1 INTRODUCTION

Domestic violence is a serious and global social problem (Devries et al., 2013; Dufort et al., 2015; World Health Organization, 2014). International studies show that men are responsible for the majority of and the most severe domestic violence, which is primarily directed at women (Devries et al., 2013; World Health Organization, 2021). Many of these men are fathers, and the victims are mothers and children (Gavell Frenzel, 2014). In Sweden, the social services are mandated to investigate, assess and provide interventions to these fathers (HSLF-FS, 2022, p. 39). Almost one-fifth of reports to Swedish child protection services involve domestic abuse (DA) (Socialstyrelsen, 2022). Since 2021, social services have been obligated to ensure that individuals who have committed DA change their behaviour (SFS, 2001, p. 453). They should be offered interventions based on a child protection investigation (CPI) (HSLF-FS, 2022, p. 39). This represents a shift towards an increased emphasis on the responsibility of the social services for prevention work with individuals who commit DA.

The child welfare system in Sweden is based on a Family Service Model that emphasizes voluntary participation, collaboration and support for the family (Gilbert, 2012), and motivating fathers to voluntarily cooperate in CPIs is therefore crucial. However, investigating DA in the context of CPIs is a complex task for SWs, because accounts of what has happened often differ between family members. Men who have committed domestic partner abuse rarely interpret their actions as abusive (Boethius, 2015; Gottzén, 2012), and several studies have shown that DA tends to be renegotiated and overlooked in CPIs (Heimer et al., 2018; Mattsson, 2017). Child protection services often fail to involve fathers in investigations, especially when they have committed DA (Brandon et al., 2019; Campbell et al., 2015; Featherstone & Peckover, 2007; Scourfield, 2006). Research shows that the focus of CPIs regarding DA is often directed at the mother's ability to protect the children rather than the father's abusive behaviour (Beth & Stefan de, 2019; Boethius, 2015; Eriksson, 2003; Mattsson, 2017; Scott & Crooks, 2006; Strega et al., 2008). However, ending DA is difficult if we only focus on those who have suffered it, without addressing those who commit the abuse and its underlying causes (Hoppstadius, 2020; Scott & Crooks, 2006). Against this backdrop, one important question involves how the focus can be shifted towards fathers, so that they become more involved in CPIs. This study examines how fathers are involved in CPIs in practice. More specifically, the aim is to examine what strategies SWs use to achieve cooperation with fathers in CPIs that have been initiated due to the father being suspected of DA against the child's mother. The term ‘strategies’ refers to the actions taken by SWs.

1.1 Relevant literature

Several studies have illuminated the challenge of engaging fathers in CPIs (Gordon et al., 2012; Pfitzner et al., 2017; Scourfield, 2006). It is therefore important for SWs to have the skills required to build alliances with fathers, for example by demonstrating empathy and maintaining a non-judgmental attitude. Schreiber et al. (2013) have argued that to engage parents in the initial phase of CPIs, SWs need the ability to overcome parents' fears of their children being placed in care and the shame of being judged or labelled as bad parents. Brandon et al. (2019) emphasize the importance of SWs establishing contacts at an early stage, containing fathers' emotions and being flexible and reliable. Scott and Crooks (2006, p. 37) have argued that a guiding principle in this process should be ‘invite, don't indict’. Additional methods that have been found to promote fathers' engagement and alliances with SWs in CPIs are strength-based and cognitive-behavioural approaches that can identify fathers' strengths, potential and skills or motivate them by highlighting the discrepancy between how they are and how they want to be (Maxwell et al., 2012; Philip et al., 2019; Scourfield, 2006).

Being able to address clients' feelings of shame has also been identified as a crucial means of addressing the barriers that shame creates in CPIs. Shame tends to produce difficulties for parents' understanding of how children are affected by abuse because they are preoccupied with managing the feelings evoked by their own shame. A shame-sensitive approach, which involves recognizing and appropriately responding to shame in order to mitigate its potential negative effects, plays a crucial role in overcoming these barriers (Dolezal & Gibson, 2022). Researchers advocate shame-reducing practices (Gibson, 2015; Salter & Hall, 2022; Walker, 2011) as well as dignity-conferring practices (Dolezal & Gibson, 2022; Salter & Hall, 2022), whereby parents can feel trust and safety in their relationship with SWs. The goal is for parents to recognize how their behaviour impacts their child's development without feeling worthless (Gibson, 2015; Walker, 2011). Research has also shown that men who have committed domestic partner abuse are often unwilling to see themselves as abusers and frequently rationalize, minimize, redefine, justify or excuse their actions. The underlying premise is that when these men face scrutiny, they need to provide acceptable accounts in which they both explain the abusive acts and attempt to present themselves as morally decent. In essence, this revolves around the men recognizing that their abusive acts are morally unacceptable, leading to a fear of how others will perceive them (Anderson & Umberson, 2001; Boethius, 2015; Gottzén, 2017; Hearn, 1998). Motivating fathers with abusive behaviour to change is therefore not easy. Research on interventions shows that methods that help fathers to reflect on their abusive behaviour and its consequences for the child's well-being can be used to promote their motivation to desist from abuse (Bourassa et al., 2017; Broady et al., 2017; Hester & Newman, 2020; Humphreys et al., 2019; Stanley et al., 2012).

1.2 Theoretical framework

The main mission of the social services is to help individuals change their behaviour to bring about an improved and more desirable social situation. In this regard, SW practice can be understood as moral work, because it involves a dimension in which the client's moral worth is categorized and judged (Hasenfeld, 1999). Working with CPIs initiated due to a father's DA is one example of this. When SWs investigate and assess the father's caregiving capacity, it becomes clear that abuse is a socially undesirable act that SWs are tasked with addressing and counteracting. At the same time, the moral dimension can complicate the task of creating a shame-reducing interpersonal relationship with the father.

According to Scheff and Retzinger (2001), interpersonal relationships are located on a continuum between secure social bonds and threatened social bonds. Both types of bonds are closely intertwined with emotions of shame or pride. If the bond is secure, pride will arise. If the bond is threatened, shame will arise. The status of the social bond is defined by the degree of attunement between individuals, in the form of mutual understanding and acknowledgement. ‘High levels of cooperation are based on effective communication patterns, which depend upon attunement and pride’ (Scheff & Retzinger, 2001, p. 38). Social bonds are constantly changing and are tested in every social situation, being alternately threatened, repaired, rebuilt, preserved or damaged, and they are constantly at risk (Scheff & Retzinger, 2001). Shame comprises a family of emotions that includes many variants such as embarrassment, guilt and humiliation (Scheff, 2003b). Several researchers distinguish between feelings of shame and guilt, suggesting that guilt is related to what one has done whereas shame is related to who one is. Guilt is seen as an ‘other-oriented’ emotion and is linked to the ability to adopt a perspective on the person harmed by a specific behaviour (Leith & Baumeister, 1998; Tangney, 1995). Guilt operates constructively in interpersonal relationships by pointing towards a change in behaviour, whereas shame operates in a self-centred manner (Tangney, 1995).

Given this, when establishing a relationship with a father suspected of abuse, it is important for SWs to understand that shame and pride are social emotions that arise based on whether the father perceives, correctly or incorrectly, that he is valued in the encounter. Shame often automatically accompanies the encounter with SWs because the father is accused of being abusive and of having crossed moral boundaries. Shame makes the individual self-conscious and signals a threat to social belonging. Shame often exists at an unconscious level and is rarely acknowledged (Scheff, 2003b). Scheff (2003a) has argued that men often tend to suppress feelings of weakness and rarely directly express shame. SWs must therefore be aware of their moral work while simultaneously balancing it against the father's experiences of both pride and shame (Scheff & Retzinger, 2001). This line of thought provides a basis for understanding what SWs encounter when they meet fathers, why it is difficult to reach them and what SWs can do to try to engage them.

2 METHOD

This is a qualitative study based on 31 semi-structured interviews with 15 SWs, conducted during their work with 12 CPIs. All interviews were conducted by the first author. The research was carried out in two medium-sized Swedish municipalities, referred to as Municipality A and Municipality B, in 2020–2021.

Individuals in leadership positions in local social services were contacted for the study. In the two municipalities in which the responsible managers showed an interest, the study was introduced to the SWs who worked there. SWs who wished to participate contacted the first author, and interview appointments were scheduled when the participants started working on a case that met the study's selection criteria: cases initiated due to the father's suspected abuse against the child's mother and involving children aged 0–12 years. The initial plan was to conduct the interviews at the SWs' offices. However, all interviews except the first two were conducted via video link due to the COVID-19 pandemic and social distancing recommendations.

The interviewees were provided with verbal and written information about the study, including that participation was voluntary, information on how the data would be stored and handled, and that all names would be anonymized and fictionalized. Verbal consent was obtained from all interviewees. The project was approved by the Swedish Ethical Review Authority (Ref/dnr 2020-03487).

To closely follow the SWs' practices, the design was based on a three-step interview series: one at the beginning, one in the middle and one at the end of the investigative process. The interviews took place over a period of approximately 4 months for each case being followed. However, the interview structure remained flexible in relation to the interviewees and their ability to participate, resulting in some interview series consisting of two rather than three interviews. Semi-structured interview guides were employed, which addressed the participants' approaches and reflections on their work with the fathers. The interviews lasted 25–53 min.

Two social workers were responsible for each case. In Municipality A, both social workers often participated in the interviews, whereas in Municipality B, it was common for only one social worker to attend. This can be understood as being due to differences in organizational and staffing stability between the two municipalities. In Municipality B, there was a high staff turnover, the municipality was undergoing extensive reorganizations and the social workers were experiencing a heavy workload. This made it more challenging to secure interviews with the social workers in Municipality B than in Municipality A. Thirteen participants were women, two were men, with an age range of 26–48 years. All participants had a bachelor's degree in social work. Four had graduated within the previous 0–4 years, seven 5–8 years earlier and four nine or more years earlier (Table 1).

2.1 The cases

The cases had been reported to the child protection services by the police, healthcare, the public or another social services unit. In two cases, the mothers had themselves contacted the social services. The abuse involved ranged from choking, attempted rape, shoving, punching, kicking, death threats, prolonged verbal abuse, control of financial resources and isolation from community life. The frequency of DA varied from isolated incidents to systematic actions. A total of 19 children were involved in the cases. Seven of the 12 mothers reported being victims of abuse; the remainder claimed there had been no abuse. Three of the 12 fathers admitted to abuse against the mothers. The CPIs did not result in any coercive measures from the social services. Interventions were offered on a voluntary basis. The four fathers who agreed to interventions aimed at preventing their abusive behaviour were living with the mothers they had been abusive towards and wanted to continue the relationship. Fathers who defined the incidents as conflicts rather than violence were either not living with the mothers who claimed to have been subjected to abuse, or if they were living together, both parties denied the occurrence of abuse. Most fathers in the study were either employed or studying. All but one father had a non-Swedish ethnic background.

2.2 Data analysis

The data were analysed using thematic analysis, inspired by Braun and Clarke (2006). The method is systematic, flexible and consists of several steps (ibid.). The empirical material was manually organized into codes, themes and sub-themes to identify patterns of meaning in the data in relation to the study's objective (ibid.). In the first analytical step, all the interviews were audio-recorded and transcribed verbatim. The transcripts were read repeatedly, and meaningful sentences and paragraphs in the text were highlighted. In the second step, the data were organized and labelled using different codes. The codes were processed and gradually merged into two main themes: safety-creating strategies, which aim to create a sense of emotional safety in the relationship, and strategies for making DA talkable. Within these themes, various sub-themes were developed. The themes guided the search for theories, and the main themes were then renamed in line with theory. The findings were discussed by the authors throughout the analytical process.

The thematic analysis was presented to the participating social services, who were given the opportunity to provide feedback on the interpretation of the empirical data. The SWs confirmed a recognition of the created themes.

2.3 Limitations

This study is a local qualitative study based on a small number of cases and SWs, which may affect its generalizability. The study is focused on the interpersonal level, where cooperation with fathers may be achieved, but other levels are also relevant and need to be addressed in this work. Further, the article's focus is directed at SWs' descriptions of their strategies in relation to fathers and not on fathers' experiences of these strategies. Fathers' experiences of their interactions with SWs will be the focus of another study in this research project.

3 RESULTS

The results are presented below based on the two main strategies that emerged from the analysis: Securing an Initial Bond and Maintaining the Bond while addressing the father's abusive behaviour.

3.1 Securing an initial bond

In the interviews, SWs often talked about their difficult role in the investigative work, in which they are supposed to investigate and support all the parties involved. They said that parents are often concerned about contacts with the social services because they are unsure what the consequences may be. There is often a fear that the children will be taken into care by child protection services if they become aware of the DA. This is particularly evident in the SWs' contacts with fathers. These are rarely initiated by the fathers themselves but are rather initiated because the fathers are suspected of partner abuse. This risks making the fathers worried and unmotivated to cooperate with SWs, which the SWs need to handle.

3.1.1 Reducing the tension in CPI proceedings

A recurring theme in the interviews involved SWs describing that they avoided being ‘harsh’ and ‘confrontational’ in their relationships with the fathers, instead trying to relax the fathers by being ‘kind’ and ‘pleasant’. They also said that they often emphasized that they do not take sides between the father and the mother regarding what has happened.

Ines: There is always some kind of balance in how kind you should be. […]. I think we have been quite flexible with the parents…

Stina: Especially with the father.

Ines: Yes, respectful also of his perspective and experience to be able to provide nuance… We get a different view from the school. A different view from the mother too, and that, well, that we are not taking anyone's side. But we get different views of it and then we try to form some kind of middle ground, or that the truth lies somewhere in between (Case A7).

Another way of reducing the tensions was to consistently inform the father about the CPI proceedings and their purpose and to emphasize the SWs' helping and supportive role.

Britt: We tried to reassure him and explain that our purpose is to offer support and nothing else in this case. That we are not threatening in any way (Case A6).

Emphasizing support and clarifying that it was not about something ‘else’ can be seen as a signal to the father that he should not worry about negative consequences of the contact with the SWs. Another SW specified the meaning of reducing the ‘threatening’ aspect of the meeting.

Ines: We repeatedly talked about our purpose. That we are not here to take his son away from him. In the beginning, he was probably afraid that he would no longer be able to see his son or that he would be deprived of his parenthood and that his wife would leave him (Case A7).

The SWs' mission is to map the father's abusive acts and assess his parenting abilities and the need for intervention, which inevitably involves a moral assessment of the fathers. Brandon et al. (2019) have emphasized the importance of initially meeting fathers in a supportive manner in the early stages of investigations to establish constructive working relationships that can last throughout the investigation. In line with this, Schreiber et al. (2013) found that one way to reduce shame and engage parents in the initial phase is by explaining the proceedings and providing emotional support. In CPIs involving DA, this is particularly challenging because acts of abuse are morally unacceptable and strongly associated with shame, which often means that individuals who have been abusive struggle to talk about and take responsibility for their actions (Boethius, 2015; Campbell et al., 2010; Gottzén, 2017; Hydén, 1995). The SWs' strategies to reduce tensions, which initially focused on helping the father understand that the child protection services were there to support him and the family, can be understood in light of this, as a means of downplaying the moral dimension in their interactions with the father. In this way, they may begin to build a secure bond through which shame is contained or reduced and threats to the relationship minimized (Scheff & Retzinger, 2001).

3.1.2 Affirming the father

Affirmation is the second strategy used to build a secure bond. Several SWs mentioned the term affirmation in relation to asking the fathers about their accounts of what had happened within the family.

Annika: I think that part of affirmation is showing that you have listened to them, being able to retell what they have said and ensuring that you have understood it correctly. I told him that he was brave and that the things we were discussing were difficult, and that it was admirable that he was putting it into words (Case A4).

Edina described affirmation in a similar manner and also how it had elicited a positive response from the father:

Edina: I want to hear your side of it, this is your chance. I want to give you an honest opportunity to tell your side of the story, please go ahead. He said it feels good that you are listening to me. I can say that it feels good (Case B2).

In addition to listening and trying to understand the fathers, the SWs also used praise to encourage them to talk about their abuse. Praise became a way of creating a secure bond and pushing shame aside. It involved praising the fathers when they talked about the abuse or described their emotions or when they described positive aspects of their parenting.

Elin: …it [praising] usually helps a lot because if you acknowledge that something they do is good, regardless of how the situation looks, they have created a beautiful child and you see that there is a lot of love and security, and so on (Case A3).

Being praised for their parenting and their feelings regarding the children, and for being part of a context, and also receiving affirmation in their role as fathers regardless of what happens was assumed to contribute to a sense of dignity and self-respect, even in a context where shame was constantly threatening to surface.

The SWs' affirmation thus entailed communicating their willingness to hear the father's story and to try to understand him without questioning. It also involved giving praise when the father showed courage in talking about abuse or when positive aspects of his fatherhood came to light. Other researchers have also emphasized the importance of establishing a respectful and empathetic relationship with fathers in CPIs, which may be achieved by initially avoiding judgement or questioning and by highlighting the father's strengths and skills (Brandon et al., 2019; Maxwell et al., 2012; Philip et al., 2019; Scott & Crooks, 2006; Scourfield, 2006; Zanoni et al., 2013). This strategy aligns with the principles of the shame-sensitive and dignity-conferring practices advocated by Dolezal and Gibson (2022), in which relationships are based on dignity, respect, empathy and trust to create a sense of emotional safety. This can help clients feel comfortable discussing their experiences without fear of judgement. Similarly, Scheff and Retzinger (2001) have noted that the degree of cooperation between individuals is determined, among other things, by the level of understanding and acknowledgement. The SWs' use of affirmation strategies can be understood as a way of fostering cooperation with the father. Through affirmation and demonstrated understanding, the father is reassured that despite the reports of the morally questionable abuse, he is valued positively by the SWs. This helps to establish the secure bond, laying the foundation for cooperation with the father.

3.2 Maintaining the bond while addressing the father's abusive behaviour

The second main strategy employed by SWs to achieve cooperation with the father involved making DA discussable without losing the secure bond they had established. Several SWs argued that sensing when and how they could address the issue of abuse involved a delicate balance. Talking about DA poses a significant challenge because it risks evoking feelings of shame and threatening the bond. To maintain the bond and cooperation while addressing the abuse, SWs negotiated with fathers about the DA and attempted to make them see the abuse from the victims' perspective, that is, to shift perspective.

3.2.1 Negotiating the naming of DA

Negotiating strategies involved the SWs attempting to find a common language with the father regarding how abuse could be named. The starting point was the mother's and the child's experiences and descriptions of abusive incidents. Based on these, the SWs explored what the fathers could accept and agree with without disrupting their cooperation. One recurring description in the interviews involved using words other than DA in discussions with the fathers. Some SWs used terms such as ‘disagreement’ or ‘high level of conflict’.

Hannah: I try not to constantly talk about physical abuse but rather focus on a high level of conflict. It is intentional on my part, as you might reach the father a bit better that way (Case B4).

By doing this, the fathers' actions become less highly charged as moral transgressions. Conflicts are something that occur in all couple relationships, whereas DA is a clear moral violation. Often, SWs tried to use the words that the fathers used themselves.

Annika: I try to use the words they use themselves and make it talkable by continuing in the same way as they express themselves and not using the term abuse (Case A4).

Following the father's choice of words thus became a way for the SW to match the father's pace. This can be understood as making abuse talkable, even if it is not explicitly labelled as such, and thus starting to address it. Shame does not then block the conversation and the relationship. In some cases where SWs and the fathers had reached agreement on certain aspects of the abusive situation, but had not yet defined it as DA, the focus shifted to other difficulties, such as the child's experience and interpretation of the situation. In one case, the child mentioned that he got ‘sad when he needs to keep dad away from hitting mom’, but in the conversation with the father, the SWs focused on the child's sadness and the parents' need for help to curb their ‘loud arguments’, rather than on the child's experience of abuse against the mother.

Another negotiating strategy used by SWs involved focusing on a less severe act of physical abuse that had been described in the referral to the child protection services.

Annika: We haven't discussed all the abuse [with the father]. Now, it's about the incidents that have happened in front of the child (Case A3).

Annika described how she and her colleague had discussed occasions on which the father ‘pushed the mother’ in front of the child, despite being aware that he also engages in more systematic abuse in the form of psychological control. They chose to start by focusing on a mild form of abuse that was limited in time and space. This became the lowest common denominator on which they and the father could agree. Several other SWs described a similar approach, which involved reaching agreement with the father regarding specific aspects of the reported abusive incidents. These agreements did not necessarily mean that the father admitted to having subjected the mother to abuse, but he might acknowledge that something had become problematic in the situation that led to the referral to child protection services. One of the SWs illustrated this way of negotiating in an investigation in which the father, according to the mother, had made a death threat against her.

Edina: You describe that you haven't threatened, and the mom describes that you have threatened her. The child has been sitting and screaming in bed during this situation. You both talk about that. There you agree, and that's what I want to focus on (Case B2).

The SW thus found common ground in relation to the consequences for the child rather than focusing on the father's threatening behaviour towards the mother. The father claimed that he had experienced a ‘panic attack’ and felt vulnerable. The SW sought agreement with the father regarding his mental health issues rather than the abuse. In this way, the focus was directed towards something less shame-inducing, the father's illness, which did not threaten the bond to the same extent as a sole focus on discussing the abuse.

Negotiating can also involve addressing the parents in a way that distributes responsibility for the abusive behaviour between them. For example, one SW described how he had spoken to a father who had taken a chokehold on the mother:

Michael: If you're going to be in a relationship like this, you two need to take measures to reduce the risk of it happening again. What do you two think you need to do? (Case A5)

Michael stated that addressing both parents was a way of talking to the father to ‘remove the shame’ while still focusing on the ‘issue at hand’. By referring to the incident using ‘you two’, the father was able to feel that he was not the only person responsible in the situation, even though he was solely responsible for the abusive behaviour.

Based on Scheff and Retzinger's (2001) work, the SWs' negotiations regarding the DA can be understood as a way of creating a mutual understanding with the father regarding certain aspects of the abuse described by the mother or the child. In this way, cooperation and a secure bond are maintained. However, there is a risk, as was evident in several of the cases, that the SWs' reformulations and negotiations with the father resulted in reducing the DA towards the end of the investigations and describing it as less serious or as not being DA at all. In line with Hasenfeld's (1999) arguments, this suggests that the SWs' moral values are influenced by their interactions with the father. When the father, or the SW, chooses different words to describe the abuse, the father appears more decent in moral terms, leading to the abuse being toned down. Thus, there is a duality in the SWs' negotiating strategies, as on the one hand, SWs need to make the abuse less charged and more discussable, whereas on the other, there is a risk of minimizing the abuse and presenting the father as more decent in moral terms. Similar findings have also emerged in other research (Anderson & Umberson, 2001; Boethius, 2015; Hearn, 1998; Hydén, 1995). For example, Gottzén (2017) found that men who perpetrate domestic partner abuse against women often try to portray themselves as more morally decent by downplaying and rationalizing their abusive actions when discussing them with others. The strategy described in this section involves approaching abuse by engaging in negotiations with the father from his perspective. The strategy described below involves integrating the perspectives of those exposed to abuse into discussions with the father.

3.2.2 Shifting perspectives

There are often different accounts about the suspected abuse in a CPI, as previous research has also highlighted (Heimer et al., 2018; Mattsson, 2017). Shifting perspectives involves the SWs, in their interactions with the fathers, emphasizing the victims' accounts and attempting to help the fathers understand how the child or/and the mother has experienced and been impacted by the abuse. To facilitate this shift in perspective, the SWs can share both the specific child's experiences and research findings on how children can experience and be impacted by DA.

Annika: I think it becomes emotional when we can use the child's words and hopefully provide an accurate description of the child that the father can recognize (Case A4).

The children's accounts of their experiences of abuse have the potential to impact fathers. This can contribute to the fathers gaining a deeper understanding of the situation, realizing that it is not solely about their relationship with their partner but also about their relationship with the child. This strategy makes the abuse more discussable, because the focus is directed at the child's experience rather than a judgement of the father as an individual. This strategy aligns with the arguments of both Walker (2011) and Gibson (2015) regarding shame-reducing strategies in social work, with the aim being to help the father understand that his actions risk harming the child and thus need to be prevented.

Two other SWs reasoned in a similar way:

Madelene: We could shift the conversation based on what happened [the sequence of abuse described by the child]. The parents managed to describe it. The father was able to reflect to some extent on how it made the daughter feel, that it was not good. […]

Amanda: The father could listen to what the daughter had said. Listen to what we said [about the daughter's experience], and he also expressed that he felt sad hearing her say that, that he wants to understand her (Case B1).

In this CPI, the child had described the father's abuse. The SWs had then relayed the girl's account to the father. Although the father did not agree to define the situation as abuse, he acknowledged the sequence of events, and his daughter's experiences evoked an emotional response in him. By conveying the child's perspective and addressing the ‘good’ parent, the SWs thus managed to maintain cooperation with the father while also addressing the need to prevent recurrence.

Similarly, several SWs described that they also conveyed the mothers' experiences of the abuse and its consequences for the couple's relationship:

Johanna: It becomes important to convey what the mother and children have described and experienced, finding ways to communicate constructively. It's a balancing act to be very clear that it is not okay, that they cannot continue this way. At the same time, you must find ways to build a relationship; otherwise, no change can be achieved (Case A4).

Michael: I have tried to be quite clear with him about the consequences of such behavior. The long-term and short-term consequences, both for the child and the fact that he wants to be with his wife. It has been about drawing his attention to the fact that his actions lead to negative consequences (Case A5).

In this last quote, the SW highlights the negative consequences that the father's actions can have, not only for his relationship with the child but also for his relationship with the child's mother. Emphasizing the consequences for the mother was particularly effective in this case because she had made ultimatums that the abuse had to stop, or she would leave him. So even if the father did not define his actions as abuse, he cooperated and accepted interventions targeted at his abuse. This constitutes a clear example of Scheff's (2003a) discussions about how seeking consensus in describing the problem facilitates a continued relationship and cooperation.

Central to the SWs' strategies in attempting to shift the father's perspective is helping him see the abusive behaviour from the victims' point of view, thus arousing his empathy and creating a mutual understanding of how the abuse may have been experienced by and impacted the victims. The risk is that becoming aware of these consequences may generate feelings of shame, which can harm the cooperation established between the SWs and the father. However, this strategy is typically employed later in the conversation with the father when the bond is assessed to have become more stable, and the risk of shame is perceived as lower. Nonetheless, shame can still arise when the father considers the perspectives of the mother and child. If the bond with the SW is strong enough however, shame may instead be transformed into guilt, opening insights into the father's actions and further strengthening the bond. This may lead the father to experience guilt for his actions rather than shame over who he is (Leith & Baumeister, 1998; Tangney, 1995). The SWs' initial strategies to avoid shame and build a secure bond enable them to later focus on the abuse and attempt to shift the father's perspective in relation to the child and the mother and help the father to better understand the consequences of his behaviour. However, as our description of the SWs' negotiation strategies has shown, this does not automatically mean that the father will agree to define his behaviour as abusive, but he may understand that these actions have negatively affected the child and should not be repeated. This means that the father may become motivated to take responsibility for his actions. Research examining the factors that motivate fathers to change their abusive behaviour has shown that emphasizing the consequences for the children can promote a motivation to desist from abuse and engage in treatment interventions (Areán & Davis, 2009; Bourassa et al., 2017; Broady et al., 2017; Featherstone & Peckover, 2007; Hester & Newman, 2020; Smith & Humphreys, 2019; Stanley et al., 2012). However, this research does not explicitly address the importance of when in the conversation perspective-shifting strategies should be employed to evoke the guilt that will enable the father to develop insights into his actions.

4 DISCUSSION

The aim of this article has been to examine the strategies used by SWs to achieve cooperation with fathers in CPIs initiated due to the father's suspected abuse against the child's mother. The study identified two overarching strategies for achieving cooperation: Securing an initial bond with the father by reducing the tension in CPI proceedings and affirming the father as an individual. The second overarching strategy involves Maintaining the bond while addressing the father's abusive behaviour. SWs do this by negotiating with the father about how the abuse can be labelled and understood and by attempting to get the father to shift perspective and thus realize how the abuse has affected and been experienced by those exposed to his actions. The moral dimension of the SWs' work is toned down, and the emphasis is placed on empathetic and supportive social work, which, according to Scheff and Retzinger's (2001) theories on social bonds, enables the father to feel positively valued as an individual in the eyes of the SWs, despite his morally questionable actions and the moral dimension inherent in CPIs.

However, achieving cooperation with the father while addressing his abusive behaviour was shown to be a challenging balancing act for the SWs. If they are too confrontational when talking about the abuse, they risk losing the cooperation of the father and, consequently, the opportunity to provide him with support to change his behaviour. By using shame-reducing practices and establishing a secure bond with the father, it becomes possible to address the abusive behaviour. The results indicate that the abuse can be addressed via a negotiation with the father about this abuse. Similarly, Brandon et al. (2019) suggests that negotiation between SWs and fathers in CPIs may be needed to engage the fathers.

The study shows that the child's perspective is a key component for achieving cooperation with fathers and has the potential to initiate a process of change. This can be done by addressing the ‘morally good’ father who has acted immorally and making him realize how his actions (even if not defined as abuse) have been experienced by and affected the child. This strategy can be understood based on the notion that there is a distinction between guilt and shame, with the former involving feelings of guilt related to the action rather than shame related to who one is. The strategy can thus be used as a resource in interactions with fathers to get them to address their actions. Other researchers have also suggested that guilt, as opposed to shame, can play a constructive role in interpersonal relationships, opening the way for changes in behaviour (Tangney, 1995). However, Gibson (2015) argues that even guilt-inducing practices risk imposing more shame. The timing and combination of strategies employed in contacts with fathers are central to achieving this distinction, as has been shown in this study. A secure bond must first be established, and even when such a bond is deemed to be in place, the strategies for building and maintaining this secure bond need to be continuously interwoven with perspective-shifting strategies to reassure the father that he is positively valued as an individual and thus avoid inflicting further shame. This is a delicate balance that the SWs need to navigate constantly, first building relationships and then addressing the abusive behaviour in a constructive manner.

At the same time, the SWs' efforts to produce cooperation also risk drawing them into fathers' rationalizations about DA, leading to a renegotiation and minimization of the abuse by comparison with the reports of the victims. The study's results show that the use of negotiation strategies may lead to a shift in focus from DA to other problems, thereby redefining the abuse and making it invisible. Similar results have been found in other studies, such as those by Heimer et al. (2018) and Mattsson (2017). Heimer et al. (2018) found that SWs' efforts to accommodate and engage parents throughout CPIs often resulted in a reframing of the issues initially described in referrals to child protection services to align these with the way parents framed the problem. This can partly be understood as a result of the Family Service Model on which Swedish social services provision is based, which emphasizes voluntariness and a partnership with parents. Concealing DA in this way may contribute to its reproduction (Berggren et al., 2020) and risks leading to misdirected interventions, which can have serious consequences for the victims of DA. The question is how much SWs can reduce tension, affirm and negotiate without negotiating away the problem of DA? SWs risk not only reducing the problem but also their own and their organization's responsibility to work towards getting the father to change his abusive behaviour (SFS, 2001, p.453). This raises the question of whether reaching agreement with the father on certain aspects of abusive acts is sufficient if the alternative is not reaching any form of agreement on abuse, which risks making it impossible to work with the father to prevent further abuse. It is therefore important for SWs to clarify to fathers that the core principles of the social services in a CPI are based on the best interests of the child (SFS, 2001, p. 453). The agency's first priority is the children's safety, which is non-negotiable. This, in turn, guides the collaborations that need to be established within the framework of a CPI.

It is a challenging task for SWs to maintain a balance between achieving cooperation with fathers and maintaining a focus on DA in CPIs. Knowledge of this balance must exist within the field of child protection practice and must include strategies for achieving cooperation with fathers suspected of abuse and an understanding of the associated challenges and opportunities. Further research and methodological development are needed regarding how fathers with abusive behaviour can be reached, not only in CPIs but also in subsequent interventions while still maintaining a focus on the acts of abuse themselves.