘small talk’, of no weight, irrele- vant even, it might rather be (as inspection of some actual materials shows) that it’s in fact ultra-critical to keeping the ‘serious’ stuff happening. (Bite 5, Bit 10) Like McHoul and Rapley’s analysis, ours problematizes any neat distinction between ‘talk at work’ and ‘talk as work’: while topic talk is clearly superflu- ous to removing hair (and, as such, may be said to be ‘talk at work’), it is also a form of ‘talk as work’ – emotional labour. Taken as a whole, then, this arti- cle has addressed both the feminist issue of emotional labour and the conver- sation analytic concern with how participants manage multiple involvements in a socially meaningful way. Offering a single-case analysis, this article is exploratory, providing sign- posts for future work, rather than a comprehensive explication of emotional labour practices. It is more common for conversation analysts to work with a collection of data fragments. However, arguing that the single-case analysis is ‘a central office for sociology, one which has not received the same attention as other of its jobs’ (p. 112), Schegloff (1987) points out that: Social action done through talk is organized and orderly not, or not only, as a mat- ter of rule or as a statistical regularity, but on a case by case, action by action, basis 657Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com … Accordingly, an analytic machinery which is meant to come to terms with the orderliness of interaction … should be able to deal in an illuminating manner with single episodes of talk taken from ‘the real world’ … Whatever concerns for macrosocial issues we entertain, our ways of dealing with them will in the end have to be compatible with a capacity to address the details of single episodes of action through talking in interaction. (p. 102) Going on to suggest that ‘much grander themes can often be … clearly seen’ (1987: 102) in single instances of real-life social action, Schegloff uses a single- case analysis to examine conflict; we have done the same for emotional labour. In doing so within a feminist agenda, this article joins a growing feminist conversation analytic project, which brings the tools of CA to bear on substan- tive feminist concerns (see Kitzinger, 2000; Speer, 2005; Stokoe and Weatherall, 2002). Recent examples include research on coming out as lesbian (Land and Kitzinger, 2005), the construction of HIV/AIDS in lesbian and gay awareness training (Kitzinger and Peel, 2005), calls to a home-birth helpline (Shaw and Kitzinger, 2007), and how psychiatrists and clients in UK gender identity clinics jointly construct the client’s passing as a woman (Speer and Green, 2007). Such work seeks to remain true to the CA principle that analytic claims must be grounded in participants’ practices. At the same time, the aim is to go beyond the ‘basic science’ of how interaction works, using CA to say something of political relevance.3 In blending these aims, feminist CA should offer something to feminists and conversation analysts. Here we have drawn out the implications of our analysis for each audience. Seeking to understand both the mechanics of social action and the socio-political consequences thereof is, in our view, a hallmark of an adequate feminist CA. Within the highly gen- dered, low-paid world of beauty therapy, using our technical analysis to argue for the recognition of emotional labour as skilled is significant. While the skills we demonstrate are hardly ‘spectacular’, that does not dimin- ish their importance. On the contrary, we would argue that their ‘everydayness’ is what makes them crucial: they are not merely required under special circum- stances (e.g. dealing with an upset client), but are fundamental to the routine work of beauty therapy. Ironically, it is their ‘everydayness’ that makes these skills so difficult to recognize. Woven into the fabric of everyday interaction, they are even subtler – and hence more hidden – than the critical literature has suggested. Acknowledgements Special thanks to our participants for allowing us to record and publish their interac- tion. We are grateful to Tony Wootton for inspiring and commenting on this analysis, to Sue Wilkinson for co-supervising Merran’s doctoral research and for feedback on an earlier draft, and to the Editors, anonymous referees and Sue Speer for helpful sugges- tions. This research was started at Loughborough University and completed at the University of York, with funding from the Emma Smith Overseas Scholarship (University of Kwa-Zulu Natal, South Africa) and the Sir Richard Stapley Educational Trust (UK). Thanks to the MRC HSRC for affording Merran time to write this article. 658 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Notes 1 Key to Jeffersonian transcription [ ] Square brackets indicate the points at which overlapping talk begins and ends. = Equal signs indicate no gap between lines/turns. (0.5) Numbers in parentheses indicate silent intervals within or between talk measured in tenths of a second. (.) Dot in parentheses indicates a silence too short to measure in tenths of a second. . Full stop indicates closing intonation. , Comma indicates slightly upward ‘continuing’ intonation. ? Question mark indicates rising intonation. ¿ Inverted question mark indicates rising intonation weaker than that indicated by a question mark. : Colon indicates extension of preceding sound – the more colons the greater the extension. ↑↑↓↓ Up or down arrow indicates marked rise or fall in intonation. here Underlining indicates emphasized talk. HERE Upper case indicates louder talk. °here° Degree signs indicate softer talk. >this< Greater than and less than signs indicate speeded up talk. (h) ‘H’ in brackets indicates audible aspirations in speech (e.g. laughter particles). 2 We are not claiming that the participants themselves are oriented to the per- formance of ‘emotional labour’ in the episode under analysis. Our goal, here, is not to examine ‘emotional labour’ as a members’ term. Nevertheless, we remain focused on members’ own practices as we seek to demonstrate some of the interactional skills involved in performing that which sociologists have dubbed ‘emotional labour’. The relationship between analysts’ and partici- pants’ categories is a contested issue across conversation analysis, discourse analysis, discursive psychology and membership categorization analysis, but unfortunately we do not have space to pursue that discussion here. 3 What makes this analysis an example of feminist CA is, in our view, the fact that we have employed the tools of CA to meet a feminist end: arguing for the recognition of invisible skills for which women are seldom remunerated. Our understanding of beauty therapy and emotional labour as gendered is based, then, on the prior literature; our focus here has not been on gender as a cate- gory in our data. A fascinating extension of our work would be to examine how gender is invoked, displayed and oriented to during the hair removal ses- sions. Unfortunately, this is beyond the scope of this article. We are grateful to an anonymous referee for highlighting this point. References Black, P. (2004) The Beauty Industry: Gender, Culture, Pleasure. London: Routledge. 659Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Bolton, S.C. and C. Boyd (2003) ‘Trolley Dolly or Skilled Emotion Manager? Moving on from Hochschild’s Managed Heart’, Work, Employment and Society 17(2): 289–308. Calhoun, C.C. (1992) ‘Emotional Work’, in E.B. Cole and S. Coultrap-McQuin (eds) Explorations in Feminist Ethics: Theory and Practice, pp. 117–22. Indianapolis: Indiana University Press. Clayman, S.E. and J. Heritage (2002) ‘Questioning Presidents: Journalistic Deference and Adversarialness in the Press Conferences of Eisenhower and Reagan’, Journal of Communication 52(4): 749–75. Drew, P. (1990) ‘Strategies in the Contest Between Lawyers and Witnesses in Court Examinations’, in J.N. Levi and A. Graffam Walker (eds) Language in the Judicial Process, pp. 39–64. New York: Plenum Press. Drew, P. and J. Heritage (1992) ‘Analyzing Talk at Work: An Introduction’, in P. Drew and J. Heritage (eds) Talk at Work: Interaction in Institutional Settings, pp. 3–65. Cambridge: Cambridge University Press. Erickson, F. (1992) ‘They Know all the Lines: Rhythmic Organization and Contextualization in a Conversational Listing Routine’, in P. Auer and A. di Luzio (eds) The Contextualization of Language, pp. 365–97. Amsterdam: John Benjamins. Fineman, S. (1996) ‘Emotion and Organizing’, in S.R. Clegg, C. Hardy and W.R. Nord (eds) Handbook of Organization Studies, pp. 543–64. London: Sage. Frith, H. and C. Kitzinger (1998) ‘“Emotion Work” as a Participant Resource: A Feminist Analysis of Young Women’s Talk-in-interaction’, Sociology 32(2): 299–320. Gimlin, D. (1996) ‘Pamela’s Place: Power and Negotiation in the Hair Salon’, Gender & Society 10(5): 505–26. Goodwin, C. and M.H. Goodwin (1992) ‘Context, Activity and Participation’, in P. Auer and A. di Luzio (eds) The Contextualization of Language, pp. 77–99. Amsterdam: John Benjamins. Heath, C. (1984) ‘Participation in the Medical Consultation: The Co-ordination of Verbal and Nonverbal Behaviour between the Doctor and Patient’, Sociology of Health & Illness 6(3): 311–38. Heath, C. and J. Hindmarsh (2002) ‘Analysing Interaction: Video, Ethnography and Situated Conduct’, in T. May (ed.) Qualitative Research in Practice, pp. 99–121. London: Sage. Heritage, J. (1997) ‘Conversation Analysis and Institutional Talk’, in D. Silverman (ed.) Qualitative Research: Theory, Method and Practice, pp. 161–82. London: Sage. Heritage, J. and J.M. Atkinson (1984) ‘Introduction’, in J.M. Atkinson and J. Heritage (eds) Structures of Social Action: Studies in Conversation Analysis, pp. ix–xvi. Cambridge: Cambridge University Press. Heritage, J. and D.W. Maynard (eds) (2006) Communication in Medical Care: Interaction between Primary Care Physicians and Patients. Cambridge: Cambridge University Press. Hochschild, A. (1983) The Managed Heart: Commercialization of Human Feeling. Berkeley: University of California Press. James, N. (1989) ‘Emotional Labour: Skill and Work in the Social Regulation of Feelings’, Sociological Review 37: 15–47. Kitzinger, C. (2000) ‘Doing Feminist Conversation Analysis’, Feminism & Psychology 10(2): 163–93. 660 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Kitzinger, C. and E. Peel (2005) ‘The De-Gaying and Re-Gaying of AIDS: Contested Homophobias in Lesbian and Gay Awareness Training’, Discourse & Society 16(2): 173–97. Korczynski, M. (2003) ‘Communities of Coping: Collective Emotional Labour in Service Work’, Organization 10(1): 55–79. Land, V. and C. Kitzinger (2005) ‘Speaking as a Lesbian: Correcting the Heterosexist Presumption’, Research on Language and Social Interaction 38(4): 371–416. LeBaron, C.D. and S.E. Jones (2002) ‘Closing up Closings: Showing the Relevance of the Social and Material Surround to the Completion of Interaction’, Journal of Communication 52(3): 542–65. McHoul, A. and M. Rapley (2000) ‘Still on Holidays Hank? – “Doing Business” by “Having a Chat”’, M/C: A Journal of Media and Culture 3(4), URL (consulted August 2006): http://journal.media-culture.org.au/0008/holidays.php Mirchandani, K. (2003) ‘Challenging Racial Silences in Studies of Emotion Work: Contributions from Anti-racist Feminist Theory’, Organization Studies 24(5): 721–42. Pomerantz, A. (1984) ‘Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes’, in. J.M. Atkinson and J. Heritage (eds) Structures of Social Action: Studies in Conversation Analysis, pp. 57–101. Cambridge: Cambridge University Press. Sacks, H. (1995) Lectures on Conversation. Oxford: Blackwell. Sacks, H., E.A. Schegloff and G. Jefferson (1974) ‘A Simplest Systematics of Turn- Taking for Conversation’, Language 50(4): 696–735. Sanders, T. (2004) ‘Controllable Laughter: Managing Sex Work Through Humour’, Sociology 38(2): 273–91. Schegloff, E.A. (1968) ‘Sequencing in Conversational Openings’, American Anthropologist 70: 1075–95. Schegloff, E.A. (1987) ‘Analyzing Single Episodes of Interaction: An Exercise in Conversation Analysis’, Social Psychology Quarterly 50(2): 101–14. Schegloff, E.A. (1996) ‘Issues of Relevance for Discourse Analysis: Contingency in Action, Interaction and Co-Participant Context’, in E.H. Hovy and D.R. Scott (eds) Computational and Conversational Discourse: Burning Issues – An Interdisciplinary Account, pp. 3–38. Heidelberg: Springer Verlag. Sharma, U. and P. Black (2001) ‘Look Good, Feel Better: Beauty Therapy as Emotional Labour’, Sociology 35(4): 913–31. Shaw, R and C. Kitzinger (2007) ‘Memory in Interaction: An Analysis of Repeat Calls to a Home Birth Helpline’, Research on Language and Social Interaction 40(1): 117–44. Silverman, D. (1997) Discourses of Counselling: HIV Counselling as Social Interaction. London: Sage. Speer, S.A. (2005) Gender Talk: Feminism, Discourse and Conversation Analysis. London: Routledge. Speer, S.A. and R. Green (2007) ‘On Passing: The Interactional Organization of Appearance Attributions in the Psychiatric Assessment of Transsexual Patients’, in V. Clarke and E. Peel (eds) Out in Psychology: Lesbian, Gay, Bisexual, Trans and Queer Perspectives, pp. 335–68. Chichester: Wiley. Steinberg, R.J. and D.M. Figart (1999) ‘Emotional Labor Since The Managed Heart’, The Annals of the American Academy 561: 8–26. 661Emotional labour in action Toerien & Kitzinger by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com Stivers, T. and J. Sidnell (2005) ‘Introduction: Multimodal Interaction’, Semiotica 156(1/4):1–20. Stokoe, E.H. and A. Weatherall (2002) ‘Guest Editorial: Gender, Language, Conversation Analysis and Feminism’, Discourse & Society 13: 707–13. Tancred, P. (1995) ‘Women’s Work: A Challenge to the Sociology of Work’, Gender, Work and Organization 2(1): 11–20. ten Have, P. (1999) Doing Conversation Analysis: A Practical Guide. London: Sage. Toerien, M. and C. Kitzinger (2007) ‘Emotional Labour in the Beauty Salon: Turn Design of Task-Directed Talk’, Feminism & Psychology 17(2): 162–72. Whalen, J., D.H. Zimmerman and M.R. Whalen (1988) ‘When Words Fail: A Single Case Analysis’, Social Problems 35(4): 335–62. Wingard, L. (forthcoming) ‘An Interactional Analysis of Emotion Work in Naturally Occurring Family Interaction’. Merran Toerien Is the RCUK fellow in Communication and Language Use in Interaction in the Department of Sociology at the University of York. Her current research centres on CA as applied to institutional talk-in-interaction, including encounters between patients and health professionals, and employment advisers and clients. She has published pre- viously on feminism, gender and the body in Feminism & Psychology, Qualitative Research in Psychology, Sex Roles and Women’s Studies International Forum. Address: Department of Sociology,The University of York, Heslington,York YO10 5DD, UK. E-mail: mgt100@york.ac.uk Celia Kitzinger Is Professor of Conversation Analysis, Gender and Sexuality in the Department of Sociology at the University of York. She has published nine books and around 100 arti- cles and book chapters on issues related to language, genders and sexualities. She is cur- rently working with a corpus of 350 calls to a helpline from women in trauma after childbirth. Address: Department of Sociology, University of York, Heslington,York YO10 5DD, UK. E-mail: cck1@york.ac.uk 662 Sociology Volume 41 ■ Number 4 ■ August 2007 by Sandra Hopps on October 14, 2008 http://soc.sagepub.comDownloaded from http://soc.sagepub.com