Managing categories: The role of social technology in kindergarten teachers’ work to promote early intervention and integration

Early intervention and integration are highly valued ideals in kindergartens in Norway. Building on two research projects informed by institutional ethnography, the authors address how kindergarten teachers ‘do’ early intervention and integration in their everyday work. They argue that this work largely revolves around managing categories, whether making categories fit people or making people fit categories. In this work, the kindergarten teachers rely on social technology that is influenced by a ‘psy-discourse’. Despite good intentions, the social technology and the professionals’ use of it ends up constructing the categories they are intended to help or ‘heal’.


Introduction
Kindergartens shall promote respect for human dignity by highlighting, valuing and promoting diversity and mutual respect.The children shall be able to discover that there are many ways in which to think, act and live.… Kindergartens shall help ensure that all children feel they are being seen and acknowledged for whom they are and highlight the place and value of each one of them within the group.(Norwegian Directorate for Education and Training, 2017: 8-9) The above text is taken from a section describing the core values of Norwegian kindergartens as set out in the 'Framework Plan for the Content and Tasks of Kindergartens' (Norwegian Directorate for Education and Training, 2017).The Framework Plan is an authoritative document outlining the values, responsibilities, objectives, working methods and learning areas with which all kindergartens in Norway must comply.Our objective in this article is to explore how kindergarten staff work to realize these ideals.
In Norway, more than 97% of all children aged three to five attend kindergarten, while the enrolment rate among children aged one to two is almost 85%.Compared to the other Nordic countries, the expenditure on kindergartens in Norway is high, at almost 2% of gross domestic product in 2015 (Nygård et al., 2019).In 2017, more than 17% of the children in kindergartens had a minoritylanguage background, defined as children with a mother tongue other than Norwegian, Sami, Danish, Swedish or English (Såheim Bjørkli, 2018).Hence, it is fair to say that kindergartens have become an important arena for social cohesion and integration in the Norwegian welfare state (see also Kuusisto and Garvis, 2020).
The ambitions to which kindergartens should aspire are manifold.They 'shall promote democracy, diversity and mutual respect, equality, sustainable development, life skills and good health' and meet 'every child's need for care, security, belongingness and respect … enabling the children to participate in and contribute to the community' (Norwegian Directorate for Education and Training, 2017: 7).Kindergarten staff are required to adapt their general pedagogical practices to suit the children's needs and circumstances, including children who may require additional support for shorter or longer periods.… For some children, early intervention could mean that staff have to work methodically and systematicallyover shorter or longer periodsto include these children in meaningful social relationships.(Norwegian Directorate for Education and Training, 2017: 40) In this article, we relate these ambitions to the concepts of 'early intervention' and 'integration', ideals that are valued highly both in policy pertaining to early childhood, as exemplified in the extract above, and in welfare policy more broadly, as expressed in different White Papers and public reports.For instance, 'early intervention' is a key concept not only in White Papers and official reports within the education and child protection sectors (e.g.Norwegian Ministry of Children and Families, 2016; Norwegian Ministry of Education andResearch, 2019-2020), but also within the financial sector (Norwegian Ministry of Finance, 2009).'Integration' is the stated goal of Norwegian immigration policy (Norwegian Ministry of Children, Equality andSocial Inclusion, 2012-2013).These concepts undoubtedly stem from good intentions.Indeed, as ideals, they are 'immune to critique' (Pettersvold and Østrem, 2012).Our aim in this article is thus not to question the ideals per se, but to explore how these ideals, as they are transformed into practice, may have unintended consequences.The question of interest, therefore, is how kindergarten staff 'do' early intervention and integration in their everyday work.
Drawing on two research projects on, respectively, early intervention (Nilsen, 2017b) and integration (Kimathi, in progress) in kindergartens, we argue that professional work to promote these ideals revolves predominantly around categorization.In this work, the kindergarten teachers rely on social technology that is influenced by a 'psy-discourse'the influential body of knowledge produced by the disciplines of psychology, psychoanalysis, psychiatry and psychotherapy (Rose, 1999).We use the term 'social technology' in a social scientific sense to refer to the application of methods and theories, such as assessment manuals and training programmes, to obtain a sciencebased analysis for specific purposes (Leibetseder, 2011).Social technology provides expert knowledge with which to define and reach solutions to social problems on the front line.We argue that the psy-discourse, mediated by social technology, may at times be at odds with the policy objectives that justify its usage.Paradoxically, therefore, with the intention of doing good and helping children and families in need, kindergarten teachers simultaneously construct the categories of people they are intending to help or 'heal'.Analytically, our discussion is informed by institutional ethnography, with theoretical inspiration from Ian Hacking's (2004) theory of how people are 'made up'.

An analytical framework for tracing categories
Our analysis resembles a 'keywords approach' (Hennum and Aamodt, 2021).Hennum and Aamodt (2021: 207) claim that keywords have two main characteristics: they can incorporate multiple meanings, rendering them 'slippery' in usage, and they are usually overwhelmingly positive and impossible to be 'against'.Keywords constitute and delimit a field, reproducing 'shared matters of concern' (Hennum and Aamodt, 2021: 208).We start our inquiry with a notion of these concepts as sensitizing (Blumer, 1954).When we, as professionals, scholars or practitioners, encounter 'early intervention' or 'integration' in a statement, we can, based on our experiential or theoretical knowledge and our familiarity with institutional discourse, deduce an interpretation of the concepts, despite their lack of agency and action (indicating who does what to/with whom).They sensitize us to a certain way of thinking and a presumptive course of events.According to Smith (2005: 111), institutional discourses, to which professionals are held accountable, are dominated by nominalization, thus dissolving the intimacy between word and experience.The words and phrases that are used therefore function as 'shells', in themselves devoid of empirical substance and agency.In contrast to definitive concepts, denoting specific actions, people, places and so on, sensitizing concepts, or 'shells', lack empirical content, although they may well be linked to certain actions or occurrences that can be explored empirically.Thus, sensitizing concepts provide an interesting entry point for empirical analysis.
Both our research projects were informed by institutional ethnography.Associated primarily with the Canadian sociologist Dorothy E Smith (2005), institutional ethnography is a method of inquiry and a research orientation dedicated to exploring the ruling relations of people's ordinary everyday lives.Ruling relations refer to objectified forms of knowledge that are abstracted from people's lives yet still inform how people think and act locally.Predominantly mediated by texts, whether authoritative, such as White Papers, frameworks and guidelines, or 'functional', such as signs, forms and manuals, ruling relations denote the concerted coordination of consciousnesses that link people across time and space.People working in kindergartens take part in ruling relations when, for instance, they invent rhymes, go for excursions or assess children's abilities.In different and often concealed ways, they relate to the overall objectives of the kindergarten, as manifested in a framework plan or White Paper, and these objectives are intricately linked to international laws and standards as set out in transnational organizations such as the Organisation for Economic Co-operation and Development and the United Nations.While the individual kindergarten teacher is unlikely to be aware of how they relate to these actors in their everyday work, they know this vast institutional complex expertly from their own standpoint.In institutional ethnography, the empirical quest is to explore the institutional complex from the standpoint of people who are situated somewhere within it.It is not the institutional complex per se that is the subject of research but the ruling relations within it that inform the everyday experience.In other words, the quest is to explore how what is being done here and now is linked to actions of other people situated in another time and place in the institutional complex of which the experience is part.The starting point of the inquiry is thus the standpoint informants' work and knowledge about their work.
In both our projects, we started the inquiry from the standpoint of kindergarten staff.Nilsen conducted interviews with 14 kindergarten teachers and assistants, asking them about their work when 'doing' early intervention.They were asked, for instance, what they looked for when they assessed children, what kinds of activities they did and what kinds of resources they used.Nilsen followed their everyday work filling in forms, attending meetings and taking part in training.She conducted interviews with people with whom the standpoint informants cooperated, such as nurses, child welfare workers and teachers.She discovered that doing early intervention in practice revolved around identifying children who, in different ways, did not conform to a certain standard, and subsequently finding out how to deal with them.These were children who were causing concern; thus, working with them may be called 'concern work' (Nilsen, 2021a).The staff used different tools for assessment.Depending on which tools they used and the outcome of the assessment, children could be categorized as being, for instance, 'at risk' or 'developmentally delayed'.The assessment was essential to justify the need for intervention.
Kimathi conducted semi-structured interviews with 13 kindergarten teachers and assistants on how they experienced the everyday integration of refugee children.The questions that were asked during the interviews included what activities were prioritized when handling the children, how the staff interacted with refugee children and their parents, and how they related to texts such as the kindergarten framework plan and other policy documents.Kimathi discovered that the integration of refugee children to a large extent revolved around children's competency with the Norwegian language and how well they absorbed dominant values, such as play, independence and respect for nature.The kindergarten staff frequently mentioned how important it was for the children to 'feel safe' (Kimathi, forthcoming).The staff appeared to be balancing the children's differences and parental expectations, on the one hand, and the institutionalized integration practices and ideals expected in Norwegian kindergartens, such as the ability to speak Norwegian, on the other.In particular, the teachers highlighted the need to do 'safety work'for instance, by making the children (and their parents) feel emotionally comfortable (Kimathi, forthcoming).This stemmed from the assumption that the children and their parents, as refugees, suffered from trauma.In providing safety, the informants relied on specific programmesnotably, the International Child Development Programme (ICDP)in which many of them had received training.
In Kimathi's and Nilsen's projects, the standpoint informants were recruited from, respectively, four and three public kindergartens in southern Norway.The informants had different educational backgrounds and positions in the kindergartens, but all were interacting directly with children in their everyday work.In Kimathi's project, one of the kindergartens was specifically designed for children from refugee or asylum-seeking families, while the other two were ordinary public kindergartens, as were those in Nilsen's project.While Nilsen conducted her interviews in Norwegian, Kimathi conducted his interviews in English, aided by a research assistant for interpretation when needed.Being non-native, he entered the research field with even more unfamiliarity.This offered the possibility to see and experience the fieldwork through an external lens.Neither of us has recent first-hand experience of working in kindergartens, but we know the policy basis of the Norwegian kindergarten system well.We acknowledge that our familiarity, or lack thereof, with the institutional discourses in the field may have an impact on our analysis.In both projects, however, the interviewing was explorative and consciously 'naive', intended to dodge professional jargon that captures predefined institutional understandings (Nilsen, 2021a).Both projects were conducted in line with the guiding norms for research ethics and were approved by the Norwegian Centre for Research Data.While Nilsen's research focused on how children who deviated from the norm in some way were sorted into manageable categories, Kimathi's research focused on how a specific categorythat of 'the traumatized refugee child'was invoked to attribute specific needs and abilities to people assigned to that category, thus rendering the categorized people manageable.The work of 'doing' early intervention and 'doing' integration is, in other words, all about managing categories, whether making categories fit people or making people fit categories.In this work, social technologies are abundant.In the next section, we introduce the theoretical perspectives that inform our analysis.Thereafter, we present excerpts and examples from the empirical data, demonstrating how social technology is part of chains of action.In the selection of excerpts, we searched the interview material purposively for clues on how kindergarten teachers use social technology in order to manage categories, inspired by McCoy's (2006: 111) recommendation to look for detailed descriptions of work that make visible institutional hooks and traces in the lived experiences of the teller.While the excerpts may not be representative of all encounters with social technology, they pinpoint tendencies that are found across cases.

How are categories constructed?
In his book Modernity and Ambivalence, Bauman (1991: 1) claims that classification is a matter of giving the world a structure and an order, where 'one knows how to go on'.Not knowing how to act creates ambivalence and discomfort, and represents a threat to the social order.The importance of classification in modern societies should not therefore be underestimated.As proponents of certain values and morality, and as the extended arm of the state (Lipsky, 1980;Zacka, 2017), professionals in modern societies play a key role in maintaining the social order and, in doing so, rely on systems of classification of people.
Categories are socially constructed.Hacking (1999) points out that categories, such as 'refugees' or 'children at risk', represent ideas and not individuals or species.When using such categories, one draws on an idea or notion of what they are about.The ideas do not exist in a vacuum, Hacking (1999: 24) asserts, but are part of and can only function in a matrix of discursive elements, formed in a complex interplay between institutions, procedures, stakeholders and media reports.When using different categories to describe people, it is the idea of the category and the matrix of which it is a part that we see, not the category as an embodied or material entity.The act of classifying someone as a specific category or type of person thus involves combining empirical observations with discursive features.Smith's (1978) article '"K is mentally ill": The anatomy of a factual account' provides a good example.Here, Smith outlines how a person (K) is constructed as mentally ill through the act of combining specific (empirical) observations of K's behaviour with specific (discursive) understandings and definitions of mental illness.By this 'cutting-out procedure', K is classified as a mentally ill kind of person.
To classify people, statistics are essential, according to Hacking (1999).A movement towards scientific thinking and praxis during the 1800s involved a rejection of determinism, as rooted in religion or tradition, in favour of a reliance on statistics and probability analysis.Equipped with statistical evidence, the population could not only be described, but also predicted.Of special concern was deviance (Hacking, 2004;Turmel, 2008).The graphical notion of normalcy was depicted by the bell curve, where people positioned near the average were considered the norm, whereas people positioned at either end of the curve represented deviance (Turmel, 2008: 81).Weight and height charts, used at public health centres in Norway, and the widely used tool to assess children's language development in kindergartens, TRAS ('Early Registration of Language Development'), are contemporary examples of the same, indicating a 'normal curve' in children's development.As the reliance on and use of statistics increased, new disciplinessuch as developmental psychology and paediatricsand new professions emerged, focused on handling the deviance.Rose (1999) points out how certain regimes of knowledge, which he refers to as 'psydisciplines' and 'psy-expertise', have played a key role in constructing 'governable subjects'.Rooted in these 'psy-disciplines', some theories have arguably gained an almost hegemonic status within professions working with childrennotably, attachment theory and neuroscience (Bjerre et al., 2021;Brodal, 2018;Wastell and White, 2017;White et al., 2019).Simultaneously, as Bjerre et al. (2021) point out, professionals have adopted a scientific language without necessarily knowing or understanding the scientific insights in which it is rooted.In line with Hacking (1999Hacking ( , 2004)), Rose (1999: 133) describes how our understanding of normalcy has grown out of a concern for the 'pathological children'that is, the disobedient, obstinate and noisy children who demand expert attention.Hence, normalcy is something that is not just observed, but also valued.
Hacking ( 2004) is concerned about the feedback effects of classification.Categories provide legitimate ways of being, not only orienting how people see themselves and how we see and value other people, but also feeding back into and reproducing the classification system itself.In other words, the categories and the people categorized come into being at the same time, thus upholding the system of classification.Once classified in a category of deviance, the person becomes the responsibility of the professionals tasked with 'healing' or dealing with the deviance and maintaining normalcy and social order.
So, how do professionals proceed in their work with categories?To an increasing extent, fuelled by the accelerating demand for evidence and accountability claims, professionals rely on social technology.Social technology may have the form of material objects, such as a chart or graph, having the capacity to mediate between social actors (Turmel, 2008: 117-118).It may also be programmes and models that rely on expert knowledge to describe specific measures and conduct.Providing expert knowledge to manage social problems, social technology serves as a connection between expertise and government.Sustained by institutional discourse, it incorporates specific understandings and knowledge.As it is used by and circulated among professionals, it can become an entity of its own in a network of translocal relations, and come to inform how people act and think.As such, it is a necessary part of accountability circuitsthat is, the work done to make performance or outcomes produced on the front line accountable in terms of managerial categories and objectives (Griffith and Smith, 2014: 340).Indeed, professionals' use of social technologies leads to the construction of social categories of normalcy and deviance, thereby invoking 'the standard child' as an ideological code (Smith, 2012).

Making categories fit people: identifying 'children at risk'
Anna is a trained kindergarten teacher and leads a unit for children aged three to five in a kindergarten in southern Norway.In the following quote, taken from an interview conducted by Nilsen, she talks about how different observations could evoke concern for a child: The first thing you notice is the contact between the parents and the child.How it works.How they approach the child and how the children approach the parents.And often you see, when you observe the children, that it relates to, well, weak language development.You see that quite quickly.If that's what it is about, then it is quite obvious, then it is visible to most people.But if it is kind of social things and suchlike, then it might be things like, say, hygiene, for instance, you can see that quite quickly, but after a while you also notice how the children, in a way, how they approach us, what sort of needs they have, and, and of course, like now, Kvello is in the frame, and of course then you also see … well, certain signs that indicate that there might be some risklike signals.Say, if the parents are divorced, if the parents, well, if you know that the parents are often in conflict.If we get, say, they divorce and the mother says this and that, and we notice that there is a conflict between mum and dad, then there is a risk factor indicating, well, maybe the child is in the middle of it all, that it kind of affects the child.… Then there are mental problemsfor instance, if you notice that mum has it or dad has it, then there is also a concern for the child.If you see that the child is afraid of things, say, that you notice that the child doesn't eat enough, for instance.Poor clothing, and yes, and, like, kindergarten attendance.Say they turn up in the kindergarten and you feel that something … And if they don't turn up at meetings, then maybe, well.(our translation) In this quote, Anna makes reference to the Kvello model.This model, which is named after its author, Øyvind Kvello, is informed by theories in developmental psychology (notably, attachment theory) and is widely used in the child welfare services in Norway.Increasingly, it is also used in kindergartens to identify children in precarious care situations.The kindergarten teachers in the municipality where Anna works had attended training by Kvello, part of which involved familiarizing themselves with his book, which incorporates a list of risk factors (Kvello, 2011).The risk factors are sorted into three categories, pertaining to the 'child', the 'nuclear family' and the 'microsystems beyond the nuclear family' (Kvello, 2011: 168-169).In assessing risk, the author says that, if three or four risk factors are present, 'the child is defined to be at risk; with five or more it is defined as highly at risk' (Kvello, 2011: 171).In the interview, Anna points out several of the risk factors that are listed, such as divorce, conflicts between parents and mental illness, as well as indicators such as poor hygiene, malnutrition and withdrawal.
'Risk' and 'risk factors' are concepts that were frequently used by the kindergarten teachers interviewed by Author 2, and they were often explicitly associated with Kvello's list.Moreover, the list was used by other professionals with whom the kindergarten teachers collaborated, and who had also attended the training.Observing at an interdisciplinary meeting, Nilsen noticed that the list of risk factors was the guiding tool used when the participants (a public health nurse, a special educational needs teacher, a child welfare counsellor, a physiotherapist, the kindergarten head and two kindergarten teachers) discussed children of concern.If several risk factors were observed, they agreed to intervene.But for that to happen, they had to decide what the problem was and whose (which institution's) responsibility it was to deal with it.In other words, having sorted out which children to help, the professionals had to sort out what kind of problem the children had in order to delegate responsibility to the right profession.
Talking about her collaboration with the kindergarten during an interview, the health nurse, Karianne , pointed out that, in order to intervene, the kindergarten teachers 'have a job to do [in the kindergarten] before they look for others to involve'.The most important job to be done in the kindergarten in that respect was systematic observation, assessment and documentation.In the kindergarten where Anna works, different assessment tools were used, such as forms to assess children's language development (TRAS) and social abilities (ALLE MED ('ALL IN')).TRAS and ALLE MED are widely used forms in Norwegian kindergartens to assess, respectively, children's language development and social abilities .They share the common feature of being based on an age-determined or stage-based conception of normalcy, and they are both developed in a collaboration between universities and municipal or state agencies.Both forms are in the shape of a circle divided into sections.The different sections of the form are supposed to be coloured in according to how well the child masters the various skillsfor instance, naming different objects or understanding prepositions.The child's shortcomings are thus presented as visual gaps on the form.In addition, the kindergarten teacher leading the unit wrote a pedagogical report for each child and, in cases of specific concern, kept a protocol in which observations were documented.Without documentation, outside intervention from professionals cannot be drawn on or justified.Hence, in order to 'do' early intervention in kindergartens, documentation is essential (Nilsen, 2017a).
The Kvello list and assessment tools are examples of social technologies that are suited for classification.Using these technologies, kindergarten teachers take part in the ruling relations of early intervention, where specific knowledge regimes are activated.Children causing concern 'fail' on some dimensions of these technologies, either because they live in families where 'three or more risk factors are present' (Kvello, 2011: 171) or because there are gaps in the TRAS or ALLE MED forms.The understanding of which children are 'at risk' or 'exhibiting delayed development' hinges to a large extent on the social technologies themselves (what elements they cover and which knowledge regimes they rely on).However, it also hinges on the kindergarten teachers' interpretation and use of the technologies.While the forms and reports may appear to give an objective representation of the child, the documented scores are, in practice, largely the result of the kindergarten staff's discretionary judgements.Knowing how the system works, they can manoeuvre the technologies in ways that render their observations interventionable or not by describing the children in ways that fit the categories.
Making people fit categories: working with 'traumatized refugee children' The participants in Kimathi's study were all working with refugee children and families.As part of building competency for this work, many of the participants in the study had undertaken training specifically aimed at enhancing knowledge of trauma care.One example is Olivia , who has 'a Bachelor's degree as kindergarten teacher … with a particular focus on special pedagogics.I've also been on … courses related to trauma.I'm also trained for ICDP'.Similarly, Ruth has 'a Master's degree in kindergarten leadership and I've also had some training in ICDP and I'm a facilitator and also a trainer in ICDP.… While I've been working here, we've been taking a lot of courses'.
As these quotes exemplify, the kindergarten teachers had attended several training programmes particularly aimed at working with refugee children, on top of their Bachelor's or Master's degrees.
In their everyday work and interaction with refugee children, they rely on this training, and the knowledge mediated by the programmes organizes their work and thinking in specific ways.Even among the kindergarten staff who had not attended the training, frequent references were made to concepts that are embedded in these programmesnotably, the ICDP.
On its website, the ICDP is described as 'a simple health-promoting and preventive programme whose goal is to enhance care and improve the experience and circumstances of children and young people.It targets caregivers, with a view to maximizing their competency in their role' (our translation). 1The ICDP is described as a psycho-social intervention and builds on a broad theoretical backgroundnotably, attachment theory.It has become an integrated part of the introduction programme for newly arrived refugees in Norway (Solberg, 2020).The ICDP is presented as 'culturesensitive' and 'sensitizing', and its focus is on empowering and supporting caregivers, in contrast to instruction-and manual-based programmes.Despite several evaluations that document a positive effect of the ICDP, some scholars warn that much of this research is related to specific interests that may prohibit critical perspectives and also fail to take parents' perspectives and experiences fully into account (Solberg, 2020;Sundsbø, 2018).
During the interviews, the participants described how refugee children are in need of safety and comfort.They argued, for instance, that building a safe feeling among children is at the core of integration in kindergartens.This is in line with 'the emotional dialogue' of the ICDP, which has four guidelines: (1) show the child positive feelings; (2) follow the child's lead; (3) talk to the child using emotional expressions, gestures and words; and (4) praise and appreciate what the child does.Ruth reported: I use a calm voice and I say, like, comforting words, because when you say comforting words, you automatically adopt a way of speaking that is calm and reassuring.I use physical contact a lot.I carry and hug them.I sing, [make] eye contact, … and a lot of talking reallyeven if they don't understandbut talking is a kind of therapy and that's a way to make them feel safe.
Assuming that many of the children 'had experienced traumatic events' (Nancy) or had 'had out-of-the-ordinary [traumatic] experiences' (Joyce) , and that there had been 'a lot of stress for the child' (Olivia) both during and after the flight from their home country, the kindergarten teachers frequently referred to both the children and their parents as traumatized.The ICDP is arguably well suited for work with traumatized children (Christie and Døhlie, 2011: 79).In the interviews, it appeared that the kindergarten teachers implicitly assumed that a child with a refugee background would be traumatizedhence, their needs were portrayed in a similar way.Lunneblad (2017) discovered a similar tendency in his research on kindergartens' work with refugee children in Sweden notably, that children's vulnerability and need for safety were emphasized when kindergarten teachers talked about these children.Hence, as a category, 'refugee children' seems to connote 'being vulnerable' and/or 'traumatized'.Moreover, it exemplifies how professional jargon is used generically, despite the ICDP's insistence on cultural sensitivity, hence producing a standardized way of seeing and interacting with the children.In contrast to Nilsen's study, the kindergarten teachers did not seem to investigate the needs of each individual child to assess if they were, indeed, traumatized or 'at risk', but rather relied on a standardized and categorical understanding of what refugee children need.

Some unintended consequences of categorization work
The two projects illustrate how social technologies organize everyday work in kindergartens in specific ways.As we have pointed out, these technologies rely predominantly on attachment theory and neuroscience, which are indicative of what Rose (1999) refers to as 'psy-disciplines'.Several scholars have pointed out how this knowledge has 'imperialized' the professions (e.g.Brodal, 2018;Wastell and White, 2017;White et al., 2019).Acknowledging the good intentions behind the intervention of such technologies, we will point out three interwoven and unintended implications that could have harmful effects.
First, employing social technologies involves 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 instance, a refugee child's crying is interpreted as emotional stress, and poor clothing is interpreted as poor parenting.Assuming that kindergarten teachers have automated their response to such observations in an almost behaviourist manner would be a misinterpretation.Indeed, our data clearly indicates that the kindergarten teachers make discretionary judgements along the way, while still taking notice of the observations that 'count' and that can be transformed into concepts fitting the psy-discourse in order to justify an intervention.In this way, their stories resemble the 'cutting-out procedure' described in the abovementioned '"K is mentally ill"' article (Smith, 1978).'Doing' early intervention or integration in kindergartens implies being attentive to signs and indications that can serve as documentation.In this work, the social technologies guide the professional gaze.Moreover, the interpretation of the empirical observations pertains to the children as representatives of a social category, not as individuals.Observations that may serve to nuance or broaden the understanding easily remain unnoticed.
Second, as pointed out by, for instance, Bjerre et al. (2021), the current demand for documentation and accountability makes the psy-language powerful, legitimizing the use of psychological concepts.When professionals reproduce the psy-discourse that is incorporated in social technologies, they rely on 'knowledge', but in a fragmented and random way.Used arbitrarily, the concepts may serve to legitimize a decision without necessarily solving the problem.Moreover, the knowledge they rely on is portrayed as 'facts' and 'evidence', whereas it should also be recognized as normative and moral (Bjerre et al., 2021: 10).
Third, interpreting the children's behaviour as representative of a specific deviant category (such as 'at risk' or 'traumatized') implies a specific understanding of normalcyfor instance, of a 'standard child' or a 'standard family'.However, the normalizing discourses also construct deviance by 'othering' those who are different, or who see or do things differently.Using collective characteristics (presenting people as representatives of a category) rather than individual characteristics (presenting people as individuals), the kindergarten teachers interpret some children's behaviours as symptoms of deviance which might otherwise be understood within a frame of normalcy.Thus, the paradox is that, with intentions of doing good and helping the 'needy', the social technology and the professionals' use of it ends up constructing the categories they are intended to help or 'heal'.This is an example of a feedback effect, whereby the category and the people categorized come into being at the same time, thus upholding the current system of classification (Hacking, 2004).
As observed in the introduction, Norwegian kindergartens have a responsibility for 'highlighting, valuing and promoting diversity and mutual respect', showing children that 'there are many ways in which to think, act and live' (Norwegian Directorate for Education and Training, 2017: 8-9).At the same time, kindergarten staff are required to 'adapt their general pedagogical practices to suit the children's needs and circumstances' (Norwegian Directorate for Education and Training, 2017: 40), and work methodically and systematically to meet this end.It appears that, to succeed in this respect, kindergarten teachers depend on social technology that has standardizing effects.Instead of representing real people, these technologies represent ideas of people (Hacking, 1999), which, in turn, rely on statistical prevalence and an authoritative psy-discourse.Acknowledging this, however, does not imply that social technology necessarily serves bad purposes.On the contrary, it may indeed strengthen professionals' work.Our point is that, despite its appeal, social technology cannot replace professional judgement and discretion, since serious problems involve people with different needs and seldom have fixed solutions.To promote integration and early intervention, kindergarten teachers should therefore be encouraged to use social technology in a critical and analytical way rather than as a standard routine.This can be achieved by incorporating reflexivity in everyday occurrences in kindergartens.In line with the thinking of Peter Moss (2008: 125-126), kindergarten teachers should be 'democratic and reflective professionals' who value the qualities of dialogue, critical thinking, researching, listening and openness to otherness, uncertainty and provisionality, subjectivity, border-crossing, multiple perspectives and curiosity.Moreover, learning to be critical of social technologies, both during training and in evaluation and staff meetings, can serve as a starting point towards understanding the potential unintended consequences of such technologies.The core argument is that kindergarten teachers and assistants working with children need to build self-awareness and professional awareness, so that they are not the subject of policies that may constrain them with demands for technicist practice