‘Ordinary decent domestic violence’: A discursive analysis of family law judges’ interviews

Abstract

This study examined judges’ constructions of the ‘best interests of the child’ in child custody and access arraignments where there were allegations of domestic violence within the context of an interview. Using interviews with six Irish District Court judges, a micro-structural discourse analysis enabled the identification of socio-cultural discourses, scientific knowledge and judges’ own values and beliefs biases about custody arraignments in cases of domestic violence. Judges’ discourses were shaped by an idealisation of the nuclear family unit which resulted in a pro-access philosophy (Theme 1). The knowledge that domestic violence had occurred challenged this ideology and, to rhetorically manage this dilemma, judges’ talk normalised or trivialised abusive parents’ behaviour, which rendered domestic violence irrelevant to child custody and access (Theme 2). Mothers who alleged domestic violence when they disputed contact between fathers and their children were pathologised through talk (Theme 3). It is recommended that systems be put in place, including judicial training, to facilitate judges in their decision-making process in this highly discretionary and complex area of the law.

Keywords

Child custody discursive analysis domestic violence family law judges interviews

Introduction

Judicial systems throughout the Western world function to interpret and implement the rules of the land. Judges, instated by governments, are charged with carrying out this process in a fair, impartial and efficient manner. However, a large body of research on social cognitive processes indicates that we all make use of cognitive schemas and heuristics in daily decision-making. In order to reduce the complex social world into something more manageable to interpret, we not only use easily accessible heuristics – shortcuts which reduce our cognitive load – but we also exhibit a tendency to pay increased attention to information which is in line with our own values and beliefs (Gilovich et al., 2002). This process can be seen as part of the explanation for the gap between laws on the books and law in action, and adds to our understanding of the relationship between formal law and everyday enactment (Krieger, 2000: 478). The aim of this study, then, was to identify if there was evidence during interviews of judges incorporating such cognitive schemas and heuristics into their constructions of the best interest of the child.

Article 9 of the UN Convention on the Rights of the Child (General Assembly of United Nations, 1989) specifies ‘that a child should not be separated from his or her parents’ unless ‘such separation is necessary for the best interests of the child’. The socio-legal concept ‘best interest of the child’ is, however, subject to interpretation (Conner, 2009), and this places power and authority in the hands of judges, as contested child custody and access adjudications are made by a judge within a family law court. The complex nature of custody and access adjudication tends to be incompatible with formal legal rules (Schneider, 1991), but is guided by legislation and dependent on the jurisdiction.

Thus, within the family law courts in Ireland where this study was undertaken, adjudications of contested cases are at the sole discretion of the presiding judge. Berger (2009) suggests that cognitive schemas may influence judges’ analysis when presented with evidence in relation to individual children. Furthermore, given that family law is particularly bound to ‘culturally embedded stories’ (Berger, 2009: 259), judges’ own values may act as a framework for their decision-making process in this area. This is particularly significant in the context of domestic violence (DV), where adjudications have important consequences for children in that they may provide protection or may alternatively place a child at risk (Conner, 2009). For example, Zeoli et al. (2013) in interviews with mothers identified a belief that family courts failed to provide protection for their children. Moreover, Holt (2011) identified Irish children’s frustration at the assumption that contact with their abusive father was in their best interest.

Research in this area is sensitive and difficult. However, discourse analysis (DA) has been used previously to good effect. Using texts from trial judgements of sexual assault cases, Coates et al. (1994) and Coates and Wade (2004) found that sexual violence was reformulated discursively as non-deliberate, non-violent acts. Furthermore, the failure of discourse to recognise the power dynamics between victims and perpetrators resulted in blame being placed with the victim. And more worryingly, use of these discursive strategies within the trial judgements was directly related to reduced sentences for the convicted perpetrator (Coates and Wade, 2004). In a similar vein, Ehrlich (2007) examined judicial decisions of the trial, appeal and Supreme Court case of the one criminal sexual assault case in Canada. This analysis demonstrated the influence of cultural gender norms around consensual sex at these multiple judicial levels, which had real consequences in the acquittal or conviction of sexual perpetrators. In fact, these gendered assumptions ‘were so powerful’ that they were interfering with Canadian law (Ehrlich, 2007: 471)

These DAs of legal adjudications enable a direct link between the use of discursive strategies employed and sentencing/judicial decisions (Cotes and Wade, 2004). However, in this study, we use interviews as an alternative source of data. In these contexts, judges are assured of anonymity and unbound by the requirements of the legal system and thus are free to speak about their views in general. Therefore, analysis of somewhat banal discursive strategies within this data corpus can be seen as an expedient means of unearthing the implicit and explicit cultural and gendered assumptions made by powerful institutional representatives about the interests of children in situations of DV. Indeed, we argue that our current focus on judges – an elite, influential group – is important as qualitative social science has a well-established record of studying powerless and marginalised groups, which has often been seen as giving voice to the voiceless (Duke, 2002). This is extremely important, but it means there is perhaps less of a tradition of studying the powerful, despite the fact that to study the powerful is to demystify power and enable challenges to privilege.

Gaining access to elites is, however, more difficult and as these elite groups do not have the same sense of being voiceless, they may be less motivated to participate in research. As a consequence, studies of the powerful, though equally worthy, are scarcer. A few examples in this area do exist nonetheless. Pond and Morgan (2008) undertook a DA of interviews with lawyers, where they identified discourses which portrayed the need to protect children via supervised access. However, they also identified discourses which portrayed mothers who alleged DV as both obstructive and ‘misusing the system’ to their own gain. Pond and Morgan argued that such discourses resulted in the reversal of perpetrator/victim status. Similarly, McCarthy et al. (2000) found that the discourses of British court-appointed mediators were shaped by what they saw as a ‘non-negotiable moral obligation’ for mothers to facilitate good quality contact with fathers, post-separation. Taken together, these studies suggest that the heuristic in operation within the judicial system may be one that is built on gendered expectations of parents and parenting behaviour, rather than justice or rights.

To date, no studies of direct interviews with judges on their understanding of child custody in DV cases are available in the published literature. As such, the aim of this study was to identify how judges talked through interviews about the ‘best interests of the child’ in child custody and access cases where there were potential risks to child welfare, posed by the presence of DV within their families. This research, then, adds significantly to the scant research in this area.

DV context

Extensive research has established the negative impact that exposure to DV has on children (Bancroft et al., 2011). The relevance of DV in child custody adjudications has also been highlighted in a growing body of research within the areas of law, psychology and sociology (Eriksson, 2011; Jaffe et al., 2003; Shea Hart, 2011). In DV situations, conflict can play out in separation and access proceedings, which can even be a site of further coercive abuse (Jaffe et al., 2003). Furthermore, the manipulation and control which encompass DV make it very difficult to prove within the courtroom setting, with judges often having to determine the credibility of parents as they present conflicting accounts of abusive events which occurred within the privacy of their own home (Jaffe et al., 2003).

Method

There are 40 District Court judges currently presiding over private family law cases in Ireland (Courts Services, 2012). Of these, six (15% of the total national group) were recruited to this study through a general request through the court email system (1), personal contacts (2) and snowballing from initial respondents (3). Ethical approval was obtained from the Education and Health Sciences Research Ethics Committee at the University of Limerick. Five of the judges presided over private family law cases at the time of interviews. Time on the bench ranged from less than 6 months to 29 years. Four women and two men were interviewed. All participation was voluntary; all judges were made aware in the initial invitation letter and on the informed consent form that the major interest in this study centred on their private custody and access adjudications in the context of DV.

An extensive literature review informed the interview schedule (see Appendix 1). However, interviews were semi-structured to enable flexibility, and to allow the researcher to probe salient areas as they arose. In line with this, questions were open-ended and designed to explicate information on judges’ incorporation of key scientific knowledge in their construction of the best interest of the child. Due to the extensive gender symmetry debate within the literature and popular media, the interviewer did not refer to gender in questioning undertaken, enabling judges themselves to construct the gender of perpetrators/victims of DV within talk. Each interview lasted approximately 1 hour, and all but one was digitally audio-recorded. Judge 1 declined to have her interview audio-taped so was excluded from in-depth analysis, but her interview informed theme formation and interpretation. Extensive notes were taken subsequent to the interviews. Digitally recorded interviews were transcribed using an abridged form of the Jeffersonian convention (Jefferson, 1984; see Appendix 1 for Glossary). The first author (C.M.N.) carried out all interviews and transcriptions; this facilitated complete immersion in the data set.

Analysis

DA lends itself to an in-depth analysis of elites. Starks and Trinidad (2007) identify that qualitative research using methods such as DA can produce rich data from relatively small data sets (Talbot and Quayle, 2010). Moreover, one way to unveil factors that influence judges’ navigation of this complex decision-making process is using a discourse analytic approach (McMullen, 2012). The identification of such discursive devices in talk can illustrate how judges work-up the factuality of their discourse, while managing their own position and accountability. This therefore allows us to identify the incorporation of socio-cultural discourses and scientific knowledge in the judges’ discourse, but also allows us to see how judges’ own values and beliefs are used, by them, to navigate these difficult decisions. Transcripts were read in their entirety several times to gain an overview of the entire data set. Each judge’s transcript was printed out on a different colour of paper to assist with the analytic process. The data set was then systematically coded into 10 main categories which remained close to the data. For example, some of the codes included talk about: children, parents, decision-making or resources. These extracts were manually cut out and organised according to codes. Memos were kept throughout the analytical process, to track patterns, interactions between these patterns and authors’ interpretations. This led to the development of themes, identified as sets of data which ‘captured’ an element deemed important to the research question (Braun and Clarke, 2006: 88). Extensive reflexivity and triangulation between authors was undertaken to minimise the impact of individual beliefs and values on the interpretation.

The analysis process was iterative; three major themes were generated which provided a ‘concise, logical and coherent’ account of the data (Braun and Clarke, 2006: 93). Extracts which best exemplified these dominant themes were then analysed using resources from Discursive Psychology (MacMartin and Wood, 2005; Potter, 1996; Wood and MacMartin, 2007). The micro-structural approach included the identification of discursive devices employed, their local function and contribution to meaning. As language use is most ‘accurately understood’ when situated in context (Cotes, 2004: 501) and since the interviewer’s questions form an integral part of that context, each extract is accompanied either by the direct question (when immediately preceding the extract) or a reference to the preceding question which is contained in the interview schedule.

Findings

Analysis of all transcripts, as outlined above, resulted in the identification of three broad interrelated themes. Judges’ talk which oriented towards an idealisation of the nuclear family and constructed a presumption of post-separation parental contact formed Theme 1, Pro-access philosophy. Judges’ talk which constructed DV as singular incidents which ended on the separation of the parties, or which minimised, trivialised or ignored abusive behaviour, thus rendering DV as irrelevant to child custody and access decisions, formed Theme 2, Irrelevance of DV to custody. Finally, judges’ talk which placed blame with mothers who contested contact with fathers or/and ignored abusive behaviour formed Theme 3, Problematic mothers, invisible fathers. Below, each of these themes, which were evident across the dataset, are presented in turn and accompanied by an exemplar quote, for which the microanalysis is detailed.

Pro-access philosophy

Judge 2’s transcript contained detailed discourses on the ‘horrendous damage’ to both women and children as a direct result of DV. Yet despite this, when the judge was requested to talk about the role of a father in a child’s life, talk in the response referred to both parents. This had the effect of decontextualising the pre-established context of abuse and enabling the idealisation of the nuclear family. This justified a pro-access philosophy, as follows:

Extract 1: Absolutely essential for a child to know both parents.

1. Int What would your opinions be on (.) the role of the father in a child’s life?

2. Judge 2 I absol (.) I (.) I (.) I really feel strongly about it (.) every child male or female (.)

3. needs a mother (.) and a father

4. (…)

5. Judge 2 it is absolutely essential (.) for a child to know that (.) they have a dad and a mam

6. in their life (.) and they have a dad and a mam who wants to be involved in their

7. lives

Here, an idealisation of the nuclear family unit was instigated by a discourse which established and assumed the need for both parents to be part of a child’s life, post-separation. Judge 2 commences with a mental state avowal, ‘a description of private thoughts and feelings’ (Edwards and Potter, 2005: 247): ‘I really feel strongly about it’. Wittgenstein (1958) suggest that avowals of feelings are difficult to challenge, thus rendering an alternative argument as less feasible. The judge also utilises a discursive device referred to as an extreme case formulation (Pomerantz, 1986); this functions to work up a pervasive argument by invoking a maximal or minimal quantity on a relevant dimension. Here, ‘every’ (Line 2) is all inclusive; it leaves no room to omit any child from any background or circumstance. This maximisation is future emphasised by the addition of both genders ‘male and female’. This discursive device also includes words which modify a description (Potter, 1996), as in ‘absolutely’ in Line 7 which also has a maximising effect. By using extreme case formulation, Judge 2’s talk communicates effectively that it is not ‘just important’ but that ‘it is absolutely essential’ (emphasised) that each parent ‘wants to be involved in their’ child’s life. In so doing, she portrays a normative assumption that the presence of two motivated and interested parents are essential to a child’s well-being.

Irrelevance of DV to custody

Extract 2: Ordinary decent domestic violence.

In response to Q2 of interview schedule (see Appendix 1)

1. Judge 3 So while the parties might have been up at high doe¹ (.) at the point of

2. separation because of a domestic violence situation (.) a flare up (.) a striking or

3. whatever (.) once the separation takes place it tends to be (.) more (.) amh (.) amh (.)

4. an issue of working out the mechanism as it were

5. Int Okay

6. Judge 3 of the custody and access (…)

7. Judge 3 You have these mechanism or these dynamic where you’re into hearing these

8. issues of abuse all over again(.) now when I say abuse I am talking about a striking

9. I’m not talking about sexual abuse (.) that that falls into an entirely seperate and

10. more severe category

11. Int Okay

12. Judge 3 so you have to differentiate between behaviours

13. Int Yeah

14. Judge 3 I’m talking about (.) amh (.) I won’t say ordinary decent (2) domestic violence

15. because obviously it can be soul destroying

The concept of DV of course challenged the idealised notion of the nuclear family unit. In their interviews, to manage their dilemma, judges’ talk oriented towards a normalisation, trivialisation and minimisation of DV. Lines 1–6 above are an example of script formation (Edwards, 1994); the discourse formulates a story which constitutes an event as commonplace, ordinary (Potter, 1996: 197). By formulating a script, Judge 3 portrays the presented scenario as factual, thus rejecting alternative versions, but also downplays the consequences of DV on its victims. Within this script, ‘a domestic violence situation (.) a flare up (.) a striking or whatever’, the description is passive; there is no agency attributed to the behaviour, for example, as in ‘he struck her’ as opposed to ‘a striking’. Therefore, the judge’s talk constructs DV as both mutual and as occurring in singular instances, as opposed to a regular pattern of behaviours perpetrated by one partner on the other, as is more often the case (Jaffe et al., 2003). Thus, talk orients towards DV as a circumscribed or acute phenomenon within a relationship rather than a chronic or on-going problem, thus trivialising the horrors of ‘everyday’ DV. The judge’s use of ‘whatever’ can be seen to portray a very minimalist construction of DV. Indeed, his talk orients towards an overriding concern to provide access to the ‘non-live-in parent’ (the father), to the extent that it is only ‘an issue of working out the mechanics’ in Lines 3 and 4. Indeed, talk which portrays DV as a minimalist concern lacking consequences makes it less relevant to decisions around child custody.

The common parlance term ‘sexual abuse’ (Line 10) is generally used to refer to rape or abuse of a child, not an adult. Thus, in Lines 8–11, talk constructs direct abuse on children, yet the need ‘to differentiate between behaviours’ in Line 13 orients towards an acceptance of DV through talk, as it is only sexual abuse on children which is portrayed as relevant to custody and access adjudication. This minimisation and normalisation of DV is reinforced by Judge 3’s description ‘ordinary decent (2) domestic violence’. There is a significant pause of 2 seconds between ‘ordinary decent’ and ‘domestic violence’. Pauses indicate that care is being taken in the choice of words (Potter, 1996). The judge still chooses to place these two incompatible terms together, but rhetorically manages the palpable inappropriateness of these juxtaposed terms with the addition of ‘because obviously it can be soul destroying’. The judge also engages in the management of the contentious statement, possibly to maintain his neutrality and objectivity, by including the disclaimer ‘I won’t say’. A disclaimer is ‘a verbal device that anticipates and rejects, potentially negative attributions’ (Willig, 2008: 103). Extract 2 illustrates how the judge’s talk oriented towards a normalisation and minimisation of the complex phenomenon of DV, which is thus categorised as irrelevant to child access. This process may be seen to rhetorically manage the challenge which DV places on the idealisation of the nuclear family.

Problematic mothers, invisible fathers

Evident throughout the interviews was the tendency for talk to sidestep both abusive behaviour and the parenting ability of an abusive partner. In this theme, talk placed mothers (DV victims) in a position of blame when they contested child–father contact. Mothers were portrayed through talk as problematic when they sought services to aid their children’s recovery, as exemplified by Extract 3, but when they failed to do so, talk oriented towards a pathologising of mothers (Extract 4):

Extract 3: We’re all just super damaged and broken.

In response to Q7 of interview schedule (see Appendix 1)

1. Judge 6 [If] they have money(.) they are probably going to tell you(.) that the child is in

2. counselling

3. Int Okay

4. Judge 6 amh as a means of point-scoring of a point or two against the other parent

5. Int Okay

6. Judge 6 you know he’s such an awful dad do you know what we’re all just getting to grips

7. with his(.) amh(.) brutality or his adultery or his meanness or his wickedness or his

8. irresponsibility or his drinking or his(.) and you know do you know what we’re all

9. just super damaged and broken (.) the children are in counselling.

As discussed above, in Extract 2, the judge’s discourse orients towards a minimisation of DV. Here, in Extract 3, the judge’s talk positions mothers who allege DV, to challenge the awarding of access/custody to the father, as manipulative and calculating. The judge engages the discursive device of footing (Goffman, 2000), where a speaker acts as an animator to increase the ‘factuality’ of their discourse (Potter, 1996: 122). This device may be used as a strategy in obtaining attributional distance from the talk, and thus functions to manage the speaker’s neutrality and give the appearance of objectivity in a sensitive and controversial area. This technique is often utilised by media interviewers who have a legal requirement to maintain an impartial stance (Wetherell et al., 2001). The mere fact that the judge chooses to change footing may be an indication that he is treating the current issue as contentious (Clayman, 1992).

Accordingly, by using the imaginary mother’s voice in talk (Lines 6–9), the judge adds factual credibility to the discourse and also obtains attributional distance. This functions to manage the judge’s own accountability when talk through a mother’s voice portrays mothers as exaggerating both the cause (‘brutality or his adultery or his meanness or his wickedness or his irresponsibility or his drinking or his’) and effect (‘just super damaged and broken’). A three-part list is thought to obtain optimum impact, by giving a sense of comprehensiveness, with additional list items adding little. Yet in Lines 7 and 8, talk includes a series of six abstract adjectives which hyperbolises the father’s behaviour, so that it is inferred as exaggerated and hence untrue. Moreover, violent descriptions are most effectively portrayed when they contain a vivid description with real consequences (Potter, 1996), for example, ‘the child being hurt, or kicked, or beaten up’ (Judge 2).

In Line 1 ‘[if] they have money’, talk implies that being ‘in counselling’ is something affluent people do. When the judge chooses to use the word ‘probably’, talk then generalises this portrayal to the majority of mothers who ‘have money’; this is not sporadic but a likely occurrence. Obtaining counselling for her children could be interpreted as a mother’s wish to aid her children’s recovery from the negative trajectory which results from exposure to DV, but when talk delegitimises the mother’s concerns as ‘point-scoring of a point or two’, the judge’s discourse orients towards a portrayal of counselling for these children as supercilious. More importantly, this discourse positions mothers as manipulative, as they are portrayed as using their children to manipulate the system to their own gain.

By positioning one parent as manipulative and self-serving, accountability is removed from the abusive partner; this facilitates the pro-access philosophy. The choice of discourse and the discursive devices employed may be considered suited to the local issue facing the judge. Hyperbolising fathers’ behaviour portrays it as untrue, which therefore works to delegitimise mothers’ concerns and imply that mothers have an embellished view of the negative impact of DV on both themselves and their children. What is more, mothers’ purpose in introducing the history of DV is portrayed as an attempt to gain the ‘upper hand’ in custody disputes:

Extract 4: Go from one abusive relationship to another.

1. Int [If] there is an abusive relationship (.) do you think that is completely separate

2. from the parents’ ability to parent?

3. Judge 3 Gosh that’s a very complex (.) amh (.) issue

4. Int u:um

5. Judge 3 because I have seen situations where (.) amh (.) some mothers particularly (.) would

6. go from one abusive relationship to another

7. Int Yeah

8. Judge 3 and the (.) you know I would ask myself the question (.) why is this happening? Is it

9. something in the make up of the person? Is it just bad luck? Amh (2) is a matter of

10. social (.) standing (.) empowerment (.) dependency? (.) it’s very very complex.

Extract 4 is an example of ‘ontological gerrymandering’ (Clayman, 1992); the practical management of a boundary such that certain phenomena are treated as problematic, while others can be assumed to be unproblematic as they are ignored. So in this example, despite being asked about parenting within DV families, the judge’s talk treats the gender-neutral question as a request to present a description which generates implications of fault and accountability for mothers while ignoring fathers’ abusive behaviour. For example, in Extract 4, Judge 3 stated in response to this question: ‘some mothers particularly (.) would go from one abusive relationship to another’. The addition of ‘particularly’ emphasises the focus on mothers. Therefore, by selecting and formulating this area (mothers’ behaviour in selecting abusive partners), the judge ignores fathers and the potentially negative effects of their parenting (Potter, 1996). By employing this discursive device, talk drew a rhetorical boundary around mothers’ behaviour, therefore making the mothers’ or victims’, rather than the perpetrators’, behaviour relevant. This may have an essential function, as introducing talk about perpetrators’ behaviour would undermine the implied explanation of mothers’ fault and compromise the pro-access philosophy.

Furthermore, talk constructs the image that mothers have in fact failed to protect their children, as it is mothers who ‘go from one abusive relationship to another’ – thus, it is mothers who repeat the pattern. This discourse orients towards a pathologising of mothers, and implies that abuse is the mother’s fault; there is in fact something inherently wrong with her: ‘is there something in the make-up of the person?’ Talk orients towards placing responsibility for perpetrators’ behaviour with victims.

This categorisation is preceded by a marker of common knowledge ‘you know’ in Line 8 (Edwards and Mercer, 1987); such devices work up the legitimacy of the category by ensuring that the claim appears rational. The discourse is embedded in uncertainty, ‘very complex’, which is strategically placed at both the beginning and end of this discourse (Lines 3 and 10). This achieves the business of maintaining the judge’s attributional distance in the complexities of the situation. The judge’s talk engages in a nature versus nurture style debate: ‘is there something in the make-up of the person?’ (nature) is rhetorically countered after a pause of 2 seconds with an environmental explanation of mother’s behaviour ‘Amh (2) is a matter of social (.) standing (.) empowerment (.) dependency?’ (nurture).

Overall, Extract 4 exemplifies the portrayal of mothers as going from one abusive relationship to another and places blame on the victim, while removing accountability from the perpetrator. Ignoring the impact that perpetrators’ abusive behaviour may have on their parenting ability thereby presents abusive behaviour as unproblematic in the context of child custody and access issues.

Deviant case analysis

Judge 4 was identified as a deviant case due to her application of scientific knowledge and her awareness of her nonconformity with her colleagues (Silverman, 2011). Judge 4 explicitly works up a category of entitlement (Potter, 1996); she positioned herself as an expert from the start of the interview with extensive scripting of membership of relevant organisations and qualifications in the area of psychology as well as law. This entitlement gives an epistemological right, therefore building up the factuality of the judge’s talk. However, it is the application of scientific knowledge which makes Judge 4 a deviant case (Silverman, 2011). Her awareness of her nonconformity with her colleagues is exemplified in Extract 5.

Extract 5: Now I am very extreme on that.

In response to Q2 of interview schedule (see Appendix 1)

1. Judge 4 I would be disinclined to have any access

2. Int Yeah

3. Judge 4 until I was assured that there was (.) a benefit of this to the children

4. Int Okay

5. Judge 4 now I am very extreme on that (…)

6. Int It’s fantastic to hear your views on this because=

7. Judge 4 = maybe my colleagues wouldn’t necessarily ah agree with me

8. Int Yeah

9. Judge 4 some colleagues ah would say that domestic violence is one issue (.) and access is a

10. different issue

11. Int Yeah

12. Judge 4 I don’t hold that view

In Extract 5, the judge is ‘talking against established ideas’ (Taylor and Littleton, 2006: 24). The judge positions herself as ‘extreme’ because she ensures that child contact is of benefit to the child. She then uses the words ‘some of’ not ‘most of’ in Line 11 which contradicts her word ‘extreme’. This may be reflective of the interactive nature of the construction. In Line 8, ‘It’s fantastic to hear your views on this because=’, the judge interrupts the interviewer to construct an explanation for her difference ‘=some colleagues ah would say that domestic violence is one issue (.) and access is a different issue’, but explicitly expresses her deviation ‘I don’t hold that view’. With this discursive work, she remains respectful to her colleagues while convincingly presenting her own views which are in line with scientific knowledge (Antaki et al., 2003).

Discussion

The current findings emphasise a pro-access philosophy in judges’ talk throughout the interviews. This position is justified, in the context of DV, by a minimisation and normalisation of DV throughout talk. Mothers’ concerns for their child’s safety are thus set aside, and talk portrays them as controlling and manipulating when they object to contact between their child and the non-live-in father.

A discourse which idealised the nuclear family was evident throughout the interviews. Moreover, a discourse acknowledging the ‘horrendous damage’ to women and children as the consequences of DV co-existed with an assumption that contact with the perpetrator was in the child’s best interest. Contra to this, available evidence suggests that children do not do well if their home lives are marked by parental conflict (Bancroft et al., 2011). The findings of a pro-access philosophy concur with previous ethnographic studies in Irish courtrooms (Coulter, 2009; Mahon and Moore, 2011) which identified joint custody as the idealised post-separation family configuration by judges.

Evidenced discourses which construct DV as singular physical incidents, which end on the separation of the parties, function to justify a pro-access philosophy. Constructing DV as a singular physically violent incident defies the world-wide understanding of DV as a pattern of coercive behaviours (Bancroft et al., 2011). The explicit portrayal of DV as ending with the separation of the parents also contradicts extensive empirical research findings. In fact, it is a well-established fact that DV will continue post-separation (Vatnar and Bjørkly, 2012) and that there is in fact an extremely high probability of the escalation of DV at the time of separation, with a high risk of femicide (Brownridge, 2006). Moreover, empirical research has demonstrated that both litigation and contact with the children can be used as a vehicle to sustain the abuse (Jaffe et al., 2003; Zeoli, 2013). Hayes (2012) identified that abusive fathers changed to more covert forms of DV (which can involve the children) post-separation, rather than ceasing the abuse.

Ontological gerrymandering had the effect of removing agency from abusive parents as their past and current abusive behaviours were ignored in talk; yet, it is established that past violent behaviour is the best predictor of future violent behaviour (Elizabeth et al., 2010). Throughout the transcripts, there was an absence of talk about the violent behaviours of the perpetrators of DV, nor were the judges overtly concerned with the negative impact witnessing DV has on children (Bancroft, et al., 2011). Furthermore, judges’ talk made no link to either the comorbidity between DV and direct child abuse (Wolfe et al., 2003), or the parenting deficits of perpetrators of DV (Bancroft et al., 2011). Kernic et al. (2005) highlighted this rendering of DV as irrelevant to child custody by family courts, when they identified no significant difference in the granting of custody between perpetrators and non-perpetrators of DV.

Rather than placing accountability with the alleged perpetrators, agency was firmly placed with mothers, who were depicted as exaggerating their situation. This is in line with arguments made by Conner (2009), who identified that judges held a belief that women had a tendency to exaggerate DV. Furthermore, the positioning of mothers as alleging DV as a means to manipulate the system concurs with previous findings; Pond and Morgan (2008) found that some lawyers’ discourses questioned mothers’ integrity and their motivations in obtaining protection orders. However, it is in direct contrast with the known cost of DV to women and children in Irish society. A recent survey conducted by the European Human Rights Federation (EU FRA, 2014), consisting of 42,000 women across 28 countries, including Ireland, identified that 1 in 3 women disclosed experiencing partner violence and that 73% of mothers identified that at least one of their children was aware of the partner violence.

The pathologising of mothers who attempted to protect their children has also been established by Harrison (2008), who identified that the family law system labelled mothers (victims of DV) as implacably hostile when they obstructed fathers’ right to a relationship with their children. Furthermore, Johnston and Steegh (2013) identified a tendency for courts to penalise mothers who do not encourage child–father contact. In line with this, Zeoli et al. (2013) found that mothers feared that advocating for their children’s safety within the court system could backfire and place the children at greater risk. This points to the extent to which values and discourses of wider society are sexist, paternalistic and misogynistic, as they may act to influence values of those operating within the system at an implicit level.

This resonates with previous research findings (Coates and Wade, 2004; Coates et al., 1994) that the passivity of a victim during a sexual attack was positioned as compliance rather than fear and facilitated the portrayal of sex as consensual. Ehrlich (2007) demonstrated that such constructions stemmed from gendered cultural norms. Our findings hinge on similar gendered ideologies. Stereotypes of victims of DV as passive, meek and vulnerable fail to acknowledge the power dynamics that may result in women appearing passive in an abusive context. In an alternate context where they feel more secure and are motivated to protect their children, such as during court proceedings, they may appear agentic. As a result of these more agentic behaviours, such mothers may be seen by our judges as incompatible with the narrative of abuse, and thus are constructed not as victims but as aggressors.

Limitations and future research

The heavy caseloads of judges and the sensitive nature of the study contributed to a difficulty in gaining access to this elite group. In any case, it would not be appropriate to generalise the current findings, as sampling was purposeful, and the judges who placed themselves at some inconvenience and took time out from a busy schedule to speak on this subject may have an increased interest in the subject area. They may also be considered more liberal than their fellow judges who, despite repeated invitation, declined to partake in this study. In line with this there was a two-thirds majority of female participants, despite the fact that they consist of less than one-fifth of the current District Court judiciary (Court Services, 2012). However, the finding of a pro-access philosophy does concur with court statistics, whereby only 2% of access applications were refused by the family law courts in both 2010 and 2011 (Court Services, 2011).

The current findings were grounded in judges’ discourse in the context of an interview; they highlight the need for future investigation in this area. These assumptions which define judges’ understanding of gender norms have been shown to play an influential role in their sense-making framework (Ehrlich, 2007) but cannot be directly linked to judges’ adjudications of individual child custody and access cases in this interview study. A DA of court files is warranted to undertake an in-depth analysis of gender differences, identify if judges’ discourse transfers to their adjudications, and the weight which is given to scientific knowledge in the determination of the best interests of the child.

Conclusion

The current findings establish that when Irish judges were questioned in recorded interviews about the extremely complex and fraught issue of child custody and access in the context of DV, their talk oriented towards a use of heuristics which portrayed mothers as manipulative and controlling. There is a clear need for a change of discourse from one which normalises DV and places agency for the impact of separation with mothers. If we are to be true to legislation and take the welfare of the child as the ‘first and paramount consideration’ (Guardianship of Infant Act, 1964, s. 3), then it beholds the family law system to place the weight on DV within their adjudications, in line with empirical research. The gap between academia and reality needs to be bridged, but first and foremost, the concept of ‘ordinary decent domestic violence’ should be relegated to a discourse of the distant past.

Footnotes

Appendix 1 Acknowledgements

We would like to thank members of the Centre for Social Issues Research, Dr Michael Quayle, Dr Susan Leahy and the anonymous reviewer for insightful comments on previous versions of this article.

Declaration of conflicting interests

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors

Notes

Author biographies

Catherine M Naughton is a doctoral candidate at the University of Limerick’s Department of Psychology and Centre for Social Issues Research. Her primary research interest is in the application of a social identities perspective to Intimate Partner Violence (IPV), specifically the mediating role of social identity in determining well-being in young people exposed to IPV in their families of origin.

Aisling T O’Donnell is a Lecturer in Psychology based in the Department of Psychology and Centre for Social Issues Research at the University of Limerick. Her broad research interests are concerned with how social identities can both impact upon, and be affected by, other aspects of our social environment.

Ronni M Greenwood is a Lecturer in Psychology based in the Department of Psychology at the University of Limerick. In the broadest sense, her research is concerned with ecological aspects of psychological well-being and distress amongst members of marginalised social groups.

Orla T Muldoon is a Professor of Psychology, Founding Chair of the Department of Psychology and Director of the Centre of Social Issues Research. Her broad research interests are concerned with the application of the social identity approach to real world social issues.

References

Antaki C Billig

Edwards

. (2003) Discourse analysis means doing analysis: A critique of six analytic shortcomings. Available at: https://dspace.lboro.ac.uk/dspacejspui/bitstream/2134/633/3/DAOL+Discourse+Analysis+Means+Doing+Analysis.pdf (accessed 17 December 2014).

Bancroft

Silverman

Ritchie

(2011) The Batterer as Parent: Addressing the Impact of Domestic Violence on Family Dynamics. Thousand Oaks, CA: SAGE.

Berger

(2009) How embedded knowledge structures affect judicial decision making: An analysis of metaphor, narrative, and imagination in child Custody disputes. Southern California Interdisciplinary Law Journal 18: 259–308.

Braun

Clarke

(2006) Using thematic analysis in psychology. Qualitative Research in Psychology 3: 77–101.

Brownridge

(2006) Violence against women post-separation. Aggression and Violent Behavior 11: 514–530.

Clayman

(1992) Footing in the achievement of neutrality: The case of news interview discourse. In: Drew

Heritage

(eds) Talk at Work: Interaction in Institutional Settings. Cambridge: Cambridge University Press, pp. 163–198.

Coates

Wade

(2004) Telling it like it isn’t: Obscuring perpetrator responsibility for violent crime. Discourse & Society 15: 499–526.

Coates

Bavelas

Gibson

(1994) Anomalous language in sexual assault trial judgments. Discourse & Society 5: 189–206.

Conner

(2009) Abuse and discretion: Evaluating judicial discretion in custody cases involving violence against women. American University Journal of Gender Social Policy & the Law 17: 163–223.

10.

Coulter

(2009) Family Law in Practice: A Study of Cases in the Circuit Court. Dublin, Ireland: Clarus Press.

11.

Court Services of Ireland (2011) Annual report. Available at: http://www.courts.ie/courts.ie/library3.nsf/WebPageCurrentWeb/646E98A7939A4C65802576D9005652F8?OpenDocument&l=en (accessed 23 April 2014).

12.

Court Services of Ireland (2012) Annual report. Available at: http://www.courts.ie/Courts.ie/Library3.nsf/0/3B21EBCDBCA7469C802575E6003DBD7FOpenDocument (accessed 23 April 2014).

13.

Court Services of Ireland (2012) Annual report. Available at: http://www.courts.ie/Courts.ie/Library3.nsf/0/3B21EBCDBCA7469C802575E6003DBD7FOpenDocument (accessed 23 April 2014).

14.

Duke

(2002) Getting beyond the ‘official line’: Reflections on dilemmas of access, knowledge and power in researching policy networks. Journal of Social Policy 31: 39–59.

15.

Edwards

(1994) Script formulations: An analysis of event descriptions in conversation. Journal of Language and Social Psychology 13: 211–247.

16.

Edwards

Mercer

(1987) Common Knowledge. London: Methuen, Routledge.

17.

Edwards

Potter

(2005) Discursive psychology, mental states and descriptions. In: Molder

Potter

(eds) Conversation and Cognition. Cambridge: Cambridge University Press, pp. 241–259.

18.

EU FRA (2014) Violence against women: An EU-wide survey. Report for European Union Agency for Fundamental Rights. Available at: http://fra.europa.eu/en/publication/2014/vaw-survey-main-results (accessed 23 April 2014).

19.

Ehrlich

(2007) Legal discourse and the cultural intelligibility of gendered meanings. Journal of Sociolinguistics 11: 452–477.

20.

Elizabeth

Gavey

Tolmie

(2010) Between a rock and a hard place: Resident mothers and the moral dilemmas they face during custody disputes. Feminist Legal Studies 18: 253–274.

21.

Eriksson

(2011) Contact, shared parenting, and violence: Children as witnesses of domestic violence in Sweden. International Journal of Law, Policy and the Family 25: 165–183.

22.

General Assembly of United Nations (1989) Convention on the Rights of the Child (United Nations, Treaty Series, vol. 1577). Available at: https://treaties.un.org/Pages/ViewDetails.aspx?mtdsg_no=IV-11&chapter=4&lang=en (accessed 23 April 2014).

23.

Gilovich

Griffin

Kahneman

(2002) Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge: Cambridge University Press.

24.

Goffman

(2000) Footing (from Forms of Talk, Oxford, Blackwell, Chapter 3). In: Wetherell

Taylor

Yates

(eds) Discourse Theory and Practice. London: SAGE Publication Ltd., pp. 93–110.

25.

Government of Ireland (1964) Guardianship of Infants Act 1964. Available at: http://www.lawreform.ie/_fileupload/Restatement/Second%20Programme%20of%20Restatement/EN_ACT_1964_0007.PDF (accessed 23 April 2014).

26.

Harrison

(2008) Implacably hostile or appropriately protective? Women managing child contact in the context of domestic violence. Violence Against Women 14: 381–405.

27.

Hayes

(2012) Abusive men’s indirect control of their partner during the process of separation. Journal of Family Violence 27: 333–344.

28.

Holt

(2011) Domestic abuse and child contact: Positioning children in the decision-making process. Child Care in Practice 17: 327–346.

29.

Jaffe

Lemon

Poisson

(2003) Child Custody and Domestic Violence: A Call for Safety and Accountability. Thousand Oaks, CA: SAGE.

30.

Jefferson

(1984) Transcript notation. In: Atkinson

Heritage

(eds) Structures of Social Action: Studies in Conversation Analysis. Cambridge, UK: Cambridge University Press, pp. 346–369.

31.

Johnston

Steegh

(2013) Historical trends in family court response to intimate partner violence: Perspectives of critics and proponents of current practices. Family Court Review 51: 63–73.

32.

Kernic

Monary-Ernsdorff

Koepsell

. (2005) Children in the crossfire: Child custody determinations among couples with a history of intimate partner violence. Violence Against Women 11: 991–1021.

33.

Krieger

(2000) Socio-legal backlash. Berkeley Journal of Employment & Labor Law 21: 476–520.

34.

McCarthy

Edwards

Gillies

(2000) Moral tales of the child and the adult: Narratives of contemporary family lives under changing circumstances. Sociology 34: 785–803.

35.

MacMartin

Wood

(2005) Sexual motives and sentencing judicial discourse in cases of child sexual abuse. Journal of Language and Social Psychology 24: 139–159.

36.

Mahon

Moore

(2011) Post separation parenting. A study of separation and divorce agreements made in the Family Law Circuit Courts of Ireland and the implications for parent -child contact and family lives. Report for the Office of the Minister for Children and Youth Affairs. Available at: http://www.dcya.gov.ie/documents/publications/researchdocs/Post_Separation_Parenting_Briefing_Note_No8.pdf (accessed 23 April 2014).

37.

McMullen

(2012) Discourses of influence and autonomy in physicians’ accounts of treatment decision making for depression. Qualitative Health Research 22: 238–249.

38.

Pomerantz

(1986) Extreme case formulations: A way of legitimizing claims. Human Studies 9: 219–229.

39.

Pond

Morgan

(2008) Protection, manipulation or interference with relationships? Discourse analysis of New Zealand lawyers’ talk about supervised access and partner violence. Journal of Community & Applied Social Psychology 18: 458–473.

40.

Potter

(1996) Representing Reality: Discourse, Rhetoric and Social Construction. Thousand Oaks, CA: SAGE.

41.

Schneider

(1991) Discretion, rules, and law: Child custody and the UMDA’s Best-Interest Standard. Michigan Law Review 89: 2215–2298.

42.

Shea Hart

(2011) Child safety in Australian family law: Responsibilities and challenges for social science experts in domestic violence cases. Australian Psychologist 46: 31–40.

43.

Silverman

(2011) Interpreting Qualitative Data. Thousand Oaks, CA: SAGE.

44.

Starks

Trinidad

(2007) Choose your method: A comparison of phenomenology, discourse analysis, and grounded theory. Qualitative Health Research 17: 1372–1380.

45.

Talbot

Quayle

(2010) The perils of being a nice guy: Contextual variation in five young women’s constructions of acceptable hegemonic and alternative masculinities. Men and Masculinities 13: 255–278.

46.

Taylor

Littleton

(2006) Biographies in talk: A narrative-discursive research approach. Qualitative Sociology Review 2: 22–38.

47.

Vatnar

SKB

Bjørkly

(2012) Does separation or divorce make any difference? An interactional perspective on intimate partner violence with focus on marital status. Journal of Family Violence 27: 45–54.

48.

Wetherell

Taylor

Yates

(2001) Discourse Theory and Practice: A Reader. Thousand Oaks, CA: SAGE.

49.

Willig

(2008) Introducing Qualitative Research Methods in Psychology. Buckingham: Open University Press, p. 108.

50.

Wittgenstein

(1958) The Blue and Brown Books: Preliminary Studies for the ‘Philosophical Investigation’. Oxford: Wiley-Blackwell.

51.

Wolfe

Crooks

Lee

. (2003) The effects of children’s exposure to domestic violence: A meta-analysis and critique. Clinical Child and Family Psychology Review 6: 171–187.

52.

Wood

MacMartin

(2007) Constructing remorse judges’ sentencing decisions in child sexual assault cases. Journal of Language and Social Psychology 26: 343–362.

53.

Zeoli

Rivera

Sullivan

. (2013) Post-separation abuse of women and their children: Boundary-setting and family court utilization among victimized mothers. Journal of Family Violence 28: 547–560.