Research on Language Teachers’ Emotion Labour and Emotional Well-Being. A Critical Analysis and Suggestions for Further Research

: This critical overview examines the body of literature on language instructors’ emotion labour, with the aim of providing insights into the aspects of teachers’ work that positively and negatively impact their emotional well-being, as well as possible strategies to help them flourish. Research in this novel area was divided into three broadly chronological categories for analysis: pioneering studies, qualitative studies with a specific focus, and more methodologically diverse studies. The findings indicated that societal attitudes influenced the institution, resulting in differing levels of pressure or support experienced by the teachers. Emotion labour was found to stem from teachers’ beliefs conflicting with those held at an institutional level or beyond, as well as from interactions with the learners, and could have positive and/or negative outcomes. One of the worst possible outcomes is burn-out. Agency, empathy, and reflection were among the strategies to emerge with the potential to neutralise negative outcomes. The article concludes with suggestions for further research.


Introduction
Having gained momentum since the start of the millennium, we have witnessed an explosion of research into language learner emotions in the last decade (Dewaele et al., 2019).As a result of this "learner-centred movement", the other key stakeholders in the language classroom, namely teachers, were for a time largely disregarded by applied linguists (Mercer & Kostoulas, 2018: 1).This imbalance is starting to be redressed, however, with attention increasingly being paid to the psychology and emotional lives of language instructors (Gkonou et al., 2020).
Extending knowledge in this area has the potential to benefit not only teachers, by enhancing their level of job satisfaction and emotional well-being, but also learners, given the impact the instructor's mental state has on the quality of his/her teaching and the students' own emotions (Mercer, 2018).
From the scarcity of research into language teacher emotions a decade ago, we now have an expanding body of literature that has already provided some useful insights.The introduction of positive psychology to applied linguistics research has been influential, with several studies exploring different aspects of teacher well-being and the role played by emotion regulation (Greenier et al., 2021;Morris & King, 2018, 2020;Talbot & Mercer, 2018) and Trait Emotional Intelligence (TEI) (Dewaele & Wu, 2021;Oxford, 2020).Much of the research has examined the internal and external threats to teachers' emotional well-being (Gkonou et al., 2020).A recent meta-analysis has shown that "social relationships seem to play a pivotal role" in teachers' emotional well-being (Hasher & Waber, 2021: 1).Teachers' emotion labour is a prime example of a strategy used to survive in situations of heavy social pressure.It is therefore this line of research that is subject to examination in this literature review.
Through the lens of emotion labour, this critical overview seeks to understand the aspects of language teachers' work that positively and negatively affect their emotional wellbeing, and to uncover possible strategies to help them flourish in their professional role.The first section summarizes the theory behind emotion labour and how the concept has evolved.This is followed by a critical synthesis of the research literature on language teachers' emotion labour, which has been divided into three categories: pioneering studies, qualitative studies with a specific focus, and more methodologically diverse studies.It concludes with an overview of the main findings and suggestions for future research.

Theoretical Background Emotional labour
The concept of emotional labour was first introduced by the sociologist Arlie Hochschild (1983), for whom it refers to commodified emotion management in the workplace that "requires one to induce or suppress feeling in order to sustain the outward countenance that produces the proper state of mind in others" (p.20).Emotional consonance, on the other hand, results when "somebody effortlessly feels the emotion that is required in a certain situation" (Näring et al., 2006, p. 304) and could thus be defined as a lack of emotional labour.Common to both constructs is the influence external beliefs or pressures exert on an individual's emotions, with emotional labour taking place when the person experiences a "pinch" between their own feelings and those mandated by the social feeling rules (Hochschild, 1983: 50).According to Hochschild (1983), a feeling rule, namely a "script or moral stance" toward feeling, "is one of culture's most powerful tools for directing action" (p.50).It is through these emotional conventions, which are often implicit, that people gauge the appropriateness of certain emotions in a particular context.Hochschild (1983) posits that when people's true feelings differ from the relevant feeling rules they manage their emotions in two ways, by surface acting, namely altering their outward appearance, or deep acting, which involves displaying "a real feeling that has been self-induced" (p.38).Some scholars have since expanded upon or modified this taxonomy, for example Näring et al. (2006), who added a third type of emotion labour, suppression of emotion, to Hochschild's original two.

Emotion Labour in Education
In the relatively novel field of language teacher emotion research, there has been a shift "from a cognitive to sociocultural and poststructuralist approach to emotions" (Her & De Costa, 2022: 1), and this is apparent in emotion labour studies.According to poststructuralists, "historical, cultural, contextual, and sociopolitical" factors are significant when theorizing emotions (Gkonou & Miller, 2021: 135) and the emphasis is therefore on the broader societal context and the relationship between emotions and power (Benesch, 2018).From this perspective, "the ways in which humans perceive and refer to emotions is a discursive process" involving local feeling rules The European Educational Researcher | 45 (Benesch, 2018: 61).This view contrasts with biological and cognitive approaches, which regard emotions as innate and universal, or as individual psychological states, respectively (Gkonou et al., 2020).
In light of the poststructural approach, one of the pioneers of language teacher emotion research, Sarah Benesch (2017), revised the conceptualization of emotional labour by retaining some of Hochschild's ideas, but modifying others.Although both concepts assume notions of tacit workplace feeling rules and the relationship between emotion labour and power imbalances, Benesch adds the idea that unequal power can be resisted.Another modification by Benesch (2020) was the term used, with her deciding to substitute "emotional" with "emotion" to avoid its negative connotations (p.56).She also posits that emotion labour better underlines the link between emotions and power relations and represents a shift away from considering emotions as individual psychological states (Benesch, 2017).For the purposes of this literature review, the term emotion labour will be used unless specifically referring to Hochschild's (1983) concept.

Pioneering Studies
Although half a century has passed since Hochschild first introduced the concept of emotional labour, scholars from the field of language education only started showing an interest in teachers' emotion labour comparatively recently.
The key early studies exploring this line of research were those by Acheson et al., King, and Loh and Liew, all published in 2016.
In their study, Acheson et al (2016) explored the emotion labour experienced by five FL teachers at a high school in the US state of Georgia.Analysis of the interview data revealed several key findings regarding contextual factors affecting emotion labour, reasons for using it, and possible outcomes.One insight was that the teachers perceived the community and institution as largely unsupportive of their subject due to the negative attitudes towards foreign languages and the mentality of English as a lingua franca found in that setting.This led to them facing a considerable burden to motivate the students and having to combat prejudicial attitudes.In this context, the main reasons the teachers gave for performing emotion labour were improving engagement, caring for the students, and maintaining a positive relationship with them.In terms of the outcomes of this emotion labour, there was evidence of emotional exhaustion and a lack of self-efficacy resulting from their work demands, which the authors posited formed a downward spiral potentially leading to burnout and/or attrition.
For his part, King (2016) interviewed five EFL teachers at a private university in Japan to discover how they perceived their emotion labour, and how their emotions were influenced by contextual factors.He also sought to understand their coping mechanisms, including the strategies used to manage their emotions when teaching.The findings indicated that the teachers valued having "caring relationships with their students" (King, 2016: 102), and suppressed negative emotions to maintain a positive learning environment.Furthermore, they reported inducing or amplifying positive emotions to motivate the students, a task rendered more burdensome due to it being "primarily their responsibility" (King, 2016: 105).Regarding the strategies used, there was evidence of surface acting combined with a sense of detachment, and emotional or even physical distancing from the students.Finally, the data indicated how a change in the university's management at a senior level and subsequent reforms in terms of the curriculum, the course structure and the administrative tasks expected of staff impacted negatively on the teachers' emotions, with them feeling vulnerable, unappreciated, and distrustful.
In their study, Loh and Liew (2016) sought to uncover the emotional labours specific to English teachers in the particular setting of multilingual Singapore, where English is the main language of instruction.Data collected from ten secondary school English teachers indicated several "areas of struggle" relating to curriculum and policy where conflict resulted in emotion labour (Loh & Liew, 2016, p. 271).Firstly, some instructors reported struggling to reconcile the importance they attached to teaching literature, culture, and values with the curriculum's emphasis on grammar and comprehension.Secondly, there was evidence of tension between attending to their beliefs and training about pedagogy, and to institutional and parental expectations.The relative importance of English grades in Singapore and its "culture of performativity" intensified the pressure to focus on preparing students for tests, rather than fostering creativity and critical thinking (Loh & Liew, 2016: 273).Grading student essays was a third area of struggle that emerged, with emotion labour relating to the burden of marking, the subjective nature of evaluating writing, and responding appropriately to any emotional issues written about in the students' essays.Finally, Singapore's one-size-fits-all curriculum diverged from teachers' training and their desire to be "culturally responsive" (Loh & Liew, 2016: 275).
Along with their methodological similarities, namely their use of semi-structured interviews, the studies adopted comparable theoretical stances based on Hochschild's (1983) emotional labour concept.This is most evident in Acheson et al. (2016) and King (2016) which both report findings related to Hochschild's (1983) emotional labour strategies.In the former study, deep acting seemed most prevalent, although there was also evidence of surface acting, including emotional suppression, and least of all emotional consonance.As for King (2016), comments by one of the participants indicate him performing surface acting by faking emotions.
Given language teachers' emotion labour being "a new conceptual direction within L2 research" (King, 2016: 98), all three studies provided novel insights into emotional aspects of language teachers' work.Their findings depended on the focus of the research and thus the questions asked of the participants, and the studies often differed in this regard.Acheson et al. (2016) examined the relationship between emotion labour and attrition and focused on emotions in the classroom, student-teacher relationships, and student motivation.For his part, King (2016) also investigated in-class emotions but chose not to limit the lines of inquiry with predetermined research questions.
Finally, Loh and Liew (2016) investigated emotions inside and outside of the classroom, with a specific focus on marking writing and instances where the teachers' values and their work-related duties diverged.
The European Educational Researcher | 47 In addition to the aims and specific foci differing overall, the contexts in which the studies were conducted also varied in terms of the country and education level.Despite the contrasting settings, some commonalities emerged from the findings related to emotion labour and teacher-student relationships in Acheson et al. (2016) and King (2016), suggesting that they might be applicable to language teachers as a population.In both studies, teachers reported not only regulating their emotions to motivate and/or show care for the learners, but also suppressing them in interactions with difficult students.Interestingly, while inattentive students triggered a negative emotional response in the two settings, in the US, needy learners also had a similar effect (Acheson et al., 2016) and in Japan, those who were simply silent and not participating (King, 2016).Another example of a contextual difference was the finding in Acheson et al. (2016) of a perceived lack of community and institutional support for foreign languages in the US due to the global dominance of English, negative parental attitudes and the importance placed on other subjects.There was no mention of a similar issue in the other two studies, where English was the language taught, presumably due to it being a global lingua franca (King, 2016;Loh & Liew, 2016).
Although there was divergence regarding the extent of support for the particular language being taught, there was common ground in terms of a perceived lack of institutional support for the language teachers, who felt the burden of motivating students (Acheson et al., 2016;King, 2016) or of raising English standards (Loh & Liew, 2016).
Furthermore, a lack of teacher autonomy and agency was evident in the findings of both King (2016) and Loh and Liew (2016).

Qualitative Studies with a Specific Focus
The three studies published in 2016 could be said to have laid the groundwork for subsequent research into language teachers' emotion labour, pursuing as it did various lines of inquiry that they had uncovered (Acheson et al., 2016;King, 2016;Loh & Liew, 2016).These included language teachers' emotional experiences of everyday classroom teaching, the relationship between institutional practices and emotion labour, and the impact of policies and reforms on teachers' emotions.The empirical studies reported on in this section are categorised accordingly.As was the case for the three aforementioned pioneering studies, the researchers used semi-structured interviews to collect their data.

Language Teachers' Emotional Experiences of Everyday Classroom Teaching:
Gkonou and Miller's (2019) study used a sample of eight EFL teachers working at a private language school in Greece to investigate the instructors' efforts to mitigate their students' language anxiety.Referring to positioning theory, the authors found that the participants "positioned themselves as caring teachers" who worked consciously to create positive relationships with the students as a means of lessening their language learning anxiety (Gkonou & Miller, 2019: 381).There was evidence of an ethic of care (Noddings, 1988) being a core aspect of both their identity as teachers and their professional responsibilities.The authors suggested that while showing care can lead to positive outcomes, such emotional aspects of teaching commonly result in emotion labour due to the teachers suppressing their own feelings in the process.
Adopting a more positive slant on emotion labour, Miller and Gkonou (2018) investigated the role of emotion labour and agency in producing emotional rewards among teachers working on tertiary-level English language programmes in the UK and the US.Data from the questionnaire, which asked the thirty respondents to select six emotion words from a possible twenty, indicated that positive ones dominated.In the interviews, while the participants reported emotional stress caused by the demands of managing relational aspects of their job, they were invested in a "teaching-as-caring" approach and benefitted psychologically from building a rapport with their learners (Miller & Gkonou, 2018: 54).Another finding was that experience contributed to the teachers being better able to manage their emotions as they learnt to distance themselves emotionally or physically from difficult situations.The authors suggested that reflexive practices, and an awareness of external expectations of caring can lead teachers to take agentive action and resist feeling rules.
The interview accounts collected for the aforementioned study were also used as data for Gkonou and Miller's (2021) research paper into the link between emotion labour and emotional capital, and the role of reflection in developing the latter.The first finding was that the teachers were aware of the classroom feeling rules and tried to act accordingly, for example through emotional or physical detachment or regulating their emotions.The authors posited that this awareness is indicative of the teachers' increasing emotional capital, which provides them with the "resources necessary to thrive as language teachers" (Gkonou & Miller, 2021: 148).Secondly, it emerged from the interviews that emotionally challenging situations gave rise to reflection, either individually or with colleagues, which enabled the teachers to find solutions.Reflection therefore assisted them in developing the capacity to orient themselves towards the feeling rules.

The Relationship Between Institutional Practices and Emotion Labour:
In Benesch's (2017) seminal work exploring language teachers' emotion labour, one focus of her research was on the teachers' beliefs and feelings about high-stakes literacy testing.As with all four of the studies in that book, the sample was composed of 15 English language instructors at a US university, 13 of whom responded to questions in interviews and the remaining two via questionnaires.The author posited emotion labour potentially stemming from several areas of conflict.The first was the "irreconcilability" between the teachers perceiving literacy testing as useful in general but having misgivings about the particular tests being used (Benesch, 2017: 71).The second was them feeling powerless and frustrated due to inequalities, such as using the same questions to test L1 and L2 English speakers.Another related to the impact of the tests on the participants' teaching, with some reportedly feeling ashamed about teaching to the test.Benesch (2017) concluded that the participants' emotion labour is primarily concerned with the impossible burden placed on them to ensure successful linguistic outcomes for the students, irrespective of the students' situation or background.
In another chapter of the same book, Benesch (2017) focused on teachers' feelings associated with responding to students' writing.Analysis of the data indicated teacher emotion labour arising from conflicts about the types of feedback to give, the teacher's role when responding, and relational aspects of commenting.More specifically, the The European Educational Researcher | 49 interviewees were uncertain whether to give written corrective feedback (WCF) and, if so, how much and what type to give (on language or content), and whether it would lead to improvements in the students' writing.As regards their role, the teachers had difficulty reconciling whether to focus on comprehension or accuracy when responding.
Finally, instances of students' work where they seemingly had not learnt "what had been taught repeatedly" nor spent sufficient time on the task were also considered to cause emotion labour (Benesch, 2017: 100).
In a later study on the same topic using the same interview data, Benesch (2020) focused on only two of the emotion labour themes which had emerged previously from the interview accounts, the first being the time spent reading and responding to student writing, and the second the type of feedback to offer.Her aim in this study was to demonstrate the usefulness of Ahmed's (2004) sticky objects and emotion labour in drawing attention to sociopolitical influences on language teachers' emotions.In her work on the cultural politics of emotion, Ahmed (2014) states that it is objects of emotion that circulate, as opposed to feelings themselves, and it is these objects that "become sticky, or saturated with affect, as sites of personal and social tension" (p.12).Thus, rather than viewing emotions as intrapersonal experiences, Benesch (2020) suggests that the concept of sticky objects shifts the focus to emotionally charged teaching activities, and that of emotion labour to working conditions in need of improvement.

The Impact of Policies and Reforms on Teachers' Emotions:
Institutional policies were also subject to investigation by Benesch in her 2017 book.In her study into language teachers' responses to the student attendance policy, she focused on three areas: the university's policies regarding lateness and absence, the teachers' implementation (or not) of these rules, and the emotions they felt when students were late or absent.The findings indicated that emotion labour stemmed from the participants feeling conflicted when there was dissonance between the university's practices and their preferred ones, notably when taking into account the challenges faced by some of the students.However, emotion labour was also apparent in related teacher-student interactions, with participants feeling displeased if late arrivals disrupted the class or uneasy when the students were absent without explanation.
The final line of research studied in Benesch's (2017) book is that of teachers' emotion labour and institutional student plagiarism policies.Subsequently, Benesch (2018) published another study using the same data, this time focusing on the role of feeling rules and emotion labour in engendering teacher agency.Analysis of the sample's responses indicated that emotion labour arose from the teachers' uncertainty about how to deal with instances of plagiarism, due to differences between how plagiarism was viewed institutionally and personally (Benesch, 2017(Benesch, , 2018)).Whereas the policy's implicit feeling rules were "hypervigilance, indignation, and retribution" (Benesch, 2018: 64), the instructors reported not only being less convinced that plagiarism constituted a deliberate and serious act, but also feeling empathy for the students given the difficulty of writing in an additional language (Benesch, 2018).Some participants even considered the policy counterproductive as it diverted attention from more pressing concerns, and felt frustrated as a result (Benesch, 2017).
In addition to emotion labour arising at a teacher-institution level, findings from the earlier study indicate that interactions with the students around plagiarism also resulted in teachers regulating their emotions (Benesch, 2017).
For instance, there was evidence of emotion labour arising from teachers not wishing to upset students when commenting on plagiarism, but also when the individual denied copying even though it was obvious.
Other scholars to have pursued a similar line of research are Her and De Costa (2022), whose case study of a community college instructor in the US investigated his response to a new state-wide language policy.This law, AB 705, signified "a shift towards student autonomy" with the learners carrying out self-guided assessment and selfplacing rather than teachers deciding on the appropriate course based on standard placement tests as had previously been the case (Her & De Costa, 2022: 5).This change provoked emotion labour for the participant in two ways: firstly, due to tension between the policy, which demanded trust in the students' judgements, and his own professional opinion, and secondly, because, as a part-timer, he felt insecure about his position and unable to resist the feeling rules.He used empathy and his spirituality to overcome these challenges, thereby accruing emotional capital.
De Costa et al. ( 2020) explored the impact of a different shift in education, in this case the emphasis on English as a medium of instruction (EMI), on two language teachers working in Nepalese public schools.The study found that emotion labour arose from feeling obliged to use only English in lessons, receiving insufficient guidance or support from teacher educators, and being unable to exercise agency and autonomy in their teaching.Analysis of the findings at three levels revealed the teachers being restricted at a personal level in terms of choice of teaching materials and language for instruction, lacking interaction with colleagues and support at an institutional level, and feeling limited pedagogically and professionally due to the importance attached to English at a societal level.These factors contributed to the participants' emotional exhaustion, which eventually led one of them to leave the profession.
While adding little to the field in terms of diversity of research design, the studies detailed in this section have contributed significantly to teachers' emotion labour research both in terms of theoretical advances and by corroborating or supplementing previous findings.Following on from the exploratory studies, the key theoretical advance was Benesch's (2017) adaptation of Hochschild's (2012) concept of emotion labour to fit the EFL context and a poststructural-discursive approach.Her notion and approach have since been adopted in most research into language teachers' emotion labour, leading to increased focus on the role of power relationships and external influences, and a move away from emotion labour strategies.
As far as findings relating to teacher-student relationships are concerned, the link between teachers' emotion labour and their personal desire and/or professional need to show care found in Acheson et al. (2016) and King (2016) was echoed in Gkonou andMiller's research (2019, 2021;Miller & Gkonou, 2018).There was, however, greater emphasis on external pressure in the later studies, reflecting the poststructuralist approach they adopted.Other commonalities were the findings that teachers perform emotion labour in order to create good relationships with their students and when faced with challenging student behaviour, whether it be in the form of anxiety (Gkonou & Miller, 2019), hostility, or emotional outbursts (Gkonou & Miller, 2021).Similarly, Benesch (2017) reported teachers' emotion labour when dealing with late or absent students, interacting with learners about plagiarism, and responding to student writing.
Regarding the role of institutions, numerous studies have continued the work started by Loh and Liew (2016), with many comparable findings being reported.For example, when investigating emotion labour and responding to student writing, Benesch (2017Benesch ( , 2020) ) found emotion labour resulting from the time spent grading and the amount of written feedback.In addition, as in King (2016) and Loh and Liew (2016), the participants in both Benesch (2017) and De Costa et al. ( 2020) reported a lack of autonomy in their teaching and feeling conflicted about the curriculum's focus on exam preparation in a high-stakes testing context.Finally, the finding of teachers feeling unsupported that emerged in Acheson et al (2016) was also evident in De Costa et al.'s (2020) study.
In addition to similar antecedents to emotion labour being found in the exploratory and ensuing research, there was some overlap in terms of possible outcomes.The negative consequences of emotion labour, such as emotional exhaustion (Acheson et al., 2016) and difficulty investing emotionally in teaching (King, 2016), were similarly apparent in later qualitative studies (De Costa et al., 2020;Her & De Costa, 2022).On the other hand, a novel perspective was adopted in several research papers, perhaps influenced by the positive psychology movement.

More Methodologically Diverse Studies
When investigating the literature on language teacher emotion labour, it emerged that out of the rather limited number of studies that had been published, the vast majority adopted a qualitative research design and relied solely on semi-structured interviews for data.Despite the scarcity of studies using longitudinal or quantitative methodologies or more than one instrument, there are a few exceptions to this, and these are detailed below.
The first of two qualitative studies employing more than one instrument is that by Humphries (2020).In this chapter, he reanalysed lesson observation and interview data collected for prior research to gain insights into the impact of new communicative language teaching (CLT) textbooks on a Japanese teacher of English (JTE).From the lesson observations, the researcher posited student behavioural issues and the teacher deviating from the textbook as suggestive of emotion labour.He also noted the participant displaying negative emotions, which could be interpreted as evidence of emotional burnout.For his part, the JTE reported low emotional demands in terms of interactions with students and institutional feeling rules but struggling with the new textbooks and a lack of direction and support.
Kocabaş-Gedik and Ortaçtepe Hart (2021) used journal entries, interviews, and field notes in their longitudinal case study of two native English-speaking teachers (NESTs) at a Turkish university.Their aim was to investigate the novice EFL teachers' trajectories in terms of identity construction and factors contributing to their emotion labour.
Despite broadly similar experiences in their first year of teaching, the findings revealed differing levels of emotion labour and outcomes, which were linked to three factors.The participant who had trained in ELT, spoke the local language, and had a more supportive mentor demonstrated a higher level of investment and participation in the community of practice.In contrast, the other NEST, who did not have the benefit of those experiences, struggled emotionally due to the divergence between her expectations and the reality of the situation.Acheson and Nelson (2020) used a quantitative research design to investigate the amount of emotion labour engaged in by a sample of FL teachers in the state of Georgia and the effect of certain demographic factors on how frequently they performed different types of emotion labour (Generating Emotions, Pretending Emotions and Hiding Emotions).The variables being explored comprised both teacher individual differences (gender, amount of teaching experience, language taught and native speaker status) and contextual factors (the school's income level and setting).
The findings suggested emotion labour, both overall and the different types, being used to a moderate extent on average.Regarding the demographic factors, although the authors had hypothesized differences in the impact on emotion labour between the subgroups for each variable, the school setting was the only one to emerge as significant.Seemingly, while emotion labour was broadly similar in rural and suburban schools, teachers working in urban settings experienced less emotional burnout.For the other variables, there was no apparent correlation between their subgroups and emotion labour overall or the three types.
In their quantitative research based on a sample of Chinese EFL teachers, Dewaele and Wu (2021) examined the relationship between a range of variables and two emotion labour strategies, Surface acting and Expression of naturally felt emotions.Unlike Surface acting, where emotions are faked or suppressed, the strategy Expression of naturally felt emotions involves someone instinctively experiencing emotions and displaying them despite the effort involved (Diefendorff et al., 2005).The study investigated numerous possible predictors related to teacher individual differences (i.e., sociobiographical, behavioural, attitudinal, linguistic, and psychological factors), as well as two institutional variables.The strategy Expression of naturally felt emotions was found to be most strongly predicted by two teacher-internal variables, the teacher's attitudes towards the students and Emotionality (a factor of Trait Emotional Intelligence (TEI) 1 ) and the two institutional factors, namely the importance ascribed to the teaching of English and the extent of the teacher's influence over the content and skills taught.As for Surface acting, the data indicated a negative correlation between it and another factor of TEI, Sociability.
The studies in this section indicate a possible trend towards more diversity in the field in terms of both research design and theoretical approach.
One methodological advance evident in the two qualitative studies described in this section is the use of triangulation, which can help strengthen a claim when the data indicate similar findings (Perry, 2017).By using journal entries and follow-up interviews in their study, Kocabaş-Gedik and Ortaçtepe Hart (2021) were able to draw upon what the participants had written to delve further into the themes that emerged.The researchers' field notes allowed them to reflect on their own thoughts and feelings about responses given.In a similar vein, carrying out lesson observations followed by interviews enabled Humphries (2020) to base his discussions with the participant around what he had observed and to compare his perspective with the teacher's.Interestingly, there were instances where the instructor's behaviour seemingly contradicted some of his comments, such as the need for teachers to enjoy lessons.This suggests that either he was unaware of his countenance or that the emotions he was feeling differed from the researcher's interpretation.
Kocabaş-Gedik and Ortaçtepe Hart (2021) also diverged from previous language teacher emotion labour research in their use of a longitudinal design, which both Gkonou and Miller (2021) and Dewaele and Wu (2021) called for in their studies.Whereas its counterpart, cross-sectional research, provides only a snapshot of a phenomenon, longitudinal research allows for studying how something changes over time (Dörnyei, 2007) and therefore adds another dimension to a field of research.It can be used "to describe patterns, and to explain causal relationships" (Dörnyei, 2007: 79), hence it being chosen for the aforementioned study which examined novice teachers' career trajectories.
Other studies which exemplify a greater variety of research design being used are those carried out by Acheson and Nelson (2020) and Dewaele and Wu (2021), both of which employed quantitative methods.They were thereby answering Acheson et al.'s (2016) call for "generalizable quantitative data" (p.533), which might allow them to "uncover facts and truths in an objective way by delineating patterns" (Wei, 2014: 13).While they converged in using a questionnaire to collect data, the scales they employed to measure emotion labour differed, with Theoretically, Dewaele and Wu (2021) was notable due to it adopting "the more encompassing perspective" (p. 2) of emotion labour, which combines professional, public, and private elements.Unlike the other studies in the field which consider emotion labour as distinct from or a type of emotion regulation, they drew on theory from both emotion labour and emotion regulation research in their paper, for example by referencing Diefendorff et al's (2005) emotional labour strategies and Gross and John's (2003) emotion regulation processes.
Despite the small number of studies in this section, they nonetheless furthered our understanding of emotion labour by supporting previous findings and offering some new ones.In the two qualitative studies, the JTE in Humphries (2020) and one of the teachers in Kocabaş-Gedik & Ortaçtepe Hart (2021) struggled in part due to a lack of support and guidance, which reflects past research.Furthermore, there was further evidence of student behavioural issues giving rise to emotion labour in Humphries (2020).New insights were the role played by the teacher's educational background and competence in the local language in the level of emotion labour experienced by NESTs teaching abroad (Kocabaş-Gedik & Ortaçtepe Hart, 2021).From the quantitative studies it emerged that the school setting may influence the extent of emotion labour, with teachers in urban schools experiencing less burnout (Acheson & Nelson, 2020).Another novel finding was the link between certain teacher-internal and two institutional variables and different emotion labour strategies (Dewaele & Wu, 2021).

Conclusion
Upon examining the body of literature on language teachers' emotion labour, several patterns emerged regarding aspects of teachers' work that positively and negatively affect their emotional well-being.From a top-down perspective, societal attitudes towards the language in question seem to play an important role as they filter down to the institution and then to the classroom both formally through policies, and informally.Depending on the importance placed on the language, there can be pressure from the institution and/or a lack of support, leading to the teachers bearing the brunt of the responsibility for motivation and achievement.In this context, emotion labour frequently results from a divergence between the teachers' values, beliefs, and training, and the local feeling rules about their professional duties.There is evidence of this conflict in areas such as the curriculum, pedagogy, and assessments, when the teachers have little autonomy or influence over institutional policies and practices or when there is a lack of empathy for the learner's situation.In everyday classroom teaching, the studies indicate that instructors often perform emotion labour in interactions with students, either to foster motivation, build positive relationships and show care, or when dealing with challenging student behaviour.
In addition to uncovering possible sources of emotion labour, research in this field provides insights into the effect it has on teachers and possible strategies which can lessen its negative consequences.While emotion labour might lead to negative outcomes, including emotional exhaustion, burnout and attrition, there is also evidence of its potential benefits, namely emotional rewards, emotional capital, and teacher agency.The type and extent of its impact depends on both personal factors, such as the teacher's personality and beliefs, and contextual ones.Possible strategies to decrease emotion labour that emerged from the studies were the teacher emotionally and/or physically distancing him/herself, using empathy or spirituality, and engaging in reflection.Experience was also shown to play a role, with the ability to manage emotions increasing over time.
While the existing research into language teachers' emotion labour has revealed numerous valuable insights into factors affecting their emotional well-being, there remains much still to be discovered.Increased methodological variation, including more quantitative and longitudinal research, would augment our understanding of patterns across groups and over time.Moreover, further investigations in differing settings and into languages other than English would allow for a greater awareness of contextual variations.In order to obtain as robust findings as possible, future studies would benefit from using triangulation, having a mixed-method research design and including students' perceptions.Theoretically, more is needed regarding emotion labour and its strategies to enable comparison between studies.Finally, given the potential risks to their well-being, it is essential that future emotion labour research provides teachers with the strategies and solutions to enable them to not only manage emotionally in the long-term, but to flourish.
Acheson et al. (2020) using a modified version ofBrotheridge and Lee's (2003) Emotional Labour Scale and Dewaele and Wu (2021) basing theirs on Yin et al.'s (2013) and Yang et al.'s (2019) emotional labour strategy scales.The strategies examined reflected this divergence, with the former studying deep acting, positive surface acting and negative surface acting (termed Generating, Pretending and Hiding emotions respectively) and the latter Surface acting, Deep acting and Expression of naturally felt emotions.The use of different scales and terminology somewhat impedes comparison between the two studies in relation to findings on emotion labour strategies.