No emotion is an island: an overview of theoretical perspectives and narrative research on emotions in sport and physical activity

Abstract

Within sport and physical activity settings emotions have typically been conceived and explored from an individualistic or intrapersonal perspective, although researchers are increasingly turning their attention to the interpersonal aspects of emotions and emotion regulation. In this paper, we provide a theoretical overview of the social or interpersonal aspects of emotions from a psychological perspective, and we also consider theoretical perspectives of emotion as intersubjective, social, performative and embodied. We then provide a review of narrative research on emotion in sport and physical activity contexts and provide suggestions for future research in this area. We suggest that narrative approaches can advance research on emotions in sport and physical activity by exploring how emotions arise within the context of social relationships; by exploring how emotional stories or narratives function and are used by athletes, coaches, and others within sport and physical activity contexts; by examining how emotions are created, recreated, and sustained through the stories people tell; by examining how collective and group-based emotions are intertwined with one’s identity and identity development; and by highlighting the ways in which social and cultural narratives within sport shape athletes’ emotional experiences. We conclude by describing some challenges we have faced in conducting qualitative research from a narrative lens, and we describe how we have navigated these issues as a way of offering some ‘lessons learned’ from our own research.

Keywords:

• Narrative analysis
• emotions
• sport
• physical activity
• affect

Introduction

The study of emotions is replete with tensions among theorists and researchers who have adopted varying, and at times conflicting, positions in their attempt to understand the nature, experience and function of emotions. For example, debates in the literatures concern the conceptualisation and nature of emotions as ‘basic’ or as comprising emotional ‘blends’ or schemas (Izard 2009). Researchers have explored neurological underpinnings of emotions (Lindquist et al. 2012), physiological activation (Lane and Nadel 2000) and facial expressions (Matsumoto et al. 2008) during emotional episodes, the socialisation and development of emotions in childhood and adolescence (Michalson and Lewis 1985), cross-cultural similarities and differences in the experience and expression of emotions (Schweder et al. 2008), emotional climates (de Rivera 1992), emotional labour and the management of emotions within social contexts (Hochschild 1983), as well as the social, cultural, and historical contexts in which emotions occur (Gergen 2009, Burkitt 2014).

This diversity of approaches to studying emotions is reflected in research within sport and physical activity, as researchers have adopted various perspectives to explore anxiety versus discrete emotions (e.g. Cerin 2003), telic and paratelic flow states (e.g. Swann et al. 2016), individualised zones of optimal functioning (e.g. Pellizzari et al. 2011), while Dupee and Werthner (2011) have explored the use of biofeedback and its associations with emotions among athletes. Additionally, researchers have explored emotional labour and emotional intelligence among coaches (Lee and Chelladurai 2015), perceptions of the team emotional climate among injured athletes (Mankad et al. 2009), and the performance of emotions among professional athletes (Gallmeier 1987). We are mindful of this diversity and the debates surrounding the study of emotions, and we view this range of approaches as useful and informative about different aspects of emotions in sport and physical activity settings. In this article, we hope to contribute to the ongoing conversation about the nature and function of emotions by providing an overview of theoretical perspectives of emotions, with an emphasis on social and interpersonal aspects of emotions. We then provide an overview of narrative research on emotion in sport and physical activity, drawing attention to ways in which this research contributes to our understanding of emotions as social and relational. We conclude by describing challenges in this area of research, and we suggest some ‘lessons learned’ from our own research endeavours in grappling with these topics and methods.

Psychological perspectives: moving from individual to social views of emotions

The terms ‘emotion’ and ‘affect’ are sometimes used interchangeably within and across bodies of literature, and it is beyond the scope of this paper to provide a review of all the ways in which these terms have been used and how they inform inquiry in sport psychology, sociology, and physical cultural areas. However, within a cognitivist psychological tradition, emotions are typically described in terms of their subjective experience, associated behaviours or action tendencies, and patterns of physiological activation (Gross 2008). Emotions are viewed as unfolding over relatively brief periods of time and in response to particular objects, events, or circumstances, whereas moods refer to higher order, more diffuse and pervasive states (Gross 1998). Gross (1998) also distinguished emotions and moods from affect, which he considers as encompassing any emotionally valenced state.

Central to many contemporary psychological views of emotion is the notion that emotions are the result of meaning-making activities (Lazarus 1999, Feldman Barrett 2006, Gross 2008). Lazarus (1999) noted that the essence of his cognitive-motivational-relational theory of emotion was the process of appraisal, which concerns the way people construe the significance of what is happening for their well-being, and what might be done about it. Thus, the interpretation of what is happening in relation to one’s goals, values, and beliefs is central to emotional experiences. Gross (2008, p. 498) echoes the work of Lazarus and noted that ‘emotions are thought to arise when an individual attends to a situation and understands it as relevant to his or her current goals’. Outlining her conceptual act perspective of emotions, Feldman Barrett (2006) also positioned meaning-making as a central process in the experience of emotions: ‘people experience an emotion when they conceptualise an instance of affective feeling. In this view, the experience of emotion is an act of categorization, guided by embodied knowledge about emotion’ (Feldman Barrett 2006, p. 20). Within this perspective, Feldman Barrett (2006) suggests that emotional life consists of two components – core affect and conceptual knowledge. Core affect is the ongoing, changing neurophysiological state that is available for conceptualisation or categorisation. Core affect is the building block of emotional life; however, it is the combination of core affect with conceptual knowledge about emotions (acquired through experience and language) which produces emotional experience (Feldman Barrett 2006). Thus, the process of meaning-making about the significance of events and one’s responses to them is generally considered to be central to the experience of emotions within most cognitivist psychological perspectives.

The emphasis on appraisal and meaning-making has typically implicated individualistic approaches to studying how people make appraisals and experience emotions (Locke 2003); however, psychology researchers are increasingly acknowledging the social or interpersonal aspects of emotional processes. Within a psychological perspective of emotion, we summarise at least four ways in which emotions can be considered as social phenomena: (a) appraisals (and consequently emotions) arise through social interactions; (b) collective emotions and group-based emotions arise as a function of appraisals with specific reference to our social relationships, and emotions appear to be ‘contagious’ between people in groups and teams; (c) emotions serve social functions within groups; and (d) emotions are experienced, displayed and managed within social contexts.

The first way in which emotions can be considered social phenomena concerns the importance of social interactions within the appraisal process. In a discussion of the history of research on stress, Lazarus (1999, p. 29) noted that ‘stress takes place at one’s job, in one’s home, and in school – in effect, anywhere people worked with each other or had close relationships as, for example, co-workers, family members, lovers, friends, students, and teachers’. Thus, appraisals occur within the context of interpersonal relationships: it is within these personal relationships and through interactions that goals, values and beliefs are made salient, giving rise to emotions. Extending this idea to sport and physical activity contexts, we would add that appraisals and emotions also rise in situations where people have relationships and interactions with others as competitors, teammates, coaches, and officials, or perhaps as physical therapists, parents, and spectators. Sport and physical activity contexts are social settings, with ample opportunity for social interactions that may include praise, support, comparisons and evaluations, punishment, and conflict. These interactions may implicate athletes’ goals, commitments, and beliefs (e.g. concerning achievement, self-worth, morality, identity, etc.) which can lead to emotions such as happiness, anger, anxiety, relief, pride, shame, guilt, and envy. Thus, many (if not all) emotions in sport and physical activity contexts could be considered to be ‘social’ in nature, as they arise from appraisals within interactions that are embedded within social relationships. Additionally, emotions such as guilt, shame, pride and embarrassment are considered a separate class of ‘social emotions’ by some theorists (e.g. Tracy et al. 2007) since they explicitly concern appraisals involving perceptions or evaluations of and by others.

Researchers have also put forth the idea that group-based emotions are experienced as a function of one’s affiliation or identity as a member of a particular group (Goldenberg et al. 2014), such that someone may feel happy or embarrassed on behalf of their team, despite not being directly involved in the situation or event itself. A related concept is that of collective emotions (von Scheve and Ismer 2013), which refers to a form of group-based emotions that are experienced collectively in the presence of others (e.g. celebrating and feeling happy with teammates after a win, feeling sad with teammates after a loss). Group-based and collective emotions are thought to arise as a function of appraisals that explicitly involve elements of individuals’ social relationships and their identification as a member of a group or team (Kuppens and Yzerbyt 2012). In sport contexts, preliminary qualitative research has revealed that athletes report experiencing group-based and collective emotions such as pride or happiness on behalf of their teammates (Tamminen et al. 2016a). Additional work by Totterdell (2000) demonstrated associations between athletes’ mood and the collective mood of their teammates, suggesting that individuals in groups or teams may experience emotional contagion or emotional linkage (Kelly and Barsade 2001). Thus, a second way in which emotions may be considered social/relational phenomena is by considering the way social relationships are implicated in the appraisal and experience of group-based or collective emotions, and in considering linkages between athletes’ emotions.

A third way in which emotions may be viewed as social phenomena concerns the social functions of emotions within social settings and groups. There is a large body of social psychology research which suggests that emotional expressions serve adaptive value by communicating messages and values to group members, and that emotions reinforce, maintain and constrain social relationships (Fisher and Manstead, 2008). Lazarus (1991) noted that social functioning is an important adaptational outcome of appraisal and coping processes, and Feldman Barrett (2006) also notes the adaptive value of the ability to conceptualise and ‘name’ emotions in communicating experiences to others efficiently. Within sport, Friesen and colleagues (2013) have drawn on the social-functional theoretical perspective of Keltner and Haidt (1999) to describe how emotions informed two athletes’ decisions to regulate the emotions of others within their team. The social functions of emotions in sport also include serving affiliative or distancing functions between teammates, communicating team values, prompting increased effort by teammates, and promoting communal coping to deal with stressors as a team (Tamminen et al. 2016a). There is also emerging evidence to suggest that leaders’ expressions of confidence in the team positively influence the confidence and performance of their teammates (Fransen et al. 2015b).

A fourth way in which emotions may be viewed socially is with regard to emotion regulation or management. Early work by Ekman and Friesen (1969) examining nonverbal behaviour and the communication of affect emphasised the role of social and cultural norms for emotion expression as well as ‘display rules’ which impact how and when people express their emotions. Lazarus (1991, 1999) also noted that emotions are regulated within the context of cultural and situational constraints; however, with a few recent exceptions, researchers in sport psychology have largely ignored the role of social and cultural influences on emotion expression, emotion regulation and coping in favour of examining emotions and coping processes at the individual level. Within sport, athletes describe regulating their emotions for social purposes and within social constraints, for example, athletes have reported up-regulating or amplifying the expression of some emotions in an attempt to motivate and increase the effort of their teammates, and athletes also report down-regulating or suppressing some emotions to conform to social norms within the team or to avoid distracting or burdening teammates during competition (Friesen et al. 2013, Tamminen and Crocker 2013, Tamminen et al. 2016a). Additional research findings have demonstrated that athletes can and do attempt to regulate the emotions of their teammates, and that efforts of adolescent athletes to regulate the emotions of their teammates are associated with their own sport enjoyment and commitment (Tamminen et al. 2016b). Thus, interpersonal emotion regulation strategies are linked to athletes’ own affect and motivation in sport, further supporting the idea that emotions and emotion regulation should be considered social phenomena.

Emotions as intersubjective, social, performative, and embodied

Despite theoretical and empirical advances within psychological traditions of inquiry regarding the social nature of emotions, researchers operating from a psychological perspective have tended to prioritise how individuals appraise, experience, feel and express emotions, and the consequences of emotions for individual well-being and performance. However, this perspective has been criticised as being reductionist, abstracting emotions from interactions and corporeal experience to treat them as variables (Spencer et al. 2012) or by implicating a cause–effect relationship between the individual and their environment (Gergen 2011, Burkitt 2014).¹ In this section, we depart from these psychological perspectives that have dominated the research on emotions in sport and physical activity, and we draw on perspectives of emotions as intersubjective, situated, and embodied to broaden our understanding of emotions in sport and physical activity contexts.

Scholars such as Denzin (1984), Ellis (1991), Gergen (2011), and Burkitt (2014) differ in their views of emotions; however, some common features of their perspectives include the ideas that emotions are intersubjective and social; that emotions are performative, constrained within cultural relations, and constrained by power and emotion rules; and that emotions are embodied and situated. Within this body of literature, the term mood is often used to refer to lingering sensations, akin to an ‘emotional hangover’ from previous experiences or situations (Burkitt 2014, p. 9; see also Denzin 1984). The term affect has been used in various ways to describe the intensity of an emotional experience, to refer to individuals being changed or moved by a feeling or emotion, or as a mental representation of how events impact a person (for an overview see Burkitt 2014). Burkitt’s position is that affect and emotion are not radically different, as both concern affective practices of ‘embodied meaning-making’ (2014, p. 14) within social life. There is a long history of research and theorisation on the nature and experience of emotions from sociological and relational perspectives, and it is beyond the scope of this paper to provide an in-depth treatment of each of these scholars’ perspectives. However, in this section we survey some key concepts that help to inform our understanding of emotions in sport and physical activity contexts.

Viewing emotions as intersubjective and social suggests that they do not arise solely within the individual, but rather that they are both individual and social (Denzin 1984, Ellis 1991) and they cannot be experienced ‘without the implicit or imagined presence of others’ (Denzin 1984, p. 3). A view of emotions as intersubjective and social counters the psychological assertion that emotions arise solely within the individual through a process of individual meaning-making about an event in relation to one’s goals, values, and beliefs (e.g. the appraisal process; Lazarus 1999). Rather, these very goals, values, and beliefs that we come to deem as important are constituted and negotiated through interpersonal, cultural and social traditions (Denzin 1984, Ellis 1991, Burkitt 2002, Gergen 2011). Consequently, our understandings of emotions such as happiness, pride, anxiety, or anger are developed within our social relationships, and the generation of emotions – even emotions experienced ‘alone’ – is inherently social. Gergen (2009) suggested that while an emotion such as anger is generally understood as having a justifiable cause, this cause is not provided by biology, but rather through cultural negotiation (p.107). Thus, what anger means and how we come to understand anger and the things that make us angry is informed by cultural, historical, and societal meanings about emotions.

Viewing emotions as intersubjective and social also suggests that they arise in patterns of relationships between the self and others, and are only understood and considered meaningful through their production within relationships (Burkitt 2014). Emotions such as happiness, pride, anxiety, anger, and shame are only considered meaningful and co-created within relationships with others (Gergen 2011) and in wider social relations (Burkitt 2014). For example, an athlete experiences pride in relation to others, or in relation to her previous accomplishments – emotions ‘express our relationship to the world and specific people and things within it’ (Burkitt 2014, p. 2). As Burkitt (2014) explained, a situation does not govern the emotions we experience, but it is our relationship to that situation and our interactions with the people in the situation which contribute to the construction and experience of emotions. Thus, emotions only occur in relation to something or someone else, and this intersubjective perspective of emotions has largely been ignored within psychological views of emotion. Denzin (1984) also discussed emotional intersubjectivity as the shared knowledge between two people, taking various forms such as common or shared feelings, fellow-feeling (akin to vicarious affect), emotional infection (akin to emotional contagion), emotional identification, emotional embracement and bonding, and spurious emotionality (presuming one’s own emotions are shared by others). These various forms of emotion illustrate the ways in which emotions occur within relationships and through patterns of interactions, and they bear striking similarities to collective emotions and group-based emotions discussed above.

Emotions can also be viewed as performative and constrained by cultural conventions and structures of power (Hochschild 1983, Gergen 2011, Burkitt 2014). To draw on an example from Gergen and apply it within a sport context, consider a coach’s expression of anger which may suggest a cause that is culturally negotiated and relevant within the context of sport – perhaps the coach perceives their athletes are not working hard enough in games. The social and cultural context of sport justifies and legitimises the coach’s expression of anger, afforded through their status as the coach, which further constitutes social and cultural norms concerning expressions of emotion by particular members of a group. Furthermore, the expression of anger is socially or culturally constrained, in that there may be more or less acceptable venues for expressing (performing) anger in the context of sport. Social and cultural conventions also provide (and constrain) options for athletes’ responses to the coach’s anger, including apologies or expressions of shame, increased effort, or perhaps rebellion, exasperation, or sarcasm. Thus, there are thought to be social norms or emotion rules (Hochschild 1983, Keltner and Haidt 1999) regarding the emotions that athletes, coaches, trainers, and even spectators might be expected to experience and express in sport and physical activity contexts.

Emotions and emotional expressions may be governed by ‘feeling rules’ (Hochschild, 1983) or performative scripts (Gergen 2009), and by power that exists within social relations (Foucault 1977). The idea that power relations constrain and shape the interactions which form the experience and expression of emotion (Burkitt 2014) is an important aspect of emotions that is largely absent from psychological studies in sport and physical activity, although there have been some exceptions. For example, the performance of emotions in professional hockey was explored by Gallmeier (1987), who described the performative and scripted nature of emotions among hockey players, as well as the intersubjectivity of emotions. Gallmeier notes that pre-game, ‘the script does not call for overt displays of emotion. Their emotional cues are being directed by the coach, the trainer, and other staff members,’ yet once the game starts, ‘the script permits, indeed demands, an overt display of excitement. The diffuse emotions experienced by the players up to this point – feelings heightened by emotion-work – are now released’. Lee Sinden (2010, 2013) also proposed that technologies of emotions effectively normalise emotions as being private and individualised experiences, and some emotions are ‘outlawed’ or considered inappropriate in sport contexts. Within this perspective, emotions are positioned as feminine, irrational, and a sign of mental weakness, contributing to notions that athletes should conform their emotions to what is ‘normal’, suppressing or disengaging from their emotional experiences. Within the context of sport, emotions are homogenised and technologies of emotion serve to maintain a particular standard of acceptable emotionality. Athletes may suppress their emotions, disregard the emotions of others, or may view emotions as unacceptable and a sign of mental weakness (Lee Sinden 2013).

A final point that we wish to emphasise is that emotions are embodied, and cannot be abstracted from the body or physical sensations that are felt within the body. Burkitt (2002) noted that ‘just as feeling and emotion cannot be reduced to the body, nor can they be explained only in terms of discourse’ (p. 152), and Denzin (1983) observed that ‘the body does not call out its own interpretations of internal somatic states’ (p. 403). Because emotions occur as physiological arousal and ‘feelings’ of bodily sensations, Ellis (1991) argued that accounts of the lived experience of emotions must take into account the physiological sensations of emotional experience. Therefore, emotions are felt as sensations in the body, yet these are only understood in relation to the contexts within which they occur, and through the use of language and words to express, describe, and name emotions (Burkitt 2014). Here, Denzin (1984), Gergen (2009), Burkitt (2014) and others have noted the importance of language and the vocabulary that is brought to bear on emotional experiences – our emotional vocabularies and what we come to understand as pride, happiness, anger, and fear are socially constructed and maintained through our interactions with others. Before reviewing some of the literature that has adopted narrative approaches to explore emotions in sport and physical activity contexts, we briefly discuss the importance of language in emotion research.

Emotions and narratives: the importance of language

The importance of language in the experience and study of emotions has been noted within many perspectives of emotion: Burkitt (2002) described the importance of ‘emotional vocabularies’ in giving cultural meaning to emotion states, while Gergen (2011, p. 119) also noted that drawing conclusions about emotions from brain states (e.g. physiological or neurological data) requires that the vocabulary to draw such inferences is already in place: ‘try as one may, one will not observe a brain condition of joy, hope, dread, or wonder’. Further, the words used to describe emotions are themselves created within relationships and cultural traditions (Ellis 1991, Gergen 2011, Burkitt 2014) and these words and emotion concepts may encourage or constrain certain expressions and experiences by invoking various cultural norms (Boiger and Mesquita 2012). From a psychological perspective, Feldman Barrett (2006, p. 37) also argued that language determines the conceptual knowledge that individuals have about emotions, stating that ‘language shapes core affective phenomena [neurophysiological states] into the emotional reality that we experience’. While Lazarus (1991, p. 192) maintained that emotions were primarily a psychobiological rather than a linguistic phenomenon, he also acknowledged the importance of language within cultures and suggested that ‘an examination of these vocabularies, aside from what they might tell us about human nature, could also be instructive about our cultural values about each emotion’. To understand emotions in sport and physical activity, then, it is necessary to understand how people talk about emotions and the words they use to discuss emotional experiences.

Thus, language is thought to shape and constrain the construction of emotions (Gergen 2011, Burkitt 2014) and the ways in which individuals learn about what emotions are (Feldman Barrett 2006). Language can also provide insight into the cultural values regarding emotions and their experience and expression in social groups (Lazarus 1991). Given the importance of language for the construction and socialisation of emotion, as well as its importance for the expression and experience of emotion, it is our position that we may better understand how people talk about emotions and what emotions ‘do’ within relationships (Ahmed 2004, Walby and Spencer 2012) by drawing on narrative approaches which attend to language in the study of emotional phenomena.

Narrative emotion research in sport and physical activity

Narrative approaches may be particularly valuable in understanding emotions as intersubjective and social, performative, and embodied. As Phoenix and Orr (2014, p. 100) noted: ‘rather than viewing feelings and emotions as isolated phenomena within the individual, this [relational] stance situates emotional life as a narrative construction, the complexities of which being best understood through the embodied narrative constructions and reconstructions of the principal social actors’. An underlying assumption of narrative work is that individuals construct stories out of the sociocultural landscape to ascribe meaning to their lives; they are storytelling animals (Polkinghorne 1988). When telling their stories, ‘people give meaning to their experiences within the flow and continuously changing contexts of life […] All this is done not only in narratives about the past and the present, but also about future times and places’ (Medved and Brockmeier 2004, p. 747). As such, narratives provide individuals with a structure to ascribe meaning to their selfhood and identities, and stories may also be created, told and re-told by individuals and groups to create identities and boundaries (Frank 2012). Individuals craft the stories they tell based on information that is available to them to explain what has and is happening in their lives, and perform these stories in certain ways and for certain purposes (Gubrium and Holstein 2009). In this section, we review research that has adopted narrative approaches to explore emotions in sport and physical activity contexts.

Researchers adopting narrative approaches to explore emotions in sport and physical activity have done so in different ways: in some cases, they have collected and analysed the narratives of athletes, adopting the role of ‘story analysts’ (Bochner and Riggs 2014, Smith 2016) to produce realist accounts of the athletes’ experiences. In other cases, researchers have adopted the role of ‘storyteller’, using athletes’ narratives or their own autoethnographic narratives to produce evocative stories about emotional lived experiences (Bochner and Riggs 2014, Smith 2016). This variation in the uses of narrative methods is important because it showcases the multitude of ways in which narratives can be adopted to study emotions in sport and physical activity, thereby making it a rich tradition from which to draw on to study emotions as situated, embodied, and intersubjective phenomena.

Injuries are often meaningful events that hold important consequences for sport and physical activity participation and they evoke various emotional experiences; thus, several researchers have explored the role of emotions in the experience of injury among athletes. For example, drawing on two years of ethnographic work, Spencer (2012) described narratives of loss and despair in the experiences of pain and injury among MMA (mixed martial arts) fighters. The athletes’ narratives of despair reflected absence and failure associated with injuries, while narratives of loss reflected missed opportunities and the loss of masculinity and social status following injuries. Spencer’s work also illustrates how emotions in sport are ineluctably tied to the body, and illuminates how cultural masculinity narratives of strength and stoicism shape the ways in which athletes construct their identities and injury experiences. Thus, the examination of athletes’ experiences of pain and injury draw attention to the notion that emotions arise due to our relationship with our body (and usually when our relationship with our body is changed, fractured, or damaged in some way). Similarly, Sutherland et al. (2014) examined narratives of young women athletes’ experiences of pain and self-compassion which frequently concerned experiences of injury in sport. These narratives highlighted how emotions arise in relation to the body and conceptions of the self: ‘Injury showed signs of weakness to Lexi … She was upset, frustrated, and disappointed with herself. The emotions seemed overwhelming. She couldn’t change what had happened and was disappointed that she let her body get the best of her’ (p. 506). Additional work examining emotions surrounding injuries in sport was presented by Laurendeau (2014), who used autoethnographic narratives to explore the construction of the self through an articulation of injury experiences, masculinity, and somatic emotional experiences in sport. In this work, Laurendeau conveys emotions arising from one’s (fractured) relationship with the injured body, and he explores how these emotions are produced through social interactions and are constructed within discourses of masculinity: ‘Out of nowhere, the ‘new guy’ hammers me. I lie on the ground, unable to move, struggling to catch my breath. My cheeks feel flushed, the surest physical indicator of the shame that consumes me’ (p. 14). This research echoes Denzin (1984), Gergen (2009) and Burkitt (2014) who noted that emotions are bodily in sensation and in their conception – they are felt within the body, and they also arise out of our relationship (and changes in our relationship) with our body.

Allen Collinson (2005) also lays bare the emotions experienced in relation to the body in her collaborative autoethnographic work documenting the rehabilitation process following injury. Through her analysis of narratives and in presenting the story of her (and her collaborator’s) injury rehabilitation process, Allen Collinson’s work illustrates how emotions are experienced in relation to the body and she also highlights the intersubjectivity of emotions between people. For example, she describes how emotions arose between her and her training partner as they each made progress in their training, and as responses to the other’s rehabilitation progress: ‘… the other partner, who was making more positive progress, experienced a kaleidoscope of shifting emotions, including sympathy and empathy, anxiety, anger that the other should have to endure more difficulties, and sometimes guilty relief that s/he was not suffering in the same way’. She also highlights how emotional ‘face work’ occurred in relation to others during a physiotherapy appointment:

In this instance, emotions were experienced immediately and strongly, and yet the norms of the social situation required that these violent emotions be managed, contained and rendered acceptable in the ‘impersonal’ public world (Lupton 1998). Although not recorded in the fieldnotes, I recall vividly that despite an almost murderous intent, I did muster sufficient self-control to proceed with the remainder of the physiotherapy encounter, and to exit the field without, I think, betraying my utter fury to the person who had provoked such a response.

Thus, in documenting the vacillating emotions that accompanied the bodily experiences of injury over a two-year period, Allen Collinson’s work illustrates the intersubjectivity of emotional experiences in a sport/physical activity context. The intersubjectivity of emotions is also evident in Sparkes’ (1996) writing on the relationship between the body and self over time. His autoethnographic narrative work revealed how emotions occurred in response to others’ assessments and comments regarding the injured body: ‘…he told me that my first-class sporting career was over. When he said this, I felt nothing, just a numbness. Then I felt a wave of hatred for this man and his arrogance at dismissing my dreams in such a matter of fact way’ (Sparkes 1996, p. 480). Collectively, these autoethnographic pieces highlight how the telling of personal narratives of pain and injury provide an avenue to illuminate the various roles and functions that emotions can play in shaping sport and physical activity experiences.

While most narrative work on emotions in sport and physical activity have not focused exclusively on the experience of a particular emotion, Phoenix and Orr (2014) used a narrative approach to undertake an in-depth exploration of the experience of pleasure within physical activity settings. The authors present four types of pleasure: sensual pleasure, documented pleasures, the pleasure of habitual action, and the pleasure of immersion. This work is important in elucidating variations in the ways that individuals may experience pleasure in physical activity, and also foregrounds the importance of situated, sensory emotion experiences, for example, by describing pleasure associated with the smell of a freshly mown golf course or a dance partner’s aftershave, or the sensation of gliding through water. The authors also draw connections between the body and the environment in producing pleasurable physical activity experiences. Another narrative study that draws attention to the notion of situated emotional experiences was conducted by Bairner (2014), who used personal autoethnographic narratives to explore how emotions are ‘prompted by and in sporting spaces’ (p. 18). Drawing on his experiences in football (soccer) grounds, Bairner notes how football grounds are sites of communal gathering and ritual that contribute to a sense of community among its members, and how shared sporting moments among spectators contribute to collective emotions and communal storytelling. This work serves to highlight the situated nature of emotions, and calls for consideration of the body in relation to other bodies (Pelias 2011).

Collectively, researchers using qualitative narrative approaches have articulated the intersubjective, performative, embodied, and situated aspects of emotions in sport and physical activity. These studies draw attention to the way emotions occur in relation to the body and often as a result of a changed relationship with the body: we note that several studies examining emotions in sport and physical activity have explored narratives of injury, because injuries change the relationship with the body and the construction of the self. We also see that emotions occur in the body as corporeal, felt sensations, and that emotions are situated and occur as lived experiences within particular times and places. Finally, we see that emotions are intersubjective, in that they occur with others, in response to the experiences and actions of others, and in relation to others’ assessments, commentaries, and observations of our bodies.

Moving forward with narrative emotion research in sport and physical activity

Smith and Sparkes (2009) suggested that narrative research is useful for advancing research in sport and exercise psychology in several ways: its interpretive foundations allow for an alternative to prevailing post-positivist approaches to qualitative research; it helps us better understand the personal and social experiences of (in)active individuals, the meanings they ascribe to these experiences, the place of (in)activity in their lives and the diversity of their experiences; it places emphasis on relatedness and focuses on examining the ways physical activity practices are shaped, enabled and constrained in social interactions; it illuminates lived experiences with a focus on emotions, feelings, motivations and sense of self and how they change over time; it highlights the ways in which sociocultural factors shape physical (in)activity attitudes and behaviours; it provides an avenue to understand the development of the self and identity; and it focuses on the corporality and embodied character of social and psychological practices.

With regard to the study of emotions, Denzin, Burkitt, Ellis, Pelias and others have articulated the benefit of adopting interpretive approaches, noting that introspective approaches presented as narrative text can ‘prompt and collect our own and other people’s stories about the lived details of socially constructed experience’ (Ellis 1991, p. 45). Even Lazarus (1999, p. 214), operating from an individualistic psychological perspective, argued for the use of narratives to explore emotions, suggesting that narratives would ‘offer a more useful approach to understanding than traditional psychological research methods, and they certainly come closer to the natural ways in which we construct meaning from our life experiences’. Interpretive, narrative approaches can contribute to understanding emotions in sport and physical activity in several ways, for example, by: (a) examining how emotions like joy, fear or anxiety arise within the context of our social relationships (e.g. how relationships shape, enable, and constrain emotions); (b) exploring what emotional stories or narratives do for athletes, coaches and others within sport and physical activity contexts (e.g. how emotions and emotional stories may be used within sport and physical activity); (c) examining how emotions are created, recreated, and sustained through the stories people tell; (d) examining how collective and group-based emotions are intertwined with one’s identity and identity development; and (e) highlighting the ways in which social and cultural narratives within sport shape athletes’ emotional experiences.

We noted several key themes from our review of narrative research on emotions in sport and physical activity, which may serve to frame future research endeavours: First, emotions are bodily experiences: they occur as physical sensations in the body, and arise from our (changing) relationship with the body (e.g. in cases of injuries where our relationship with the body is fractured, changed in some way, re-negotiated). In this way, emotions are embodied; they are experienced in and through the body as individuals move through the world (Merleau-Ponty 1962, Strauss 1963, Crossley 1995). Future research on emotions in sport and physical activity may seek to explore emotions arising from changes in relationships with the body that occur in other ways – for example, examining emotions associated with changes in the body following retirement from sport (Gairdner 2015), emotions associated with physical changes accompanying ageing in master’s athletes and physically active older adults (Phoenix and Sparkes 2009, Phoenix and Smith 2011), or during athletes’ experiences of and recovery from eating disorders (Papathomas and Lavallee 2012).

We feel that further research may explore connections between emotions and the construction of the self and identity in sport and physical activity, as emotions and one’s relationship with the body contribute to constructions of the self (Denzin 1984, Burkitt 2014). It is valuable to consider narrative work by Douglas and Carless (2015; see also Carless and Douglas 2008, Douglas 2009) as well as by Smith et al. 2016 on identity in sport, which highlights emotions within the stories that athletes tell as they construct and maintain their identities and understanding of the self. For example, in their examination of the narratives of elite athletes with impairment, Smith et al. 2016 highlight the role of embodied emotional regulation in the participants’ constructions of activist identities. As such, future research may seek to explore the various ways in which athletes experience and embody specific emotions arising from and contributing to these changing constructions of the self, such as anger and anxiety. Research may also benefit from exploring not only negative emotions, but also the embodiment of positively toned emotions and their influence on constructions of the self. Finally, the investigation into self-conscious emotion narratives of shame, guilt and pride may be useful in further illuminating the relational role of body-related emotions in relation to identity construction in sport and physical activity settings.

Since narratives are told about, in and through the body (Smith and Sparkes 2009), and since language and the body are integral to emotion experiences, narrative methods might be useful to examine intersections between bodies, language, storytelling, and sociocultural narratives. The challenge of using narrative methods to capture the corporeal experience of emotions has been raised previously, as there is a ‘gap’ between feeling and the language that may be used to capture or describe the emotions one experiences (Gergen 2009, Burkitt 2014, Phoenix and Orr 2014). Burkitt (2014) argues that feeling and emotion are not automatically or easily articulated, and people may struggle to express exactly what they are feeling, but ultimately we would not have feeling or sense without words or meanings themselves (p. 69). Thus, words and language are used to shape or structure the emotions we experience, and these emotional experiences are informed by the language we have available to us. Emotional experiences, in this sense, are viewed as enactments of pre-given stories and narratives about what is enjoyable or unpleasant (Phoenix and Orr 2014).

The meanings that are made regarding emotional experiences are shaped and constrained by the availability of narrative resources (see Smith and Sparkes 2009, Frank 2012), and narrative methods may be used to discover the range of narrative resources available to athletes in understanding emotions. Therefore, it is also worth examining what emotion narratives do for athletes and individuals participating in sport and physical activity. In particular, the products of narrative research would be valuable for participants and practitioners within sport and physical activity. Recently researchers have demonstrated the potential usefulness of narrative vignettes among mentors for sport participation for individuals with a spinal cord injury (Perrier et al. 2015) and narrative therapy has been used as a therapeutic approach in sport psychology practice to improve body image among young female athletes (Leahy and Harrigan 2006). Narrative therapy invites individuals to develop alternative scripts or ‘re-author’ experiences in their lives and relationships, and it encourages examination of discarded, unspoken, or unnoticed narratives to change relational meanings of events and stories (Freeman and Coombs 1996, Leahy and Harrigan 2006). Thus, it may be possible to adopt narrative approaches and use stories to broaden athletes’ narrative resources to change the meanings and understandings of emotions within relationships and groups. For example, athletes or physically active individuals may develop alternative meanings associated with particular events or stories within their lives, thereby changing athletes’ emotional experiences associated with particular events.

The broadening of narrative resources and the use of stories may also have value within relationships and in social groups. Stories create identities and boundaries within groups (Frank 2012), thus narratives and stories may be purposefully crafted to foster athletes’ identities in groups and teams by producing positive emotions among group members. While Tamminen et al. (2016a) have explored some of the social functions of emotions in sport contexts, this research did not explicitly adopt a narrative approach and there is little research examining what emotional narratives ‘do’ in social and emotional life within a sport context. Within sport, stories are told and re-told, and in doing so they become emotional experiences, or they are told purposefully to construct or alter emotions. There is also evidence that leaders and coaches impact the emotional experience and performance of athletes: for example, Fransen et al. (2015a, 2015b) provided evidence that leaders’ expressions of confidence led to increases in athletes’ social identification with the team and greater confidence in the team, as well as better performance on an experimental soccer task. Furthermore, athletes’ and coaches’ emotional displays are perceived to impact others around them (Vargas-Tonsing and Guan 2007, Tamminen and Crocker 2013, Tamminen et al. 2016a), and athletes’ emotional displays have been associated with their team’s subsequent success (Moll et al. 2010). There is much work to be done in exploring exactly how coaches and leaders structure their stories, and to examine how these stories contribute to the emotions, identities, and performance of athletes. Moving forward, it would be valuable to explore how coaches, athletes and others construct stories to engender emotions by using elements of suspense, drama and tension, by recalling past victories or failures, detailing past rivalries with opponents, and by foreshadowing possible futures in their storytelling.²

Challenges, lessons and suggestions

We have argued the potential merit of narrative methods for the study of emotions in sport and physical activity. However, we are both relative newcomers to narrative inquiry and in this final section we offer some suggestions to researchers based on our attempts at navigating our narrative research experiences. We ground our reflections in examples from our ongoing work on the social aspects of emotions in sport and physical activity. The first author draws on data from interviews and diaries with high-performance athletes, as well as interviews with parents and their children participating in sport, and the second author draws on life history data from interviews and field notes with older physically active women. The sections that follow outline the strategies we found useful in examining emotional experiences in narrative forms, and we offer suggestions to researchers interested in pursuing similar work.

We began by reading Reissman (2008), Holstein and Gubrum (2012) and Andrews, Squire, and Tamboukou’s (2013) pieces on narrative methods to get a broad sense of (some) narrative analysis techniques for working with storied data. We then explored some additional theoretical and empirical pieces in the narrative tradition (for some examples see Ricouer 1990, Bruner 2002, Smith and Sparkes 2005, Loseke 2007, Bochner 2014), some in sport and exercise psychology but most from outside our discipline. At the same time, we met frequently to discuss our processes of analysis and challenges we were facing; while we were both working with different data-sets and while our discussions occurred over Skype as a result of our geographical locations, these steps proved instrumental in shaping our understanding of narrative methods.

Based on our readings and discussions, and on the premise that we were interested in exploring what participants said and how they were saying it, we initially employed a combination of thematic and structural narrative analysis techniques to explore our data (Reissman 2008). Through a thematic analysis, we examined the content of the participants’ stories, the core patterns within them, as well as the context in which they took place (Reissman 2008). Drawing on Sparkes and Smith (2014), we found it useful to write down initial thoughts and comments regarding the plot of one participant’s story with a focus on significant events, the peripetia (Bruner 2002) and tuning points throughout the story. We identified characters within the story and asked ourselves about their traits, beliefs, actions, as well as their connections to the context and purpose of the story. We identified key themes across the participants’ stories as well as explored the meanings the participants ascribed to their experiences, while also identifying connections between themes as well as the contexts in which they occurred. These steps were taken for every participant, followed by a comparison and contrasting of themes across participants. Drawing on these experiences, we would suggest that researchers consistently write conceptual notes throughout the analysis process so as to keep track of thoughts regarding the participants’ stories, and the connections occurring across and within themes. Novice researchers may also find it useful to remind themselves of the iterative nature of narrative work; it is not comprised of sequential ‘steps’, but rather of a back and forth process that moves to and from the extant literature, immersion in the data, as well as the identification of plots and storylines, themes, and contextual factors that shape the participants’ stories.

To add further analytical depth, we moved beyond a thematic analysis to consider the structure of the participants’ stories, with a focus on how the stories were pieced together. Informed by Sparkes and Smith (2014), we found it helpful to immerse ourselves in their participants’ stories and the findings of our thematic analyses, while making note of how different events within each story were connected and how the participants told their stories in relation to their past, present, and future. Drawing from the extant research, prominent cultural representations of sport and physical activity, and from reflexive thoughts regarding the embodiment of cultural narratives and the emotions they elicit, we also made note of the broader societal narratives shaping how participants told their stories. For example, in analysing life history interviews with physically active older women, we identified that some participants embodied cultural narratives surrounding idealised femininity; they expressed discontent with their changing bodies and engaged in physical activity to retain their thin and youthful appearances. Self-conscious body-related emotions of embarrassment and guilt permeated their narratives. Others drew on counter-cultural narratives of body appreciation; they took pride in their ageing identities, and embodied attitudes that privileged overall well-being. Researchers may find it useful to name and cluster the various storylines from each of their participants’ accounts into broader narratives types. The narrative types can then be written up, with a description of their underpinnings and characteristics, coupled with detailed participant accounts to highlight how the stories are framed by these typologies.

Throughout the thematic and structural narrative analysis process, we also found ourselves engaging in other genres of narrative analysis, such as dialogic/performance analysis (Reissman 2008) to explore the social–relational aspects of our data by paying attention to how, in what contexts and to what end participants told their stories. For example, in analysing interviews with parents of athletes involved in competitive sport, we identified instances in the interviews where parents appeared to perform the role of ‘the good sport parent’ by contrasting their own actions against ‘bad sport parents’ (e.g. commenting that their actions are not as bad as other sport parents) and by emphasising their supportive involvement in their child’s sport activities. We reflected on the idea that within the context of an interview with researchers who may be perceived as judges or experts, parents (and athletes) may perform particular roles and identities that shape their narrative performances.

While we would suggest that researchers attempt to embrace the complexity of the analytic process, they may find themselves (as we did on many occasions) wondering whether they are ‘missing something’ in the stories. We found it helpful to employ various analytic tools to explore the social–relational aspects of participants’ stories and to advance our interpretations of the data, including:

•	Looking for metaphors in the data (e.g. Smith and Sparkes 2004)
•	Writing fictional stories based on the data (e.g. Sparkes 1997)
•	Attempting poetic representations of the data (e.g. Richardson 1992)
•	Attending to the plot and turning points in the stories told by participants (e.g. Reissman 2008)
•	Considering the typologies and genres of stories told by participants (e.g. Eco 1994)
•	Exploring the characters present and the roles of narrators within participants’ stories (e.g. Koven 2012)
•	Attending to the role of time and place in the participants’ accounts (e.g. Ricouer 1990)
•	Engaging in reflexive journaling and discussions with colleagues

We found that these analytic tools helped us consider what cultural narrative resources participants drew on, took for granted, and/or overlooked, to what end and for which audiences participants constructed their stories, how events were sequenced, if any counter-narratives occurred within participants’ stories, and which emotions were elicited when reading the stories.

We would again like to stress the importance of documenting the analysis process effectively, succinctly and credibly without undermining its complexities. For example, an audit trail documenting the decisions made throughout the research process may help researchers better understand how they have conducted their analyses, and to reflect on their role in the interpretive process. In addition, keeping track of discussions with colleagues and collaborators, turning points in understanding of theoretical and methodological concepts, as well as of difficulties occurring throughout the analysis process may help researchers further tease out the nuances of the data within the final written piece. Novice researchers may also find it helpful to provide collaborators with frequent updates on the analytic process. However, this may be challenging since the analysis may not move in a sequential progression across thematic, structural and performative stages (Reissman 2008, Holstein and Gubrum 2012, Sparkes and Smith 2014).

A key element in gaining confidence in our analyses was the instrumental guidance and support from established narrative scholars who encouraged us to take ‘risks’ and to think deeply and creatively about alternative ways in which to explore and present the data.³ Ongoing mentorship and opportunities to develop methodological and analytic skills through meetings at conferences, workshops, graduate courses⁴ and via email were instrumental for advancing our understanding of the complexities of narrative approaches. The narrative analysis process has also led us to wonder how to go about training future narrative researchers, and worry that many may shy away from engaging with this important work given its complexity, a notion already discussed by many scholars (e.g. Lazarus 1999, Sparkes and Smith 2014). It may be helpful for researchers to seek guidance from colleagues employing narrative methodologies in different fields of study to their own, since conflicting and challenging discussions surrounding epistemology and ontology may occur given that conceptualisations of the self and narrative and the nature of reality vary within the tradition itself (Smith and Sparkes 2006). These discussions, albeit potentially difficult, may push researchers to better understand the assumptions underlying their work (e.g. their view on reality, how they define the self and narrative, Smith and Sparkes 2008) as well as how they locate themselves within the narrative tradition as researchers who study emotions from psychological perspectives. As such, we suggest that researchers grapple with and attempt to clearly articulate the assumptions underpinning their research when navigating narrative work.

Conclusions

Narrative approaches to studying emotions are not currently privileged in sport and exercise psychology and as such, one of the difficulties we have encountered as a new faculty member and PhD student embarking on this work has been resistance to these approaches due to a perception that projects of this nature can take ‘too long’, which may negatively affect junior researchers’ training opportunities and career trajectories. We are reminded that stress and emotion theorist Lazarus (1999) foreshadowed this notion as he discussed the reluctance to adopt narrative, idiographic approaches in psychology:

We are such a contentious discipline that it is difficult for any radical departure to get a respectful hearing, much less to lead to programmatic change. Some of our problems lie in the reward structure of academe, wherein rapid publication, rather than programmatic research and replication, is coin of the realm.

In this article, we have outlined ways in which emotions can be considered as interpersonal phenomena from a psychological perspective, although we noted that this view positions the individual at the heart of the study of emotions. We have also drawn on some perspectives which consider emotions as constructions that are inherently intersubjective and social, performative, and embodied. From the intersection of these perspectives, we advocated for the use of qualitative narrative methods to advance our understanding of social/relational aspects of emotion in sport and physical activity domains. We concluded by attempting to articulate some of our own challenges in proceeding with narrative analyses and raised some issues with conducting narrative sport and physical activity research. Whether one adopts an individualistic position that emotions are bounded, arising from appraisals of the (social) environment, or if one adopts a position that emotions emerge and are constituted through patterns of interaction embedded in social relationships, narrative approaches can usefully contribute to the study of the social aspects of emotions in sport and physical activity.

Notes on contributors

Katherine A. Tamminen, PhD, is an assistant professor in the Faculty of Kinesiology and Physical Education at the University of Toronto, Toronto, Ontario, Canada.

Erica V. Bennett, MA, is a PhD candidate in the School of Kinesiology at the University of British Columbia, Vancouver, British Columbia, Canada.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This manuscript was supported by a Social Sciences and Humanities Research Council of Canada (SSHRC) Insight Development Grant awarded to the first author, and a SSHRC Joseph Armand Bombardier Doctoral Award awarded to the second author.

Notes

1. The first author admits to being complicit in this tradition and has only recently begun to turn towards examining the social and cultural influences on emotional phenomena in sport (e.g. Tamminen and Crocker 2013, Tamminen et al. 2016a).

2. Thank you to the anonymous reviewer who encouraged us to elaborate on this point.

3. The authors are indebted to Brett Smith and Cassie Phoenix for their mentorship regarding the use of narrative methods in sport and physical activity.

4. The second author is indebted to Sandra Mathison for offering a seminal graduate course on narrative inquiry.