Introduction

Across the globe, significant numbers of children are exposed to intimate partner violence (IPV) (Hamby et al., 2010). Experiencing such violence as a child has been found to be associated with a variety of problems during childhood and in later life (e.g. Vu et al., 2016). However, the nature and magnitude of the negative consequences of exposure to IPV vary significantly between children. Some children react strongly to such experiences, while others seem more resilient. To better understand the processes of harm and recovery, researchers have begun to seek explanations of these differences in the children’s understandings of their fathers’ violence.

Children’s appraisals of their fathers’ use of IPV are a robust predictor of behavioral and emotional problems (Figge et al., 2018). For example, appraising the violence as a threat to one’s own safety mediates the risk of problems after having experienced IPV during childhood (Grych et al., 2000; Kim et al., 2008). However, self-blaming appraisals also affect the relationship between childhood experiences of IPV and childhood problems (Grych et al., 2000; Kim et al., 2008). Children’s self-blame may reflect a belief that they are responsible for preventing or ending the violence (Øverlien et al., 2009). In addition, childhood appraisals of the parent who used violence seem important. For example, adults who reported that as a child they believed that the parental IPV happened because the perpetrator was debilitated (e.g. by mental illness or substance abuse) and/or cruel were found to have greater mental health problems and poorer relationship quality in adulthood (Graham-Bermann et al., 2017).

Furthermore, children may adapt to perpetrators’ experiences and construct a positive view of a violent parent with the aim of protecting the bond with that parent, thus enhancing their safety (Lahav et al., 2021; Howell, 2014). In doing so, they may maintain a positive view of the parent, while diminishing or ignoring problematic or disappointing aspects (Lahav et al., 2021; Howell, 2014; Cater, 2007) found that children make sense of their fathers’ violence in relation to general conceptions about fathers and violence in three different ways. These are (1) acknowledging the violent father as one of several kinds of fathers, (2) acknowledging the father’s violent behavior as one part of his multifaceted personality, and (3) contrasting the violence with their father’s relative goodness. DeBoard-Lucas and Grych (2011) found that children’s appraisals of IPV centered on its consequences and their efforts to understand why the fights occurred. Children generally thought that IPV occurs because of the perpetrators’ personal characteristics or lack of control over their anger. However, approximately one third of the children viewed the victims as provoking the aggression, thus attributing some responsibility to both parents. Cater and Sjögren (2016) found that the children’s descriptions of three types of fathers’ IPV (obedience-demanding, chronic and mean, and parenthood-embedded violence) all involved the fact that the perpetrator is the child’s father. Thus, children actively attempt to understand the causes and consequences of IPV and to handle the question of responsibility. Their perceptions and interpretations may have consequences for their health and wellbeing.

Lamb et al. (2018) found that children and young people described elements of reparation as important for their recovery. It was important that their father addressed the past, admitted the harm he had caused, and apologized for it. He also needed to commit to changing his attitudes towards the child/young person and the mother, to cease the violence and rebuild trust. Although many researchers have stressed the importance of fathers explicitly taking responsibility for the violence they have committed (Lamb et al., 2018), there is little research focusing on children’s own perspectives on their fathers’ attitudes about the violence. To develop support for these children, we therefore need a fuller understanding of children’s meaning-making with regard to their experiences of IPV. Specifically we need to know more about how children understand their fathers’ views on responsibility and guilt after having engaged in IPV.

Addressing the knowledge gap described above, the current study investigates children’s sense-making about their fathers’ attitudes about their use of violence against the child’s mother. More specifically we examine various ways in which the children reflect upon whether, and if so how, their father showed regret. A better understanding of such talk and sense-making may be helpful for practitioners working with children and IPV.

Discourses on Family and Violence in a Swedish Context

Understanding and making sense of one’s own father’s violence may be difficult for several reasons. First, in Sweden—the setting of this study—the traditional discourse of family is based on a heteronormative, two-parent family (Andersson & Carlström, 2020). Conceptions of family have however changed in Sweden in recent decades (e.g. Larsson Sjöberg 2012). Still, notions of family are constrained by societal and legal boundaries (Grümüscü et al., 2014). Moreover, the discourse of a “good family” includes harmonious, caring, and supportive relationships (Johnsson & Regnér, 2006). Within the discourse of a “good family,” fathers generally evoke positive expectations and shared responsibility. It is also the norm for parents to have joint custody of children after separation, and most children remain in contact with their father even if he has used IPV (Forssell & Cater, 2015). However, if children are exposed to a father’s IPV, the expected security within the family may be replaced by discomfort, fear, and unpredictability. Thus, discourses on violent or distant fathers, for example, can be understood in contrast to the discourse of the “good family” and the caring father. Accordingly, a father’s violence can be difficult for a child to understand and explain, as he is expected to stand by the child and provide safety and wellbeing (Cater, 2007; Henze-Pedersen, 2021).

Secondly, how violence is understood depends on social, cultural, and historical contexts (Nilsson & Lövkrona, 2015). In some contexts, violent acts can be considered legitimate or be viewed as unproblematic or understandable. In other contexts, violence is constructed as problematic, troubling, or illegal. The prevailing discourse on violence in Sweden stresses that it is wrong to use violence against women and children. Rape within marriage is criminalized since 1965, woman abuse is subject to public prosecution since 1982 and since 2021 exposing children to IPV is criminalized. Within this discourse, fathers’ IPV against mothers and children is accordingly understood as bad and problematic. However, most people actively seek an explanation of the violence, as a way to make it comprehensible. Isdal (2017) argues that this can be understood as a way to avoid positioning the perpetrator as a bad or evil person.

Finally, Dobash and Dobash (2004) note that there is a significant difference between justifying the violence that has taken place, as men who use violence often do (Isdal, 2017; Stanley et al., 2012), and apologizing for it. Offering a justification can be understood as an attempt to legitimize the violence. Apologizing for one’s violent actions, or showing regret in other ways, can instead be understood as a way to excuse the violence without denying its blameworthiness (Dobash & Dobash, 2004). The emotion of regret can be understood as feeling pain or dissatisfaction with yourself because you know, believe, or feel you have done something wrong that you would rather not have done. Against this background, it is vital to further understand children’s reflections on whether, and if so how, their fathers showed regret, in relation to discourses on violence and fatherhood.

Method

This study used data from an evaluation study (Swedish Interventions for Children Who Have Witnessed Violence Against Their Mother – SICVAM; Broberg et al., 2011) that was commissioned by the Swedish National Board of Health and Welfare and conducted during 2008–2010. The SICVAM study aimed to investigate the effectiveness of different treatment programs for child witnesses of IPV. The participants were recruited through eight units offering interventions for children exposed to IPV within the health or social services or associated NGOs, and nine regular units in the health or welfare services. Criteria for inclusion in the study were that (a) the mother had experienced IPV, and (b) the child was aged 3–13 years at inclusion.

Participants

The current study uses interviews led by five interviewers, including the second author of this article. In all, 219 mothers with 315 children were included in the study. Of the 315 children, 133 were nine years of age or older and thus are regarded as old enough to be interviewed. By the age of nine, children generally have experience of communicating with adults and can reason inductively and flexibly (Lutz & Sternberg, 1999). The inclusion procedure began with professionals working at the unit asking the children’s mothers whether they wished to participate in the evaluation. If the mother had sole custody, she was then informed about the possibility of the child participating in interviews. If the mother consented on her child’s behalf, the child was informed and could choose whether to be interviewed. If the parents had joint custody, the mother was asked whether the father should be informed. If she agreed, information was then sent to the father, who could give passive consent to the child’s participation by not opposing it. The child was then informed and could choose whether to be interviewed. In both scenarios, the child received both written and verbal information explaining the purpose of the interview. This was to inform researchers about: (a) the problems children can suffer if they have experienced violence against their mothers, (b) whether the support children receive from different agencies helps, and (c) what mothers and children think about the support they receive.

Many mothers did not want the researchers to contact the father because this might fuel their conflict. We were thus unable to obtain consent for many of the children. Primarily for this reason, only 31 of the children could be interviewed. Of these, 14 were boys and 17 were girls, all of whom are included in this study. At the time of the interview, they were between 10 and 14 years old. The men responsible for the violence against the mothers in this study were biological fathers (22) and stepfathers (8) of the children. In one interview, both the biological father and a stepfather had committed IPV against the mother.

Interviews

The children were interviewed using a semi-structured interview guide that included four main questions. The current study relies primarily on answers to the question: “What do you think [the perpetrator(s)] think(s) about what he/she/they has/have done?” This was followed, when necessary, by one or more of the following sub-questions: “Do you believe [the perpetrator] regrets what he has done?”, and “How does [the perpetrator] show this?” In addition, depending on what the child said, general questions were asked about what the child remembered, who had taken the initiative to talk, and what the child thought and felt about it. The wording and sequence of any follow-up questions depended on each child’s previous answers. This part of the interview lasted up to 25 min.

An interview can be understood as a special situation and social context. It is also important to understand the interview as a joint conversation (Potter & Wetherell, 1987/2014), one that is affected by different expectations, social norms, power relations, and available discourses on violence and fatherhood, for example. Each of these influences the interaction, what is considered possible to talk about, and how the children answer the questions. In this research setting, the prevailing discourse on violence stresses that IPV is problematic and a form of wrong-doing against the mother and child. A violent father is furthermore most often positioned as a person in need of help to resolve his problems with violence and his relationship with the family.

The children’s narratives about regret will also be affected by the questions asked. One of the sub-questions was about regret, but it is important to highlight that in most cases it was the children themselves who raised the issue of the father’s regret. In summary, the children’s perceptions of the interviewer, the interview situation, and the questions asked will all have an effect on the children’s narratives and how they talk about regret and their fathers’ attitudes towards their use of violence.

Ethics

The project was approved by the Gothenburg Regional Ethical Review Board (Dnr. 565-08). Interviewing children about their experiences and understandings of parental violence requires heightened ethical sensitivity. This includes giving careful consideration to issues of consent, whether the knowledge being sought could be obtained from other sources, the risk of unintentional disclosure, and how to deal with information that may signal that the child is at risk (cf. Cater & Överlien, 2014). This study gathered as much data as possible from the mothers and then complemented it by gathering knowledge about their children’s understandings from the children themselves.

One ethical dilemma concerned how to handle consent if the parents had joint custody. On the one hand, children should not have to keep their participation in research a secret from their (violent) fathers. On the other hand, they should not be hindered from participating by parents who may not prioritize their best interests. After consulting with family law experts, we decided that the best way to prioritize the mothers’ and children’s safety over sample size was to let the mothers decide whether the father should be contacted, as described above.

The interview format was designed to minimize the risk of unintentional disclosure. All children were asked if they wanted to participate. Before the interview began they were informed that they could choose not to answer any question, and that they could terminate their participation at any point without giving a reason. They were also informed that we would not pass on any information they disclosed in the interview to their parents or professional contacts with one exception: if we received information indicating that the child might be at risk. This made it possible for us to take action if we received such information.

Discursive Psychology as a Theoretical Framework

To examine various ways in which the children reflect on their fathers’ expressions of regret, we used discursive psychology as a theoretical framework. This starts from the premise that meaning is produced through language, and serves as a way to explain our experiences and argue for a specific standpoint in relation to existing discourses (Potter & Wetherell, 1987/2014). Although there is no uniform definition of discourse, it can be said to constitute the ways in which one speaks about a phenomenon. Discourses influence our understanding and define what counts as a meaningful statement. They constrain and enable what can be said (Wetherell & Potter, 1992). However, discourses need to be understood as complex and changing (Billig, 1987/1996). What sorts of things are possible to talk about can differ from one context to another, depending on the various discourses that are present and prevalent. Wetherell and Potter (1992) also highlight that discourses have “real” substance, affecting our understanding of reality. Narratives are thus understood as constructions and as ways to make sense of thoughts and experiences in a specific context.

Also important for this study are the concepts of interpretative repertoires and (un)troubled subject positions. Interpretative repertoires can be understood as smaller discourses used as resources for storytelling in a specific context and conveying one’s standpoint (Wetherell & Potter, 1992; Wetherell, 1998). More specifically, interpretive repertoires function as rhetorically effective resources that are actively used in meaning-making in relation to a broader discourse (Billig, 1987/1996). Repertoires are not used consistently, however, but rather in cases where one feels a need to reinforce or clarify one’s own reasoning. Looking at which interpretative repertoires are used and how they are used thus enables examination of how people explain and make sense of their experiences, and how meaning is created in relation to a prevailing discourse. Interpretative repertoires can thus be a useful theoretical concept for discursively analyzing and understanding children’s narratives about whether and how their father showed regret.

Within discursive psychology, the subject is understood to play an active role in the processes of meaning making and the way in which we position ourselves or someone else (Wetherell & Potter, 1992) argue that a person is not to be understood as “being” a certain identity; subject positions are instead formulated and expressed in different ways. Subject positions are thus not understood as static or given, but as changing and constructed, albeit limited by available discourses. An untroubled subject position is in agreement with discursively normative expectations and ideals. This differs from a troubled subject position, which can be understood as a position that is considered contradictory, negative, or problematic in some way (Wetherell, 1998).

Analytical Procedure

We have used Potter and Wetherell’s analytical steps as a guide in our analytical process (1987/2014:167–174), starting with the process of coding. The first step in our analytical procedure was thus to read the transcripts of the interviews. The first author then did the initial coding. This can be described as an inductive process of getting to know the material and sorting through the body of text. The focus of this process was understanding each child’s narration in relation to the narrative as a whole. A mind map was used to structure the overall picture of the narratives. This first step is described by Potter and Wetherell as “initial coding,” however it was at this stage that we found that fatherhood and violence were the two prevailing discourses that influenced both the narratives and whether they emphasized regret or lack of regret.

The second step described by Potter and Wetherell (1987/2014) is to sort the narratives in relation to the themes identified during the initial coding. In this step, both authors returned to the interview transcripts to conduct a deeper reading of the narratives, and we jointly discussed the results from the initial coding made by the first author. This step of the analysis also involves looking for patterns of both variability and consistency, as well as analyzing the arguments used to make sense of the topic at hand. During this deeper reading we found that the children were talking about regret in various ways. The theoretical concept of interpretative repertoires was therefore brought to bear, and the first author performed the next step of the analysis.

The interpretative repertoires enabled us to analyze how the children explained and made sense of their experience. As a result we found that the children used three different interpretative repertoires when arguing for their standpoint. The analysis also showed that the children positioned their fathers in different ways – as untroubled or troubled fathers. These findings were then discussed between the two authors, after which we returned to the interview transcripts in a back-and-forth process. This resulted in our finding not only the subject positions of an untroubled regretful or a troubled non-regretful father, but the third subject position of an inconsistent father.

The final two steps, the validation and the report, involved focusing on what things were seen as compatible or different. We accordingly analyzed how the interpretative repertoires were used and how they allowed for or restricted certain narratives and subject positions. This was discussed between the authors to get a better understanding of what the different subject positions and interpretative repertoires could tell us. We selected extracts that clearly illustrate the children’s arguments and also illuminate the various ways in which the children talk about regret and their fathers’ attitudes towards violence. The analysis was performed directly on the Swedish interview transcripts and the excerpts were later translated into English.

Findings

The aim of the study was to examine the children’s various ways of reflecting on whether, and if so how, their fathers showed regret. Their narratives can be understood as influenced by two prevailing discourses, on fatherhood and on violence, which in this context construct fathers as good and caring and violence as problematic, bad, and something one should regret. The analysis reveals that some children argued that their father did regret the violence he had inflicted on their mother, while others said they did not believe their father regretted his use of violence, and instead positioned him as inconsistent or not regretful. When the children argued for what they believed in, the analysis also reveals the use of three different interpretative repertoires. These were: (a) a repertoire of communication, (b) a repertoire of feelings, and (c) a repertoire of actions. The three repertoires can be understood as resources upon which the children could draw when they needed to reinforce or clarify their reasoning so that their argument did not appear incomprehensible.

In the following, we illustrate the ways in which the children talked about their fathers as regretful, inconsistent, or not regretful, and how the children used the three repertoires in relation to the discourses on violence and fatherhood.

Children’s talk About a Regretful Father

We start by showing how the children’s talk about regret was linked with the prevailing discourse on violence and the idea that violence is bad and problematic. More precisely, we look at how one of the children argued that her father must regret his use of violence simply because violence is wrong.

Interviewer: What do you think he thinks about what he did?

Girl 1: He regrets it, I think (Mm).

Interviewer: How does he show it then?

Girl 1:. . Uh, what do you mean?

Interviewer: If you think that, like, you say he regrets it, why do you think he regrets it, then?

Girl 1: Mm, because it’s wrong to commit a crime (Mm).

Answering the question about what she believed her father thought about his use of violence against her mother, this girl gave prominence to the feeling of regret. She said she believed he regretted his actions. When asked to elaborate further on this, the girl hesitated, but then argued that her father regretted it because violence is wrong, and is a crime. Hence, the discursive understanding of violence as bad and something one should regret made it possible for this girl to give meaning to her belief that her father regretted his actions and to explain why she believed this.

Positioning one’s father as regretful may not be easy, however. The interviews indicate that this is experienced as a subject position that needs to be explained, even though the discourse on violence emphasizes that violence is bad and problematic and thus is something one should regret. Hence, to be able to position their fathers as regretful, some children in the study used interpretative repertoires as resources for making sense of their fathers’ attitudes. One example is shown in the following excerpt from the interview with one of the girls using the repertoire of communication.

Interviewer: What do you think your dad thinks about what he did?

Girl 2: I think he thinks it’s bad and that he regrets it (Mm).

Interviewer: What, uh, how does he show that he regrets it?

Girl 2: He… I actually don’t know, you can’t see it in him (No). I just think, (Yes but) that he did say it once (Mm).

Interviewer: When did he say that?

Girl 2: I don’t Remember!

Interviewer: Do you remember what he said?

Girl 2: No!

Firstly, this extract shows that the girl believed that her father did regret the violence he had inflicted on her mother. As in the example above, the discourse on violence made it possible for this girl to argue that violence is bad and something that one should regret. Then, when asked to explain why she believed that her father regretted his use of violence, the girl said that the regret was not visible, but she then used a repertoire of communication and highlighted that he once said that he regretted his actions. Even though she could not remember when this was, or what her father said, the use of the repertoire of communication made it possible for the girl to argue that her father did say something to show his regret. Thus, the girl could position her father as regretful, which can be understood as an untroubled subject position despite his use of violence.

Other children also described their fathers as regretful, but the example below shows how rather than using a repertoire of communication, one of the boys used a repertoire of feelings to argue why he believed his father did regret his use of violence.

Interviewer: Do you think your dad regrets what he did?

Boy 1: Yeah! I think he regrets it a lot.

Interviewer: How does he show it? Do you have an example?

Boy 1: He may look a little sad afterwards.

Interviewer: You can see it in him?

Boy 1: Yeeah…

Interviewer: Are there any…

Boy 1: He kind of stands there mumbling and then says “uh” and then he has a very sad look, I think.

In the above excerpt, the boy used the repertoire of feelings, emphasizing an appearance of sadness to explain his belief that his father regretted his use of violence. The boy said that his father mumbled and looked sad. As with the repertoire of communication, the use of the repertoire of feelings and the child’s talk about his father’s sadness can be understood as a way to make sense of the violent actions by viewing them as mistakes he regrets. This creates makes it possible to position his father as a regretful father. Another example was given by the girl in the extract below, but she used the repertoire of action instead:

Interviewer: Mm, what do you think that, that he kind of, he thinks about what he’s done?

Girl 3: I think he regrets what he’s done. (Mm). That he shouldn’t, doesn’t want to do it anymore, that he doesn’t want to do (Mm) it any longer.

Interviewer: How does he show then, that he thinks that way? Or why…?

Girl 3: He doesn’t, sort of. He ju-, is just doing… like treatment and everything (Mm) drugs and stuff and he becomes nice (Yes).

As in the previous excerpts, this girl talked about her father’s regret. She also said that she believed her father did not want to use violence again. By using the repertoire of actions, she not only argued that her father would not use violence again but also that he was taking action to change by getting treatment, and thus would become a kind (untroubled) person.

These examples illustrate how the children in this study talked about and made sense of their fathers’ attitudes regarding the violence against their mothers by positioning them as regretful. The examples reveal that the children used three different repertoires to strengthen their arguments and clarify for the interviewer why they believed their fathers regretted their actions. They were thereby able to position their fathers as regretful, which is an untroubled subject position, in relation to the discourse on fatherhood and the prevailing idea that fathers are supposed to act in a caring way and provide safety. However, the children’s arguments and their use of the various repertoires do not necessarily diminish the meaning of violence as bad or wrong. Instead the repertoires can be understood as resources that enable the children to position their (violent) fathers as regretful.

Children’s talk About an Inconsistent father—regretful or not?

Not all of the children positioned their fathers as regretful, however. Some talked in a more complex way, expressing doubt and ambivalence about whether their father regretted the violence. In this section, we show how the subject position of an inconsistent father was constructed in one boy’s narrative.

Interviewer: And if you think about, like, your dad and what he did, do you think he regrets what he did?

Boy 2: Yes, he says so, but I don’t know if I really believe him (No). No that’s, well, I don’t know.

Interviewer: Mm, he says that, but you don’t really believe him?

Boy 2: No, I don’t know what to think, he always says so, and then he does it again and (Mm). I don’t know if I should believe it (No, okay).

This boy began answering the question about regret by saying that his father has said he regrets using violence. He then stated that he did not believe his father’s words, or at least he was not sure that his father was telling the truth. Thus, when explaining his thoughts, he used a repertoire of communication to explain that his father said he was sorry. However, by using the repertoire of actions to highlight that his father had used violence again, the boy was able to argue that it was difficult to know what to think. In this way, although the boy talked about his father communicating regret, at the same time he could argue that he did not believe his father, because he repeated the violence after saying he was sorry.

The repertoires are thus used in multiple ways, and accordingly need to be understood as resources for storytelling and ways to argue for one’s standpoint in more complex ways. Accordingly, the discourses on violence constrain the possibility of positioning one’s father as regretful when he repeatedly uses violence. What remains is the position of a violent and possibly untruthful father who might regret his actions but continues to be violent. This inconsistent and contradictory father can be understood as occupying a troubled subject position in relation to the prevailing discourses on fatherhood and violence.

Children’s talk About a non-regretful Father

There are also examples in which the children talked about and positioned their fathers as different from other people or troubled, and as therefore not regretting the violence. In the first example below, a boy characterized his stepfather’s thoughts about the violence in terms of a disease.

Interviewer: Uh… what do you think he… thinks and feels about what he’s done?

Boy 3:. .. Uh, I think he doesn’t think he did anything wrong because he has kind of a disease (Mm).

Interviewer: Do you think he regrets anything?

Boy 3: No.

This boy argued that his stepfather did not think he did anything wrong. But, in order to make this argument comprehensible in relation to the discourse on violence stressing that violence is bad, he referred to a disease. In this way, the boy was able to describe why he believed his stepfather did not regret his use of violence. The boy was accordingly able to position his stepfather as a violent, ill, and non-regretful father. This can be understood as a troubled subject position. However, it does not necessarily indicate a bad or mean stepfather, but rather an ill father, even though his use of violence was constructed as wrong.

Below, we give another example which shows how a girl used the repertoire of feelings to talk about her violent father and argue for positioning him as a non-regretful father.

Interviewer: What do you think dad thinks about what he’s done?

Girl 4: He really should regret it, but… I don’t really know how he thinks because he’s a bit… different, if you can call it that.

Interviewer: What do you mean?

Girl 4: Well, he sort of thinks violence is fun. You see it when he watches movies and such, because he laughs when someone dies or something, and it’s not even funny.

Interviewer: You don’t think it’s funny?

Girl 4: No. But he kind of laughs at it, so, I don’t know how he perceives violence and such.

This excerpt shows that, in line with the discourse on violence and the construction of violence as bad, the girl argued that violence was something that her father was supposed to regret. But rather than positioning him as a regretful father, she described him as different, as not viewing violence in the same way as other people. Using the repertoire of feelings, the girl argued that her father laughed and considered violence amusing rather than problematic. Thus, in this case, the repertoire of feelings was used to argue that her father showed or felt the wrong kind of emotion in relation to the use of violence, or violence in general. The girl’s narrative thus illustrates that violence is assumed to be something a father is supposed to regret if he has subjected the mother to it. Hence, the narrative illustrates how the father’s laughter was constructed as deviant in relation to violence, and also that the girl positioned her father as different from other people. He was a non-regretful father, which can be understood as a troubled subject position, but not necessarily a bad father. Rather, it was the father’s understanding of violence as something funny that was constructed as the problem in the narrative.

In the two examples above, the children were able to explain why their fathers did not regret their use of violence. However, instead of positioning their fathers as bad fathers, they could argue that their fathers were different from other people or ill. In their talk, the children could thus argue for an understanding of violence as bad and something that fathers are not supposed to use. Hence, the children were making a complex argument for why their violent fathers did not regret their use of violence. They did so by positioning them, not as bad fathers, but rather as different, ill, and troubled individuals, which in these two examples is explained by illness and a non-normative understanding of violence, respectively.

Discussion

Our findings support previous research indicating that children make sense of their fathers’ attitudes towards violence in varied ways. They show that the ways in which the children interpreted their fathers’ violence, and their attitudes and behaviors afterwards, are related to two discourses: one about violence, with the assumption that it is bad to use violence, and one about fatherhood, which understands fathers as good, protective, and caring. The main finding is that the children positioned their fathers as either regretful, not regretful, or inconsistent. Also, in order to be able to argue for their beliefs, the children sometimes used one or more of the three interpretative repertoires we found: a repertoire of communication, a repertoire of feelings, and a repertoire of actions.

A regretful father can be understood as an untroubled subject position in line with the prevalent discourses on violence and fatherhood in the interview context. Violence is understood as bad and therefore something that a caring father does regret. As described by Wetherell (1998), an untroubled subject position is in line with discursively normative expectations and ideals, in contrast to a troubled position, which is considered contradictory, negative, or problematic in some way. Arguing that one’s father regrets his violence enables the creation of an untroubled (or less troubled) subject position. This makes it possible for a child to position their father as regretful and untroubled parent, by differentiating between the bad violent acts and the possibly good, or at least not bad, regretful father. Further, the children’s narratives can be understood not only as ways to explain their fathers’ actions, but also as related to an understanding of regret as a way of feeling sorry about something one has done and would rather not have done (Dobash & Dobash, 2004; Lamb et al., 2018) also argue that for children, one important component of reparation with their fathers is that the fathers admit to the harm they have caused and apologize for it. This is also important in relation to children’s wellbeing and their behavioral and emotional problems (Figge et al., 2018). Hence, reparation might be easier for children who understand their fathers as regretful; which is shown, for example, by the fathers admitting the harm they have caused or changing their behavior.

However, the subject position of a non-regretful father can be understood as a troubled subject position, in relation to the discourses on violence and fatherhood, but without this necessarily meaning that the father is cruel or bad. By positioning one’s father as ill or as having a different (i.e. problematic) understanding of violence, it seems possible to negotiate an understanding of the violent, non-regretful father. By describing him as ill or different from others, the child instead constructed the position of a non-regretful and violent father as a troubled father whose understanding or perspective on violence differs from the prevalent discourse on violence within both the interview situation and the broader Swedish context, namely that violence against women and children is bad. Perhaps these children were differentiating between a non-regretful violent bad father and a non-regretful ill or different violent father who needs help. Previous research has further shown that cases where children believe that the perpetrator was suffering from mental illness or substance abuse and/or was cruel and took pleasure in violence are associated with greater mental health problems and poorer relationship quality in adulthood (Graham-Bermann et al., 2017). It might be easier, however, for a child to position their father as an ill or different father than as a bad and violent one, and thus to maintain a positive or untroubled view of the parent while avoiding problematic or disappointing aspects (cf. Howell, 2014). As shown by Cater (2007), children also try to distance the violence from their fathers’ relative goodness and thus to position them as good fathers despite their actions.

However, some of the children also positioned their fathers as inconsistent, because they perceived their fathers’ attitudes as contradictory (cf. Lamb et al., 2018). The inconsistent subject position, where the father is described as both showing regret and at the same time repeatedly using violence, is not as common in our participants’ narratives as the other subject positions. However, the example given here indicates that the inconsistent subject position is a troubled one in relation to discourses on violence and fatherhood. It is troubled both in relation to the violent acts and because the father may be understood as not telling the truth about regretting his violence. Taking responsibility and apologizing, or showing regret in other ways, are commonly expected in the Swedish context (Isdal, 2017:177–180). Continuing to use violence can thus be understood as problematic and as transgressing normative expectations of how violence should be handled, especially as your father is expected to stand by you and provide safety and wellbeing (Cater, 2007; Henze-Pedersen, 2021). Hence, our findings add nuance to previous findings by showing that some children either acknowledge their violent father as a different kind of father, or argue that his violent behavior is something he regrets or is part of his multifaceted, and sometimes problematic, personality (cf. Cater, 2007; Henze-Pedersen, 2021). Altogether, these results thus inform us about what living with an adult (biological parent or not) who has subjected one’s mother to violence might be like for children.

Lastly, we would also like to highlight the children’s use of the three interpretative repertoires of communication, feelings, and actions. To be able to argue for the subject positions discussed above—a regretful, an inconsistent, or a non-regretful father—our analysis reveals that the children used the repertoires as resources for clarifying their arguments about why they believed or did not believe their fathers regretted their violence. It also indicates that the discourses on fatherhood and violence constrain and enable what can be said, and that they have “real” substance, affecting how children make sense of their fathers’ violence (cf. Wetherell & Potter, 1992).

Limitations and Suggestions for Future Studies

As with many other qualitative studies, a key limitation is the small number of interviews, and there is a risk of selection bias. However, compared to most other studies based on interviews with child witnesses of IPV, the sample is relatively large. Also, interviews can only give a glimpse of the meanings that people give to complex phenomena. What the children expressed in these interviews is likely to have been colored by what other people have said and done, for example their fathers, mothers, and professionals from various organizations. It is also important to highlight that the children’s narratives are constructed within a particular interview context. In another context, other discourses and constructions of violence and fatherhood might be prevalent, enabling or constraining the narratives. In this study, however, these various influences are not seen as methodological problems, but rather as highlighting the intrapersonal dimensions of discourses. Nevertheless, it would have been valuable to ask about the children’s own feelings in relation to how they understood their fathers’ violence and whether this had affected their relationships with their fathers. This is an aspect that future studies ought to explore.

The sample size and the nature of the interviews did not allow us to distinguish biological fathers from stepfathers in the analysis. Whether the likelihood of children acknowledging their fathers’ regret depends on whether he is the biological father, whether they live with or have contact with their biological father, or the quality of their relationship would be an interesting topic for future research. Also, it would be valuable to study how children make sense of their fathers’ minimizations of the violence and whether, and if so how, their appraisal of their father’s regret changes if their father participates in an intervention or treatment program.

A further potential limitation is that these interviews were conducted ten years ago. Although we find little reason to believe that this has affected the findings, it does raise the ethical question of how long after collection interview data can still be used. In this case, we judged that the benefit of being able to further understand children’s perspectives on IPV was significant enough to revisit the interview data. This is especially apparent if we consider the emotional costs of collecting new data from children with such experiences. Although a study such as this one has its limitations, our findings elucidate some important aspects of the challenges that child witnesses of IPV can face.

Implications for Policy and Practice

Keeping the study’s limitations in mind, our findings have some important potential implications for policy and practice. Overall, the study demonstrates the importance that social workers and other practitioners keep in mind the variation in children’s experiences and narratives of IPV. Specifically, our findings indicate a need, when working with children who have been exposed to IPV, to problematize the normative ideal regarding parenthood and notions of a caring father who provides safety and wellbeing. If this knowledge about how discourses of violence and fatherhood affect children’s narratives is overlooked, there is a risk that the consequences of the fathers’ use of IPV will not be given enough attention.

In addition, we consider it crucial that social work, therapy, and interventions for these children acknowledge not only the violence, but also what the father’s attitude towards it means for the child in a longer-term perspective. Fathers’ expressions of regret are of great value for children’s well-being and reparation. This is especially true because, in contrast to justifications of violence, regret can be understood as a way of describing the regrettable action without removing the blame for the violence itself (Dobash & Dobash, 2004). This is particularly important in at least two types of situation: (1) when deciding whether, and if so how, a parent who has used violence but does not acknowledge it should be involved in the support given to the child, and (2) when deciding whether a child should continue to live with or have regular contact with a father who has used violence.

Overall, our results suggest that practitioners and service providers who come into contact with children exposed to IPV and have responsibility for their safety and wellbeing must understand the discourses on families and violence, and the ways in which most people try to seek an explanation that makes the violence comprehensible (Isdal, 2017), as these affect children’s descriptions of fathers who have subjected the mother to violence.

Concluding Remarks

Our results indicate that the prevalent discourses on violence and fatherhood influence children’s talk about their fathers’ violence, and we understand the normative expectations about fathers as caring and violence as bad to be constraining, and hence as influencing what can be constructed as meaningful and appropriate within the Swedish context. The interpretative repertoires of communication, feelings, and actions are thus understood to be helpful resources in the children’s narratives, making it possible to position one’s father as a regretful father, and thus as an untroubled and possibly good and caring father who regrets his violence against the mother. A position as an inconsistent or non-regretful violent father appears instead as a troubled subject position, which the children try to avoid or negotiate. They do so by using the different repertoires and positioning the violent father as a different kind of troubled person, and thus not necessarily bad or mean.