Contents

• How do parents become abusive or neglectful?
• Methodologies
  • Retrospective studies
  • Prospective studies
• Research evidence
  • Prospective studies
• Australian data
• Causes of intergenerational transmission of maltreatment
• Types of abuse
• Physical abuse
  • Physical punishment
• Sexual abuse
  • Sex offenders
  • Child and adolescent sex offenders
• Gender differences
• Domestic violence
• Criminal behavior
• Substance abuse
• Psychological factors
• Mediating factors
• Breaking the cycle
• Prevention
  • Taking a holistic approach
  • Sexual abuse
• Future directions
• References

Children depend on parents and caregivers to look after them through childhood. Being protective is considered natural and instinctive (Bowlby 1971, as cited in Corby 1993), and most children are well cared for by their parents. However, there are some children for whom a parent's 'instinctive' desire to protect is unrealised and children are thereby exposed to the risk of child maltreatment.

How do parents become abusive or neglectful?

Determining and moderating the causes of maltreatment have been central objectives in all research, theory and methods of intervention in this complex area (Gelles and Loseke 1993). A recurrent theme in the child abuse literature for some time has been the notion that it is maltreated children who become abusive parents (US National Research Council 1993). It is hypothesised that children may learn to be abusive from parents who model abusive behavior. Alternative explanations have been that some children may have a genetic predisposition for aggressive behavior which is transformed into child maltreatment when the child becomes an adult. Or it may be that intergenerational transmission is caused by a combination of genetic and social influences (Kaufman and Zigler 1993).

This belief that maltreatment can be transmitted across generations, or that a cycle of maltreatment could be set up across generations was one of the earliest and most widely accepted theories of the causation of maltreatment (Egeland 1993). Widom (1989) reported 'intergenerational transmission' to be the 'premier' hypothesis in the child maltreatment field.

Current evidence suggests that while some parents who have been abused as children become abusive parents, the majority do not (Kaufman and Zigler 1993). It is the objective of this paper to review the literature on the intergenerational transmission of maltreatment. Two questions are addressed: How likely is it that today's maltreated children will become tomorrow's abusive parents? What factors influence whether an abused child goes on to become an abusive adult?

Methodologies

Lewis (1988) contended that there may be no area of child development or child abuse that is better understood than the intergenerational transmission of maltreatment. However the research to date has not produced conclusive 'proof' for such a contention. Two main methods have been employed to test the intergenerational transmission hypothesis - retrospective and prospective studies.

A retrospective approach tests the question: 'Given that a child has been identified as maltreated, what is the probability that the child's parents were themselves maltreated in childhood?'. In contrast, prospective studies ask: 'Given that people are maltreated or not maltreated as children, what is the probability that they will maltreat their own children?' (Lewis 1988). These are similar questions, but they involve the estimation of different base rates of intergenerational transmission.

Retrospective studies

As the name would suggest, retrospective studies involve looking backward in time by studying maltreating parents after they have been identified as having maltreated their own children (Gelles and Loseke 1993). Demographic factors are identified in families known to be abusive which differentiate them from non-abusing families matched on demographics (for example, age, sex of child, parental occupation, ethnic group, social class) (Browne and Saqi 1988). In the case of intergenerational transmission, once abusive families are identified it is a simple task to determine what percentage of them involve parents who were maltreated as children (Lewis 1988).

Investigators using the retrospective approach, with or without comparison groups of non-abusive parents, usually find that the majority of abusing parents were themselves abused as children (Steele and Pollock 1968; Herrenkohl, Herrenkohl and Toedter 1983, as cited in Egeland 1993). For example, Steele and Pollack (1968) found that all 60 abusing parents in their study had been abused as children.

The problem with the retrospective approach is that it does not enable a determination of the proportion of adults who were maltreated, but who subsequently provide adequate care for their children (Egeland 1993). The retrospective approach only provides information about those adults who were abused in childhood who then go on to abuse their own children. While the majority of maltreating parents may themselves have been maltreated, the majority of abused parents may not go on to maltreat (Egeland 1993).

Retrospective studies have also been criticised as being beset with methodological flaws. First, it is difficult to determine whether maltreating parents have provided distorted accounts of their childhood. Thus, some parents may falsely report a childhood history of maltreatment leading to an over-estimation of the rate of intergenerational transmission (Belsky 1992, as cited in National Research Council 1993). However, it is also possible that a reliance on the self-reports of maltreaters may instead result in an under-estimation of the rate of intergenerational transmission (Egeland 1993). Zeanah and Zeanah (1989, as cited in Egeland 1993) noted that some parents maltreated as children may normalise their abusive experiences, and thus when asked, will not report a history of maltreatment. Alternatively, some parents may idealise their childhood experiences and their relationship with their parents in order to cope with the trauma of abuse. This may also lead to an under-identification of those parents who have a history of maltreatment (Egeland 1993).

Second, when comparing current and past concepts of maltreatment, it has been concluded that in general, the broader, or more encompassing the definition of maltreatment used, the greater the apparent link between a history of maltreatment and current maltreatment (Kaufman and Zigler 1993). Kaufman and Zigler (1993) argued for the use of restrictive definitions of maltreatment which encompass only the more severe forms of maltreatment, those requiring protective intervention, in order to prevent the misapplication or misinterpretation of research results.

Overall, despite the potential for under-estimation, retrospective studies are commonly perceived as providing over-estimations of the rate of intergenerational transmission of child maltreatment (Belsky 1992, as cited in National Research Council 1993).

It has been argued that because of these methodological and measurement problems (National Research Council 1993), the findings of retrospective studies cannot be considered as conclusive proof of the intergenerational transmission of maltreatment (Kaufman and Zigler 1987). In order to determine logically if the intergenerational transmission of maltreatment occurs, it is necessary to conduct prospective studies of children (Lewis 1988).

Prospective studies

Prospective analyses are more difficult to perform (Lewis 1988). Ideally, two groups of children, from families where children are adequately cared for, and from families where the children have been maltreated are matched on demographic factors, and followed until they themselves have children. Information is then obtained on the number of children who were maltreated, and the number of children who were not maltreated, who go on to maltreat their own children.

Rather than compare previously identified maltreating families with matched non-maltreating families, prospective studies collect data on risk factors (such as a childhood history of maltreatment) present in families, and follow the families forward over time to determine the proportion who go on to maltreat their child.

Prospective longitudinal studies that have examined the later caregiving practices of people who are known to have been abused as children (Hunter, Kilstrom, Kraybill and Loda 1978; Egeland, Jacobvitz and Papatola 1987, as cited in National Research Council 1993), have produced significantly smaller rates of intergenerational transmission than those estimated from retrospective studies.

Prospective studies have been criticised for failing to conduct follow-ups which extend for an adequate length of time (Kaufman and Zigler 1993). This may result from a failure to adequately consider the interaction of parental characteristics, the child's age and developmental stage (Hunter and Kilstrom 1979). The onset of maltreatment is not limited to infancy, and prospective studies need to continue over time to enable the investigation of additional maltreatment which may be identified at later developmental stages (for example, toddlerhood, adolescence) (Belsky 1992, as cited in National Research Council 1993).

Research evidence

Estimations of the rate of intergenerational transmission of child maltreatment have ranged from 7 per cent (Gil 1970) to 70 per cent (Egeland and Jacobvitz 1984, as cited in National Research Council 1993). Because of the significant resources required for longitudinal designs, the majority of studies investigating intergenerational transmission have taken a retrospective approach (Lewis 1988). The over-reliance on retrospective studies has seriously hampered the academic community's acceptance of the intergenerational transmission hypothesis (National Research Council 1993). As mentioned previously, it is only by employing prospective designs that the occurrence of intergenerational transmission can be logically determined.

Prospective studies

However, where prospective studies have been undertaken, few have followed a group of abused children through into adulthood (Egeland 1993). Most prospective studies have involved monitoring parents (with/without a self-reported abusive childhood) and their young children over time to determine the proportion who subsequently maltreat their child.

One exception has been a study conducted by Quinton, Rutter and colleagues (Quinton and Rutter 1985, as cited in Youngblade and Belsky 1990). Quinton and colleagues conducted a follow-up of English girls who had been raised in care. They found that in comparison with girls from the same neighborhood who had been raised in their own families, the ex-care girls were more likely to show insensitivity to their 2 to 4 year old children, were more prone to exhibit irritability and frequently spanked their children. Consequently they were more likely to be categorised as poor parents (Quinton, Rutter and Liddle, Quinton and Rutter 1985, as cited by Youngblade and Belsky 1990).

Hunter and Kilstrom (1979) conducted a commonly cited, experimentally controlled, prospective study that produced one of the lowest transmission rates (18 per cent). They followed the mothers of 282 premature or ill infants for one year. Information about the parents' childhoods (including childhood history of abuse), mothers' pregnancies, children's health and families' social networks was obtained by semi-structured interview. At the time of interview, 49 parents reported childhood histories of maltreatment. At follow-up, 9 of 49 parents in the sample who initially reported a childhood history of maltreatment were identified as maltreating parents, giving an intergenerational transmission rate of 18 per cent. In contrast, only one of the 233 families who had no parental history of childhood maltreatment was subsequently identified as having maltreated their children.

This transmission rate may be an under-estimation, especially since the follow-up of the families only continued for 12 months, and because the definition of maltreatment was only based on official agency reports. Thus, parents who maltreated their child as she/he got older, and those who maltreated but were not officially reported to an agency would not have been identified as maltreaters in the study.

Importantly, if the data from Hunter and Kilstrom's study was analysed retrospectively, using only those parents identified as maltreaters, the link between child abuse and a parent history of childhood maltreatment would have appeared much stronger, as nine out of ten maltreaters reported a childhood history of abuse, producing a transmission rate of 90 per cent. Thus, Hunter and Kilstrom's (1979) study clearly illustrates how the selection of subject groups (identified abusers or high risk families) and research design (prospective versus retrospective) can affect the outcome of research findings (Kaufman and Zigler 1993).

In Kaufman and Zigler's (1987) literature review, the results of three prospective studies were used to produce an estimated intergenerational transmission rate of 30 per cent (with a plus or minus 5 per cent error). They thus contended that the issue of intergenerational transmission of maltreatment needs to be approached with caution, not only because of methodological problems with previous research studies, but also because the majority of parents who were themselves maltreated, do not go on to maltreat their children.

Nevertheless, a transmission rate of 30 per cent is still considerably higher than the overall societal rate of maltreatment. For example, using the Australian national statistics on officially substantiated cases, the rate of maltreatment is estimated to be approximately 0.6 per cent for children aged 16 or less (Angus and Woodward 1995). However, this figure is an under-representation of the rates of maltreatment, given that it does not include reports to non-statutory agencies, or the many occurrences of maltreatment which are not reported (Angus and Woodward 1995).

Despite the lack of a precise estimate of the societal incidence of maltreatment, the apparent strength of the association between experiencing child maltreatment and the subsequent maltreatment of children as an adult, as evidenced by the 30 per cent transmission rate, indicates a need for the further exploration of this relationship.

Australian data

Data dealing with the intergenerational transmission of maltreatment in Australian populations is limited. Most of the studies that have been done have employed retrospective designs, though the Victorian case tracking studies presented below involved the collection of data at the time of case presentation in an attempt to avoid some of the pitfalls associated with assessing data retrospectively.

First, in 1977, Ryan, Davies and Oates reviewed the background of parents whose children were brought to a Sydney hospital with indications of maltreatment. Overall, 64 (37 per cent) of 173 mothers, and 26 (17 per cent) of 151 fathers, reported being physically abused in their childhood.

In a more recent Victorian study by Hiller, Goddard and Diemer (1989), 206 cases labelled as physical and sexual abuse by hospital professionals were tracked through a hospital setting and the child protection and criminal justice systems. For the purposes of the study, cases were defined as abusive where it was known, or strongly suspected by medical and/or social work staff that a child was maltreated by a caregiver or another member of the child's household.

The most commonly identified stress factor in the families was the reported maltreatment of at least one parent during childhood. Overall, 41 of 57 cases of sexual abuse (72 per cent) and 40 of 54 cases of physical abuse (74 per cent) involved families where at least one parent disclosed a childhood history of abuse

Similar findings were reported in another hospital-based tracking study, where in 30 of 42 families (71 per cent) labelled as sexually abusive, and in 67 of 89 families (79 per cent) labelled as physically abusive, at least one parent disclosed prior childhood maltreatment (Goddard and Hiller 1992).

Tomison (1994) reported on the results of a large-scale tracking of suspected child abuse and neglect cases involving a number of agencies and professions in a Victorian regional child protection network. A valid case of child abuse or neglect was defined as: 'any child under the age of eighteen years, who presented at one or more of the participating agencies or professionals, where it was known, or strongly suspected by the worker(s) involved that the child was subject to sexual, physical, emotional abuse or neglect, and that the worker(s) had felt sufficient concern to investigate, refer and/or treat/counsel the child' (Tomison 1994:23).

Overall, 344 alleged maltreaters were reported for the 295 cases that were collected, 88 per cent of which were identified as members of the victim's immediate or extended family. Most of the non-family, or unidentified maltreaters occurred in cases labelled as sexual abuse.

A disclosure of childhood maltreatment by at least one parent was reported in 66 of the 295 cases (22.4 per cent). Using measures of association, it was found that a reasonably strong relationship existed between a disclosure of parental history of maltreatment and the type of maltreatment which the family was suspected of committing (Cramer's V=0.19, p<0.05) (see Table 1). Cases of physical abuse, emotional abuse, or neglect appeared to be more likely to involve a parent with a history of childhood maltreatment, while cases of sexual abuse were under-represented.

The finding that a parental history of abuse was under-represented in cases of sexual abuse may be caused, in part, by a parental reticence to disclose such abuse, due to the perceived stigma of being sexually abused. This may have been particularly true of male caregivers, given the added stigma of having to deal with the homosexual nature of sexual abuse, the vast majority of abusers also being male (Tomison 1994)

Approximately 70 per cent of the cases collected in this study (206 of 295) were officially labelled by the various professionals as either sexual abuse, physical abuse, emotional abuse or neglect. The remaining 89 cases were labelled as a combination of the four types of abuse, or given agency-specific labels such as 'welfare check' (used by the police). An analysis was conducted of the constituent elements of the suspected child abuse cases in comparison with the official labels applied by the workers. The 89 cases with a 'combination' or 'agency specific' label were re-coded to provide a label based on only one of the four main categories of abuse.

The re-coding was done on the basis of which label had been applied by the worker concerned, and also by the actual case content (as evidenced by the types of abusive actions reported by the workers). This classification system was used on the assumption that it approximated the weighting given to different types of child maltreatment by the various professionals involved in child protection case management. Thus a case involving sexual, physical and emotional abuse became a sexual abuse case; a case of physical abuse, neglect and 'child at risk of abuse' became physical abuse; and so on.

Following re-coding, a further breakdown of the type of maltreatment suffered by parents in their childhood was produced (see Table 2). A mother's history of childhood maltreatment was reported in 85 of the 288 cases (29.5 per cent) where a mother resided with the child victim . A father's history of maltreatment was reported in 54 of the 216 cases (25 per cent) where a father or father figure resided with the child victim.

Overall, families in which one or both parents had a history of childhood maltreatment were associated with children who exhibited a higher number of presenting problems (more than two), a high number of family stressors (five or more), and a greater likelihood that professionals would substantiate the maltreatment concerns (Tomison 1994).

It should be noted however, that while a history of maltreatment in the family was shown to be associated with the maltreatment of children, none of the studies presented here have explicitly investigated the extent to which substantiated maltreatment was actually committed by the parent (or parents) who had a history of maltreatment. This is therefore identified as an area in need of further research.

Finally, recent investigations into the prior sexual abuse of identified sex offenders have estimated the rate of intergenerational transmission to between 30 per cent (Freeman-Longo 1986) and 100 per cent (Briggs and Hawkins 1996). Briggs and Hawkins (1996) using an Australian sample of incarcerated sex offenders noted that 93 per cent reported being sexually abused as children. This research is discussed in greater detail below (see section on sex offenders).

Causes of intergenerational transmission of maltreatment

There are three main theories proposed to account for the intergenerational transmission maltreatment. First, under the principles of social learning theory, it is assumed that if a physically punitive parent ends up with an aggressive child, it is because the child has learnt that pattern of response (Muller, Hunter and Stollak 1995). That is, aggressive or violent actions are learnt via patterns of positive reinforcement and punishment which shape behavior, and/or from the observation and imitation of significant others who model aggressive behavior (Bandura 1977).

The parent role is one of the most nurturant, meaningful and powerful positions in a child's life, and each of these qualities has been shown to enhance the potential for the imitation of a model (Mischel 1970, as cited in Peterson 1984). It is therefore not surprising that under social learning theory, parents are perceived to be very powerful models for children.

Second, it is argued that there is a biological or genetic component to aggressive behavior, where aggressiveness is assumed to be an individual characteristic based in temperament (Muller, Hunter and Stollak 1995). Under this theory a parent's predisposition for aggression or violence is thought to be inherited by a child. The punitive response of a parent is then assumed to be a response to the child's inherent character, rather than the aggressive behaviors the child has learned (Muller, Hunter and Stollak 1995). The child's inherited predisposition also acts as the factor in the perpetuation of the cycle of maltreatment, producing an increased probability of the child subsequently using corporal punishment or maltreating his/her own child.

Kaufman and Zigler (1993) contended that most studies of family violence have failed to consider the biological or genetic component to aggressive behavior. Many investigations of intergenerational transmission, particularly studies of corporal punishment and physical abuse, have assumed the underlying operation of social learning principles without directly testing the assumptions against alternative perspectives (Muller, Hunter and Stollak 1995).

A third explanation put forward to explain the intergenerational transmission of violence is the interaction of environmental (social learning) and genetic factors. There does appear to be some evidence suggesting a genetic component to the expression of antisocial behavior (Kaufman and Zigler 1993), but that component alone, does not appear able to adequately explain the process by which corporal punishment is transmitted intergenerationally (Muller, Hunter and Stollak 1995). Kaufman and Zigler (1993) contended that a genetic predisposition merely puts an individual at risk for the expression of violent behavior, and that it is the interaction of genetic and environmental factors which produce the greatest risk of acting violently.

Cloninger, Sigvardsson, Bohman and von Knorring (1982, as cited in Kaufman and Zigler 1993) in an investigation of the intergenerational transmission of criminality, demonstrated the effects of environmental and genetic effects in isolation and interactively. Cloninger et al. studied 862 Swedish men adopted at an early age by non-relatives. Data about criminality, alcohol abuse, occupational status, and medical and social history was obtained for the adoptees and each genetic and adoptive parent. Cloninger et al. defined environmental risk as the number of placements experienced by the child prior to adoption, the length of time with genetic mother prior to placement, and indices of adversity in the adoptive family (for example, low socioeconomic status). Genetic risk was defined as evidence of antisocial behavior or alcoholism in the genetic parents.

Cloninger et al. found that adoptees with low environmental and genetic risk had a base rate level of criminality of 2.9 per cent. Adoptees with a high environmental risk, but low genetic risk had a 6.7 per cent rate. Adoptees with a low environmental risk, but high genetic risk had a rate of 12.1 per cent, while those with high environmental and genetic risk had a 40 per cent criminality rate. Thus, while environmental and genetic effects each appeared to significantly affect the intergenerational transmission rate of criminality, it was the interaction of environment and biology that appeared to particularly increase the rate of criminality. Studies of the intergenerational transmission of maltreatment should therefore assess the interaction between biological and social learning factors via longitudinal research projects (Muller, Hunter and Stollak 1995).

It is generally acknowledged that no single factor can fully explain how maltreatment is transmitted intergenerationally. The intergenerational transmission of maltreatment is affected by a large number of factors (Kaufman and Zigler 1993). Just as no one variable has been shown to explain the causes of maltreatment in general (Browne 1988). An approach which appears to offer a way forward for research into child maltreatment, and intergenerational transmission in particular, is Belsky's (1980) multi-factor approach.

Based on the work of Bronfenbrenner (Bronfenbrenner 1979), Belsky's (1980) model of child maltreatment integrates a number of diverse single factor approaches, including psychological disturbance in parents, abuse-eliciting characteristics of children, dysfunctional patterns of family interaction, stress-inducing social forces, and abuse-promoting cultural values.

Belsky (1980) offers a conceptualisation of child maltreatment that operates on the assumption that maltreatment is determined by the mutual influences of the individual, family, community and culture. Using a similar framework, Langeland and Dijkstra (1995) report that attempts to predict which individuals are most likely to perpetrate the cycle of violence are undermined by the complexity of interaction between factors at various 'levels' which may affect the probability of intergenerational transmission.

The essence of Belsky's multi-level modelling approach to the etiology of child maltreatment is the interaction of protective and risk factors. This approach is just as valid for the study of intergenerational maltreatment. As Youngblade and Belsky note, it 'is generally acknowledged that it is principally in interaction with other etiological factors . . . that the risk associated with child-rearing history is or is not "realized"' (1990:137).

The following sections outline the evidence for individual factors which may, in combination and interaction, influence the intergenerational transmission maltreatment.

Types of abuse

Investigations of the intergenerational transmission of maltreatment have frequently concentrated on physical abuse and physical punishment (Muller, Hunter and Stollak 1995). While the transgenerational nature of physical abuse has been generally accepted (Goddard and Hiller 1993), Finkelhor (1979) and others have suggested that child sexual abuse is also likely to be intergenerationally transmitted within families. However, most studies of sexual abuse and intergenerational transmission have limited their investigations to the study of incarcerated sex offenders. This is problematic given that convicted sex offenders are a minority of all sexual offenders and are much more likely to represent the chronic, most severe forms of offending (Finkelhor 1984).

Hunter and Kilstrom's (1979) study defined a parental history of maltreatment in terms of both neglect and child abuse. The inclusion of neglect is a notable exception to much research in the field, which historically, has paid scant attention to the transmission of neglect. Child neglect is often minimised via its links with child abuse - in particular, by the perception that child abuse is a more serious societal problem than neglect Tomison (1995b). Yet neglect alone for example, has been found to be significantly related to later episodes of violent criminal behavior (Widom 1992).

Given that neglect cases make up a sizeable proportion of cases in the child protection system, a fact particularly evident in the United States (National Research Council 1993), researchers need to remedy the current dearth of information on the intergenerational transmission of neglect.

Vondra and Toth (1989) stated that there is 'ample evidence that a high proportion of parents who maltreat and/or have a child removed from their care have experienced disturbances and disruptions in relations with their own parents, without necessarily having suffered the identical form of maltreatment they themselves perpetrate' (Vondra and Toth 1989:11).

They contended that it is the emotional suffering underlying child maltreatment, and not necessarily the actual maltreatment suffered by the parent in childhood, that is passed down from parents to children in a significant proportion of families (Vondra and Toth 1989). If this contention is accepted, then research into intergenerational transmission should be focusing on maltreatment in general, rather than limiting itself to the current predominant focus on physical violence. Given the dearth of existing research investigating the intergenerational transmission of neglect or emotional abuse, existing research is incorporated into the following sections reporting on physical and sexual abuse data, and other factors which may influence intergenerational transmission.

Physical abuse

The most frequently cited characteristic of physical abuse perpetrators in the literature is a childhood history of physical abuse (Milner and Chilamkurti 1991). Apart from the repeated suggestion of a relationship between a childhood history of physical abuse and adult physical abuse, studies have reported a relationship between the seriousness and chronicity of childhood abuse and the potential to physically abuse as an adult (Straus, Gelles and Steinmetz 1980). Merely witnessing physical abuse as a child has also been associated with an increased abuse potential as an adult (Milner, Robertson and Rogers 1990).

The crux of the intergenerational transmission hypothesis is that abusive parents have experienced maltreatment as a child. However, it may be that a childhood history of abuse is merely an identifier for other family problems such as inappropriate parental expectations, poor parental communication, or a lack of parental emotional support, all of which may mediate the transmission of maltreatment (Milner and Chilamkurti 1991). Milner, Robertson and Rogers (1990) contend that despite the attention paid to the transmission of physical abuse, there is still a need for more in-depth assessment of the specific factors which enhance or reduce the potential for adults to physically abuse. For example, additional data is required on the relationship between different types of childhood assault at different ages and later parental physical abuse.

Physical punishment

It is often asserted that there is a strong relationship between the childhood experience of severe corporal punishment and the subsequent use of such techniques on one's own children (Webster-Stratton 1985, as cited in Muller, Hunter and Stollak 1995). Weiss, Dodge, Bates and Petit 1992 (as cited by National Research Council 1993) contended that physical discipline, or more specifically, harsh disciplinary practices have been causally linked with a maladaptive social information processing style which in turn leads to aggression. That is, it is argued that severe discipline fosters more aggression in a child via the learning of aggressive patterns of response.

Sexual abuse

Clinical and empirical studies of incestuous families have indicated that both parents, mother and usually the father/perpetrator, frequently report growing up in a chaotic, dysfunctional family often described as involving substance misuse, marital conflict, parental divorce and childhood sexual abuse (Hanson, Lipovsky and Saunders 1994). The individual effects of a childhood history of sexual abuse and the other family characteristics on the potential to offend have not been clearly delineated to date.

In the Issues paper 'Update on Child Sexual Abuse' (Tomison 1995c), data from the Australian studies by Hiller, Goddard and Diemer (1989) and Tomison (1994) were used to flesh out the relationship between sexual abuse, physical abuse and domestic violence. As mentioned previously, Goddard and Hiller (1993) reported that almost three quarters of parents in a hospital study who abused their children reported that they themselves were abused in childhood - with virtually identical proportions among sexual and physical abuse cases (72 and 74 per cent respectively). Higher rates of parents with a childhood history of abuse (in both sexual and physical abuse cases) were reported in families where domestic violence was also present.

Child sexual abuse, like child physical abuse, was perceived to arise as a function of the misuse of personal power, and was taken as another example of male attempts to control others through the use of violence. Under this assumption, both physical and sexual abuse occur as a result of a violent, coercive environment, and the similar proportions of parental childhood history of abuse in sexual and physical abuse cases in this study were taken as affording some empirical evidence for Finkelhor's contention that sexual abuse can also be intergenerationally transmitted (Goddard and Hiller 1993).

Sex offenders

Knowledge about the perpetrators of child sexual abuse has largely been gathered through studies with incarcerated offenders. As mentioned previously, this is problematic given that convicted sex offenders are not representative of the sexual offender population, being more likely to represent the chronic, most severe forms of offending (Finkelhor 1984). Data from perpetrator studies should thus be treated with caution (Wurtele and Miller-Perrin 1993).

Fattah (1992, as cited in Briggs and Hawkins 1996) noted the failure of researchers to take account of the effects that a childhood history of sexual abuse may have in later sex offending. Fattah contended that the search for a typical sex offender profile has been a futile exercise because of the failure to recognise that the profile of a typical offender is identical to that of a typical victim. That is, victim and offender populations are not mutually exclusive, rather they are homogenous populations which overlap to a large extent (Fattah 1992, as cited in Briggs and Hawkins 1996).

In order to investigate the effects of prior childhood sexual abuse on subsequent offending, Briggs and Hawkins (1996) conducted an Australian study of incarcerated sex offenders and of self-reported, non-offending males who admitted being sexually abused as children. While a specific definition of child sexual abuse was not provided, sexual abuse appeared to be defined as any form of physical contact of a sexual nature occurring between the child and an older child or adult. Overall, 93 per cent of the incarcerated offenders reported being sexually abused as children.

Briggs and Hawkins found that the sex abusers often regarded their own abuse as 'normal, sometimes enjoyable, and were abused by significantly more perpetrators than the non-offending sample. In contrast non-offenders were more likely to report their own abuse as negative. Briggs and Hawkins speculated that men who normalised their own experience of sexual abuse may be more likely to perpetrate sexual abuse themselves. Ryan (1989, as cited in Briggs and Hawkins 1996) suggested that the normalisation of sexual abuse via denial or repression of the traumatic aspects and acceptance of the pleasurable aspects may enable the future denial of the harm caused by sexual abuse. Offenders who rationalise away the effects of the abuse they suffered may also rationalise away the damage they do when sexually abusing children (Briggs and Hawkins 1996).

However, Finkelhor (1984) provides a note of caution. Returning to a theme running throughout this paper, Finkelhor argued that single factor theories are unable to fully explain maltreatment and noted the need for multi-factor explanations of sexual offending.

Child and adolescent sex offenders

Like adult perpetrators young offenders are a heterogeneous group (Vizard, Monck and Misch 1995) with few broad characteristics. Among the most widely reported of which, is a history of sexual abuse and/or other maltreatment. While it has been contended that the majority of male children who are sexually assaulted do not become sexual offenders (Becker 1988), estimations of the rate of sexual victimisation for young offenders have varied from 30 to 70 per cent (Watkins and Bentovim 1992).

The recent Victorian Parliamentary Crime Prevention Committee report (1995) noted that practitioners had reported that adolescents displaying the early signs of sex offending tended to grow up and commit sex offences unless they were provided with treatment. The need to break the cycle of offending at an early stage was echoed by many professionals and highlights the targeting of young sex offenders as a special population for intervention.

Gender differences

Until recently, theory and research have understated gender differences in the occurrence and transmission of maltreatment, and the inter-relationships between different types of violence (for example, child abuse and domestic violence) (Goddard and Hiller 1993, Tomison 1995a). Yet there is evidence implicating gender in the intergenerational transmission of aggression (Langeland and Dijkstra 1995).

Straus, Gelles and Steinmetz (1980) found that victims abused by a same-sex parent were more likely to use aggression in later life than are victims abused by a parent of the opposite sex. Social learning theory posits that children perceive themselves as being more similar to the same-sex parent. Under this approach, a child 'identifies' strongly with a same-sex parent, and has a tendency to copy and internalise all the attitudes, behaviors and idiosyncracies of the same-sex parent (Peterson 1984).

Hunter and Kilstrom (1979) indicated a much stronger relationship between a mother's reported history of childhood maltreatment and the subsequent maltreatment of a child, than for a father's history and subsequent maltreatment. However this finding needs further exploration, given that there is very little research into the antecedents of male adults who maltreat children (Corby 1993). The vast majority of studies, with the exception of incest research, have focused investigations of the transmission of violence and abuse on mothers, despite the fact that to the best of our knowledge, men are responsible for at least half of all physical abuse (Corby 1993).

The failure of workers and researchers to engage father figures in casework or research has been attributed to sexist, cultural and legal assumptions whereby mothers having any role in the care of their children are virtually automatically assumed to be accountable (Hiller, Goddard and Diemer 1991, Corby 1993, Langeland and Dijkstra 1995). It appears researchers and practitioners have a tendency to rely too readily on mothers as informants in families because of assumptions about the mothering role, and their greater availability (Corby 1993). Thus, as Stark and Flitcraft (1988) concluded, the 'abusing parent' is frequently a euphemism for mother.

Domestic violence

Using a retrospective sample of marital or cohabiting heterosexual couples, Kalmuss (1984) found that adults who had been hit by their parents, or who had witnessed domestic violence were more likely to be involved in marital aggression themselves. It was found that witnessing domestic violence was actually the better predictor of later violence. Straus, Gelles and Steinmetz (1980) found that the rate of domestic violence was dramatically higher for the sons of batterers than for the sons of non-violent fathers.

There is a growing body of research suggesting that child abuse and domestic violence are linked within families (McKay 1994). Research examining battered women populations in refuges have indicated that domestic violence and child abuse are strong predictors of each other. In an assessment of child abuse cases, Stark and Flitcraft (1988) found that almost two-thirds of such cases occurred in families where battered women were caring for the child. In contrast, research on domestic violence (Stacey and Shupe 1983, cited in McKay 1994) conservatively estimated that some form of child abuse was up to 15 times more likely in families where domestic violence was occurring.

As mentioned previously, data from the Australian studies by Hiller, Goddard and Diemer (1989) and Tomison (1994) has been used to flesh out the relationship between domestic violence, physical and sexual abuse. In both studies it was contended that a violent, coercive environment was almost as likely for child sexual abuse cases as it was for physical abuse cases, particularly with the more severe cases of physical and sexual abuse.

Both studies also reported an association between the presence of domestic violence in families and the likelihood that a parent had disclosed a history of childhood maltreatment. In addition, domestic violence was associated with higher rates of both child sexual and physical maltreatment (Hiller, Goddard and Diemer 1989, Tomison 1994). Other research has indicated a greater propensity for domestic violence in boys who have witnessed domestic violence, or who had been assaulted themselves (Kalmuss 1984, Straus, Gelles and Steinmetz 1980). Therefore, witnessing domestic violence, or being assaulted is associated not only with the intergenerational transmission of spousal violence, but also with the transmission of a propensity for child sexual or physical abuse. Further research is required to elaborate on these relationships.

Criminal behavior

In a classic study by Widom, first released in 1989, the long-term consequences of child maltreatment were assessed in a longitudinal study conducted in the United States. Approximately 1500 children (900 victims of substantiated child maltreatment, 667 matched comparison group, not known to have been maltreated) were followed from childhood through adolescence to adulthood, in order to assess the criminal arrest records for each group over time. Widom (1992) reported that while the majority of subjects in both groups had no juvenile or adult criminal record, being maltreated as a child increased the likelihood of arrest as a juvenile by 53 per cent, arrest as an adult by 38 per cent, and arrest for violent crime by 38 per cent.

Maltreated children were significantly more likely to be arrested for juvenile crime, adult crime, and for violent crimes. In addition, maltreated children were a year younger on average when first arrested (16.5 versus 17.3 years), committed nearly twice as many offences (2.4 per cent versus 1.4 per cent), and appeared to be arrested more frequently (17 per cent versus 9 per cent with 5 or more arrests).

Note: All differences are significant

A history of child maltreatment also had a substantial impact on the likelihood of females engaging in crime. A history of child maltreatment increased the probability of arrest for females by 77 per cent. Being abused or neglected as a child did not increase the likelihood of a woman being arrested for violent crime.

An analysis of the relationship between the type of maltreatment experienced in childhood and subsequent criminal behavior revealed that approximately 16 per cent of children who were physically abused were arrested for violent crime. Of particular interest, 12.5 per cent of neglected children were also arrested for violent crime. After controlling for the effects of age, sex and race, there was still a significant relationship between neglect and subsequent violence.

Widom (1992) concluded that the notion that only childhood victims of physical violence resort to violence in later years may be rather limited. She argued that a continuum of family situations may contribute to violence, incorporating both physical abuse and neglect. Both forms of maltreatment appeared to be equally potent predictors of later violence and criminal behavior.

Substance abuse

Another factor commonly associated with the occurrence of child maltreatment is substance abuse (Hayes and Emshoff 1993). Studies have suggested that chemical dependence is present in at least half of the families known to the child welfare system because of child maltreatment, some place this figure nearer to 80 per cent (Morrison Dore, Doris and Wright 1995).

The frequently identified association between substance abuse and aggressive behavior has suggested that the use of alcohol and drugs may be a significant risk factor for all types of child maltreatment (National Research Council 1993, Morrison Dore, Doris and Wright 1995). However, the associations between the misuse of alcohol, drugs and child abuse are not well understood (Widom 1992, as cited in National Research Council 1993), particularly the relationship between severity, chronic patterns of substance misuse and the risk of child maltreatment (Widom 1992, as cited in National Research Council 1993).

Parental substance abuse has also been linked to family violence and as a factor in the intergenerational transmission of maltreatment (Hayes and Emshoff 1993). The relationship or interaction between individual, familial and societal risk factors, substance misuse and family violence is not clear, but it has been asserted that the children of substance abusers are more likely to engage in antisocial, delinquent, and other problem behaviors, and to be arrested with greater frequency for criminal offences (Hayes and Emshoff 1993).

While not all of the children's 'acting out' is violent in nature, it is likely that as a group, children with substance abusing parents become predisposed towards violent behavior, and are likely to perpetuate familial cycles of violence and substance abuse (Hayes and Emshoff 1993). The general relationship between substance misuse and child maltreatment will be extensively reviewed in a further Issues Paper.

Psychological factors

The childhood experience of abuse is not only traumatic at the time, but may negatively affect victims' mental and physical wellbeing in the long term (Browne and Finkelhor 1986). Mullen, Martin, Anderson, Romans and Herbison (1996) investigated the long-term impact of severe physical, sexual and emotional abuse in a community sample of women in New Zealand. They reported an association between a history of abuse (all types) and an increased vulnerability to a range of mental health, personal and social problems in adult life, but concluded that child abuse was merely one of a range of other developmental traumas and disadvantages which may have been acting as potential causal agents.

They did report, however, that women who suffered severe abuse as children were not randomly distributed throughout the population; rather they occurred 'more often in those from disturbed and disrupted home backgrounds' (Mullen, Martin, Anderson, Romans and Herbison 1996:18).

Mullen et al. (1996) noted that poor parent/caregiver mental health was associated with both an increased risk of being maltreated as a child and increased risk of psychopathology later in life. The relationship between mental disorder and child maltreatment will be explored more fully in another Issues Paper. However, it is clear that there is a need to flesh out the effects of parent mental disorder and the transmission of psychopathology and/or child maltreatment.

Mediating factors

While support for a relationship between a childhood history of maltreatment and the subsequent maltreatment of children is well-documented, a relatively small number of studies have explicitly investigated the relationship between other childhood risk factors and adult child abuse potential (Milner, Robertson and Rogers 1990). Straus, Gelles and Steinmetz (1980) for example, found a relationship between a chronic, long-term pattern of childhood maltreatment and later child abuse potential.

Other researchers have investigated the lack of economic resources as a central factor in child maltreatment and intergenerational transmission (Milner and Crouch 1993). Lower income commonly occurs in families with lower intelligence, lower educational status, and greater levels of mental disorder and stress (National Research Council 1993).

Such negative demographic conditions may set the stage for the intergenerational transmission of maltreatment because they make it difficult for parents to teach the child how to adequately function (Gabinet 1983, cited in Milner and Crouch 1993).

Milner, Robertson and Rogers (1990) explored the effects of age at which abuse was experienced, and the effects of actually experiencing abuse compared with only witnessing maltreatment, on adult physical abuse potential. Using an undergraduate sample, they found that a significant positive association was found between a childhood history of maltreatment and predicted adult potential to maltreat. In addition it was found that experiencing maltreatment prior to puberty produced a significantly higher adult potential for abuse.

Greater severity of maltreatment (particularly the presence of physical maltreatment) was found to produce a greater abuse potential. Being sexually abused as a child produced a modest association with adult potential for physical abuse. The study indicated that both witnessing maltreatment and being maltreated were associated with adult potential to abuse. Overall, however, it was not determined which variables were responsible for the relationships, or what moderator variables may have affected the potential to maltreat.

Finally, it has been demonstrated that several cognitive biases commonly held by high risk parents are also displayed by their children (Milner and Crouch 1993). Physically abusive parents often perceive their child's 'bad' behavior as being due to internal factors (child's character) while 'good' behavior is attributed to external and unstable factors, such as luck or particular situations (Larrance and Twentyman 1983, as cited in Milner and Crouch 1993). High risk parents have been found to have unrealistic expectations of their child in general (Kravitz and Driscoll 1983, as cited in Milner and Crouch 1993), and in particular, for the child's compliance after discipline (higher expectations for compliance after punishment for minor transgressions, lower expectations for compliance after punishment for serious transgressions) (Chilamkurti and Milner 1993, as cited in Milner and Crouch 1993).

Chilamkurti and Milner (1993, as cited by Milner and Crouch 1993) contended that such parental beliefs may be learned by children, and this may mediate, in part, the intergenerational transmission of abuse. Thus, the transmission of cognitive biases may enhance the potential for intergenerational transmission of maltreatment.

Breaking the cycle

As mentioned previously, Milner and Chilamkurti (1991) suggested that in cases of intergenerational transmission, a childhood experience of abuse may act merely as a marker or indicator variable for other unidentified factors, such as inappropriate parental expectations, poor parental communication, or a lack of parental emotional support, which may more directly affect the subsequent perpetration of abuse. There may be other factors which buffer or attenuate the effects of experiencing maltreatment as a child (Caliso and Milner 1994). Though there is a paucity of controlled research investigating ameliorating factors (Caliso and Milner 1994), the available evidence is summarised below.

First, non-abusing mothers in the prospective studies conducted by Hunter and Kilstrom (1979) and Egeland, Jacobvitz and Papatola (1987 as cited in Egeland 1988) reported having extensive social supports, were less likely to have been maltreated by both of their parents, and were more likely to report a supportive relationship with one parent while growing up. These mothers acknowledged and gave a coherent account of the maltreatment they suffered as children, and were more openly angry about their experiences. Having a supportive spouse/partner (Quinton, Rutter and Liddle 1984; Crockenberg 1987, Egeland, Jacobvitz and Papatola 1987, as cited in Youngblade and Belsky 1990), fewer stressful life events, and making a conscious resolve not to repeat their history of maltreatment also characterised non-maltreating mothers (Egeland, Jacobvitz and Papatola 1987, as cited in Youngblade and Belsky 1990).

That parents who experience more social support were less likely to continue the cycle of violence, compared with those who have a poor support system, is congruent with the literature on physical abuse which indicates that physically abusive or neglectful parents are often isolated, lack family and peer support, and are not involved in community activities (Young 1964, as cited in Caliso and Milner 1994, Tomison 1995b).

It has also generally been found that for parents with a history of abuse, those who do not become physically abusive have less rigid expectations about their child, and are happier in their interpersonal relationships, than parents who are physically abusive (Caliso and Milner 1994). Within a social information processing model (Milner 1993 as cited in Caliso and Milner 1994), a greater flexibility in expectations towards children may reduce the probability that parents will engage in coercive exchanges with their children.

Finally, a strong, supportive religious affiliation (Kaufman and Zigler 1989, as cited in Langeland and Dijkstra 1995), and positive school experiences as a child (Rutter 1989, as cited in Langeland and Dijkstra 1995) may also ameliorate the transmission of violence.

Prevention

Fry (1993) proposes two levels of approach to breaking the cycle of violence. First, taking a cultural perspective, Fry advocates the promotion of attitudes and beliefs that run counter to the use of physical force. The Scandinavian countries have already begun this process by outlawing corporal punishment by adults some years ago (Cashmore and de Haas 1995). However Australia, like most of the western world, is a country where attitudes and beliefs favoring physical chastisement are pervasive (Cashmore and de Haas 1995). Fry (1993) argues that in order to stop intergenerational transmission, violence in the community, from corporal punishment to domestic violence needs to be portrayed as unacceptable.

Second, explicit training on non-violent conflict resolution, problem-solving and childrearing techniques needs to be widely available. This would have two benefits, a reduction in corporal punishment and abuse, and the disruption of the intergenerational cycle of violence, as children would acquire problem-solving and conflict resolution techniques which did not involve violence (Fry 1993). Such training could be incorporated into existing life skills programs in schools and counselling agencies, indicating the ineffectiveness of corporal punishment (Fry 1993, Cashmore 1995), and providing instruction in alternative childrearing and conflict resolution strategies.

Given that early intervention strategies are one of the most effective ways to ameliorate the effects of maltreatment, there needs to be increased recognition of the signs of maltreatment and serious attempts to intervene as early as possible (Widom 1992). Particular attention should be paid to children exhibiting behavioral problems. Widom (1992) for example, found that these children had the highest risk of later juvenile and adult arrest, and for engaging in violent criminal behavior.

In the United States, Wolfe (1991) established an approach designed to address a broad population with a wide variety of parenting problems. In his program parents were taught child management and parental sensitivity techniques, and encouraged to participate in peer support groups. Specific attention was focused on educating parents about alternate forms of discipline and appropriate forms of anger management.

In Australia, similar approaches have been set up by non-government and government bodies across the nation. A number of programs currently stored on the National Child Protection Clearing House Prevention Programs database exemplify the prevention objectives outlined above. It should be noted, however, that the majority of the family support programs target parents or the family as a whole, rather than focusing on a maltreated or 'at risk' child. It is argued that in order to break the intergenerational cycle of maltreatment, programs should aim to address the potential for maltreatment in both the parents and the children. While parent-focused initiatives may break the cycle of abuse in the parent, prevention should also address the child's needs, ameliorating any negative attitudes and behaviors already experienced, and further reducing the likelihood of intergenerational transmission.

Three programs from the National Child Protection Clearing House programs database* are presented as examples of programs attempting to prevent intergenerational transmission.

Centacare Respite Camp for Boys, New South Wales

Respite is provided for boys aged 7 to 12 years exhibiting challenging behavior. The objectives of the program are to give the boys and their families 'time out', and to provide a fun, safe environment for the boys. Group leaders model socially acceptable behavior, provide positive non-violent role models for the boys and teach effective problem solving and conflict resolution skills.

Centaceare Changing Abusive Behaviours (CAB) Program, Tasmania

Adult males who voluntarily attend the program are assisted to deal with violent or abusive behavior by helping them to understand their motivations and by providing alternatives to intimidation, anger and abuse. The program includes raising awareness of sociocultural values and offers conflict resolution, self-esteem and communication skills training. A support group for partners of the men is offered concurrently with the CAB group. Thus, the CAB program attempts to address the violent and/or abusive behavior of men, and support women to overcome their experiences of violence.

Parent Aide Program, Caloundra Community Centre, Queensland

Based on successful programs operating from the Mater, and Childrens Hospitals in Brisbane, trained, supervised, volunteer parent aides assist stressed parents at risk of maltreating their children. The program aims to reduce stress, boost parental self-esteem, independence, and feelings of control. Parents are taught effective communication and problem solving skills and appropriate expressions of anger. Parent-child relationships are enhanced through parents' gaining greater knowledge of child development, appropriate expectations of children, reduced stress and wider social involvement/social support network. Home visit/parent aide programs such as this have been shown to be particularly effective in preventing maltreatment (Hamilton 1989, Tomison 1995b); a number of such programs operate across Australia.

The Centacare Respite program is focused on preventing intergenerational transmission in children already acting out via the modelling of appropriate social behavior. The other two programs attempt to address intergenerational transmission through programs which predominantly target parents. The Changing Abusive Behaviours Program is specifically focused on breaking the cycle of violence, while the Parent Aide Program prevents intergenerational transmission as a function of enhancing protective factors which reduce the likelihood of future maltreatment.

Taking a holistic approach

Interventions aimed at preventing child maltreatment may be ineffective in changing an abusive situation if other forms of family violence operating in the family are not understood and addressed (Stanley and Goddard 1993). A holistic approach to violence is needed in order to prevent intergenerational transmission.

First, any holistic approach should also seek to address all types of child maltreatment experienced by a child. As previously mentioned, the intergenerational transmission of neglect has been, for the most part, neglected. Policies and prevention programs should therefore be developed that recognise the high risk of neglect as well as abuse. As Widom states, '[research] shows that today's victim of neglect may well be a defendant in tomorrow's violent criminal case' (1992:6).

Second, observing significant others being maltreated (especially siblings and parents) by someone the child closely identifies with is reported to be more intimately related to intergenerational transmission than actually being maltreated (Brown 1983, as cited in Hamilton 1989). Therefore, children who have witnessed their siblings being physically assaulted need to be incorporated into any treatment program in order to prevent the development of problematic behavior and to reduce trauma.

Third, women's refuges, accommodation programs and domestic violence groups focus by definition, on supporting battered mothers. However, there must be recognition of the damage done to the child by a violent homelife, and where possible attempts should be made to also address the child's needs. The children of battered women need access to preventative programs, whether they are modelled by refuge or other domestic violence workers, or available by referral to external support agencies.

Sexual abuse

Briggs and Hawkins (1996) provided new information for offender rehabilitation programs. They contended that abusers may deny the negative effects of sexual abuse on their lives, and in turn discount the negative effects on their victims. Treatment should therefore emphasise the harm done to victim and attempt to counter the cognitive distortions evident with some perpetrators The prevention of the intergenerational transmission of sexual abuse may be prevented in part by overcoming the perpetrator's denial of the harm he/she does to a child.

Treatment programs for offenders should therefore incorporate methods of confronting abusers in order to make them accept responsibility for their actions (Morrison 1989, as cited in Briggs and Hawkins 1996).

As has been mentioned previously, the effects of maltreatment, and the potential for intergenerational transmission can be ameliorated by early intervention programs (Widom 1992, Crime Prevention Committee 1995). With regard to young sex offenders it has been unanimously concluded that the professional community must 'work to identify and treat child (and adolescent) perpetrators and not deny the potential risks to themselves and the community' (Johnson, 1988, as cited in Children's Protection Society 1995:14).

Future directions

In order to come to a better understanding of the intergenerational nature of maltreatment, and to break the cycle of abuse, greater attention needs to be paid to the broader family context within which maltreatment may occur (for example, father figures, grandparents, parent siblings) (Langeland and Dijkstra 1995). Research to date has been overly focused on mothers without in-depth investigation of the roles of other adult family figures.

Trigenerational studies, where maltreatment is investigated across three generations of families offer one way forward. To date, studies on more extended intergenerational patterns of maltreatment have focused on the continuation of the maltreatment cycle (Oliver and Taylor 1971, Oliver and Buchanan 1979, as cited in Langeland and Dijsktra 1995). However, studies of three generations can also shed light on the protective factors and coping strategies which may enable parents to avoid repeating the abusive cycle (Langeland and Dijkstra 1995).

In many cases where parents maltreat, all of the children in the family experience some form of maltreatment (Jones, Pickett, Oates and Barbor 1987). Investigations of the adult siblings of abusive parents may provide an important source of data on factors which promote a break in the abusive cycle. Such studies have important implications not only for practice, but at a theoretical level, in guiding and improving prevention and protection efforts (Langeland and Dijkstra 1995).

Finally, what is apparent from the literature is that the prevention of intergenerational transmission of maltreatment is possible. Experiencing child maltreatment need not inevitably doom a child to repeating that violence. If 'therapeutic experiences can be provided which instil a sense of trust, of self-worth, while providing behavioral skills to deal with others, particularly in affectively charged situations, then there is every reason to expect that the effects of child maltreatment can be ameliorated' (Youngblade and Belsky 1990:139).

Rather than spend another decade debating the proportion of parents with a childhood history of maltreatment who will become abusive, it would be more positive to focus on the factors, conditions and interventions which can prevent the cycle of violence from continuing through another generation.

References

• Angus, G. and Woodward, S. (1995), 'Child abuse and neglect Australia 1993-94', Child Welfare Series No. 13, Australian Institute of Health and Welfare, AGPS, Canberra.
• Bandura, A. (1977), Social Learning Theory, Prentice-Hall, Englewood Cliffs, NJ.
• Belsky, J. (1980), 'Child maltreatment: an ecological integration', American Psychologist, vol.35, no.4, pp. 320-335.
• Briggs, F., Hawkins, R.M.F. (1996), 'A comparson of the childhood experiences of convicted male child molesters and men who were sexually abused in childhood and claimed to be nonoffenders', Child Abuse and Neglect, vol.20, no.3, pp.221-233.
• Bronfenbrenner, U. (1979), The Ecology of Human Development, Harvard University Press, Cambridge, MA.
• Browne, K. (1988), 'The nature of child abuse and neglect: an overview', in K. Browne, C. Davies and P. Stratton, (eds.), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• Browne, K. and Finkelhor, D. (1986), 'Impact of child sexual abuse: a review of the research', Psychological Bulletin, vol.99, pp.66-77.
• Browne, K. and Saqi, S. (1988), 'Approaches to Screening for Child Abuse and Neglect', in K. Browne, C. Davies and P. Stratton, (eds.), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• Caliso, J.A. and Milner, J.S. (1994), 'Childhood physical abuse, childhood social support, and adult child abuse potential', Journal of Interpersonal Violence, vol.9, no.1, pp.27-44.
• Cashmore, J. and de Haas, N. (1995), Legal and Social Aspects of the Physical Punishment of Children, AGPS, Canberra.
• Children's Protection Society (1995), Adolescent Sex Offender Treatment Program, Children's Protection Society, Melbourne.
• Crime Prevention Committee, Parliament of Victoria (1995), Combating Child Sexual Assault: An Integrated Model, First Report, LV North, Government Printer, Melbourne.
• Corby, B. (1993), Child Abuse: Towards a Knowledge Base, Open University Press, Buckingham, PA.
• Egeland, B. (1993) 'A history of abuse is a major risk factor for abusing the next generation', in R.J. Gelles and D.R. Loseke (eds) Current Controversies on Family Violence, Sage Publications, Newbury Park, California.
• Finkelhor, D. (1979), Sexually Victimized Children, The Free Press, New York.
• Finkelhor, D. (1984), Child Sexual Abuse: New Theory and Research, The Free Press, New York.
• Freeman-Longo, R.E. (1986), 'The impact of sexual victimisation on males', Child Abuse and Neglect, vol.10, pp.411-414.
• Fry, D.P. (1993), 'The intergenerational transmission of disciplinary practices to conflict', Human Organization, vol.52, no.2, pp.176-185.
• Gelles, R.J. and Loseke, D.R. (1993), 'Issues in causes', in R.J. Gelles and D.R. Loseke (eds) Current Controversies on Family Violence, Sage Publications, Newbury Park, California.
• Gil, D.G. (1970), Violence against Children: Physical Child Abuse in the United States, Harvard Univeristy Press, Cambridge, MA.
• Goddard, C. R. and Hiller, P. C. (1992), Tracking Physical and Sexual Abuse Cases from a Hospital Setting into Victoria's Criminal Justice and Child Protection Systems: A Report for the Victorian Law Foundation, Volumes 1-3, Department of Social Work and Department of Anthropology and Sociology, Monash University, Melbourne.
• Goddard, C.R. and Hiller, P.C. (1993), 'Child sexual abuse: assault in a violent context', Australian Journal of Social Issues, vol.28, no.1, pp.20-33, February.
• Hamilton, L.R. (1989), 'Variables associated with child maltreatment and implications for prevention and treatment', in J.T. Pardeck (ed) Child Abuse and Neglect: Theory, Research, and Practice, Gordon and Breach Science Publishers, New York.
• Hanson, R.F., Lipovsky, J.A. and Saunders, B.E. (1994), 'Characteristics of fathers in incest families', Journal of Interpersonal Violence, vol.9, no.2, pp.155-169.
• Hayes, H.R. and Emshoff, J.G. (1993), 'Substance abuse and family violence', in R.L.Hampton, T.P. Gullotta, G.R. Adams, E.H. Potter III, and R.P. Weissberg (eds) Family Violence: Prevention and Treatment, Issues in Children's and Families' Lives, vol. 1, Sage Publications, Newbury Park, California.
• Hiller, P.C., Goddard, C.R. and Diemer, K.M. (1989), 'The nature of and institutional reaction to child abuse: continuities and discontinuities in the comparative analysis of cases of physical and sexual abuse', in Proceedings of the Second Annual Conference of the Victorian Society for the Prevention of Child Abuse and Neglect: Twelve Months On: Achievements and Future Directions, pp.111-134, Vicspcan, Melbourne.
• Hiller, P.C., Goddard, C.R. and Diemer, K.M. (1991), 98 Cases of Child Abuse Presenting at a Major Melbourne Hospital: Sociodemographic factors and Institutional Reactions, Volumes I and II, Department of Anthropology and Sociology, Monash University, Melbourne.
• Hunter, R.S. and Kilstrom, N. (1979), 'Breaking the cycle in abusive families', American Journal of Psychiatry, vol.36, October, pp.1320-1322.
• Jones, D.N., Pickett, J., Oates, M.R. and Barbor, P. (1987), Understanding Child Abuse (2nd ed), Macmillan Education, Houndsmills, Basingstoke.
• Kalmuss, D. (1984), 'The intergenerational transmission of marital aggression', Journal of Marriage and the Family, vol.46, pp.11-19.
• Kaufman, J. and Zigler, E. (1987), 'Do abused children become abusive parents?', American Journal of Orthopsychiatry, no.57, pp.186-192.
• Kaufman, J. and Zigler, E. (1993), 'The intergenerational transmission of abuse is overstated', in R.J. Gelles and D.R. Loseke (eds) Current Controversies on Family Violence, Sage Publications, Newbury Park, California.
• Langeland, W. and Dijkstra, S. (1995), 'Breaking the intergenerational transmission of child abuse: beyond the mother-child relationship', Child Abuse Review, vol.4, no.1, pp4-13.
• Lewis, M. (1988), 'What can child development tell us about child abuse?', in K. Browne, C. Davies and P. Stratton, (eds.), Early Prediction and Prevention of Child Abuse, John Wiley and Sons, Chichester.
• McKay, M.M. (1994), 'The link between domestic violence and child abuse: assessment and treatment considerations', Child Welfare, vol.73, no.1, pp.29-39, January/February.
• Milner, J.S. and Chilamkurti, C. (1991), 'Physical child abuse perpetrator characteristics: A review of the literature', Journal of Interpersonal Violence, vol.6, pp.345-366.
• Milner, J.S. and Crouch, J.L. (1993) 'Physical child abuse', in R.L.Hampton, T.P. Gullotta, G.R. Adams, E.H. Potter III, and R.P. Weissberg (eds) Family Violence: Prevention and Treatment, Issues in children's and families' lives, vol. 1, Sage Publications, Newbury Park, California.
• Milner, J.S., Robertson, K.R. and Rogers, D.L. (1990), 'Childhood history of abuse and adult child abuse potential', Journal of Family Violence, vol.5, no.1, pp.15-34.
• Morrison Dore, M., Doris, J.M. and Wright, P. (1995), 'Identifying substance abuse in maltreating families: a child welfare challenge', Child Abuse and Neglect, vol.19, no.5, pp.531-543.
• Mullen, P.E., Martin, J.L., Anderson, J.C., Romans, S.E. and Herbison, G.P. (1996), 'The long-term impact of the physical, emotional, and sexual abuse of children: a community study', Child Abuse and Neglect, vol.20, no.1, pp.7-21.
• Muller, R.T., Hunter, J.E. and Stollak, G. (1995), 'The intergenerational transmission of corporal punishment: a comparison of social learning and temperament models', Child Abuse and Neglect, vol.19, no.11, pp.1323-1335.
• National Research Council (1993), Understanding Child Abuse and Neglect, National Academy Press, Washington DC.
• Peterson, C.C. (1984), Looking Forward Through the Life Span: Developmental Psychology, Prentice-Hall of Australia, Sydney.
• Ryan, M.G., Davis, A.A. and Oates, R.K. (1977), 'One hundred and eighty-seven cases of child abuse and neglect', Medical Journal of Australia, vol.2, pp.623-628.
• Stanley, J. R. and Goddard, C.R. (1993), 'The association between child abuse and other family violence', Australian Social Work, vol.46, no.2, pp.3-8, June.
• Stark, E. and Flitcraft, A. (1988), 'Women and children at risk: a feminist perspective on child abuse', International Journal of Health Services, vol.18, no.1, pp.97-118.
• Steele, B. F. and Pollack, C.B. (1968), 'A psychiatric study of parents who abuse infants and small children', in R.E. Helfer and C.H. Kempe (eds.), The Battered Child, University of Chicago Press, Chicago.
• Straus, M.A., Gelles, R.J. and Steinmetz, S.K. (1980), Behind Closed Doors: Violence in the American Family, Anchor Books, New York.
• Tomison, A.M. (1994), An Evaluation of Child Abuse Decision Making in the Barwon Region: A Report for the Victorian Health Promotion Foundation, Volumes I and II, Department of Social Work and Human Services, Monash University, Melbourne.
• Tomison, A.M. (1995a), 'Linking child abuse and other family violence: findings from a case tracking study', Family Matters, no.41, Winter, pp.33-38.
• Tomison, A.M. (1995b), Spotlight on Child Neglect, Issues Paper No.4, National Child Protection Clearing House, Australian Institute of Family Studies, Melbourne.
• Tomison, A.M. (1995c), Update on Child Sexual Abuse, Issues Paper No.5, National Child Protection Clearing House, Australian Institute of Family Studies, Melbourne.
• Vondra, J.I. and Toth, S.L. (1989) 'Ecological perpsectives on child maltreatment: research and intervention', in J.T. Pardeck (ed) Child Abuse and Neglect: Theory, Research, and Practice, Gordon and Breach Science Publishers, New York.
• Vizard, E., Monck, E. and Misch, P. (1995), 'Child and adolescent sex abuse perpetrators: a review of the research literature', Journal of Child Psychology and Psychiatry, vol.36, no.5, pp.731-756.
• Watkins, B. and Bentovim, A. (1992), 'The sexual abuse of male children and adolescents: a review of current research', Journal of Child Psychology and Psychiatry, vol.33, pp.197-248.
• Widom, C.S. (1989), 'Does violence beget violence? A critical examination of the literature', Psychological Bulletin, vol.106, no.1, pp.3-28.
• Widom, C.S. (1992), 'The cycle of violence', Child Protection Seminar Series No.5, NSW Child Protection Council, Sydney.
• Wolfe, D.A. (1991), Preventing Physical and Emotional Abuse of Children, Guilford Press, New York.
• Wurtele, S.K. and Miller-Perrin, C.L. (1993), Preventing Child Sexual Abuse: Sharing the Responsibility, University of Nebraska Press, Lincoln.
• Youngblade, L.M. and Belsky, J. (1990), 'Social and emotional consequences of child maltreatment', in R.T. Ammerman and M. Hersen (eds.), Children at Risk: An Evaluation of Factors contributing to Child Abuse and Neglect, Plenum Press, New York.

* The Prevention Programs database is available for access in the Australian Institute of Family Studies library.