The construction of a ‘traumatized’ refugee child in need of safety in Norwegian kindergartens

This paper explores how kindergarten teachers relate to the concept of safety in the integration of refugee children. My research findings reveal that the concept of safety, or trygghet in Norwegian, is understood as emotional support and comfort accorded to children. The notion of trygghet emerges as an important value-laden concept that shapes and largely influences teachers’ everyday work, and how they relate to children and their parents. The idea of safety is traceable within an institutional discourse mediated by social technology, such as the International Child Development Programme and the Resource Centre for Violence and Trauma programmes, which have their roots in psy-discourses. My findings indicate that the construction of a potentially ‘traumatized’ refugee child is interconnected with the assumed need for safety. When using this discourse in their daily work, the kindergarten staff contribute to constructing and reproducing a specific category that I refer to as the ‘traumatized’ refugee child in need of safety. This standardized understanding risks categorizing refugee children, and highlights how professionals can get caught up in dominant discourses that universalize their routine practices.


Introduction
Kindergartens and childcare centres can provide a safe, stabilizing experience for children and families from refugee backgrounds who have experienced trauma, and can provide a platform for them to overcome their vulnerability (Signorelli et al., 2017). Therefore, access to a good kindergarten experience has been deemed essential to support children's holistic development, and alleviate the impact of potential childhood trauma (Park & Katsiaficas, 2018;Vandekerckhove & Aarssen, 2020). However, some studies point out that more training and professional support for kindergartens, to help increase staff potential to identify and respond to the early signs of trauma in children, is needed (Lamb, 2020;Park & Katsiaficas, 2018). This is the case in Norway. For instance, a study in Norway on parents with children living in asylum centres preferred to enrol their children in kindergartens to help provide them with opportunities for interaction with other children, and to get a break from the monotonous life in the asylum centres (Lauritzen & Sivertsen, 2012).
In my research on how Norwegian kindergarten teachers work with the integration of refugee children, the term safety was uttered consistently by participants as they spoke about their everyday work for refugee children. In Norwegian, the concept of safety is referred to as trygghet. The participants proactively used this term without being prompted. The kindergarten teachers emphasized that the integration of refugee children involves ensuring that the children 'feel safe' by providing emotional comfort and calmness to the children. Talking, comforting, hugging and holding children are synonymous with what I refer to as 'safety work'. The focus is on emotional stability and other kinds of protection, but not necessarily overlapping with the English notions of security (Gullestad, 1997).
Arguably, the notion of trygghet is linked with the Norwegian kindergarten's long tradition of creating a home-like environment, emphasizing intimacy, warmth and safety (Korsvold, 1998;Gullestad, 1997). This is anchored in the Norwegian Kindergarten Framework Plan (2017, p. 20), which outlines that kindergarten staff shall 'ensure that all children find safety, belongingness and well-being in kindergarten'. Trygghet, thus, emerges as an important ideal in the everyday life of children in the kindergartens including refugees. In this article, I foreground the concept of safety as a core part of educational and professional training for kindergarten teachers in their work with refugee children.
This article discusses how the notion of safety organizes daily work in the kindergarten as mediated by knowledge acquired from the International Child Development Programme (ICDP) 1 and the Resource Centre for Violence and Trauma (RVTS), 2 which the participants had attended. This study aims at contributing to debates on trauma and the power of categorization in the kindergarten that emerges from well-intended programmes such as the ICDP and RVTS. This investigation scrutinizes the relationship between assumed trauma among refugee children and its role in the understanding of the safety discourse, as that informs kindergarten teachers' work with children.
In this light, this paper seeks to answer the following research questions: (1) How does the notion of doing safety work emerge as a coordinator of everyday social relations among kindergarten teachers?, and (2) How does the discourse of trauma bring about the categorization of refugee children by kindergarten teachers?

ICDP and RVTS
The International Child Development Programme (ICDP) was founded in 1992, and has its roots in developmental and cultural psychology, which aims at strengthening the conditions for the upbringing of children through the supervision of their caregivers (Bråten & Sønsterudbråten, 2017). Moreover, it borrows ideas from attachment theory, in particular the emphasis on the need for a safe emotional base for a child as they develop (Bowlby, 1969), by reactivating existing positive patterns of care. The overarching understanding of the safety discourse is that a child needs a safe emotional base (Bowlby, 1969).
The ICDP is used for training parents and professionals on caring for children up to the age of six within contexts such as kindergartens and schools, not only to help improve the interaction between staff and children, but to also curb neglect and abuse, and attend to vulnerable children in childcare settings that deal with abuse, 1 https://www.icdp.info/about/training/ 2 https://rvtsmidt.no/kompetanseheving/utdanningsprogrammer/traumebevisst-omsorg/#del_0 trauma and abandonment (The Directorate for Children, Youth and Families, 2016, p. 12). According to the ICDP guidelines: The teacher should try to create a trustful and intimate atmosphere by showing emotional warmth, giving praise and confirmation to each child, and as far as possible, within the scope and limits of the classroom situation, try to respond to the children by establishing dialogues and activities along the lines of their initiatives and interests, and by giving them praise for what they have done well. 3 The Resource Centre for Violence and Trauma (RVTS) offers courses, programmes and information geared towards trauma-conscious care for people working with children and adults, especially those who are likely to have experienced trauma. 4 The main purpose of RVTS is not to offer treatment, but contribute to the professional development of services within refugee health, forced migration, psychological traumas and psychosocial preparedness. 5 Like ICDP, RVTS is rooted in psychological disciplines and shares similar ambitions, especially concerning the training of professionals working with the reception of refugees in Norway.

Previous research and theoretical framework
The discourse of how refugees are seen and treated is widely observable, and connected with the discourse of mental health interventions within psychiatry and psychology, the so-called 'psy' discourse (Bjerre et al., 2021;Lunneblad, 2017), and less so from social science disciplines such as sociology. This is reflected in studies on refugee children, in which the dominance of knowledge from the psy-disciplines on the mental health and well-being of refugees contributed to a significant focus being placed on traumatic experiences (Lunneblad, 2017). In the Nordic and other Western countries, the general refugee population has become identified with the dominant discourse of portraying refugees as victims of war, traumatized, suffering and in need of care (Eastmond, 2014).
A study on Bosnian refugees in Sweden indicates that the 'traumatized' refugee as a category became an object of interest used to mobilize for acceptance, protection and in political debates, in which lobbying for funding for refugee mental health and care especially by professionals working with refugees in different welfare 3 https://www.icdp.info/wp-content/uploads/2020/12/Essay-on-ICDP-and-education.pdf 4 https://rvtssor.no/aktuelt/294/de-tre-pilarene-i-traumebevisst-omsorg/ 5 https://rvtssor.no/dette-er-oss/about-rvts-sor/ programmes (Eastmond, 2014). Seemingly, trauma awareness has come to permeate the reception of refugees across diverse welfare institutions that help to facilitate the integration of refugees.
Similarly, the trauma discourse (Rutter, 2006) is gaining increasing attention, particularly in the strategies various professionals employ to establish close relations with children. In a study on the reception of refugee children in Swedish kindergartens, Lunneblad (2017) asserted that the children's vulnerability and need for safety were dominant among the teachers when they talked about the children, especially the emphasis on the image of the refugee child as traumatized. In this light, the integration of a refugee child at the kindergarten is premised on the idea that the teacher will create a close relationship with the child. Kindergarten teachers, hereby conceptualized as 'street-level bureaucrats' (Lipsky, 1980), are responsible for putting policy into practice by delivering care and safety to refugee children on behalf of the state. In so doing, they activate the concepts, categories and discourses embedded in research and policy in their daily work.
The trauma discourse needs to be understood not in isolation, but also in considering the significant influence of attachment theory (Bowlby, 1969), as well as its linkage with neuroscience and psychology in general (Tobin, 2016). Neuroscience studies reveal that a traumatic childhood experience is linked with a failure in optimal brain development, which, in turn, has potential negative consequences for the holistic development of the child (Tobin, 2016). Recent developments reveal that the common treatment methods for trauma are Trauma Focused Cognitive Behaviour Therapy (TFCBT), Eye Movement Desensitization and Reprocessing (EMDR) (Bisson et al., 2013), Sensorimotor Psychotherapy (Ogden & Minton, 2000) and Narrative Exposure Therapy (NET), to mention a few trauma-specific methods (Lie et al., 2014). Launched in 2013, trauma care and support for refugees are part of Norway's national strategy for the health of immigrants (2013-2017), 6 which recognizes trauma care as a public care responsibility.
It is a common perception that when a child's secure attachment is disrupted, the potential for more profound developmental consequences exists, particularly in future learning outcomes (De Bellis, 2001). Nevertheless, neuroscience studies have shied away from conclusively establishing a direct relationship between childhood trauma and developmental outcomes because it is difficult to distinctively separate the causes and effects (Veltman & Browne, 2001). Neuroscience has therefore advanced more focus on attachment theory, particularly the link between emotional experiences with caregivers and later neurobiological, emotional and cognitive functioning (Tobin, 2016).
However, attachment theory is criticized for being normative and narrow due to its focus on only a small set of interactions the child experiences early in life. Moreover, it is claimed to be totalizing with little room for alternate views of relations, while at the same time promoting a diagnostic mind-set among street-level bureaucrats such as kindergarten teachers (White et al., 2019). Psychological knowledge, such as attachment theory, has contributed to the formation of government policies and practices in which some experts gain authority over laypeople, and their authority supports preferred ways of what it entails to be a human being (Bjerre et al., 2021).
Such policies and practices shape and define how refugees should be viewed and treated. Moreover, the increasing demand for documentation and accountability makes the use of 'psy' knowledge powerful, and legitimizes the use of underlying concepts, such as trauma. Attachment theory operates as a powerful 'psy' discourse in a way that reflects prevailing social, cultural and political beliefs (Keddell, 2017).
When refugee children are viewed as vulnerable, traumatized and in need of safety, a generalized 'adult' understanding of children and childhood is likely to influence the professionals who work with them (Warming, 2011). Rose (1999) argued that certain knowledge regimes, arguably emanating from developmental psychology and paediatrics, play a key role in constructing 'governable subjects-. Finding inspiration in the works of Foucault, Rose argued that the 'psy-disciplines' have had a profound impact on how we understand and categorize people, including ourselves.
These categories do not represent individuals or groups, but ideas about them according to Canadian Sociologist Ian Hacking's (1999) work on the construction of people. The ideas in this case 'functions within a matrix of discursive elements that are part of an interplay between different processes, institutions, people and technology' (Hacking, 1999 p. 24). When categories are used to refer to people, both the category and the matrix within which it is part of becomes visible. Ideas from Hacking are important in this study, which seeks to challenge the hegemony of psydiscourses, particularly within professions working with children by questioning how categories are made to fit people, which in turn legitimizes the dominant knowledge regimes in use.

Analytical approach
My study is informed by institutional ethnography (IE), widely associated with Canadian sociologist Dorothy Smith (2005). Institutional ethnography has been growing within sociological studies because of the methodological commitment to go beyond the individual experience or narrative, i.e., the local, into the institutional complex, i.e., the trans-local (Walby, 2013). In this light, the individual experience and activities serve as the point of entry into the investigation of the institutional processes in which the kindergarten teachers partake (DeVault, 2021). This means beginning a research inquiry from what kindergarten teachers know, and moving 'further' to find out how what they are doing is connected with others' doings in ways 'they cannot see' (Kearney, 2019).
In IE, this means 'keeping the institution in view' by exploring texts such as the ICDP, the RVTS and the Framework Plan for Kindergartens, which mediate between the everyday experiences in the kindergartens, and how integration work is organized and coordinated (McCoy, 2006). In this instance, the integration of refugee children is an institutional process that is part of ruling relations. Ruling relations hereby refer to the social relations that organize work outside of what is going on in a specific scenario (Smith, 2005). Therefore, doing an IE is about tracing the ruling relations, as embedded in the descriptions of everyday work by those who experience it from their standpoint, to reveal what is concealed as seen or experienced from the knower's location (DeVault, 2021).
In my study on the role of Norwegian kindergartens as arenas for the integration of refugees, I conducted research beginning with the standpoint of the kindergarten staff. The ambition was to explore the ruling relations that organized their knowledge of doing the integration of refugee children. By taking the standpoint of kindergarten teachers, I position them as the knowers of integration work. The kindergarten is the immediate site of experience and activities, which embody integration from the standpoint of the teachers. The analytical attention that this article focuses on is on how and why kindergarten teachers engage in promoting safety for refugee children as a core element of integration work.

Data collection
The findings and discussion in this paper are based on interviews with 13 kindergarten teachers who worked in three kindergartens in southern Norway, which were carried out between April and September 2019. The kindergartens sampled included a reception kindergarten exclusively for refugees and two general kindergartens: one public and one private. The kindergartens rely on funding from the state, and are accountable to the Norwegian national framework plan for kindergartens. The overarching consideration for selection was that participants worked in kindergartens that received refugee children.
Participants included pedagogical leaders, teacher assistants and language and diversity support teachers. Diversity support teachers are employed by the municipality to routinely visit kindergartens to offer consultative and practical assistance, predominantly to minority children. Ethical considerations made before, during and after the study were in tandem with the guidelines from the Norwegian Ethical Committee on Social Science Research (NESH, 2006). The project was approved by the Norwegian Centre for Research Data (NSD) to help ascertain that data protection procedures were followed. To comply with the guidelines, comprehensive information was provided via email and orally, with consent forms signed by each participant at the beginning of their interview. All kindergartens and participants remain anonymized, and the information collected remains confidential and was only utilized for my project (Israel & Hay, 2006). I used a non-standardized snowball and purposive sampling technique to recruit participants, through which I gained access to additional participants through networks of those already recruited.
The criterion was that the participant worked in a kindergarten with refugee children.
All participants were female, which reflects the general population within kindergartens and childcare centres in Norway.
All interviews were semi-structured and took between 45 to 60 minutes on average.
Most were conducted within kindergarten premises, as only three were held on the university premises. The interviews were audio recorded, and then transcribed by a research assistant. The same assistant translated the interviews conducted in Norwegian into English. Since I am not a native Norwegian, I was flexible with my choice of language during the fieldwork. Six interviews were conducted in Norwegian, while the rest were done in English. The decision on which language to use depended on the choice and convenience of individual participants. For the interviews conducted in Norwegian, a native research assistant joined me as a translator and an intermediary in the research process. The native research assistant was initially meant to assume an invisible role in the research process, but that changed as it became increasingly important that the translator served as a cultural decoder during the interviews (Munday, 2008).
Reflecting on the lack of local language competency meant that I was not accustomed to the institutionalized discourses that an 'outsider' (Wolf, 1996) can perceive as taken-for-granted representations (Lund, 2015). Being an outsider provided me with a chance to go with the tag of the 'ignorant' researcher, while the participants were placed in a relatively empowering expert position (Lund, 2015).

Data analysis
The data in this study were analysed through the transcription and translation of interviews, followed by the in-depth task of describing typical accounts of safety work as described by the kindergarten teachers. IE researchers usually 'interview' the interview material (McCoy, 2006) to learn about the individual work experience of the participants. I searched for the transcribed material to purposively identify clues on how kindergarten teachers refer to ICDP and RVTS to do safety work. The effort was to look for detailed descriptions of safety and circumstances surrounding safety, as narrated by the participants in the transcriptions (McCoy, 2006: 111).
The interview excerpts were therefore handpicked because they show how kindergarten teachers are connected to an institutional way of knowing and doing safety work using ICDP and RVTS as forms of social technology. Used in a social scientific sense, the term 'social technology' refers to the application of methods and theories, such as assessment manuals and training programmes, to obtain a science-based analysis for specific purposes (Leibetseder, 2011;Kimathi & Nilsen, 2021). Social technology hereby offers expert knowledge with which professionals like kindergarten teachers define and reach solutions to everyday social problems in their line of work.
The analysis therefore takes on a descriptive perspective. It is through these descriptions that it is possible to uncover the social relations that influence safety work practices in kindergartens, as well as how those experiences become part of a wider complex of institutions (Devault & McCoy, 2006;Smith, 2005). My interest was in using the interviews to unpack the concepts, categories, and wider institutional discourses that kindergarten workers have been socialized and trained to use, as they further reveal how the workers become accustomed to professional ruling relations (Nilsen, 2021).
In addition to tracing and identifying the social processes that connect the safety work across the participant interviews, I followed up with a back-and-forth exploration of the described work considering the dominant texts, and mapped the connections between them. These texts included the Kindergarten Framework Plan, and training programmes like the ICDP and RVTS, as cited by the participants. The importance of texts was to locate how they mediate and govern the processes in routine safety work, as the activities of the kindergarten teachers are coordinated through such objectivized systems of knowledge like text documents, laws and discourses (Smith, 2005). In the final step, I used the transcripts to determine how the refugee children were categorized and conceptualized in the kindergartens, and how existing texts facilitated the coordination of how refugee children were categorized.
The findings from this study cannot be generalized, and the research sample is not representative of all kindergartens and their work with refugee children. While I am familiar with the Norwegian kindergarten policy, I have no first-hand work experience in the kindergarten, and therefore cannot relate to the institutional discourses in the field. However, from an institutional ethnographic perspective, this is an ideal situation for researching to avoid professional jargon (Nilsen, 2021). In the presentation of the empirical results, I describe how the safety discourse emerges in the everyday work of the kindergarten teachers, first focusing on how refugee children are understood, and then on how the kindergarten teachers work to provide safety. Next, I turn to the ICDP and the RVTS programme that the kindergarten teachers consistently cited, tracing how their daily work is connected to a textually mediated institutional discourse on safety.

Tracing the safety discourse in kindergarten teachers' everyday work
This project set out to discover how kindergarten teachers work with refugee children.
I asked the teacher participants to describe their everyday work, but I did not ask questions specifically relating to safety during the interviews. However, my attention was drawn to the narrative of safety, which appeared as a common theme, and decided to explore where this discourse emanated from, and how it is activated in kindergarten teachers. The use of the notion of safety, which is a contextual translation of trygghet in Norwegian, is used to refer to emotional calmness and comfort for refugee children, unlike the commonly acknowledged understanding of safety as an avoidance of harm and risk.
During the interviews, the kindergarten teachers referred to both the ICDP and the RVTS programme that they had attended. Involvement in both appeared crucial to the teachers' understanding of their work with refugee children who were perceived to be potentially traumatized. Olivia, a participant in the research, related her work experiences as shaped by ICDP and RVTS: Olivia: We had a child who came from [a country in the Middle East] two months ago. We were sitting with the parents talking about what we are doing here in the kindergarten. I have attended RVTS courses and other courses related to trauma. I also attended ICDP training. I now understand that a feeling of safety is important when I am working with younger refugee children. We tell the parents to be there for as long as the children need them to be. One 1-year-old has been lying here, next to the mother's chest during the flight, feeling the heartbeat. There can be a lot of stress for the child, so it can be a hard transition. It is important to take the children and their needs seriously.

Interviewer:
When the parents bring their children here, what do you think is their impression of your work with the children?
Olivia: They feel a need for safety and that we meet their child's needs. Many of them are new to kindergartens, and so they do not know what to expect, so they feel insecure, especially when the child is only one year old. Understandably, it is hard for parents to leave their children here. It is not easy to express their needs either because everything is new. We are always working with making them feel safe -we are filming [the children], taking pictures and sending them to the parents, writing messages. Especially here we know much of what they are going through as refugees, and we have a big focus on making them feel safe.
In this quote, Olivia describes the situation and needs of refugee children, asserting that the children and their parents need to 'feel safe', especially for those who have recently arrived in Norway. Olivia does not refer to any specific individuals but, rather, to a socially constructed idea of refugees as a category. Olivia elucidates how training from the ICDP and RVTS is significant to her understanding of the importance of safety work in kindergartens, and it is evident that her work knowledge is informed by these programmes. Olivia's reference to the ICDP and RVTS helps account for her work knowledge, and the need for interventional support to provide safety in the kindergarten. The need for safety applies to parents and children, respectively.
Nancy and Nora are kindergarten teachers and participants in this study. During the interviews, they both discussed why the ICDP is important to their work. They highlighted that the potential trauma that refugees may have faced creates an inherent need for trauma-conscious training. In the following, Nancy and Nora explain why the ICDP and RVTS are important to them. We have a big focus on that here, so I have been to Oslo for RVTS training. I also use a lot of personal experience. I grew up with an English mother here in Norway, and she struggled, so I felt different. We all want the same thing, which is to feel safe.
While Nancy and Nora referred to the ICDP and RVTS, they emphasized the significance of bringing an awareness about trauma and children's mental health.
Nancy, in particular, noted that her background in psychiatry gave her the privileged knowledge that she needed to work with refugee children, and cites experiences with refugees who were perceived to have trauma. Joana and Ruth have no ICDP training, but equally highlighted the focus on trauma in their daily work.
Joana: In kindergartens, the biggest challenges are finding ways to make children feel safe, how to welcome children with trauma, and seeing them as a resource and not a burden to the kindergarten.

Ruth:
Working with colleagues, we are serious about the refugee children and safety.
Yeah, more about their feelings and how to make children feel safe. Some of the children have special traumatic experiences, so it's a very important part of our work, and how we relate with them.
The participants refer to the need for safety work, and why awareness about it is crucial. Indeed, these quotes reveal how the trauma discourse has become an integral part of the daily work life in the kindergartens. It is quite visible that their participation in ICDP and RVTS programmes may have contributed to this understanding of what to focus on as part of integrating refugee children. This awareness about safety work is indeed crucial. The essence here is not to dismiss the importance of safety, but to highlight covert discourses that become accessible from professional language and challenge some of their implications. In so doing, we can connect the descriptions of the safety work, to the trans-local institutional discourses upon which the work descriptions are founded.

Doing safety work
The interviews indicate that as a social category, refugee children are associated with trauma and, hence, are perceived to be in critical need of safety. In explaining how they handle refugee children in the kindergarten, participants commented on the importance of sensitivity, calmness and providing comfort through physical touch and hugging, as exemplified in the quotes below. Put differently, kindergarten teachers perceive this form of interaction as key to connecting to-and promoting emotional safety with refugee children.
Ruth: I use a calm voice, and I say, like, comforting words because when you say comforting words, you automatically use a way of speaking that is calm, and I very much use physical contact, carrying them tight. I am usually singing, [having] eye contact, and doing things together, and a lot of talking really -even if they do not understand -but talking is a kind of therapy, and that is a way to make them feel safe.
Sophie: I must sit down, be quiet, be calm... ..but I must watch them if they need something. Maybe one of the children is alone, so I need to maybe go to the child and sit down and try to find out how the child is. Is he/she sad, or maybe he wants to play alone? We don't know. We must be at the level where the children are. And of course, we have activities like painting, beading. They are fond of that at this time According to the participants, a kindergarten teacher equipped to do safety work must be emotionally sensitive, offer comfort, use physical touch (hugging) and partake in play with children. These behaviours can be traced back to the ICDP training that focuses on emotional dialogue, showing loving feelings and praising and acknowledging the child. When the teachers spoke about emotional dialogue, their dialogue resonated with the language used by the ICDP, which is concerned with early emotional-expressive communication between the caregiver and child.
The empirical data in this study emphasize that teachers feel responsible for collaborating with the children, and finding what is best for them as part of making them feel 'safe', as the quote from Ruth illustrates: Ruth: In the kindergarten where I worked first, we thought of integration as very exciting, and were aware of how we should work with refugees, like ….how we could make them feel safe through collaboration with both the children and parents. We also focused on trauma because that's important, if the children aren't feeling safe, they need support to cope with their strong emotions. I often see the children who are struggling and think, how is this going to be in school when there is so much anger and aggressive behaviour?
The findings in this study reveal that the kindergarten teachers perceived and assumed that the children and their parents were traumatized, e.g., the quotes from Nancy highlighting that refugee children 'had experienced traumatic events', or 'had been abused'. Similarly, Ruth asserts the 'need to focus on trauma if the children are feeling unsafe', and further claims that some children 'have experienced special traumatic events'. According to the findings, the image that is portrayed in the professional language is that of 'traumatized' refugees in need of safety.

Discussion
The textual mediation of the safety discourse In my findings, the centrality of the discourses on trauma and safety visibly illustrated the unseen complex experiences kindergarten teachers engage in when working with refugee children. Through the interpretation of authoritative institutional texts, such as the ICDP and RVTS training programmes, safety work is textually mediated, which, as the findings indicate, the kindergarten teachers were not necessarily aware of, nor had they paid attention to the ruling relations of safety work. However, most had attended ICDP training to improve their competency to work with refugee children.
The construction of the 'traumatized' refugee child in need of safety as a social category can be linked to the discourses of the ICDP and the RVTS programmes, which advance that refugee children have potential trauma, and that teachers therefore ought to intervene through the provision of safety work. The category is care (Hundeide, 2010). While the ICDP and RVTS programme appears to have improved teachers' competency in their work with refugee children, and the refugee parent's awareness of their parenting, they are also examples of how psychological discourses continue to shape the construction of the image of refugees who are treated as a category. Categories are socially constructed and represent ideas, and not individuals or species (Hacking, 1999). The use of categories exemplifies how professional language can be utilized generically, despite the ICDP's insistence on cultural sensitivity, hence producing a standardized way of seeing and interacting with the children (Kimathi & Nilsen, 2021). Anchored on psy-discourse, the concept of safety can therefore be seen as a token of accountability, revealing a scientific power that goes uninterrogated at times (Nilsen, 2021).

Implications of the trauma discourse
My findings indicate that the kindergarten teachers were immersed in a safety discourse as a form of intervention against potential trauma during the integration of refugee children. Previous studies reveal that research on refugee children, as well as practitioners' perspectives, have been dominated by a trauma discourse (Lunneblad, 2017;Rutter, 2006;Watters, 2011) that has been criticized for homogenizing refugee children as weak and vulnerable. Seeing children as traumatized is part of a wider narrative through which refugees are seen in terms of presenting 'problems' (Rutter, 2006), rather than for the gifts and human potential inherent in them. The professional training of staff on trauma and trauma intervention is central to their understanding of what they should prioritize.
Kindergarten teachers are therefore becoming increasingly more accountable to 'psy knowledge' discourses (Rose, 1999) that mediate the social relations of their typical safety work. The dominance of 'psy knowledge', and the widespread unquestioned acceptance into professions such as childcare and kindergarten education, was evident in my study. In this light, the discourse of trauma has increasingly become an ideology that is shaping policy and practice within kindergartens in Norway, and as such, it is shaping the prevailing ideas and construction of refugee children and their needs.
In addition, the discourse of trauma connects different kindergarten professionals in an institutional complex, in which the construct of the 'traumatized' refugee child is made functional for the professionals to make sense of their everyday work. This aligns with previous studies that reveal that vulnerability, deprivation and traumatic experiences have been at the centre of recent research on refugee children, mainstreamed in psychology and psychiatry (Watters, 2011;Lunneblad, 2017). The implication is that if left unquestioned, psy knowledge can become a powerful knowledge regime (Rose, 1999) that influence how professionals understand and categorize refugee children.
This categorization serves as the basis for the implicit standardization of the work kindergarten teachers do (Downey, 2007;Lawrence et al., 2019). Social technologies involve a broad categorization of children, in which the professionals rely on standardized descriptions embedded in the technologies. Empirical observations are combined with discursive concepts to assign specific 'pathologies' to the children. For example, a refugee child's crying is interpreted as emotional stress, and poor clothing is interpreted as poor parenting.
Despite the failure of existing studies to conclusively show a direct relationship between trauma and developmental outcomes among children (Veltman & Browne, 2001), the assumed relationship between safety work and the trauma discourse in kindergartens must be scrutinized. Arguably, trauma-based education for kindergarten staff is important in the sense that children who may have experienced trauma are handled appropriately from an informed point of view. However, it is also equally important to consider that refugee children can feel unsafe without carrying trauma, i.e., not all refugee children who are unsafe have been traumatized.
The application of knowledge designed for trauma intervention not only assigns refugees a traumatized and vulnerable category, but also risks victimizing children with a refugee background, who, like other children, have agency and are resourceful in their routine encounters in and out of kindergartens. The psy discourse that is visible in social technologies such as the ICDP and RVTS is increasingly becoming powerful and legitimizing professional practice, but when used in a fragmented manner it can legitimize, without necessarily solving apparent problems or creating others such as being normative and moralizing (Bjerre et al., 2021: 10). Through safety work, kindergarten teachers are fulfilling the ambition of the authorities (through policy) to ensure that resources and training provided by the state and other interrelated agencies are utilized to help produce the desired outcomes, all with good intentions. As such, they are meeting the objectives of the institutional framework that organizes the processes within kindergartens, such as the Kindergarten Framework Plan and programmes like the ICDP and RVTS. Policymakers, kindergarten teachers and relevant agencies would thus find it meaningful to revisit the theoretical and categorical presumptions that inform classification systems, since they are founded on a theoretical knowledge that is rarely interrogated and implicitly embedded in discourses and texts (Bjerre et al., 2021).
Indeed, uncovering what knowledge has been privileged and how such knowledge becomes authoritative and legitimized, even when not well understood by those who embed it in their practice, would be beneficial. The implication is that in their ambition to promote integration in kindergartens, teachers should use social technology in a critical way that incorporates reflexivity rather than as a standard routine.
Kindergartens can create experiences that offer warmth and affection for refugee children, among others, to help them feel safe and develop a sense of belonging without necessarily putting labels on them (Kalkman et al., 2017).
My findings indicate that when children are seen as traumatized, kindergarten teachers may interact with them as representative of a category, thereby overlooking their individuality. In this case, the individuality is lost not because the children are not catered for individually, but rather because the discourse that emerges out of the work of kindergarten professionals advances the category of the 'traumatized' refugee children. Consequently, this becomes the label that acts as an umbrella for refugee children. The individuality therefore becomes lost in the discourse and not essentially in practice, since my empirical data showed that kindergarten teachers made discretionary judgments that catered to children's individual needs for care and safety.
Each category carries with it a preconceived meaning (Nilsen, 2017). When the category is used in an institutional setting, it becomes the foundation upon which certain practices and outcomes are justified. As a result, the category of the 'traumatized' refugee children serves as the basis upon which kindergarten teachers engage in safety work. The category here becomes significant because it calls for teachers to be accountable, but it can also be used normatively to reveal deviant behaviour among children, e.g., looking unsafe, thereby othering them. The use of social technologies by front-line professionals can consequently end up constructing the categories they are intended to 'help'.

Concluding remarks
Using institutional ethnography, I have explored how kindergarten teachers relate to the concept of safety in the integration of refugee children. The concept of safety in this paper is represented as both a discourse and a practice framed within an institutional complex that includes kindergarten teachers. The article's contribution is to reveal how safety work as a practice is implicitly mediated by authoritative knowledge regimes embedded in texts such as the ICDP, the RVTS programme, and the Kindergarten Framework Plan.
The article also examines how the discourse of trauma is increasingly becoming a dominant ideology to which kindergarten teachers subscribe, and in turn, it has resulted in a categorization of refugee children as potentially traumatized. The construct emerging here, 'traumatized' refugee child in need of safety, is therefore a result of well-intended intervention efforts in the kindergartens. Kindergarten practitioners and policymakers must investigate in more depth the effects of knowledge that, on the one side produces desired outcomes, but on the other, may counterproductively victimize refugee children.

Acknowledgments
This article was supported by the University of Agder, Norway. I would like to thank Professor Ann Christin Nilsen and Professor Ove Skarpenes for their valuable input and assistance in the preparation of this article. I am equally grateful to the kindergarten teachers who participated in the fieldwork during this study.

Conflict of interest
The author declares that there is no conflict of interest.