Did you know that a number of BJPsych Advances articles are translated into other languages? Follow this link...

BJPsych Advances
Domestic violence: working with perpetrators, the community and its institutions
Neil Blacklock

Every year one in ten women in the UK is subjected to physical abuse from a partner or ex-partner, while one in four experiences such abuse at some point in their lives (Mooney, 1993; Morley & Mullender, 1994a). Domestic violence is one of the most pervasive of all social problems, affecting most of the population directly or indirectly (Box 1). I cannot think of a single close friend or colleague whose life has not been touched by this largely hidden problem.

Box 1.

Pervasiveness of domestic violence against women

One in four women experiences domestic violence at some time in their life, with two women per week killed by their current or ex-partner. Women are victims of 70% of domestic incidents (Mirlrees-Black et al, 1998)

Of women subjected to domestic violence in any year, one in nine sustains injuries serious enough to require medical attention, 10% report being knocked unconscious and 5% sustain broken bones (Stanko et al, 1997)

Domestic violence is a factor in one in four suicide attempts by women (Stark et al, 1979)

A study by the National Children's Homes Action for Children (1994) found that 75% of mothers subjected to domestic violence said their children had witnessed it

It is striking that for such a far-reaching social problem our responses as a community have been so slow. The first refuges for women escaping domestic violence were set up in the early 1970s, but refuge provision remains inadequately funded. This would appear to indicate, at best, inertia and at worst, collusion with such abuse at both a social and an institutional level.

Against such a background, it is hardly surprising that direct work with perpetrators of domestic violence is a new field. The Domestic Violence Intervention Project (DVIP) has been running programmes for perpetrators of domestic violence since 1992 and it is one of the longest running projects of its kind in the UK. It is based in Hammersmith, London, with a branch in Peterborough.

The DVIP was developed at a time when there was virtually no specialist provision in the UK for addressing men's systematic use of abusive behaviours within relationships. We had to look to other countries for much of our early influences and learning. In particular, we used the experiences of projects in the USA (Domestic Abuse Intervention Project in Duluth; Emerge in Boston; Manalive in San Rafael, California) and New Zealand (the Hamilton Abuse Prevention Project). Alongside learning how to construct and deliver perpetrator programmes, we learnt that intervening with perpetrators of domestic violence was about intervening in both the social and the institutional context in which the abuse occurred.


Many of the principles underpinning the work of DVIP are shared by other projects working in this area, and they are enshrined in the ‘Statement of principles and minimum standards of practice’ of RESPECT (the National Association for Domestic Violence Perpetrator Programmes and Associated Support Services). Such projects within the UK deliver programmes based on cognitive–behavioural theory, informed by social learning theory and feminist understanding of domestic violence. Of course, this is not the only way to construe domestic violence, and in such a new field it is important that practitioners are able assess the value of other approaches for their particular clients. However, they should not lose sight of the clarity of the RESPECT statement concerning responsibility (Box 2).

Box 2.

The RESPECT opening statement of principles (2000)

Men's violence towards partners can include physical, sexual, emotional and other forms of abuse. It is the direct consequence of a fundamental structural inequality in the relationship between men and women rooted in the patriarchal traditions that engender men's beliefs in the need to secure and maintain power and control over their partners. From this perspective, men's violence is defined as learned and intentional behaviour rather than the consequence of individual pathology, stress, alcohol abuse or a dysfunctional relationship.

Who uses violence towards partners?

The vast majority of those who use violence and other abusive behaviours to control and dominate in relationships are heterosexual men. The majority of those arrested for domestic violence are men who have assaulted women, and domestic violence accounts for 25% of all violent crime in the UK (Mirlrees-Black et al, 1998). The majority of those injured and killed in domestic assaults are women: every week two women are killed by their current or former partner (Mirlrees-Black et al, 1998).

The work of DVIP focuses on men's violence towards women with whom they have or have had an intimate relationship. For the purpose of this article the terms perpetrator and abuser are given a masculine pronoun.

From our experience and that of specialist programmes in other countries, it is obvious that offering programmes for perpetrators influences more people than just those men involved. It affects their partners and other family members (not always in ways that increase their safety) and it affects the responses of other professionals (not always in ways that support our contention that the perpetrator is held accountable).

The wider responsibilities of perpetrator programmes

Offering a service that appears to give the hope that the perpetrator may change his abusive behaviour has significant safety implications for his ex-partner. Most of us will have stayed in a relationship beyond the point where we realised that it was not right to do so, often on the basis that if we both worked at it, things would get better. Similarly, when a man starts attending a perpetrator programme his partner may decide to stay in or return to the relationship. However, her hope that he will change may not be realised, and consequently she will find herself in a relationship that is dangerous. A study looking at factors affecting women's decisions to return to an abusive partner found that whether he was in some form of counselling carried a greater weight than any other single factor (Gondolf, 1988).

A domestic violence perpetrator seeking control over his partner may use a range of tactics to achieve it. These include manipulation of his attendance at a programme and misrepresentation of the programme's contents and messages. Through working with perpetrators we are trying to increase the safety of, and reduce the risks to, their partners and children. It is the responsibility of those working with perpetrators to ensure that information and support is provided to the partners and others affected by the abuse (Box 3). It is also the responsibility of practitioners and their agencies to challenge institutional collusion with abusive behaviour. Failure to provide support and information to the partners and failure to ensure that the intervention does not reduce the perpetrator's accountability for the violence undermines the aim of the work.

Box 3.

The responsibilities of perpetrator programmes to the victims of abuse

The support offered to the female partner or ex-partner of a man attending a programme should:

  • increase her safety in practical ways

  • improve her mental and emotional health and well-being

  • give clear messages about the abuse and develop her understanding of it

  • promote realistic expectations of the perpetrators' programme

  • promote in general women's empowerment and working for change

Practitioners considering whether to begin working with perpetrators of domestic violence therefore need to consider whether they and their agency are willing and able to take responsibility for the essential collateral work. Practitioners are working to end violence and therefore share a responsibility for creating an environment that supports this goal. It is common for practitioners to avoid responsibility by saying that they do not have the resources to undertake partner support and interagency work. As I see it, either you have the resources to carry out all asepcts of this work (including partner support and interagency work) or you simply do not do it at all. Given the dangerousness of many domestic violence perpetrators, this is not an area in which practitioners should be half-heartedly dabbling.

Work with perpetrators of domestic violence

It is DVIP's experience that there are four key elements to any effective programme. First, the programme must challenge the perpetrator's denial mechanisms and the gender-based assumptions supporting his abusive behaviour. Second, programmes can be delivered safely only when accompanied by safety-oriented support and empowerment work with those affected by the perpetrator's violence. Third, other professionals must give the message that domestic violence is unacceptable and hold perpetrators accountable for their behaviour. Finally, programmes cannot bring about change in the men attending them if there is not a growing intolerance of the abuse of women within the community as a whole. A most important and effective part of the work of running groups for perpetrators is the interlinking of a range of interventions on a number of fronts working towards safety and change.

It has become increasingly evident (Shepard, 1999) that the safety of those experiencing abuse depends on how the above four factors intertwine. The wider impact of perpetrator programmes seems as important, if not more so, than any change they bring about in individual abusers.

Characteristics of male domestic violence perpetrators

The motivation and rationale of each domestic violence perpetrator is a complex interplay between power and gender on an individual, family, social, institutional and cultural level. Yet perpetrators attending the DVIP demonstrate a number of common factors (minimisation, denial of responsibility and a sense of entitlement – see below) that appear to be central to their abusive behaviour. These factors all have representations at a social and institutional level as well as within the individual perpetrator.


The perpetrator plays down or does not face up to (compartmentalises and ignores) aspects of his abusive behaviour, minimising its extent, frequency, seriousness, impact and consequences. Minimisation is encountered in the following forms.

Complete denial

The perpetrator denies the abuse entirely, saying that it did not happen.

Exclusion and inclusion (Hearn, 1995)

The perpetrator excludes from his account of his behaviour:

  1. particular acts because they are not included in his internal definition of violence and abuse; most frequently this will mean omitting actions such as pushing, holding, blocking and throwing things;

  2. sexual abuse or sexual violence, which perpetrators frequently view as different from domestic violence;

  3. abusive behaviour directed towards children;

  4. non-physical abusive behaviours, which he is unwilling to address.

He includes (acknowledges) only abuse that has become public, often through the criminal justice system.

Forgetting, blanking out and not knowing

The perpetrator begins his account of the violence with phrases like “It all happened a bit quick”, “I can't really remember” or “I don't know what we were arguing about”. This vagueness is most frequently an attempt to obscure his violence and its meaning both from himself and from others.

“It's not really me”

Here the perpetrator's image of someone who uses violent behaviour is of someone whose acts of violence are more life-threatening than his own acts. This form of denial presupposes that it is other people who behave abusively (and therefore need to seek help), but not himself. It is revealed in phrases such as “I'm not a violent person”, “I'm not one of those wife beaters” and “Everyone's overreacting – it's not like I beat her up every week”. A variation of this particular form of distancing occurs when men have recently left the relationship or attended a perpetrators' programme. Such men will often say “I used to be violent” or “before I came to DVIP I would have been violent, but now…”.

Normalising (Trew, 1979)

He presents his violence as if it were of no great significance, using phrases such as “it was only” and reducing a set of violent actions to “we had a row”.

Denial and minimisation of the impact of his behaviour on others (Dutton, 1995)

Often the perpetrator will objectify his partner, denying her humanity. This is undoubtedly a central process in domestic violence. He sees his partner as ugly or evil and thinks of her in terms of (often gendered) derogations such as a bitch, slag and cunt. This is part of a process that allows him to use violence by reducing his empathy for his partner. With many perpetrators this objectification constantly pervades their view of their partners.

Loss of partner empathy also occurs when the perpetrator's expectations of services (physical and emotional dependence) and narcissism are such that it becomes increasing difficult for him to see her as having experiences and needs separate from his own. For example, he cannot see that she called the police for her own protection rather than to punish him. If men in the early stages of the DVIP programme are asked to name one quality that their partner or ex-partner has which is not about her relationship to them, most find it difficult to reply. They struggle to see her beyond her role in relation to themselves and their home, and give responses such as “she stands by me”, “ she puts up with me”, “she keeps the place tidy” and “she's a good mother”.

Denial of the effects on children

Even when a perpetrator has systematically abused his children's mother for many years he may still claim that he is a good father, unable to see the hypocrisy in this statement.

In 90% of domestic assaults, children are in the same or next room (Hughes, 1992). Furthermore, we know that where domestic violence occurs there is an increased likelihood of child abuse in the same family (and vice versa) (Morley & Mullender, 1994b).

Owing to the taboos and consequences of disclosing direct physical and sexual abuse of children, perpetrators of domestic violence rarely talk about it. This means that this area must be discussed, regardless of whether such abuse is disclosed.

Denial of responsibility

The perpetrator denies control over his actions, placing control elsewhere, external to himself. The ways this is most commonly encountered are as follows.


The actions or inactions of the partner or victim are most frequently cited as the cause or provocation of the violence. This is heard in many different ways in perpetrator's accounts of their abusive behaviour: “She pushes me too far”, “She winds me up”, “She knows what I'm like”, “If she didn't do x I wouldn't do y”, “She goes on and on (nagging)”.

Sometimes the violence may be presented as a last-ditch, regrettable step to protect the perpetrator from his partner's verbal or physical assault, but in fact it was either retaliation (punishment, not self-defence) or the silencing of his partner's criticism.

Attributing the reason for abusive behaviour to the woman can become entrenched in domestic violence perpetrators, to the point where remorse and culpability are extinguished. While the perpetrator continues to place responsibility for his behaviour with his partner, motivation to work for change within himself will remain low. Often, partner-blaming is linked to an exaggerated sense of entitlement and anger at her resistance to meeting his demands. He will want to talk about her behaviour and his feelings, and any useful intervention needs to reverse this focus to look his behaviour and her feelings.

Substance misuse

This is most commonly heard as “I'm alright when I haven't had a drink”. It is undoubtedly true that drink and drugs affect people's perception and behaviour, otherwise there would little point in using them. However, their relationship to domestic violence is far from simplistically causal (McKenry et al, 1995).

Drink lowers inhibitions and can be a contributing factor to the severity of assaults, increasing the likelihood that a perpetrator will use greater force than he would if sober (e.g. a slap becomes a punch). Alcohol can also be a precursor to or a part of the act of abusing. Experiencing feelings of hostility and anger causes the perpetrator to start drinking, which becomes part of the build-up to an assault. Alcohol both provides a readily available excuse for denying responsibility and acts as a disinhibitor. The ritual of drinking before violence can be seen clearly in football-related violence, as well as in bars up and down the country. In perpetrators of domestic violence, alcohol and drugs are not themselves causal factors in their aggression. Perpetrators with substance misuse problems have two problems and both should treated in tandem.


Perpetrators usually start an account of their assault of their partner with words like “Well, I'd had a really difficult day at work” or “We've been having lots of money problems”. This scene-setting of the pressure in the abuser's life is an attempt to make his actions understandable, given the circumstances. It presents a picture of the man as a pressure cooker that blows when the pressure reaches a certain point: this is both simplistic and mechanistically misleading. The question we must ask is what it is about this man's construction of his relationship with his partner that permits him to use abusive behaviour towards her when he feels himself to be under pressure. When looked at closely, the ‘excuse' of stress/pressure is either that he perceives her as its cause (partner-blaming) or as the person who should make him feel better (sense of entitlement).

Temporary or long-standing psychological disturbance

Temporary This is reflected in men's accounts of their violence as “I've always had a short fuse” and its many variations (“I just snapped”, “saw red”, “lost it”, etc.). The perpetrator presents his violence as an overwhelming explosion of adrenaline and emotion, which is a commonly held understanding of violence and one that is frequently enmeshed with notions of provocation (partner-blaming). In this form of denial of responsibility the violent or abusive behaviour is seen as a ‘momentary insanity’, out of character and out of control. Yet examination of the actions of almost all perpetrators reveals control in the level and type of violence used and clear choices in where, to whom and in what circumstances it occurs.

Perpetrators of domestic violence find comfort in the momentary insanity explanation for their behaviour. It is more acceptable within constructions of masculinity to be seen as having a bit of a temper (“I don't put up with any shit”) than to have a problem in relating respectfully to women. Theories of anger management and impulse control play into this and are therefore an inappropriate response to domestic violence perpetrators.

Long-standing Here the perpetrator presents his violence as rooted in his experience of abuse in childhood and not related to his current construction of his relationship with his partner. It is true that domestic violence is learned and that one powerful site of learning will have been the man's own family of origin. Many of the men attending DVIP have experienced or witnessed domestic violence as children. But so have many of the staff team and most of the project's women clients. It is certainly true that experiencing abuse is damaging, and the more severe and frequent the abuse the greater the likelihood of future psychological disturbance. However, the link between experiencing abuse (including witnessing violence between parents) as a child and using abusive behaviours towards a partner as an adult is far from clear. Theories about cycles of abuse with domestic violence perpetrators are again simplistic and unhelpful in designing perpetrator programmes.

Some perpetrators of domestic violence will have experienced severe abuse during their childhood that was frequent and damaging of their self-esteem and ability to develop intimate relationships in later life. Often these men have long-term substance misuse problems and a history of criminal activity and violent assaults, both within and outside of close relationships (Saunders, 1993). They present as low in minimisation, empathy and remorse, and require a high level of risk management control. Fortunately, they are just a small percentage of the population of domestic violence perpetrators.

The vast majority of perpetrators do not fall into this subgroup, although they frequently perceive themselves as victims and view their abusive behaviour as a response to their own ‘persecution’. Their wish to explore their experience of being ‘wronged’ is a mechanism to avoid seeing their own abusive behaviour for what it is and to shift responsibility for that behaviour.

If the perpetrator has experienced abuse himself, it is useful briefly to explore the impact of this on him in order to draw parallels between this and the experience of those whom he abuses. More in-depth discussion of his experience of abuse should be avoided, otherwise there is a danger of playing into his sense of victimhood. When all physical violence has stopped for some time (1 year) and other abusive behaviours have reduced dramatically, it might be approprate then for the client to address how his experience of abuse has affected his responses in intimate relationships.

Denial of responsibility for consequences for his behaviour

Here the perpetrator blames others when he experiences consequences of his abusive behaviour. This is a variation of partner-blaming that extends to others who challenge his behaviour, including the police, courts, social services and, frequently, his partner's family (for siding against him or never having liked him).

Sense of entitlement (masculinity)

Violence and other abusive behaviours are used to control. Domestic violence perpetrators use abusive behaviour to support their sense of entitlement by:

  • punishing the woman for something she has done wrong (“teaching her a lesson”)

  • forcing her to do something she does not want to do (e.g. “shut up”)

  • stopping her from doing something she wishes to do (e.g. leave the relationship).

This is establishing the power to set the further rules of the relationship (e.g. he can drink, spend money, come and go as he pleases without criticism). Through this the perpetrator also obtains services from his partner (e.g. sex, housework, emotional care) and restricts her movements and autonomy. These expectations are deeply enmeshed with gender and identity, with what it means to be a man and, in particular, a man in a relationship with a woman.

Despite the fact that the perpetrator knows that his violence is wrong, this sense that entitlement is bound up with received ideas about gender identity allows him to see his behaviour as ‘reasonable’ given his partner's ‘unreasonable’ resistance to his expectations. This further fuels the process of partner-blaming.

There is a danger here that we see domestic violence perpetrators as caricature sexist males. Every man I have worked with has had a unique combination of expectations of his partner. Many perpetrators are socially skilled and present as wanting only what is best (in their own estimation) for themselves and their partner. However, all perpetrators use abusive behaviour to enforce the ‘rightness’ of their position.

Programme design

The above characteristics are present in an ever-changing mix of minimising, partner-blaming and expectations of entitlement, and programmes must accommodate this variation. They must also be designed to engage with men, enabling them to face the reality of their behaviour, accept responsibility for it and critically examine their expectations of service from and authority over their partners. For the reasons given in Box 4, group work is the preferred setting for this to take place.

RESPECT recommends that programmes should be a minimum of 75 hours over a minimum of 30 weeks. Considering that we are asking perpetrators to change what is frequently a long-standing way of behaving and to reassess a major aspect of their identity, this minimum duration is optimistic.

Box 4.

Reasons for working in groups

Group leaders can challenge members‘ attitudes to women and set boundaries for the group. Members are often resistant to this challenge, and the group setting allows it to be made without it becoming adversarial, while the the other group members can maximise support for change

Group members learn from each other, both directly and indirectly

When among peers, men learn and have re-enforced many of their ideas about what it is to be a man. The therapeutic group provides the ideal environment in which to deconstruct and challenge these social constructions

Denial is aided by the hidden/private nature of most domestic violence. Through the use of drama techniques in the group setting, some of the reality of the perpetrator's dangerousness can be brought into the room and reflected back to him, through the responses of group members

Process of working with perpetrators

Iwi & Todd (2000) have developed a model of working with perpetrators that involves a series of stages that deconstruct most of the levels on which domestic violence is supported. These stages are designed to be worked through in order. But each step need not be completed before moving on to the next and many will need to be addressed concurrently.

Step one involves helping the client to acknowledge his violence and abusive behaviour, challenging minimisation and creating within him the motivation to stop abusing. A full history of the client's violence and abuse is elicited, to enable him to accept the seriousness and impact of his behaviour.

Step two is to increase the client's awareness of his build-up towards violence, teaching him to recognise his growing hostility and anger towards his partner and to intervene with himself to stop it.

Step three is to reduce denial in all its forms (including denial of responsibility), so that it becomes harder for the man to avoid feelings such as guilt, shame and remorse. This involves exploring the barriers to honest disclosure, increasing empathy through hearing women's and children's experiences and accepting the gains and losses of giving away responsibility.

Step four is to develop the man's understanding that his violence is part of a system of abusive behaviour. This involves extending his definition of abusive behaviour to include non-physical abuse, and making him understand that the same belief systems and processes underlie both physical violence and non-physical abuse.

Step five focuses on helping the man to reassess and change expectations that underlie his use of violence and abuse. He is encouraged to question whether his expectations are universal, rational, fair or realistic. The group process is particularly useful in this. During this step we would look in detail at the social and familial aetiology of the man's belief system.

Step six is to increase his ability to use new non-controlling behaviours. This step involves rehearsing situations in his life, developing non-abusive ways for him to deal with his feelings and enabling him to feel angry yet remain respectful.

Step seven is to work with the client to enable him to acknowledge and express vulnerable feelings, which are often submerged by his anger, and to encourage him to develop support systems that do not collude with his abusive behaviour.

Step eight, the final stage, is to increase the man's ability to reflect on how and what past experiences contribute to his current feelings, reducing his projection of past experiences onto his current relationship.

This model of working is applicable to most, but not all, domestic violence perpetrators. Individuals with sado-masochistic behaviour, psychosis, morbid jealousy and the excessive use of projection as a defence mechanism pose difficulties for the model.

Do perpetrator programmes work?

If the success we are looking for in perpetrator programmes is that the men attending become less abusive then it is hard to say. The number of variables makes it difficult to obtain clean data. Many evaluations have suffered from problems of small sample sizes, lack of random assignment and control groups, short follow-up periods and inadequate follow-up data. Among the evaluations considered to be methodologically sound, the majority have found modest but statistically significant reductions in recidivism among men participating in perpetrator programmes (Healey et al, 1998, p. 14).

In the quest for the most effective programmes, it is important that we understand what enables abusers to move away from controlling to more egalitarian relationships. However, current attempts to find better programmes, including Home Office research (Pathfinder Programmes), are failing to recognise adequately that it is the layering and interlinking of interventions that produce the conditions in which individual change is fostered and sustained.

As stated earlier, interventions with perpetrators do not occur in isolated experimental conditions. If the criminal justice system fails to hold men accountable, if women-blaming attitudes exist within agencies and if there is insufficient support and protection for abused women, then any attempts to get individual men to end their abuse are undermined. If, on the other hand, we are able to create effective multi-layered responses to domestic violence, looking at the effectiveness of perpetrator programmes in isolation will be missing the point. These factors have led to calls for evaluations of perpetrator programmes that address their wider impact in fostering interlinking interventions (Tolman, & Edleson, 1995; Gondolf, 1997; Shepard, 1999).

The cost of domestic violence to our community and to the quality of all our lives requires that we take very seriously the need to bring about change in those who use such behaviour.

Multiple choice questions

  1. Most perpetrator programmes, including DVIP, are influenced by:

    1. feminist theory

    2. family systems approaches

    3. social learning theory

    4. cognitive–behavioural approaches.

  2. A domestic violence perpetrator's sense of entitlement is:

    1. an expectation that his partner will not challenge his opinion

    2. an expectation of authority over his partner

    3. a gender-based assumption of superiority

    4. an attempt to compensate for poor self-image.

  3. The risk of providing programmes for perpetrators without adequate support for their partners is that:

    1. she may return to or stay in a dangerous relationship because he is in a programme

    2. he will misinform her about the programme in ways that undermine her

    3. she will not be able to support him in the changes he is trying to make

    4. an opportunity to increase her safety will have been missed.

  4. Group work is preferred for perpetrators because:

    1. it reduces their isolation

    2. the public acknowledgement of their private abuse reduces their denial

    3. groups enable deconstruction of socially constructed attitudes towards domestic violence

    4. groups enable maximisation of support systems.


View this table:

MCQ answers