How are secure attachment relationships fostered through talk between teachers and students who have been adopted? A conversation analysis

Abstract

Despite the growing recognition of the importance of supportive teacher–student relationships to create safety for young people who have experienced early adversity and trauma, there is not a clear understanding of what factors make for positive school relationships and how these can be fostered. The aims of the study were to explore how children with challenging emotional backgrounds are supported by their key adult in school and how this occurs in the process of conversations between them. Three student–teacher pairs from a specialist school took part in a semi-structured interview about their relationship. Data was analysed using conversation analysis. Analysis found how teaching staff use several conversational markers in talk with children with attachment difficulties when emotional experiences are raised and when troubles occur in navigating difficult conversations. Recommendations for clinical practice and future research are made.

Introduction

Positionality statement

I am neither a teacher nor an adopted child, and am fortunate to not have experienced significant childhood trauma. Thus, I have not had the same experiences as those participating in the interviews and as a White middle-class woman bring a different perspective. However, through my personal connection, and in my professional work as a trainee clinical psychologist, I have a deep understanding of the difficulties adoptive families face, and with schools who support these children. Therefore, I undertook three bracketing interviews over the course of data collection and analysis to become more aware of my own assumptions. Namely, my hope that teachers are skilled in creating safe environments and meaningful relationships with children who have experienced challenging backgrounds was particularly attended to, as to not overshadow the findings from the participants. Furthermore, to minimize the impact of these biases, the design and analysis of the research was undertaken with co-authors who do not have the same personal connections to the subject.

Background

Teachers play an important role in the lives of children, and relationships have been described as the 'infrastructure of school success' (Pianta, 1999). The influence of teachers escalates with the increasing amounts of time young children spend in formal educational settings and are particularly important at periods of critical development (Silver et al., 2010). It has been advanced that education staff have the power to create and foster a culture of warmth, nurture, kindness, and positive relationships within their school environments (Public Health Scotland, 2020). Students who have close relationships with their teachers are more likely to experience academic interest, engagement, achievement, self-efficacy, and motivation compared to students with more distant relationships (Raufelder et al., 2016; Roorda et al., 2011). Some qualities of connected student–teacher relationships have been found to be feeling liked (Wang and Eccles, 2013), feeling known, trust, respect, and mutuality (Gillespie, 2005).

Better understanding the quality of student–teacher relationships has the potential to improve children’s developmental outcomes (Myers and Pianta, 2008). This is particularly important for looked-after and adopted children who are extremely likely to have experienced early abuse and neglect (Dann, 2011), and are highly at risk of academic and social difficulties (Daignault and Hebert, 2009). Yet, the presentation and aetiology of this trauma in the classroom environment is often misunderstood (O’Neill et al., 2010).

It is thought that when children have experienced pervasive and prolonged trauma involving multiple events, which may be both intra-familial and interpersonal, it can have significant impacts on physiological, social, and cognitive development (Luxenberg et al., 2001). This early adversity causes impairments to children’s attachment patterns, difficulties with identification and regulation of core biological processes, affect regulation, dissociation, behaviour control, and self-concept (Cook et al., 2005). The extreme stress of early trauma changes the structure and physiology of the brain, resulting in long-term effects on functioning and behaviour (Beers and DeBellis, 2002; Gabowitz et al., 2008).

When children have experienced early trauma, their cortex will be less evolved, as their brainstem and limbic systems have not developed a strong foundation for the cortex and pre-frontal higher-order thinking to regulate these more primitive systems (Van der Kolk, 2003). Therefore, these children are hypervigilant and have a lower tolerance to perceived threat, meaning they are quicker and more likely to enter into a stress response of fight, flight, or freeze, as a survival strategy to protect themselves (Perry et al., 1995). This is important when understanding the school environment, as children with a less developed cortex, operating at a brain-based survival level, will not be able to access processes in this top region of the brain as easily (Barr, 2018). These processes include inhibition, the ability to receive and retain new information, problem-solve, rationalize, and use logical thinking; essentially the mechanisms needed for a child to be able to learn (Bomber, 2007).

The classroom environment, and the opportunity to interact with additional supportive adults in a stable environment, can mitigate the negative effects of trauma (Dorado et al., 2016). School connectedness has been shown to be one of the strongest predictive protective factors for young people who have reported physical or sexual abuse, or violent or unstable home lives (Saewyc et al., 2006). A key aspect of repair from childhood trauma and attachment difficulties occurs within the context of a relationship, and as children spend a significant amount of time in the school setting, teachers can act as subsequent key attachment figures for children where they can be provided with safe, positive, intersubjective experiences (Kim and Cicchetti, 2010).

One study of nursery school children showed that it was not the mother’s absence in itself that increased stress hormones such as cortisol, but the absence of an adult figure who was responsive and alert to their states moment by moment. If there was a member of staff in the nursery school who took on this responsibility, infants cortisol levels did not rise. Without such a figure, the child became stressed (Dettling et al., 2000). Additionally, Koomen and Hoeksma (2003) demonstrated the role of kindergarten teachers as a safe haven and a secure base by showing high initial scores for child security seeking at school entry and sharp decreases in the weeks thereafter.

Although there is growing recognition of the importance of supportive relationships with teachers for this population of children (Bomber, 2007), much of the literature is offered through anecdotal evidence and theoretical consideration of the relevance of attachment and trauma in the school setting (Barr, 2018; Dann, 2011; O’Neill et al., 2010; Wright and Ryan, 2014). Furthermore, research has not focused on a specific group of children, such as children who have been adopted, but rather have focused on attachment patterns.

Helpfully, Verschueren and Koomen (2012) give an overview of the research that is available on teacher–child relationships from an attachment perspective. They identify four possible contributions from research in this field. First, relationship qualities have been conceptualized as closeness, conflict, and dependency. These qualities are most frequently assessed using the Student–Teacher Relationship Scale (STRS; Pianta, 2001). Using the STRS, O’Connor et al. (2012) found elevated levels of teacher–child conflict during childhood were associated with externalizing behaviours in late childhood, whereas low levels of teacher–child closeness were associated with internalizing behaviours. Furthermore, the effects of insecure/other attachment on externalizing and internalizing behaviours in late childhood were mediated through teacher–child relationships during childhood and early externalizing and internalizing behaviours.

Second, research has highlighted the importance of teacher sensitivity or responsiveness to children’s needs as a central proximal determinant or antecedent of relationship quality. Buyse et al. (2011) found with high teacher–child closeness, less securely attached children are no longer at risk for more aggressive behaviour. They also found that with high teacher sensitivity, less securely attached children are no longer at risk for developing less close relationships with their teachers.

Third is the contribution research makes to understanding the consequences of teacher–child relationship quality and the intervening mechanisms explaining these effects. For example, Kobak et al. (2012) showed how negative interactions with teachers predicted increased sexual risk-taking behaviours and females’ early romantic involvement. Preoccupied states of mind increased risk for early romantic involvement and the likelihood that females would engage in risky sexual behaviour. The findings demonstrate how adolescents’ school experiences contribute to adaptation in romantic relationships in mid to late adolescence above and beyond representations of parent–child attachment.

Fourth is the guidance literature gives in developing interventions aimed at improving teacher–child relationships and interactions. Spilt et al. (2012) evaluated the effects of a relationship-focused reflection program for teachers. The program yielded changes over time in closeness for about half of the teacher–child dyads. In addition, teachers with high efficacy beliefs were more likely to report declines in conflict than low-efficacy teachers. Lastly, significant increases were found in observed sensitivity.

Due to a profound mistrust of others, children with significant attachment difficulties are highly prone to emotional dysregulation, where various forms of anxiety, mistrust, withdrawal, and expressions of anger are frequently triggered (Hughes, 2017). When teachers and schools are unaware of the effects of trauma, they might unintendedly interpret behaviours associated with survival strategies as disrespect and defiance, rather than symptoms of developmental trauma, responding with behavioural punishment (Day et al., 2015; Sweeney et al., 2016). This can reinforce the child’s survival strategies and push them farther from embracing school as a safe and emotionally protective space (Wright and Ryan, 2014).

Therefore, it is important to understand how teachers foster secure relationships with this population of young people in school. Not only because interactions with their teachers can potentially represent an important opportunity for re-learning and repair from childhood trauma, but also as by helping children develop a better sense of safety with their key adult in school to be co-regulated will enable them to better access learning in the school environment. Empirical research into teacher–student relationships for children who have experienced early trauma in childhood is sparse. Despite this, the research that is available has made important contributions to better understanding that educators do play an important role in supporting children with attachment difficulties through their relationships but is limited in terms of its focus on quantitative data and priority of teacher’s perspectives. There is a gap in the literature of research from a constructivist paradigm, exploring both child and teacher’s perspectives. Research in this area has identified some factors of supportive teacher–child relationships but has not attended to how in their interactions these relationships are fostered.

Research question

From a social constructivist position, selves and identities are constructed through the interactions we engage in with others, particularly in conversations. Thus, this research chose to use conversation analysis to explore how in their moment-to-moment interactions important aspects of relationships are fostered with teaching staff. Children who have experienced trauma and subsequent significant attachment difficulties may benefit considerably from the opportunity to develop a sense of trust and emotional containment in their interactions with teaching staff who can represent important stable figures in their lives.

Specific questions are as follows:

(1) What conversational markers are used by teachers and students with attachment difficulties in their moment-to-moment interactions?

(2) Through their talk, how do teachers contain emotions which may be triggered by the children’s prior experience of adversity and trauma?

(3) Through their talk, how do student–teacher pairs navigate difficult episodes in their conversations?

Methodology

Design

This study employed a cross-sectional design, where each pair participated in one joint interview. Interview questions encouraged student–teacher pairs to discuss how they understood positive, negative, and neutral aspects of their relationship, to promote material to be used for detailed conversation analysis. Semi-structured interviews concentrate conversation on a specific topic of interest so that the way in which participants introduce and manage talk about their relationship can be considered. There is clinical relevance to this design, as this format is not dissimilar to clinical settings where participants are asked to respond to a series of questions by a third-party member in the room (Crix et al., 2012).

Conversation analysis perceives the in-depth study of fragments of data to be helpful in revealing properties of the broader picture through focusing on ‘naturalistic’ social–cultural–moral phenomena in everyday interaction (O’Byrne et al., 2008). Therefore, it can reasonably be expected that some patterns that emerge in these student–teacher interactions around their relationship to be similar in other student–teacher interactions surrounding similar circumstances of children who are adopted being taught in a specialist setting.

Ethical considerations

Ethical approval was granted through University of Plymouth Research Ethics Committee. Vulnerability of participants was carefully considered during the design of the study; the aims of the research were made transparent to all participants from the initial point of contact. Although not explicitly talked about at interview, it was essential to confirm with parents prior to meeting with the child that they understood they were adopted as this language was used in the information sheet. This ensured that the researcher or teacher did not reveal any information to the child that was unknown to them. A plan in the event of the child becoming upset during the interview was agreed upon with the school.

Recruitment

Eleven headteachers of secondary schools in South England were contacted via email, inviting them to take part in the research. One specialist trauma-informed school agreed to participate. All other secondary schools who were contacted declined or did not respond. Due to the nature of being removed from their birth family, children in the care system and who are adopted are extremely likely to have been subject to abuse and neglect and experience attachment difficulties. There is a robust legal framework around children under the care of the local authority (e.g. children in foster care), and so children who are adopted were chosen as the sample for this study.

The inclusion and exclusion criteria for recruitment were as follows:

Inclusion criteria

• Children legally adopted in the United Kingdom. It can be assumed that due to the circumstances of needing to be removed from their birth home and placed for adoption, this population of children are highly likely to have experienced developmental trauma.

• Children taking part must be 11–16 years old. Due to the impact of developmental trauma, children are often developmentally younger than their chronological age; thus, interviewing children younger than this age bracket may not have yielded sufficient data for this type of design.

• Both teachers and teaching assistants could take part in the study.

Exclusion criteria

• If the caregiver did not hold parental responsibility for the child.

• If the identified key worker declined to take part in the study, the child would not be able to take part either, even if parental consent had been granted.

Participants

The participating school is an independent school, based in South England, supporting young people with Special Educational Needs (SEN) who have experienced trauma and adverse childhood experiences. The school accepts students from years one to seven who have an Education Health and Care Plan with Social, Emotional, and Mental Health as the primary need. The school has a therapeutic approach guided by trauma-informed principles and has a clinical multi-disciplinary team attached to the school. The multi-disciplinary team works closely with teachers and support staff to ensure that students grow socially and emotionally alongside their educational development. Each child has a tailored curriculum to meet their learning needs. Alongside the National Curriculum, the school also use the Dimensions, ‘Learning means the world’ curriculum. There are 5 classes at the school, with approximately 10 to 15 students each. In this trauma-informed specialist setting, children and their teachers spend more time together and there is more of a focus on building relationships and supporting pupil’s emotional wellbeing, compared to a mainstream school.

Three staff–student pairs took part in the study. All sets of pairs confirmed that the staff member was the key adult for the pupil and also had the best relationship with them compared to other members of staff. Demographic information was collected prior to interviews and all participants were White British. To protect confidentiality and anonymity, no information was collected about children’s birth, developmental, or adoption history. All names used throughout the article are pseudonyms.

Pair 1: Naomi (13 years old) and Catherine. Naomi has attended the school for two years and Catherine has been one of her key adults for the duration of this time. Catherine has worked for the school for four years as a teaching assistant and has recently transferred job role to the Pastoral Co-Ordinator. This is Catherine’s first job role in education.

Pair 2: Kyle (13 years old) and Sasha: Kyle has attended the school for five years and Sasha has been his teacher for the last two years. Sasha has been a teacher for three years and has solely worked for this school since qualification.

Pair 3: Jacob (11 years old) and Sebastian: Jacob has attended the school for five months and Sebastian has been his teacher since he came to the school. Sebastian has been a qualified teacher for four years, but has only been teaching for three years, all of which have been at the participating school.

All of the teaching staff interviewed reported receiving ongoing training around trauma as continuing professional development throughout the year, predominantly from the multi-disciplinary team. Sasha is also currently undertaking the Trauma Informed Schools certification.

Procedure

Informed consent was collected from all participants (parental consent was gathered for children) prior to interviews. Interviews took place in the participating school, during school hours, and were audio-recorded. On the day of the interview, the research was explained again to both the child and adult, and both were given an opportunity to read the information sheets and ask any further questions. The researcher sat with the pair, asking them a series of 10 semi-structured questions.

Method of analysis

After orthographic transcription, the main researcher (R.P.) used ‘unmotivated listening’ to identify initial collections of patterns of interaction, which were then refined by continuing to check the corpus of data for consistency. These patterns centred around conversation about positive, negative, and neutral aspects of the participants relationship. Samples from the different collections were then analysed in a more detailed Jeffersonian-transcribed format (Jefferson, 1988) by the main researcher and secondary researcher (R.D.). These practices were examined in relation to the functions that they served and refined according to the research aims.

Transcript sections were chosen that best answered the research question, using one transcript from each pair. This was based on a consensus between the authors that the sections exemplified both positive and non-problematic interactions compared to more difficult or delicate interactions. In particular, the later contained instances where negative emotions were expressed by the children. These problematic instances appeared to indicate ‘intrusions’ of negative memories and feelings for the children from past events in that they were not a response to the teachers’ prior speech acts. Finally, analysis was checked with all of the co-authors.

Analysis

Managing difficult emotional experiences: A dance of moving away and towards expression of emotion

Extract 1: Naomi and Catherine

Characteristic of all pairs at points during the interview there was a process of moving away from and then back in a possible safer position towards expressions of emotions. Teachers frequently responded to children’s emotional experiences by initially moving away from and de-escalating talk about difficulties. This was a recurring strategy in the interview whereby delicate emotional talk is postponed for a safe time and place in the teacher’s view.

In extract 1, there is an oscillation between Naomi and Catherine of curiosity about Naomi’s experience and reinforcing the narrative that things are good now. This pattern of Catherine directing the talk to positive aspects is evident throughout the interview of the pair. In this sequence, it appears that Naomi attempts to engage in telling a story of being unhappy at a previous school. Catherine partly opens this conversation before moving onto a more positive story about the current school.

16:30

1 Catherine: hm ((slight giggle))

2 Naomi: and my dad used to work at that school

3 Catherine: YOUR DAD used to work ↑at that school?

4 Naomi: go to that school

5 Catherine: oh ↑go to that scho::ol ↑ah:::

6 (.) you were there for quite a while we::ren’t you (.) yeah?

7 Naomi: ↓when I got to ye↑ar five

8 Catherine: when you got to year ↑five (.) and what was year ↓five ↑like

9 Naomi: ◦er:: it was rubbish◦

10 Catherine: rubbish w(h)hy w(h)as it r(hh)ubbish

11 Naomi: every↑one in the class expect molly was #mean to ↑me#

12 Catherine: aw::: that’s ↓not very ↓nice

13 Naomi: except molly

14 Catherine: aw:: >everyone was mean to you apart from molly<

15 ↑you’re still friends with molly now ar:en’t you

16 Naomi: = ↑yeah

17 Catherine: = ye↓ah that’s good

18 Naomi: (.) ◦yeah◦

19 Catherine: and then at this school::l people (.) aren’t (.) so mean?

20 Naomi: ◦yeah◦

21 Catherine: = no so you’ve got lots LOTS of more friends here haven’t you

22 Naomi: yeah

17:09

Extract 1 follows after a staff member accidently walks into the interview room and Catherine comments that 'it’s so busy all the time'. Prompted by the interruption to the interview, the researcher asks Naomi how this school is different to her previous school. There is then a sequence characteristic of the pair, where Catherine asks a series of rhetorical and tag questions ('wasn’t it') about Naomi’s previous school, to which Naomi responds as expected by conversational ‘rules’ of confirming these statements made by her key adult.

However, on line 1, Catherine changes the interaction pattern, no longer using rhetorical or tag questions, and slightly giggles as if she no longer knows how to continue the conversation. Now, on line 2, without the expectation of how to answer the question predefined by Catherine, Naomi makes an open statement in the conversation that 'my dad used to work at that school'. Catherine is somewhat surprised by Naomi’s open statement, shown by her increased volume (line 3) and repeats Naomi’s statement back to her with a rising intonation, making it into a question. Naomi corrects herself (line 4) that her dad attended her previous school as a pupil. Catherine almost exaggerates her response, emphasizing 'oh go' and elongating 'ah' to show to Naomi she has understood what she is saying. There is a slight pause (line 6), before Catherine returns to the pattern of rhetorical questioning 'you were there for quite a while' plus two tag questions 'weren’t you' and 'yeah'.

Naomi gives Catherine another invitation to thinking about her experiences at her old school (line 7). This time, Catherine again repeats Naomi’s statement back to her, and also gently asks 'what was year five like' (line 8). This soft curiosity by Catherine with no qualifier or tag gives rise to Naomi tentatively telling Catherine 'it was rubbish' (line 9). There is a slight mismatch to Naomi’s emotion in Catherine’s response (line 10), as she has a suppressed laughter to her voice, perhaps somewhat taken off guard that Naomi has given an insight into her emotional experience. Despite this, there is still genuine curiosity in Catherine’s response. This helps to open up the conversation further and Naomi tells Catherine that 'everyone' was mean to her but qualifies this with 'expect Molly'. Naomi’s voice also becomes creaky when saying 'mean to me' indicating this is a delicate subject. On line 12, Catherine validates Naomi that it is not a nice experience when people are mean to you.

On line 14, Catherine creates connection and paraphrases back to Naomi that 'everyone was mean to you apart from Molly', showing she has understood what Naomi has said. Catherine then moves away from thinking about this negative emotion by use of rhetorical questioning that Naomi and Molly are still friends (line 15). This closing off to the negative emotion is furthered by the tag 'aren’t you'. Naomi confirms Catherine’s statement (line 16), and on line 17, Catherine says, 'that’s good', reinforcing the need for positive talk and avoidance of internal experiences. This act of closing off to the emotional aspect leaves Naomi to briefly pause (line 18) and become quieter, but again confirming Catherine’s statement. Catherine furthers this need for confirmation that things are good and invites Naomi to confirm that 'at this school people aren’t so mean' (line 19). The answer is already preset and so Naomi again conforms to this by confirming with a quiet 'yeah', suggesting Naomi is somewhat shutting down too to the conversation. On line 21, Catherine again closes off and emphasizes the need for talk to be positive, repeating strongly that Naomi has 'lots lots of more friends' with another tag question which Naomi repeats the pattern of confirming (line 22).

Navigating difficult conversations

Extract 2: Jacob and Sebastian

The interaction between Jacob and Sebastian in extract 2 shows features that were characteristic of the three pairs but appeared in different ways, in how potentially negative escalations in the relationship was managed in their talk. The pair repeatedly exchanged in ‘joking’ banter initiated by mildly mocking remarks by Jacob. Part of the delicate task of managing this for the teacher was to maintain a positive connection with Jacob but at the same time maintain the boundary of his authority as a teacher. Extract 2 comes at the beginning of the interview, during which the pair are discussing how well they would rate their relationship and where potential disagreements originate.

03:07

1 Researcher: and how do you and sebastia::n get on?

2 Jacob: (.) ◦hmm◦ not gr::eat.

3 Researcher: not gre::at?

4 Jacob: (1.6) >↑what would you say<

5 Sebastian: (1.3) ↑I think pe::rsonally we get on quite well, (1.05) I [I think

6 Jacob: APART]

7 >from when < I get ann#oyed# at you

8 Sebastian: (0.7) but > why is it < you > get annoyed at me<.

9 Jacob: = because y(h)ou make too many d(h)ad j(h)okes hhh.

10 Sebastian: I make t::oo many dad jo:kes

11 Jacob: = ah(h)a ((laughs)) a::bsolutely rubbish.

12 Sebastian: = a::nd I’m also the one that (.) kind of makes you do the work

13 ↓aren’t I?

14 Jacob: (.) ◦mmm◦

15 Sebastian: = and I think mo::st of our most of o::ur disagreements > I’ll call ↓them<

16 (0.9) come fro::m (1.6) me going ↓co::me on it’s time to do some

17 learning↓ (1.1) but ↑I think ↑I think personally on the who::le (1.0) we

18 get on quite well I’ve got a lot of respect (.) and a lot of time for yo::u \

19 Jacob: (.) mhmm

20 Sebastian: and (1.0) >the same as all the other children in the class< and I I ↑fheel

21 personally (.) that yo::u have a lot of time and respect for me ↓as well↓

22 (1.6) but (0.7) I:: can only sa- I can only go off [what I see

23 Jacob: second chances] ohhnly (.) no thi::rd. ((slight laugh))

04:06

In response to the researcher’s initial question about how the pair get on, Jacob has a slight hesitation before responding 'not great'. This slightly enigmatic answer draws the researcher in to question Jacob (line 3). Sebastian does not respond, and so Jacob defers to asking Sebastian directly on line 4 what his teacher thinks of their relationship, clarifying that he was not serious in his criticism (line 2). Sebastian begins by hedging, and also softening, his response in line 5 with the use of 'I think'. This softening is continued a further two times in line 5 through 'personally' and 'quite well'. This repetitive softening is a strategy to manage the delicacy of this conversation.

Jacob partially agrees to Sebastian suggesting the pair get on quite well (line 6) and continues the playfulness teasing (line 7). Jacob uses a creaky voice (line 7) when saying 'annoyed'. In response to Jacob introducing this potentially critical comment, Sebastian neither agrees nor disagrees with Jacob (line 8), but questions why Jacob gets annoyed with him. This is reasserting his authority gently and without directly confronting Jacob. On line 9, Jacob quickly responds by a much more obvious joke to Sebastian reasserting authority as the teacher. On line 10, Sebastian repeats what Jacob has said that he makes 'too many dad jokes', again neither agreeing nor disagreeing with the statement. Jacob cuts in quickly (line 11), increasing the playfulness, now laughing at Sebastian’s 'rubbish' jokes.

On line 12, Sebastian continues where Jacob cut him off, reasserting his role as the teacher, and making an epistemic claim that Jacob gets annoyed at him for making him do schoolwork. Sebastian does this very gently, qualifying the statement with 'kind of' and a tag question of 'aren’t I' (line 13). Jacob becomes quieter (line 14), as he has been taken away from his joking banter, but also partially confirms Sebastian’s statement with 'mmm'. Sebastian continues (lines 15–18) by taking charge of the conversation. This is done delicately by Sebastian, softening with the repeated phrase 'most of' and by qualifying that the 'disagreements' aren’t really disagreements by saying 'I’ll call them'. Sebastian also takes part of the responsibility of these 'disagreements' (line 16), instead of putting the blame to Jacob. Sebastian’s statement is softened even further by now turning the delicate subject of 'disagreements' to a more positive 'on the whole we get on quite well'. The use of three qualifiers 'I think I think personally' (line 17) gently implies that this is Sebastian’s opinion, warding off potential disagreement from Jacob. Jacob continues neither agreeing nor disagreeing to Sebastian (line 19) but does not dispute Sebastian’s claims. Sebastian perhaps takes this as permission from Jacob to continue talking about this delicacy in their relationship and maintains control of the conversation. On line 20, Sebastian pauses before going on to talk about other children, implying that this is not a personal attack on Jacob about their 'disagreements' over schoolwork. Sebastian attempts to draw Jacob into confirming that he also has 'a lot of time and respect' as a mutual feeling towards Sebastian (line 21), further qualifying this with 'I can only go off what I see' (line 22). Jacob now responds to this attempt of validating their relationship is positive, returning to his playful stance (line 23).

Extract 3: Kyle and Sasha

Extract 3 gives an example as to what happens when talk about troubles is less contained, and the teacher becomes more defensive and authoritarian. This extract follows after the researcher asks the pair to give an example of a time they have not got on so well. Kyle verbally growls at this question, and so Sasha tentatively seeks permission from Kyle to tell the story of a particular incident between the pair, to which he agrees.

05:13

1 Sasha: um because he had a very very wobbly day (.) a:nd he decided that he

2 was going to rea::lly hur::t me < which landed me in hospital didn’t it

3 (.) with my knee>

4 Kyle: (1.4) not the o::nly time you’ve landed in hospital

5 you got your (.) head hit by a ston:e by Bobby

6 Sasha: ↓yes::

7 Kyle: =>mm mm < then NEARly got hit (.) the second time

8 >I wish she did hit you<

9 Sasha: = uh

10 Kyle: = I would have ↓laughed.

11 Sahsa: = uh ↑umm(.) have we got stuff in our mouth?

12 Kyle: = uh nuh nuh nuh yeah yeah yeah

13 Sasha: [I don’t want to see anything aroun::d it

14 Kyle: >yeah yeah yeah] yeah<

15 Sasha: ◦it makes me feel◦

16 Kyle: = SEE food.

17 Sasha: see food mmm (.) UM (.) so that’s one of (.) 1 day ◦do not do that◦

18 Researcher: ↑you didn::t get on: that day

19 Sasha: we didn’t get on that da:y (.) kyle wasn’t making the correct choices

20 which he he’s ↑really good at doing the right thing

21 but that day he was just (.) very very wobbl::y

06:01

On line 1, Sasha emphasizes how much the pair did not get along, by repeating the word 'very' and using strong language of 'decided', implying there was an intention to Kyle’s actions to hurt her. Sasha further conveys this message to Kyle about how much he hurt her by elongating her speech (lines 2 and 3) and using 'didn’t it' to prompt Kyle to confirm this. Kyle takes a fairly long pause before answering, suggesting he is unsure how to respond to this claim of intending to hurt his teacher, perhaps unable to mentalize Sasha’s feelings, and so defends himself that he is not the only child who has hurt Sasha (lines 4 and 5).

Sasha confirms this is not the only time she has been in hospital (line 6); however, possibly Kyle has already become dysregulated from recalling this negative memory, shown by quicker and louder speech (line 7) and a shame response 'I wish she did hit you' (line 8). Sasha immediately responds 'uh' (line 9) as if to warn Kyle, reminding him of her position as the teacher. Kyle somewhat ignores this and continues by further antagonizing Sasha (line 10). Sasha makes a second non-word prompt to Kyle, 'uh umm', that he is crossing a boundary in their relationship, and turns to commenting on speaking with his mouth full (line 11). This redirection stays with the claim to authority by Sasha to remind Kyle of manners, but in a way that is more de-escalating than directly telling Kyle off for saying he wishes his teacher got hit a second time. Furthermore, Sasha uses 'we' although she is not eating herself, attempting to bring the pair back together.

Kyle now regresses into a more ambivalent response to this conflict (line 12) and indicates an attempt to dismiss Sasha’s criticism by dismissing comments 'yeah yeah yeah'. Sasha further engages with attempting to gain some control over Kyle and the conversation (line 13). One interpretation is that Sasha is angry about Kyle’s lack of empathy to her being hurt and projects this anger into the present moment of not wanting to see food in Kyle’s mouth (line 13). Kyle ignores Sasha’s attempts to contain the conversation (line 14) and raises his voice to her to 'see food' (line 16), defiantly doing what his teacher has asked him to stop doing. Sasha now appears to become somewhat lost in the conversation, flitting between the event she is recalling when Kyle hurt her, and asking Kyle to stop speaking with his mouth full (line 17). Sasha attempts to initiate a repair by stating tentatively that this was 'one day' but this is overtaken by another claim to her authority saying, 'do not do that'. It is at this apparently failed repair that the researcher steps into the conversation to help contain dysregulation for both Sasha and Kyle, orientating Sasha back to the initial question which started the conversation, and also remind the pair that it was 'one day' not always that they do not get along (line 18).

On line 19, Sasha continues this vacillation. She initiates another attempt at repair by taking the responsibility for both of them that 'we didn’t get along that day', but then oscillates back to make a blaming claim about Kyle, indicating again that it was a 'choice' to hurt her. Yet again this is followed by an attempt at repair by adding that Kyle can be thoughtful, qualifying it with him being 'very wobbly' on that day (line 21). Here, 'very' emphasizing that this was and needs to be understood as an extreme state for Kyle.

Kyle continued to be dysregulated throughout the remainder of the interview (a few more minutes), calling Sasha a 'dumbass', 'loser', and 'ugly' and saying 'will you shut up that’s a good question' to the researcher.

Discussion

This research used conversation analysis to explore how in their moment-to-moment conversation relationships are fostered between a child who has been adopted and their teacher or teaching assistant. The conversational markers found across the three extracts in this context of student–teacher relationships in a specialist trauma school were epistemic claims, claiming authority, rhetorical questions, tag questions, softening, change of narrative, institutional authority, and hedged questions.

Findings indicate that a critical dilemma for the teaching staff is that they have a knowledge of children experiencing trauma and attachment difficulties, and that troubles in talk may need to be responded to in empathetic and attuned ways. However, they also have to strike a balance of maintaining boundaries as a teacher. The analysis suggests that teaching staff have developed impressive skills in using these markers to manage this complex dilemma in talk with these children.

It is suggested that these markers found for how teachers foster relationships with these pupils may be an important part of the mechanisms of previously identified factors of student–teacher connection. For example, they indicate how closeness and conflict (Pianta, 2001), and sensitivity and responsiveness (Buyse et al., 2011) are managed in patterns of conversation. Also, how trust, respect, mutuality, and a sense of feeling known (Gillespie, 2005) are fostered.

Extract 1 highlights how through their talk, teachers provide containment for students when emotional experiences are raised. Here, there was seen to be a dance in and out of emotional talk to prevent dysregulation for the child. The teaching assistant attempted to ground the child in the here and now as being safe and positive compared to the negative past. As the students’ key workers, these staff members appear to know that these children have experienced various adverse childhood events and consequently are likely to be experiencing trauma and attachment difficulties. Consequently, they also know that difficult feelings and memories may intrude triggering ‘troubles’ or ruptures in their conversation when difficult subjects come up in conversation. Therefore, as is seen, the teaching staff position themselves in taking management of the talk to avoid the child becoming overly upset.

Extracts 2 and 3 show how the pairs navigate difficult episodes in their conversation as well as manage emotions which may be triggered by the children’s prior experience of adversity and trauma. Extract 2 shows how when teachers are aware and responsive to children’s states, they can appropriately align with students to deescalate potential rupture and navigate these difficulties well. However, extract 3 demonstrates how when teachers respond in a more authoritarian manner, perhaps less attuned to the pupil’s trauma history, or triggering emotional memories for the teacher, troubles in talk are managed less well, and dysregulation escalates.

The results from extracts 1 and 2 mirror existing findings that closeness between the pair and teacher sensitivity to the child contain or reduce emotional dysregulation for children with attachment difficulties which may present as both internal and external (Buyse et al., 2011; O’Connor et al., 2012). Furthermore, extract 3 relates to previous findings that when there is conflict and a lack of closeness between teachers and children with attachment difficulties, this increases the risk of internal and external dysregulation for the child (Buyse et al., 2011; Kobak et al., 2012; O’Connor et al., 2012).

Overall, these findings support amassed anecdotal evidence that teachers can play an important role in fostering positive relationships with students who have experienced early adversity and disruption to their attachment bonds to support their recovery from trauma. Importantly, they also offer some suggestions of how teachers develop impressive skills in negotiating the delicate balance between empathy towards these children to foster their emotional healing and also maintaining discipline and a clear structure for learning to take place.

Limitations

The findings from this research should be considered within the specific context in which the data was collected. The research interviews took place within a private room at the school, where pairs were asked to focus talk on the specific topic of their relationship. Although the pairs are familiar with the environment and spending time together outside of the classroom, the specific context of an interviewer asking questions and audio-recording their conversation may not be generalizable to naturally occurring day-to-day interactions between the pairs.

Furthermore, the presence of the researcher in these interactions should be noted as having a performative influence on the participants, possibly wanting to show the positive aspects of their relationship. This is known as the Hawthorne effect (Landsberger, 1958) where when participants are observed, they change aspects of their behaviour to be seen in a more positive light, compared to their performance in unobserved settings.

Lastly, issues of the generalizability of the study should be considered from the sample of participants who took part in the research. Due to the detailed analysis of CA, a sample size of three student–teacher pairs was sufficient for this study. However, the specific context of the sample consisting of all White British participants, children who are adopted, teachers who have had some training on trauma, and a specialist trauma-informed school means that findings and implications of the results perhaps cannot be generalized to mainstream school settings and children who are not adopted and who have not experienced childhood trauma.

Implications

The extracts in this article illustrate how some teachers may be more skilled in fostering trauma-informed relationships with pupils than others. Due to these children’s attachment difficulties, the complex ability to negotiate and respect boundaries in relationships will be limited, and it is common for these children to have a strong need to take control over relationships and situations when they feel unsafe. Therefore, it is difficult for the teacher to stay regulated and respond in a trauma-informed way, not only because teachers are trained in the idea of children complying and respecting the teacher but also because their own attachment patterns may be activated to become shut down or more authoritarian. Thus, it can be suggested that training be provided to teaching staff in similar settings using examples of transcripts of talk between pairs, helping them in understanding how talk contributes to their interactions and relationships with children.

For children who have experienced trauma, it may be helpful for their key workers in school to have some understanding of their history, how this has impacted on their relational security and how they interact with other people and the world, and how to not only respond in a trauma-informed way but also create a whole-school trauma-sensitive ethos. Clinical psychologists are potentially well placed in this role, and additionally through supervision can support teachers to understand their own attachment patterns and when their own coping strategies are triggered. This could be helpful for schools in identifying which staff members may be best placed in being a key adult for which children.

These findings also lead to further research questions. Additional research needs to be done in this area to strengthen our understanding of how teachers foster connected relationships with young people who experience attachment difficulties, and the impact these relationships have on repairing children’s attachment security. Lastly, as in the participating school where there is a multi-disciplinary team attached to the school, future research could better understand the implications of this service structure for this population of children. This additional research has the potential to inform policy around quality of education for children with trauma and attachment difficulties.