Over the past 20 years, kinship care placements have been given high priority internationally (Connolly, Kiraly, McCrae, & Mitchell, 2016). Marie Connolly writes: 'Governments across the industrialised world now prioritise kinship care as the preferred care option for children who cannot live with their parents' (Connolly et al., 2016, p. 88). This also applies to the Scandinavian countries, including Denmark. Relevant legislation: The Care Reform of 2005 (LOV nr 1442 of 22/12/2004), the Children's Reform of 2010 (LOV of 628 af 11/06/2010). Section 47 of the Consolidation Act on Social Services states that local authorities shall systematically identify and involve the child's family and network, and consider kinship care placements in the first instance (Servicestyrelsen, 2011). There are good value-based, economic, professional and research reasons for doing this. Positive research results have not changed in the last 20 years (Winokur, Holtan, & Valentine, 2007, Winokur, Holtan, & Valentine, 2009, Winokur, Holtan, & Batchelder, 2014, Winokur, Holtan, & Batchelder, 2018; Washington et al., 2018; Kanisha & Wu, 2016; O’Higgins, Sebba, & Gardner, 2017; Connolly et al., 2016). It is still true that children placed in family and network care do better than children placed in foster care (Connolly et al., 2016; Winokur et al., 2018). Research results show that children in kinship care placements have fewer mental and behavioural problems, greater general well-being, experience fewer interruptions of placement and maintain closer contact with their families compared to children placed in foster care (Connolly et al., 2016; Winokur et al., 2018). These results are based primarily on major quantitative investigations, and obviously do not mean that all kinship care placements work well, are unproblematic, or of a high quality, but simply that, statistically speaking, children do better in these settings.

The political and legal prioritising of kinship care internationally has variously affected the number of such placements in western countries. Kinship and network placements in Denmark only increased from about 5% in 2004 to 7.5% in 2018 (Rasmussen, Søbjerg, Jæger, & Rasmussen 2020), while in Norway they increased to about 27%, in Sweden to about 15–20%, in Finland about 15%, in Australia about 50% and in New Zealand about 50%. The figures have been collected from the official national statistics and are approximate in relation to the total numbers of placements in the various countries and are validated as approximations by Scandinavian researchers in the field. The large differences may be due to differences in family values, legislation, the organisation of social work, etc. However, it is noteworthy that such large differences in numbers, and thus in placement practice, can be identified even among the Scandinavian countries, where culture, organisation and legislation in the field are very similar (Mehlbye, Bonfils, & Andersen, 2011; Sundt, 2012).

Since all the industrialised countries have prioritised kinship care both politically and in terms of legislation, we assume that what particularly affects the number and quality of network placements must be differences in the practice of field workers (that is, social workers, their institutions and case procedures) (Lipsky, 2010; Brodkin, 2013). This assumption is partly supported by the fact that decisions concerning placements are made by social workers and their managers according to a legally determined procedure, and partly by a small number of Scandinavian studies focused on social workers: Vinnerljung (1993) finds a sceptical approach to kinship care among social workers, as does Moldestad (2003) in Norway. Scepticism among social workers is mostly based on the fact, that they judge the families to be dysfunctional and therefore not capable of rising to the task, and because they think that the children need a new family. Three years later, Linderot (2020) identified what is in theory a positive attitude towards kinship care in Sweden, where this kind of placement is seen as a 'natural law', though matched by a very cautious approach in practice, where social workers very rarely recommend kinship care. She further comments that social workers only have a limited knowledge of children's networks. Ponnert (2016) shows in her discourse analysis that there is a conflict between discourses: emotional overtones in the case of kinship care; emotional neutrality in the case of placements in 'a real' family. So emotions are either viewed as 'emotions as a glue that binds', or 'emotions as obscuring the perspective of the child'. She uses an approach to discourse analysis inspired by Fairclough (1992), in which discourses both constitute and create practice with a special focus on changes of discourse at nodal points, showing how emotions determine the central premises on which decisions are based. In their historical outline of the practice and attitudes of childcare authorities to kinship care placements in Norway, Holtan, Thørnblad and Skoglunds (2020) show that attitudes have gradually shifted from opposition to regarding such placements as something of intrinsic value. They discuss many reasons for this change of attitude, including positive research results, problems related to the breakdown of foster families, cases of young people who have lost their network, etc. In many respects, the official historical development in Norway with regard to legislation, case procedures and changes in attitude closely resembles the Danish situation, but in practice social workers in Denmark have not internalised such developments, since the number of kinship care placements has not increased significantly.

Our study does not compare the Scandinavian countries, yet they provide a reference point that puts this study into perspective. We examine the Danish practice in two municipalities and use the initial comparison to argue that explanations and solutions are both to be found at the micro level of interaction between field workers and service users. Our assumption is that Danish field workers have more or less continued their previous placement practice, unfazed by research, legislation or political priorities. We are interested in the discourses they use, their professional judgment, their case procedures and their interaction with end users, hoping to find information about what limits or promotes the number of kinship care placements. From our previous research (Rasmussen & Jæger, 2019; Rasmussen et al., 2020) we have formulated the assumption that if the family and its network have not been identified and involved early in the proceedings, before any placement decision has been made, then a kinship care placement is not an option, due to time factors—for example, the time it takes to identify people and get them involved, and the time it takes for a family member to decide whether or not to accept a child, including negotiations between the person who has custody of the child and the social worker. Time pressure is one of the arguments for Lipsky's (2010) defence mechanisms. Lipsky (2010) explores the double-sided position of frontline staff, caught between the end users and the structure and aims of the organisation—a situation in which they can exercise considerable discretion in relation to the distribution of benefits and drawbacks. Professional judgement is a condition of social work, which is a job involving cooperation with people, and cannot be regulated down to the smallest detail. He sees the use of professional judgement as part of the job, but points out that it entails a power shift from management to staff. This means that in reality the formulation of policy is left to the field worker, who, it should be noted, is an independent operator who can use their autonomy and freedom of action to pursue their own interests, those of particular end users, of the profession, of the organisation, or simply follow the intentions of the law concerning a greater use of kinship care placements—but as we have said, this does not happen in Denmark. The social worker's work falls in many ways within the political sphere (Brodkin, 2013), because it involves interpreting and translating legislation into practice. Brodkin interprets Lipsky in this way: On one level, social workers follow rules and regulations; at another level they interpret the regulations; and at a third level they make policy. It is perhaps a little radical to say that social workers make policy, but it is clear that they and other field workers affect the implementation of national and local policy. There may well be a number of defence mechanisms that reduce the pressure on them and facilitate their work, since some cases can be difficult and these mechanisms may offer more control over end users (Lipsky, 2010). After all, involving the child's network means more work, coordination and complexity—and less control (Rasmussen & Sandgaard, 2018, 2019; Rasmussen & Jæger, 2019; Rasmussen et al., 2020). Lipsky uses the concept of 'creaming', which is all about social workers choosing the 'easiest' citizens with the greatest chance of success. There is no question of that here, but it is possible that they choose those placements which at first sight would seem the easiest, and which they believe will be most successful. He has also used the concept 'working bias', which is about the social worker taking a special liking to a particular end user, which in relation to kinship care placements could be a custodial parent who does not want the child in a kinship care placement.

In one of our earlier studies about involving networks (Rasmussen & Jæger, 2019), we saw tendencies towards several of these defence mechanisms, for example: time constraints, a lack of resources, a parental perspective, the fear of losing control, etc., but this does not explain why Danish social workers' practice should be so different from that of social workers in the other Scandinavian countries. It is also possible that they have developed a particular collective professional culture, or discourse that produces meaning, which calls for not involving networks. We need to look at actual practice in more detail, in order to be able to identify any discourses, defence mechanisms—or in general anything that inhibits or promotes the integration of networks and kinship care placements in Denmark—with a view to contributing research-based knowledge to improve the quality and range of kinship care placements.

Method

This study of kinship care placements in an administrative perspective was carried out in 2017–2019 in two Danish municipalities: one with 90,000 inhabitants and approximately 5% kinship care placements; the other with 40,000 inhabitants with about 7% kinship care placements (Rasmussen et al., 2020). Our investigation of the practice of kinship care placements is based on the conduct of specific cases, the general framework of legislation, etc., and the experiences of those involved. A mixed methods case study was used (Creswell, 2003), involving document analysis, interviews, observations and dialogue meetings. The starting point is that a social worker's case follows a specific sequence laid down in law, and a chronology to which are associated various tasks assigned to public authorities in social matters, as well as dialogues with children, parents, family care consultants, managers, other professional groups and the child's family and network. We chose seven current cases—three from one municipality and four from the other –involving kinship care placements that had been running for at least a year, and which corresponded to approximately 50% of such placements in the two municipalities. We have also investigated the other 50% of such placements in both municipalities and have interviewed all the social workers involved. These cases do not differ significantly from the selected cases.

However, the seven cases are very different and quite complicated. These kinship care placements have come about on the initiative of family members—for example, a maternal grandmother who had looked after her grandchild when the mother could not care for her, or a father or paternal grandmother who has insisted that the child not be placed with strangers. Only one of the placements had been made on the initiative of social workers who had investigated and involved the family network, and our study indicates that this also applies to the cases we have not investigated. The two municipalities are similar to other Danish municipalities in terms of organisation and the range of staff employed in the social administration department.

All those actively involved in each case were interviewed (the social worker, the family care consultant, the child, the kinship care family, and the top administrative authority). Subsequently, we conducted seven group interviews involving nearly all the social workers, managers and family care consultants in the two municipalities, during which participants responded to a short questionnaire. The entire corpus of data has been thematically analysed (Andersen, 1990) based on theories of field worker behaviour, discourse analysis and path dependence (Lipsky, 2010; Brodkin, 2013; Fairclough, 1992). The analyses and results were validated and quality assessed by staff of the two municipalities at a subsequent dialogue meeting. The respondents gave their consent to the study and the researcher has complied with UCSYD research ethics.

Results

In all seven cases from the two municipalities, all the private individuals and professional workers involved were to a large extent satisfied by the kinship care placements. We are looking at seven very different case trajectories, some of which have been very complicated, but when assessing the current status of all those involved in the placements, the children seem to have benefited from them. In almost all of the seven cases, kinship care placements have meant a placement within the family at the insistence of the family. As we have pointed out, almost none of the placements were arranged on the initiative of social workers who had investigated and involved the family network. Like all municipalities in Denmark, the municipalities investigated have followed the prescriptions and standard procedures laid down in the Danish Social Service Act, and their organisation and staffing do not differ significantly from other Danish municipalities. We have structured the thematic case analysis around the mindset of the social workers, the defence mechanisms and other challenges related to the case trajectory.

The Mindset of the Social Workers; Defence Mechanisms and Challenges in Practice

Social workers inform us that they find it hard to work systematically with family network involvement. There is a lot of talk about family network involvement and kinship care placements, but closer look at the case trajectories reveals that not much work was done on uncovering family networks. The social workers find it difficult to talk with families about their networks, and typically they find them dismissive about involving their family network in any assessment of the problems or, finding a solution to them. They found it hard to ask questions that might open up a positive dialogue, and equally hard to keep on insisting that the family network be involved: We have cases where we talk and talk about involving the network, but the parents clearly so refuse to do so. It is not our custom to involve the network without the consent of the parents.

Management, social workers and family care consultants were all very open about, and interested in, the question of involving family networks, but this is more like a mindset to be talked about a lot, rather than a real mindset lived out in practice. None of our respondents found that it was a really integrated mindset to be used systematically, in a carefully considered and justified manner, nor a mindset that permeated their work. It seems rather to be something that pops up by chance, or when a family presents a network of its own accord. As one social worker put it: It may well be that a network pops up when we're holding a Signs of Safety meeting. This is one way of figuring out who can be brought into play here, but there is nothing certain about it.

Social workers feel most at home with clear policies and procedures, and since kinship care placements are not an integral part of normal case procedure, then obviously neither colleagues nor management offer any systematic help towards including this element in case procedure. As a result, many cases are rounded off, and many decisions reached, solely on the basis of standard meetings and interviews involving the child, the family and professional workers. As one social worker remarked: Actually, I don't really think I've done anything to motivate the network because I've regarded that as one of the tasks of our family consultants. Adding: It's pretty much a matter of chance who gets involved in this, and when. In one of the municipalities, there was a general feeling that networks are only involved late in the process, and then only because the network itself has put pressure on, so to speak—insisting on being part of the child's life and on being taken into consideration in terms of offering a place where the child could live.

Staff did not express any opposition to network involvement and kinship care placements as such. Social workers are generally positive, and feel on the basis of their experience that the child's family and the rest of its network are important factors in terms of ensuring the development and well-being of the child. They know the law and they discuss the mindset it indicates, but this remains something to be discussed and reflected on, not something to be insisted on and practised systematically. This corresponds in many ways to the findings of Linderot (2006) in Sweden back in 2006.

Social Workers' Defence Mechanisms and Challenges in Practice

We have found examples of several different defence mechanisms or challenges in the course of our interviews, but have chosen to focus on the most prevalent and the underlying reasons behind them. Conflicts within the family and between the family and the local authority constitute a challenge that is often mentioned. Working in such a complicated field as placements—whether we are looking at kinship care placements or ordinary foster families—is always something of a conflict minefield. Professional workers seem to be working on the assumption that there are more conflicts in kinship care families than in ordinary foster families, and that the conflicts are of a different nature: in the former case, the emotional elements of the conflicts are more dominant than usual.

The Meetings Themselves

Meetings are particularly experienced as conflict-ridden and challenging. As one social worker explains: Clarifying people's expectations is a very important thing, and also a dialogue about roles. It is also very important to get a grip on the way the biological parents see things, because they can easily feel very alone and isolated at meetings.

Having to manage and guide emotionally charged processes is seen to be very difficult. Some of these emotions are experienced as conflicts and as chaotic elements in the social worker's task, right from the first interview with the parents about the involvement of the network to the ongoing cooperation required when organising a kinship care placement. It would seem that social workers find it challenging to navigate in cases with heavy emotional content which can come to dominate the whole process and shift the focus away from practical arrangements and from generally moving the case forwards. There seems to be extra vigilance on the part of the social workers with relation to these emotional conflicts, and this is a factor that deters them from convening network meetings, as well as being one of the predominant reasons why social workers readily accept the conclusion that there is no network, and do not press for meetings that involve network participation.

Parental Bias

If the parents have been reluctant about involving the network, or have rejected the matter in the first interview, social workers find it difficult to take the topic up again. The social workers do not feel that they have a solid professional foundation for this aspect of their work: Things sometimes crop up when we are doing a child development study and uncover the network. We usually ask both the parents and the child if there is a network, but I think we hit a brick wall when we have asked three times and nobody has provided useful information. There is also the time factor.

In a situation like this, the social worker would have to keep on trying, which is experienced as something difficult and unfamiliar and takes them out of their comfort zone. Furthermore, family responses to network issues often make it even more difficult to integrate them as a natural focus in the case trajectory.

Social workers speak of parents who may perceive their attempts to activate a network as intrusive, as a sign of insecurity, or as an attempt to avoid responsibility. One social worker informed us in an interview that he had been given the answer: We pay our taxes and expect to get our money's worth, while others recount being greeted by the answer: No, we don't have a network that can be involved in this. In these cases, a very closed family culture seems to be the reason why parents do not want to get their network involved in both the identification of, and finding a solution to, the problem—which may discourage social workers from any further attempts to involve the network.

The Network

Networks respond very differently to getting involved in the process. Some people are ready to attend meetings but have no suggestions when it comes to finding solutions. Others do offer their help, but this can also be a challenge because there are many emotions at stake. A kinship care mother had this to say: It's been very tough for us, because some people in the family have been accusing us of stealing the family's children and others are saying that we're making a lot of money out of it… Obviously it would have been easier for the family if it had been some stranger's place. But if you look at the way the children have developed, it's gone really well—after a few months they really began to develop and to thrive.

It is undoubtedly difficult to be a kinship care family, and especially difficult when the biological parents do not want their child to be placed in care at all. So, there will usually be disagreements and hurt feelings when the parents do not want their child to be placed in care, not even within their own family network.

The picture seems to be emerging that in the vast majority of cases someone in the network itself steps in to suggest themselves as a possible kinship care family. Either the parents tell the social worker about someone in the network who would like to be considered, or—in the vast majority of cases—someone in the network contacts the social worker, or even the head of department, to inform them that they would be willing to take in the child. In such situations, there is not necessarily any consensus between the parents and the network with regard to a possible kinship care placement.

Being Together is Often Difficult

Spending time together is a central issue for all concerned: the child, biological parents, network families and professional workers. Formal visitation rights have typically been laid down by the social authorities in cooperation with the parties involved. There are different types of visitation rights: unrestricted, supported and supervised. In the cases we have investigated, there has been no support or supervision, but there are relationships or visitation situations that children opt out of for longer or shorter periods. One kinship care family reports: Two out of our three foster children have rejected one of their parents in terms of visiting. One has rejecting the father and the other, the mother. This caused a lot of conflict and was not pleasant at all. The parents withdrew their consent when the children refused to see them, so the cases ended up before the Young Persons' Committee. In connection with another case, a social worker reported: We hold meetings when the issue is visitation rights. In this case that I'm thinking about, the fact is that the mother simply does not understand what children need… In the end I had to stop the meeting when the mother's boyfriend got angry during the meeting and said he hated the child (the child in care) and wanted to wring her neck. It got way too violent and we stopped it. The difficulty here was that everyone in the family got steamed up whenever visiting rights were being discussed, and balloons round because in this case the girl was living with her paternal grandmother. Although everyone is uncomfortable when a meeting develops in this way, the social worker concluded: It is so important that the children attend network meetings because they are forced to face reality and become part of that reality, even if it can be tough; and: I think people leave network meetings with a sense of having been heard and given responsibility.

In principle, a child in care has fixed visitation times with its parents, and these visits are no less difficult in the case of kinship care placements. There are many strong feelings at stake and the social worker is often handed a mediating function that can be difficult to tackle—both at the meetings themselves and in the frequent phone calls from frustrated family members in between meetings. It would appear to be very difficult to navigate in this heavily conflicted, emotional space. Visiting rights in kinship care cases take up more time than when children are placed with professional foster parents. Apart from the cumbersome timescale involved, and paying special attention to the wishes of the parents rather than those of the children, these good reasons are indeed many: All interviews involve feelings and conflicts that are to a certain extent unmanageable, chaotic and difficult to cope with, so it would seem that the central defence mechanism is all about not entering this heavily conflicted, emotional and 'private' space, rather than any arguments about lack of time and resources.

Discourses: Children Needing Therapeutic Care; 'Notorious' Families; and Kinship Care Families Must be 'Just as Good'

When a child is to be placed in care, most often some assessments are made as to whether the child is in need of treatment or therapeutic foster care. Typically, such assessments are made by professional workers, often psychologists. There can be variations in how specifically the child's needs are described, why the individual child needs therapeutic or treatment care, and how these needs can be met. When for one reason or another the child is deemed to require therapeutic or treatment foster care, it is as if some doors are opened and others are closed. The doors that are opened are typically the doors to a more professional placement environment, such as institutions, social-educational residential units, or municipal foster families. The doors that are closed are typically those to the family's own network and the solutions offered by such a network: When we talk about children requiring treatment, we tend to look at professional residential units. I fear that focusing on a child's special needs for treatment may exclude network involvement (social worker).

The exclusion of kinship care and other network placements in the case of a child who requires therapeutic care is not an articulated, well-considered discourse, but rather a silent reaction that is perhaps fully automated, or perhaps part of a culture of silence: If a young person, for example, is in need of treatment, we don't think about networks at all. Though in fact kinship care placements have proved possible on occasion, despite the fact that a young person has been requires special treatment (social worker).

Another discourse that can complicate network involvement is that some families may be 'notorious'—that there are thoughts and judgements about them which are not expressed, but which mean that network placements are more or less ruled out, without anyone actually talking or doing anything about it. When a family is 'notorious' in a municipality, you tend to think that there are probably not many resources available, if a lot has gone wrong in the family, like in this case. It's great if people do investigate the situation more closely, but it's easy for us to understand if they don't (social worker).

A third discourse is that kinship care families must be able to manage on a par with professional foster care families. There seems to be an assumption (on the part of the municipality) that the relationship they have to the child and the love they feel for it mean that they want to have the child living with them, and can more or less cope with anything. In an interview, one social worker mused about whether more children could perhaps be placed in a network family, if there were greater focus on offering help and support: “It's an interesting thought—what IF a kinship care family can manage, for example, 80% of the assignment, but not the last 20%, and we might be able to step in and help here, but nobody seems to think that way”. For many kinship care families, accepting and loving the child is not a major challenge. Kinship care families face many other challenges, for example: the child's moving in, the child's grief and feelings of loss, visitation issues, agreements with the biological parents and cooperation with public authorities about the child, covering expenses, courses, supervision and ongoing skills development. Both social workers and kinship care families agree that they do not receive that much support, and they also agree that an inherent (and unstated) attitude within the system makes it difficult to meet the needs of kinship care families in terms of support, skills development and financial compensation.

Institutional Challenges and Collaboration with Other Professional Groups

Most municipalities have an organisational division of labour, which means that the authority responsible for the conduct of the case makes decisions about the placement of children, while the family care consultants prepare, match and supervise care families. There appear to be some interfaces in this cooperation that create difficulties, and which are not necessarily conducive to a focus on getting networks involved. One social worker explained: Family care consultants work in very different ways and there are a lot of differences from person to person. Sometimes, as a social worker, I find myself treading water, because I don't know exactly what's going on in the case during the waiting period between when we have handed the placement case over to them, and them finding an appropriate match. We don't get hold of the network in this period to let them take over a bit. It might be an unfortunate discourse if we pulled in the network when trying to place a child, and then ended up deciding that the child should be placed in professional care rather than with the network.

In principle, it is the responsibility of the social worker to uncover a child's network and therefore the possibility of a kinship care placement. Once a child assesment study has been done and a decision made to place a child in care, the placement aspect of the case is handed over to a family care consultant, who must then match the child to a suitable foster home—be it a kinship care family, an ordinary foster family or a social-educational residential unit. If there is a relevant kinship care family, the family care consultant starts a possible approval process: The family care consultants approve the network families, but that is at the request of ourselves or our line manager. The two family care consultants go off together and approve or disapprove of the placement. As a social worker, I am not part of the process.

So there is a waiting period from when the decision to place the child in care has been made to when a foster family has been approved and matched with a child. It is obvious that network involvement is not an integral part of case procedure, so if the preplacement child welfare investigation indicates a placement, the social workers will be pressed for time, and will not have the time to investigate and involve the network with a view to a possible kinship care placement. So if the possibility of such a placement has been indicated very early in the case, or just before the actual placement, then it will simply be too late. Moreover, a placement is often a painful, delicate process, during which social workers, on their own admission, try to take into account family feelings and conflicts. The upshot is that they feel they have to protect the family network and therefore do not make use of it to relieve or support the family during the matching period. This can be interpreted as a defence mechanism, so that in the guise of concern for the child and the family, what is actually happening is that the social worker is trying to avoid entering into the emotionally charged and conflicted family space.

Discussion

Our findings confirm the initial hypothesis that when there has been no network involvement, or involvement of the family and its network in the decision regarding placement, then it seems too late to find suitable placement opportunities within the kinship network. We are dealing with an emotionally saturated and fragile matching process. During this period, the care and safety of the child could be taken on by someone in the family, for example grandparents with whom the child has a relationship, but there is some indication that this option is often dropped for the sake of the child and the parents. However, it might be the case that it is dropped for the sake of the social worker and other professional staff, who simply cannot face entering this private space rife with emotions, chaos and conflicts, in which they risk losing their grip and control of the case and will have to resort to playing an authoritative role. It can be uncomfortable, conflict-ridden, time-consuming and cumbersome to run meetings and be in contact with so many, often very emotionally involved, people around a child. In her article, Ponnert (2016) draws attention to the confrontation between the emotional discourse and the neutral professional family care discourse used by social workers, pointing out that the so-called nodal point (Fairclough, 1992) of the emotional discourse is whether this is good or bad for the child and its parents. If it is considered bad, then the choice will be a neutral placement in a professional foster care family. She does not focus on the defence mechanisms (Lipsky, 2010) used by social workers, but demonstrates how an assessment of the emotional burden involved has a major influence on the decision reached, so this is perhaps not just a nodal point with regard to the welfare of the child, but also with regard to the professional workers involved.

There are other discourses that are equally crucial to whether a kinship care placement becomes an option. The first is whether the child is assessed as in need of therapeutic care, because the attachment of this label to the child becomes a nodal point from where the professional solution becomes the only option. When a child requires treatment or therapeutic care a set of very complicated discussions is called for. Our study suggests that the concept of 'in need of therapeutic care' is used by professionals more as a floating signifier both to determine and legitimise decisions concerning the type of placement. This discourse is related in part to the discourse that a kinship care foster family must live up to the same standards as a professional foster care family, but as they are 'amateurs' it is argued that they will not be able to look after of a child in need of therapeutic care, and that it therefore makes no sense to examine the possibility of a kinship care placement. The cases covered in this investigation are very complicated and quite a number of the children are described by the social workers as in need of treatment care and other studies show that they do receive less support than professional foster families and that they are often are poorer and less well educated (Holtan et al., 2020). Finally, the same applies to the discourse on social inheritance and dysfunctional families—cases where the child needs a different and better family. In the case of all three discourses they make sence, but we could be looking at them as conscious or unconscious defence mechanisms set up to avoid entering the family's private emotional space to investigate the possibility of kinship care placements. Maybe when social workers ask the parents and get a negative response they are not willing to shoulder the conflicts and move on to investigate and involve the network. This is a bias in favour of the parents (Lipsky, 2010). The social workers do not want to disappoint anyone, if it should turn out that no one in the family can be approved, nor do they wish to enter into a conflict between the families of the father and mother in cases where one side of the family might be able to take care of the child. And this corresponds with the whole question of responsibility and control: Generally speaking, it is far easier and less time consuming to cooperate and share responsibility with other professionals—something that Lipsky mentions in connection with defence mechanisms (Lipsky, 2010).

Concluding Comments

The social workers are not as such opposed to the official policy, nor are they pursuing their own political interests, such as Brodkin, and in part Lipsky (Lipsky, 2010; Brodkin, 2013), would insinuate, but they certainly do not implement this policy and therefore, in a way, formulate their own policy in practice. Our study indicates that it is far more a question of avoiding anything 'nasty'—all that is uncontrollable, troublesome and time-consuming about the family's private emotional space—and about taking and sharing responsibility with non-professionals, than about insuring professional or political interests. We began our investigation with the assumption that this was due to the behaviour of social workers in the field. This has been partially confirmed by our case study, but there are indications that it should be seen in the context of time-consuming and vulnerable matching processes, and the absence of any systematic involvement of the network before matching is completed. One of our main findings—combining discourse theory and the behaviour of fieldworkers—in our analysis is that behind the social workers discourses and behaviour, there seem to a reluctance to enter an emotionally charged family space. If that is the case, then it would be very interesting to take an even closer look at this phenomenon and compare with other countries especially Norway, that has many more kinship care placements.