The identification of implicit theories (ITs) in certain offender groups, that is, offence-related schemas of their view of the world which support their offending behaviour (Ward 2000), have led to the development and implementation of successful offence specific interventions (Beech et al. 2012; Collie et al. 2007. In recent years, the cognitive characteristics of intimate partner violence (IPV) perpetrators has received considerable attention, however, empirical research regarding implicit theories among these individuals is an area which has, only recently, begun to be explored (Gilchrist 2009; Dempsey and Day 2010; Dobash and Dobash 2011; Weldon and Gilchrist 2012). Moreover, as of yet, the cognitions of IPV offenders who also sexually offend (SO) against their partners has received no empirical investigation. This is surprising given that IPV risk assessment and screening tools in addition to intervention strategies are aimed at addressing attitudinal differences thought to be present in the IPV offender (Gilchrist 2007; Rabin et al. 2009; Morgan and Gilchrist 2010).

Defining implicit theories for IPV offenders who sexually offend against their partner would be extremely beneficial in terms of intervention and treatment as the specific schemas present in this group of offenders could be targeted. Furthermore, intimate partner rape has been found to be a recognised risk factor in intimate partner homicide (Campbell 2007). Distinct implicit theories which have been identified in sex offenders, and the current ongoing work relating to violent offenders, have increased our understanding of specific offending behaviours and led to the design and implementation of intervention programmes targeting specific criminogenic needs (Ward 2000; Polaschek and Gannon 2004; Collie et al. 2007. Schema-based therapy is now being established as an appropriate treatment for sex offenders (Beech et al. 2012; Fisher and Beech 2007) and so the same could be utilised in the distinct group of IPV and IPV SO offenders. In addition to intervention, the current research may also be useful in the development of appropriate risk assessment tools which take specific schemas into account.

When considering the implicit theories established in other offender groups, Polaschek and Ward (2002) identified five potential implicit theories in rapists by applying the implicit theory perspective to existing scales that measured rape related cognitive distortions. The implicit theories extrapolated were: “women are unknowable;” “women are sex objects;” “male sex drive is uncontrollable;” “entitlement;” and “dangerous world.” These implicit theories were later corroborated by Polaschek and Gannon (2004) who examined transcripts of sex offender interviews and coded them for the presence or absence of the five implicit theories proposed by Polaschek and Ward (2002). While not a distinct implicit theory from the analysis of interview transcripts, Polaschek and Gannon (2004) also noted the importance of denial in rapists’ accounts of their offences, given that males who are in denial of the offence may not be suitable for treatment. Therefore, Polaschek and Gannon (2004) defined deniers as men who admit that they had sexual contact with the victim but insist it was not offensive and so minimising the negative impact this would have had on the victim.

Beech et al. (2005)investigated the presence of implicit theories in sexual murderers by applying a grounded theory analytical approach to the murderers’ account of their offence. Their findings corroborated the same five ITs found to be present in rapists by Polaschek and Ward (2002) and Polaschek and Gannon (2004) indicating that sexual murderers are not qualitatively distinct from rapists with reference to the underlying schemas that they have regarding the world. Beech, Fisher, and Ward concluded that this would have intervention implications and suggested that a schema approach would allow offenders’ cognitions to be challenged.

When considering IPV offenders, Weldon and Gilchrist (2012) investigated implicit theories from the perspective of the IPV perpetrators, by examining their own accounts of their background, relationship history, offending behaviour history, and role ideology. They proposed 11 potential implicit theories in IPV offenders using interpretative phenomenological analysis (IPA; Smith 2004). The implicit theories proposed were “violence is acceptable;” “grievance/revenge;” “need for control;” “real man;” “entitlement/women are objects;” “male sex drive is uncontrollable/policing partner;” “women are provoking;” “rejection/abandonment;” “women as supportive;” “external factors responsible;” and “nature of harm.”

The established implicit theories in rapists and sexual murderers greatly overlap with those proposed by Weldon and Gilchrist (2012). In addition to the apparent commonality in cognition in IPV and sexual offenders, there is also increasing recognition that IPV often co-occurs with child sexual abuse (Morgan and Gilchrist 2010). Hester et al. (2007) note that familial settings in which IPV occurs are the most common context for child abuse and that male IPV perpetrators are more likely to be abusive to children and more extremely so. This has been recognised in government policy with the Action Plan on Sexual Abuse (HM Government 2007) calling for an integrated approach to tackling child abuse and IPV (Morgan and Gilchrist 2010). When considering cognition in child sex offenders, Ward and Keenan (1999) proposed five implicit theories based on measures of cognitive distortions in empirical literature. These implicit theories are as follows: “children as sexual objects;” “entitlement;” “dangerous world;” “uncontrollability;” and “nature of harm.” These findings somewhat overlap with the implicit theories presented in IPV offenders and sex offenders however were not derived from qualitative exploration of the child molesters offence process. Indeed Ward and Keenan (1999) note that, for effective intervention, the child sex offenders’ offence process should be explored on an individual basis to gain access to the core implicit theories.

The findings of Polaschek and Ward (2002); Polaschek and Gannon (2004), and Weldon and Gilchrist (2012) suggest that it is possible to identify an offender’s underlying motivations without the offender explicitly acknowledging them. This is invaluable given the difficulty of assessing offenders using standard psychometric measures and interviews, where they are frequently defensive and denying (Fisher and Beech 2007). Given the difficulty of accessing deep-level schemas and obtaining honest answers from IPV offenders and sex offenders, approaches other than direct psychometric questionnaires need to be used. Qualitative approaches, as utilised by Beech et al. (2005) and Weldon and Gilchrist (2012), where offenders are encouraged to give a free narrative about their offending, can be beneficial in determining the implicit theories held by the offender.

Aims of Current Study

The current research study aims to explore sex offending and IPV from the abusers’ perspective and identify schemas which could be representative of distinct implicit theories in this specific group of offenders. Therefore, interpretative phenomenological analysis shall be utilised to explore the participant’s lived experience (Smith 2004), with the intention of identifying implicit theories which may account for cognition in IPV sex offenders. Identification of distinct implicit theories in IPV sex offenders would have implications for empirical research, clinical intervention, and risk assessment.

Method

Interpretative Phenomenological Analysis

Interprative Phenomenological Analysis (IPA) is widely used in the domains of clinical, health, and social psychology and is particularly appropriate to investigating unexplored territory where a theoretical or empirical pretext may be lacking Reid et al. (2005). The goal of IPA is to study how people make meaning of their lived experience. This approach is particularly pertinent in the current study given the aim of identifying implicit theories in IPV sex offenders, an area which has little empirical support.

When considering the analytical process in IPA, the themes identified are representative of a core commonality and structure of the experience (Starks and Brown-Trinidad 2007). Therefore, this methodology is particularly suited in the current study as it is hypothesised that the themes identified will be representative of core implicit theories held by IPV sex offenders.

Recently, the empirical utility of IPA with forensic populations has been established (Brown and Beail 2009; Blagden et al. 2011; Weldon and Gilchrist 2012). The access to latent cognitions is significant in a forensic population where the demand factor, that is the desire to be seen in a positive manner, is more pronounced which may have been reflected in self-report quantitative measures.

Participants

Full ethical approval for the current study was sought and granted from the Scottish Prison Service (SPS) and this was forwarded to the School of Health University of Edinburgh Ethics Committee. Participants were recruited by a Strategy Analyst, employed by the Scottish Prison Service, who was asked to conduct a search for potential recruits by the SPS ethics board as part of his role. The inclusion criteria were that the participant be presently serving a custodial sentence and have a current or previous IPV conviction in addition to a current or previous conviction of a sex offence. The Strategy Analyst ran a statistical report using a desktop intelligence package which identified all participants who met the inclusion criteria across all prisons in Scotland. Participants were identified by having a Sex Offender Marker and Domestic Violence Aggravation Code flagged on their profile. The analyst focused on three particular sites due to their geographical viability of access for the researcher and as these three sites would pertain to sufficient numbers for interview for a qualitative study. Twelve potential participants were identified across the sites who were each sent a letter inviting them to take part and explaining what this would entail. Of these 12 participants, 11 agreed to be interviewed for the study. The guidance offered by Smith et al. (2009) states that a sample size of between four and 10 is sufficient when utilising IPA, therefore the current study is exceeding this guidance by n = 1. Furthermore, Smith et al. (2009) note that it is important not to see higher numbers as being indicative of better work as successful analysis requires time, reflection, and dialogue with larger datasets tending to inhibit all of these things hence why 11 participants was deemed appropriate in the current study.

Of the 11 participants interviewed, 10 had a current or previous conviction of IPV. The remaining participant had been arrested on three occasions for an IPV charge and served a custodial sentence on remand. Thereafter, the charges were dropped. Ten participants had a current or previous sexual offence conviction. The remaining participant had six untried offences for the Sexual Offences Act over a two year period. All charges of rape committed by the participants in the current study, of which there were seven, occurred in the context of an intimate relationship. A further four of the participants were convicted of a sexual offence against a child and not an adult female. It was felt that their inclusion in the current study was warranted given the significant links between child sex abuse, IPV, and family violence (Hester et al. 2007; Mullender 2006). The final participant was convicted with both attempted rape (against an adult female) and sexual assault against a child.

The offence history and relevant demographic information for each of the participants is outlined in Table 1. Pseudonyms have been used to ensure confidentiality.

Table 1 Demographic Data of Participants
Full size table

Due to the sensitive nature of the material covered in the interviews, each participant was given an advice sheet with telephone numbers of internal and external support lines should the interview evoke thoughts or feelings which they would benefit from talking through with a professional. Additionally, with consent from the participant, a prison officer in their wing was informed of their participation in the study so as to provide support after the interview if the participant deemed it necessary.

Procedure

All 11 interviews were conducted in interview rooms within each of the three prisons. All of the interviews were conducted on a one-to-one basis and were digitally-recorded with permission from the participant. In order to ensure safety for both the participant and the researcher, a prison officer was available outside each of the interview rooms across all sites and local security procedure was followed. The prison officer did not hear the content of the interviews. The length of each interview varied from 36 min to 1 h and 8 min. In accordance with IPA methodology (Smith et al. 2009), interviews were semi-structured and non-prescriptive to allow the participant to reveal their own cognitions, beliefs, and feelings. This enabled participants to have ownership of the direction of the conversation while the researcher implemented non-directive triggers to aid exploration of relevant topics as they emerged (Brown and Beail 2009). This method of interview has been shown to be especially appropriate when investigating personal subject matters (Silverman 2000). In order to gain insight into the participants’ perception of their world, the semi-structured interview explored childhood, relationships with prominent attachment figures, as well as intimate relationships and their perception of male and female roles within these. Where possible, the interviews also explored the participants’ experience of offending behaviour and the context in which this occurred.

All data were anonymised and stored confidentially. The digitally-recorded interviews were transcribed by the researcher and kept securely thereafter. At all times, every effort was maintained to ensure confidentiality of the participants.

Data Analysis

Data were analysed in accordance with IPA methodology in order to make sense of and explore the participant’s lived experience from their perspective (Smith and Osborn 2003; Smith et al. 2009). Interpretative phenomenological analysis recognises the importance of the researcher in the analytic process which is described by Smith (2004) as a dialectical interpretative relationship between the researcher and the text. Consequently, the construct of the participant’s world is not readily available but rather accessed through sustained engagement and interpretation of the text (Smith and Osborn 2003). Analysis is, therefore, an interactive process open to influence from the researchers’ beliefs and cognitions. These must be made explicit, as far as possible, to put the analysis into context. Thus, the role of the female researcher in the process is prominent given that all participants have a charge of violence and sexual assault against a female. Every effort was made to minimise any impact of the female researcher on what the participant chose to discuss by adopting a non-judgemental approach and adhering to the semi-structured interview procedure.

When considering analysis, transcripts were first read in detail several times in order to commence a sustained engagement with the text. Subsequent to detailed reading of the text, notes are made by the researcher in the left hand margin to identify prominent points, associations, and contradictions. Thereafter, these notes are used to establish potential emergent themes which are documented in the right hand margin. These themes are recorded in the form of phrases attempted to capture the essence of what was being discussed and what sense the participant was making of their world (Smith and Osborn 2003). Participants’ phenomenologies begin to be interpreted at this stage through a dialect between the text and theoretical constructs underpinning these. Emerging themes were listed separately and thereafter reviewed in relation to each other to identify those which could be clustered together as subordinate themes. Superordinate themes were then identified to encapsulate these subordinate themes.

Quality Assurance

When considering what constitutes a theme, in accordance with Smith (2011) a theme was evidenced as such if it was present in extracts from at least three of the participants. Furthermore, as the transcripts of 11 participants were included in the current study, a measure of prevalence of each subtheme is included as a quality assurance measure.

In order to ensure the process from raw data to emergent themes could be ensued, three independent researchers familiar with implementing IPA methodology analysed three separate transcripts and were found to have a high level of agreement with the researcher in both subordinate and superordinate themes (75, 100, and 92 %, respectively). All three independent researchers had limited knowledge of IPV literature unlike the original researcher. In order to ensure confidentiality, pseudonyms were used throughout all transcripts, including when names of victims and/or family members were mentioned.

At interview, each participant was offered the opportunity for the researcher to provide feedback regarding their transcript and analysis at a later stage in the process to ensure the researcher has adequately captured their perception of their world. All participants, other than Jonathon, declined this offer. Jonathon was no longer serving a custodial sentence when his transcript and analysis was complete. Subsequently, he was contacted through Prison Social Services (at his request at interview) at which point he declined to meet with the researcher to discuss his part in the research process.

Results

Analysis in accordance with IPA methodology and validated by three independent researchers revealed the emergence of five superordinate themes composed of 14 subthemes which are presented as Appendix 1, Table 4. A brief illustrative extract of each subtheme is provided in addition to the prevalence of each subtheme across all participants and a description of each subtheme. This is done to demonstrate the presence of each subtheme across at least three participants as highlighted by Smith (2011).

The extracts provided in Appendix 1, Table 4 and in the subsequent analysis were chosen to illustrate the superordinate and subthemes from the raw data as it was thought they best represented the experience of the participants from their perspective providing insight into their world and their interpretation of this. Analysis is presented in the format of extracts of transcripts supporting superordinate themes and detailed analysis and interpretation of the extracts.

Three of the transcripts were rated by three independent researchers familiar with IPA methodology for the purposes of validity. This involved each independent researcher reading a transcript, picked at random by the original researcher, completing analysis, and then conceptualising sub-themes. Thereafter, the primary researcher compared these subthemes to the original analysis. The results of the inter-rater reliability process are outlined in Table 2.

Table 2 Concordance of Subthemes Between Researchers
Full size table

While all researchers used slightly different terminology to capture the themes, it was thought the themes identified were representative of themes established by the original researcher. For example, “violence as part of growing up” was thought to be representative of “violence is normal/acceptable” and was coded as such. The theme of “feeling emasculated” identified by one researcher was accurately represented in the “real man” subtheme and “external locus of control” and “blaming on substances” were captured under the “uncontrollability” subtheme. All 14 subthemes, or variations of these capturing the same implicit concept, were identified by at least one of the independent researchers.

The 14 subthemes identified are representative of 14 potential implicit theories held in IPV SO offenders. The emergent subthemes correspond to five superordinate themes which are presented and analysed in accordance with IPA methodology below. The themes are representative of a descriptive and exploratory account of IPV sex offenders from their perspective. It is thought the themes are representative of implicit theories of IPV sex offenders as, throughout the interview process, participants construct narrative identities which can be understood as an active information processing structure, a cognitive schema or a construct system that is both shaped by and later mediates social interaction including that with the researcher. The superordinate themes are as follows: Violence is Normal; Need for Control; Women’s Role in Violence and Sexual Assault; Diminishing Personal Responsibility; and Sense of Self.

Violence is Normal

The superordinate theme “violence is normal” was evident in all participants’ accounts and is representative of violence being a habitual developmental, social, and interpersonal construct pervasive in all domains of their world and may be present in all violence offenders not specifically IPV. The superordinate theme is composed of three subthemes: “violence is acceptable;” “grievance/revenge;” and “dangerous world.”

I can’t remember what happened, somebody had done something wrong or something like that and I was only about two or three, maybe a toddler, he lined the four of us up and he put down a slipper, a belt, a bat, a cane, things like that and asked us what one we wanted to be hit with as a punishment. Then that’s like a bit bad like mental torture it’s bad enough hitting someone but asking them to pick the weapon they get hit with that’s like sadistic. -Malcolm.

Malcolm’s account of his father’s violent behavior at such a young age demonstrates that he was accustomed to violence as an acceptable punishment for “bad” behaviour. Furthermore, Malcolm’s inability to recall the incident which caused the need for such a punishment shows the incident itself was not particularly unusual or extreme behaviour to warrant such a severe punishment. His statement that it’s a “bit bad” shows a minimisation or desensitisation to such a punishment which is somewhat contradicted by his reference to “mental torture” and “sadistic.” The use of these words demonstrates that, when reflecting on the violence used, he is aware of the severity of it but that it was a standard response to wrong doing in his family home. The perception of violence being acceptable within the family home was consistent with Paul’s experience:

My mum was quite bad to me, know what I mean. She used to take me and lock me in garages and cupboards and that when she was drunk. My dad worked away most of the time so he didn’t know she was doing it and then he got made redundant and it just continued. I used to wet the bed and every morning, I got put in a cold bath and leathered every single day if you know what I mean. Sometimes I was terrified to go to sleep at night in case I wet the bed and when I got put in care I stopped wetting the bed you know what I mean. I don’t know why it just stopped maybe it was just my nerves or something. - Paul.

As in Malcolm’s account, Paul minimises his mother’s violent behaviour by stating she was “quite bad to me” and then contradicts this account but recalling severely abusive behaviour. The statement “leathered every single day” highlights the frequency of the abuse showing that, for him, violence was a daily occurrence. This was to the extent he was scared to go to sleep at night as a young child and so seems to have had no experience of a nurturing home environment. Paul’s account is also consistent with the subtheme of “rejection/abandonment” as it seems he felt rejected by his female primary caregiver which may have caused him to develop cognitions around women being unknowable and/or dangerous at a young age. Paul’s experience further corroborates the “dangerous world” subtheme, in which the experience of participants is that people will behave in a rejecting manner to promote their own needs. This leads to the development of the implicit theory that the world is dangerous, meaning you must be perceived as a threat or you will be victimized.

Participants had a shared experience that violence was a standard means of practice in the family home and a form of punishment. For some participants, it went further and was viewed in a positive manner in their peer group:

it (violence) was usually planned for a Friday or a Saturday, Friday or a Saturday night so it wasn’t like violence every day of the week you know what I mean it was... we were all kinda close. Em, I don’t think, I don’t think anything would’ve been done any different. Even if I knew then what I know now just with the kind of environment I was in, I don’t think it would’ve been any different. I’m still the same person, grew up around the same people and I think it would’ve just happened anyway. - Steve.

Steve’s account of violence was that it was a positive experience for him and seemed to provide him with peer support and a sense of camaraderie that was something which he valued and even reflecting on this time now does not regret it nor view it as a negative experience.

The subtheme of “grievance/revenge,” that is violence as an acceptable response to a perceived slight and an effective means of problem solving, was evidenced by Malcolm:

I ended up having to hit him. I confronted him and told him to stop it you’d better stop your carry on it’s pointless and then he bit me, well he tried to bite me. If he hadn’t tried to bite me I probably wouldn’t have hit him but the fact that he tried to bite me the red mist came o’er and I just hit him a couple of times and that solved the problem eh.

This exerpt shows that Malcolm perceived violence to be the only means available to respond to an altercation with a fellow prisoner after having tried to verbally reason with him. Furthermore, it seems his use of violence reinforced his application of it in future occasions given his perception that it “solved the problem.” The subtheme of “dangerous world” was evidenced by James:

In the children’s home there was a wee bit of bullying going on in there, me and my brother went into the children’s home, the Officer in Charge tried to hit us, almost hit me in the arm with a hammer eh there was a bit of, I felt really scared cos the woman was quite aggressive she would make you stand in the corner for hours, you werenae getting your supper or tea and obviously the guys that were living in there were older than us so they gave us a bit of a hard time. Em I got hit by a brick aff somebody in there...

James experience of early family life was one of neglect hence why he was placed in a children’s home. The above extract shows his experience of both peers and caregivers in the home was that they were dangerous and could not be trusted. This experience was consistent throughout James’s life:

I was only 16 and I was in a hostel. As I say Mary Jones, Social Worker, started a relationship with us and it was a bit more friendly and familiar than it should have been. We were kissing and cuddling and stuff and she used to travel once a fortnight to visit me.

This exerpt shows James’ experience of a female attachment figure who held a position of trust and exploited him to promote her own interests. This may have initiated a threat response in James at a time when he may have felt particularly vulnerable given his previous abusive experiences and lack of a positive attachment figure. This theme is especially pertinent for James as Mary Jones was the victim of his rape offence suggesting that people perceived as threats may become victims themselves.

Desire to Remain in Control

The superordinate theme of desire to remain in control is representative of a need for control in intimate relationships and patriarchal and misogynistic views of societal and familial constructs. The superordinate theme is made up of four subthemes: “need for control;” “real man;” “entitlement/women are objects;” and “male sex drive/policing partner.” The importance of being in control and being a “real man” were pervasive themes across participants:

My business went up in flames and she would throw that in my face, I lost money, she threw that in my face, she’d throw her previous relationships in my face, saying they had done this well, they had done that well. I think she just generally, she really put me down, put me down as badly as possible. I mean, as I say, I’d worked all my life kind of thing, even with drugs, I made a lot of money with that, it was like a business. Eh but since I started working at 23 I had my own wee company so I was hard working and she just kinda, made me feel really bad about no working, not earning more. – Greg.

Greg’s account of his relationship was that his partner made him feel like “less of a man” and inadequate and, perhaps more importantly, that he was made to feel like this, thereby clearly indicating that he believes these thoughts are not self-produced but rather forced upon him. Greg appears to be very proud of having his own business, albeit a criminal business, and having made his own money and so it appears that what it means to be a man to him is to be in control and have power in intimate relationships. Greg was convicted of rape and attempted murder of his partner and so it seems his response to feeling emasculated was to exert the ultimate exploitation and violation of his partner and yet he continues to portray himself in the victim role.

In accordance with need for control, themes of entitlement and women requiring policing from the partners emerged from the analysis:

There was trust there. If she said she was at home I’d accept that, if she said she was out with her mum I’d have to accept that. I wouldn’t question it because if that’s where you say you are that’s probably where you are. There were one or two occasions where I was kinda sitting thinking well I’m not sure you are right where you say you are who’s that guy’s voice in the background, why won’t you just answer your phone. –Ryan.

Ryan describes a positive aspect of an intimate relationship in which he states there was “trust,” yet he contradicts this by voicing his concern that he does not believe his partner and feels he is entitled to “police” her. Additionally, it is evident that if his partner does not answer her phone his automatic thought is that she is being unfaithful, further reinforcing his belief that his partner requires his policing to negate advances from other men.

Women’s Role in Violence and Sexual Assault

It was apparent that all participants experienced women as having a role to play in the violence and/or sexual assault that was perpetrated on them. The superordinate theme was composed of three subthemes: “women are provoking;” “rejection/abandonment;” and “women are supportive.”

She kept going on “you bought it (an apartment) so you are leaving me” and all that. So I said “well if you are going on like that I will leave you” and went away from her and maybe a week later I committed the index offence. –Kyle.

Kyle here describes a situation in which he feels his partner initially provoked him and that, ultimately, led him to committing the index offence, raping and assaulting his partner, a week later. It was Kyle himself who subjectively made the link between his partner provoking him and the index offence. The phrase “she kept going” and “all that” suggests he perceived his partner to be consistently provoking him and challenging his authority to make decisions as a “real man” and it seems he ruminated on this altercation for a week, at which point he committed the index offence. Interestingly, when describing the index offence later in the interview, Kyle denies it happened referring to it as “farewell sex” yet his use of language here “I committed the offence” contradicts this denial.

Women’s provocation and perceived challenge of the males dominant position was the experience for other participants also. Greg, who described feeling that nothing he did was “good enough” for his partner, explained an incident in which he learned from a female friend that his partner had been unfaithful to him:

when I got to her house, she still denied it, I then tried to drag her out of the house up to the neighbours house cause one of the boys she slept with is the neighbour and I was taking her to his house and in between there was a serious incident as such.

In this exerpt, Greg experiences his partner’s continued emasculation of him in addition to the ultimate insult and female provocation, having a sexual relationship with another male, which further fragments his image of himself as a “real man.” In order to assert his masculinity, he perpetrates violence and sexual assault to dominate his partner. This exerpt is also in line with the “nature of harm” subtheme as the “serious incident” which Greg refers to is the rape and attempted murder of his partner.

Diminishing Personal Responsibility

The theme that violence and sexual violence was in some way out of the control of participants was evident across all transcripts. This superordinate theme encapsulated two subthemes: “uncontrollability” and “nature of harm.”

The red mist had set in and the actual events of it are a bit of a blur. I remember hitting him a lot and I remember hitting her a lot and then I stormed off... I got sentenced for that, got 9 months. Got out was still very angry and very bitter about the whole situation. Went back to the flat to pick up some stuff that had been left behind and the red mist set in again and I ended up hitting her with a curtain pole. Again, I stopped. I had every intent of killing her at that point, I though “I don’t want you to live any longer” and then stopped. I couldn’t physically do it but I was thinking throughout the whole thing I want to kill you, I want to kill you and then thought wait a minute I don’t want to kill you. I don’t want to be in this house with a dead body, how am I going to explain this and oh shit what have I done. –Ryan.

The reference to “red mist” was pervasive in four of the transcripts and is representative of the participants’ perception that they have lost control of the situation and the experience that they themselves are victims to the violence and that it is no longer under their control. This contradicts Ryan’s experience of serving nine months for his previous act of violence and choosing to go to the home he shared with his partner knowing an altercation was most likely to take place. The phrase “I ended up hitting her” implies it was not his choice to do so, that it was somehow out of his control. Again, this challenges his account of his internal world, that he “had every intent of killing her” and so it seems it is more likely that this was his intention upon going to her home. It is also apparent that, in this incidence, it was fear of the negative consequences of her murder that stopped him as opposed to any concern for the victim.

The minimisation or outright denial of violence and sexual assault was present in all participants and seemed to serve a protective function in that it was linked to the “I am not like them” subtheme. This allowed participants to view themselves in a positive light:

I didn’t think I would hit her back because she was pregnant with my daughter by that time so... but I, I maybe slapped her about the face to try and wake her up because I knew when she was in that phase all she could see was her abuser it wasnae me she was hitting or a pillow or anything it was her abuser. And so it was getting that bad that the only way, I couldn’t punch her in the stomach or anything cos she was pregnant with my daughter and that, and I didn’t know what to dae so I gie her a slap in the face or something to try and bring her out of it but it wasn’t working. –Malcolm.

Here Malcolm describes his response to his partner experiencing flashback symptomotology in relation to being sexually abused as a child. He states that the only way he could manage this situation was through violence and does not explore other options. He minimises this violence by stating, “I maybe slapped her” and “giev her a slap in the face” as it is evident he does not consider these acts as particularly violent. Furthermore, it is apparent that he believes he was protecting his daughter, who his partner was pregnant with at the time, by only using minimal force to control the situation. The phrase “I could not punch her in the stomach” implies that this is something he would have done had she not been pregnant.

Perception of Self

The participants’ view of themselves and how they will improve their situation in the future and not engage in violent or sexually assaultive behaviour was pervasive in all transcripts. This view of themselves as “the new me” or “I am not like them” (the two subthemes composing the superordinate “perception of self” theme) seemed to serve as a coping mechanism while participants served a custodial sentence:

I just feel out of place. I don’t know how to explain it to yourself 100 % bit I just feel uneasy in here, I just feel different. I don’t know. Everybody says “you’ve done a sex offence the same as everybody else” but I dunno I cannae see it like that. I think there is a bigger picture and different people are going to take a different view when the truth comes out. –James.

James’ perception that he feels “out of place” and “different” when comparing himself to other sex offenders in the prison implies he views himself as internally distinctive from the others and this serves a protective function for him that he is not inherently “bad.” Interestingly, none of the participants expressed the view that they were intrinsically “bad” or that they should be punished for what they have done suggesting that they do not view themselves as abusers and that this gives them hope that they will experience positive interpersonal relationships in the future.

Discussion

The current research aimed to explore cognitions present in the IPV SO offender representative of their view of themselves, others, and the world. Analysis in accordance with IPA methodology revealed the emergence of 14 themes thought to be representative of implicit theories in IPV SO offenders. These are: “violence is acceptable;” “grievance/revenge;” “dangerous world;” “need for control;” “real man;” “entitlement/women are objects;” “male sex drive/policing partner;” “women are provoking;” “rejection/abandonment;” “women are supportive;” “uncontrollability;” “nature of harm;” “the new me;” and “I am not like them.” The majority of these themes correspond to implicit theories proposed in IPV offenders, sex offenders, and child sex offenders as outlined in Table 3 suggesting that there is great overlap in the cognitions of specific offending behaviour groups. The emergence of additional themes, limitations of the current study, and clinical implications are discussed.

Table 3 Implicit Theories identified in Child Sex Offenders; Rapists; Sexual Murders and IPV Offenders
Full size table
Table 4 Table of Superordinate and Sub Themes and Prevalence of Subthemes
Full size table

The themes “rejection/abandonment;” “women as supportive;” “need for control;” and “real man” seem to be cognitions explicit to IPV and IPV sex offenders. The theme of abandonment is consistent with Mitchell and Gilchrist (2006) who found that fear of rejection is a pertinent risk factor in triggering abusive behaviours in male offenders of IPV, especially those with a diagnosis of borderline personality disorder (Mitchell and Gilchrist 2006). This theme, therefore, has implications for risk assessment and management of male offenders of IPV, as when they perceive their partner to be rejecting them or attempting to leave to seek safety, their partner will be at most risk (Dutton 2002).

The theme of “dangerous world” was not previously identified as an implicit theory in the Weldon and Gilchrist study (Weldon and Gilchrist 2012). It was however, an emergent theme in previous sex offender literature (as outlined in Table 3) and by Dempsey and Day (2010) in a paper focussing specially on ITs in domestic violence perpetrators. This theme relates specifically to the view that an individual must be protective of oneself from possible exploitation by others and that no-one should be trusted. It was evidenced in all but one of the participants in the current study and it was apparent that it was a schema which had developed at a young age due to adverse environmental, familial, and attachment experiences.

Two themes emerged in the current study which were consistent with the implicit theories proposed by Dempsey and Day (2010), that of “the new me” and “I am not like them.” The “new me” theme postulates a protective function for the participants in the current study, an idealistic view of the future, and is consistent with Dempsey and Day’s “create a better life” sub theme. This theme relates to a sense of wanting to be redeemed and striving for a second chance and is in line with the “wanting to change theme” which emerged in Blagden et al. (2011) when qualitatively exploring denial in sex offenders. This theme, therefore, has potential implications for intervention given its concordance with the Good Lives Model of offender rehabilitation (GLM; Ward and Stewart 2003). The GLM is a strength-based approach and proposes that, from a humanist perspective, people offend as they are trying to secure a valued outcome in their life. As such, people offend when trying to achieve an inherently human goal: however the desire or goal is achieved in a harmful and anti-social way due to dynamic interaction between a range of deficits and weaknesses in the offender and their environment (Ward and Stewart 2003).

The “new me” theme was closely linked with the “I am not like them” and “nature of harm” themes which similarly seem to serve as a coping mechanism for participants and was recognised as such by Dempsey and Day (2010) whose analysis revealed a similar theme of “I am a good person.” This theme seems to provide a means of cognitively distancing themselves from the acts they committed with the belief that they are not “like” other sexual and violent offenders and that they are not “bad” people. Consistent with this view of themselves as not like “other” sexual or intimate partner offenders, all participants in the current study greatly minimised or completely denied their sexual offence conviction. This was often in contrast to admitting and describing violent assaults in detail when this involved other males and hence reflecting the “grievance/revenge” and “real man” subthemes. This may reflect a desire for the participants to be seen in a certain way by the female researcher, i.e. “I am not bad.” Perhaps more likely, however, and consistent with Blagden et al. (2011) is that minimisation and denial serves as a protective function for the participants. Therefore, denial could be seen as a narcissistic coping mechanism that, if removed, may lead to a challenge of their cognitive dissonance, that is their belief that they are “not bad”. To overcome this, participants must want to address their denial and so may benefit from a readiness to change framework as that offered by Burrowes and Needs (2008). Finally in relation to the emergence of the themes related to perception of self, participants may have a personal motivation for wishing to appear as “changed” in fitting with the “new me” construct. This motivation may be driven by their position in the custodial system and their desire to be seen in a positive light. Participants’ motivation to engage in the research process is unknown. However, it may be related to the opportunity to construct their own narrative identity to a person they perceive to be “part of the system.” Conversely, Dobash and Dobash (2011), when investigating cognition in intimate partner murderers, found that complete denial was used as a means of avoiding intervention and treatment while serving a custodial sentence. This seems to contradict the current study in which participants self –selected to participate in the research and had completed various offending behaviour programmes and so is perhaps linked to the desire to be perceived as “the new me.”

Limitations of the Current Study

Firstly, with regard to limitations of the current study, the sample size was small and self-selecting (given that each participant volunteered to be interviewed) which makes it difficult to generalise the results to wider IPV SO population. The small sample number also means it is difficult to comprehensively establish if saturation of the data has been achieved. There was, however, a marked consistency in the participants’ narratives which was evidenced by the prevalence of ratings of each theme with 12 of the 14 themes identified in at least eight of the participants (66 %). Smith (2011) noted that, for a theme to be classed as present, it should be evident in at least three of the participants when the number of participants is eight or above. Consequently, this suggests that homogeneity does exist among men who perpetrate IPV, intimate partner sexual assault, and child sexual assault. It should also be noted that, while participants in the current study had convictions for IPV, intimate partner sex offending, and child sex abuse, there may have been more offences in this context for which they were not charged. This is especially pertinent given that IPV and intimate partner sexual violence continue to be under-reported (Office for National Statistics 2015).

Secondly, there is the issue of the subjective role of the researcher in the research process (Smith 2004). This was especially pertinent in the current study given the researcher’s familiarity with the domain of IPV. This was minimised, however, by three independent researchers familiar with IPA methodology, but with limited knowledge of IPV, analysing three distinct transcripts and coding for themes to increase validity and for the purposes of triangulation. There was a high inter-rater reliability with all themes with two raters identifying 11 of the subthemes outlined by the original researcher and one identifying nine from one transcript.

Participants’ motivation to engage in the research process may have been a factor in what they choose to discuss in their interview given, at the time of interview, given that they were serving a custodial sentence. Participants may have used the interview as an opportunity to portray themselves in a positive light resulting in the themes of “new me” and “I am not like them.” Every effort was made to minimise any such issues by assuring the participants their participation was entirely voluntary and had no implications on their sentence length or participation in offending behaviour groups. Furthermore, the issue of the demand factor may be more prominent in this setting given participants may have anti-authority attitudes and perceive the researcher as part of the system contributing to their sentence. This is in fitting with the implicit theory of “uncontrollability” given participants voiced an external locus of control with regard to their current custodial sentence. They did not attribute it to their own offending behaviour which they perceived to be out with their control and due to external factors.

Finally, with regard to limitations, there is consideration as to the role of the female researcher in the process and participants’ responses to this given that they all have a charge of violence and sexual assault against a female. As described in the analysis section, every effort was made to minimise any impact of the female researcher on what the participant chose to discuss by adopting a non-judgemental approach and adhering to the semi-structured interview procedure.

Clinical Implications and Future Research

Four of the participants in the current study were convicted of sexual assault against a child highlighting the link between the aetiology of IPV sex offenders and child sexual abusers which has been emphasised in the empirical literature by Morgan and Gilchrist (2010) and Hester et al. (2007). This link, therefore, has implications for risk assessment as to who may be at risk when working with an offender presenting as an IPV SO/IPV or child sexual abuse (CSA) offender. Other potential familial victims should be considered when assessing such an offender. The Spousal Assault Risk Assessment (Kropp et al. 1999 is routinely used to assess for risk of intimate partner assault. However, given the links between IPV and CSA (Hester et al. 2007) the additional use of the Child Abuse Risk Evaluation (CARE; Agar 2002) may have clinical utility. In the future, the development of a risk assessment tool incorporating both risk of IPV and child abuse/sexual assault may prove clinically useful and increase efficiency.

When considering intervention with IPV sex offenders, it is apparent that this distinct group of offenders hold implicit theories which make them mistrustful and suspicious of others’ intentions. This may have a bearing on the development of a positive and supportive therapeutic alliance. Consequently, the implementation of schema therapy with this group of offenders may allow for implicit theories to be identified by the therapist early in intervention to improve the development of the alliance from a relational and cognitive perspective. Ward (2000) points out that, recognition of underlying implicit theories associated with offending, as opposed to simply identify individual beliefs, is essential in treatment of child sex offenders. Schema based therapy is currently being employed by the UK prison service in England and Wales with child sex offenders and so this could potentially be implemented with IPV sex offenders also. There is evidence that schema-focussed therapy can reduce distorted schema-driven beliefs (Beech et al. 2012). In order to do this successfully, the aetiology of problematic implicit theories in IPV sex offenders could be incorporated into pre-treatment assessment and the intervention itself. The implicit theories themselves can be translated into dynamic risk factors and treatment targets in a meaningful and empirically defensible way (Dempsey and Day 2010). This is also a possibility when considering the emergent theme “the new me” which could clinically adhere to the Good Lives Approach, a strengths based model. This model focuses on developing pro-social goals that are achievable for offenders and the development of skill acquisition that would make these goals attainable.

Future empirical research would benefit from establishing implicit theories held by non-violent non-sex offending males to allow for a comparison and potentially identify further treatment targets and skill acquisition. As previously mentioned, little is known about the distinct cognitions present in female perpetrators of IPV and so future research would gain from exploration of this empirically and aid in risk assessment and interventions with this offending behaviour group.