Domestic violence as a form of child abuse: Identification and prevention


You are in an archived section of the AIFS website 


Content type
Policy and practice paper

June 1994


Marianne James


The issue

Child abuse and domestic violence are not new social problems, nor are they haphazard or deviant activities. Both are considered an accepted part of our culture. However, while domestic violence is now politically recognised as one of the most entrenched and pervasive forms of violence in our society today, child abuse issues have yet to receive the same degree of attention. This is despite the fact that every year in Australia, thousands of children, as well as women, suffer physically, psychologically and sexually as a result of acts of violence against them in the home. Not surprisingly, these issues have essentially been regarded as a private matter. This, together with the fact that the family has traditionally been regarded as a source of love and support, has led to an element of denial at both a community level and an individual level. However, the family, as the most important institution within our society, is extremely complex. It is both an agency of social control and an agency of social support, and that its internal dynamics can have both positive and negative effects on relationships.

One negative aspect of family life is the effect of domestic violence on children, either as witnesses to, or victims of, the conflict. The increasing recognition of this specific aspect of domestic violence owes much to the work of the women's refuge movement. It was through the reports of refuge workers that people first became aware that children could be severely traumatised by witnessing domestic violence and, indeed, that the children may also be victims of this violence. Concern has also been expressed by health and welfare professionals, who have felt the frustration of not being able to intervene legally to protect child witnesses from extreme incidents of domestic violence, unless there was also evidence of child abuse.

In Australia, there has been an historical separation of domestic violence and child protection issues. Domestic violence was brought out into the open by the women's movement in the late 1970s, and has commonly been applied to various forms of violent and abusive behaviour which occur in a marriage or de facto relationship. At an institutional level, domestic violence has been regarded as a matter for the police, courts, women's refuges and other women's support services. In contrast, child abuse refers to the physical, sexual or psychological damage caused to the child by the abusive behaviour of others, or the failure of others to protect a child from such damage and has been more of a health and welfare issue. Child protection has therefore involved an additional group of people as well as additional legislation.


It is difficult to establish with any accuracy the extent of domestic violence. This is primarily the result of two factors. The first is that very little official data has been collected, while the second is the low rates of reporting by those affected. In fact, some estimate that only one in ten report, whilst those who work in refuges for women subjected to violence would place that proportion much lower at one in 50 or one in 100. (These reporting rates were presented informally at the Second National Conference on Violence held in Canberra in 1993 as cited in Easteal 1994).

This reluctance to report incidents to others, especially authorities, stems from a concern to avoid the legal consequences of criminal behaviour as well as the guilt and shame usually associated with the experience. In many relationships, there are also powerful feelings of denial, together with a minimisation of the true nature of the abuse (Queensland Domestic Violence Task Force 1988). However, by piecing together various statistics, some idea of the seriousness of the problem can be gleaned.

During the month of November 1993, there were over 2200 calls to the New South Wales Police that related to domestic violence, and in the same month in Victoria about 1200 sought assistance from the police in relation to violence within the home (Easteal 1994). From August 1987 to August 1988, women's refuges accommodated approximately 11,000 women and children, and turned away approximately 23,000 because of lack of space (NSW Domestic Violence Committee 1991). In the Australian Capital Territory, out of a population of 100,000 women, about 3,000 (3 per cent) contact the Domestic Violence Crisis Service every year and the police call-outs to such incidents number more than one thousand (Mugford et al. 1993). However, in the United States, Canada and the United Kingdom (Jaffe et al. 1990, MacLeod 1987, Dobash and Dobash 1979) it has been estimated that as many as ten per cent of all women are victims of domestic violence.

Child Witnesses to Domestic Violence

In Queensland, 88 per cent (580) of the 856 respondents to the phone-in conducted by the Queensland Domestic Violence Task Force reported the presence of dependent children in the household during the course of the violent relationship. Ninety per cent of these respondents reported that the children had witnessed the domestic violence, and a further 74 per cent of these respondents had spoken with their children about the violence (Queensland Domestic Violence Task Force 1988). In Western Australia, the Domestic Violence Task Force found that 84 per cent of the 420 respondents to a newspaper survey had children living in the same household as the abusive partner. In a phone-in conducted at the same time, almost 87 per cent of the 297 respondents with children reported that their children had witnessed them being abused (WA Domestic Violence Task Force 1986).

The figures illustrating a high incidence of child witnesses to domestic violence are reinforced by Walker (1984) who also reported that 87 per cent of children were aware of the violence between adult partners, while Dobash and Dobash (1984) in a study of 314 first, worst and last attacks of violence recalled by victims, found that 58 per cent of the attacks took place in front of the children. Sinclair's research (1985) based on clinical experience has suggested that if children are in a violent family, 80 per cent of them will witness an episode of wife assault. What they witness may range from a fleeting moment of abusive language to a homicide (Bowker, Arbittel and McFerron 1988).

A review of Victoria's domestic violence legislation between 1987 and 1990 has also shown some alarming results. For instance, during 1989- 90, of the 3003 violent domestic incidents reported to the police, 92 involved the threat or use of a gun. Sixty-five per cent of these cases were witnessed by children under the age of 5, and 35 per cent were witnessed by children aged between 5 and 9. A further 84 incidents involved the use of a weapon (usually a knife) where 79 per cent were witnessed by children under 5, and 25 per cent were witnessed by children between the ages of 5 and 9. Children under the age of 5 were also present at more than two-thirds of domestic disputes in which property was damaged. Over the three-year-period, an analysis of domestic disputes dealt with by the Magistrate's Court shows that children were assaulted or molested in 25 per cent of domestic disputes; and in 4 per cent of cases children were held in unlawful custody by the perpetrator (Wearing 1992).

Child Victims of Domestic Violence

Some children who witness domestic violence are also victims of the abusive behaviour. Studies have shown an overlap between violence towards women and violence towards children of at least 40 per cent (Straus, Gelles and Steinmetz 1980; Hughes 1988). The Queensland Domestic Violence Task Force (1988) phone-in revealed that, of the 88 per cent (580) of respondents who reported the presence of dependent children, 68 per cent (392) said that their children had also suffered at the hands of the perpetrator of domestic violence. Of these, 68 per cent reported their children being physically abused, 70 per cent reported emotional abuse, and 8 per cent reported sexual abuse. Research in the United States has also shown that the rate of child abuse and neglect of children in violent homes has been found to be fifteen times greater than the national average (Peled and Davis 1992). In a New Zealand study, Church (1984) stated that half of the children surveyed had to be protected by their mother during the confrontation.

Significantly only 6 per cent (23) of the respondents with abused children (392) in the Queensland Domestic Violence Task Force Report contacted the Department of Family Services. This is similar to research conducted by Roy (1977) who stated that 95 per cent of her sample of adult family violence victims did not report the husband to the authorities for child abuse. Reasons cited for this ranged from fear of reprisals to counter charges by the husband.

Walker (1987) concludes as a consequence of her research with a sample of 453 abused women that they were eight times more likely to hurt their children while they were living in a violent relationship, than when they were safe from violence. This is supported by Straus, Gelles and Steinmetz (1980) who found that mothers and fathers in violent marriages are both more likely than their counterparts in non-violent marriages to be child abusing parents.

The effects of domestic violence on children

There is now a small, slowly emerging literature on the effects of witnessing violence on children's psychological development. Initially the literature was limited to clinical descriptions of children's behav ioural and emotional problems elicited primarily from assessment of children in women's shelters. These assessments used a standardised checklist which measures internalising problems (depression, somatic or physiological complaints, anxiety and withdrawal) and externalising problems (disobedience, destructiveness and aggression). Recent studies have improved methodologically by including appropriate comparison groups and additional standardised measures, and by examining a wider range of children's dysfunctional and adaptive behaviour. These studies represent beginning efforts to document the effects domestic violence has on children's behaviour, their cognitive and social problem-solving abilities, as well as their coping and emotional functioning. A discussion of this literature in terms of age, stage of development and gender is outlined below.


Infants, by definition, are the most limited of all children in their cog- nitive abilities and resources for adaptation. In terms of behaviour, however, infants who witness spousal violence are often characterised by poor health, poor sleeping habits and excessive screaming (Jaffe et al. 1990). It is also possible that they may suffer serious, unintended consequences when their basic needs for attachment to their mother may be significantly disrupted. Routines around sleeping and feeding often become far from normal. A mother living in fear of her husband may be unable to handle the stressful demands of an infant. Clearly, any rejection from lack of availability to their principal caretaker, which is likely to continue for the duration of the domestic violence, would be felt by the child and could have long-term effects in the form of emotional deprivation (Hart and Brassard 1987).

Developmental evidence suggests that children begin to learn the importance of emotions for communication and regulation early in the first year of life. They look for cues in their principal caregiver in order to recognise the appropriate emotion. They are therefore aware of others negative emotions and mirror these in their own responses (Cummings et al. 1981).


By the second year of life, children are developing basic attempts to relate causes to emotional expressions (Jaffe et al. 1990). A research study (Cummings et al. 1981) assessed toddlers' reactions to naturally occurring and simulated expressions of anger and affection by others in the family. It was found that the expressions of anger caused distress in the young children. This distress became even more appar- ent when verbal expressions were accompanied by the physical attack of another family member. They also found that repeated exposure to anger between their parents increased the likelihood of these stress reactions in the children, and as a result the children made more efforts to become actively involved in the conflict. Based on these initial data, the researchers hypothesised that exposure to harsh emotions threatens children's sense of security in relation to their social environment.

In a second study Cummings et al. (1985) targeted children in their third year of life. They found that these children readily distinguished between warmth and anger, and that the children responded to angry adult interactions with significantly greater displays of distress and subsequent increases in aggression with their peers. When the children were exposed a second time to negative emotional exchanges a month later, the researchers found even higher levels of distress and aggressive behaviour. Interestingly, boys showed more aggressive behaviour than girls following the simulation, whereas girls showed more distress than boys during the simulation.

The short-term, immediate effects of domestic violence on toddlers can often be portrayed by behavioural problems such as frequent illness, severe shyness, low self-esteem, and trouble in daycare as well as social problems such as hitting, biting or being argumentative (Blanchard et al. 1992). Generally, the behaviour of boys tends to be externalised, while the behaviour of girls tends to be internalised. For instance, Carlson (1984) describes girls as having an increasing assortment of physiological symptoms and being more likely than boys to become withdrawn, passive, clinging and anxious.


In a third study, this time on children aged 4 and 5, Cummings et al. (1987) recorded similar distress reactions as previously noted on the younger age groups. In addition, the researchers were able to identify the following three types of behavioural reactions to adult arguments. Forty-six per cent of the children displayed negative emotions during the time that the anger was being acted out, but afterwards they reported feeling sad and wanting to intervene. Seventeen per cent showed no evidence of emotion, but later reported that they were angry. Over a third showed high emotional feelings (both positive and negative) during the arguments. Later, this latter group reported feeling happy, but they were also the most likely to become physically and verbally aggressive with their peers. It appears from this study that children's reactions to adult arguments and anger vary considerably, ranging from strong displays of emotional distress to very hidden emotional reactions. Also, the type of immediate reaction shown by each child was found to be associated with his or her own degree of anger, sadness, or aggression following the violent incident.

Davis and Carlson (1987) found that, through clinical testing of 77 children, those who displayed their reactions aggressively were pre-school boys. This same group of boys also demonstrated a higher lever of somatic complaints, with twice as many pre-school boys as girls showing their emotional difficulties through such symptoms. On the basis of this study, pre-school boys had the highest rating for aggressive and somatic difficulties of any group in terms of age and gender. In another comprehensive study (Hughes 1988), using reports obtained from mothers and children, abused and non-abused child witnesses to domestic violence were compared to other children from a similar economic background on measures of self-esteem, anxiety, depression and behavioural problems, using reports obtained from mothers and children. Consistent with previous studies, results indicated much higher distress levels in the children who had been both witnesses and victims than in the comparison group, with the non-abused witness children's scores falling somewhere between the two. Although Hughes did not perform any analysis by gender, the results of this study are in partial agreement with those of Davis and Carlson in that, on examination of the behavioural problem scores, it was revealed that the pre-school group had the highest rates of any group.

Children of this age interpret most events in relation to self. They see themselves as the cause of the anger. They do not have the cognitive competence to take into account the whole situation. Placing blame for adult anger on oneself, therefore, is a developmentally defined common occurrence for preschoolers (Jaffe et al. 1990). There is also a relationship between the anxiety levels of this age-group and the mother's own anxiety levels. In fact, Hughes (1986) has suggested that shelter children, may particularly associate their own feelings very closely with their mother, so that as the mother's anxiety level rises and falls, so does their own. It was also observed by deLange (1986) that exposure to domestic violence may affect pre-school age children's social-cognitive developmental competence; they were often socially isolated from their peers and did not relate to the activities or interests of their age group and they had some problems relating to adults.

Primary School Age

By the time children reach school age, they look to their parents as significant role models. Both boys and girls who witness domestic violence quickly learn that violence is an appropriate way of resolving conflict in human relationships (Jaffe et al. 1990). They are more able to express their fears and anxieties regarding their parents' behaviour. Like pre-schoolers, many feel partially responsible as participants in the family conflicts, and sex differences consistent with traditional sex-role stereotypes are likely to manifest themselves at this age (Hilberman and Munson 1978). Hughes (1986) found that children of this age often had difficulties with school work, including poor academic performance, not wanting to go to school, and difficulties in concentration. Similarly, McKay (1987, cited in Jaffe et al. 1990) described such children as constantly fighting with peers, rebelling against adult instruction and authority, and being unwilling to do school work. In terms of gender, Davis and Carlson (1987) observed that girls in this age group showed high levels of both aggression and depression on the clinical behaviour checklist and, in fact, they had the highest scores for problems compared with the other groups.

Studies of both boys and girls in this age group which have compared children living in refuges with community control groups matched on the basis of socio-economic status (Wolfe et al. 1986; Jaffe et al. 1986; Christopoulos et al. 1987) have tended to confirm that children from domestic violence family backgrounds had significantly more behavioural problems and lower social competence than children from non-violent backgrounds. The research conducted by Wolfe et al. (1986) found that 35 per cent of boys and 20 per cent of girls fell within the clinical range of behavioural problems, with boys tending to have more elevated scores than girls. The pattern was replicated by Jaffe et al. (1986) who confirmed that girls tended to show more internalising than external- ising problems (depression, anxiety and withdrawal), while boys showed both internalising and externalising problems (increased activity levels and aggression). However, Christopoulos et al. (1987) found that, while children from domestic violence backgrounds had significantly more behavioural problems than the control group, and there were no differences in terms of gender, both boys and girls had increased internalising problems.

An Australian study has compared the psychological functioning of 22 children aged from 6 to 11 years who came from violent backgrounds against a matched group of children who had no history of domestic violence (Mathias et al. 1995 in press). This research found that more than half of the group who had been exposed to violence showed borderline to severe levels of behaviour problems; and had below average adaptive skills whilst over 40 per cent had reading ages over a year below their chronological ages; and moderately high to high levels of anxiety were evident in only 15 per cent of children. These children also chose significantly more assertive responses and fewer aggressive responses. For these aggressive responses, however, boys rated higher. This was the only obvious gender difference in the study.

One study (Rosenberg 1984) which does offer some explanation for different gender outcomes in research findings, points to a possible interaction between the amount of violence the child has witnessed and the type of behavioural adjustment shown by males and females. Rosenberg found that when there was a relatively lower occurrence of parental violence, boys selected aggressive coping strategies whereas girls reacted passively. Alternatively, when there was a higher occurrence of violence, girls chose aggressive methods for solving problems and boys became more passive. Rosenberg hypothesises that a child's predominant method of problem solving in interpersonal situations, which is gender-related, becomes exaggerated following exposure to parental violence. When the violence is more extreme, however, the children may attempt to escape or avoid the problem situations, or even draw attention away from the parents to themselves by resorting to more unusual and dramatic coping responses.


By the time children reach adolescence, their cognitive skills and resources for adaptation have usually reached a stage of development which encompasses both their own family dynamics and outside social networks such as peer groups and school influences. In other words, they are becoming aware that there are different ways of thinking, feeling and acting in the world from those to which they have been exposed. However, the question is whether the behavioural and social-learning processes of adolescents, who have been exposed to domestic violence, have become so entrenched that they find it difficult to engage in more positive ways of social interaction. For instance Davis and Carlson (1987) concluded that growing up in a violent family increases the likelihood of becoming an abused wife, while Hughes and Barad (1983) found that a high incidence of violent men and their victims have been raised in violent homes and witnessed domestic violence as children. However, it is emphasised that not all children who have lived with abusive relationships will repeat the experience (Rosenbaum and O'Leary 1981).

Given the important developmental tasks associated with adolescence, it would be expected that an ongoing stressor, such as inter-parental conflict, would have a profound influence on adolescent development (Hetherington and Anderson 1988). Indeed, there have been several studies which have revealed significant relationships between inter-parental conflict and anxiety, depression, stress, and/or hostility in adolescents. For instance, Forsstrom-Cohen and Rosenbaum's (1985) research of those who witnessed violence in the home, revealed that adolescent females were significantly more depressed than their male counterparts. Additionally, adolescent females who witnessed parental violence were significantly more depressed and aggressive than females from non-violent homes, whereas no similar interactions were found for males. Schwarz and Getter (1980) found support for their hypothesis that the level of inter-parental conflict, parental dominance, and the gender of the adolescent were predictive of severe adolescent problems. In fact, conflict between parents in combin- ation with a dominant opposite-sex parent was significantly predictive of major adolescent psychopathology.

In another analysis, Widom (1989) revealed that exposure to continued violence was the strongest predictor of violent delinquent behaviour. Based on research with other delinquent populations, Wexler (1990) estimates that between 20 per cent and 40 per cent of chronically violent adolescents had witnessed extreme parental conflict. Kalmuss (1984) found that observing aggression and violence between parents was more strongly related to future involvement in severe marital violence than was being the victim of abuse. Furthermore, the problem of marital violence in adulthood increased dramatically when both types of family violence were experienced. Similarly, the study of Miller et al. (1991) indicated that a history of witnessing domestic violence is a very high risk factor for potential child abuse, as well as being associated with increased psychological stress. Also, some adolescent boys handle their frustration with the behaviour that has been most clearly modelled for them by assaulting their mother or siblings (Straus et al. 1980).


A broad summary of the effects on children of either witnessing and/or being victims of domestic violence is as follows. Infants are reactive to their environment; when distressed they cry, refuse to feed or withdraw and are particularly susceptible to emotional deprivation. They are extremely vulnerable. Toddlers, who are beginning to develop basic attempts to relate causes to emotional expressions, can often be seen to have behavioural problems such as frequent illness, severe shyness, low self- esteem and trouble in daycare as well as social problems such as hitting, biting or being argumentative. Gender differences can emerge at this stage. By preschool age, children believe that everything revolves around them and is caused by them. If they witness violence or abuse, they believe they have caused it. Some studies have shown preschool boys to have the highest ratings for aggressive behaviour and the most serious somatic difficulties of any age group. Primary school age children, particularly in the latter stage, begin to learn that violence is an appropriate way of resolving conflict in human relationships. They often have difficulties with schoolwork and girls in this age group have been found to have the highest clinical levels of both aggression and depression. Adolescents see the violence as their parents' problem and they often regard the victim as being responsible. Ongoing conflict between parents has a profound influence on adolescent development and future adult behaviour, and can be the strongest predictor of violent delinquency.

However, it must be emphasised that, while there is no doubt that children who either witness and/or are victims of domestic violence are all affected in terms of their behavioural, cognitive and emotional adjustment, the research is still not conclusive enough to determine that there is a definitive set response for gender, age or stage of development. The sample sizes of some of the studies are also often not large enough to warrant firm conclusions being drawn. The inconsistencies suggest that there are still many more factors to be taken into account, including: the extent and frequency of the violence; the role of the child in the family; the number of repeated separations and moves; and economic and social disadvantage.

Children's Coping Abilities

One area which needs to be considered in more detail is the specific coping abilities of individual older children. Researchers have begun to investigate reasons which may account for the level of resilience shown by some children. Following from this, a clinical and research consensus is forming in favour of viewing this resilience as being influenced by more than one factor (Jaffe et al. 1990). In a review of the stressors of childhood, Garmezy (1983) found children's coping abilities could be divided into three categories. These are: dispositional attributes of the child (for example, ability to adjust to new situations); support within the family system (for example, good relationship with one parent); support figures outside the family system (for example, peers, relatives).

There is further evidence that children's coping abilities can vary as a function of their developmental stage (Hetherington 1979). Research with preschool children has demonstrated that disruptions in their normal family functioning is associated with maladaptive behaviours, both in the home and other social situations (Hess and Camara 1979; Wallerstein and Kelly 1975). Similarly, Kurdek (1981) suggests that young children are generally more negatively affected than older children as a result of the dependence on their caregivers and the younger children's lack of sufficient cognitive development to allow them to interpret surrounding events accurately. While family disruption certainly has a negative influence on older children's social interactions, it has been suggested (Hetherington 1979; Kurdek 1981) that they are better able to cope with the stress because of the additional support of peers and schools.

It has also been suggested that a sense of empowerment may be useful to some children. Rosenberg and Rossman (1990) in their research found that children who believed they had control over their own thoughts and behaviour during their parents' fights were less anxious and had lower reports of delinquent behaviours. They perceived themselves more positively in terms of their behavioural and social competence, as well as their sense of good worth. When the same study measured control over self and control over parents' beliefs together, children with higher control beliefs in both areas showed fewer problem behaviours; the children who believed they had control over their parents' thoughts and behaviour, but little control over their own, were the most powerless. In fact, children who held this pattern of beliefs had the highest depression scores, displayed low adaptive functioning, and showed evidence of many behavioural problems including aggression. However, with this model, it is important to emphasise that children be not made to feel responsible for the violence in any way.


Procedures to raise awareness of the effects on children who witness and/or are victims of domestic violence are necessary so that the long term harmful effects are ameliorated. In fact, the National Strategy for the Prevention of Child Abuse and Neglect which has been developed for the National Child Protection Council (Calvert 1993) is specifically concerned with changing attitudes, beliefs and behaviours. This involves, in the first instance, primary prevention programs such as community education campaigns involving use of the mass media, education through the publication of information packages, pamphlets and videos as well as school programs for children.

To address the needs of children affected by domestic violence, school programs would need to cater for the various child developmental stages. For instance, the Perry Preschool Program which is now used in many areas of the United States, is based on a preschool curriculum which emphasises positive role models, the teaching of alternative conflict resolution strategies and co-operative play (Schweinhart 1987). Long term follow-up has indicated that the Perry Preschool Program participants, when compared with a group who had not been exposed to the program, committed only half as much juvenile delinquencyincluding only one-fifth as many acts of property violence and half as many acts of drug abuse. In addition, they Perry Preschool group participated more in sports and extracurricular activities at school and had higher educational and employment earning rates (Schweinhart et al. 1986).

Another school program, called Preventing Abuse in Relationships (PAIR) is targeted at adolescents, and its main objectives are:

  • to highlight the relevance of violence in relationships to young people by exploration of violence in dating relationships;
  • to raise awareness of the nature and extent of violence against girls and women in relationships;
  • to increase the understanding of students of the effects of domestic violence and dating violence;
  • to educate young people to be aware of the unacceptable nature of violence and abusive actions by men/boys against women/girls with a view to ending/preventing the prevalence of these actions;
  • to encourage girls to be aware of the indicators of violence and to take steps to prevent themselves from being at risk of this violence;
  • to assist students in gaining an understanding of how stereotyped expectations of men and women contribute to domestic violence and thus encourage attitudes and skills that will enable them to develop non-violent, equitable relationships;
  • and to familiarise students with resources available to them, their families and their friends (Domestic Violence Crisis Centre 1992).

A program similar to the PAIR program would also benefit older primary school children who are at a stage in their cognitive develop- ment to be able to confront the issues of violence in their own families and could be built into existing personal safety programs. Children between the ages of 8 and 12 are at a stage when intervention could be successful, before the impact of violent role modelling has become entrenched in their behavioural and social-learning processes.

The introduction of programs such as the PAIR Program into school curricula would have the added benefits of targeting all children through a non-stigmatising process. It would also help promote new attitudes against violence and positive attitudes towards women and equality in relationships.

Secondary prevention programs, would also benefit high risk groups of children who may or may not demonstrate symptoms, but who are known to have witnessed violence. These could include children from refuges, as well as those identified by police officers, family lawyers, marriage counsellors, staff of hospital casualty services and other agencies (Menzies 1993). An important component of secondary prevention programs are home visitor and parenting programs which have already identified at-risk populations. By extending these to include children who witness and/or are victims of domestic violence, more protection would be guaranteed.

Indeed, this has been recognised by the Talera Centre in Queensland which provides a range of therapeutic, educational, accommodation and support services for families, with the aim of increasing the possibility that children can live in safety within their own family. Talera maintains that effective intervention with child witnesses to domestic violence requires a range of child, family, school and community responses. To incorporate these multiple interventions, the following goals have been developed.

  • To empower child witnesses to disclose domestic violence and to enhance their ability to develop safe, non-violent relationships.
  • To heighten the awareness among the parents of the impact of domestic violence on their children and to support them in developing non-violent, nurturing relationships with their children.
  • To improve the current response to the needs of child witnesses through raising the level of knowledge and skill among human service providers.
  • To encourage and facilitate the development of community responses to the issue of children and domestic violence (Talera Centre 1993).


Children's responses to witnessing and/or experiencing domestic violence between parents vary considerably. No typical reactions emerge, although there is ample evidence that exposure to domestic violence can and often does influence children's behaviour detrimentally. However, at various stages of their development, children are differentially able to understand and cope with what is happening between their parents.

So that this issue receives the attention it deserves, linkages must be made between domestic violence and child abuse. However, changes in definitions and parameters of both child abuse and domestic violence, as well as legislative reform, though necessary, are not sufficient to bring about a fundamental shift in attitude by the community at large. Real change must involve an integrated response from a wide variety of agencies: police, lawyers, health and welfare professionals, teachers and the community need to be informed and willing to be part of a concerted effort to bring about such change.

However, to fully understand the extent of the effects of domestic violence on children, there are several areas which need to be considered for future research. First, more needs to be known about the incidence and prevalence of child witnesses, and about the interrelationship between witnessing violence and experiencing other forms of child abuse and neglect. Secondly, child witnesses to domestic violence tend to be a heterogeneous population. It is therefore necessary to further identify factors that mediate children's reactions rather than assuming that all children will be equally affected by witnessing violence. For instance, these variables could be child related, such as gender, temperament or intelligence; adult related, such as identification of individual characteristics; or family related, such as the quality of the parent- child relationship or parenting practices, as well as determining the intensity and length of the violence. (Ammerman and Herson 1990).

It is only by developing a broader definitional and research base for child abuse and domestic violence issues, and placing them within the framework of family violence generally, that we can hope to direct government policy, with corresponding adequate resources, in a way which will ensure a better future for children who live with inter- parental conflict.


  • Alexander, R. 1988, The Crimes (Family Violence) Act 1987. Law Institute Journal. March pp 166-169.
  • Ammerman, R.T. and Herson, M. (eds) 1990, Treatment of Family Violence, John Wiley and Sons, New York.
  • Blanchard, A., Molloy, F. and Brown, L. 1992, 'I Just Couldn't Stop Them': Western Australian Children Living with Domestic Violence: A Study of Children's Experiences and Service Provision, Curtin University School of Social Work, Western Australia.
  • Bowker, L.H., Arbittel, M. and McFerron, J.R. 1988, 'On the relationship between wife beating and child abuse' in: Feminist Perspectives on Wife Abuse, eds K. Yllo and M. Bograd, Sage, California.
  • Calvert, G. 1993, Preventing Child Abuse and Neglect: the National Strategy, National Child Protection Council, Canberra.
  • Carlson, B.E. 1984, 'Children's observations of inter-parental violence' in: Battered Women and Their Families, ed. A.R. Roberts, Springer, New York.
  • Christopoulos, C., Cohn, D., Shaw, D., Joyce, S., Sullivan-Hanson, J., Kraft, S. and Emery, R. 1987, 'Children of abused women: adjustmenet at time of shelter residence', Journal of the Marriage and the Family, vol. 49, pp. 611-19.
  • Church, J. 1984, Violence Against Wives: Its Causes and Effects, John Church (Publisher), Christchurch, New Zealand.
  • Cummings, E.M. 1987, 'Coping with background anger in early childhood', Child Development, vol.58, pp.976-84.
  • Cummings, E.M., Zahn-Waxler, C. and Radke-Yarrow, M. 1981, 'Young children's responses to expressions of anger and affection by others in the family', Child Development, vol.52, pp.1274-82.
  • Cummings, E.M., Iannotti, R.J. and Zahn-Waxler, C. 1985, 'Influence of conflict between adults on the emotions and aggression of young children', Developmental Psychology, vol.21, pp.495-507.
  • Davis, L. and Carlson, B. 1987, 'Observation of spouse abuse: what happens to the children?', Journal of Interpersonal Violence vol.2, no.3, pp.278- 91.
  • deLange, C. 1986, 'The family place children's therapeutic program', Children's Today, pp.12-15.
  • Dobash, R.E. and Dobash, R. 1979, Violence Against Wives: A Case Against the Patriarchy, Free Press, New York.
  • ----------- 1984, 'The nature and antecedents of violent events', British Journal of Criminology, vol.24, no.3, pp.269-88.
  • Domestic Violence Resource Centre 1992, Preventing Abuse in Relationships, Brisbane.
  • Easteal, P.W. 1994, 'Violence against women in the home', Family Matters, No. 37, April, pp. 86-93.
  • Forsstrom-Cohen, B. and Rosenbaum, A. 1985, 'The effects of parental marital violence on young adults: an exploratory investigation', Journal of Marriage and the Family, May, pp.467-72.
  • Garmezy, N. 1983, 'Stressors of childhood', in: Stress, Coping and Development in Children, eds N. Garmezy and M. Rutter, McGraw-Hill, New York.
  • Hart, S.N. and Brassard, M.R. 1987, 'A major threat to children's mental health: psychological maltreatment', American Psychologist, vol.42, pp.160-5.
  • Hess, R.D. and Camara, K.A. 1979, 'Post-divorce: family relationships as mediating factors in the consequences of divorce for children', Journal of Social Issues, vol.35, pp.79-96.
  • Hetherington, E.M. 1979, 'Divorce, a child's perspective', American Psychologist, vol.34, pp.851-8.
  • Hetherington, E.M. and Anderson, E.R. 1988, 'The effects of divorce and remarriage on early adolescents and their families', in: Early Adolescent Transitions, Lexington, MA, pp. 46-57.
  • Hilberman, E. and Munson, K. 1978, 'Sixty battered women', Victimology, vol.2, pp.460-70.
  • Hughes, H. 1986, Research with children in shelters: implications for clinical services, Children Today, vol.15, no.2, pp.21-5.
  • -----------, 'Psychological and behavioural correlates of family violence in child witnesses and victims', American Journal of Orthopsychiatry, vol.58, no.1, pp.77-90.
  • Hughes, H. and Barad, S. 1983, 'Psychological functioning of children in a battered women's shelter: a preliminary investigation', American Journal of Orthopsychiatry, vol.53, no.3, pp.525-31.
  • Kalmuss, D. 1984, 'The intergenerational transmission of marital aggression', Journal of Marriage and the Family, vol.46, pp.11-19.
  • Kurdek, L.A. 1981, 'An integrative perspective on children's divorce adjustment', American Psychologist, vol. 36, pp. 856-66.
  • Jaffe, P., Wolfe, D., and Wilson, S.K. 1990, Children of Battered Women, Sage Publications, California.
  • Jaffe, P., Wolfe, D., Wilson, S. and Zak, L. 1986, 'Family violence and child adjustment: a comparative analysis of girls' and boys' behavioural symptoms', American Journal of Psychiatry, vol.143, no.1, pp.74-7.
  • MacLeod, L. 1987, Battered but not beaten...Preventing wife battering in Canada, Canadian Advisory Council on the Status of Women, Ottawa.
  • Mathias, J., Mertin, P. and Murray, B. (forthcoming), 'The psychological functioning of children from backgrounds of domestic violence', Australian Psychologist, vol.30, no.1 (March).
  • Menzies, T. 1993, Nurturing Non-Violence: A Handbood to Help Children Who Witness Domestic Violence, Lower North Shore Domestic Violence Liaison Committee, Lane Cove, NSW.
  • Miller, T., Handal, P., Gilner, F. and Cross, J.F. 1991, 'The Relationship of Abuse and Witnessing Violence on the Child Abuse potential inventory with black adolescents', Journal of Interpersonal Violence, vol. 6, pp. 353-63.
  • Mugford, J. 1989, Domestic Violence, Violence Today No. 2, National Committee on Violence, Australian Institute of Criminology, Canberra.
  • Mugford, J., Easteal, P., Edwards, A. 1993, ACT Domestic Violence Research. Report to the ACT Community Law Reform Committee, Canberra.
  • New South Wales Domestic Violence Committee, 1991. Costs of Domestic Violence, Premiers Department, Women's Coordination Unit, Sydney.
  • Queensland Domestic Violence Task Force 1988, Beyond These Walls, Report of the Queensland Domestic Violence Task Force to the Minister for Family Services and Welfare Housing, Brisbane.
  • Rosenberg, M.S. 1984, Inter-generational family violence: a critique and implications for witnessing children, paper presented at the 92nd annual convention of the American Psychological Association, Toronto.
  • Rosenberg, M.S. and Rossman, B.B. 1990, 'The child witness to marital violence', in: Treatment of Family Violence, eds. A. T. Ammerman and M. Herson, John Wiley and Sons, New York.
  • Rosenbaum, A. and O'Leary, K.D. 1981, Children: the unintended victims of marital violence', American Journal of Orthopsychiatry, vol. 51, pp.692-9.
  • Roy, M. 1977, Battered Women: A Psychological Study of Domestic Violence, Van Nostrand, New York.
  • Schweinhart, L.J. 1987, 'Can preschool programs help prevent delinquency', From Children to Citizens, vol. 3. eds J. Wilson and G. Loury, Springer Verlag, New York.
  • Schweinhart, L.J., Weikart, D.P., and Larner, M.B. 1986, 'Consequences of three preschool curriculum models through age 15', Early Childhood Resarch Quarterly, vol. 1, no. 1, pp. 15-45.
  • Schwarz, J.C. and Getter, H. 1980, 'Parental conflict and dominance in late adolescent maladjustment: a triple interaction model', Journal of Abnormal Psychology, vol. 89, pp. 573-80.
  • Sinclair, D. 1985, Understanding Wife Assault: A Training Manual for Counselors and Advocates , Ontario Governrment Bookstore, Toronto, Canada.
  • Straus, M.A., Gelles, R.J. and Steinmetz, S. 1980, Behind Closed Doors, Anchor Books, New York.
  • Talera Centre 1993, Issues and Opportunities in Responding to Child Witnesses to Domestic Violence . A Discussion Paper Arising from the Talera Centre's Children and Domestic Violence Program.
  • WA Domestic Violence Task Force 1986, Break the Silence , Perth, W.A.
  • Walker, L.E. 1984, The Battered Woman Syndrome , Springer Publishing Company, New York.
  • ----------- 1987, 'Identifying the wife at risk of battering', Medical Aspects of Human Sexuality , Anchor Books, New York.
  • Wallerstein, J.S. and Kelly, J.B. 1975, 'The effects of parental divorce: experiences of the child in later latency', American Journal of Orthopsychiatry , vol.50, pp.256-69.
  • Wearing, R. 1992, Monitoring the Impact of the Crime (Family Violence) Act 1987, Report , La Trobe University, Victoria.
  • Wexler, R. 1990, Wounded Innocents: the Real Victims of the War against Child Abuse , Prometheus Books, Buffalo, New York.
  • Widom, C.S. 1989, 'Does violence beget violence? A critical examination of the literature', Psychological Bulletin , vol. 106, pp.3- 28.
  • Wolfe, D.A., Zak, L., Wilson, S. and Jaffe, P. 1986, 'Child witnesses to violence between parents: critical issues in behavioral and social adjustment', Journal of Abnormal Child Psychology , vol.14, no.1, pp.95- 104.

Marianne James is Senior Research Officer at the Australian Institute of Criminology.


0 642 20336 9