Emotion talk in the context of young people self-harming: facing the feelings in family therapy.

This article describes the use of emotion talk in the context of using a manualised approach to family therapy where the presenting problem is self-harm. Whilst we understand that there is an internal aspect to emotion, we also consider emotions to be socially purposeful, culturally constructed and interactional. We found that within the presenting families, negative emotions were often talked about as located within the young person. Through using 'emotion talk' (Fredman, 2004) in deconstructing and tracking emotions and exploring how emotions connected to family-of-origin and cultural contexts, we developed an interactional understanding of these emotions. This led to better emotional regulation within the family and offered alternative ways of relating. The article discusses the use of relational reflexivity, and using the therapist and team's emotions to enable the therapeutic process, encouraging reflexivity on the self of the therapist in relation to work with emotions. PRACTITIONER POINTS: Emotions can be seen as both a reflection of feelings experienced by the individual and as a communication.An interactional understanding of emotions can be used therapeutically.Therapists should explore emotional displays and track the interactional patterns within the therapeutic system.Therapists should self-reflexive about ways of doing emotions and use this awareness in practice.

Introduction

The systemic approach to self‐harm informed by the SHIFT manual was concerned with the ways in which problems are embedded in relationships alongside the felt experiences, meanings and narratives that have become attached to and shape these problems. Self‐harm is described in the manual as being a

response to and having meaning in relation to particular issues, including emotional regulation, a sense of hopelessness … and family communication patterns (Boston et al., 2010 p. 11).

The adolescent's emotional regulation through self‐harm was assumed to be relationally connected to that of the family members. Thus, the manual supported exploration of emotional responses of family members in the sessions, as a reflection of the feelings experienced by individuals and as means of communication between them. Emotions become embedded in family narratives, as well as the narratives co‐created in the therapeutic system. It was believed that changes in these patterns of management and understanding of emotions could be significant.

The manual further advocated considering the self of the therapist, with a particular focus on reflexive abilities:

The principle of reflexive ability is an important component enabling the therapist to make decisions about the choice of specific interventions and includes emotional, conceptual and relational reflexivity (Burnham, 1993, 2005). (Boston et al. 2010 p. 12)

While attention to attachment theory and patterns is mentioned in the manual, this is not a central theme throughout. We interpreted the manual's aims of improving communication, connectedness and awareness within close relationships as aligned with thinking about attachments and we have found ourselves retrospectively augmenting some of our theorising about our work with inclusion of attachment theory. Understanding what causes young people to self‐harm is complex and multiple factors are usually involved. Whereas the SHIFT manual outlined a number of factors, of particular interest to us in this article are: difficulties in parent‐child relationships, including those related to early attachment, perceived low levels of parental caring and communication with the young person, as well as parental conflict.

This paper will provide a brief theoretical review of what can be understood by emotions, and how working with emotions fits for us into systemic practice. In the first part of this article, which shares our experiences of ‘facing the feelings’ in family therapy, we will look at how we addressed the communication of emotion within the presenting families. Four practices include: (a) exploring the family's understanding of self‐harm; (b) attending to beliefs and behaviours; (c) tracking interpersonal patterns and trigger events; (d) deconstructing constraining discourses or contexts. The second part will look at how the team and therapist used their own emotions in working with the families. This includes: (a) increasing awareness of, and using our own, emotion talk to inform the therapy; (b) co‐constructing a shared language of emotion; (c) facing heightened emotions; (d) using therapists’ own emotional self‐reflections in conversation with the family.

The families

We illustrate this article with some composite case material taken from four families, mainly referred to in the dialogue boxes (Figures 1 to 5). Each of the families comprised an adolescent girl, aged between 14 and 16, who had presented with at least two episodes of self‐harm, and was living at home with siblings and parents. These families were varied in terms of class, and from a wide variety of cultural backgrounds, including second and third generation immigrant communities. There were different family forms represented, including step‐parents and adoptive parents.

Figure 1

Imagined family's understanding of self‐harming behaviour.

Figure 5

Example of use of therapist's own emotional self‐reflexions.

The Authors

The SHIFT team was based within a Tier 3 generic Child and Adolescent Mental Health Service (CAMHS) This is a highly diverse multi‐cultural inner city context, which includes both poverty and wealth. The authors are two of a team of three female systemic psychotherapists, one of the team members was not available to contribute to the paper. The therapists were familiar with one another but had not worked as a team before. In accordance with the SHIFT manual, the therapist was in the room with the other two therapists as observers and reflecting team members (Anderson, 1987). We were all of a similar age, two had children, two were in intimate committed relationships and our cultural backgrounds were varied, coming from Northern England, Germany and Colombia.

What is an emotion?

A traditional view of an emotion is an internal, subjective experience influenced by psychobiological processes and influenced by temperament, mood and cognition. Neuroscience differentiates between emotion and feeling; emotions are considered non‐conscious and embodied, feelings are a conscious awareness of emotions (Fishbane, 2007). We use the terms interchangeably in this article.

Harre and Gillet (1994) challenge the orthodoxy of psychology that emotions are a state of the individual by describing emotions as being psychologically equivalent to statements. They suggest that an ‘emotion is socially purposeful and cultural’. From an interactional, rather than an interpersonal perspective, emotions are created between people and hence are a social form of action (Fredman, 2004). Tsai (2013) finds that there is some level of universality in the recognition of core facial expressions, but there is also cultural variability in reading the cues. Emotion is an invitation to others to respond rather than solely an internal experience. But this invitation is not always so clear. Emotions as communication also need to be ‘read’ accurately; their display may be misinterpreted or only partially available for a response (Pace and Sandberg, 2012).

Any emotion felt is a consequence of, and response to, an emotion displayed by another. This is the case even if the rehearsal of the social interaction is internal, or the cognitions that come in response are to do with the self (e.g. depressed cognitions) (Bertrando, 2010). The social relationships that one is embedded in give rise to emotion and these are displayed in interaction, either overtly or subtly. Analogical cues, such as tone of voice, facial expression, body positioning, and eye contact are particularly important in communication of emotion (Johnson, 1998). Fredman (2004) expands on this: the communication of emotions always involves the body. Language cannot be understood as separate from the body because of the analogical cues, many of which are not conscious.

Relationships and interactions do not take place in a vacuum. The social interactions we take part in are in the context of the history of the relationship, and our other relationships, as described in communications theory. They take place within a cultural and historic context that shapes our understanding and interpretation of emotion and our resultant response (Matt and Stearns, 2014). Emotions are shaped by cultures and our embedded experiences, as we learn how to do and describe our ‘internal’ feelings. The naming and classification of emotions differs between cultures, and which emotions are privileged either positively or negatively by particular cultures at particular times in history (Fredman, 2004). Within the client group of adolescents presenting in an inner city CAMHS clinic, anger could be privileged positively by the adolescent with the peer group, and negatively by the family. Gendered meanings are given to anger. For example, in one family we saw the brother's anger was viewed positively as protective of the sister, even though his anger entailed smashing up the flat when she self‐harmed, where the sister's anger was viewed as highly problematic. We attach a value to what we feel according to our cultural heritage, and a value to what we perceive in another's emotional display – these value judgments have an effect on emotional interchanges (Bertrando, 2010).

Attachment theory is a relational theory that can be linked to how people adapt to the challenges of life. Our early attachments are where we first learn how to understand and process our emotions. Children who have had their inner world responded to effectively by parents, internalise this ability and develop greater capacity to understand their own needs and feelings, as well as those of others (Fonagy et al., 2005). In the case of a self‐harming adolescent, there can be an association with attachment needs not being currently (or perhaps previously) met. Johnson (1998) describes emotional expression as defining the nature of attachment relationships; in these close relationships emotions are a ‘primary signaling system that organizes interactions … emotional cues pull for responses from others’. If the response to the emotional cue is inaccurate, this can lead to further breakdown in relational processes and communication patterns. These are attachment ruptures.

Emotional processing is the primary goal in a manualised family therapy treatment for adolescents with depression (Diamond et al., 2014). Consideration of the attachment relationships is fundamental; and the manual suggests that adolescents may have attachment ruptures which are infused with difficult emotions but that their attachment style, which had developed as self‐protective, actually prevents successful emotional processing, for example, an adolescent not wanting to talk (‘I'm fine’) but cutting instead. Addressing this is core to the therapeutic process.

As protection of the self is central, the safer one feels, the less one is concerned to hide from others and oneself; there is a ‘perception of emotional acceptability and non‐acceptability’ that is influenced by sense of safety (Pocock, 2005). Within the presenting families our hypothesis was that there was a perception of danger in talking about emotions. This led to lack of emotional communication. Negative emotions were perceived as non‐acceptable and hence hidden, or displayed and not talked about.

Pocock (2011) uses Crittenden's (2008) model of attachment to describe the mechanism of self‐harm in preserving interpersonal relationships, through either the inhibition of negative affect (regulating the self so as not to dysregulate others – a Type A strategy), or the escalation of affect in order to regulate others (Type C). One of the young people we saw kept her self‐harm hidden other than the parent finding tissues with blood on them sometimes. In this family there was a fear that if the father got upset by family arguments, he would walk out of the door and never contact them again, as he had threatened to do in the past. The young person inhibited negative emotion in case it dysregulated her father (Type A). In another family, emotional display was vivid, rapidly moving and outwardly exhibited. On coming into therapy the father said that both his brothers had died due to drug addiction. This was not spoken about at home, nor was the father's low mood. Highly fluctuating mood in the family was experienced but not spoken about. The young person's highly fluctuating mood had the effect of regulating the mood of the parents (Type C) by keeping father active and preventing him from becoming too depressed.

A different young person had suffered significant bereavements in her close family, where the family had a powerful script of not ‘crying over spilt milk’. The young person's explanation of the self‐harm in this case was to manage the self, and in doing so, protecting the relationships that she had with the family she had left (Type A). She described cutting as helpful for the mourning, as she felt it powerfully connected her to those she had lost, without risking arousing emotions in others by talking about it.

It was key to our work to create a sense of safety for the families, so that emotion talk was possible at a time when people felt very unsafe. This created a therapeutic ‘safe base’ where exploration was possible (Byng‐Hall, 1995). We did this by making sure we were available to the families outside sessions, and being relationally reflexive (Burnham, 2005) in the work with the families. We did this by ‘talking about talking’ (Burnham, 2005), creating a shared language, and by talking openly about self‐harm, to indicate that the therapists were not afraid of it. It was necessary for the therapist to stay attuned to the family and reflect on their own experiences, thoughts and feelings in the moment, but not necessarily share them at this point. This is what Rober (2005) calls the ‘inner conversation’ of the therapist (Rober, 2005). This guides when to hold back and when to pursue emotional topics that could potentially be scary for the family.

Systemic family therapy and emotions

In recent years, systemic family therapists have taken account of emotions more than previously and are encouraged to make use of them in the process of therapy (e.g. Fredman, 1997, 2004; Flaskas, 1989, 2005; Pocock, 1997, 2005; Bertrando, 2007, 2010). We agree with Dallos and Draper (2010), that systemic therapists of all schools are likely to pay attention to the emotional atmosphere of a family as they discuss important issues, such as in our case self‐harm. They suggest that working alongside family members and helping them to manage their emotional reactions in more productive ways may be one of the core outcomes of family therapy.

Emotional regulation has been an important aspect of work with self‐harm within other therapeutic approaches, in particular Dialectical Behavioural Therapy (Linehan and Dimeff, 2001). It is usually considered in an individualistic paradigm, but a systemic approach explores the interactional expression of emotion. In one family a young person described self‐harming whilst listening to her parents argue with her brother outside her door. Initially the family had great difficulty in thinking about how these things connected, but by tracking the incident, the family were able to think about how the emotional atmosphere in the house contributed to the internal emotional state of the young person. Making links between the emotional regulation of the family and the emotional regulation of the young person, placed both the self‐harm and the emotional experience in a more interactional frame.

Facing the feelings in family therapy: addressing the communication of emotion within the family

As a team we became interested in what we perceived as commonalities across the presenting young people and families in the early phase of therapy. In the sessions we noticed that in many of the families the emotions of the young person were the only ones that were readily talked about; all of the negative emotion was attributed to the young person. What was described as happening at home, was altogether avoiding the articulation of emotions. We developed the following working hypotheses.

Within these families that there was a difficulty in talking about emotions; it did not feel safe enough within the family at this time to talk about emotions, particularly ‘negative’ ones.

Often subsequent to a young person self‐harming, family members found it even more difficult and were less likely to talk about emotions, due to a fear of making things worse.

Following an act of self‐harm there was a sense of helplessness, not knowing what to do, and being overwhelmed by a range of different feelings, which were unable to be talked about. One set of parents described not even daring to ask their daughter, ‘How are you?’ in case the resultant emotion triggered an episode of self‐harm.

Exploring the family's understanding of self‐harm

As suggested by the manual, in the engagement phase we explored what different family members thought was behind the act. We asked questions, for example:

(To parent/s) ‘What do you think are the situations that make it more likely for your daughter to harm herself? What do you think the self‐harm is about?’

(To young person) ‘Do you agree? What situations make it more likely to harm yourself? Do you have any particular thoughts or feelings that will make you more likely to harm yourself?’

Figure 1 shows an imagined example of a family's understanding of the self‐harming behaviour that we would map with pen and paper, and then deconstruct with the family.

In the early phases, different family members’ explanations for the self‐harming behaviour could be discordant, resulting in polarisation. A parent may attribute self‐harming behaviour as part of the child having been adopted, whilst the young person accounted for it as a response to the family arguments. We found that a position of respectful interest in alternative stories was both validating and allowed for people to step back from the emotionally driven interchanges and allow for multiple understandings of the self‐harm.

Attending to beliefs and behaviours

Dallos and Draper (2010) describe how families have sets of beliefs and explanations about the role of emotions in their lives. With this in mind, we were curious about where, how and in which context family members developed their unique ways of experiencing, doing and articulating particular emotions. Asking questions that linked to childhood experiences, cultural and gender stories connected behaviours to beliefs. Using circular questioning to examine current contexts and rank emotional interchanges enabled family members to develop an observational position on their own emotion display, and have an understanding of the interactional effects. This helped them to regulate themselves better. Examples of questions to explore a family's beliefs and behaviours around a particular emotion are shown in Figure 2.

Figure 2

Questions to explore beliefs and behaviours.

Tracking interpersonal patterns and trigger events

As therapy progressed, we linked emotions, interpersonal patterns and trigger events for self‐harm through questions that tracked patterns. Figure 3 is an excerpt from a therapy session.

Figure 3

Transcript of tracking an act of self‐harm.

The tracking of this event helped the family to see that rising emotional exchanges, even when not directly involving the young person, contributed to rising internal emotion in the young person, which led to self‐harm. The tracking of the interchanges in several of the families opened up understanding of the ways in which an ‘atmosphere’ was created within the home, that then led to an escalation of affect within the young person. By developing a detailed sequential account, opportunities arose for the interruption of sequences and the development of alternative ways of relating.

Deconstructing constraining discourses or contexts

Families and individuals in western societies are positioned within constraining discourses, for example, that teenagers go through a phase of being miserable. As described earlier, negative affect was often located in the young person and self‐harm was perceived as the display of negative effects of adolescence. In their concern about their teenager, the emotion that others in the family felt was often not noticed or dismissed. The location of the negative emotion does not necessarily have to be the young person; it could also for example be a parent with a mental health diagnosis.

Opening up conversations in family therapy that allowed understanding of how different contextual discourses constrain action and expression of emotion of not only the young person but of others was important. The young person did not have to be the only one who was allowed to express depression. In one of the families, the young person was described as always being depressed since she had been a teenager (as teenagers ‘go through depression’), whilst her father sat in the room crying and shaking. The therapist asked who was most depressed. All members of the family pointed to the mother. This opened up the possibility for externalising conversations about depression and its influence on all members of the family.

Another constraining idea that families could present, was that cutting meant that the young person was potentially ‘totally lost’ or on her way to suicide. One father spoke catastrophically about the self–harm, augmented by a rotating forefinger to head ‘going mad’ hand gesture. From this belief, the father found it difficult to respond with calmness. We deconstructed the storylines surrounding this belief: what they meant, the wider narratives they were part of, and where they came from, e.g. from media representations of suicide/mental health. We explored how these meanings informed the parents’ actions in their relationship with their daughter. Exploring the past, present and imagined future was important in the co‐construction of alternative narratives, which opened up space for calmer response and alternative behaviours in response to self‐harm.

When the different family members could regulate interactional emotions, rather than just believe that the young person needed to be able to self‐regulate, heightened emotion and self‐harm became more manageable.

Facing the feelings in family therapy: team and therapist's emotions and ways of working with them

The manual suggested that the ‘therapeutic system’ may find itself reflecting the strong emotions felt by the family. The emotions engendered in the therapist and team needed to be regulated, understood and, when relevant, used as information to guide the therapy or to open up new ways of talking. Speed (1996) describes how in systemic writing about therapeutic encounters often the emotional experience is absent.

Increasing awareness of, and using our own, emotion talk to inform the therapy

The team and therapist reflected on and used their emotions therapeutically. The experience for the therapist of working with young people and families where there is a high level of emotion is a powerful one. Understanding our own cultural and historic experiences of emotion talk was a key aspect of practice.

In order to think about this, we had to revisit emotion talk in our family of origin and cultural contexts. The therapists’ current family contexts, agency context and team processes also needed consideration.

Figure 4 shows questions we devised to ask ourselves and invite self‐reflexivity.

Figure 4

Questions to invite self‐reflexivity.

Co‐constructing a shared language of emotion within the therapeutic system

In order to talk about emotions we needed to develop a shared language with the families, but equally importantly within the team. We needed to think about what words were used and whether we fully understood each other. We also needed to consider the coherence of analogical cues (facial expression, posture etc.) and what was being said.

Team and therapist

With regards to team and therapist, this required taking the time to speak about our own emotions, and asking ourselves whether we were communicating with one another clearly about how we were feeling. For example, we pointed out to the therapist incongruities between spoken language and analogical cues of the therapist in the room or in the pre‐/post‐session:

‘You look annoyed, what is that about?’

‘I wouldn't describe it as annoyed, I feel irritated.

It could be easy to make assumptions that we were talking about the same thing when we used words that denoted emotion. Admitting our own difficult feelings about family members to each other offered us opportunities to use the information to inform the therapeutic interventions. Being able to say this to each other openly came only over time, and was a part of the development of the team process.

Family and therapist

When working with the families it was also important to deconstruct what words meant and develop a shared language. The spoken language of emotions is inextricably intertwined with the body. Analogical cues, such as tears, would be used within sessions. For example, when a mother was crying asking what the tears were saying avoided assumptions being made. There may be cultural variations in analogical cues (Tsai, 2013), and our own experiences can cause mis‐readings.

Facing heightened emotions

Episodes of serious self‐harm during the course of the therapy created moments of heightened emotion in the family, young person, therapist and team. What we found in such situations was that heightened emotions (e.g. of anger, horror, panic, fear) shut down the emotion conversation within the family even further. We could find ourselves talking a lot about the emotions we saw in the family, but little about what we were feeling. We thought that there was an isomorphic process, whereby we did not initially look at our own emotions but focused on the emotions of the families, just as we had thought that the families tended to locate all of the negative emotion in the young person. Hence the heightened emotions shut down emotion talk throughout the family therapist and team therapeutic system.

The manual supported emotion talk in supervision; the felt experience of the therapist, self‐reflexivity and prior team discussions formed a significant part of team supervision. Once again, ‘listening with our bodies’ (Fredman, 2004) was an essential aspect of this. The supervisor would ask what we saw in each other, and also pick up on the atmosphere in the supervision room. Talking about our felt experiences opened up avenues for new ways of thinking and of talking with the families. An example of emotion talk in supervision came about when a parent shouted at a therapist to take the young person out of the room, talk to her and ‘help her!’ The therapist described feeling useless, not good enough, upset and angry. She also felt embarrassed in front of the team and that they would also think she was not good enough. Through talking about this in supervision we were able to think about the experience of the parent in not feeling good enough, useless and that the therapist would think they were not good enough. Taking this isomorphic process back to the family, and the therapist describing how she felt ‘anxious’ to talk about it more, led to a situation where these emotions could be talked about.

Using therapist's own emotional self‐reflections in conversation with the family

Use of the therapist's own emotions in conversation with the family was also highlighted in the SHIFT manual. The manual invited active engagement with the therapists’ own responses, consideration of the therapeutic relevance, and whether/how to use them. This could happen within the therapy session, or through pre‐ and post‐session discussion. Guided by the idea that emotions are simultaneously a feeling, a thought and a behaviour (Furlong, 1996), we introduced our own emotions into the therapy room by using relational reflexivity (Burnham, 2005). For example, talking to the families about how we felt cross, or sad, or impatient allowed these emotions to be openly spoken about and the interactional aspects explored (see Figure 5). This therapeutic risk (Roberts, 2005) had to be balanced in the ‘inner conversation’ (Rober, 2005).

This would then open up space for discussion of emotions in response to the self‐harming behaviour that might be harder to talk about at home. It helped to highlight the interactional patterns that create emotions and thoughts and lead to action. Using the emotions of the family, but also of the therapist and team, within the sessions created a shared understanding that allowed safety around emotional expression.

Conclusions

In this article we have focused on emotion talk in the context of young people self‐harming. We are aware that there are other areas that we could have chosen to focus on when working systemically with self‐harm. The manualised family therapy treatment of the SHIFT trial was helpful in keeping the therapy focused and purposeful. We particularly liked the centrality of addressing self‐harm in the therapeutic process, as directed by the manual. The manual instructed emotion talk in the team and in supervision; we found this explicit permission important. The three‐phased approach (engagement, middle, end), what to address at what point in the treatment, and the time‐limited aspect of the treatment was crucial in our view. It maintained the integrity of the treatment, kept the therapeutic intent of interventions purposeful, and helped the therapist and team stay with the goals and the family's desired outcomes (the young person may not always have as their primary goal for the self‐harm to stop).

The manual covered many aspects of what we wanted to use in our work, but was not detailed enough in terms of attachment theory, given its relevance to the area. We also thought that the manual could have paid more attention to power and social GRRRAAACCEEESSS (Burnham, 2012) in the therapeutic system, and the wider context of the presenting families, and thought more about how this relates to self‐harm.

This article also offers some suggestions for practice, particularly in the deconstructing and tracking of emotions, and helping families and therapists to see interactional patterns that create emotions, hence challenging the orthodoxy of emotions being wholly internal felt experiences. We found a value in asking different types of questions to elicit emotion talk: those intended to influence the client and family to first notice patterns, to introduce news of difference or alternative ways of being and thinking, and to amplify those alternative behaviour patterns and beliefs. There was also a value in using statements concerning the emotional atmosphere, especially those including our own emotional state. Temporal arcs that explored emotional experience and display in family of origin, different life stages, the here and now, and the potential future were central to the process of change. Using self and relational reflexivity, getting to know ourselves and each other, and understanding more about our beliefs about emotions helped us to work more effectively with the topic of high emotional expression within the therapeutic system. Our own experiences and beliefs about self‐harm, including its use and perceived dangerousness, were important and connected to the emotional reactions that we had to young people self‐harming. Taking our own emotions and responses to self‐harm to supervision helped us to understand and explore our own reactions and the impact of our reactions on the work. We are aware of the impact of one's agency discourse, and its current emotional climate, local and wider societal contexts on the emotional experience of working in the area of self‐harm. For example, a Serious Untoward Incident will affect the work, or press reports about CAMHS not responding well enough to an increase in self‐harm, has an emotional impact on the worker. This is perhaps a topic for a future article.

Last, and perhaps most important, the authors hope to have encouraged systemic practitioners to pay further attention to emotions. We believe it is of use to open up discussion in and outside sessions about emotions in the therapeutic system, particularly when risk is high and it feels more difficult to do so. We invite practitioners to use their own emotions to elicit emotion talk in the therapeutic system.

In family therapy training and systemic practice, we believe that there is a move towards emotion talk and the emotion of the therapist being more openly discussed and thought about. However we would see this as an ongoing process that therapists, supervisors and training institutions should be highlighting and developing. In our view, it is easy to get caught up with discussion and curiosity about the family from a distance, instead of being able to keep the focus on the emotional processes that occur between family and therapist, team and supervisor.