Cossins, Annie --- "Cross Examination in Child Sexual Assault Trials: Evidentiary Safeguard Or An Opportunity To Confuse?" [2009] MelbULawRw 3; (2009) 33(1) Melbourne University Law Review 68

CROSS‑EXAMINATION IN CHILD SEXUAL ASSAULT TRIALS: EVIDENTIARY SAFEGUARD OR AN OPPORTUNITY TO CONFUSE?

ANNIE COSSINS^[*]

[A central feature of the Australian court system is the use of cross‑examination as the main means by which eyewitness evidence is tested. The ability to test evidence by cross‑examination has come to be viewed as a right of an accused. This right, however, is not absolute — it is qualified by the interests of the community, which include the protection of victims of child sexual assaults. Recent studies have shown that cross‑examination, far from ensuring that the truth is revealed, often causes inaccuracies in the evidence of children. This is due to the strange language of the courtroom (usually completely foreign to children), the linguistic techniques and other tactics employed by defence counsel, and the true purpose of cross‑examination in child sexual assault cases: an attempt by the defence to create confusion and inconsistencies. Studies also show that jurors tend to apply their preconceived views on sexual assaults when evaluating the evidence of children. Despite the power to do so and despite training, judicial officers are often reluctant to intervene to protect child sexual assault victims when giving evidence. All of this means that cross‑examination in child sexual assault trials can be as traumatic for the victim as the sexual assault itself. This article thus argues that the questions that can be asked of child sexual assault victims should be limited so as to make their experience less traumatic and maximise the accuracy of their evidence. It concludes by suggesting provisions that could be enacted in all Australian jurisdictions to achieve these aims, through the elimination of repetitive and suggestive questions, limits on accusing child witnesses of lying, and the use of professional intermediaries who evaluate children’s ability to answer a question.]

I INTRODUCTION

One of the key features of the adversarial criminal trial is the giving of oral evidence by witnesses and the testing of that evidence through cross-examination. The ‘primacy of the oral tradition’,^[1] within a culture of adversarialism, has produced entrenched patterns of testing oral evidence through leading questions that utilise complex vocabulary, sentence construction and syntax.^[2] Such techniques have been described as ‘legitimated bullying.’^[3]

For children, cross‑examination is that part of court proceedings where their ‘interests and rights … are most likely to be ignored and sacrificed.’^[4] Because of this, the child sexual assault trial has been described as a ‘legally sanctioned’^[5] forum in which children can be emotionally traumatised by the unregulated behaviour of defence counsel and by questions that children do not understand or cannot answer.^[6] Children, as a group, are therefore disadvantaged by a criminal justice process that ‘does not allow them to participate on a full and equal basis’,^[7] even though their evidence is central to the prosecution’s case.

Numerous historical and recent analyses show that trials involving sex offences are governed by specific rules of evidence, judicial warnings and methods of cross‑examination that were originally based on cultural beliefs about women’s and children’s propensity for promiscuity and lying.^[8] These beliefs are well entrenched, not least because of the judicial repetition throughout the centuries^[9] of Lord Hale’s assertion that rape is ‘an accusation easily to be made yet hard to be proved,’^[10] a view that is still used as the basis for justifying protections for the accused.^[11]

However, little work has been done on identifying and preventing the inaccuracies produced during cross‑examination, which arise partly because of the way these prejudicial beliefs play out in the adversarial trial. While the last decade has seen major reforms in all Australian jurisdictions to address children’s vulnerabilities through the use of closed‑circuit television (‘CCTV’), remote rooms and prerecorded evidence,^[12] the process of cross‑examination remains the least regulated part of the adversarial trial.

The centrality of cross‑examination to the adversarial trial reflects the belief that it is the most effective method for testing a witness’s truthfulness and the accuracy of their testimony. As Louise Ellison explains:

Inordinate faith is placed in the capacity of the skilful cross‑examiner to expose the dishonest, mistaken or unreliable witness, and to uncover inconsistency and inaccuracy in oral testimony. Consequently, it is viewed as a fundamental right of the accused in a criminal trial to have the evidence of prosecution witnesses tested by live cross‑examination.^[13]

Yet the focus of the adversarial trial on two parties locked in verbal battle, where one party wins and the other loses, is likely to have a considerable impact on the wellbeing of a child who is the Crown’s chief witness. In this context, ‘the contest between a child … and a defence advocate [will be] far from equal’,^[14] with the child’s vulnerability and the adversarial nature of cross‑examination combining to produce a relationship of power easily exploited by defence counsel. This power discrepancy will be heightened where the traumatic effects of cross‑examination trigger similar feelings of powerlessness to those experienced by the child as a result of the original abuse.^[15]

Many lawyers will argue that ‘[e]vidence which has not been subjected to the conventional safeguards of the adversarial trial process [should be] viewed with considerable suspicion.’^[16] However, the answers given by a child in response to cross‑examination within this power relationship are highly suspect, not because of the qualities of the child, but because of the risk that confusion, suggestion and/or a child’s deference to an authority figure will produce inaccurate evidence.^[17] Whilst a child’s evidence during cross‑examination may be the best evidence for the defendant, this article contends that the best evidence given by a child will be evidence that is not subject to standard cross‑examination techniques.

This article argues that ‘[t]he extent to which truth is prejudiced by the use of these language tactics’^[18] in cross‑examination must be addressed on the basis that the fairness of a trial is to be measured by reference to not only the rights of the accused, but also the rights of victims.^[19] Before examining a number of reform options, this article canvasses the following issues:

• to what extent the right of the defence to cross‑examine is absolute;

• the real purpose of cross‑examination in the child sexual assault trial;

• how the ‘strange language’ of the courtroom produces unreliable evidence;

• what the studies about the impact of cross‑examination on children’s evidence tell us; and

• why reforms so far have been inadequate.

With reference to the public interest in the proper administration of justice, the article concludes by making five recommendations for reforming the cross-examination process in child sexual assault trials.

II THE RIGHT TO CROSS‑EXAMINE: AN ABSOLUTE RIGHT?

It is widely accepted that cross‑examination is ‘[t]he primary evidentiary safeguard of the adversary trial process’^[20] and that the defendant’s inability to effectively test the prosecution’s case will infringe the fair trial principle.^[21] Yet commentators are rarely able to articulate how that right would be affected if cross‑examination were more tightly regulated.^[22]

The rights of an accused under the fair trial principle are not absolute and are to be balanced against ‘the interests of the Crown acting on behalf of the community’.^[23] Indeed, the concept of fairness is not fixed and immutable and ‘may vary with changing social standards and circumstances’^[24] — it is inextricably ‘bound up with prevailing social values.’^[25] The concept of fairness can take into account the interests of the victim,^[26] including the desirable goal of minimising the re‑traumatisation experienced by sexual assault complainants during the trial process. Being informed by the public interest, it can and should be applied in a way that encourages victims to report sexual offences to the police.

In recent times, both Parliaments and other bodies have begun to accept that the right to cross‑examination by the defendant is not an absolute right. For example, in most Australian jurisdictions the right to cross‑examine has been modified where the accused is unrepresented so that the accused is unable to directly cross‑examine a complainant in a sexual assault trial.^[27] In New South Wales, this modification was introduced in response to a gang rape trial in which the defendants had rejected legal representation.^[28] Before this case was set down for trial, the modification had been the subject of a reference to the New South Wales Law Reform Commission (‘NSWLRC’), which discussed the balance between the rights of the accused and other public interests:

the public interest in … the accused [being] fairly tried … does not mean … that the interests of the accused take priority over all other interests … There is a substantial public interest in ensuring that … actual witnesses are not bullied into giving untrue or inaccurate evidence, … because such conduct must undermine public confidence in the administration of justice. Without these protections … the court would be an instrument of injustice rather than an instrument of justice. … [T]o accommodate the accused’s wish to cross‑examine the complainant personally is to confer an inappropriate advantage on the accused. Leaving aside those cases in which the accused is refused legal aid … the most likely motive for refusing representation is the desire to obtain an advantage by virtue of the intense character of direct personal confrontation. This advantage has never been part of the function of a trial or an element of fairness. … It is the view of the Commission that the benefit to complainants and to the community in general [of preventing cross‑examination by the accused] outweighs any perceived detriment to accused persons.^[29]

In this way, the legitimate expectations of an accused under the fair trial principle were balanced against the needs of sexual assault complainants and the benefits to the community at large. This means that the right to cross‑examine prosecution witnesses is not necessarily absolute and can be subject to controls where necessary, a view recently echoed by Spigelman CJ in R v TA:

The difficulties encountered by complainants in sexual assault cases … has [sic] been a focus of concern for several decades. Judges play an important role in protecting complainants from unnecessary, inappropriate and irrelevant questioning by or on behalf of an accused. That role is perfectly consistent with the requirements of a fair trial, which requirements do not involve treating the criminal justice system as if it were a forensic game in which every accused is entitled to some kind of sporting chance. … In a sexual assault matter, it is appropriate for the court to consider the effect of cross‑examination and of the trial experience upon a complainant when deciding whether cross‑examination is unduly harassing, offensive or oppressive.^[30]

Before considering whether or not the right to cross‑examine ought to be subject to specific controls in child sexual assault trials, it is necessary to examine the extent to which cross‑examination is used as an oppressive tool for intimidating and confusing children, rather than as a forensic tool for exposing a dishonest witness. In other words, what is the extent of the problem and are reforms really necessary?

III INQUIRIES INTO THE ROLE OF CROSS‑EXAMINATION IN THE CHILD SEXUAL ASSAULT TRIAL

Over the past 14 years, a number of inquiries have been conducted into the prosecution of child sex offences and the experiences of children as witnesses within the criminal justice system. These inquiries, together with other reports, provide the most detailed information about the way children are treated as witnesses in the Australian criminal justice system. They have found that:

• cross‑examination is one of the worst parts of testifying for children;^[31]

• children are often subject to aggression, humiliation, harassment and accusations of lying from defence counsel^[32] and sometimes have been berated to the point of breakdown;^[33]

• the most hurtful part of cross‑examination for children is being accused of lying;^[34]

• children are subject to lengthy cross‑examination, sometimes without breaks;^[35]

• children are subject to questions that are complex, developmentally inappropriate, repetitive and deliberately designed to confuse and create inconsistencies;^[36]

• the linguistic style of defence lawyers, in comparison to that of other investigative professionals, is least likely to match that of the child;^[37]

• when comparing the professional manner of lawyers and interviewers towards children, only defence lawyers were rated by the researchers as being aggressive, sarcastic or accusatory towards the child;^[38]

• when comparing the fairness of all court participants towards them, defence lawyers were rated as the least fair by child complainants (the 43 child complainants in a study by Judy Cashmore gave this rating);^[39]

• children are commonly discredited because of delay in complaint or continued association with the offender, despite the fact that these are recognised responses to sexual abuse in the literature;^[40] and

• the powers of judicial officers to intervene in order to prevent improper questioning are either ‘exercised sparingly’^[41] or sometimes have no effect on defence counsel questioning.^[42]

The findings of these reports suggest that evidence in child sexual assault trials is ‘derived not out of truth but as a result of bullying or coercion’.^[43] The extent of the problem means that the criminal justice system may be perpetuating injustice in the name of fairness to the accused ‘by inducing fear and bewilderment and by taking unfair advantage of children’s emotional and intellectual stage of development.’^[44] In fact:

As a result of inappropriate cross‑examination, a child’s evidence may be distorted and the child may wrongly be perceived as an unreliable and untruthful witness. … [This] may have … an adverse impact on the child’s emotional state, and a child who has been abused may feel that he or she has been re-victimised by the court system. The welfare of the child may therefore be put at risk. Moreover, if the child is traumatised by cross‑examination at a committal proceeding, he or she may refuse to testify at trial or the prosecution may decide not to proceed with the case in order to spare the child the ordeal of further cross‑examination.^[45]

Some might argue that these problems only existed in child sexual assault trials before technological changes were made to the way children give evidence in court. Yet even with the advent of CCTV and remote rooms for giving evidence, cross‑examination remains one of the most traumatising aspects of giving evidence for children, with recent studies showing that the treatment of children by defence counsel and the styles and types of cross‑examination questions remain unchanged.^[46] In fact, the submission of the New South Wales Director of Public Prosecutions, Nicholas Cowdery, was taken by the Standing Committee on Law and Justice of the Parliament of New South Wales (‘NSWSCLJ’) to indicate that ‘the cross‑examination process as a whole runs contrary to known best practice for dealing with child sexual assault victims’.^[47] Cowdery (on behalf of the Office of the Director of Public Prosecutions) submitted that:

Following disclosure … children become involved with … police, Department of Community Services … officers, sexual assault doctors and counsellors. These professionals … convey to the child the message that they have done the right thing to tell someone, that they are not in trouble and that they are not responsible for the assault occurring. … [Yet] children under cross‑examination have things put to them which seem to go against everything else that a child has been told. … The child can have it suggested that they are making up stories … [because] they do not like the accused, they want some money, [or] they have been persuaded to make this up by someone else. It can also be put to children that they actively encouraged or participated in the sexual assault.^[48]

While there is a common belief that greater regulation by judicial officers will solve the problems associated with the cross‑examination of children,^[49] Christine Eastwood and Wendy Patton reported that:

Both prosecutors and defence lawyers commented that there is a wide variation as to what judges allow during cross‑examination. Many named specific members of the judiciary who are ‘known’ to be either pro‑defence or pro‑prosecution. … [M]any judges and magistrates would not even recognise if questioning is … oppressive or intimidating for a child … [and] even when judges and magistrates try to control brutal defence lawyers, a core [group] of defence lawyers simply refuse[s] to be controlled …^[50]

Similarly, a ‘common theme’ reported by the NSWSCLJ was ‘the failure of judges and magistrates to intervene to curtail harsh or confusing cross‑examinations of children’,^[51] which may partly arise because many judicial officers are appointed with ‘a defence background, so they do not see a problem with the process.’^[52] It will also be much more difficult for a judge to know when to intervene in a line of questioning that is ‘intentionally confusing’ — designed to confuse or to create inconsistencies in evidence — rather than one which is aggressive or intimidating.^[53] Furthermore, there is a conflict for judges between their role in controlling cross‑examination and their duty to protect the rights of the accused, which is likely to produce judicial resistance to making more interventions in cross‑examination.

The NSWSCLJ noted that there were opposing views about how aggressive defence counsel actually are in their cross‑examination of children.^[54] Yet the concerns about cross‑examination of children are just as relevant to those defence counsel who use the ‘[s]oftly, softly, catchee, monkey’ approach,^[55] since repetitive, accusatory, suggestive and misleading cross‑examination is just as possible using a quiet approach as it is with a more aggressive approach. Indeed:

it is perhaps a much more effective tactic to diminish a child’s testimony to engage in … very calm, very quiet, very patient and very caring sorts of cross‑examination that are devastating in the use of tactics which undermine the credibility of the child … One can do that by focusing on peripheral events. One can focus on minor inconsistencies … One can … jump backwards and forwards to different events and different times so that the child is not clear on what you are talking about. All of that can be done in the most pleasant way without upsetting a jury.^[56]

The most common response from the reports and inquiries into how children are treated by the criminal justice system has been to recommend the education of judges and lawyers about the dynamics of child abuse and child development.^[57] A study by Cashmore and Trimboli found, however, that educating judges had no discernible effect on the cross‑examination process in the child sexual assault trials that were evaluated.^[58] In fact, it confirmed anecdotal evidence that judges are reluctant to intervene to protect a child witness from badgering during cross‑examination. The study evaluated trials conducted in both the New South Wales child sexual assault ‘specialist jurisdiction’ and in the Sydney District Court (the ‘comparison registry’).^[59] In the 16 trials that were fully observed, judicial intervention was least common ‘to protect the child witness from badgering or oppressive questioning, [or] to support and encourage the child’,^[60] with only 20.6 per cent (60 out of 291) of judicial interventions being made to control cross‑examination,^[61] compared to interventions made to clarify questions or answers or to support the witness. Although judicial intervention varied considerably according to the particular trial judge hearing the case, it did not vary according to the complainant’s age or the style of questioning by the defence.^[62] In fact, judges intervened less frequently in the trials involved in the specialist jurisdiction compared to those in the comparison registry, both to protect the child from badgering (20 versus 40 interventions) or to support the child (4 versus 46 interventions).^[63] Although Cashmore and Trimboli concluded that the effect of the educational packages delivered to judicial officers and prosecutors in the specialist jurisdiction was unclear,^[64] it is arguable that the relative lack of judicial intervention in the specialist jurisdiction indicates that the education received by judges had little or no impact on their control of proceedings.

IV THE ‘STRANGE LANGUAGE’ OF THE COURTROOM: ARE LEADING QUESTIONS AND CONFUSION THE WAY TO THE TRUTH?

The impairment of a child’s ability to give their best evidence can occur in a number of ways. For example, Mark Brennan has described the features of the ‘strange language’^[65] of the courtroom, by categorising the types of questions used in cross‑examination that take advantage of a child’s developmental vulnerability. These include the use of:

• ‘[t]agging’^[66] — a statement with a question tagged on at the end that encourages the witness’s agreement with the tag (for example, ‘It wasn’t him, was it?’);^[67]

• ‘[n]egative tagging’^[68] (for example, ‘Now this happened on a Friday, was it not?’)^[69] and the ‘[n]egative rhetorical’^[70] question (for example, ‘Now you had a bruise, did you not, near one of your breasts, do you remember this?’)^[71] — statements with a negative that creates a question;^[72]

• the ‘[m]ulti‑faceted question’^[73] — a question that contains more than one proposition so that it is not clear which question should be answered (for example, ‘[D]id he take hold of you and make you do anything? Did he grab hold of your hand or do anything with your hand?’);^[74]

• a question that ‘[lacks a] grammatical and/or semantic connection’^[75] (for example, ‘At any stage whilst you were in the bathroom did he ever enter the bathroom that previous week?’);^[76]

• the ‘[j]uxtaposition of topics’^[77] — ‘[t]opics of unequal significance’ or ‘with no obvious sequential ties’ are ‘placed alongside one another’.^[78] Although meanings between topics are not connected, topic changes, parenthetical statements or repetitions of the witness’s answers are all ‘made to appear of equal importance and what sometimes emerges as a list of unrelated details serves to create a “linguistic fog”.’^[79] An example is:

Q: On that occasion when Mum went to, being that night that Mum [went] to Youth Group, you were at Clareville?

A: I have made a mistake there it wasn’t Clareville, it was West Hampton.

Q: It should be West Hampton. You did not see the defendant at any time when he put his penis in your bottom, did you?^[80]

If these types of questions exemplify the linguistic realities of cross‑examination for child complainants, then it is not surprising that ‘children six to fifteen years of age fail to hear as sensible language around half of what is addressed to them during cross‑examination.’^[81] Although Brennan’s work is now 14 years old, the recent studies by Rachel Zajac, Julien Gross and Harlene Hayne, and by Judy Cashmore and Lily Trimboli have confirmed the lack of a linguistic match between cross‑examiner and child.^[82]

Other defence tactics involve the use of repetitive questions, which may induce a child to give inconsistent evidence ‘because they believe that the first answer was wrong or somehow unsatisfactory.’^[83] Judges and laypeople with no knowledge of the impact of leading and age‑inappropriate questions on a child’s testimony may be beguiled by the fact that the child continues to give answers of some kind, even though they may not understand what is being asked of them.^[84]

However, the nature of cross‑examination is such that there is no way for judges and juries to tell when a child’s testimony is affected by the social factors associated with being interviewed by an authority figure using leading questions. There is a distinction between errors of reporting due to internal factors, such as memory changes, and errors of reporting due to external or social forces.^[85] The belief that a cross‑examiner has uncovered a dishonest and inconsistent witness could, in the case of a child witness, actually mean that cross‑examination has produced a confused and/or psychologically stressed child who has ‘succumb[ed] to the effects of complex, misleading, or aggressive questioning even when he or she was originally telling the truth.’^[86]

‘[T]here is no simple “Pinocchio” test’ for determining whether a child is lying or telling the truth and there are ‘no widely accepted criteria to differentiate true from false memory’ in the psychological literature.^[87] For example, an assent to a negative tag or negative rhetorical question does not necessarily mean the child agrees with the statement — it may mean the child does not have the capacity to refute it. Yet the adversarial system has no method for ascertaining whether the child understood the question to which they have agreed and, therefore, no way of ascertaining the reliability of answers given in cross‑examination.

Other assumptions that guide the cross‑examination of children include the belief that delay in complaint means fabrication;^[88] that children have vivid imaginations and cannot distinguish fantasy from reality;^[89] that adults (in particular police investigators, friends, family members and counsellors) commonly persuade children that they have been sexually abused;^[90] that children have an ulterior motive for deliberately lying, such as revenge;^[91] or that inconsistencies in evidence mean that the child is lying about all of their evidence.^[92] Through skilful cross‑examination, these untested assumptions produce a particular ‘truth’ that the child and the court are unable to counter or correct.

Another assumption is that the dishonest witness will betray themselves by their demeanour.^[93] However, a number of psychological studies show that facial and behavioural expressions are an unreliable indicator of veracity and that untrained professionals (including judges and police officers) are no better than lay persons at predicting veracity through observing a person’s demeanour.^[94] These studies have shown that judges and law enforcement officials misinterpret behavioural cues at or below chance levels (that is, at or below 50 per cent accuracy).^[95]

Cross‑examination can also exploit the difficulties that children have in relation to identifying specific times and dates. These difficulties are ‘particularly problematic for younger children who have not yet learned to tell time on a clock, who may confuse calendar dates or who have trouble reporting events in exact chronological order.’^[96] If children report events out of sequence or if they are unable to give a particular date or time in relation to the alleged abuse, this can be exploited by the defence as bearing on the accuracy of the child’s complaint, even though the inability to give such details has been shown not to have any bearing on the accuracy of the allegation.^[97]

Leading and suggestive questions are known to be the most unreliable method for eliciting information from children.^[98] Researchers warn against the use of multiple interviews of children and leading questions, as both increase the chances that a child will ‘respond affirmatively to leading questions’^[99] or change their answers to please an interviewer.^[100] Yet in the adversarial trial, multiple cross‑examination (where there is more than one defendant) is permitted and leading questions are the order of the day. This is so despite the fact that ‘[o]ver the past three decades, researchers in the field of eyewitness testimony have identified ways of interviewing children that maximize their accuracy and decrease their level of stress while testifying’.^[101] In particular, interviewing protocols emphasise the use of non‑leading, open‑ended questions in order to ensure the accuracy of children’s reports and have been endorsed by various professional organisations.^[102] In Australia, there has been a move towards using a child’s investigative interview as their evidence‑in‑chief,^[103] so whilst there have been attempts to improve ‘the direct examination process, little or no change has been made regarding cross‑examination.’^[104] Since the basis of cross‑examination is to put suggestions to witnesses that contradict their testimony, cross‑examination of children highlights one of the great inconsistencies in the adversarial system: the type of suggestive and leading questioning that would make a child’s police interview inadmissible on the grounds of unreliability is the very type of questioning adopted by the cross‑examiner.

There is considerable tension between the attitudes of defence lawyers and efforts to protect children from suggestive and intimidating cross‑examination,^[105] and care must be taken to ensure that in the desire to preserve adversarialism we do not sacrifice the welfare of children. Clearly, cross‑examiners are aware, strategically speaking, that cross‑examination is not a method for the elicitation of the truth: ‘defence lawyers will admit that if it is necessary to break a child down, they are willing to do that in the interests of their client.’^[106] For example, the following opinion of a defence counsel is an instructive insight into the adversarial model of justice:

You’ve got to get around the idea that the criminal justice [system] is about the child. It shouldn’t be about the child, and hopefully will never be about the child. … [If] I am defending a bloke I want to make life difficult for [the prosecution’s] witnesses. … I’m not there to find the truth … no‑one’s there to find the truth.^[107]

The view that the adversarial system is not about the child or the truth, and is designed to make life difficult for witnesses, should ring alarm bells for the administration of justice and the welfare of children.

V STUDIES EXAMINING THE IMPACT OF CROSS‑EXAMINATION ON CHILDREN’S EVIDENCE

Because many of the objections to cross‑examination are made without any empirical basis, it is important to engage with the research literature to better inform the criminal justice reform agenda. Three key empirical studies are relevant to informing this agenda in relation to the cross‑examination of children. The first, by Cashmore and Trimboli, showed that, although jurors are aware of some of the difficulties faced by children during cross‑examination, jurors’ verdicts are influenced by the consistency of the evidence and the credibility of the complainant — both of which are the subject of attack during cross‑examination.^[108]

In a survey of 277 jurors that analysed their perceptions of how child complainants were treated,^[109] Cashmore and Trimboli reported that jurors ‘perceived the questions asked by defence lawyers to be less well understood by the child complainants than those asked by crown prosecutors.’^[110] In addition, ‘[o]nly

15 jurors, from eight trials, commented that the defence lawyers asked age‑appropriate questions’,^[111] whilst one in five jurors said that defence counsel ‘were inappropriate in either their language or their behaviour towards the child complainants’.^[112]

When it came to the treatment of children, judges were perceived as ‘very fair’ by 85.1 per cent of jurors, prosecutors as ‘very fair’ by 58.0 per cent and defence lawyers as ‘very fair’ by only 33.9 per cent of jurors.^[113] Defence lawyers were rated as either ‘very unfair’ or ‘quite unfair’ (12.9 per cent of jurors) more frequently than either judges (0.7 per cent of jurors) or prosecutors (2.2 per cent of jurors) in relation to how children were treated.^[114]

Not surprisingly, there was a positive correlation between juror perceptions of consistency and credibility, so that children who were considered to give consistent testimony were rated as more credible. Jurors ‘focused on inconsistency and lack of detail, referring to clothing, times and dates’ when rating a child’s testimony as ‘not at all convincing’ and ‘[c]redibility was also strongly correlated with the child’s perceived confidence in answering both the crown prosecutor’s … and the defence lawyer’s questions’.^[115] This suggests that deliberate questioning about detail and weakening a child’s confidence through various cross‑examination tactics are likely to contribute to decreased perceptions of credibility by jurors. The study found that:

both the perceived consistency and credibility of the child complainant were significantly associated with the verdict. Juries which returned a guilty verdict on some or all charges rated the child complainant as significantly more consistent and more credible than those which acquitted the defendant.^[116]

Not only does this study support the findings made by Eastwood and Patton and Cashmore and Trimboli about treatment of children by defence counsel,^[117] it also provides empirical evidence to confirm what has long been suspected: that there is a positive correlation between juror perceptions of complainant consistency, credibility and verdict. There is, however, a distinct mismatch between what lawyers and jurors believe children’s inconsistency indicates (unreliability and lack of veracity) and the findings of the research literature. In other words, ‘the emphasis on consistency may be misplaced’ since ‘inconsistencies in children’s accounts of sexual abuse do not indicate that their claims are false’.^[118]

Two studies have specifically analysed the phenomenon of inconsistent evidence given by children in response to cross‑examination style questions. The first is by Zajac, Gross and Hayne,^[119] who undertook an analysis of the transcripts of 18 examinations‑in‑chief and re‑examinations as well as 21 cross-examinations involving child complainants between the ages of 5 and 13 years in order to compare the questioning styles of defence lawyers and prosecutors.^[120] This led to an analysis of 600 questions put in examinations‑in‑chief and re‑examinations as well as 2935 questions put in cross‑examination.^[121] Each question was coded into at least one out of seven different categories representing questioning styles, whilst the children’s responses were coded according to the cognitive strategies they used to respond to the lawyers’ questions, including adaptive strategies, such as silence, compliance or asking for clarification.^[122]

The first set of analyses compared the styles of questioning used by prosecutors and defence lawyers, and assessed whether or not that style ‘differed as a function of the child’s age.’^[123] As expected, defence lawyers asked a ‘greater proportion of complex, … grammatically confusing, … credibility‑challenging, … leading, … and closed questions’^[124] compared to prosecutors. The most common types of defence questions were leading and closed questions.^[125] Zajac, Gross and Hayne concluded that the questioning style did not differ as a function of the child’s age.^[126]

The second set of analyses assessed children’s responses to the questions and found that that these responses varied as a function of the type of lawyer (prosecutor or defence).^[127] Children showed more uncertainty, misunderstandings, compliance with leading or closed questions and changes to earlier testimony in response to defence questions.^[128] In other words, children answering defence questions ‘exhibited high rates of misunderstanding and compliance with leading and closed questions, and a low rate of clarification seeking.’^[129] In particular,

[Seventy‑six per cent] of children … made changes to one or more of their earlier statements under cross‑examination. … Ninety‑five per cent of changes … were made in response to leading questions, credibility‑challenging questions, or combinations of both.^[130]

Yet these types of inconsistencies in a child’s evidence can have significant ramifications for the prosecution’s case, since

children who make one or more changes to peripheral details … yet still allege abuse may be perceived as unreliable in the eyes of the jury, … [resulting] in a not guilty verdict. [Additionally], witnesses who make substantial changes or recant altogether may lead a case to be dismissed. In either case, if the child has actually been abused, justice has not been served and the child remains vulnerable to further abuse.^[131]

Overall, the results reported by Zajac, Gross and Hayne confirm the findings in other studies^[132] that there is ‘a clear mismatch between courtroom language and the language capabilities of the children being questioned’ and that, cognitively speaking, ‘children do not cope well with the cross‑examination process.’^[133] This was particularly so for children under the age of 10 years, who were less likely than older children to seek clarification of complex, confusing or ambiguous questions, possibly because ‘children below the age of 8 years cannot reliably detect [their own] non‑comprehension.’^[134] Yet the age of the child was found to have had no effect on the types of questions asked by defence lawyers, indicating that they did not ‘make allowances for the child witness’s developmental level.’^[135]

The study also found that children’s responses depended on the questioning style of the lawyers and the types of questions asked,^[136] which Zajac, Gross and Hayne found ‘encouraging, as it suggests that cross‑examination can be made more appropriate for children merely by changing the types of questions deployed.’^[137]

Yet the fact that 76 per cent of the children in the study changed their testimony during cross‑examination led the authors to ask whether the changes made by children constituted changes ‘towards or away from the truth’.^[138]

To answer this question, Zajac and Hayne conducted a second study to examine the effect of cross‑examination on children’s reports of a contrived event (a trip to a police station) in which the children actually participated.^[139] After the event, but before the first interview, some of the children were exposed to misleading information in order to attempt to mimic the situation where a child has been misled or coached prior to testifying.^[140] All children (46 five‑ and six‑year‑olds) were interviewed six weeks after the event using open‑ended questions and then more specific questions about two actual things that had happened (their photo was taken and they were shown a police car) and questions about two things that did not happen (trying on handcuffs and seeing a lady report a stolen bike).^[141] Eight months after this direct examination, they were interviewed by a different interviewer using cross‑examination style questions and language in order to try to persuade the children to change their accounts of what happened or to admit that their answers might have been wrong.^[142]

Zajac and Hayne reported that 85 per cent of the children ‘changed at least one of their original responses during cross‑examination, and one third … changed all of their original responses.’^[143] There were no significant differences in the number of changes made between the control group and the group of children who had been given misleading information.^[144] These findings were found to be highly consistent with the study by Zajac, Gross and Hayne of actual court transcripts in which 76 per cent of children were found to have made one or more changes to their evidence during cross‑examination.^[145]

When comparing the accuracy of the control group with the misled group, Zajac and Hayne reported that the misled group was significantly less accurate during direct examination than the control group.^[146] However, ‘the errors made by children in the misled group were largely restricted to the false events [which] suggests that the effect of prior misinformation was highly specific to those events.’^[147] In other words, the children in the misled group were as accurate as the children in the control group when asked, during direct examination, about events that had actually happened.^[148]

The next question addressed by Zajac and Hayne was whether the changes made by 85 per cent of the children were changes that corrected earlier mistakes, or whether the children made changes to answers that were originally correct.^[149] While the children’s overall accuracy during direct examination differed according to whether they were in the control group or the misled group, after cross‑examination, the children’s accuracy scores did not differ significantly as a function of which group they were in:

For children in the misled group, their accuracy after cross‑examination was not significantly different from the low level that was seen during direct examination [due to responses relating to false events] … For children in the control group, on the other hand, their accuracy actually declined after the cross‑examination interview …^[150]

In addition, ‘children were just as likely to change a correct answer under cross‑examination … as they were to change an answer that had initially been incorrect’.^[151] Further, the misled group’s prior exposure to misinformation did not make them more susceptible to probing cross‑examination.^[152] This means that children who had given accurate answers were somehow influenced or persuaded during cross‑examination to change their previous accurate accounts of what had happened at the police station. This may have been due to a weakened memory of the event and greater susceptibility to suggestion (since cross‑examination occurred nine and a half months later) and/or because of ‘compliance with suggestions that [the children knew to be] incorrect’.^[153] Zajac and Hayne concluded that ‘cross‑examination not only proved to be unsuccessful in discrediting inaccurate children (ie, misled children questioned about false events), it also decreased the accuracy of children who were initially correct.’^[154]

Confirming what a number of inquiries have suggested,^[155] this study demonstrates that cross‑examination, ostensibly designed to flush out the dishonest witness,^[156] is incapable of achieving that task with five‑ and six‑year‑old children. Rather, in a majority of the children who participated in the study, it achieved what has long been thought to be its main aim in sexual assault trials — producing inconsistencies or retractions. In other words, cross‑examination was not able to discriminate between true and false testimony, and produced inaccuracies in children who had previously given accurate accounts. The question is whether these inaccuracies are produced within the actual courtroom. Since the cross‑examination questions used in the study were based on the questioning styles used by defence counsel in child sexual assault trials, and because children are cross‑examined under much more intimidating conditions than those used in Zajac and Hayne’s study (in a public forum, by people unknown to them and for considerably longer periods of time), it is likely that cross‑examination will produce similar inaccuracies during child sexual assault trials.

VI JUROR BELIEFS AND BIASES: THE IMPACT OF RAPE MYTHS AND STEREOTYPES DURING CROSS‑EXAMINATION

Whilst each child sexual assault trial will be characterised by a number of unique factors — the particular facts of the case, the nature of the evidence, the age of the child, and the relationship between the child and the accused — a jury’s decision‑making takes place within a cultural framework of gender relations that is reproduced by specific patterns of cross‑examination. Where these patterns have the effect of constructing the child as an unreliable witness by tapping into common myths and stereotypes about the reliability of children, this may mean that a jury is more likely to acquit the accused by relying on such myths and stereotypes.

Recently, the Australian Institute of Criminology undertook two studies that showed that:

• the pre‑existing attitudes of mock jurors about sexual assault influenced their judgements about the credibility of the complainant and the guilt of the accused;^[157]

• these pre‑existing attitudes influenced their judgements more than the facts of the case and the manner in which the evidence was given;^[158] and

• mock juror attitudes are reflective of the attitudes and beliefs of the general community about rape myths and stereotypes.^[159]

These results — and the fact that jurors are drawn from the general community — mean that we cannot discount the effect of pre‑existing attitudes in jurors’ decision‑making:

Juror beliefs and attitudes about what a sexual assault case looks like and how a victim of sexual assault would behave therefore become critical to understanding why complainants may or may not be believed by jurors, and whether a particular sexual assault case is likely to achieve a guilty verdict.^[160]

Cross‑examination, therefore, is likely to play a central role in confirming pre‑existing attitudes and beliefs of jurors since

[t]here is considerable evidence that people, including professionals …, tend to be biased towards information that confirms their initial beliefs and to reject information that challenges their established views … [so that once] formed, impressions and beliefs can be resistant to contradictory evidence …^[161]

A perusal of the trial transcripts of the cross‑examination of sexual assault complainants, whether adult or child, shows that defence counsel construct scenarios that mirror the commonly believed rape myths and stereotypes, arguably ‘creat[ing] a confirmation bias that may result in faulty perception and incorrect interpretation’^[162] on the part of the jury. The jury system ‘is based on the assumption that jurors are able to judge and evaluate the evidence and to make rational decisions based on information presented in court’,^[163] yet the work of Natalie Taylor and her colleagues suggests that jury decision‑making is, unfortunately, affected by ‘the tendency to deal with information [such as evidence presented in court] by incorporating it within an existing schematic framework’.^[164]

Sometimes defence counsel construct particular stories containing elements of the myths that are commonly associated with sexual assault. A useful example is the published cross‑examination of Tegan Wagner who, at the age of 14, was gang-raped by three brothers known as MSK, MAK and MMK. At the age of 17, she was cross‑examined by three defence barristers for three days, who were described as ‘three barristers operating as a hunting pack … with little intervention by the judge or the Crown.’^[165] The first cross‑examination by counsel for MSK focused on Tegan’s consumption of alcohol, supposedly for the purposes of releasing her inhibitions and seducing one of the brothers. Whilst it is entirely appropriate for defence lawyers to put their client’s version of events to the complainant, the content of the questions are remarkable because they create an entirely passive male and an entirely active female sexual ‘aggressor’ (who in this case was a 14‑year‑old virgin), presumably to elicit sympathy for the accused and disapproval towards the complainant. Some of the suggestions used to create this story (all of which were refuted by Tegan) are set out below:

And I suggest to you that you continued to pour yourself drinks … during the course of the evening … What do you say to that?

And that you don’t want people to think … that you were excited about drinking vodka this evening. That’s the truth, isn’t it?

I suggest that you … went and sat next to [him] and you placed your hand on his leg?

You put your right hand on his left thigh and began to stroke it?

And I suggest to you that you moved your hand to his crutch area, what do you say about that?

He put his hand on your hand, removed it from the crutch area, and you said to him: ‘Let’s go to your bedroom’, correct?

And you said to [him]: ‘I want to do it 69 style’?

And he said: ‘I can’t do that’, and you said: ‘Can I suck your dick’?

[A]nd that’s what you did, you started sucking his penis, is that correct?

You then lay back down on the bed and produced a condom, correct?

And [when you] took the condom off the floor where he’d placed it, and it appeared that it was broken …

You’d gone from passionate, willingness, wanting, to hysterical crying because you thought: ‘My God, I could get pregnant’, correct?^[166]

For reasons of space it is not possible to include the cross‑examination by the other two barristers,^[167] although at no stage did Tegan veer from her version of events.^[168] Despite the attempts to portray Tegan as the sexual aggressor who became deliberately drunk in order to overcome her inhibitions and then cried rape due to fears of pregnancy, it appears that her credibility remained intact because of her consistency in giving evidence, her immediate report to the police, the medical evidence and the bizarre and disruptive behaviour of the three accused during the trial.^[169]

Although the above cross‑examination did not achieve its desired effect, the type of questions asked of Tegan are likely to confirm juror beliefs and prejudices where the complainant becomes confused, changes their testimony or retracts their complaint, or where issues such as drunkenness and flirtatiousness are not able to be denied.^[170]

VII JUDICIAL INTERVENTION: THE BEST CONTROL OF CROSS‑EXAMINATION?

So far this discussion has identified five key problems with the cross-examination of children:

• mental abuse, intimidation and humiliation by defence lawyers are relatively common occurrences in child sexual assault trials in Australia;

• cross‑examination questions have been shown to produce inaccurate testimony from a majority of children who had previously given accurate testimony;

• defence counsel deliberately use complex and leading questions to produce inaccuracies and inconsistencies in children’s testimony;

• mock jurors’ pre‑existing beliefs about sexual assault influence their decisions more than the facts of the case and the manner in which the evidence was given; and

• these beliefs are likely to be reinforced by cross‑examination that taps into common myths and stereotypes.

It is thus possible to conclude from both the research literature and the many inquiries conducted in Australia that cross‑examination is, at times, both used and allowed to be used as an instrument of injustice that may produce inaccurate evidence favouring the defence case. The question is whether it is in the public interest for jurors to make decisions based on what amount to manufactured inaccuracies, particularly since other evidence strongly suggests that cross‑examination that confirms pre‑existing beliefs on the part of jurors will influence jury decision‑making.^[171] If it is accepted that a fair trial necessarily involves preventing the contamination of a child’s evidence, then there is a need to consider the ways in which the cross‑examination process can be reformed.

It may have been hoped that the introduction of measures allowing children to prerecord their evidence or give live evidence from another location via CCTV (with support persons present) would protect them ‘from the full rigour of adversarial’ proceedings.^[172] Yet despite these improvements, a child’s emotional welfare is not prioritised during cross‑examination, even if individual judges might display concern or ensure a child has sufficient breaks. There are no court personnel responsible for monitoring the welfare of the child during their court appearance. In addition, the style and patterns of cross‑examination have remained unchanged,^[173] and there is no evidence to show that children’s susceptibility to suggestive styles of cross‑examination is reduced through vulnerable witness protections.

However, ‘[t]he difficult issue in relation to child witnesses is to what extent the normal methods of cross‑examination to attack a witness’s credit should be restricted to give proper recognition of the child’s particular vulnerability.’^[174] The first issue to consider is whether or not judicial intervention is the best way of regulating improper cross‑examination of children.

It has been recognised that judicial officers are very often reluctant to intervene to protect witnesses in criminal trials^[175] and may ‘tolerate, or even perpetuate, child abuse by the legal system.’^[176] Although some judicial officers may be ‘more willing to intervene … [to] prevent intimidatory, hostile, badgering tactics’,^[177] they are ‘less likely to intervene in confusing cross‑examination tactics’.^[178] It is therefore necessary to consider the extent to which judges can be expected to intervene to protect vulnerable witnesses and whether they should be relied upon as the only control on cross‑examination in child sexual assault trials.

A recent survey reported that Swedish judges were more sceptical about the reliability of children’s evidence than Swedish police officers.^[179] Although there are no similar studies in Australia, such data raise questions about the attitudes and beliefs of Australian judges in relation to the reliability and credibility of children as witnesses and the possibility that judges who are sceptical about children’s evidence will be unlikely to intervene in cross‑examination that confirms their beliefs.

In addition, the expected neutrality of judicial officers hinders the extent to which they are capable of intervening in cross‑examination, since undue interference is likely to lead to an appeal if the defendant is convicted:

the trial judge’s real dilemma is that to halt cross‑examination … is a hazardous exercise because the appeal court is very likely to say, ‘Well, the accused was not given an adequate and fair trial because this was a relevant line of questioning and counsel for the accused was not given adequate opportunity to pursue it’.^[180]

The Australian Law Reform Commission (‘ALRC’), NSWLRC and Victorian Law Reform Commission (‘VLRC’) also recognised that judges have traditionally been reluctant to intervene in relation to cross‑examination because of ‘concern about jeopardising a fair trial … and/or concern regarding the approach to be taken by the Court of Criminal Appeal.’^[181] As discussed above,^[182] Cashmore and Trimboli’s analysis of 16 child sexual assault trials, as part of their evaluation of the New South Wales Child Sexual Assault Specialist Jurisdiction Pilot Program, confirmed this anecdotal evidence (that judicial intervention to protect children during cross‑examination is unusual), with only 20.6 per cent of interventions in the study being made to control cross‑examination.^[183] Even with intervention, defence counsel can and do refuse to be controlled, all of which leads to the conclusion that legislation that requires judges to control cross‑examination ‘has not worked.’^[184]

VIII REFORMS TO CROSS‑EXAMINATION

Recommendations for reform in this area have tended to focus on both educating judges with the expectation that they will become more interventionist, as well as changing professional conduct rules so that they ‘specifically proscribe intimidating and harassing questioning of child witnesses.’^[185] Such approaches are, however, unlikely to have an effect on the way that child sexual assault trials are conducted, since every jurisdiction in Australia already has legislation which allows a trial judge to disallow certain types of questions put to a witness in a criminal trial and barristers’ professional associations contain similar rules, as discussed below.^[186]

In 2002, the NSWSCLJ considered that these controls were insufficient and recommended that a duty should be imposed on trial judges to control improper questioning. As a result, s 275A of the Criminal Procedure Act 1986 (NSW) was enacted,^[187] which, for the first time in Australia, imposed a positive duty on judges to intervene in relation to a range of improper questions, irrespective of whether or not the other party raised an objection.

In 2005, the ALRC, NSWLRC and VLRC agreed that the uniform Evidence Acts^[188] should be amended to impose such a duty on judges because of the need to protect witnesses who are vulnerable to improper questioning, as well as the need to ensure that ‘the best evidence is received by the court.’^[189] It was considered that this type of duty would be necessary to overcome judges’ long‑standing reluctance to intervene in cross‑examination. As a result, s 41 (the provision governing improper questions under the uniform Evidence Acts) was amended to adopt the terms of s 275A of the Criminal Procedure Act 1986 (NSW).^[190]

Although there was an argument by the VLRC to make the amendments to s 41 applicable only to vulnerable witnesses, the ALRC and the NSWLRC disagreed with this approach so that, in New South Wales, the Australian Capital Territory and federally, the new s 41 has been drafted to apply to all witnesses, not just vulnerable ones.^[191] This general applicability of s 41 was based on the belief that it would be difficult for trial judges to determine the mental, physical or other vulnerability of a particular witness and that lengthy argument might ensue about that vulnerability.^[192] Section 41, as it applies in New South Wales, the Australian Capital Territory and federally, provides:

(1) The court must disallow a question put to a witness in cross‑examination, or inform the witness that it need not be answered, if the court is of the opinion that the question (referred to as a disallowable question):

(a) is misleading or confusing; or

(b) is unduly annoying, harassing, intimidating, offensive, oppressive, humiliating or repetitive; or

(c) is put to the witness in a manner or tone that is belittling, insulting or otherwise inappropriate; or

(d) has no basis other than a stereotype (for example, a stereotype based on the witness’s sex, race, culture, ethnicity, age or mental, intellectual or physical disability).

(2) Without limiting the matters the court may take into account for the purposes of subsection (1), it is to take into account:

(a) any relevant condition or characteristic of the witness of which the court is, or is made, aware, including age, education, ethnic and cultural background, gender, language background and skills, level of maturity and understanding and personality; and

(b) any mental, intellectual or physical disability of which the court is, or is made, aware and to which the witness is, or appears to be, subject; and

(c) the context in which the question is put, including:

(i) the nature of the proceeding; and

(ii) in a criminal proceeding — the nature of the offence to which the proceeding relates; and

(iii) the relationship (if any) between the witness and any other party to the proceeding.

(3) A question is not a disallowable question merely because:

(a) the question challenges the truthfulness of the witness or the consistency or accuracy of any statement made by the witness; or

(b) the question requires the witness to discuss a subject that could be considered distasteful to, or private by, the witness.

(4) A party may object to a question put to a witness on the ground that it is a disallowable question.

(5) However, the duty imposed on the court by this section applies whether or not an objection is raised to a particular question.

(6) A failure by the court to disallow a question under this section, or to inform the witness that it need not be answered, does not affect the admissibility in evidence of any answer given by the witness in response to the question.^[193]

As can be seen, s 41 now expands the type of improper questions to be disallowed, taking into account the tone and manner of the cross‑examiner and questions based on particular stereotypes. (Section 41 and the duty to disallow an improper question has also been incorporated, albeit in a slightly reorganised fashion, into s 25 of the Evidence Act 1929 (SA)).^[194] A large degree of discretion still remains under provisions imposing a duty to disallow improper questions, however, since they require a subjective judgement as to whether a question is misleading or confusing; whether a question is put to a witness in a manner or tone that is belittling, insulting or otherwise inappropriate; or whether a question is based on stereotype. Questions that are annoying, harassing, intimidating, offensive, oppressive, humiliating or repetitive have to be unduly so, requiring another value judgement by the trial judge.

In New South Wales, new Bar Association Rules have also been gazetted that are designed to prevent barristers from asking sexual assault complainants improper questions and to encourage them to take into account ‘any particular vulnerability of the witness’^[195] in the way questions are put.^[196] Richard Ackland has observed that these new rules represent ‘a determined attempt by the Government to wrest control of the criminal trial process away from the barristers and give it back to the judges’,^[197] which, for the reasons discussed above, may turn out to be futile.

In common law jurisdictions, the regulation of improper questions varies from jurisdiction to jurisdiction, although most do not place a duty on the trial judge to mandatorily disallow an improper question, as is the case under s 41 of the uniform Evidence Acts and in South Australia. For example, Evidence Act 1906 (WA) s 26(1), Evidence Act 1977 (Qld) s 21(1), Evidence Act 1939 (NT) s 16(1) and of the Evidence Act 2001 (Tas) s 41(1) (even though Tasmania has adopted in large part the uniform Evidence Acts) state that a court may disallow an improper question. Like the uniform Evidence Acts, these jurisdictions stipulate the matters that the court may take into account in disallowing an improper question, such as age or mental disability.^[198]

In Victoria, the VLRC concluded in its 2004 Sexual Offences: Final Report that the powers of judicial officers to intervene ‘to ensure that questions asked of witnesses are fair, comprehensible and appropriate … [are] exercised sparingly’, and that greater controls were needed to ‘explicitly deal with questions that are age‑inappropriate or misleading and repetitive’ in order to protect child witnesses.^[199] The VLRC stated:

It is vital to ensure that children … are questioned fairly and appropriately. … [This] will improve the quality of children’s evidence because … [they] will understand and be able to respond effectively to the questions they are asked. There is a public interest in supporting children to give evidence in sexual offence cases. In some cases their evidence will lead to the conviction of offenders who, if they were not apprehended, would go on to abuse other children.^[200]

As a result, a specific recommendation was made by the VLRC to impose a duty on trial judges to intervene to protect witnesses with a cognitive impairment and those under the age of 18 years^[201] — in contradistinction to the approaches taken by the Commonwealth, New South Wales and South Australia. This was subsequently enacted as s 41F of the Evidence Act 1958 (Vic).^[202]

As of 17 September 2008, Victoria is a uniform Evidence Act jurisdiction,^[203] and this provision was incorporated into the new Evidence Act 2008 (Vic). The Victorian version of s 41, therefore, only imposes a duty to disallow improper questions in relation to ‘a vulnerable witness’,^[204] unlike the Commonwealth, New South Wales and South Australian legislation. Despite the opposing views of the ALRC and NSWLRC, the VLRC (in the subsequent Uniform Evidence Law: Report) remained ‘convinced that a specific duty in relation to vulnerable witnesses offers the best prospect of changing the culture of judicial non‑intervention’ because it expected that ‘judicial concerns about the effect on a fair trial of provisions requiring intervention where questioning is inappropriate are less likely to arise if the witness is a vulnerable witness.’^[205]

The VLRC also suggested that, at the time the child is instructed to tell the truth, the trial judge should also inform the child that they:

• may not know or … be able to remember some things that … [they are] questioned about, and … should tell the court if this is the case;

• will be asked questions that may make suggestions that are true or untrue;

• should agree with true statements, but should not feel under pressure to agree if the statement is incorrect, according to the child’s understanding of what happened.^[206]

The VLRC believed that the duty to disallow questions combined with the requirement to tell children how they may answer questions would be sufficient to ensure that child witnesses will be treated fairly.^[207] Because the developmental and cognitive abilities of children, as well as their vulnerability to suggestion and confusion, are matters of specialised knowledge — of which few judges and prosecutors will be aware — the VLRC also recommended education programs for judges and prosecutors, since ‘legislative change in isolation from attitudinal change is not effective.’^[208]

The question is whether these mandatory requirements to prevent improper questions will improve the way in which children are cross‑examined. The results from Cashmore and Trimboli’s study show that there is no evidence that judges are more likely to intervene to protect vulnerable witnesses from improper cross‑examination.^[209] Prior to the enactment of a positive judicial duty in New South Wales, a report by the NSW Adult Sexual Assault Interagency Committee noted that s 41 (before its amendment) was ‘under‑utilised’.^[210] The ALRC, NSWLRC and VLRC also concluded that the use of the old s 41 of the uniform Evidence Acts ‘to control improper questions during cross‑examination is patchy and inconsistent’^[211] since a ‘significant number of consultations and submissions indicated that the section is seldom invoked by judges, and its use often depends on the particular judicial officer and prosecutor.’^[212] Recently, a senior defence barrister effectively confirmed that this practice is continuing when he stated that the new positive duty in New South Wales was not making much difference in practice because of its discretionary nature, based as it is on a judge’s subjective assessment about the improper nature of cross‑examination questions.^[213] Terese Henning also notes that ‘[f]urther evidence of the small role accorded in practice to s 41 is provided by the fact that it has been considered and applied in only one significant case’.^[214]

In relation to the scope of the duty to be imposed under s 41, the ALRC, NSWLRC and VLRC observed that ‘[a]t the heart of the difference between the Commissions is a debate about the most effective way of addressing the competing policy concerns and how best to change the adversarial culture’.^[215] Yet, ‘[i]f it is accepted that misleading, confusing, intimidating, harassing, offensive, oppressive, humiliating and insulting questions negatively affect the reliability of evidence and bring the criminal justice process into disrepute,’^[216] reform of the cross‑examination process ought to be guided by the public interest in ensuring verdicts are based on the most accurate evidence, such that any discretionary approach to disallowing improper questions needs to be removed. If, as the research literature indicates, admissible evidence can become inconsistent, skewed or inaccurate because of the method used to elicit that evidence during cross‑examination, the legitimate public interest in the goals of justice in relation to child sexual assault is undermined.

Improvements to the cross‑examination process are also required to preserve the child’s emotional and mental health,^[217] as well as to improve the child’s and the community’s perceptions of the fairness of the trial, the willingness to report and the child’s overall participation in the trial process.

IX CONCLUSION: RECOMMENDATIONS FOR THE WAY FORWARD

The above findings from 14 years of reports and inquiries into the way children are treated within the criminal justice system lead to the inevitable conclusion that cross‑examination is inappropriately used to exploit a child’s age and developmental vulnerability, and may impede a child’s psychological recovery.^[218] The above discussion shows that aggressive, confusing, intimidating and repetitive styles of cross‑examination are commonly used in child sexual assault trials, including the use of age‑inappropriate language, sentence structure and grammatical styles. The research literature also shows that the techniques of cross‑examination are an important defence method for impugning the accuracy of a child’s evidence, constructing children as unreliable and, thus, influencing a jury’s decision as to the guilt or innocence of the accused, with empirical evidence showing that there is a positive correlation between juror perceptions of consistency, credibility and verdict.^[219]

In order to produce real change to the cross‑examination process, reforms will need to go further than a judicial duty to disallow improper questions, given the subjectivity that is inherent in provisions that require a judge to assess whether questions are unduly annoying, harassing, intimidating, offensive, oppressive, humiliating or repetitive. In other words, ‘where cross‑examination will cross the line from due intimidation (whatever that is) … to undue intimidation … is anybody’s guess.’^[220]

Education of judges is also unlikely to provide a comprehensive solution to the problems of protecting children from the rigours of cross‑examination since there are no consequences for judges if they do not intervene to protect vulnerable witnesses. Because there is some evidence to show that, even with education programs, judges may not be encouraged to change their approach to protecting children during cross‑examination,^[221] other protections for vulnerable witnesses during cross‑examination are required. Further, lawyers and judges who are untrained in child development will not necessarily recognise that a child is stressed or confused or is responding to questions they do not understand.

Since it is not the goal of a cross‑examiner to ask age‑appropriate questions (nor are they likely to possess the skills to do so), this requirement ought to be imposed upon them. A package of five recommendations for legislative reform is set out below. They are based on the earlier recognition that the legitimate expectations of an accused under the fair trial principle must be balanced against the needs of complainants and the expectations of the community at large.^[222] This means that the right to cross‑examine prosecution witnesses is not absolute and can be subject to controls where necessary in the public interest.

It is acknowledged that these reforms are radical, yet, arguably, they are required to meet the equally radical phenomenon that the type of questions permitted during cross‑examination in child sexual assault trials are the same type of questions that studies have shown produce inaccurate evidence from children^[223] and the type of questions that police and other professionals are discouraged from using for fear of contaminating a child’s evidence.^[224]

The first recommendation is to prohibit suggestive questions or statements that are designed to persuade the child to agree with the proposition or suggestion put to them.^[225] In other words, what is not permitted during the investigative interview of the child — in order to ensure the child’s evidence is not contaminated — ought not be permitted during cross‑examination. This recommendation would need to be accompanied by improved training of prosecutors and judges about the differences between suggestive and non‑suggestive questions asked of children in order to ensure that they are able to object to, and disallow, such questions. However, as training may be insufficient, this reform, along with Recommendations 2–4, needs to be implemented with the use of court‑appointed intermediaries, who will assess the appropriateness of the questions put to the child, as outlined in Recommendation 5 below.

Recommendation 1: Suggestive questions during the cross‑examination of child witnesses in sexual assault trials

(1) This section applies to prescribed sexual offence proceedings.

(2) During cross‑examination, the court must disallow a question put to a witness under the age of 18 years, or inform the witness that it need not be answered, if the question suggests a particular answer to which the witness is asked to agree.

Secondly, it is recommended that a prohibition be imposed on asking the same question or making the same statement more than once because of the risk that the child’s answer will be a product of suggestion or a desire to please an authority figure, rather than the child’s memory and experience.

Recommendation 2: Repetitive questions during the cross‑examination of child witnesses in sexual assault trials

(1) This section applies to prescribed sexual offence proceedings.

(2) During cross‑examination, the court must disallow a question put to a witness under the age of 18 years, or inform the witness that it need not be answered, if the question is repetitive.

(3) For the purposes of sub‑section (2), a question is repetitive where the content of the question is largely the same as that of a previous question asked of the same witness.

The third recommendation is to prohibit questions or statements made by the defence that directly accuse the child of lying or being a liar, given the damaging effects of such allegations on a child’s mental health.^[226] Such questions can be rephrased without directly accusing the child of lying,^[227] as in the example, ‘Are you sure you’re telling the truth?’

Recommendation 3: Questions about lying during the cross‑examination of child witnesses in sexual assault trials

(1) This section applies to prescribed sexual offence proceedings.

(2) During cross‑examination, the court must disallow a question or statement put to a witness under the age of 18 years, or inform the witness that it need not be answered, if the question directly suggests that the witness is a liar, or is lying, or has lied.

(3) For the purposes of sub‑section (2), examples of questions and statements that are prohibited include but are not limited to:

(a) You’re lying, aren’t you?

(b) You’re a liar, aren’t you?

(c) It’s all lies.

(d) What you’ve said is a pack of lies.

These first three recommendations are aimed at ensuring that inaccuracies are not produced by the typical questions used by defence counsel during cross‑examination. This would have the effect of enhancing the fairness of the trial from the perspective of the complainant, the defendant and the community since it would reduce the extent to which juries base their verdict on inaccurate testimony. All three recommendations are likely to decrease the length of cross‑examination, and hence the trial, by preventing the use of repetitive questions and producing confusion in child witnesses. The restriction of questions that directly accuse a child of lying may also result in less stress and trauma being experienced by the child, which is likely to have a positive effect on the child’s ability to give their best (that is, most accurate) evidence.

The fourth recommendation places restrictions on the use of prior inconsistent statements by the defence.^[228] This would mean that, before the defence can cross‑examine a child on a prior inconsistent statement, the judge must be satisfied that the statement is a prior statement about a matter central to the facts in issue, that it is in fact inconsistent, and that it does not relate solely to a trivial or irrelevant matter. This would prevent a child from being cross‑examined on a prior inconsistent statement that is only relevant to the child’s credibility, on the grounds that there is no evidence that inconsistencies in children’s accounts are the result of lying and are more likely to be the result of cognitive development.^[229] This recommendation is also likely to reduce the length of cross‑examination by restricting the number of questions put to the child about inconsistencies in their evidence.

Recommendation 4: Questions about a prior inconsistent statement during the cross‑examination of child witnesses in child sexual assault trials

(1) This section applies to prescribed sexual offence proceedings.

(2) During cross‑examination, the court must disallow a question put to a witness under the age of 18 years, or inform the witness that it need not be answered, if the question is about a prior inconsistent statement made by the witness, unless the court is satisfied that the prior statement:

(a) is in fact inconsistent with the witness’s evidence given in court;

(b) is a prior statement about a matter relevant to a fact in issue in the trial; and

(c) does not relate solely to a trivial or irrelevant matter.

The fifth recommendation is for a court‑appointed intermediary (social worker, psychologist or other relevant professional) trained in child cognition, language, and development to assess defence questions during the cross‑examination of a child complainant. The role of this intermediary would be to advise the trial judge on whether or not each question put in cross‑examination is age‑inappropriate, suggestive in content, misleading, confusing, oppressive, intimidating, humiliating, repetitive, or unable to be understood by a child of the particular complainant’s age (the intermediary’s training providing an objective basis for this advice).^[230] In addition, the intermediary would be able to advise the court about how to rephrase a particular question so that the child can understand it. This recommendation would also ensure that Recommendations 1–4 are able to be put into practice and can be justified on the grounds that

[i]n order to maximise the ability of children to give accurate evidence, it is essential that they are asked questions appropriate to their cognitive level. This is not fundamentally different from ensuring that witnesses who cannot communicate in English are questioned via an interpreter, in a language they can understand.^[231]

This recommendation is, therefore, akin to appointing an interpreter to ensure not only that cross‑examination questions are able to be understood by a particular child according to their cognitive development, but also that cross‑examination questions do not breach provisions which are designed to prevent improper questions. There is anecdotal evidence that children do not seek clarification when they misunderstand a question — ‘this situation is not picked up by lawyers or the judge who often have their heads down, writing or reading, and who are not observing the child.’^[232] The Manager of the New South Wales Witness Assistance Service has also stated that:

Witness Assistance Officers who are present in the remote witness room while the child is giving evidence by CCTV often have a sense of great frustration because they can see problems that are not necessarily being picked up in court … [O]ur first‑hand experience of seeing children under those circumstances is that children are reluctant to say that they do not understand. Even though [we assist children so that they are able to say they do not understand,] I think that children still believe that they might be in trouble if they do not answer questions.^[233]

This lack of seeking clarification by children was also one of the key findings in the study by Zajac, Gross and Hayne discussed above.^[234]

Recommendation 5: Court‑appointed intermediaries in child sexual assault trials

(1) This section applies to prescribed sexual offence proceedings.

(2) A question must not be put to a witness under the age of 18 years during cross‑examination unless the question has been assessed by a court‑appointed intermediary, trained in the linguistic skills and cognitive development of children and the intermediary has stated their opinion to the court as required by this section.

(3) The intermediary must state their opinion to the court as to whether the question is:

(a) able to be understood by a person of the witness’s particular age and cognitive development; and

(b) an improper question.

(4) For the purposes of sub‑section (3)(b), a question is improper if it:

(a) is misleading or confusing; or

(b) is annoying, harassing, intimidating, offensive, oppressive, humiliating or repetitive; or

(c) is put to the witness in a manner or tone that is belittling, insulting or otherwise inappropriate; or

(d) has no basis other than a stereotype (for example, a stereotype based on the witness’s sex, race, culture, ethnicity, age, or mental, intellectual or physical disability).

(5) Where the intermediary concludes that a question put to a witness during cross‑examination is not able to be understood by the witness or is an improper question, the intermediary must advise the court about whether or not the question can be rephrased and, if so, how it can be rephrased. If the court accepts the opinion of the intermediary, it must disallow a such a question unless the question can be rephrased.

Arguably, the use of an intermediary during cross‑examination is likely to extend the length of cross‑examination, although, on balance, the benefits to a child from the use of an intermediary to protect him or her from inappropriate questions are likely to outweigh any disadvantage due to an increase in the length of the trial (as is the case with an interpreter). On the other hand, the use of an intermediary might save time because there would be less need for trial judges to intervene to clarify questions and answers which arise out of confusion on the part of child witnesses.

Some will argue that the above reforms would impede the rights of an accused to a fair trial. But what is the nature of those rights? Whilst there is a fundamental right to cross‑examine witnesses, previous reforms show that this right is not absolute. Nor is there any case law stating that the right to cross‑examine extends to attempts to produce or manufacture inaccurate evidence. Even if cross‑examination is generally considered the best method for testing the accuracy of a witness’s evidence, the present methodology and cross‑examination techniques used in the specific context of child sexual assault trials are clearly not appropriate to that task. Given that it has been acknowledged that judges play a legitimate role in ‘protecting complainants from unnecessary, inappropriate and irrelevant questioning’,^[235] these reforms, informed as they are by the research literature, merely limit the right to cross‑examine in ways that will improve the quality of the evidence given by children in child sexual assault trials.

This article has documented the evidence that justifies the package of reforms set out above. Such a package is required to achieve real cultural change in the courtroom. If lawyers themselves would not allow their own child to be a complainant within the adversarial system,^[236] then the Australian community and governments owe it to all children to change that system.

[*] BSc (Hons), LLB, PhD (UNSW); Senior Lecturer, Faculty of Law, The University of New South Wales. I would like to express my thanks to the two anonymous referees for their very helpful comments and to my research assistant, Rachael Fraher, for her research work.

^[1] Jill Hunter, ‘Battling a Good Story — The Failure of Evidence Law’ (Paper presented at the International Evidence Colloquium, University of Nottingham, 2004) 1.

^[2] See generally Mark Brennan, ‘The Discourse of Denial: Cross‑Examining Child Victim Witnesses’ (1995) 23 Journal of Pragmatics 71; Rachel Zajac, Julien Gross and Harlene Hayne, ‘Asked and Answered: Questioning Children in the Courtroom’ (2003) 10 Psychiatry, Psychology and Law 199.

^[3] Terese Henning, ‘Control of Cross‑Examination — A Snowflake’s Chance in Hell?’ (2006) 30 Criminal Law Journal 133, 136.

^[4] Brennan, ‘The Discourse of Denial’, above n 2, 73.

^[5] Christine Eastwood and Wendy Patton, The Experiences of Child Complainants of Sexual Abuse in the Criminal Justice System (2002) 4 <http://www.criminologyresearchcouncil.gov.au/

reports/eastwood.pdf>.

^[6] Australian Law Reform Commission (‘ALRC’) and Human Rights and Equal Opportunity Commission (‘HREOC’), Seen and Heard: Priority for Children in the Legal Process, ALRC Report No 84 (1997) 343.

^[7] Reid Howie Associates, Scottish Executive Central Research Unit, Vulnerable and Intimidated Witnesses: Review of Provisions in Other Jurisdictions (2002) 16.

^[8] See generally Judith A Allen, Sex and Secrets: Crimes Involving Australian Women Since 1880 (1990); Carol Smart, ‘Law’s Truth/Women’s Experiences’ in Regina Graycar (ed), Dissenting Opinions: Feminist Explorations in Law and Society (1990) 1, 8; Dorne J Boniface, ‘Ruining a Good Boy for the Sake of a Bad Girl: False Accusation Theory in Sexual Offences, and New South Wales Limitations Periods — Gone but Not Forgotten’ (1994) 6 Current Issues in Criminal Justice 54; Department for Women (NSW), Heroines of Fortitude: The Experiences of Women in Court as Victims of Sexual Assault (1996) 169; Anne Cossins, Masculinities, Sexualities and Child Sexual Abuse (2000) 17–19.

^[9] See, eg, Kelleher v The Queen [1974] HCA 48; (1974) 131 CLR 534, 543 (Barwick CJ); R v Rosemeyer [1985] VicRp 90; [1985] VR 945, 963 (Ormiston J); Question of Law Reserved on Acquittal Pursuant to Section 351(1A) Criminal Law Consolidation Act (No 1 of 1993) [1993] SASC 3896; (1993) 59 SASR 214, 227 (Perry J); Leoni v State, 44 Ala 110, 113 (Peters J) (1870); see also at 114 (Peck CJ); Bute v Illinois[1948] USSC 47; , 333 US 640, 681 (Douglas J for Douglas, Murphy and Rutledge JJ) (1948); Johnson v Alaska, 501 P 2d 762, 766 (Boney CJ for Boney CJ, Rabinowitz, Connor and Erwin JJ) (Alaska, 1972); United States v Wiley[1974] USCADC 5; , 492 F 2d 547, 554 (Bazelon CJ) (DC Cir, 1973). See also R v Sherrin [No 2] (1979) 21 SASR 250, 254 (King CJ); R v Henry (1968) 53 Cr App R 150, 153 (Salmon LJ).

^[10] Sir Matthew Hale, Historia Placitorum Coronae (first published 1736, 1971 ed) 635.

^[11] Christine Eastwood, Sally Kift and Rachel Grace, ‘Attrition in Child Sexual Assault Cases: Why Lord Chief Justice Hale Got It Wrong’ (2006) 16 Journal of Judicial Administration 81, 82,

85–6. For example, in a submission to the Victorian Law Reform Commission (‘VLRC’), the Criminal Bar Association commented that ‘the old adage that an allegation of rape is easy to make and hard to disprove still holds true in some cases’: Criminal Bar Association, Submission No 42 to VLRC, Sexual Offences: Final Report, 14 October 2003, 3, cited in VLRC, Sexual Offences: Final Report, Report No 78 (2004) 89.

^[12] See, eg, Crimes Act 1914 (Cth) ss 15YI–15YJ, 15YM; Evidence (Miscellaneous Provisions) Act 1991 (ACT) ss 8, 40F, 40Q; Criminal Procedure Act 1986 (NSW) ss 306U, 306ZB; Evidence Act 1939 (NT) ss 21B, 21E; Evidence Act 1977 (Qld) ss 21AK, 21AQ; Evidence Act 1929 (SA) s 13A; Evidence (Children and Special Witnesses) Act 2001 (Tas) ss 5–6; Evidence Act 1958 (Vic) ss 37B, 41G–41H; Evidence Act 1906 (WA) ss 106HA, 106N. A ‘remote room’ is one that is located away from the courtroom, but is generally within close proximity. Evidence is to be given by the child from the remote room through the use of CCTV.

^[13] Louise Ellison, ‘The Protection of Vulnerable Witnesses in Court: An Anglo–Dutch Comparison’ (1999) 3 International Journal of Evidence and Proof 29, 35.

^[14] Reid Howie Associates, above n 7, 16.

^[15] See Helen L Westcott and Marcus Page, ‘Cross‑Examination, Sexual Abuse and Child Witness Identity’ (2002) 11 Child Abuse Review 137, 139, 147–8. In relation to trauma triggers, see generally Craig L Donnelly, John S March and Lisa Amaya‑Jackson, ‘Posttraumatic Stress Disorder’ in Mina K Dulcan and Jerry M Wiener (eds), Essentials of Child and Adolescent Psychiatry (2006) 479, 483.

^[16] Ellison, above n 13, 35.

^[17] See below Parts IV and V.

^[18] Brennan, ‘The Discourse of Denial’, above n 2, 73.

^[19] Standing Committee on Law and Justice, Parliament of New South Wales (‘NSWSCLJ’), Report on Child Sexual Assault Prosecutions, Parl Paper No 208 (2002) 78 (‘NSWSCLJ Report’).

^[20] Ellison, above n 13, 35.

^[21] Dietrich v The Queen [1992] HCA 57; (1992) 177 CLR 292, 363, 370 (Gaudron J) (‘Dietrich’); R v McHardie [1983] 2 NSWLR 733, 739 (Begg, Lee and Cantor JJ); R v McLennan [1999] 2 Qd R 297, 303 (Davies JA). This principle has been applied in different ways in several cases: see, eg, S v The Queen [1989] HCA 66; (1989) 168 CLR 266, 275 (Dawson J), 285 (Gaudron and McHugh JJ); R v McLennan [1999] 2 Qd R 297, 303 (Davies JA), 305 (Ambrose J); Stack v Western Australia [2004] WASCA 300; (2004) 29 WAR 526, 534 (Murray J), 545–6, 555–6 (Steytler J).

^[22] Eastwood and Patton, above n 5, 127; NSWSCLJ Report, above n 19, 69, 78.

^[23] Dietrich [1992] HCA 57; (1992) 177 CLR 292, 335 (Deane J), quoting Barton v The Queen [1980] HCA 48; (1980) 147 CLR 75, 101 (Gibbs ACJ and Mason J).

^[24] Dietrich [1992] HCA 57; (1992) 177 CLR 292, 328 (Deane J) (citations omitted). See also McKinney v The Queen [1991] HCA 6; (1991) 171 CLR 468, 478 (Mason CJ, Deane, Gaudron and McHugh JJ).

^[25] Dietrich [1992] HCA 57; (1992) 177 CLR 292, 364 (Gaudron J).

^[26] Ibid 357 (Toohey J).

^[27] Crimes Act 1914 (Cth) s 15YF(1); Evidence (Miscellaneous Provisions) Act 1991 (ACT) s 38D; Criminal Procedure Act 1986 (NSW) s 294A; Sexual Offences (Evidence and Procedure) Act 1993 (NT) s 5; Evidence Act 1977 (Qld) s 21N; Evidence Act 1929 (SA) s 13B(1)(a); Evidence Act 1958 (Vic) s 37CA(4); Evidence Act 1906 (WA) s 106G(1). Under Commonwealth and New South Wales legislation, the complainant can only be cross‑examined by a person appointed by the court (not necessarily a lawyer) who will ask the complainant questions requested by the accused: Crimes Act 1914 (Cth) s 15YF(2); Criminal Procedure Act 1986 (NSW)

ss 294A(2)–(3). In Queensland and Victoria, Legal Aid must provide legal representation for the accused for the purposes of cross‑examination where the accused has not obtained representation for that purpose: Evidence Act 1977 (Qld) s 21O; Evidence Act 1958 (Vic) s 37CA(6).

^[28] New South Wales, Parliamentary Debates, Legislative Council, 3 September 2003, 3026 (John Hatzistergos, Minister for Justice), 3026–7 (Greg Pearce). The validity of Criminal Procedure Act 1986 (NSW) s 294A was upheld by the New South Wales Court of Criminal Appeal in R v MSK [2004] NSWCCA 308; (2004) 61 NSWLR 204, 212–13, 215, 218 (Mason P), 218, 220–1 (Wood CJ at CL), 221 (Barr J).

^[29] New South Wales Law Reform Commission (‘NSWLRC’), Questioning of Complainants by Unrepresented Accused in Sexual Offence Trials, Report No 101 (2003) 45–8 (emphasis in original).

^[30] [2003] NSWCCA 191; (2003) 57 NSWLR 444, 446–7. Spigelman CJ thus applied s 41(1)(b) of the Evidence Act 1995 (NSW), holding that questions asking the complainant in a rape case to interpret a transcript of a video (showing sexual intercourse between the complainant and the accused, of which the complainant had no memory) were unduly harassing, offensive and oppressive, and therefore inadmissible: at 446.

^[31] Judy Cashmore, The Evidence of Children (1995) 32; Eastwood and Patton, above n 5, 4, 128. Children have suggested that seeing the accused is the worst part of testifying: see Cashmore at 29–30, 64; Eastwood and Patton, above n 5, 54–6.

^[32] Cashmore, above n 31, 32–3; Crime Prevention Committee, Parliament of Victoria, Combating Child Sexual Assault: An Integrated Model, Parl Paper No 47 (1995) 191; New South Wales, Royal Commission into the New South Wales Police Service, Final Report (1997) vol 5, 1086, 1135 (‘Paedophile Inquiry’); ALRC and HREOC, Seen and Heard Report, above n 6, 335, 344; Eastwood and Patton, above n 5, 59–62; VLRC, Sexual Offences: Final Report, above n 11,

309–10; Judy Cashmore and Lily Trimboli, An Evaluation of the NSW Child Sexual Assault Specialist Jurisdiction Pilot (2005) 48–51.

^[33] Crime Prevention Committee, above n 32, 191–2; ALRC and HREOC, Seen and Heard Report, above n 6, 344.

^[34] Eastwood and Patton, above n 5, 123.

^[35] Cashmore, above n 31, 33–4.

^[36] Ibid 33–6; Paedophile Inquiry, above n 32, 1086; Eastwood and Patton, above n 5, 59–62,

67–8; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 46–8, 62; Crime Prevention Committee, above n 32, 207.

^[37] Zajac, Gross and Hayne, above n 2, 206–7; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 46–7.

^[38] Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 48–9.

^[39] Cashmore, above n 31, 28. See also ibid 56–7.

^[40] NSWSCLJ Report, above n 19, 8–9, 68–9; Anne Cossins, ‘Children, Sexual Abuse and Suggestibility: What Laypeople Think They Know and What the Literature Tells Us’ (2008) 15 Psychiatry, Psychology and Law 153, 155–6, 161. In the context of New Zealand, see also Emma Davies, Emily Henderson and Fred W Seymour, ‘In the Interests of Justice? The Cross‑Examination of Child Complainants of Sexual Abuse in Criminal Proceedings’ (1997) 4 Psychiatry, Psychology and Law 217, 219–20, 225–6.

^[41] VLRC, Sexual Offences: Final Report, above n 11, 314. See also Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 51–3.

^[42] Eastwood and Patton, above n 5, 126.

^[43] Paedophile Inquiry, above n 32, 1086.

^[44] NSWSCLJ Report, above n 19, 78.

^[45] Queensland Law Reform Commission, The Receipt of Evidence by Queensland Courts: The Evidence of Children, Report No 55 (2000) part 2, 266 (citations omitted).

^[46] Eastwood and Patton, above n 5, 59–62; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 48–51. However, the use of CCTV and remote rooms has minimised the trauma experienced by child witnesses by minimising their chances of seeing the accused: see Eastwood and Patton, above n 5, 96, 119–21; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 39.

^[47] NSWSCLJ Report, above n 19, 61.

^[48] Nicholas Cowdery, Submission No 27 to NSWSCLJ, Report on Child Sexual Assault Prosecutions, 13 February 2002, 10, cited in ibid.

^[49] See below n 57 and accompanying text.

^[50] Eastwood and Patton, above n 5, 125–6 (emphasis in original).

^[51] NSWSCLJ Report, above n 19, 71.

^[52] Evidence to NSWSCLJ, Sydney, 19 April 2002, 3 (Judy Cashmore), cited in ibid.

^[53] See NSWSCLJ Report, above n 19, 71.

^[54] Ibid 63.

^[55] Evidence to NSWSCLJ, Sydney, 19 April 2002, 5 (Judy Cashmore), cited in ibid.

^[56] Evidence to NSWSCLJ, Sydney, 19 April 2002, 21 (Patrick Parkinson), cited in NSWSCLJ Report, above n 19, 64.

^[57] See, eg, NSWSCLJ Report, above n 19, 75–7, 204; Criminal Justice Sexual Offences Taskforce, Attorney‑General’s Department (NSW), Responding to Sexual Assault: The Way Forward (2006) 118; Crime Prevention Committee, above n 32, 208.

^[58] Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 61–4.

^[59] The study was an evaluation of the ‘specialist jurisdiction’ which was established in 2003 after recommendations made by the NSWSCLJ to address the obstacles associated with child sexual assault prosecutions. Cashmore and Trimboli observed 17 trials held in the Campbelltown, Parramatta, Penrith and Sydney District Courts from March to December 2004. Eleven of these trials were heard in the ‘specialist jurisdiction’ and six were held in the ‘comparison registry’ (Sydney). One trial (in the specialist jurisdiction) was only partially observed and was therefore not included in some findings: see ibid 2–8.

^[60] Ibid 52.

^[61] Ibid 52, 54.

^[62] Ibid 52.

^[63] Ibid.

^[64] Ibid 61.

^[65] Brennan, ‘The Discourse of Denial’, above n 2, 74.

^[66] Ibid 86.

^[67] The tag encourages the answer ‘no’, but in order to disagree the child would need to have the cognitive ability to understand that it is necessary to say in reply, ‘It was him.’

^[68] Brennan, ‘The Discourse of Denial’, above n 2, 88.

^[69] Ibid.

^[70] Ibid 75.

^[71] Ibid.

^[72] Both are suggestive, although they create ‘the illusion of choice’ by asking the witness to agree: see ibid. The child must have the cognitive ability to dissect the question to be able to disagree with it.

^[73] Ibid 76.

^[74] Ibid. Where a child gives an answer, ‘the control to interpret which sub question it most easily serves remains with the cross‑examiner.’

^[75] Ibid 77.

^[76] Ibid. The ‘form of delivery’ implies that the question makes sense, and ‘[t]he responsibility for finding sense in nonsense is shifted to the witness, and as the child does not recognise that this doesn’t make sense to adults either, they assume their own inadequacy and failure’: at 77–8.

^[77] Ibid 78.

^[78] Ibid.

^[79] Ibid 79. Because such a questioning style is confusing and suggestive, particularly if it continues over a number of questions, the child may not understand what exactly is being asked, with the possibility that the child will agree with the suggestions put to them.

^[80] Ibid 78.

^[81] Ibid 71.

^[82] Zajac, Gross and Hayne, above n 2, 206–8; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 46–8.

^[83] VLRC, Sexual Offences: Final Report, above n 11, 313 (citations omitted).

^[84] Anne Graffam Walker, ‘Questioning Young Children in Court: A Linguistic Case Study’ (1993) 17 Law and Human Behavior 59, 67.

^[85] See Gail S Goodman and Annika Melinder, ‘Child Witness Research and Forensic Interviews of Young Children: A Review’ (2007) 12 Legal and Criminological Psychology 1, 13.

^[86] Rachel Zajac and Harlene Hayne, ‘I Don’t Think That’s What Really Happened: The Effect of Cross‑Examination on the Accuracy of Children’s Reports’ (2003) 9 Journal of Experimental Psychology: Applied 187, 187. See generally Brennan, ‘The Discourse of Denial’, above n 2.

^[87] Goodman and Melinder, above n 85, 8.

^[88] Davies, Henderson and Seymour, above n 40, 219–21; Anne Cossins, ‘Defying Reality: Child Sexual Assault and the Delay in Complaint Rule’ (1999) 11 Current Issues in Criminal Justice 17, 22–3; Anne Cossins, ‘The Hearsay Rule and Delayed Complaints of Child Sexual Abuse: The Law and the Evidence’ (2002) 9 Psychiatry, Psychology and Law 163, 164–5.

^[89] Mark Brennan, ‘The Battle for Credibility — Themes in the Cross Examination of Child Victim Witnesses’ (1994) 7 International Journal for the Semiotics of Law 51, 53.

^[90] Davies, Henderson and Seymour, above n 40, 221–2; Anne Cossins, ‘Recovered Memories of Child Sexual Abuse — Fact or Fantasy?’ (1997) 3 Judicial Review 163, 163.

^[91] Davies, Henderson and Seymour, above n 40, 224.

^[92] Judy Cashmore and Lily Trimboli, ‘Child Sexual Assault Trials: A Survey of Juror Perceptions’ (Crime and Justice Bulletin Issue No 102, NSW Bureau of Crime Statistics and Research, 2006) 9, 14.

^[93] Ibid 9.

^[94] See, eg, Paul Ekman, Maureen O’Sullivan and Mark G Frank, ‘A Few Can Catch a Liar’ (1999) 10 Psychological Science 263, 265; Stephen Porter, Mike Woodworth and Angela R Birt, ‘Truth, Lies, and Videotape: An Investigation of the Ability of Federal Parole Officers to Detect Deception’ (2000) 24 Law and Human Behavior 643, 650.

^[95] See, eg, Ekman, O’Sullivan and Frank, above n 94, 265; Porter, Woodworth and Birt, above n 94, 650. Ekman, O’Sullivan and Frank found that some professionals, such as law enforcement officials and judges, have high levels of accuracy in lie detection. However even in these groups there was ‘a substantial number who performed at or below chance’: at 265. Cf Samantha Mann, Aldert Vrij and Ray Bull, ‘Detecting True Lies: Police Officers’ Ability to Detect Suspects’ Lies’ (2004) 89 Journal of Applied Psychology 137, 144, where police officers detected lies and truths at above chance levels, however errors were frequently made.

^[96] ALRC and HREOC, Seen and Heard Report, above n 6, 307.

^[97] John R Spencer and Rhona H Flin, The Evidence of Children: The Law and the Psychology (2^nd ed, 1993) 289–91; Karen J Saywitz, ‘Improving Children’s Testimony: The Question, the Answer, and the Environment’ in Maria S Zaragoza et al (eds), Memory and Testimony in the Child Witness (1995) 113, 120–2.

^[98] Pamela C Snow and Martine B Powell, Getting the Story in Forensic Interviews with Child Witnesses: Applying a Story Grammar Framework (2007) 34, 40–3 <http://www.

criminologyresearchcouncil.gov.au/reports/200405-04.pdf>.

^[99] Goodman and Melinder, above n 85, 9–10.

^[100] Ibid 10.

^[101] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86, 187 (citations omitted).

^[102] Goodman and Melinder, above n 85, 10–13; Snow and Powell, above n 98, 7–8.

^[103] See, eg, Crimes Act 1914 (Cth) s 15YM(1); Evidence (Miscellaneous Provisions) Act 1991 (ACT) s 40F; Criminal Procedure Act 1986 (NSW) s 306U; Evidence Act 1977 (Qld) s 93A; Evidence Act 1929 (SA) s 34CA; Evidence (Children and Special Witnesses) Act 2001 (Tas) s 5; Evidence Act 1958 (Vic) s 37B; Evidence Act 1906 (WA) ss 106H–106HB.

^[104] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86, 187.

^[105] See, eg, NSWSCLJ Report, above n 19, 59–68; Attorney‑General’s Department (NSW), above n 57, 118.

^[106] Evidence to NSWSCLJ, Sydney, 19 April 2002, 5 (Judy Cashmore), cited in NSWSCLJ Report, above n 19, 63.

^[107] Eastwood and Patton, above n 5, 76.

^[108] Cashmore and Trimboli, ‘Child Sexual Assault Trials’, above n 92, 9, 14.

^[109] A total of 277 jurors from 25 juries completed a questionnaire in the jury deliberation room in the presence of court officers immediately after their verdicts had been delivered. Of the 300 possible juror participants, this represented a response rate of 92.3 per cent: ibid 4.

^[110] Ibid 7.

^[111] Ibid.

^[112] Ibid 11; see also at 10.

^[113] Ibid 10.

^[114] Ibid.

^[115] Ibid 9.

^[116] Ibid.

^[117] Eastwood and Patton, above n 5; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32.

^[118] Cashmore and Trimboli, ‘Child Sexual Assault Trials’, above n 92, 14 (citations omitted).

^[119] Zajac, Gross and Hayne, above n 2.

^[120] Ibid 201.

^[121] Ibid.

^[122] Ibid 201–2.

^[123] Ibid 202.

^[124] Ibid 203.

^[125] Ibid 204.

^[126] Ibid 203, 206.

^[127] Ibid 203–4.

^[128] Ibid 204.

^[129] Ibid 206.

^[130] Ibid 204. A further study by Zajac and Hayne showed that older children were also not immune from the negative consequences of cross‑examination style questions: see Rachel Zajac and Harlene Hayne, ‘The Negative Effect of Cross‑Examination Style Questioning on Children’s Accuracy: Older Children Are Not Immune’ (2006) 20 Applied Cognitive Psychology 3. The authors reported that ‘the cross‑examination interview significantly decreased the accuracy of

9‑ and 10‑year‑old witnesses who were initially highly accurate’: at 13. Over 40 per cent changed their initial correct responses about a staged event: at 12.

^[131] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86, 188.

^[132] See Mark Brennan and Roslin E Brennan, Strange Language: Child Victims under Cross‑Examination (2^nd ed, 1988) 7, 41–2, 59–77; Brennan, ‘The Discourse of Denial’, above n 2, 73–5; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 46–9.

^[133] Zajac, Gross and Hayne, above n 2, 207.

^[134] Ibid.

^[135] Ibid 206.

^[136] Ibid 207.

^[137] Ibid 207–8.

^[138] Ibid 208 (emphasis added).

^[139] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86, 188. The event involved a tour of a New Zealand police station during which the children had their right thumbprint recorded and ‘mug shot’ taken. They were also shown a jail cell and a police car with the lights and siren turned on.

^[140] Ibid 188.

^[141] Ibid.

^[142] Ibid 189. The eight month period is the average delay in New Zealand between when a child first reports and when the matter goes to trial.

^[143] Ibid 190.

^[144] Ibid.

^[145] Zajac, Gross and Hayne, above n 2, 204. See above n 130 and accompanying text.

^[146] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86, 190.

^[147] Ibid.

^[148] Ibid 191.

^[149] Ibid 192.

^[150] Ibid 190 (emphasis added).

^[151] Ibid 191.

^[152] Ibid.

^[153] Ibid 192.

^[154] Ibid 191.

^[155] See, eg, Cashmore and Trimboli, ‘Child Sexual Assault Trials’, above n 92, 14; Zajac, Gross and Hayne, above n 2, 204, 206; Zajac and Hayne, ‘The Negative Effect of Cross‑Examination Style Questioning on Children’s Accuracy’, above n 130, 12–13.

^[156] See Brennan and Brennan, Strange Language, above n 132, 3; Ellison, above n 13, 35.

^[157] Natalie Taylor and Jacqueline Joudo, ‘The Impact of Pre‑Recorded Video and Closed Circuit Television Testimony by Adult Sexual Assault Complainants on Jury Decision‑Making: An Experimental Study’ (Research and Public Policy Series No 68, Australian Institute of Criminology, 2005) 59–60, 67–8. This study involved a demographically diverse group of 210 mock jurors who were randomly assigned to 18 mock sexual assault trials, which used different modes of victim testimony (face‑to‑face, CCTV or video) and were conducted to simulate actual trial conditions. The jurors completed pre‑deliberation and post‑deliberation questionnaires: at 22–5, 78–92.

^[158] See Natalie Taylor, ‘Juror Attitudes and Biases in Sexual Assault Cases’ (Trends and Issues in Crime and Criminal Justice No 344, Australian Institute of Criminology, 2007) 2. Taylor commented that in the study she conducted with Jacqueline Joudo on juror decision‑making (at 4):

Even within the same jury, opinions about [the complainant’s] credibility differed, which could not be attributed simply to how she presented her testimony or what she said since all jury members watched exactly the same trial …

^[159] Natalie Taylor and Jenny Mouzos, ‘Community Attitudes to Violence against Women Survey 2006: A Full Technical Report’ (Violence against Women Community Attitudes Project No 1, Australian Institute of Criminology, 2006) 65.

^[160] Taylor, ‘Juror Attitudes and Biases in Sexual Assault Cases’, above n 158, 2.

^[161] Goodman and Melinder, above n 85, 2 (citations omitted).

^[162] Ibid.

^[163] Ibid 5 (citations omitted).

^[164] Ibid.

^[165] Paul Sheehan, Girls Like You: Four Young Girls, Six Brothers and a Cultural Timebomb (2006) 216.

^[166] Transcript of Proceedings, R v MSK (Supreme Court of New South Wales, Hidden J, 18 May 2005) 112, 143, 151, 156. See also ibid 219–28.

^[167] See Sheehan, above n 165, 231–4.

^[168] Ibid 231–5.

^[169] Ibid 217–56. This behaviour included numerous interjections as well as physical violence directed against the mothers of the victims and the jury: at 249–53.

^[170] See Douglas D Koski, ‘Jury Decisionmaking in Rape Trials: A Review and Empirical Assessment’ (2002) 38 Criminal Law Bulletin 21, 85, 89–91.

^[171] Taylor and Joudo, above n 157, 59–60, 67; Goodman and Melinder, above n 85, 5.

^[172] Ellison, above n 13, 34.

^[173] See above n 46 and accompanying text.

^[174] Judge Kevin Sleight, ‘Commentary on the Video “A Case for Balance” — The Issue of Judicial Control of Cross‑Examination’ (Paper presented at the Biennial District and County Court Judges’ Conference, Fremantle, Western Australia, 27 June – 1 July 2007) 3.

^[175] Cashmore, above n 31, 19–20, 33–4; ALRC and HREOC, Seen and Heard Report, above n 6, 346; NSWSCLJ Report, above n 19, 70–5; Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 51–3.

^[176] ALRC and HREOC, Seen and Heard Report, above n 6, 346.

^[177] Evidence to NSWSCLJ, Sydney, 19 April 2002, 5 (Judy Cashmore), cited in NSWSCLJ Report, above n 19, 71.

^[178] NSWSCLJ Report, above n 19, 71.

^[179] Pär Anders Granhag, Leif A Strömwall and Maria Hartwig, ‘Eyewitness Testimony: Tracing the Beliefs of Swedish Legal Professionals’ (2005) 23 Behavioral Sciences and the Law 709,

721–2.

^[180] Evidence to Crime Prevention Committee, Melbourne, 27 October 1994 (Judge Waldron), cited in Crime Prevention Committee, above n 32, 188.

^[181] Judge Roy Ellis, ‘Judicial Activism in Child Sexual Assault Cases’ (Paper presented at the National Judicial College of Australia Children and the Courts Conference, Sydney, 5 November 2005) 11.

^[182] See above Part V.

^[183] Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 54.

^[184] Eastwood and Patton, above n 5, 126.

^[185] ALRC and HREOC, Seen and Heard Report, above n 6, 347 (recommendation 112).

^[186] See below nn 193–204 and accompanying text.

^[187] Criminal Procedure Further Amendment (Evidence) Act 2005 (NSW) sch 1 item 4.

^[188] Evidence Act 1995 (Cth); Evidence Act 1995 (NSW); Evidence Act 2001 (Tas); Evidence Act 2008 (Vic). Section 4(1) of the Evidence Act 1995 (Cth) applies that Act to federal courts and the courts of the Australian Capital Territory. At the time of publication, the substantive provisions of the Victorian Act had not yet commenced.

^[189] NSW Adult Sexual Assault Interagency Committee, A Fair Chance: Proposals for Sexual Assault Law Reform in NSW (2004) 4, cited in ALRC, NSWLRC and VLRC, Uniform Evidence Law: Report, ALRC Report No 102, NSWLRC Report No 112, VLRC Report No 186 (2005) 153.

^[190] Evidence Act 1995 (Cth) s 41, amended by Evidence Amendment Act 2008 (Cth) sch 1 item 13; Evidence Act 1995 (NSW) s 41, amended by Evidence Amendment Act 2007 (NSW) sch 1 item 12; Evidence Act 2008 (Vic) s 41. The relevant section of Evidence Act 2001 (Tas) is also s 41, however this section has not been amended in accordance with the recommendations of the ALRC, NSWLRC and VLRC.

^[191] The Commonwealth and New South Wales amendments to s 41 of the uniform Evidence Acts are the same: Evidence Amendment Act 2008 (Cth) sch 1 item 13; Evidence Amendment Act 2007 (NSW) sch 1 item 12. Victoria, soon to be a uniform Evidence Act jurisdiction, has s 41(2) of the Evidence Act 2008 (Vic), which is as follows (emphasis added):

The court must disallow an improper question or improper questioning put to a vulnerable witness in cross‑examination, or inform the witness that it need not be answered, unless the court is satisfied that, in all the relevant circumstances of the case, it is necessary for the question to be put.

Evidence Act 2008 (Vic) s 41(3) replicates s 41(2) of the Commonwealth and New South Wales Acts. See below n 204 and accompanying text.

^[192] ALRC, NSWLRC and VLRC, Uniform Evidence Law: Report, above n 189, 151–2.

^[193] Evidence Act 1995 (Cth) s 41; Evidence Act 1995 (NSW) s 41.

^[194] Amended by Statutes Amendment (Evidence and Procedure) Act 2008 (SA) s 14.

^[195] New South Wales Bar Association, New South Wales Barristers’ Rules (2008) r 35A(b).

^[196] See New South Wales Bar Association, New South Wales Barristers’ Rules (2008) rr 35A, 35B. Both rr 35A and 35B were gazetted on 30 May 2008: New South Wales, Government Gazette, No 61, 30 May 2008, 4083.

^[197] Richard Ackland, ‘Things Become Very Heated at the Bar ‘n’ Grill’, The Sydney Morning Herald (Sydney), 25 April 2008, 33.

^[198] Evidence Act 1939 (NT) s 16(2); Evidence Act 1977 (Qld) s 21(2); Evidence Act 1929 (SA) s 25(4); Evidence Act 2001 (Tas) s 41(2); Evidence Act 1906 (WA) s 26(3).

^[199] VLRC, Sexual Offences: Final Report, above n 11, 314.

^[200] Ibid 316.

^[201] Ibid 318 (recommendation 143), 333 (recommendation 158).

^[202] Amended by Crimes (Sexual Offences) Act 2006 (Vic) s 38.

^[203] But see above n 188.

^[204] Evidence Act 2008 (Vic) s 41(2) (emphasis added).

^[205] ALRC, NSWLRC and VLRC, Uniform Evidence Law: Report, above n 189, 155.

^[206] VLRC, Sexual Offences: Final Report, above n 11, 296 (recommendation 137). The Commonwealth, New South Wales and Victoria have enacted these instructions as ss 13(5)(b)–(c) of the uniform Evidence Acts: Evidence Act 1995 (Cth) ss 13(5)(b)–(c), amended by Evidence Amendment Act 2008 (Cth) sch 1 item 3; Evidence Act 1995 (NSW) ss 13(5)(b)–(c), amended by Evidence Amendment Act 2007 (NSW) sch 1 item 3; Evidence Act 2008 (Vic) ss 13(5)(b)–(c). These instructions will be given by a trial judge where a person (often a child) is only capable of giving unsworn evidence. Cf Evidence Act 2001 (Tas) s 13(2), which remains unamended.

^[207] VLRC, Sexual Offences: Final Report, above n 11, 316.

^[208] Ibid 316–17 (citations omitted).

^[209] Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 51–3.

^[210] NSW Adult Sexual Assault Interagency Committee, above n 189, 3.

^[211] ALRC, NSWLRC and VLRC, Uniform Evidence Law: Report, above n 189, 147.

^[212] Ibid 147–8 (citations omitted).

^[213] Stephen Odgers, ‘Witness Questioning: The Role of the Judge and the Role of the Barrister’ (Lecture delivered at the Student Forum, University of New South Wales Law School, Sydney, 29 August 2008) (notes of speech on file with the author).

^[214] Henning, above n 3, 134. See R v TA [2003] NSWCCA 191; (2003) 57 NSWLR 444, 446–7 (Spigelman CJ).

^[215] ALRC, NSWLRC and VLRC, Uniform Evidence Law: Report, above n 189, 157.

^[216] Henning, above n 3, 136.

^[217] The legal process has a detrimental effect on a significant proportion of child complainants: Reid Howie Associates, above n 7, 6–7. In one Australian study, 6 out of 9 complainants interviewed in New South Wales, 12 out of 18 complainants interviewed in Queensland, and 12 out of 36 complainants interviewed in Western Australia reported detrimental effects on their education as a result of being involved in the legal process: Eastwood and Patton, above n 5, 68.

^[218] See, eg, Eastwood and Patton, above n 5, 59–62, 67–8.

^[219] Cashmore and Trimboli, ‘Child Sexual Assault Trials’, above n 92, 9.

^[220] Henning, above n 3, 136 (emphasis in original).

^[221] Cashmore and Trimboli, Evaluation of Specialist Jurisdiction Pilot, above n 32, 51–3.

^[222] NSWLRC, Questioning of Complainants by Unrepresented Accused Report, above n 29, 47–8.

^[223] Zajac and Hayne, ‘I Don’t Think That’s What Really Happened’, above n 86; Zajac and Hayne, ‘The Negative Effect of Cross‑Examination Style Questioning on Children’s Accuracy’, above n 130.

^[224] Snow and Powell, above n 98, 8–10, 34.

^[225] As described by Brennan, ‘The Discourse of Denial’, above n 2, 75–6, 86–8. Under the uniform Evidence Acts, a court is to disallow a leading question, or direct the witness not to answer it, where the court is satisfied that the facts concerned would be better ascertained if the leading questions were not used: Evidence Act 1995 (Cth) s 42(3); Evidence Act 1995 (NSW) s 42(3); Evidence Act 2008 (Vic) s 42(3).

^[226] See above n 34 and accompanying text.

^[227] Sleight, above n 174, 10.

^[228] Cf ibid. See generally Evidence Act 1995 (Cth) s 43; Evidence Act 1995 (NSW) s 43; Evidence Act 2008 (Vic) s 43.

^[229] Helen Dent, ‘The Effects of Age and Intelligence on Eyewitnessing Ability’ in Helen Dent and Rhona Flin (eds), Children as Witnesses (1992) 1, 6–11; Lane Geddie, Sasha Fradin and Jessica Beer, ‘Child Characteristics Which Impact Accuracy of Recall and Suggestibility in Preschoolers: Is Age the Best Predictor?’ (2000) 24 Child Abuse and Neglect 223, 225, 231–2; Simona Ghetti et al, ‘Consistency in Children’s Reports of Sexual and Physical Abuse’ (2002) 26 Child Abuse and Neglect 977, 987, 990.

^[230] There is provision for the use of an intermediary under the Youth Justice and Criminal Evidence Act 1999 (UK) c 23, ss 16–17, 29, which gives a court the power to order the examination of certain types of vulnerable witnesses (including all sexual assault victims) using an intermediary during the witness’s live or prerecorded evidence‑in‑chief, cross‑examination and re‑examination. In Western Australia, a person may be appointed by the court to communicate and explain the questions put to the child and to explain the evidence given by the child to the court: Evidence Act 1906 (WA) ss 106F(1)–(2). In the context of the Republic of South Africa, see Criminal Procedure Act 1977 (RSA) s 170A, which is a similar provision to those of the United Kingdom and Western Australia. In these jurisdictions, the intermediary acts as an interpreter for the child, rather than simply advising the court as to the appropriateness of questions. Recommendation 5 below is designed to provide a greater role for the intermediary than is envisaged in these provisions, which have been under‑utilised in the United Kingdom.

^[231] VLRC, Sexual Offences: Final Report, above n 11, 312.

^[232] Nicholas Cowdery, Submission No 27 to NSWSCLJ, Report on Child Sexual Assault Prosecutions, 13 February 2002, 11, cited in NSWSCLJ Report, above n 19, 64.

^[233] Evidence to NSWSCLJ, Sydney, 26 March 2002, 8 (Lee Purches), cited in NSWSCLJ Report, above n 19, 64–5.

^[234] Zajac, Gross and Hayne, above n 2, 207.

^[235] R v TA [2003] NSWCCA 191; (2003) 57 NSWLR 444, 446 (Spigelman CJ).

^[236] Eastwood and Patton, above n 5, 89–94; NSWSCLJ Report, above n 19, 25.