Parkinson, Patrick --- "Family Law and Parent-Child Contact: Assessing the Risk of Sexual Abuse" [1999] MelbULawRw 15; (1999) 23(2) Melbourne University Law Review 345
Family Law and Parent-Child Contact: Assessing the Risk of Sexual Abuse
PATRICK PARKINSON[*]
[Child sexual abuse allegations cause great difficulties for the Family Court, not only because so often there is no corroborative evidence, but because of the way that lawyers approach the notion of proof, and restrict the evidence which may be used to support a finding that a parent has sexually abused the child. The High Court held in M v M (1988) that the Family Court could restrict or deny contact where there is an unacceptable risk of abuse. However, there are differences between judges of the court on the application of the test and, in particular, whether the assessment of risk needs to be based upon findings of fact from which that risk may be inferred. In assessing the risk to the child of abuse in the future, more attention needs to be paid to risk factors other than the contested accounts of past abuse, based upon what is known in the social science literature about intrafamilial child sexual abuse and, in particular, what is known about the way offenders ‘groom’ their child victims. This article proposes a four stage process for decision-making in the Family Court which will require the court to examine a range of possible risk factors, and also factors which may reduce the risk of abuse in the future. In this way, courts may be able to make findings based upon facts, even though the facts relied upon may not involve findings about past abuse.]
CHILD SEXUAL ABUSE ALLEGATIONS AND THE FAMILY COURT
One of the most difficult dilemmas which face the judges of the Family Court of Australia is how to determine whether to suspend, restrict or to entirely deny contact where it is alleged that the parent seeking contact has sexually abused a child. Similar issues arise where a parent is seeking a residence order and it is alleged that he or she, or a new partner, has sexually abused the child.
There is evidence now emerging from a number of studies that resolving child protection issues in children’s cases has become a significant part of the Family Court’s workload. The largest number of such cases involve allegations of domestic violence which may raise issues about the safety and emotional well-being of the children.[1] However, sexual abuse allegations are also made in a significant number of cases concerning children. The Family Court’s study of 294 judgments in defended cases from all over Australia heard in 1990 found that seven percent of cases involved allegations of sexual abuse by the child’s father and, in another three percent of cases, it was alleged that the child had been abused by another adult such as a stepfather or other relative. [2] Thus in 10 percent of the defended hearings in the Family Court, the Court had to deal with the question of alleged sexual abuse. In a further 11 percent of cases, there were allegations of physical abuse or neglect.[3]
A review of over 700 cases awaiting pre-hearing conferences in the Melbourne registry of the Family Court in 1997 found that more than 40 percent of children’s cases involved allegations of some form of child abuse.[4] Research by Professor Thea Brown and her colleagues in Melbourne and Canberra found a similar pattern. Their analysis of cases in Melbourne and Canberra between January 1994 and June 1995 found that one half of all the cases which went to a pre-hearing conference involved allegations of some form of child abuse.[5] Of the cases which went to court, one quarter involved allegations of child abuse. In their detailed analysis of cases in Melbourne and Canberra, the researchers found that of all the cases in which child abuse was alleged, 24.1 percent involved allegations of sexual abuse. In Canberra, the percentage was 48.6.[6]
II EVIDENCE OF CHILD SEXUAL ABUSE
As has been widely observed, cases in which there are allegations of child sexual abuse often present perplexing problems for the Court.[7] If the Court finds that the child has been sexually abused by one of the parents, or perhaps by a new partner of one of them, in all but the most exceptional circumstances, it would have a decisive effect on the result of the case and would almost certainly lead to an order denying that parent contact with the child.[8] However, such findings are uncommonly made. There are numerous reasons for this.
A Corroborative Evidence
A major problem is the inherent difficulty in obtaining corroborative evidence which can support the allegations of a parent or the purported disclosure of the child.[9] In the majority of cases, clinical evidence is not available to confirm the abuse.[10] Most sexual abuse does not involve acts of penetration or violation sufficient to cause physical damage to the body of the child. The abuse consists of the perpetrator touching, groping, fondling and exploring genitalia, with or without digital penetration, or persuading or coercing the child to perform acts upon the perpetrator, such as fondling of the penis or masturbation.[11] Where there has been penetration of the vagina or anus of the child it may or may not leave indications of the abuse depending upon the age of the child and the degree of violation.[12] Even where there is medical evidence of penetration which supports the child’s account of the sexual abuse, that evidence may not be corroborative of the identity of the perpetrator unless traces of semen, pubic hair or other such evidence is found. Many signs of penetration are in any event equivocal. Medical indicators such as a reddened and sore vulva or anus are suspicious signs when taken together with the child’s statements of sexual abuse, but ultimately, medical practitioners may only be able to say in these circumstances that such medical indicators are consistent with abuse rather than conclusive of abuse.[13]
B Beliefs About False Allegations
A further difficulty in assessing allegations of child sexual abuse is that in the context of the breakdown of a marriage or de facto relationship it is widely believed by lawyers and others that there is a greater likelihood of unfounded allegations of sexual abuse than is otherwise the case.[14] The extent of this problem is not as great as many legal practitioners believe. In two Australian studies, the rate of unsubstantiated cases of sexual abuse, as assessed by state child protection workers, was no higher among couples involved in family law disputes than in relation to notifications of abuse generally.[15] A large US study reached the same conclusion.[16] Studies in other countries, involving small samples, have found a greater number of unsubstantiated allegations.[17] However, even if this is so, the research does not bear out the view that there are a large number of lies told by children. As Nicholson CJ, Baker and Tolcon JJ stated in D and Y, ‘finding an allegation to be false is not the same as finding a child to have lied.’[18] In one large study in the United States, 551 cases were reviewed. In 14 (2.5 percent) of these cases, erroneous reports of abuse emanated from children. In three cases, the child was being influenced by the parent; in a further three cases, an innocent event was misinterpreted as sexual abuse; and, in the remaining eight cases (1.5 percent), the reports from the children were assessed as fictitious.[19] Only the last group ought to be considered as lies told by the children, and it is a very small percentage of all reports.
There are many other reasons why unfounded allegations may be made in the context of divorce. The breakdown of a marriage often generates considerable hostility between the parents, particularly where one feels abandoned, abused or betrayed by the other.[20] In a climate of hostility such as this, innocent behaviour by the other parent may be misinterpreted as abusive, especially if the mother was the victim of domestic violence during the marriage or had very negative sexual experiences.[21]
C Legal Concepts of Proof
The problems in this area are not only problems about the paucity of probative evidence as lawyers understand it. Difficulties also arise from the way in which lawyers approach the issue of whether or not child sexual abuse has occurred. Fogarty J articulated this problem in his dissenting judgment in N and S and the Separate Representative:
The secrecy which usually surrounds sexual abuse, the nature of the offences which it involves, and the nature of the relationship between the perpetrator and the child, all militate against the furnishing of the type of evidence with which lawyers like to work. Of course, the lack of that type of evidence may say less about whether an alleged event did or did not occur, than it says about the inappropriateness of the legal concepts used to test the allegations, especially in the context of a case which centres on the welfare of the child.[22]
Before it can make a positive finding that a parent or other caregiver has sexually abused a child, the Family Court needs to be satisfied according to the civil standard,[23] as interpreted in Briginshaw v Briginshaw.[24] In this case, Dixon J said:
The seriousness of an allegation made, the inherent unlikelihood of an occurrence of a given description, or the gravity of the consequences flowing from a particular finding are considerations which must affect the answer to the question whether the issue has been proved to the reasonable satisfaction of the tribunal. In such matters ‘reasonable satisfaction’ should not be produced by inexact proofs, indefinite testimony, or indirect inferences.[25]
It has been held, applying the Briginshaw test, that given the gravity of the offence, a finding of child sexual abuse can only be justified by evidence satisfying a standard of proof towards the stricter end of the civil spectrum.[26] In making findings of fact on matters of this nature, the Family Court is also bound by the Evidence Act 1995 (Cth). The rules of evidence restrict the evidence which is admitted or the use to which certain evidence may be put in proving the allegation of abuse. For example, evidence of abusive behaviour towards another child in the family is not in itself probative of the abuse of the child who is the subject of the relevant proceedings, unless that evidence falls within the rules admitting tendency evidence, coincidence evidence or other evidence by way of background to the alleged abuse.[27] The rules of evidence thus adopt a narrow definition of relevance and a restrictive view of what evidence may be probative of abuse in a given case. This has the effect of isolating the alleged incident or incidents of abuse from much of its context as regards the behaviour of the alleged perpetrator and their relationship with the victim.[28]
A further problem in relation to the way lawyers view ‘proof’ is that, in order to justify a positive finding that a parent or other caregiver has sexually abused a child, what has to be proven is that one or more discrete events constituted a sexual assault. These events, treated as discrete happenings, need to be proven with a sufficient degree of specificity enabling the court to say whether, on the balance of probabilities, a particular event did or did not occur.[29] The cumulative picture of abuse, attested to generally by the child, is insufficient. Children’s experience of intrafamilial sexual abuse is, by way of contrast, typically one of an ongoing abusive relationship in which the various specific incidents of sexual abuse may well be largely undifferentiated from one another. A consequence of this is that a child may have great difficulty describing each event of assault with enough specificity and detail to satisfy the legal test of proof for each individual occurrence.[30]
It is thus not surprising that judges of the Family Court have been able to make a positive finding that there has been sexual abuse in so few cases, although there are a few examples.[31] Indeed in M v M, the High Court discouraged such findings,[32] saying that there are ‘strong practical family reasons why the court should refrain from making a positive finding that sexual abuse has actually taken place unless it is impelled by the particular circumstances of the case to do so.’[33] There is not the same reluctance to make findings on the balance of probabilities in other jurisdictions.[34]
In most cases which go to trial in the Family Court, the evidence will be insufficient to justify a finding that a parent or other caregiver has sexually abused the child. As Fogarty and May JJ noted in Re C and J,[35] the cases which Family Court judges have to hear are generally the ones in which the evidence is most unclear as to whether or not there has been sexual abuse. Cases where investigation or other evidence indicate that the allegations are without foundation or where they indicate the likelihood that they have validity tend to be resolved at an earlier point.[36] Frequently, the cases which go to hearing involve pre-school children.[37] Cases involving pre-school children are especially problematic because children under five or six years of age do not have the linguistic skills, knowledge and capacity for accurate expression to articulate precisely what has happened to them in the way most adults can.
In these cases, as in some cases where older children have allegedly been abused, there may be enough evidence to indicate grounds for serious concern but not enough evidence for the Court to reach a confident finding either that sexual abuse has occurred or that the parent or other caregiver was the perpetrator. Conversely, the Court may not be in a position to make a finding that there has been no sexual abuse.
III BINARY REASONING AND THE UNACCEPTABLE RISK TEST
A Criticisms of the Unacceptable Risk Test
The test laid down by the High Court in M v M in 1988 is that contact should be restricted or denied where it would expose the child to an ‘unacceptable risk’ of abuse. Much has been written about this test. It has been criticised as indeterminate and circular. As Richard Chisholm noted, ‘arguably it breaks down into the following tautology: the court should not order ... access where the risk of abuse is such that the court should not order ... access.’[38]
Perhaps the most serious criticism is that the test is fundamentally inconsistent with the fundamental premise of all adjudication — that legal determinations ought to be made on the basis of facts, not suspicions.[39] To some extent, of course, this is a false dichotomy. In the context of the evidence available to support an allegation of child sexual abuse in legal proceedings, there is a spectrum between ‘facts’ and ‘suspicions’, rather than a contrast. Even where the abuse has not been proven to the satisfaction of the Court, the Court may well have much more than a mere suspicion.
Nonetheless, the issue is an important one. All decisions of courts ought to be based upon findings of fact, including discretionary decisions. Even where the task of the court is one of assessing risk, which necessarily has elements within it of prediction, the court ought to make its assessment on the basis of a substratum of fact as found by the court.[40] This issue was raised, but not adequately addressed, by the High Court in M v M.[41] In that case, the High Court stated that
in deciding what is in the best interests of a child, the Family Court is frequently called upon to assess and evaluate the likelihood or possibility of events or occurrences which, if they come about, will have a detrimental impact on the child’s welfare.[42]
However, sexual abuse cases involving the application of the unacceptable risk test are different in nature to cases involving residence decisions in which the Court must make an assessment of whether a child will be better off living with the mother or the father. In such cases, the Court must weigh up possibilities and evaluate the future benefits of competing alternatives, but it does so on the basis of findings of fact which provide an evidential basis for predictions about the future. As Fogarty J put it:
While the assessment of future possibilities and risks is a familiar component of the judicial role in cases involving the welfare of children, it differs in many cases of sexual abuse to the extent that the possibilities being examined are not future possibilities whose evaluation derives from a known factual basis, but possibilities which relate to an unestablished series of facts.[43]
This is the central dilemma in applying the unacceptable risk test. There is no such thing in law as an unestablished fact, for if it is unestablished, then in law it does not represent a fact at all. Findings of fact are binary in nature.[44] Either a fact is proven or it is not. An event occurred, or it did not. The abuse happened, or it did not. The parent or step-parent either molested the child or they did not. In law, a finding of ‘not guilty’ is equated with innocence. In traditional modes of legal reasoning, findings of fact, expressed in this stark binary way, are a necessary basis from which legal conclusions about future risk need to be drawn.
The binary nature of legal fact-finding does not readily accommodate a way of thinking which allows an assessment of what might have happened to be the basis of such severe consequences as denying a father contact to his children perhaps for the duration of their childhood.
The approach laid down by the High Court in M v M stands in contrast to the approach taken by the House of Lords in Re H (Minors) (Sexual Abuse: Standard of Proof).[45] In this case, the House of Lords had to interpret s 31(2) of the Children Act 1989 (UK) which allows the court to make a care order, inter alia, if it is satisfied that the child ‘is likely to suffer significant harm’. The Nottinghamshire County Council, which initiated the care proceedings, relied upon statements by the eldest daughter that she had been sexually abused by her stepfather, Mr R, as evidence that the three younger girls in the household were at risk. There was no evidence that these younger children had been molested. Mr R had been charged with sexual assault but acquitted. At the trial of the care proceedings, the judge held that he could not be satisfied according to the high standard of proof that Mr R had sexually abused his stepdaughter, although he considered that it was a real possibility that the eldest daughter’s statements were true. Hence, he declined to make a care order. The Council’s appeal to the Court of Appeal and the House of Lords failed. Lord Nicholls, for the majority of the House of Lords, held that since the Council had not proven, on the balance of probabilities, that Mr R had sexually abused the eldest daughter, there was no evidential basis on which a court could form the view that the younger children were likely to suffer significant harm. An alleged, but unproven ‘fact’ could not be the basis for a conclusion that there is a real possibility that the younger children might be abused in the future.
B Applying the Unacceptable Risk Test
What then is the role of the court when it is actively discouraged from making findings of fact which would indicate that one of the parties has committed a criminal offence, and when in any event the evidence may be sufficiently equivocal to leave the judge with doubts in his or her mind as to whether or not the abuse has occurred? In response to this dilemma, two irreconcilable views have emerged. The first is that the decision of the court must be based upon findings of fact which are relevant to the assessment of risk. Thus, if no positive findings of fact can be made which indicate a risk of abuse, then no order should be made adverse to the party against whom allegations are made on the sole basis of such allegations or reports. The second view is that the unacceptable risk test comes down in practice, in many cases, to an assessment of the degree of concern which the evidence evokes. This view gains support from the decision in M v M itself, in which the High Court upheld the decision of a trial judge to deny access based upon his ‘lingering doubts’ that the allegations may be true.[46]
The first view is illustrated by the decision of the trial judge in N and S and the Separate Representative.[47] This case demonstrates the impact of the binary approach to legal fact-finding in the context of assessing an ‘unacceptable risk’ of child sexual abuse. In this case, a very young child made numerous, detailed statements which may have been indicative of sexual abuse. However, the trial judge concluded that the case was not proven on the balance of probabilities, not least because the expert witness appointed to assist the Court[48] was able to offer a plausible alternative explanation for at least some of the child’s statements and had concluded that the child had not been sexually abused. The trial judge, Coleman J, then continued:
If one is not satisfied, as I am not, that the allegations of sexual abuse or any of them have been proved on the balance of probabilities, it would seem logically to follow that an unacceptable risk of sexual abuse would be found to exist, only if there was evidence implicating the (husband) but being insufficient to support a finding on the balance of probabilities that sexual abuse had in fact occurred. In this case there is really not a middle course reasonably open. The evidence either establishes the fact of abuse or it does not. Only by proceeding on a basis similar to that indicated earlier could one realistically find that there was an unacceptable risk. There may be cases which, though not involving a positive finding of sexual abuse do involve a finding of unacceptable risk but this is not such a case.[49]
Thus, in Coleman J’s view, if the judge concludes that the abuse is not proven, then logically he or she ought not to act as if it were proven and consequently conclude that contact should be denied, unless there are other proven facts which, independently of the judge’s conclusions on sexual abuse, lead to a determination that it would be an unacceptable risk to the child’s wellbeing to allow contact.
The alternative view is that the Court needs to consider all the factors involved in a qualitative assessment and determine whether in all the circumstances there is an unacceptable risk. This in turn requires the Court to make an assessment about how strong the concerns are within a spectrum of proof from proven to be true according to the Briginshaw standard and proven to be false. For example, in A and A and the Separate Representative[50] the mother had been the victim of a very violent attack but could not remember her assailant. She believed, and had reason to believe, that the father was the assailant. The trial judge declined to go into an assessment of the probable identity of the attacker, seeing it as the Court’s role only to be concerned whether the mother believed the father to be the assailant, and the impact that belief might have on issues concerning the best interests of the children. The Full Court disagreed with this approach. Fogarty, Kay and Brown JJ stated:
Whilst it is correct to say that the Family Court is not a criminal court, and that its primary task is not to determine guilt or innocence, that is entirely different from an approach which declines to examine that issue at all ... The task which his Honour was required to perform was to determine whether the evidence was such as to establish that there would be an unacceptable risk to the children if they were to have contact or supervised contact with the husband ... In reaching a conclusion on that issue, it is necessary for the Court to form some opinion about the connection between the assault and the husband. It would not be necessary in this exercise to reach a positive conclusion that he was the assailant. On the other hand, if the Court reached a comfortable conclusion that the husband was not the assailant, that would be likely to have a profound effect upon the approach to the question of contact. In cases of this sort often it is not possible for the Court to form a positive view at one end or the other end of this scale of persuasion and it is not necessary for it to do so.[51]
This translates into an evaluation of the possibility or probability that the father was the assailant without requiring the Court to be satisfied according to any standard of proof or reasonable satisfaction that indeed he was the assailant. The focus on ‘assessing risk’ is an attempt to change the question. However, ultimately it cannot do so if the Court is reliant only on evaluating the strength of the evidence in relation to a past event without having any broader inquiry into risk factors which might provide further evidence of concerns about the person’s suitability to have continuing parental responsibility.
The unacceptable risk test on this analysis becomes a test about the extent to which the judge feels comfortable with their acceptance of one body of evidence over another. In no other context are judges called upon to evaluate the chances that something may have happened, rather than deciding whether or not it did happen.
IV THE NEED FOR A NEW APPROACH TO RISK ASSESSMENT
The challenge then for the Family Court in determining cases in which child sexual abuse is alleged is a challenge not only of how to assess the available evidence, such as it is, but also of how to reconcile the task of applying the unacceptable risk test with fundamental legal values and modes of thought. In cases where sexual abuse has not been proven according to the Briginshaw standard, the question becomes how to base an assessment of unacceptable risk on findings of fact when the central ‘fact’ is in dispute and cannot be proven to the requisite standard of proof.
The dilemma may in part be resolved if the unacceptable risk test is seen as a test which requires an assessment of the factors which might indicate risk to the child in the future rather than only as an evaluation of the extent to which there is evidence that sexual abuse may have occurred in the past. The two questions, of course, are not entirely distinct. If the Court is satisfied that no abuse has occurred in the past — or to put the question differently, it is not satisfied that there are grounds for serious concern that the child has been abused in the past — then it is most unlikely that there could be a finding that there is an unacceptable risk in the future.
In Re H (Minors) (Sexual Abuse: Standard of Proof), Lord Nicholls, for the majority in the House of Lords, identified a number of factors which might raise the possibility of future harm to the child even where there has been no proven abuse in relation to that child.[52] These include: the history of members of a family; the state of relationships within a family; proposed changes within the membership of a family; parental attitudes; omissions which might not reasonably have been expected; actual physical assaults, threats, abnormal behaviour by a child; unsatisfactory responses to complaints or allegations; and facts which may seem minor or trivial if considered in isolation, but which, when taken together, may satisfy the court of the likelihood of future harm.
This precise list of factors given by Lord Nicholls is perhaps less important than the fact that he sought to identify risk factors on which evidence might be adduced and accepted, and that he saw the assessment of risk as being based upon the cumulative picture. This requires a quite different mode of assessing the available evidence from the mode of thought involved in determining whether an incident of sexual abuse has already occurred. Factors which taken in isolation might not give rise to significant concerns about a child’s wellbeing, may cumulatively build up to a picture which indicates that the child might be seriously at risk of harm.
Risk assessment is a familiar feature of child protection work. Risk assessment is necessary in order to prioritise responses to notifications of abuse and neglect at a time when public awareness of the issue of child abuse is considerable, and willingness to notify is quite high. It is also necessary in case planning, in making such important decisions as to whether to remove a child from the home pending an application for a care order, and in determining whether an application for a care order should be made placing the child in the care of the state. Consequently, considerable research has gone into identifying risk factors for child abuse and neglect and in devising models for prediction.[53]
In assessing the risk of sexual abuse in a given case, there may well be evidence available to the Family Court which, while not probative that abuse has occurred, may indicate cumulatively that there is a serious risk that the child may be abused in the future, whether or not he or she has been abused in the past. There is now extensive literature on sex offending against children both within the family and outside of it. There is also a growing literature on risk factors. Social science evidence of this kind may assist the Court in making appropriate requests to expert assessors to report on such risk factors, and may assist judges in understanding the significance of the evidence which is being presented. One may see and hear and be told much, without realising the significance of those observations and that information.
A major problem for judges in utilising social science information to come to a conclusion about the risk of harm is that, according to traditional views of the matter, the Court is dependent on such evidence being adduced by counsel, although there are exceptions such as the doctrine of judicial notice.[54] Even if judges are aware of the social science research, this does not mean that relevant evidence which would allow them to apply such knowledge will be presented to the Court. However, the Full Court has indicated that judges may take note in a general way of the findings of social science research as background information.[55] This is an important step in the right direction. Indeed, many judges do make reference to such research.[56] It is submitted that the Full Court should go much further and encourage judges to become well acquainted with social scientific research relevant to the decisions they have to make, and to refer to it in decisions, so that their decisions are founded upon evidence — not evidence presented to the Court necessarily (that may depend on the knowledge and diligence of counsel), but evidence of what is known about children’s wellbeing generally. The Family Court is a specialist court exercising a discretionary jurisdiction in which legal rules have a very limited role to play either in the process or the outcome of adjudicating cases involving children. If judges are not encouraged to read and cite the relevant social science literature, whether or not it is presented to them in evidence, then the dangerous alternative is that decisions will be based upon incorrect assumptions about the welfare of children, anecdotal evidence about what promotes their welfare, and beliefs about children which have no basis in scientific research. The traditional ideas, derived from the adversarial system, that judges should base their decisions only on the evidence presented in court, needs to be rethought in light of the best interests principle[57] which makes children’s welfare the paramount consideration, not the interests of the adult litigants.
V A FOUR STAGE PROCESS FOR ASSESSING RISK
One of the striking features of the reported cases on child sexual abuse in the Family Court since M v M[58] is that no attempt has been made by the Full Court to lay down guidelines on how to approach the exercise of discretion or factors which it should consider relevant in coming to its conclusion. This contrasts with the very clear guidance on structuring the assessment of applications for property interests under s 79 of the Family Law Act 1975 (Cth)[59] and, to a lesser extent, the guidance given by the Full Court in relation to residence and contact orders generally.[60]
It is suggested that, as a way of structuring the Court’s assessment of all the relevant factors which may fall to be considered in a case of alleged sexual abuse, the Court should examine four distinct questions as part of a cumulative assessment of the magnitude of the risk:
- What evidence is there before the Court which gives rise to serious concerns about the child’s welfare if contact is allowed with the alleged perpetrator or a residence order is made in his or her favour?
- What other evidence is there before the Court which might lead the Vourt to conclude that there is a heightened risk of abuse if contact is allowed with the alleged perpetrator?
- Are there any factors in the case, or options available to the Court, which might operate to reduce the magnitude of the risk to the child in the future?
- Are there other factors relevant to the best interests of the child which might affect the ultimate decision of the court?
The advantage of this four stage approach is that it clearly differentiates the issue of whether the child has been abused by the alleged perpetrator in the past from the question of whether there is a serious risk, in the light of that history, that the child might be sexually abused in the future. The evidence which might be adduced in answer to these four questions necessarily needs to be assessed cumulatively rather than in isolation. Thus, the presenting of evidence of past sexual abuse remains highly relevant to an evaluation of the other factors even where, on the balance of probabilities, the Court is not satisfied that the parent or step-parent has abused the child and, even where, as in cases such as N and S and the Separate Representative, there are plausible explanations for the child’s statements or behaviour other than that sexual abuse has occurred.
A Evidence of Past
Abuse Giving Rise to Serious
Concerns About the Child’s
Welfare
The first question is to assess the evidence of alleged past abuse to determine how seriously the Court ought to be concerned about the possibility of abuse. In this assessment of the strength of the evidence of past abuse which falls short of the Briginshaw standard, two questions might, in some cases, be conflated. The first is whether there is convincing proof that the alleged perpetrator has sexually abused the child and the second is whether the child has been sexually abused.
In my submission a finding in relation to this second issue does not need the same level of convincing proof as the Briginshaw standard requires for the first question, since a finding that the child has been sexually abused does not involve any grave findings of misconduct in relation to any other person. It is understandable that the High Court should have discouraged the Family Court from making positive findings against one of the parties in relation to child sexual abuse since the function of declaring guilt or innocence in relation to the committal of criminal offences is that of the criminal courts.[61] This requires a particular standard of proof, to which the Briginshaw standard, in this context, may well approximate.[62] But it also requires a particular process. It is fundamental to the defendant’s procedural rights in a criminal matter that the complainant’s evidence is able to be tested through cross-examination. In Family Court cases, children are generally not called to give evidence.
The same concerns do not necessarily arise in making a finding as to whether or not a child has been sexually abused. Such a finding leaves open the identity of the perpetrator. It is not an uncommon defence for the alleged perpetrator to concede that the child may have been abused by someone, but to deny that he or she is the perpetrator.[63]
Determining whether or not the child has been sexually abused is nonetheless of great importance to the assessment of risk because it provides an evidential foundation upon which the Court must assess the risk that the perpetrator is someone who is seeking a residence or contact order. If there is evidence on which the Court may confidently base a finding that a child has been sexually abused, and the child’s statements appear to indicate that the perpetrator was a parent or step-parent, then the Court is in the position of having to weigh up the likelihood that the perpetrator is indeed that parent against the likelihood that it is another unknown person. This narrows the focus considerably and allows the parties to make submissions, or to adduce evidence, which goes to the question of whether someone else could have abused the child and why that possibility should be preferred to the possibility that the child’s own identification of the perpetrator is reliable.[64]
Another way of expressing this is that, where the Court is in a position to make a finding that the child has been sexually abused by someone and the child is indicating that this is the parent, a prima facie case is set up which that parent needs to answer. In this situation, the failure of the parent to satisfy the Court that the legitimate concerns of the other parent for the safety and wellbeing of the child are unfounded would lead to a conclusion that there is an unacceptable risk. This is not the same as saying that the parent has sexually abused the child. Even concluding that the child has been sexually abused and has identified the parent as the perpetrator might not be sufficient to satisfy the Court if the allegations are not detailed with sufficient specificity.
B Evidence of Heightened Risk of Abuse
There are numerous cases in which the evidence as to whether the child was abused by the parent or other caregiver is unclear, but where evidence of other behaviour by the alleged perpetrator may provide grounds for serious concern about ongoing risk. Many of these factors may have been adduced in evidence to buttress allegations that the child who is the subject of the proceedings has been sexually abused. However, it is important to identify the independent role these risk factors might play in determining whether there is an unacceptable risk. In most cases they will not be of great probative value in determining whether abuse has occurred. However, at least some of these warning signs ought to raise concerns about a future risk of sexual abuse. Furthermore, they raise issues about the parenting of the person against whom allegations have been made. Too often it appears that, in the adversarial contest, the judge is presented with a choice on the evidence between two stark alternatives. The parent against whom the allegations have been made is either a monster or a devoted and caring parent. Often the truth is more complex than the adversarial alternatives represent.
What then are the factors which might lead the Court to conclude that there is a risk that the child will be abused in the care of the perpetrator, whether or not abuse has occurred in the past? There are three categories of fact which might be considered in this process of risk assessment:
- Behaviour of the alleged perpetrator towards another child which might indicate that the child who is the subject of the proceedings is at risk;
- Behaviour and attitudes of the alleged perpetrator of the abuse towards the child who is the subject of the proceedings which fall short of sexual abuse but which are warning signs;
- Characteristics of the alleged perpetrator which might indicate statistically a greater risk of sexual abuse occurring than in the general population.
1 Behaviour of the Alleged Perpetrator Towards Another Child
The first instance where there would be a very serious risk to a child if contact is allowed is where there is strong evidence to indicate that the applicant for a residence or contact order has seriously abused another child or has behaved in a manner which raises serious concern about the possibility that they might sexually abuse children. For example, in the 1987 case of M v M,[65] there was an appeal by the husband against an order providing that he should only be allowed supervised daytime access to his five year old daughter. Evidence was given in the case by the 17 year old stepdaughter who alleged incidents of molestation by the husband over a three year period. The trial judge accepted this evidence as substantially true and, on this basis, held that there was a risk that he might abuse his own daughter. The husband’s appeal against this decision was dismissed. The Full Court criticised the trial judge’s decision to make a finding of fact that the allegations were true since it was not necessary for the judge to have made such a finding.[66] Nevertheless, he was entitled to rely on the evidence of the older stepdaughter in making a finding that there was a risk to the child which warranted restriction on the father’s access.[67]
The evidence which might be considered under this category need not amount to sexual abuse. For example, in WK and SR[68] important evidence was given by the father’s oldest stepdaughter, AR, in a case concerning residence and contact orders in relation to two younger children, whom the mother alleged had been sexually assaulted by the father. The evidence of AR was that, on a number of occasions, the father had offered her money or marijuana if she would allow him to touch her breast or kiss her. At the time that the father made these offers, AR was about 15 years old.
The trial judge relied upon AR’s evidence to support the finding of abuse in relation to the two younger children. The Full Court overturned this decision and ordered a new trial, finding an appealable error in the fact that the trial judge had relied on AR’s evidence in making a positive finding of abuse of the younger children. Her evidence did not form part of a chain of events, or logical series of actions, which would constitute a part of the res gestae of the father’s alleged abuse of the two younger children. The only basis on which this evidence could be found to be ‘significant’ in respect of a finding of abuse towards the younger children would be if the Court were persuaded that it established that the father was of such a character, or had such a history of actions, that it was likely that he had abused the two younger children. The evidence did not display an underlying unity with the evidence pointing to the abuse of the younger children nor any similar features. Nonetheless, the Court went on to state that such evidence may be relevant and probative in relation to the question of an unacceptable risk of abuse.
Where abuse of another child is proven, there is a sufficiently strong risk that the perpetrator may offend again to justify the denial or restriction of contact. The greatest risk of reoffending is with perpetrators who have a fixated sexual interest in children. One well-known typology is to distinguish between fixated and regressed offenders.[69] In this typology, fixated offenders are said to be those who have a primary sexual attraction towards prepubescent children, while regressed offenders have a primary sexual orientation towards adults, but also abuse children episodically, particularly when certain situational factors are present, such as marital discord or other stressors. This clear division into two categories of offenders is falling into disuse now, since research has indicated that child molesters are distributed on a continuum between these two poles, rather than necessarily being able to be classified in one way or another.[70] Some offenders, indeed, are sexually abusive towards both children and adults, so there are problems in identifying them as having a clear sexual orientation to one group or the other.[71]
Nonetheless, offenders who have a primary paedophilic sexual orientation are the ones most likely to reoffend and present the gravest risk to the safety of other children in their care,[72] since they generally describe their own orientation as a compulsive one.[73] It is not clear what proportion of incest offenders have such a primary or fixated orientation towards children. In a review of various studies conducted on this issue in the 1970s and 1980s, Williams and Finkelhor estimated that between a fifth and a third of incest offenders showed signs of general sexual arousal towards children.[74]
If there was a clearer dichotomy between fixated and regressed offenders, it might be possible to say that a ‘regressed offender’ who has addressed the circumstances of his offending, could be an ‘acceptable risk’. The research evidence suggests, however, that such a conclusion ought only to be reached with great caution and after expert assessment. Indeed, in the absence of successful completion of a reputable treatment program,[75] such a risk might well not be justified.
2 Behavioural Indicators Consistent with ‘Grooming’
There is now a considerable body of research literature on sex offending against children and one of the important findings which has emerged is an understanding of the process of victimisation for many children who are sexually abused. Information has come from both offenders[76] and victims,[77] and both studied together.[78] Understanding this process of victimisation helps in some cases to explain how it can be that a child cooperates in or acquiesces to abusive sexual activity and how it could be that the mother remains unaware of the abuse — at least for some time.
The sexual abuse of children is usually carefully planned and stage-managed. In the process of grooming, the perpetrator creates the conditions which will allow them to abuse the child while remaining undetected by others, and the child is prepared gradually for the time when the offender first engages in sexual molestation. It is a process which has been acknowledged frequently by offenders in treatment programmes, and it may take weeks or even months, although there is some evidence that incest offenders may progress more quickly towards serious sexual contact than extra-familial offenders.[79]
The offender may groom the child through a variety of means. One means of grooming is by forming an alliance with the child. Typically, a father or stepfather will show special attention to a particular child, giving them treats and, in subtle ways, isolating them from their mother. One offender acknowledged in interviews with researchers that he went along with anything his stepdaughter wanted. Another said he bought his daughter presents, let her stay at friends’ houses, and showed her favours which he did not show to any other of the children.[80] In this way, the offender forms a particular bond with the child. The offender develops a pattern of spending a lot of time alone with the child. He or she may well treat the child emotionally like an adult friend, sharing intimate details about their sex life and adult relationships.[81] The child thus, quite inappropriately, becomes the offender’s confidant. In Budin and Johnson’s study of incarcerated offenders, both incest offenders and non-incest offenders said that the chief way in which they won their child’s trust was by being the child’s friend.[82] This has different meanings depending on whether one is a parent. The parental role requires appropriate parent-child differentiation. It is this differentiation which is frequently eroded in the grooming process as the perpetrator and child form a toxic bond founded upon the shared secret of a hidden activity.
This grooming may occur in many different ways. One way is through the gradual sexualisation of the relationship. For example, Elliott, Browne and Kilcoyne, in interviews with 91 child sex abuse offenders, found that the majority of offenders carefully tested the child’s reaction to sex by bringing up sexual matters or having sexual materials around, sexualised talking, and by subtly increasing sexual touching.[83] Conte, Wolf and Smith’s interviews with 20 adult sexual offenders revealed some similar deliberate strategies.[84] Asked to write a manual on how to sexually abuse a child, for example, one offender responded:
[Find] some way to get a child living with you. If you have a repertoire of jokes that move from risky to pornographic, have porno magazines lying around. Talk about sex. Watch the kids’ reactions. Stick your head in their bedrooms while they are in their bedclothes. Act like it’s a natural thing. Be sympathetic. Try a lot of compliments. Have accidental contact with their breasts [sic]. [85]
Children who have been sexually abused indicate similar things which, in retrospect, they identify as warning signs. In a survey of 23 children, for example, Berliner and Conte found that there were numerous ways in which, before the abuse began, the perpetrator was thinking of them in a sexual way. Verbal clues included statements, such as telling the child that they had beautiful legs or that they looked sexy, or talking about and showing the child pornographic pictures. The parent might also talk to the child about their own sexual activities at an age when this is inappropriate.[87]
Another form of grooming is that the sexual abuse occurs through a gradual process in which the offender moves from innocent to sexual touching. At times, such as bath times when it is natural for the child to be unclothed or scantily clothed, the perpetrator may begin getting the child used to the touching of the child’s genitals, perhaps by a special washing routine or by playing games which involve ‘accidental’ contact with sexual organs. The perpetrator may be partially undressed themselves as they splash water together, or they may get into the bathtub with the child. Similar kinds of activities might occur at the child’s bedtime or on occasions when the child gets into the bed of the parent. Gradually, over a period of time, the sexual contact becomes more explicit and invasive, until it clearly crosses the line between appropriate and inappropriate contact. Because the grooming occurs gradually, the child may not realise that the boundary line of appropriate behaviour has been crossed, even though the sexual contact may make them feel very uncomfortable. By blurring the boundary lines between normal and abnormal parent–child interactions, the parent causes the child to be confused about what is appropriate behaviour and what is not, and the extent to which the child ought to be able to have a realm of bodily privacy which the parent should not invade. Phelan, in her interviews with 14 natural fathers and 26 stepfathers who had abused their daughters, found that most fathers initiated the sexual activity using non-verbal means as an extension of some routine family activity, such as entering the child’s room at night to check on her, watching TV, and cuddling or other innocent contact. When Phelan interviewed the daughters, she found that typically they reacted with confusion and disbelief to what was happening and this hindered their ability to resist or to tell someone.[88]
Not all offenders groom the children. Some use force or a variety of threats. Others molest the children when they are asleep or when the offender thinks the child is asleep.[89] There is no one pattern of child molestation. However, since the key for the offender is that the child will not tell, grooming provides a more subtle means of achieving compliance than coercion or threats.
It is easier, of course, to identify grooming processes in hindsight than in the forensic context when at best judges may see through a glass darkly what might have been occurring in the household. Nonetheless, there are indicators suggestive of grooming which place the evidence that the child may have been sexually abused into a broader context. Three aspects of the parent–child relationship may indicate grooming or, if they do not, reveal aspects of the parent–child relationship which are matters for concern, irrespective of whether they point to a future risk of abuse. These are:
- a poor sense of boundaries;
- the erosion of the parent–child differential;
- isolation.
(a) A Poor Sense of Boundaries
First, there may be evidence that the parent figure has a poor sense of boundaries. Children who have been sexually abused sometimes, with the benefit of hindsight, describe behaviours which break down the normal and appropriate boundaries between parent and child. Such indicators may include ‘accidentally on purpose’ touching the child’s private parts or coming into the bedroom or bathroom when the child is undressed at an age when the child has a sufficient sense of modesty that this is not appropriate, or ‘accidentally on purpose’ showing the parent’s naked body to the child, regularly coming into the bedroom at night, or wanting to inspect the child’s private parts. Children’s descriptions of such behaviours which sexualise the parent–child relationship accord with offenders’ descriptions of deliberate strategies used to accustom the child to activities of a sexual nature as a normal or accepted part of what the parent and child do, or talk about together.
Of course, taken in isolation, none of these behaviours would justify the denial of contact, and most could have quite innocent explanations. Parents often have an entirely unfounded fear that if they touch a child’s private parts over a certain age or come into their bedroom while they are undressing that this could be wrongly construed. However, in assessing the level of risk it is the cumulative picture which needs to be considered. If courts take an atomistic approach to evidence as occurs at common law, and each action or behaviour which does not directly go to the proof of the offence is strictly scrutinised for its probative value, independently of the others, then it will be difficult for courts to gain the full picture. This is where the application of evidential rules for the assessment of risk must necessarily be different from those rules which apply in determining whether a person has committed a grave offence according to the civil standard as interpreted in Briginshaw.
There are also significant problems in interpreting this evidence. The court is in no position to work out whether individual incidents were accidental or deliberate. Cumulatively, however, the court may build up a picture of a poor sense of boundaries which is itself a relevant factor in assessing risk.
(b) The Erosion of the Parent–Child Differential
Secondly, there may be indicators of the erosion of the parent–child differential. There may, for example, be evidence that the father against whom allegations of abuse have been made does not really treat the child like a child but has formed another kind of emotional bond with them. In a normal and healthy family, the relationships take the form of a strong parent–parent dyad and a clear parent–child differential. The relationship between parent and child involves a different kind of love than that between husband and wife.
In a family where there is sexual abuse, other patterns may be evident. As part of the process of grooming the child, a father or stepfather may develop a different kind of intimacy with the child. In this process, the whole relationship between father and child becomes distorted as the father treats the child in a similar way to a lover, forming an emotional parent–child dyad rather than an appropriate parent–child differential.
Where there is evidence of the erosion of the parent–child differential it indicates, at its most benign, a lack of awareness of appropriate and inappropriate relationships with children. It may also support other evidence which may indicate grooming of the child.
(c) Isolation
A third warning sign of sexual abuse is isolation, which is frequently associated in the research literature, with sexual abuse. For example, in an Australian study, Fleming, Mullen and Bammer found that social isolation was one of three factors associated with sexual abuse before the age of 12.[90] Social isolation was understood to mean not doing well socially and having few friends at school, and being dissatisfied with one’s social life as a teenager. Isolation may represent three different possibilities. First, it may be merely an environmental factor which increases the risk that sexual abuse will occur.[91] Secondly, it may be a characteristic of the child’s situation which makes him or her more vulnerable to abuse, since offenders report targeting lonely and vulnerable children.[92] Thirdly, the isolation may be deliberately created by the perpetrator of the abuse. It is common in households, where sexual abuse is occurring, that the child is not allowed to go out to social activities or their friends are rarely allowed to come and play. Some sexually abusive fathers are domineering fathers who keep the child isolated from other supports in order to reduce the risk that the secret will be disclosed. It is this third form of isolation with which courts ought to be most concerned since it may be a risk factor which is observable in the behaviour of an alleged perpetrator rather than being a risk factor generally as a matter of statistical probability.
Another aspect of this isolation is that sometimes the child is alienated from the mother and other siblings. This can occur in many different ways. It is not uncommon, for example, for the father to claim that his wife’s sexual unresponsiveness is a justification for the sexual relationship with the daughter, and the daughter may come to resent the mother for this. Another false message is that the mother knows what is happening and does not care, so there is no point turning to her for comfort.[93] Alienation may also occur as a result of the special attention shown to the child. Siblings may be alienated because of the special favours and privileges that the target child has received. Child victims may also be alienated from their friends because the perpetrator dominates the child’s life and discourages normal interactions with peers.[94]
Evidence of isolating behaviour by the perpetrator is not, on its own, an indicator of risk of abuse. Taken together with other evidence, however, it is part of the picture which the court will need to assess.
3 Characteristics Associated With Statistically Greater Risk
A third fact which might be examined in any assessment of risk is whether the alleged perpetrator has characteristics which might indicate a greater statistical risk of sexual abuse occurring than in the general population. Such evidence ought to be treated with the greatest of caution in determining whether there is an unacceptable risk of abuse. The fact that, in a given case, there is a constellation of risk factors related to the perpetrator’s characteristics could not possibly justify such a grave outcome as the denial or severe restriction of contact between a parent and a child. Being a member of a class which has a tendency to do x or y does not mean that the individual will do x or y.
A further reason for great caution about the use of perpetrator characteristics is that there is no one profile of an incest offender.[95] The most that can be said is that there are certain characteristics which are more commonly found in perpetrators. Williams and Finkelhor summarise the evidence as follows:
[W]hile it is not possible to sketch a single profile of the incestuous father, there are characteristics that seem relatively common. Many incestuous fathers appear to be passive, dependent, isolated, somewhat paranoid, and lacking a core masculine identification. Many have been maltreated in their families of origin and report rejection, particularly by their fathers. Many have poor marriages and a low level of satisfaction and arousal with adults. Many have difficulty feeling empathy, particularly with their children, and they report being uninvolved in child caretaking. However, here again, few of these characteristics apply to a majority of offenders. Moreover, the studies have not indicated whether or how these characteristics tend to coalesce. There are also many offenders who do not have these characteristics.[96]
In a similar vein, Smith and Saunders warn:
Attempts to determine if a particular man is likely or not to be an incest offender by using personality profiles to classify him are contradicted by the available empirical evidence, unwarranted clinically, and represent unethical practice. [97]
Nonetheless, used properly, an understanding of the research on perpetrator characteristics may be of assistance as background information which can be taken into account in forming the cumulative picture of evidence of risk where there is other evidence which points to the very strong possibility that the father or other caregiver has already abused the child. In assessing the risk that a mother may abuse her daughter, for example, where the evidence is quite unclear, it is not an irrelevant factor to note that very few women sexually abuse their children and that the abuse is more likely to be of boys than girls.[98]
Another fact which is well documented in the research literature is that stepfathers are statistically much more likely to abuse their children than biological fathers. More fathers abuse their daughters than stepfathers, but fewer children grow up with a stepfather. In Diana Russell’s landmark study of 930 women in San Francisco, in which in-depth interviews were conducted, one in six girls who had grown up with a stepfather were sexually abused by him. In contrast, one in forty girls were abused by their natural father.[99] Similar results have emerged from other studies.[100]
Williams and Finkelhor have sought to examine why being a biological parent, as opposed to a step-parent, may be an inhibitor of sexual abuse. In a study of 116 incestuous biological fathers, compared with a control group of non-abusive fathers, they found that the incestuous fathers who were not greatly involved in the care of their daughters in the first few years of life were much more likely to abuse them sexually.[101] More significant than bodily contact such as changing nappies, bathing and dressing, was reading the child stories and watching her play. Many stepfathers, who often come into a family later in the child’s life, will not have been present during these formative years.
C Factors or Options Which Reduce the Magnitude of Risk in the Future
The third stage is to ask whether there are any factors which might reduce the risk? Typically, in this context, the focus is on whether supervised contact is or is not appropriate, and whether it will operate to reduce the risk sufficiently.[102] There is an ongoing debate about the utility of supervised contact, especially as a long-term arrangement.[103]
However, there are also other factors to consider. One is the effect of the court proceedings on the likelihood of future abuse. Child sexual abuse is a crime that thrives in the darkness of childhood silence. Once detected, the crime is much harder to repeat. It may well be that, in many cases where the concerns about sexual abuse are in fact valid, the child’s disclosure of the abuse and the subsequent court case will act as a powerful deterrent to further abuse of the child. It would be dangerous for a sex offender to continue the abuse of a child where the abuse has been disclosed and where the child has been made aware by the subsequent interviews with police, social workers and therapists that the offender’s behaviour is treated with enormous concern.
Finkelhor has conceptualised the process of offending against children in terms of four preconditions for the occurrence of sexual abuse.[104] Finkelhor suggests that, in order for an abusive act against a child to take place, the perpetrator needs to have overcome four ‘hurdles’: motivation, internal inhibitors, external inhibitors and the resistance of the victim. The offender has to be motivated by a deviant sexual attraction to children, be prepared to overcome their conscience telling them that this behaviour is unacceptable (the internal inhibitor), create the environment which will allow sexual abuse to take place (for example, isolating the mother who would otherwise act as the external inhibitor) and, finally, overcome the child’s resistance to an abusive act.
The fact that the secret is out in the open may well act as an inhibitor at a number of levels. It may create an internal inhibitor, as the perpetrator fears the consequences of a further disclosure. The other parent’s vigilance after contact visits might also act as a deterrent. The likelihood of suspicion and discovery of abuse is many times greater in the case of further abuse than when the original abuse occurred. The child’s resistance may also act as an inhibitor, particularly given that the child will have undoubtedly become aware that the perpetrator’s previous behaviour was wrong and is regarded as serious by the other parent and many other adults.
The child might also be taught more about how to protect himself or herself from abuse through the many educational programmes which exist for children to encourage them to feel safe and to recognise the warning signs indicating when they may not be safe.
It is not being suggested here that a risk should be taken with the safety of the child only on the basis that the secret is now out, and that the child may be taught protective behaviours. Where abuse has occurred, there are many other factors to consider, in particular, the therapeutic issues for the child who may need a significant amount of time away from the perpetrator in order to heal.[105] However, the Family Court is often confronted by cases in which the evidence is far from clear and in which the Court cannot confidently make a finding either that abuse has occurred or that it has not. In cases of grave doubt such as this, where the consequences of denying contact between a father and a child may well be as serious as the risk of harm from sexual abuse, it is a relevant factor to consider that if the parent has offended against the child, it will be much more difficult for them to do so in the future, with or without supervised contact.
D Other Factors Relevant to the Best Interests of the Child
The issue of sexual abuse is, of course, not the only issue which the court needs to consider. The other parent’s attitude to contact and the impact contact might have on their parenting are also key issues.[106] Furthermore, there may well be a range of other issues which have implications for decision-making about residence and contact orders. Brown et al’s study of child abuse cases in the Family Court indicated that the families had a great many other problems, including psychiatric illness, drug and alcohol abuse, unemployment, and domestic violence.[107] All of these may impact upon the decision which the court must make. As the High Court observed in M v M:
The Family Court’s wide-ranging discretion to decide what is in the child’s best interests cannot be qualified by requiring the Court to try the case as if it were no more than a contest between the parents to be decided solely by reference to the acceptance or rejection of the allegation of sexual abuse on the balance of probabilities. [108]
There may also be issues about domestic violence or other matters which go to the question of contact. There is a significant overlap between sexual abuse, physical abuse and domestic violence.[109] Thus the decision about the risk of sexual abuse will often be part of a broader picture that the court must assess.
VI CONCLUSION
The Family Court, in dealing with questions of sexual abuse, constantly needs the wisdom of Solomon. Its task is made more complex rather than more simple, by the way in which the law and lawyers go about the task of determining whether something is, or is not, a fact.
Approaching the task would be easier if there was a clearer structure for decision-making and the court was presented with all the evidence which might be relevant to the assessment of the risk of sexual abuse, going beyond an examination of past ‘events’. A risk assessment of the kind proposed might lead courts to ask particular questions of experts in their assessments, or other professionals involved with the case.
[*] MA (Oxon), LLM (Illinois); Associate Professor, Faculty of Law, The University of Sydney.
[1] Patrick Parkinson, ‘Custody, Access and Domestic Violence’ (1995) 9 Australian Journal of Family Law 41.
[2] Interview with Sophy Bordow, Research and Evaluation Unit, Family Court, Sydney, personal communication, May 1994.Court (Sydney, May 1994). Other results from this study were published in Sophy Bordow, ‘Defended Custody Cases in the Family Court of Australia: Factors Influencing the Outcome’ (1994) 78 Australian Journal of Family Law 252.
[3] Ibid.
[4] Alistair Nicholson, ‘The Approach of the Family Court of Australia to Child Abuse Matters’ (Paper presented at the 12th International Congress on Child Abuse and Neglect, Auckland, September 1998).
[5] Thea Brown et al, ‘Child Abuse and the Family Court’ (1998) 91 Trends and Issues in Crime and Criminal Justice 1, 2.
[6] Thea Brown et al, Violence in Families: Report No 1 — The Management of Child Abuse Allegations in Custody and Access Disputes Before the Family Court of Australia (1998) 56.
[7] Patrick Parkinson, ‘Child Sexual Abuse Allegations in the Family Court’ (1990) 4 Australian Journal of Family Law 60; John Myers, ‘Allegations of Child Sexual Abuse in Custody and Visitation Litigation: Recommendations for Improved Fact Finding and Child Protection’ (1989–90) 28 Journal of Family Law 1.
[8] M v M (1988) 166 CLR 69, 77; Bieganski and Bieganski [1993] FamCA 143; (1993) 16 Fam LR 353 (‘Bieganski’).
[9] N and S and the Separate Representative [1995] FamCA 139; [1996] FLC 92-655 (Fogarty J dissenting); S v S [1993] NZFLR 657.
[10] Allan De Jong, ‘Impact of Child Sexual Abuse Medical Examinations on the Dependency and Criminal Systems’ (1998) 22 Child Abuse and Neglect 645.
[11] See, eg, the breakdown of the forms of sexual abuse in Goldman and Goldman’s prevalence study of sexual abuse in Australia: Ronald Goldman and Juliette Goldman, ‘The Prevalence and Nature of Child Sexual Abuse in Australia’ (1988) 9 Australian Journal of Sex, Marriage & Family 94. The researchers surveyed nearly 1000 post-secondary social science students in Victoria. Of these, nearly 28 percent of the females indicated that they had experienced some form of sexual abuse before the age of 16. The figure for males was nine percent. Of the girls who were sexually abused, the great majority reported sexual abuse falling short of sexual intercourse. 5.3 percent of the abused group of girls reported that intercourse had taken place. The figure for boys was 15 percent. See also Jillian Fleming, ‘Prevalence of Childhood Sexual Abuse in a Community Sample of Australian Women’ (1997) 166 Medical Journal of Australia 65, who found that 10 percent of women who reported childhood sexual abuse had experienced vaginal or anal penetration.
[12] David Kerns, ‘Triage and Referrals for Child Sexual Abuse Medical Examinations: Which Children are Likely to Have Positive Medical Findings?’ (1998) 22 Child Abuse and Neglect 515.
[13] See generally Jan Bays and David Chadwick, ‘Medical Diagnosis of the Sexually Abused Child’ (1993) 17 Child Abuse and Neglect 91.
[14] See generally Kee McFarlane, ‘Child Sexual Abuse Allegations in Divorce Proceedings’ in Kee McFarlane and Jill Waterman (eds), Sexual Abuse of Young Children: Evaluation and Treatment (1986) 121; G Blush and K Ross, ‘Sexual Allegations in Divorce: The SAID Syndrome’ (1987) 25 Conciliation Courts Review 1; Kathleen Faller, ‘Possible Explanations for Child Sexual Abuse Allegations in Divorce’ (1991) 61 American Journal of Orthopsychiatry 86.
[15] Brown et al, ‘Child Abuse and the Family Court’, above n 5; Marie Hume, ‘Study of Child Sexual Abuse Allegations Within The Family Court Of Australia’ (Paper presented at the 5th Australasian Conference on Child Abuse and Neglect, Melbourne, October 1995). See also Marie Hume, Child Sexual Abuse Allegations in the Family Court (Masters thesis, University of South Australia, 1997).
[16] N Thoennes and P Tjaden, ‘The Extent, Nature and Validity of Sexual Abuse Allegations in Custody/Visitation Disputes’ (1990) 14 Child Abuse and Neglect 151.
[17] E P Benedek and D H Schetky, ‘Allegations of Sexual Abuse in Child Custody and Visitation Disputes’ in E P Benedek and D H Schetky (eds), Emerging Issues in Child Psychiatry and the Law (1985); A Green, ‘True and False Allegations of Sexual Abuse in Child Custody Disputes’ (1986) 25 Journal of American Academy of Child Psychiatry 44; D Jones and A Seig, ‘Child Sexual Abuse Allegations in Custody or Visitation Cases: A Report of 20 Cases’ in B Nicholson and J Buckley (eds), Sexual Abuse Allegations in Custody and Visitation Cases (1988) 22. For a critical review, see Catherine Humphreys, ‘Child Sexual Abuse Allegations in the Context of Divorce: Issues for Mothers’ (1997) 27 British Journal of Social Work 529.
[18] [1995] FamCA 146; [1995] FLC 92-581, 81,766.
[19] K Oates et al, ‘Erroneous Concerns About Sexual Abuse’ in K Oates et al (eds), Child Abuse and Neglect (forthcoming). In a review of other studies, the authors found that two to eight percent of allegations were judged fictitious, with the highest number of fictitious allegations being made by adolescents.
[20] Mary Power, ‘Stages of Separation’ (1996) 10 Australian Journal of Family Law 143.
[22] [1995] FamCA 139; [1996] FLC 92-655, 82,709–10.
[23] M v M (1988) 166 CLR 69, 76.
[24] [1938] HCA 34; (1938) 60 CLR 336 (‘Briginshaw’). See also Evidence Act 1995 (Cth) s 140.
[25] Briginshaw [1938] HCA 34; (1938) 60 CLR 336, 362.
[26] WK v SR [1997] FamCA 57; [1997] FLC 92-787, 84,694.
[27] See, eg, ibid, in which the Full Court of the Family Court held that the trial judge had been wrong to rely on the evidence of an older stepchild to support a finding of sexual abuse of two younger children, discussed below at pp 15-16.
[28] For discussion of some of these issues in the context of the criminal law, see Andrew Palmer, ‘Child Sexual Abuse Prosecutions and the Presentation of the Child’s Story’ [1997] MonashULawRw 10; (1997) 23 Monash University Law Review 171.
[29] This is also the case if a cross-vested claim is made for damages for domestic violence: Kennon and Kennon [1997] FamCA 27; [1997] FLC 92-757.
[30] See, eg, the problem that the complainant had in recalling specific incidents of sexual abuse in S v The Queen [1989] HCA 66; (1989) 168 CLR 266.
[31] See, eg, M v H [1996] FLC 92-695; Bieganski [1993] FamCA 143; (1993) 16 Fam LR 353.
[32] In M v M (1988) 166 CLR 69, 76, the court stated that:
[T]he resolution of an allegation of sexual abuse against a parent is subservient and ancillary to the court’s determination of what is in the best interests of the child. The Family Court’s consideration of the paramount issue which it is enjoined to decide cannot be diverted by the supposed need to arrive at a definitive conclusion on the allegation of sexual abuse. The Family Court’s wide-ranging discretion to decide what is in the child’s best interests cannot be qualified by requiring the court to try the case as if it were no more than a contest between the parents to be decided solely by reference to the acceptance or rejection of the allegation of sexual abuse on the balance of probabilities.
[33] Ibid 77.
[34] Parkinson, above n 7, 80.
[35] [1996] FamCA 86; [1996] FLC 92-697, 83,334.
[36] Ibid.
[37] In the study by Brown et al, the average age of the child in the Melbourne cases where there was only one child in the family, was five. Where there was more than one child, the average age was six. In Canberra, the average age of the children was higher: Violence in Families: Report No 1, above n 6, 50.
[38] Richard Chisholm, ‘Child Sexual Abuse: The High Court Rules on Onus of Proof’ (1989) 3 Australian Journal of Family Law 184, 189.
[39] See, eg, Joanne Roebuck, ‘Allegations of Child Sexual Abuse: Should “Unacceptable Risk” Be the Only Criterion for Refusal of Contact?’ (1996) 3 James Cook University Law Review 116. See also Frank Bates, ‘Evidence, Child Sexual Abuse and the High Court of Australia’ (1990) 39 International and Comparative Law Quarterly 413.
[40] Re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, 89–90 (Lord Nicholls).
[42] Ibid.
[43] N and S and the Separate Representative [1995] FamCA 139; [1996] FLC 92-655, 82,712 (Fogarty J).
[44] For a discussion of this in the context of autopoietic theory, see Niklas Luhmann, A Sociological Theory of Law (Elizabeth King and Martin Albrow trans, 1st ed, 1985) 281–8; Günther Teubner, Law as an Autopoietic System (Anne Bankowska and Ruth Adler trans, 1993); Michael King, ‘The Truth About Autopoiesis’ (1993) 20 Journal of Law and Society 218; Michael King and Christine Piper, How the Law Thinks About Children (1990).
[45] [1996] AC 563.
[46] M v M (1988) 166 CLR 69. See also B v B (1988) 63 ALJR 112.
[47] [1995] FamCA 139; [1996] FLC 92-655.
[48] This witness is known as the Order 30A expert: Family Law Rules O 30A.
[49] N and S and the Separate Representative [1996] FLC 92–655, 82,704 (Coleman J quoted in the appeal judgment of Fogarty J). For another example of the binary approach see the trial judge’s finding in Re C and J that the father ‘has never sexually abused the child’ when he found that the case was not made out that the parent had sexually abused the child: [1996] FamCA 86; [1996] FLC 92-697, 83,336 (Purdy J cited on appeal).
[50] [1998] FamCA 25; [1998] FLC 92-800.
[51] [1996] AC 563 atIbid 84,995–6.
[52] Re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, 591.
[53] See, eg, Virginia Murphy-Berman, ‘A Conceptual Framework for Thinking About Risk Assessment and Case Management in Child Protective Service’ (1994) 18 Child Abuse and Neglect 193; Howard Doueck et al, ‘Decision-Making in Child Protective Services: A Comparison of Selected Risk-Assessment Systems’ (1993) 72 Child Welfare 441.
[54] For a full review of the ways in which social science literature can be used, see G Mullane, ‘Evidence of Social Science Research: Law, Practice and Options in the Family Court of Australia’ (1998) 72 Australian Law Journal 434.
[55] Patsalou and Patsalou [1994] FamCA 118; [1995] FLC 92-580; Mitchell and Mitchell [1995] FamCA 32; [1995] FLC 92-601, 81,997–8. See also B and B; Family Law Reform Act 1995 [1997] FLC 92-755, 84,230.
[56] Mullane, above n 58, 438.
[57] Family Law Act 1975 (Cth) s 65E.
[59] Pastrikos and Pastrikos [1979] FamCA 56; (1980) 6 Fam LR 497; Lee Steere and Lee Steere [1985] FamCA 57; [1985] FLC 91-626; Efthimiadis and Efthimiadis [1993] FamCA 15; [1993] FLC 92-361.
[60] B and B: Family Law Reform Act 1995 [1997] FLC 92-755, 84,219–20; Maday and Maday [1985] FamCA 37; [1985] FLC 91-636, 80,142–3. On guidelines for deciding about interim residence orders, see Cowling and Cowling [1998] FamCA 19; [1998] FLC 92-801.
[61] This view was taken explicitly by the Full Court in B and B [Suspension of Access] [1988] FamCA 46; [1988] FLC 91-957 and M v M [1988] FamCA 47; [1988] FLC 91-958. These are the two cases which went on to be heard by the High Court in B v B (1988) 63 ALJR 112 and M v M (1988)166 CLR 69 respectively.
[62] WK v SR [1997] FamCA 57; [1997] FLC 92-787.
[63] See, eg, the husband’s defence in B v B (1988) 63 ALJR 112. A full summary of the evidence is given in the Full Court judgment: B and B [Suspension of Access] [1988] FamCA 46; [1988] FLC 91-957.
[64] See also the discussion of the New Jersey case, Re DT, 229 NJ Super 509, 552 (1988) in Patrick Parkinson, ‘Unacceptable Risks and the Investigation of Child Sexual Abuse’ (1990) 4 Australian Journal of Family Law 254, 259.
[66] Ibid 76,240 (Evatt CJ); 76,241–2 (Kay J).
[67] See also B v B [Access] [1986] FamCA 52; [1986] FLC 91-758. In that case, the husband had pleaded guilty to a charge of committing an act of indecency on his stepdaughter. This, together with the fact that the two younger children had made statements indicative of sexual interference by the father and that the mother was convinced that the children had been sexually abused, led to a denial of access.
[68] WK v SR [1997] FamCA 57; [1997] FLC 92-787.
[69] N Groth, W Hobson and T Gary, ‘The Child Molester: Clinical Observations’ in J Conte and D Shore (eds), Social Work and Child Sexual Abuse (1982) 129.
[70] Leonore Simon et al, ‘Characteristics of Child Molesters: Implications for the Fixated-Regressed Dichotomy’ (1992) 7 Journal of Interpersonal Violence 211.
[71] J Conte, ‘The Incest Offender: An Overview and Introduction’ in A Horton et al (eds), The Incest Perpetrator: A Family Member No One Wants to Treat (1990) 19.
[72] Many offenders against children have multiple victims, especially those who have a sexual orientation towards boys. Some men offend against children both within the family and outside of it: Gene Abel et al, ‘Self-Reported Sex Crimes of Nonincarcerated Paraphiliacs’ (1987) 2 Journal of Interpersonal Violence 3; M Weinrott and M Saylor, ‘Self-Report of Crimes Committed by Sex Offenders’ (1991) 6 Journal of Interpersonal Violence 286.
[73] As Jenkins, a leading therapist, puts it (Alan Jenkins, Invitations to Responsibility (1990) 196–7):
Many men who experience their offending as somewhat compulsive in nature, describe cyclical patterns of behaviour in terms of an escalating experience of self-intoxication in which they become increasingly pre-occupied with their own fantasies and feelings of excitement, arousal and urgency at the expense of other rational thoughts.
[74] Linda Meyer Williams and David Finkelhor, ‘The Characteristics of Incestuous Fathers’ in W Marshall, D Laws and H Barbaree (eds), Handbook of Sexual Assault: Issues, Theories and Treatment of the Offender (1990) 231, 252.
[75] In NSW, eg, there is a Pre-Trial Diversion of Offenders Scheme which was established by the Pre-Trial Diversion of Offenders Act 1985 (NSW). The scheme is to divert those who plead guilty to intra-familial sexual abuse into an intensive two year program in which counselling is provided to the whole family.
[76] See, eg, Lee Budin and Charles Johnson, ‘Sex Abuse Prevention Programs: Offenders’ Attitudes About Their Efficacy’ (1989) 13 Child Abuse and Neglect 77; Jon Conte, Steven Wolf and Tim Smith, ‘What Sexual Offenders Tell Us About Prevention Strategies’ (1989) 13 Child Abuse and Neglect 293; John Christiansen and Reed Blake, ‘The Grooming Process in Father–Daughter Incest’ in A Horton et al (eds), The Incest Perpetrator (1990) 88; Michele Elliott, Kevin Browne and Jennifer Kilcoyne, ‘Child Sexual Abuse Prevention: What Offenders Tell Us’ (1995) 19 Child Abuse and Neglect 579.
[77] Lucy Berliner and Jon Conte, ‘The Process of Victimisation: The Victims’ Perspective’ (1990) 14 Child Abuse and Neglect 29.
[78] Patricia Phelan, ‘Incest and Its Meaning: The Perspectives of Fathers and Daughters’ (1995) 19 Child Abuse and Neglect 7.
[79] Donald Fischer and Wendy McDonald, ‘Characteristics of Intrafamilial and Extrafamilial Child Sexual Abuse’ (1998) 22 Child Abuse and Neglect 915.
[80] Conte, Wolf and Smith, ‘What Sexual Offenders Tell Us About Prevention Strategies’, above n 76, 297.
[81] Berliner and Conte, ‘The Process of Victimisation’, above n 77.
[82] Budin and Johnson, above n 76, 81.
[83] Elliott, Browne and Kilcoyne, above n 76, 585–6.
[84] Conte, Wolf and Smith, ‘What Sexual Offenders Tell Us About Prevention Strategies’, above n 76.
[85] L Berliner and J Conte, above n 82.
86 Above n 83.Ibid 298.
[87] Berliner and Conte, ‘The Process of Victimisation’, above n 77.
[88] Phelan, above n 78.
[89] See, eg, Phelan’s study in which the offenders and the children gave quite different accounts: ibid. Sometimes the offender thought the daughter was asleep when she was not. See also Jenkins, above n 73.
[90] Jillian Fleming, Paul Mullen and Gabriele Bammer, ‘A Study of Potential Risk Factors for Sexual Abuse in Childhood’ (1997) 21 Child Abuse and Neglect 49. The other two factors were mother’s death and physical abuse. For other research on risk factors, see David Finkelhor, ‘Epidemiological Factors in the Clinical Identification of Child Sexual Abuse’ (1993) 17 Child Abuse and Neglect 67; Jocelyn Brown et al, ‘A Longitudinal Analysis of Risk Factors for Child Maltreatment: Findings of a 17 Year Prospective Study of Officially Recorded and Self-Reported Child Abuse and Neglect’ (1998) 22 Child Abuse and Neglect 1065.
[91] See, eg, Goldman and Goldman’s study of nearly 1000 students which found that living in a country area gave rise to a greater risk of sexual abuse for girls (the figures for boys were too small to make any statistical inferences): Goldman and Goldman, above n 11, 103.
[92] See the offender studies cited in above n 81. See also Sue Boney-McCoy and David Finkelhor, ‘Is Youth Victimisation Related to Trauma Symptoms and Depression After Controlling for Prior Symptoms and Family Relationships? A Longitudinal, Prospective Study’ (1996) 64 Journal of Consulting and Clinical Psychology 1406.
[93] L Laing and A Kamsler, ‘Putting an End to Secrecy: Therapy with Mothers and Children Following Disclosure of Child Sexual Assault’ in M Durrant and C White (eds), Ideas for Therapy with Sexual Abuse (1990) 159.
[94] Christiansen and Blake, above n 81, 90.
[95] Daniel Smith and Benjamin Saunders, ‘Personality Characteristics of Father/Perpetrators and Non-Offending Mothers in Incest Families: Individual and Dyadic Analyses’ (1995) 19 Child Abuse and Neglect 607.
[96] Williams and Finkelhor, ‘The Characteristics of Incestuous Fathers’, above n 79, 249.
[97] Smith and Saunders, above n 94, 615. For a rejection of profile evidence, see Gee J’s judgment in K and K (1991) 15 Fam LR 285.
[98] On female offenders, see Michele Elliott (ed), Female Sexual Abuse of Children: The Ultimate Taboo (1993); David Finkelhor and Diana Russell, ‘Women as Perpetrators: Review of the Evidence’ in David Finkelhor (ed), Child Sexual Abuse: New Theory and Research (1984) ch 11.
[99] Diana Russell, ‘The Prevalence and Seriousness of Incestuous Abuse: Stepfathers vs Biological Fathers’ (1984) 8 Child Abuse and Neglect 15.
[100] Kathleen Faller, ‘Sexual Abuse by Paternal Caretakers: A Comparison of Abusers Who Are Biological Fathers in Intact Families, Stepfathers, and Noncustodial Fathers’ in A Horton et al (eds), The Incest Perpetrator (1990) 65; Lesley Margolin, ‘Child Abuse by Mothers’ Boyfriends: Why the Overrepresentation?’ (1992) 2 Child Abuse and Neglect 541.
[101] Linda Williams and David Finkelhor, ‘Paternal Caregiving and Incest: Test of a Biosocial Model’ (1995) 65 American Journal of Orthopsychiatry 101. The researchers cautioned however that a high level of caregiving is not necessarily an inhibiting factor if the perpetrator is otherwise oriented towards sexual abuse. Indeed, it may disguise grooming.
[102] See, eg, Bieganski [1993] FamCA 143; (1993) 16 Fam LR 353; K v B [1994] FamCA 171; [1994] FLC 92-478; Re C and J [1996] FamCA 86; [1996] FLC 92-697.
[103] See generally Eric Baker, ‘Assessing and Managing Allegations of Child Sexual Abuse: An Australian Perspective’ (1997) 33 Family and Conciliation Courts Review 293.
[104] David Finkelhor, ‘Preconditions to Sexual Abuse’ in David Finkelhor (ed), Child Sexual Abuse: New Theory and Research (1984) ch 5.
[105] Elizabeth Jones and Patrick Parkinson, ‘Child Sexual Abuse, Access and the Wishes of Children’ (1995) 9 International Journal of Law and the Family 54.
[106] A and A and the Separate Representative [1998] FamCA 25; [1998] FLC 92-800; Re Andrew [1996] FamCA 43; [1996] FLC 92-692; Russell and Close (Unreported, Family Court, Williams J, 25 June 1993).
[107] Brown et al, Violence in Families — Report No 1, above n 6, 74–6.
[108] M v M (1988) 166 CLR 69, 76.
[109] See Adam Tomison, ‘Child Abuse and Other Family Violence: Findings from a Case Tracking Study’ (1995) 41 Family Matters 33; Maura O’Keefe, ‘Predictors of Child Abuse in Maritally Violent Families’ (1995) 10 Journal of Interpersonal Violence 3.