Aust J Soc Issues. 2021;00:1–22.
|
1
wileyonlinelibrary.com/journal/ajs4
Received: 3 April 2021
|
Revised: 8 June 2021
|
Accepted: 11 June 2021
DOI: 10.1002/ajs4.171
ORIGINAL RESEARCH ARTICLE
Allegations of child sexual abuse: An empirical
analysis of published judgements from the Family
Court of Australia 2012– 2019
NolaWebb
1
* | Lawrence J.Moloney
2
| Bruce M.Smyth
3
|
Robyn L.Murphy
4
This is an open access article under the terms of the Creative Commons Attribution- NonCommercial- NoDerivs License, which
permits use and distribution in any medium, provided the original work is properly cited, the use is non- commercial and no
modifications or adaptations are made.
© 2021 The Authors. Australian Journal of Social Issues published by John Wiley & Sons Australia, Ltd on behalf of Australian
Social Policy Association
*Retired.
1
Barrister- at- Law, Supreme Court of NSW,
Lismore, NSW, Australia
2
La Trobe University, Melbourne, VIC,
Australia
3
Australian National University, Canberra,
ACT, Australia
4
Technologies Consultant, Melbourne, VIC,
Australia
Correspondence
Bruce M. Smyth, ANU Centre for Social
Research & Methods, Research School of
Social Sciences, RSSS Building, 146 Ellery
Cr, Acton, ACT 2601, Australia.
Email: Bruce.Smyth@anu.edu.au
Abstract
Allegations of child sexual abuse pose agonisingly
difficult issues for families, family law professionals
and the courts. We present data from the population
(N=521) of Family Court of Australia judgements
containing allegations of child sexual abuse published
in the Australasian Legal Information Institute's
Australian database. Our data cover all in- scope
judgements published between mid- 2012 and mid- 2019,
of which 71 dealt with cases that were uncontested.
A further 70 were contested but the allegations were
abandoned before the end of the trial. We classified
the remaining 380 cases as “fully contested”. Of this
group: (a) in 14% of cases, judicial officers expressed
a direct or clearly implied belief that the allegations of
child sexual abuse were true; (b) risk of sexual harm to
a child was found in 12% of judgements; (c) when no
risk of sexual harm was found, judges were more than
twice as likely to regard the allegations as genuine but
mistaken rather than to have been deliberately mis-
leading; (d) just under two- thirds of allegedly unsafe
2
|
WEBB Et al.
1 | INTRODUCTION
Allegations of child sexual abuse made after parental separation raise highly consequential
and generally long- term issues for family members. In an atmosphere frequently plagued by
uncertainty and with high stakes for all, anecdotes about outcomes in litigated cases prompt
claims ranging from judicial objectivity and fairness to excessive judicial subjectivity and sys-
temic bias. In this article, we analyse all published in- scope judgements heard in the Family
Court of Australia over a recent seven- year period. By providing a contemporary empirical
benchmark of court outcomes in this difficult area of decision making, our core aim is to stim-
ulate much- needed discussion about children alleged to be at risk of sexual harm in the care
of a separated parent.
We begin by noting that mainstream definitions of child sexual abuse continue to lack preci-
sion (Mathews & Collin- Vézina, 2019), ranging from the broad (e.g. Butchart Harvey, Mian, &
Fürniss, 2006) to the more detailed and focused (e.g. Quadara, Nagy, Higgins & Seigel, 2015).
We note too that there are cultural and legal variations in the literature with respect to issues
such as what constitutes “childhood” and what defines a “sexual act”. Notwithstanding these
limitations, there is broad agreement in the literature that the key ingredients of child sexual
abuse include children's lack of comprehension and consent alongside a profound betrayal of
the trust placed in the perpetrator who, in the eyes of the child, typically holds a position of
power and “prestige” (Commonwealth of Australia, 2017:12).
1.1 | Impact and frequency
Multiple reviews in the past two decades have demonstrated that child sexual abuse is associ-
ated with a plethora of negative emotional and physical outcomes (MacGinley, Breckenridge
& Mowll, 2019).
1
Important but less extensive research points to factors associated with mod-
erating these impacts (MacGinley et al., 2019).
2
In addition, the broader consequences of child
sexual abuse have been shown to reach far into families and communities, with significant
costs associated with primary and rehabilitative healthcare, education and welfare assistance,
child protection and the justice system (Fang, Derek, Brown, Florence & Mercy, 2012).
Estimates of the extent of child sexual abuse have varied according to a range of factors, in-
cluding the representativeness of data and the child sample's gender ratio, the country in which
the sample was derived and within- country ethnic background of subsamples, how narrowly
or broadly child sexual abuse is defined,
3
how the data were gathered (e.g. face- to- face inter-
views typically yield differing results to other survey methods), the nature and sequencing of
questions (i.e. order effects) and reporter source (e.g. self- report vs carer) (e.g. see Peters, Wyatt
& Finkelhor, 1986; Stoltenborgh, Van Ijzendoorn, Euser & Bakermans- Kranenburg, 2011;
Wyatt & Peters, 1986). There is also persistent evidence that for a variety of complex reasons,
child sexual abuse has been – and almost certainly remains – significantly under- reported by
parents had the time they spent with their child(ren)
increased by the court; and (e) in 17% of judgements,
children's living arrangements were changed to the al-
legedly unsafe parent.
KEYWORDS
child sexual abuse, decision making, family law, social policy,
social welfare
|
3
WEBB Et al.
agencies, carers and other professionals, as well as by the victims themselves (London, Bruck,
Ceci & Shuman, 2005).
4
Considered on a spectrum – from exposure to unwanted touching, through to penetrative
assault before the age of 18 years – a Cochrane meta- analysis of data collected from retro-
spective studies of adults in countries and cultures worldwide (Walsh, Zwi, Woolfenden &
Shlonsky, 2015) reported evidence that 10% to 20% of female children, and 5% to 10% of male
children, have experienced child sexual abuse.
5
Broadly consistent with the above estimates, a
summary statement on child maltreatment published by the World Health Organization has
included among its “key facts” that one in five women (i.e. 20%) and one in thirteen men (~8%)
are reported to have been sexually abused between the ages of 1 and 17years (World Health
Organization, 2020).
Finally, it is now well- documented that childhood sexual abuse is most often perpetrated
by individuals known to the child (e.g. Snyder, 2000).
6
It has also been suggested (e.g. Quadara
et al., 2015) that the most common form of child sexual abuse alleged in post- separation par-
enting disputes – intra- familial abuse (formerly known as incest) – is associated with partic-
ularly severe outcomes. In addition, although – or perhaps because – incest has traditionally
been the subject of long- standing and near- universal taboos (Nakashima & Zakus, 1979), it
was generally assumed to be uncommon until well into the twentieth century.
7
Consistent with
this belief, most paedophiles were thought of as especially deviant, even monstrous, individ-
uals who exploited the vulnerability of children with whom they had little or no ongoing rela-
tionship. Besides its frequency, the shocking truth of child sexual abuse is that perpetrators are
generally known to their victims and present not as monsters, but as ordinary, well- respected
citizens.
1.2 | Legal responses: punishing the perpetrator; protecting the child
When allegations of child sexual abuse are heard in criminal justice systems, the primary
focus is on the past behaviour of the alleged perpetrator. Convictions are uncommon under
the “beyond reasonable doubt” standard of proof (e.g. Cashmore, Parkinson & Taylor, 2018).
8
Benefits to the victim/survivor are likely to be indirect at best. The low conviction rates, along
with the emotional demands and systemic limitations of adversarial processes, also act as a
disincentive with respect to making or following through with allegations in criminal jurisdic-
tions (McLellan, 2017).
9
By contrast, the primary focus of proceedings that fall outside of criminal law is future pro-
tection of the child. Proof of the occurrence of child sexual abuse is judged on the lower stan-
dard of the “balance of probabilities” (see Higgins & Kaspiew, 2008; Kelly & Fehlberg, 2002).
In Australian State, jurisdictions charged with considering allegations of child abuse and ne-
glect, those to whom care has been entrusted (typically a parent, stepparent or foster parent)
are frequently the very persons alleged to be a danger to the child. If a threshold of abuse or
neglect has been demonstrated, a court's task is to consider alternative care possibilities.
In post- separation parenting disputes involving allegations of child sexual abuse, the most
common judicial outcome is that the child remains in the primary care of one of the parents.
In assessing the level of risk to the child, the decision maker must make a call on the extent to
which the allegations are likely to be true, or otherwise genuine but mistaken, or else deliber-
ately misleading. Decision making is especially demanding because one or both of the follow-
ing questions must frequently be addressed: “What, if any, occasional or ongoing relationship
should the child have with (a) a parent who is alleged to be unsafe, or (b) with a parent who
makes deliberately misleading allegations?”.
O’Donohue, Cummings and Wills (2018) provide a recent, detailed review of the literature
on the apparent frequency of false allegations of child sexual abuse. After careful consideration
4
|
WEBB Et al.
of sampling and methodological issues attached to multiple studies, they conclude that a large
majority of allegations of child sexual abuse are likely to be true. Though they found that no
study reported even a sizeable minority of claims that are likely to be false, O’Donohue and his
colleagues acknowledge that allegations in adjudicated settings remain particularly challeng-
ing for decision makers.
This latter observation takes us to the heart of a critical tension between decisions in in-
dividual cases and results derived from research drawn from random samples or population-
based data. On the one hand, individual decisions that rely too readily on aggregate data and
statistical probability raise significant methodological, philosophical and ethical questions.
On the other hand, careful consideration of context and evidence, the lifeblood of legal pro-
cesses, cannot be entirely detached from consideration of established probabilities that an
alleged event might have occurred.
10
When allegations of child sexual abuse are made, judges in Australian family law courts
follow a train of thought articulated by the High Court of Australia in a case known as
M & M.
11
This case directs judges to assess the extent to which a parenting arrangement
poses an “unacceptable risk”. We now know, as noted above, that child sexual abuse in
general and intra- familial sexual abuse in particular are considerably more common than
was previously believed. In the context of parental separation, the best available literature
also suggests that allegations are considerably more likely to be true than not true. A key
question here is: What if any impact should this knowledge have on decision makers in
individual cases?
In assessing “unacceptable risk”, the decision maker's traditional tools are (a) astute
observation of human behaviour and (b) a capacity to assess the credibility, consistency
and logic of the arguments and evidence proffered by protagonists, witnesses, experts
and lawyers. With respect to allegations of child sexual abuse, the tortuous process of
determining “unacceptable risk” outlined in M & M has been the subject of considerable
published commentary (e.g. Chisholm, 1989, 2011a,b; Fogarty, 2006; Parkinson, 1990a,b,
1999, 2014; Young, Dhillon & Groves, 2014). Of some relevance is that the earliest of these
analyses (Chisholm, 1989), expressed concern from the very outset that the High Court's
underpinning logic, might itself be fundamentally flawed.
12
Since then, further attempts
at clarifying the High Court's intentions have been addressed in at least 12subsequent
judgements.
13
The extensive academic and judicial commentary on the judgement in M & M points to the
ongoing struggle to prioritise the safety of children at the same time as ensuring a just result for
parents alleged to have perpetrated child sexual abuse. In addition, however, we would argue
that besides being assessed against the criteria outlined by the High Court, the sum total of
decisions must be assessed against the extent to which they are broadly commensurate with
research- based findings about frequency of occurrence of child sexual abuse as well as the
statistical likelihood that allegations are both genuine and correct.
As noted, the focus of the present research is on key outcomes from 521 FCoA judge-
ments in cases containing allegations of child sexual abuse. The present study follows
the other recent Australian AustLII- based investigation into a broad range of outcomes
arising out of child sexual abuse allegations in 156 parenting disputes (Ferguson, Wright,
Death, Burgess & Malouff, 2018). Of particular relevance to the present study were find-
ings by Ferguson and her colleagues that allegations of child sexual abuse were substan-
tiated by judges in 10% of decisions (n = 16) and unsubstantiated in 73% (n = 113). Of the
113 unsubstantiated cases, 7% (n = 10) were nonetheless suspected to be true, while 62%
(n = 95) were disbelieved. No determination appears to have been made in another 16% of
decisions (n = 25).
|
5
WEBB Et al.
1.3 | Aims and research questions
We sought to document the outcomes of 521 published judgements in which allegations of
child sexual abuse were made in the Family Court of Australia over a seven- year period. In all
such judgements, at least one child was alleged to be at risk of sexual harm while in the care
of the other parent.
Though quantitative in nature, the data raise questions about what happens when judges
wrestle with their primary obligation to protect children in such cases in juxtaposition with
their other obligation to support ongoing meaningful parent– child relationships when it is
appropriate to do so.
The following seven research questions (RQs) guided our analysis:
RQ
1
. How often did judges indicate a belief that allegations of child sexual abuse were true
or likely to be true?
14
RQ
2
. How often was a risk of sexual harm found in contested hearings?
RQ
3
. In cases in which a judge did not find that a risk of sexual harm existed, how often
were the allegations found to be genuine but mistaken or deliberately misleading?
RQ
4
. How was parental responsibility allocated following contested hearings?
RQ
5
. How were parenting arrangements allocated following contested hearings?
RQ
6
. How often were restraining orders or warnings imposed on the parent making sexual
abuse allegations?
RQ
7
. Were any allegations abandoned during a hearing, and what effect – if any – did this
have on any of the above questions?
2 | METHODOLOGY
Court judgements are the definitive data source for exploring legal outcomes to allegations
of child sexual abuse. The research presented below is based on analysis of court judgements
published in the Australasian Legal Information Institute's (AustLII) Australian database of
legal materials. AustLII allows access to the full population
15
of Family Court of Australia
(FCoA) judgements on a wide range of family law topics. Analysis of AustLII data provides an
ethically responsible method of exploring the especially sensitive and complex topic of child
sexual abuse allegations.
2.1 | Selection of cases
Our study sought to analyse Family Court of Australia judgements in a substantial number
of consecutive cases in which there were clearly defined allegations of child sexual abuse.
We aimed for a statistically credible population size of at least 500 contemporary FCoA
judgements.
16
When the search term “sexual*” was applied, 1704 judgements for the period 7June 2012
to 31May 2019 were identified.
17
Preliminary analysis of these judgements suggested that 863
were unrelated to allegations of child sexual abuse – that is, they were not cases in which alle-
gations of child sexual abuse could be identified and were thus out of scope. Of the remaining
841 judgements involving an allegation that a child was at risk of sexual harm in the care of
one parent or the other, a further 320 were excluded because they involved one or more of the
following characteristics:
6
|
WEBB Et al.
1. the allegation(s) were one of a multitude of criticisms of the other parent, serving to
possibly minimise their seriousness and compromise their believability.
2. chaotic parenting for most of a child's life was evident, such as a history of involvement by
child protection authorities, mental health issues, violence, abuse of alcohol or other drugs,
on the part of both parents, sometimes in conjunction with allegations by both parents of
sexual abuse in the care of the other parent;
3. interim hearings in which a judge's belief or non- belief as to the truth of the allegations could
not be confidently determined;
4. interim hearings for which the judgement in the full hearing was also available (in which
case only data from the full hearing were analysed). Exceptions to this rule were cases in
which (a) the hearing was adjudicated by a different judge (in which case both judgements
were considered); or (b) more than 2years had passed since allegations were last heard (the
presumption here was that circumstances were likely to have changed to such a degree that it
was legitimate to consider both judgements).
The final dataset thus comprised 521 Family Court of Australia judgements involving an
allegation of child sexual abuse published in AustLII between 7June 2012 and 31May 2019.
18
These cases consisted of mainly contested hearings and a minority of uncontested hearings
(86% vs 14%); and mainly hearings listed for final trials and a minority of interim hearings
(70% vs 30%). Of the contested hearings, we also noted that in a minority (16%) of cases, the
allegations were not pursued through to the end of the trial. Our reasons for including (a) un-
contested cases; (b) interim and final hearings; and (c) judgements in which allegations of child
sexual abuse were abandoned before the end of the trial were as follows.
2.1.1 | Contested and uncontested cases
Our initial hypothesis was that in uncontested cases, outcomes linked to variables such as un-
acceptable risk would be uniformly predictable. This did not prove to be the case, suggesting
that even when matters were uncontested, judicial officers felt a continued sense of obligation
to address questions of safety and fairness. For this reason, we decided to include uncontested
cases, even though our focus is on hearings that are fully contested – that is, those in which al-
legations of child sexual abuse continued to be prosecuted and opposed throughout the length
of the trial. These cases impose a particularly heavy burden on the judge to weigh the compet-
ing evidence and to make the best determination possible based on an assessment of the truth
or otherwise of the relevant allegations.
2.1.2 | Interim and final hearings
For understandable reasons, judges are reluctant to rule on disputed facts or make final orders
in interim hearings. It might therefore be argued that such cases should not be in our dataset.
However, the judgements in this category not infrequently refer to previous interim cases in
which judicial officers were acutely aware of a need to respond realistically, urgently, and in
the best way they can when serious allegations are presented.
19
For this reason, we included
interim hearings in our contested category of cases wherever there was sufficient material in
the judgement to permit an analysis of judicial thinking. We also included these cases because
from the point of view of the parent making the allegation, the outcome for the child rather
than the process leading to the outcome is the critical issue.
Having distinguished between these interim hearings and final hearings, we examined dif-
ferences in two key outcome variables of “risk of sexual harm” and “belief” in the allegation.
|
7
WEBB Et al.
As the differences between the two types of hearing were negligible, the present article does not
address differences between interim and final hearings.
2.1.3 | Allegations that were abandoned during the course of the hearing
In the majority of the 450 contested cases in our sample (84%, n = 380), the allegations were
pursued for the full length of the hearing – that is, “fully contested”. Less frequently (16%, n
= 70), allegations were abandoned at some point during the hearing. That is, they were either
formally withdrawn or not pursued. These cases were included because inspection of their
contents reveals a complex set of dynamics. They range from cases in which the alleging parent
genuinely recognised that the allegation had been misguided, sometimes apologising for the
distress caused to family members; to cases in which though the allegations were not pursued,
the parent remained concerned that sexual abuse had occurred; to cases in which the evidence
strongly suggested that the motivation behind the allegations had been mainly vindictive.
2.2 | Coding procedure
Twenty- eight variables of interest were chosen by the research team on a largely a priori basis.
Each variable was then coded in each of the 521 judgements. The variables fell into one of five
broad categories: (a) procedural (n = 9 variables); (b) allegations (n = 4 variables); (c) engage-
ment of independent investigator (n = 2 variables); (d) judicial findings, or judicial sentiment
(i.e. mental attitude or feeling) (n = 5 variables); and (e) court orders (n = 8 variables).
Each in- scope judgement was read and coded by the first author (Webb). The coding was
then cross- validated by a group of law students (n = 12), working individually or in pairs to re-
fine the exclusion or inclusion criteria, and add their comments to assist in subsequent reviews
of coding decisions. Student cross- validation of the coding for each judgement, cell by cell,
was compared with the findings of the lead author using the Excel VLOOKUP function.
20
The
resultant comparison sheet was returned to the same student for review; the student was then
paired with another student to discuss any discrepant codes.
21
Students were asked to come to
consensus as a pair, and provide an agreed explanatory note for any remaining discrepant cells
on either of their sheets.
22
The final stage of the validation process once again involved using the Excel VLOOKUP
function to compare findings (now agreed by two students) with those of the lead author.
Each remaining mismatch was subject to further discussion between another two people until
consensus was reached. This process was applied to ensure that (a) exclusions were only made
according to the nominated criteria to validate the coded data for all in- scope judgements
(N = 521) and (b) all coded data for each in- scope judgement had been agreed by at least two
researchers.
23
3 | RESULTS
Each of the tables below is disaggregated by three types of hearings: those that were (a) fully
contested; (b) initially contested but the allegation(s) of child sexual abuse were abandoned
during the course of the hearing; and (c) uncontested. The present study should be seen as a
first step toward future work that would explore our findings in more detail – including bivari-
ate and multivariate analyses, and contextually grounded qualitative analysis.
Tables 2– 8 address the seven research questions noted above. Table 1sets out the character-
istics of the cases in the final 521 in- scope FCoA judgements delivered over a seven- year period.
8
|
WEBB Et al.
It includes the gender distribution of judges which, it should be noted, closely resembles that of
the gender distribution of all FCoA judges in 2018– 2019 (see Family Court of Australia, 2019).
Table 1shows that (a) most of the judgements arose from hearings in which the allegations
were “fully contested” (73%, n = 380), while the remaining judgements arose from initially
contested hearings in which the allegation(s) were either abandoned (13%; n = 70) or were not
TABLE 1 Characteristics of Family Court of Australia judgements involving an allegation of child sexual
abuse (2012– 2019) published in AustLII: contested and uncontested hearings (N = 521)
Attribute of Interest
Contested (n = 450)
Uncontested
(n = 71)
Fully
contested
(n = 380)
Contested but
allegation(s)
abandoned (n = 70)
n % n % n %
Parent making allegation of child sexual abuse
Mother 342 90 63 90 60 85
Father 38 10 7 10 11 15
Applicant parent vis- a- vis allegation direction
Applicant parent made the allegation 147 39 25 36 45 63
Applicant parent had the allegation made
against them
233 61 45 64 26 37
Legal representation
Both parties legally represented 236 62 51 73 44 62
Both parties self- represented 24 6 3 4 2 3
Father only self- represented 68 18 7 10 16 23
Mother only self- represented 52 14 9 13 9 13
Engagement of independent children's lawyer (ICL)
Yes 343 90 70 100 63 89
No 37 10 0 0 8 11
Gender of judge
Male 232 61 36 51 38 54
Female 148 39 34 49 33 46
Stage of judicial determination
Interim 124 33 13 19 12 17
Final (i.e. Trial) 256 67 57 81 59 83
Independent expert report
a
Yes 331 87 66 94 48 68
No 49 13 4 6 23 32
Judgement on Magellan list
b
Yes 108 28 19 27 31 44
No 272 72 51 73 40 56
Notes: % may not sum to 100 due to rounding.
a
Independent expert reports were supplied by court- appointed professionals, internal and/or external to the court
b
Magellan is a case management model implemented by the Family Court of Australia for responding to cases in which one or
both parties have raised serious allegations of sexual abuse or physical abuse of children in a parenting dispute; Magellan was
designed as a fast- track programme, involving protocols for receiving reports from State and Territory child protection authorities
(see Higgins, 2007).
Source: AustLII in- scope judgements 2012– 2019.
|
9
WEBB Et al.
contested (14%; n = 71); (b) most allegations of child sexual abuse were raised by the mother in
both types of contested hearings and in uncontested hearings (90% and 85%, respectively); (c)
the applicant in both categories of contested hearings tended to be the parent against whom
an allegation was made (61% and 64%, hereafter referred to as the “allegedly unsafe parent”)
– whereas the applicant in uncontested hearings tended to be the parent who had alleged the
child to be at risk of sexual harm (63%, hereafter referred to as “allegedly protective parent”);
(d) both parties were legally represented in a majority of contested hearings (62% and 73%
respectively) and uncontested hearings (62%); (e) male judges substantially outnumbered fe-
male judges in fully contested hearings (61% vs 39%), but the differences were small in both
“contested- but- allegations- abandoned” and uncontested cases (51% vs 49% and 54% vs 46%,
respectively); (f) the majority of hearings were final judgements (67% of fully contested, 81%
TABLE 2 Judicial belief in the veracity of the allegations in contested and uncontested hearings in the Family
Court of Australia: Frequency and percentage (N = 521)
Outcome
Contested (n=450)
Uncontested
(n=71)
Fully contested
(n=380)
Initially
contested but
allegation
abandoned
(n=70)
N % n % n %
Belief in truth of allegations noted
a
55 14 1 1 52 73
No indication of belief in truth of allegations 325 86 69 99 19 27
Tot al 380 100 70 100 71 100
Note: % may not sum to 100 due to rounding.
a
In assessing matters on the balance of probabilities, judges in this study frequently cited Dixon J (as His Honour then was) in
Briginshaw v Briginshaw [1938] HCA 34; (1938) 60 CLR 336 at 362. “The truth is that, when the law requires the proof of any fact,
the tribunal must feel an actual persuasion of its occurrence or existence before it can be found. It cannot be found as a result of
mere mechanical comparison of probabilities independently of any belief in its reality”. In Murphy & Curtis [2016] FamCA 474
(20June 2016) at par 69, Faulks J noted that in order to properly find that the father sexually abused the child, “I must feel an
actual persuasion that the sexual abuse occurred”.
Source: AustLII in- scope judgements 2012– 2019.
TABLE 3 Risk of sexual harm to a child found in contested and uncontested hearings in the Family Court of
Australia: Frequency and percentage (N = 521)
Outcome
Contested (n = 450)
Uncontested (n = 71)
Fully contested (n = 380)
Initially contested but
allegation abandoned (n = 70)
n % n % N %
Judge found risk of sexual harm
a
Yes 47 12 0 0 46 65
No 333 88 70 100 25 35
Tot al 380 100 70 100 71 100
Note: % may not sum to 100 due to rounding.
a
Included 16 interim hearings which did not permit findings to be made but in which the risk of sexual harm was clearly
recognised and acknowledged by the judge. This applied in 9 (2%) and 7 (10%) of judgements arising from contested and
uncontested hearings, respectively.
Source: AustLII in- scope judgements 2012– 2019.
10
|
WEBB Et al.
of “contested but abandoned”, 83% of uncontested hearings); and (g) hearings involved an
Independent Children's Lawyer (ICL) in 89– 90% of fully contested hearings and uncontested
hearings, and in all cases in which allegations were contested but abandoned (70%).
Further analysis revealed that allegations of child sexual abuse were more likely to be aban-
doned where there was an expert report (15%) than without an expert report (5%) and also more
TABLE 4 Whether the judge regarded the allegations as genuine but mistaken or deliberately misleading in
contested and uncontested hearings in the Family Court of Australia when no risk of sexual harm was found:
Frequency and percentage (N = 428)
Allegations regarded by
judge as:
Contested (n = 403)
Uncontested (n = 25)Fully contested (n = 333)
Initially contested but
allegation abandoned (n = 70)
n % n % n %
Genuine but mistaken
a
163 49 22 31 10 –
Deliberately misleading
b
76 23 26 37 2 –
Unable to be determined
c
94 28 22 31 13 –
Tot al 333 100 70 100 25 –
Note: Excluded cases (i.e. those in which risk of sexual harm was found) = 93 (47 fully contested hearings, 46 uncontested
hearings); % may not sum to 100 due to rounding.
a
Judges made their views clear in all these cases though not always as an explicit finding.
b
Judge's view in all these cases was an explicit finding.
c
There was no explicit finding and the judge's view could not be determined.
Source: AustLII in- scope judgements 2012– 2019.
TABLE 5 Allocation of shared parental responsibility in judgements containing an allegation of child sexual abuse
in contested and uncontested hearings in the Family Court of Australia: Frequency and percentage (N = 486)
Outcome
Contested (n = 417)
Uncontested
(n = 69)
Fully contested
(n = 349)
Initially contested
but allegation
abandoned
(n = 68)
n % n % n %
Parental responsibility orders
Shared parental responsibility
a
121 35 36 53 3 4
Sole parental responsibility to allegedly
protective parent
149 43 11 16 64 93
Sole parental responsibility to allegedly
unsafe parent
79 23 21 31 2 3
Tot al 349 100 68 100 69 100
Note: Indeterminate cases = 35 (31 fully contested hearings, 2 uncontested hearings, 2hearings in which allegation abandoned); %
may not sum to 100 due to rounding.
a
Most but not all “shared responsibility orders” were equal shared responsibility. Our data distinguish between a parent being
given sole responsibility and orders requiring the parents to share parental responsibility.
Source: AustLII in- scope judgements 2012– 2019.
|
11
WEBB Et al.
likely in a final hearing (15%) than an interim hearing (8%), whereas being on the Magellan list
had virtually no effect (13% Magellan vs. 14% not Magellan) (data not shown).
TABLE 6 Change vs no change to parenting time, and direction of any change, in judgements containing
an allegation of child sexual abuse in contested and uncontested hearings in the Family Court of Australia:
Frequency and percentage (N = 506)
Outcome
Contested (n = 437)
Uncontested (n = 69)Fully contested (n = 368)
Initially contested but
allegation abandoned
(n = 69)
n % n % n %
Orders increasing parenting time with:
Allegedly protective parent 60 16 4 6 46 67
Allegedly unsafe parent 233 63 56 81 8 12
No change to parenting time 75 20 9 13 15 22
Tot al 368 100 69 100 69 100
Note: Indeterminate cases = 15 (12 fully contested hearings, 2 uncontested hearings, 1hearing in which allegation abandoned); %
may not sum to 100 due to rounding.
Source: AustLII in- scope judgements 2012– 2019.
TABLE 7 Continuity and change to “live with” orders in judgements containing an allegation of child sexual
abuse in contested and uncontested hearings in the Family Court of Australia: Frequency and percentage (N = 514)
Outcome
Contested (n = 444)
Uncontested (n
= 70)
Fully contested (n
= 375)
Initially
contested but
allegation
abandoned (n
= 69)
n % n % n %
Residence (“live with”) orders
Continued living with allegedly protective
parent
260 69 37 54 62 89
Continued living with allegedly unsafe parent 34 9 8 12 3 4
Changed from allegedly unsafe to allegedly
protective parent
6 2 0 0 5 7
Changed from allegedly protective parent to
allegedly unsafe parent
63 17 20 29 0 0
Equal care orders remained in place 8 2 3 4 0 0
Changed to equal care 4 1 1 1 0 0
Tot al 375 100 69 100 70 100
Note: Indeterminate cases = 7 (5 fully contested hearings, 1 uncontested hearing, 1hearing in which allegation abandoned); %
may not sum to 100 due to rounding.
Source: AustLII in- scope judgements 2012– 2019.
12
|
WEBB Et al.
3.1 | Extent to which judges expressed belief in the truth of the allegations
How often did judges find that allegations of child sexual abuse were true or likely to be true
(RQ
1,
RQ
7
)? Table 2 addresses this question.
Almost three- quarters (73%) of judges in uncontested hearings expressed a belief in the truth
or likely truth of the allegations. This compared with 14% in fully contested hearings and only
one case (1%) in hearings in which the allegations were initially contested but abandoned.
24,25
Our coding for this variable was binary. That is, we did not explore meanings underpinning
“no indication of belief”. However, our results suggest a clear association between determina-
tion of belief/likely belief in the truth of the allegations on the one hand and findings of risk
of sexual harm on the other. Where risk of sexual harm was not found, we noted that judges
regarded the allegations as “genuine but mistaken” or “deliberately misleading” or made no
determination either way (see Table 4). The findings in Table 4 are likely to be a close proxy
for judicial expressions of belief in the veracity or otherwise of allegations as well as for the
sizeable percentage of cases in which no determination was made.
3.2 | Risk of sexual harm
How often was a risk of sexual harm found by a judge (RQ
2,
RQ
7
)? The data in Table 3 address
this question.
Judges found there was a risk of sexual harm to a child in 12% of all fully contested hearings
compared with almost two- thirds (65%) of uncontested hearings (Table 3). No risk of harm was
found in cases in which allegations were abandoned.
3.3 | “Deliberately misleading” and “genuine but mistaken” allegations
There is a persistent perception in some quarters that a substantial number of allegations of
child sexual abuse are made by vindictive mothers determined to remove fathers from the day-
to- day lives of their children.
26
In order to cast some light on the extent to which this assump-
tion might be reflected in family court judgements, we asked the following research question:
in cases in which a judge did not find that a risk of sexual harm existed, how often were the
TABLE 8 Presence of restraining orders or warnings made against the allegedly protective parent in contested
and uncontested hearings in the Family Court of Australia: Frequency and percentage (N = 513)
Outcome
Contested (n = 443)
Uncontested (n = 70)Fully contested (n = 375)
Initially contested but
allegation abandoned (n = 68)
n % n % n %
Restraining order/warning against allegedly protective parent
Yes 93 25 21 31 2 3
No 282 75 47 69 68 97
Tot al 375 100 68 100 70 100
Note: Indeterminate cases = 8 (5 fully contested hearings, 1 uncontested hearing, 2hearings in which allegation abandoned); the
coders noted five restraining orders imposed on the allegedly protective parent in judgements which the judge nonetheless believed
the allegations to be true.
Source: AustLII in- scope judgements 2012– 2019.
|
13
WEBB Et al.
allegations judged to be deliberately misleading rather than genuine but mistaken (RQ
3,
RQ
7
).
Table 4 addresses this question.
Allegations were regarded by judges as genuine but mistaken in about half (49%) of fully
contested hearings compared with almost one- third (31%) of hearings in which the allegations
were initially contested but abandoned (Table 4). This pattern is reversed in relation to alle-
gations regarded as deliberately misleading (fully contested = 23% vs initially contested but
abandoned = 37%). A sizeable proportion of judgements – over one quarter (28%) of fully con-
tested hearings and just under one- third (31%) of hearings in which an allegation was initially
contested but subsequently abandoned – could not be coded on this variable.
3.4 | Basis on which parental responsibility was allocated
How was parental responsibility
27
allocated following contested hearings in which an allega-
tion of child sexual abuse was raised (RQ
4,
RQ
7
)? Table 5 addresses this question.
Table 5shows that in the 417 judgements arising from contested hearings for which parental
responsibility was able to be coded, an order for shared parental responsibility was more likely
in contested hearings where the allegations had been abandoned than if they had been fully
pursued (53% vs 35%). Sole parental responsibility was almost always awarded to the allegedly
protective parent where allegations were uncontested (93%), while it was awarded to less than
half of the fully contested cases (43%) and only a small minority of contested cases where the
allegations were abandoned (16%).
3.5 | Changes in parenting arrangements in the context of alleged child
sexual abuse
How were parenting arrangements allocated following contested hearings in which an allega-
tion of child sexual abuse was raised (RQ
5,
RQ
7
)? Tables 6 and 7 address this question.
At the time of the court hearings, a large majority (85%) of children in these cases lived
with the parent making the allegations (data not shown). Table 6 indicates that in the 506
judgements which were able to be coded, parenting time with the allegedly unsafe parent was
increased in 81% of the contested cases in which allegations had been abandoned, in 63% of
the fully contested cases and in 12% of the uncontested cases. Time was increased with the al-
legedly protective parent in: only 6% of contested cases in which allegations were abandoned;
in 16% of fully contested cases; and in 67% of uncontested hearings.
Table 7shows the proportions of various types of living arrangements that either continued
or changed as a result of parenting orders in the contested and uncontested cases. In total, 514
“live with” orders were able to be coded (including 444 of all the contested cases).
Table 7 shows that in the majority of cases, court orders were for children to continue
living primarily with the allegedly protective parent – especially in uncontested hearings.
Specifically, such orders occurred in almost 9 out of 10 uncontested hearings (89%), around 2
out of 3 fully contested hearings (69%) and a little over one half of cases where the allegations
had been abandoned (54%).
In contested cases, children's living arrangements were more likely to be changed in favour
of the allegedly unsafe parent when allegations were abandoned than when they were fully
contested (29% vs 17%); no such change occurred for any of the uncontested cases. In no case
where allegations had been abandoned was children's residence changed from the allegedly
unsafe to the allegedly protective parent. This change occurred in only 2% (n = 6) of fully con-
tested cases and in only 7% (n = 5) of uncontested cases.
14
|
WEBB Et al.
Further examination (data not shown) revealed that in those instances in which residence
was changed away from the allegedly protective parent, deliberately misleading allegations
were cited as a reason 41% of the time. In a further 42% of judgements in this category, reasons
given for the change included a judicial assessment of the alleging parent's “unshakeable” or
“entrenched” belief in the truth of her or his allegations despite the Court having made a con-
trary finding. In the remaining 18% of cases in this category, the judge determined that the
alleging parent was unwilling or unable to support the child(ren)’s relationship with the other
parent.
3.6 | Restraining orders or warnings associated with child sexual abuse
allegations
How often were restraining orders or warnings imposed on the parent making sexual abuse
allegations (RQ
6,
RQ
7
)? Table 8 examines this question.
In total, 513 judgements could be coded for the presence or absence of a restraining order
or warning directed at the allegedly protective parent. Warnings or restraining orders were
imposed in almost one- third (31%) of cases in which allegations were abandoned, one quarter
of judgements arising from fully contested hearings and only 3% (n = 2) uncontested cases
(Table 8).
The most common restraining order was one that prohibited the parent who made child
abuse allegations from taking a child to sexual abuse counselling. Some of these orders al-
lowed for such counselling with prior written permission from the other parent. Another com-
mon restraining order prohibited any further allegations of this nature being made except to
the Family Court or the Independent Children's Lawyer. In some cases, the allegedly protec-
tive parent was warned that any further making of such allegations would result in having her
or his children removed.
The extent to which there is a causal relationship between warnings or restraining orders di-
rected at the protective parent and abandoning the allegation of child sexual abuse is unclear.
4 | DISCUSSION
Allegations of child sexual abuse pose agonisingly difficult issues for families, family law pro-
fessionals and the courts. The present study sought to document the outcomes of judgements
published in AustLII in which allegations of child sexual abuse were litigated in the Family
Court of Australia. Over a seven- year period between mid- 2012 and mid- 2019, 521 judgements
met our inclusion criteria, of which 450 (86%) arose from contested hearings. Our central aim
was to provide new data at a time when the social and political context, as well as the legally
informed methodology for determining the believability of allegations of child sexual abuse,
has come into sharp focus.
4.1 | Limitations
Before summarising our key findings, several study limitations warrant mention. First, while
we adopted a primarily qualitative context- sensitive approach to coding, our results have been
largely reduced to quantitative data – the counting of bald numbers. This somewhat positivist
reductionist approach could be seen as a crude way to treat such important, rich and complex
data. We believe, however, that there is merit in exploring higher- order patterns in judgements
in specified areas of litigation such as child sexual abuse cases. It is important, we contend, to
|
15
WEBB Et al.
consider the extent to which these aggregate data “fit” with what is known about the preva-
lence and dynamics of child sexual abuse, and what is known about genuine but mistaken or
deliberately false allegations in the context of family law parenting disputes. That said, we
recognise that intensive qualitative examination of the thinking behind the outcomes in these
cases would shed additional light on how such allegations are considered. The cases we exam-
ined are on the public record and we are happy to share our list of cases and coding frame with
other researchers.
A second limitation is that while our method of coding called for independent assessments,
cross- validation through inter- rater agreement and the resolution of potential ambiguities
through intensive discussions, there were inevitable elements of subjectivity in the process –
especially with respect to variables such as indication of judicial belief in the truth of the al-
legation, and whether the allegation was genuine but mistaken or deliberately misleading. We
worked to minimise such uncertainties and, again, would welcome replication by others using
our inclusion and exclusion criteria.
The impact of likely selection effects within a specialised court population creates a third
issue deserving of consideration. For instance, our sample did not include families who ap-
peared to be chaotic and dysfunctional. Though we are confident that we have captured a
population of cases that met our selection criteria, there are likely to be important differences
between specialised court samples and national random samples in which allegations of child
sexual abuse had been made. Our brief review of prevalence and incidence figures serves as a
reminder that child sexual abuse is unfortunately all too common. But we are unable to make
formal links between this observation and expectations of incidence in a particular setting
such as the Family Court. Overviews such as that of O’Donohue et al (2018) would suggest that
allegations of child sexual abuse in this setting are more likely to be true than otherwise. But
we are also aware that powerful competing narratives persist regarding the juxtaposition of
parental separation and such allegations.
Finally, the data presented speak to a population of judgements made over a seven- year
period up until May 2019. Since that time there has been continued refinement of public and
research- based understandings of the circumstances in which child sexual abuse occurs and of
its impact. The present analysis does not address any developments that might have occurred
within the Family Court of Australia from mid– 2019 onwards.
It is important therefore that our analysis should not be regarded as definitive. Its contribu-
tion is that of an early response to an increasing level of interest in the believability of sexual
abuse claims in the context of parenting disputes and the impact of making such claims on
behalf of children.
4.2 | Key findings
In the majority of judgements examined, children lived primarily with their mother at the
beginning of the hearing (85%) and most allegations of child sexual abuse were raised by moth-
ers. The applicant in contested hearings tended to be the father against whom an allegation of
child sexual abuse was made. Typically, in these cases the applicant father alleged a breach of
parenting orders because the mother was withholding the child/ren on the basis that the child
was being sexually abused while in the father's care.
Among all judgements arising from fully contested hearings: (a) judicial officers expressed
belief that allegations of child sexual abuse were true in 14% of the cases; (b) risk of sexual
harm to a child was found in 12% of judgements; (c) when no risk of sexual harm was found and
a judicial view could be determined, judges were more than twice as likely to regard the allega-
tions as genuine but mistaken rather than to have been deliberately misleading; (d) just under
two- thirds of allegedly unsafe parents had the time they spent with their child(ren) increased
16
|
WEBB Et al.
by the court; and (e) in 17% of judgements, residence arrangements were changed to the al-
legedly unsafe parent.
When considered against uncontested hearings, judges expressed belief in 73% of judge-
ments, while risk of sexual harm was found in 65% of judgements. A risk of sexual harm was
not found in any of the judgements in which the allegations were abandoned during the hear-
ing although, in one such case, the allegations were believed. Allegations were more likely to
be abandoned where there was an expert report rather than when no such report existed, and
during a trial rather than an interim hearing. Being on the Magellan list, on the other hand,
made no difference to whether allegations were abandoned.
4.3 | Triangulation of findings
Our findings are broadly in line with those reported by Ferguson et al. (2018) in Australia and
more recently by Meier (2021) in the United States. As noted, Ferguson et al found that only
10% of the allegations of child sexual abuse made in the Family Court of Australia were “sub-
stantiated”. They found that 73% were “unsubstantiated” and 16% resulted in no determina-
tion. Of the “unsubstantiated” cases, 84% were coded as disbelieved.
Meier (2021) reported on 2,189 judgements resulting from mothers’ allegations of family vio-
lence and child maltreatment made to a wide range of courts in the United States. With respect
to allegations of child sexual abuse, Meier (2021) found, among other things, that mothers were
believed in 19% of the judgments.
Like the present study, the data from Ferguson and her colleagues were derived from a pop-
ulation of AustLII judgements. Their population of 156 judgements was considerably smaller
than ours, and many of the coding categories differed. However, two key outcomes – a low
percentage of findings of unacceptable risk, and a low percentage of cases in which judges
indicated belief in the truth of the allegations – are consistent with our own.
In addition, among the 73% of judgements in which allegations were categorised as unsub-
stantiated, Ferguson et al. (2018) found that almost one in four were thought by the judge to
contain elements of “parental alienation”;
28
and among the 16% of judgements in which no
determination was made, one in five judges nonetheless identified parental alienation as a
factor to be considered.
The present study did not code for suggestions of “parental alienation” but did code for an
overlapping variable: that the allegedly protective parent did not hold a genuine belief in the
truth of his or her allegations (i.e. the parent made allegations that the judge regarded as delib-
erately misleading). Of those judgements arising from a contested hearing in which the judge's
view could be discerned, 25%
29
contained a formal finding that the allegations were deliber-
ately misleading. This figure is consistent with the parental alienation data coded by Ferguson
and her colleagues. It is also consistent with the findings that restraining orders or warnings
were directed toward the allegedly protective parent in 25% of the fully contested hearings.
4.4 | Two contextual observations
Before concluding, two contextual observations warrant brief mention. The first speaks to
reflections by Meier (2021), whose key findings have been noted above. The second speaks to
reflections by Middleton et al. (2014) on the “dynamics of silence”.
30
Meier (2021:33) asks the question, “Why are mothers’ claims of abuse so widely denied in
court?”. Meier is clearly of the view that many such denials are ill- founded. As she puts it,
“[s]ocial media and professional reports abound in which clear evidence of almost certain
abuse is ignored, minimized and sidestepped, or its existence denied” (Meier, 2021:44). Meier’s
|
17
WEBB Et al.
(2021:44) belief is that “human brains are hard- wired with defenses against awareness of hor-
rific realities, especially those inflicted by humans against others”. She suggests that “simply
inferring that too many judges are ignorant or biased is both questionable and unfair to the
many conscientious judges doing their best to achieve what they believe is right for children”
(Meier, 2021:44).
Our reading of the judgements, many of which go to considerable lengths to carefully weigh
the evidence presented, is consistent with this suggestion. It may also be, as Meier believes,
that we have inherited a neurological tendency or perhaps an ongoing social disposition to
minimise both the impact of child sexual abuse and the frequency with which it occurs. A com-
peting hypothesis, however, is that we have indeed become increasingly aware of the key issues
around child sexual abuse but that our adversarially informed decision- making processes are
failing families and require revision, reform or replacement (see, e.g. Cossins, 2020).
Finally, we note the views of Middleton et al. (2014). Like Meier (2021), these authors believe
that the existence of child sexual abuse continues to be denied or minimised but link the pri-
mary problem to more global issues of gender and power. Theirs is a darker perspective, which
suggests continued widespread ignorance about what they see as deliberate attempts to silence
victims of child sexual abuse.
Whether it is the controlling incestuous father, the politically connected pedophile
ring, or a hierarchical church that actively avoids the reporting of life- destroying
crimes to law enforcement authorities, the mechanisms for ensuring silence are
very similar. These are, threat and the manipulation of shame, discrediting the
victim’s testimony, isolating, rejecting and dispossessing those who try to speak
out, the ‘buying of silence’, misuse of legal powers to intimidate, death threats,
and in some cases, attempted murder (including pressure to suicide) or even actual
murder … [P]erpetrators – irrespective of their social standing and economic sta-
tus – generally use every means at their disposal to ensure that the veil of silence
continues
(Middleton et al. 2014:582).
Readers will no doubt make up their own minds about the value of these two perspectives. We
have seen our main task not as providing commentary but as that of extracting and reporting
on the data. In a spirit of allowing the data to “do the talking”, we conclude by asking fourteen
questions. Though not exhaustive, we hope these questions might help shape future engagement
with this important issue. We would welcome responses from readers – particularly those on the
frontline of child protection and decision making about children.
4.5 | Some questions arising from the data
1. To what extent are the findings from the present study surprising, unsurprising, or
both? Why?
2. Which patterns and assumptions in our data require further investigation? Ideally, what
would that research strategy, related method(s) and analytic approach look like?
3. How adequate are current litigation processes for responding to, and making determina-
tions about allegations of child sexual abuse made in the context of post- separation parent-
ing disputes? Should alternatives to adversarial processes be explored?
4. How affordable, accessible and adequate are current investigative processes in child sexual
abuse cases? What level of assistance do they provide to courts?
5. Should all cases in which child sexual abuse is alleged be referred to the Family Law Courts’
Lighthouse Project and/or considered as candidates for the Evatt list or the Magellan list?
31
18
|
WEBB Et al.
6. In the light of what is known about the impact of child sexual abuse, to what extent are
judges, barristers, solicitors and Court staff, as well as independent experts employed
within and outside the Court aware of and trained in trauma- informed practice?
7. To what extent is equal shared parental responsibility feasible and viable in cases in which
child sexual abuse is suspected or alleged?
8. How workable are the presumptions, limitations and internal logic inherent in the “unac-
ceptable risk” test set out in the 1988 High Court Judgment inM & M?
9. Why do so few child sexual abuse cases result in a finding of unacceptable risk?
10. Why do so few child sexual abuse cases attract a formal statement of belief in the allegations?
11. Should parents expect an indication from judges as to whether allegations of child sexual
abuse are credible or otherwise?
12. Why do some parents who allege child sexual abuse abandon their allegations during con-
tested court proceedings? Are there lessons to be learned from a close examination of these
cases?
13. Should lawyers consider aligning advice to parents wishing to make allegations of child
sexual abuse with findings from the present study? Would it be prudent to await evidence
from further studies? Or do such studies fall outside the ambit of individual advice?
14. Across all jurisdictions concerned with the protection of children, how can child sexual
abuse investigations be better coordinated?
ACKNOWLEDGEMENTS
We are indebted to the following University of Wollongong law students who cross- validated
the target sample of judgements or coded in- scope cases under the supervision of one of the
authors (Webb): Gabriella Abraham, Jay Daniels- Takagi, Isabella Diener, Rubaiyat Evans,
Vanessa Harley, Ashley Howard, Jessica Jobberns, Lukas Love, Nicholas Marin, Jana
Massingham, Bethany McGhie, Dallas Mojanovski, Samantha Moon, Danielle Moran,
Alysha Tan and Sophia Tao. We are also grateful to Angelo Webb for his help consolidating
the coding data; Tracy Stewart, Sue Hayes and Molly Stewart for administrative support coor-
dinating the work of the University of Wollongong law students; our colleague, Ruth Weston,
for her incisive feedback on several drafts of the paper; and two “critical friends” and the two
anonymous referees for their helpful comments. The ethical conduct of this study was ap-
proved by the Australian National University Human Research Ethics Committee (Protocol
no. 2020/035).
ORCID
Lawrence J. Moloney https://orcid.org/0000-0001-8171-5822
Bruce M. Smyth https://orcid.org/0000-0003-3514-5325
ENDNOTES
1
In their recent review, MacGinley et al. (2019) reported that impacts included signicantly higher risks of medical,
psychological, behavioural, interpersonal and social difculties. In particular, child sexual abuse has been found to
be a risk factor for depression, anxiety, feelings of shame, post- traumatic stress disorder, sexual exploitation, inti-
mate partner violence, suicidality, substance abuse and educational difculties.
2
According to MacGinley et al. (2019), reviews have demonstrated a range of psychosocial protective factors that can
help reduce risk, including family and social supports, personal attributes, active coping style, a sense of personal
inuence and externalising blame for the abuse (see also Higgins et al., 2019).
3
For example, some denitions include the experience of exposure of genitals; some do not.
4
London et al. (2005) concluded at the time that two- thirds of individuals never disclose their victimisation.
5
There are limitations inherent in the retrospective nature of the many studies that rely on adult memories (see, e.g.
Sanci 2019). On the other hand, like many others (e.g. London et al., 2005) these researchers regard the gures as an
underestimate.
|
19
WEBB Et al.
6
Snyder’s (2000) study of nearly 120,000 cases of sexual assault reported to U.S. law enforcement agencies concluded
that 84% of sexual victimisation of children under age 12 occurred in a residence, typically that of the victim or the
perpetrator.
7
In 1955, Weinberg cited estimates of incest – based on popular belief rather than data – as one in a million. Freud's
initial acceptance of his patients’ accounts of incest later turned to incredulity. More recently, Freud's reformulation
of these accounts as fantasies has been widely criticised: see, for example, Masson (1984) and Wolff (1995). Early re-
searchers into sexual behaviour, such as Kinsey et al. (1948), estimated prevalence of child sexual abuse to be about
5000 cases in a million – far short of what we now know to be the case.
8
Numerous studies have demonstrated a substantial gap between rates of notication of child sexual abuse and even-
tual conviction (see, e.g. Cashmore et al., 2018).
9
The Hon Justice McLellan (2017), who provides an excellent summary of these limitations, also cites evidence indi-
cating that notwithstanding greater recognition of the dynamics and prevalence of child sexual abuse in recent years,
the gap between rates of notication and rates of conviction has increased.
10
For example, theCrimes Act (Victoria) 1958 – Sect 37Bwas amended bytheCrimes (Sexual Offences) Act 2005 to
include the following guiding principles. It is the intention of Parliament that in interpreting and applying Subdivi-
sions (8A) to (8G), courts are to have regard to the fact that: (a) there is a high incidence of sexual violence within
society; (b) sexual offences are signicantly under- reported; (c)a signicant number ofsexual offences arecommit-
ted against women,children andother vulnerable persons including persons with a cognitive impairment ormental
illness; (d) sexual offenders are commonly known to their victims; and (e) sexual offences often occur in circumstanc-
es where there is unlikely to be any physical signs of an offence having occurred.
11
M & M (1988) 166 CLR 69.
12
The standard of “unacceptable risk” to be followed according to the High Court's judgement in M & M is one which
may prompt the lay reader to ask what constitutes an acceptable risk of sexual harm. This apparent non- sequitur
was addressed by Chisholm (1989:9) in his early view that the High Court's judgement might possibly be reduced to
a tautology, which he interpreted as, “[T]he court should not order custody or access where the risk of abuse is such
that the court should not order custody or access”. Chisholm also noted that the judgement could be “… one of
those formulas that the late Julius Stone might have called, ‘a category of illusory reference’”.
13
Marriage of B & B (1993) FLC 92– 357; N & S & Separate Representative (1996) FLC 92– 655; WK v SR (1997) FLC
92– 655; A v A (1998) FLC 92– 800; Re W & W Abuse allegations; expert evidence (2001) FLC 93– 085; Re W (Sex
abuse: standard of proof) (2004) FLC 93– 192: W v W (Abuse allegations; unacceptable risk) 2005 FLC 93– 235;
Napier & Hepburn (2006) FLC 93– 303; Potter & Potter (2007) FLC 93– 326; Johnson v Page (2007) FLC 93– 344;
Partington v Cade (no 2) (2009) FLC 93– 422; Nikoladis v Nikoladis [2010] FamCAFC 52.
14
Coders were asked to read the judgements carefully and make a determination with respect to whether or not judges
indicated a belief that the allegations were true or likely to be true. Among cases that did not reach this threshold,
judges either clearly indicated they did not believe the allegations or remained silent on the issue. Based on an
absence of any indication by a judge of the likely truth of the allegations, these cases were also placed in the “no in-
dication of belief category (see in addition, note 25 below)”. Our results suggest a strong co- relation between coding
a determination of belief or likely belief in the truth of the allegations and a risk of sexual harm found by judges.
15
The FCoA effectively makes all of its judgements available to AustLII once they have been made non- identiable.
16
We began our analysis of judgements delivered in May 2019 and worked backwards. By 7June 2012, (the month in
which Family Law Legislation and Amendment (Family Violence and Other Measures) Act 2011 – commonly known
as the “safety reforms” – came into effect), we had identied 521 judgements. We identied a further 525 earlier in-
scope judgements from that point back to May 2006. These data remain available for further scrutiny; but the focus
of the present analysis is on the 521 post- “safety reform” judgements.
17
One of the authors (Webb) did this for the rst 200 judgements, after which all remaining judgements with the term,
“sexual*” were randomly assigned to and cross- checked by six University of Wollongong Law students.
18
The nal list of extracted judgements and our coding frame is available on request.
19
Rhoades et al (2000) identied interim hearings as an especially problematic area for judges attempting to balance
applications for a continued relationship with children against applications to reduce or terminate involvement on
the grounds of alleged violence or abuse. Rhoades et al (2000:76) cite one judge's description of decision- making
processes at the interim stage as more like “artful dodging” than a judicial exercise.
20
The VLOOKUP function in Excel was used to identify cell mismatches between different Workbooks that had been
consistently formatted.
21
The student supervisor's (Webb) case comments were provided to students at this point in the review process.
22
The cross- validation material is available on request.
23
The total 521 judgements included four post- trial interim hearings for which no trial was listed on the AustLII data-
20
|
WEBB Et al.
base for the research period of interest, but trial data were obtained from the interim judgement.
24
Our coding for this variable (see note 14) was informed by our assumption that litigants might expect both an out-
come and an indication of belief or otherwise regarding the veracity of the key allegation(s). We subsequently came
to appreciate that family lawyers’ views on the circumstances in which a judge might or should declare such a belief
are far from unanimous.
25
In some cases (data not shown), allegations were believed but no risk of sexual harm was found. The reasons for this
varied. For example, a judge might regard the child as no longer at risk because the new court orders were thought
to provide sufcient protection; or the allegedly unsafe parent was in jail or otherwise not a current threat to the
child.
26
The most widely cited literature promoting this concept derives from the work of Gardner (e.g. Gardner 1990; 1998)
who coined the highly contested phrase The Parental Alienation Syndrome. As a default explanation for resistance to
children's contact with the “other” parent – especially when linked to allegations of child sexual abuse – Gardner's
largely self- published work continues to attract widespread criticism in scholarly journals (e.g. Lubit, 2019; Rathus
2020).
27
Under Australian family law, there is a presumption that both parents will have a role in making decisions about
major long- term issues (e.g. where a child goes to school or major health issues) – that is “shared parental respon-
sibility”. This presumption, however, does not apply if it is not in the best interests of the child (e.g. where a parent
has engaged in abuse of the child or family violence).
28
See above note 26.
29
25% can be derived from Table 4 as follows: n in fully contested hearings = 76 + n in abandoned hearings = 26 = 102;
group totals = 333 + 70 = 403; thus 102/403 = 25%.
30
Middleton and his colleagues are psychiatrists and related professionals who work with child sexual abuse and its
consequences across a range of countries.
31
The Lighthouse Project is a recent innovation in Australia's family law courts. It aims to provide early screening of
cases, identication and management of safety concerns, assessment and triage, and referral of high- risk cases to
a court list known as the Evatt list. “The Evatt List is an initiative of the Court. A highly qualied team of judges,
Registrars, Family Consultants and Court staff are allocated to help progress a case considered to be high risk.
There are various Court events that take place for these cases, with a nal trial taking place as quickly as possible”
(http://www.feder alcir cuitc ourt.gov.au/wps/wcm/conne ct/fccwe b/repor ts- and- publi catio ns/publi catio ns/famil y- law/
evatt - guide - parties). See Table 1 notes for an explanation of the Magellan list.
REFERENCES
Butchart, A., Harvey, A.P., Mian, M. & Fürniss, T. (2006) Preventing child maltreatment: a guide to taking action and
generating evidence. Geneva: World Health Organization.
Cashmore, J., Parkinson, P. & Taylor, A. (2018) Trends in prosecutions for child sexual abuse in South Australia
1992– 2012. Adelaide Law Review, 39, 99– 124.
Chisholm, R. (1989) Child sexual abuse: the High Court rules on onus of proof. Australian Journal of Family Law,
3, 184– 189.
Chisholm, R. (2011) Child abuse allegations in family law cases: a review of the law. Australian Journal of Family
Law, 25, 1– 32.
Chisholm, R. (2011) Unacceptable risk. A comparison of the family law and care jurisdictions. NSW Children’s Court
‘Children’s Law News. June. Available at: https://www.child rensc ourt.nsw.gov.au/child rens- court/ publi catio ns/
child ren- s- law- news/cln- 2011.html (accessed 2 April, 2021).
Commonwealth of Australia (2017) Royal Commission into Institutional Responses to Child Sexual Abuse Final
Report: volume 2, Nature and Cause. Canberra: Commonwealth of Australia.
Cossins, A. (2020) Closing the justice gap for adult and child sexual assault: rethinking the adversarial trial. London:
Palgrave Macmillan.
Family Court of Australia (2019) Annual Report 2018– 2019. Canberra: Family Court of Australia.
Fang, X., Derek, S., Brown, D.S., Florence, C.S. & Mercy, J.A. (2012) The economic burden of child maltreatment in
the United States and implications for prevention. Child Abuse & Neglect, 36, 156– 165.
Ferguson, C., Wright, S., Death, J., Burgess, K. & Malouff, J. (2018) Allegations of child sexual abuse in parenting
disputes: an examination of judicial determinations in the Family Court of Australia. Journal of Child Custody,
15(2), 93– 115.
Fogarty, J. (2006) Unacceptable risk: a return to basics. Australian Journal of Family Law, 20(3), 249– 295.
Gardner, R. (1990) Sex abuse hysteria: Salem witch trials revisited. Creskill: Creative Therapeutics.
Gardner, R. (1998) The parental alienation syndrome, 2nd edn. Creskill: Creative Therapeutics.
|
21
WEBB Et al.
Higgins, D. (2007) Cooperation and coordination. An evaluation of the Family Court of Australia Magellan case man-
agement model. Canberra: Family Court of Australia.
Higgins, D.J. & Kaspiew, R. (2008) Mind the gap…’: protecting children in family law cases. Australian Journal of
Family Law, 22(3), 235– 258.
Higgins, D., Sanders, M., Lonne, B. & Richardson, D. (2019) Families– private and sacred: how to raise the cur-
tain and implement family support from a public health perspective. In: Lonne, B., Scott, D., Higgins, D.
& Herrenkohl, T.I. (Eds.) Re- visioning public health approaches for protecting children (Vol 9). Switzerland:
Springer, pp. 127– 143.
Kelly, F. & Fehlberg, B. (2002) Australia’s fragmented family law system: jurisdictional overlap in the area of child
protection. International Journal of Law, Policy and the Family, 16, 38– 70.
Kinsey, A.C., Wardell, A., Pomeroy, R. & Martin, C.E. (1948) Sexual behaviour in the human male. Philadelphia: W.
B. Saunders Company.
London, K., Bruck, M., Ceci, S.J. & Shuman, D.W. (2005) Disclosure of child sexual abuse: what does the research
tell us about the ways that children tell? Psychology, Public Policy, and Law, 11(1), 194– 226.
Lubit, R. (2019) Valid and invalid ways to assess the reason a child rejects a parent: the continued malignant role of
“parental alienation syndrome”. Journal of Child Custody, 16(1), 42– 66.
MacGinley, M., Breckenridge, J. & Mowll, J. (2019) A scoping review of adult survivors’ experiences of shame fol-
lowing sexual abuse in childhood. Health & Social Care in the Community, 27(5), 1135 – 1146.
Masson, J.M. (1984) Freud, the assault on truth: Freud’s suppression of the seduction theory. London: Faber & Faber.
Mathews, B. & Collin- Vézina, D. (2019) Child sexual abuse: toward a conceptual model and definition. Trauma,
Violence, & Abuse, 20(2), 131– 148.
McClellan, P. (2017) Criminal justice issues for the Royal Commission. Speech delivered to the ODPP Victims’
Voices Conference, Making stronger connections, Sydney. Available at: https://www.child abuse royal commi
ssion.gov.au/speec hes/2017- odpp- victi ms- voice s- confe rence - makin g- stron ger- conne ctions (accessed 2 April,
2021).
Meier, J.S. (2021) Denial of family violence in Court: an empirical analysis and Path Forward for Family Law. GW
Law Faculty Publications & Other Works (No. 1536) at 16. Available at: https://schol arship.law.gwu.edu/facul
ty_publi catio ns/1536 (accessed 2 April, 2021).
Middleton, W., Stavropoulos, P., Dorahy, M., Kruger, C., Fernandez, R.L., Martinez- Toboas, A. et al. (2014)
Institutional abuse and societal silence: an emerging global problem. Australian and New Zealand Journal of
Psychiatry, 48(6), 581– 583.
Nakashima, I. & Zakus, G. (1979) Incestuous families. Pediatric Annals, 8(5), 29– 42.
O’Donohue, W., Cummings, C. & Willis, B. (2018) The frequency of false allegations of child sexual abuse: a critical
review. Journal of Child Sexual Abuse, 27(5), 459– 475.
Parkinson, P. (1990a) Child sexual abuse allegations in the Family Court. Australian Journal of Family Law, 4(1),
60– 84.
Parkinson, P. (1990b) Unacceptable risk and the investigation of child sexual abuse. Australian Journal of Family
Law, 4(3), 254– 260.
Parkinson, P. (1999) Family law and parent- child contact: assessing the risk of sexual abuse. Melbourne University
Law Review, 23(2), 345– 372.
Parkinson, P. (2014) Child sexual abuse: assessing the risk. Presented at the Independent children’s lawyers: Honouring
the role. Paper for the Independent Children’s Lawyer’s Conference, Sydney, New South Wales, October (pp. 1–
36), Sydney: School of Law, University of Sydney.
Peters, S.D., Wyatt, G.E. & Finkelhor, D. (1986) Prevalence. In: Finkelhor, D. (Ed.) A sourcebook on child sexual
abuse. Beverly Hills: Sage, pp. 15– 59.
Quadara, A., Nagy, V., Higgins, D. & Seigel, N. (2015) Conceptualising the prevention of child sexual abuse (Research
Report No 33). Melbourne: Australian Institute of Family Studies.
Rathus, Z. (2020) A history of the use of the concept of parental alienation in the Australian family law system.
Contradictions, collisions and their consequences. Journal of Social Welfare and Family Law, 42(1), 5– 17.
Rhoades, H., Harrison, M. & Graycar, R. (2000) The Family Law Reform Act 1995: the first three years. University
of Sydney & Family Court of Australia.
Sanci, L. (2019) Understanding and responding to the long- term burdens of childhood sexual abuse. Lancet
Psychiatry, 6(10), 795– 797.
Snyder, H.N. (2000) Sexual assault of young children as reported to law enforcement: victim, incident, and offender
characteristics: a statistical report using data from the national incident- based reporting system. US Department
of Justice, Office of Justice Programs, Bureau of Justice Statistics.
Stoltenborgh, M., Van Ijzendoorn, M.H., Euser, E.M. & Bakermans- Kranenburg, M.J. (2011) A global perspective
on child sexual abuse: meta- analysis of prevalence around the world. Child Maltreatment, 16(2), 79– 101.
Walsh, K., Zwi, K., Woolfenden, S. & Shlonsky, A. (2015) School- based education programmes for the prevention of
child sexual abuse. Cochrane Database of Systematic Reviews, 4.
Weinberg, S.K. (1955) Incest behaviour. New York: Citadel Press.
22
|
WEBB Et al.
Wolff, L. (1995) Child abuse in Freud’s Vienna: postcards from the end of the world. New York: New York University
Press.
World Health Organization (2020) Online fact sheet: child maltreatment (8 June). Switzerland: World Health
Organization. Available at: https://www.who.int/news- room/fact- sheet s/detai l/child - maltr eatment (accessed 2
April 2021).
Wyatt, G.E. & Peters, S.D. (1986) Methodological considerations in research on the prevalence of child sexual abuse.
Child Abuse & Neglect, 10(2), 241– 251.
Young, L., Dhillon, S. & Groves, L. (2014) Child sexual abuse allegations and s 60CC (2A): a new era? Australian
Journal of Family Law, 28(3), 233– 265.
AUTHOR BIOGRAPHIES
Nola Webb began her long career as a barrister and legal researcher. She eventually became
a consultant to the Australian Public Service, with responsibility for commissioning, man-
aging and contributing to major socio- legal projects that changed the face of parental sepa-
ration services to focus on children. Her research publications cover topics ranging from tax
law to the effects of child abuse and neglect. She has a special research interest in the effect
of trauma on the developing brain of a child.
Lawrence J. Moloney is a psychologist, and an adjunct professor in the School of Psychology
and Public Health, La Trobe University. He has more than 200 publications, mainly in
the area of children, parenting and separation. Lawrie was Director of Family Court
Counselling in the Family Court of Australia for 10years; he taught counselling psychology
at La Trobe University for over 20years, before spending six years as a senior research fel-
low at the Australian Institute of Family Studies.
Bruce M. Smyth is Professor of Family Studies with the Australian National University
Centre for Social Research and Methods, Canberra. In 2018 in Washington DC, he received
the Stanley Cohen Distinguished Research Award from the American Association of Family
& Conciliation Courts in recognition of outstanding research and research achievements
in the field of family and divorce. He has published widely in the area of post- separation
parenting.
Robyn L. Murphy is a data analyst and technology consultant. She runs her own consul-
tancy business in web design and development, specialising in complex data systems and
user interfaces. She loves working with complex datasets and has worked across a diverse
range of technical roles in software companies and educational institutions.
How to cite this article: Webb, N., Moloney, L. J., Smyth, B. M., & Murphy, R. L. (2021).
Allegations of child sexual abuse: An empirical analysis of published judgements from
the Family Court of Australia 2012– 2019. Australian Journal of Social Issues, 00, 1– 22.
https://doi.org/10.1002/ajs4.171