Rethinking cross-examination: Reducing retraumatisation for victims and survivors of sexual assault

Abstract

Over the last 40 years, complainants in sexual assault cases have consistently reported that defence questioning during cross-examination is one of the most retraumatising aspects of going to trial. The recent Australian Law Reform Commission (ALRC) inquiry into justice responses to sexual violence recommended education and training in trauma-informed, culturally safe practice for all legal actors and law students. Education and training forms just one part of the large ALRC package of recommended reforms. However, if implemented well, it may change the trajectory of a long history of ineffective reform efforts.

The prevalence of sexual violence in Australia is now a ‘national emergency’. ¹ In 2023, the highest-ever sexual assault prevalence rate was recorded (136 per 100,000), representing an 11 per cent increase on the previous year. ² Only 7.7 per cent of victim-survivors engage with the justice system and contact police ³ and, of these reports, 10 per cent or less result in a conviction. ⁴ For victims and survivors of sexual violence, the justice system offers little in the way of ‘justice’.

Against this backdrop, in 2024 the Australian Law Reform Commission (ALRC) was tasked with an inquiry into the justice responses to sexual violence. The ALRC was to take a trauma-informed approach to promote ‘just’ outcomes for victims and survivors, including minimising retraumatisation, ⁵ and build on prior reports in a ‘transformative’ way. ⁶ The final report, Safe, Informed, Supported: Reforming Justice Responses to Sexual Violence (the report) was tabled in the Australian Parliament in March 2025. ⁷

The report is over 600 pages long and covers many problematic features of the criminal justice system. One critical reform issue is ‘improper’ questions by defence counsel during cross-examination – that is, questions designed to insult or annoy or that are needlessly offensive. This is a key factor in complainant retraumatisation in the justice system, as well as case attrition, and low reporting rates. ⁸

This article discusses the longstanding problem of improper cross-examination questions and the ALRC’s core recommendations targeting the issue – education and training on trauma-informed, culturally safe practice. The article first provides an overview of how cross-examination impacts on victim-survivors and outlines previous attempts at reform. It then considers the complex nature of training for judges, defence counsel ⁹ and law students. ¹⁰ It posits that the prospects of gaining traction on improper questions will hinge on both judicial leadership and an upcoming generation of legal professionals who, from the earliest stage in their legal training, are informed about the need for trauma-informed, culturally safe practice. This leadership and training can then be applied – not just in the courtroom, but in all interactions with victim-survivors.

The brutal experience of cross-examination

Cross-examination by defence counsel aims to discredit the witness, reveal evidential inconsistencies or weaknesses, and suggest alternative explanations to the witness’ account. This can be difficult for anyone, but for traumatised sexual assault complainants, it can be particularly distressing. The intense focus on credibility can ‘mimic’ the experience of powerlessness in sexual assault, ¹¹ risking retraumatisation and exacerbating symptoms, such as post-traumatic stress disorder (‘PTSD’), anxiety and dissociation. ¹² Other factors that can contribute to the potential for retraumatisation include the deeply private subject matter, the extent of the trauma, that the complainant often has a prior relationship with the accused, and that sexual assault cross-examinations often last longer than in assault trials. ¹³

In 1980, in The Politics of Rape – A Feminist Perspective, Marjorie Levis opined: ‘It should be of major concern to us to change the situation for the individual rape victim and hopefully to end such statements as “I felt I had been raped all over again by the court”; “... if I had known, I wouldn't have reported it”; “... the trial was worse than the rape …”’. ¹⁴ Yet, by 1996, little had changed. The Heroines of Fortitude study exposed the deeply concerning level of complainant retraumatisation during cross-examination: 65 per cent of sexual assault trials were stopped due to distress, as ‘complainants dry-retched, claimed to feel nauseous in the witness box, were unable to answer questions or had to take regular breaks.’ ¹⁵

Fast forward to the 2020s, and multiple state and territory reports continue to describe cross-examination as ‘unnecessarily hostile’, ¹⁶ ‘brutal and apparently irrelevant’ or ‘badgering and intimidating’, ¹⁷ using ‘misconceptions about sexual violence’, ¹⁸ and using questions to ‘confuse or “shake” a complainant’s account’. ¹⁹ Indeed, one study of child sexual abuse trials shows that cross-examination today is ‘much worse’ than questioning during cross-examination in the 1950s. ²⁰

Reform efforts over the years

The ALRC’s Evidence (Interim) report in 1985 noted that the then state and territory legislation on improper questions provided limited protection for witnesses. ²¹ The ALRC’s 1987 final report therefore recommended a legislative discretion to disallow improper questions, or to inform the witness that the question need not be answered. ²² Almost a decade later, the Evidence Act 1995 (Cth) was enacted; s 41 provided a discretion for the court to disallow questions if they were misleading or ‘unduly annoying, harassing, intimidating, offensive, oppressive or repetitive’, and a discretion to take into account any relevant ‘characteristic’ of the witness in deciding the matter. ²³

In 2005, New South Wales (NSW) tightened the judicial discretion into a legislative duty to disallow. ²⁴ Subsequently, a joint report by the ALRC, NSW Law Reform Commission (‘NSWLRC’) and Victorian Law Commission (‘VLRC’) noted that the discretionary protections were ‘seldom invoked’, so the Evidence Act 1995 (Cth) was also tightened to a duty, with expanded witness factors and more detail on the types of improper question. ²⁵ It was hoped this would flag the provision for judicial attention and provide greater guidance. However, the change was resisted by some members of the legal profession, such as Victoria Legal Aid, which maintained that most counsel already ‘refrain[ed] from harassing and offensive questioning in criminal trials as such tactics generally alienate the jury’. ²⁶

A Model Uniform Evidence Bill (‘Uniform Evidence Law’) was subsequently endorsed by the Standing Committee of Attorneys-General in 2007, but has not been uniformly implemented. ²⁷ The VLRC recommended a discretionary approach, concerned that a duty might encourage excessive judicial interference, compromising the fact-finding process. ²⁸ By 2016, it noted that these concerns had ‘not materialised elsewhere’. ²⁹ Notwithstanding opposition from the Law Institute of Victoria, Victorian Bar and Criminal Bar Association, ³⁰ a legislative duty was introduced in 2018. ³¹

Despite 40 years of well-intended legislative tinkering, multiple studies, judicial papers ³² and the reports discussed earlier have consistently observed that s 41 of the Uniform Evidence Law and its equivalents have failed to protect witnesses from improper questions. The VLRC described the problem in a nutshell: ‘there is clearly a gap between what victims and the legal profession consider appropriate questioning’. ³³ This is highlighted by the Victorian Criminal Bar Association’s 2021 observation that there had been a ‘major cultural shift’: improper questions had ‘long gone and can no longer occur, as the law does not permit it’ and judicial officers ‘will not countenance’ it. ³⁴ Yet our 2023 integrative literature review, which underpinned the ALRC’s inquiry, found that cross-examination was still a ‘profound locus of retraumatisation’. ³⁵

Some have mused that the judicial reluctance to intervene might be due to legislative flaws, lack of awareness, or simple unwillingness to apply s 41 (potentially due to fear of appeal). ³⁶ However, there are no significant studies of judicial sentiment on the issue and it remains something of a 40-year mystery. It seems a large part of the problem is cultural, as one judicial officer observed: ‘[i]t’s very hard if you are all from the bar and you’ve got a bar culture, … obviously it’s not enough to give the judge power [to stop improper questions].’ ³⁷ And therein lies the challenge: changing bar culture.

Experts in leadership attest that cultural change can be extraordinarily difficult, ³⁸ but the ALRC’s recommended top-down and bottom-up approach may provide an answer. It builds on 20 years of previous reports that have identified education and training as a means of leveraging change. ³⁹ In view of past efforts, one might dismiss present recommendations as incapable of achieving the ‘transformative’ change the ALRC was tasked with. Yet there is now a groundswell of national and international recognition regarding the need for trauma-informed, culturally safe legal practice ⁴⁰ that adds both impetus and weight to the argument that effective implementation of the ALRC recommendations may generate traction. Of course, training measures will not be ‘transformative’ on their own but may represent a potential pathway to sustained improvement on one aspect of the many systemic failures.

Judicial education and training

Harvard Professor John Kotter noted that the top three steps for successful organisational change are establishing a sense of urgency, creating a vision, and forming a powerful guiding coalition. ⁴¹ So it makes sense, as previous reports have done, ⁴² for the ALRC to ‘start at the top’ – recommending judicial education and training. ⁴³ All magistrates, trial and appellate judges who sit on sexual violence matters must necessarily be included, because cross-examination before magistrates in committal proceedings is ‘often more rigorous and intimidating’ than at trial. ⁴⁴ And, anecdotally, one of the main reasons for trial judges’ reluctance to apply s 41 is fear of creating grounds for appeal (and an unreceptive appellate bench). ⁴⁵

The ALRC recommends the three judicial training organisations (National Judicial College of Australia, Judicial Commission of New South Wales, Judicial College of Victoria) be funded to provide education that encompasses rape myths and misconceptions, utilising research on trauma, memory, complainant behaviour, sexual offending, grooming and coercive control. This training should also encompass trauma-informed, culturally safe practices and be evidence-based and developed with experts, people with lived experience, sexual assault services, and representatives of groups disproportionately reflected in sexual violence statistics. This co-design and/or co-production will be important in producing programs that can create the necessary vision and sense of urgency; as Eastwood and Patton posited more than 20 years ago, a paradigm shift will only be possible where there is a ‘real recognition of the damage that can be done … through brutal cross-examination’. ⁴⁶

The ALRC anticipates that the above judicial education and training will go hand-in-hand with two more targeted recommendations: nationally harmonised adoption of s 41 of the Uniform Evidence Law, ⁴⁷ and additional judicial education on s 41, to ensure it is understood and applied consistently. ⁴⁸

Clearly, legislative harmonisation is a pre-requisite for nationally consistent treatment by judges who undergo the same nationally consistent education and training. But it is important not to underestimate the level of the challenge. Theoretical knowledge is one thing, but applying it in practice can be extraordinarily complex, given the legal and social matrix of a sexual assault trial involving multiple conflicting interests. This is exacerbated by the subjective nature of the assessment required by the legislative term ‘unduly’ (annoying, harassing, intimidating, etc) in s 41(1)(b) and the fact that it would be difficult, if not impossible, to arrive at any bright-line guidance – all the facts and circumstances of the individual case would (and must) continue to hold sway in judicial decision-making.

Perhaps the most that can be achieved by the recommended education process is creating Kotter’s vision and sense of urgency via a thorough grounding, possibly by lived-experience experts, in what it is like for a complainant on the witness stand. Indeed, the ALRC suggested that ‘realistic examples’ and experiential training would help judges identify when intervention was necessary and mandated. ⁴⁹ In this way, we might find in the ‘powerful guiding coalition’ a more consistent readiness to intervene. ⁵⁰

Of course, success will depend entirely on judicial sentiment towards government-funded education. Historically, it has been cautious, given concerns around judicial independence, including: (i) perceived adverse impacts on the separation of judiciary and government; and (ii) potential scrutiny of a judicial officer’s performance if evaluations proceed on an individual basis rather than an overall assessment of the judiciary. ⁵¹ For these reasons, conventional wisdom is that judicial education must be ‘judge-led’. ⁵² Nevertheless, if the ALRC’s recommendations are implemented, as noted, the development process will involve co-design and/or co-production with stakeholders. ⁵³ An understanding of current levels of judicial knowledge, sentiment and where perceived shortfalls occur would be an advantage in this co-design and/or co-production process.

Against this background, the ALRC, as an independent government agency, did not recommend mandatory judicial training. Similarly, state reports have not done so, noting that ‘neither the legislature nor the executive must dictate judicial education’. ⁵⁴ Generally it is left to heads of jurisdiction to require ⁵⁵ or encourage training. ⁵⁶ However, the ALRC does suggest training should be ‘strongly encouraged’ by heads of jurisdiction and (perhaps controversially) would require levels of attendance to be recorded and included in annual court reports. ⁵⁷ If the recommendation is implemented, this prospect of transparency may be sufficient to attract the desired participation levels.

Defence counsel

One ‘feminist’ defence counsel describes her duty in the words of Lord Brougham (1821) – that she must

save that client by all means and expedients, and at all hazards and costs to other persons … and in performing this duty [she] must not regard the alarm, the torments, the destruction which [she] may bring upon others. ⁵⁸

While this perspective may seem anachronistic, echoes of it remain at the heart of defence practice, as reflected in the Legal Profession Uniform Conduct (Barristers) Rules. A barrister must ‘promote and protect fearlessly and by all proper and lawful means the client’s best interests … without regard to … any consequences to the barrister or to any other person’. ⁵⁹

Barristers also owe an overriding duty to the court to act with independence in the interests of the administration of justice ⁶⁰ and a duty to the ‘opponent’ (in sexual assault trials, the State) not to make false or misleading statements. ⁶¹ Of course, defence counsel owe no duty to any witness (including complainants) in a criminal trial; witnesses appear to rank with the collateral damage that may be generated by ‘fearless’ representation, as long as it is independent in the administration of justice and not misleading.

Again, it will be no small task to shift this cultural positioning, or its metaphors of violence, steeped as they are in a long history of rarefied practice at the bar. This is also true for comparable jurisdictions such as Canada: Professor Elaine Craig’s study meticulously documents the ‘celebration of courtroom aggression as the indicia of a heroic criminal lawyer’. ⁶² If a sizeable proportion of the bar do not believe there is a problem in practice or a knowledge gap in training, there is little incentive to allocate otherwise billable hours to an exercise in perceived futility. Indeed, there is a growing defence and media narrative insisting that the pendulum of court reform has already swung too far towards the complainant’s interests, endangering an accused’s right to a fair trial. ⁶³

However, as discussed, the preponderance of literature over the past four decades, as well as the current state of the criminal justice response, demonstrate otherwise. In addition, there is growing judicial acknowledgment that a fair trial is not just about an accused’s rights: it must balance the interests of the accused and the complainant and the public interest in the rule of law. ⁶⁴ As Chief Justice Spigelman (as he then was) noted, effective restriction of improper questioning is ‘perfectly consistent’ with a fair trial – the criminal justice system is not a ‘forensic game in which every accused is entitled to some kind of sporting chance’. ⁶⁵

A fair trial requires that witnesses give their best evidence, to enable the jury or finder of fact (in a judge-alone trial) to properly weigh the evidence and arrive at a just outcome, in accordance with the law. Distressed witnesses do not give their best evidence. Thus, the literature increasingly conceptualises improper questions as a tool to generate unreliable ⁶⁶ or inaccurate evidence. ⁶⁷ Even 20 years ago, the VLRC observed:

Examples of defence counsel asking multiple questions designed to confuse witnesses or lead them into a statement which suits the needs of the defence are commonplace. It is a technique which accomplished legal counsel perfect. ⁶⁸

Rapid-fire or repetitive questioning on irrelevancies is another tactic which, ‘far from being an engine for the discovery of truth, produces a large amount of unreliable information’. ⁶⁹ Such tactics may be misleading, contrary to the administration of justice. ⁷⁰ If judicial intervention via s 41 is unsuccessful or absent, counsel might also be deterred by the prospect of regulatory intervention. Professor Craig suggests counsel engaging in improper questions be referred to the relevant Law Society or Bar Association for discipline:

Encouraging regulatory action to reduce the mistreatment of sexual assault complainants without creating a chilling effect on the ability of lawyers to advocate resolutely for their clients is a delicate calculation. Nevertheless, law societies bear some of the responsibility for doing this disciplinary work. ⁷¹

The prospects of such regulatory discipline are remote, although not impossible. ⁷² A duty to witnesses is highly unlikely to be countenanced, but a Hippocratic-like ethos of doing ‘no further harm’ would effectively underpin trauma-informed, culturally safe education and training. ⁷³ Thus, the ALRC’s recommended education and training may be a potential catalyst for change, or the beginnings of Eastwood and Patton’s anticipated paradigm shift. Certainly, a greater emphasis on the ethical obligation not to manufacture false or misleading evidence by way of improper questioning should be communicated. Yet such training seems at odds with the current drafting of the barristers’ rules that articulate the beating heart of the profession, and a centuries-old culture highly resistant to change. ⁷⁴

Perhaps surprisingly, the ALRC did not recommend mandatory training for defence lawyers, ⁷⁵ although this approach was adopted in recent reports in Victoria and Queensland. ⁷⁶ The VLRC heard evidence that mandated training could have unintended consequences, including reducing the availability of counsel (especially in regional areas) and being less effective (ie, undertaken simply as a box-ticking exercise). ⁷⁷ The ALRC echoed these sentiments, adding that there could be ethical and legal difficulties, ⁷⁸ as well as practical obstacles in imposing training requirements on all barristers and solicitors practising in sexual offences matters. ⁷⁹

Nevertheless, both Victorian and Queensland reports recommended mandatory training for Legal Aid-funded counsel, ⁸⁰ with the VLRC noting evidence from judges in Aotearoa New Zealand’s pilot specialist court that mandatory training was key: ‘even lawyers who resisted it found it helpful’. ⁸¹ The VLRC additionally recommended increased fees be paid to those who had completed the necessary training and accreditation, as an incentive. ⁸² In contrast, the ALRC did not recommend even limited mandatory training; it considered ‘strong’ encouragement from bar associations and law societies would suffice. ⁸³ This makes educating law students – generational change – even more important, as discussed below.

Unfortunately, the conduct of prosecutors is beyond the scope of this article. However, it is noted that multiple studies show questioning by prosecutors is generally less problematic than defence counsel, but can be an issue where they fail (often for tactical reasons) to object to improper questioning by defence counsel. ⁸⁴ Accordingly, the ALRC recommends mandatory education and training for prosecutors. ⁸⁵

Law students

There is immense value in law students being educated as to the necessity for, and practice of, trauma-informed, culturally safe lawyering – and not just for the sake of sexual assault complainants. There is a clear need for trauma-informed lawyering, given that around 75 per cent of Australian adults have experienced a traumatic event. ⁸⁶ Further, the concept of cultural safety has evolved to broadly encompass a variety of marginalised populations, including those with a culturally and linguistically diverse background – and more than a quarter of the Australian population were born overseas. ⁸⁷ Of the ALRC measures, this kind of succession planning is possibly one of the most fundamental reforms.

Previously, the Queensland report recommended the Attorney-General and relevant Ministers ‘request’ that the Law Admissions Consultative Committee (‘LACC’) ‘reconsider’ its prescribed areas of knowledge for admission to practise law, to include family violence and victim impacts:

Given the broad-ranging impacts these complex social issues have across so many areas of legal practice, the Taskforce is satisfied that more should be done to embed these topics within undergraduate law programs and practical legal training courses. Universities and schools of professional practice … are also equipped to develop and maintain courses that address the impact of laws and colonisation on Aboriginal and Torres Strait Islander peoples, Indigenous perspectives, and cultural competency. ⁸⁸

The report suggested that the Attorney-General and relevant Ministers encourage other Attorneys-General to make a similar request. ⁸⁹ The final report recommended such training be expanded to include ‘sexual violence as appropriate’. ⁹⁰

Despite this, the ALRC did not recommend mandatory education in trauma-informed, culturally safe practice in the Australian undergraduate curriculum; instead, it recommended that the LACC ‘should ensure’ this is included in discussions between the six peak bodies on curriculum reform. ⁹¹ Government-imposed mandatory educational content would face significant structural, institutional and political hurdles. While the federal government influences education programs via funding levers, universities are generally autonomous statutory bodies, so any government influence does not extend to dictating educational content. The law curriculum content reflects admission requirements regulated by LACC, the peak bodies, state and territory legislation, and admission boards. In any event, the legal profession is typically regarded as self-regulating; mandatory imposition of educational content would face political pushback from legal stakeholders.

Thus, the possibility for significant generational change hinges on a consensus between these six peak bodies. Given that undergraduate legal education is the vital ‘bottom-up’ component of the ALRC’s package, its importance cannot be overstated – particularly in light of the ALRC’s reluctance to mandate training for defence counsel. Reaching consensus will be a challenge, given the already crowded curriculum, but careful design of legal education content could also yield further benefits. For example, studies show that Australian law students’ distress levels are higher than medical students or the general population; ⁹² critical self-reflection on trauma could highlight the need to adopt positive wellbeing strategies and overcome student resistance to help-seeking. ⁹³ Further, a grounding in vicarious trauma would equip graduates with the knowledge and tools to engage in healthy help-seeking when in practice, given that practitioners – including judges – often experience significant burnout and secondary trauma. ⁹⁴

Conclusion

In addressing persistent complainant reports of improper cross-examination questions, the ALRC’s report recommends education and training in trauma-informed, culturally safe practice for judges, defence counsel and law students. Carefully designed education and training has the propensity to improve the wellbeing of legal actors, while furthering their understanding of the impacts of trauma more broadly for clients and complainants. Education and training forms just one part of the ALRC package of reforms to reduce victim-survivor retraumatisation during cross-examination, but – if effectively implemented – it is one that will hopefully go some way to changing the 40-year trajectory of law reform that has achieved limited traction to date.