1 Introduction
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 the prosecution witness tested by live cross-examination. (Ellison, 1999: 35)
However, this oral tradition within a culture of adversarialism combines to produce entrenched patterns of manipulating a vulnerable witness’s oral evidence through leading questions using complex vocabulary, sentence construction and syntax (Brennan, 1995; Hanna, Davies, Crothers, & Henderson, 2012; Davies, Henderson, & Hanna, 2010; Zajac, Gross, & Hayne, 2003), all of which has been described as ‘legitimated bullying’ (Henning, 2006: 136).
Apart from the requirement based on the old common law rule in Browne v Dunn (1893) 6 R 67 (HL) that a cross-examiner must actually put challenges to an opponent’s witness to allow that witness, in fairness, to respond, the defence is entitled to raise reasonable doubts, with historically few legal limits on the methods for doing so and no limits on the use of stereotypes or rape myths. In Chapters 4 and 5, research studies revealed that the more that jurors endorse rape myths which are based on common stereotypes, the more likely they will blame the complainant, perceive the defendant as low in culpability and return not guilty verdicts.
It is a long-held objective of cross-examination to use witness questioning to communicate with the jury rather than the witness (Rook & Ward, 2016: 1555), thus resulting in zealous control of a witness’s answers. This may be contrasted with the approach of forensic questioning which aims to obtain reliable evidence from a witness through open-ended, non-leading questions.
Because cross-examination is most likely to impact the quality of children’s evidence, this chapter begins with an analysis of the problems facing child complainants within adversarial trials, followed by an analysis of the effect of cross-examination on adult witnesses.
2 A Culture of Adversarialism: ‘Legitimated Bullying’?
Cross-examination of witnesses is based on the logically flawed assumption that cross-examination style questions will not affect the evidence of a witness who is telling the truth and that any change to a witness’s examination-in-chief is a change towards the truth. However, for children, cross-examination is that part of court proceedings where their interests and rights ‘are most likely to be ignored or sacrificed’ (Brennan, 1995: 73). Because of this, the CSA trial has been described as ‘a legally sanctioned’ forum (Eastwood & Patton, 2002: 4) in which children can be emotionally traumatised by the unregulated behaviour of defence counsel and by questions children do not understand or cannot answer (Australian Law Reform Commission and Human Rights and Equal Opportunity Commission [ALRC & HREOC] 1997: 343; Davies et al., 2010).
More so than adults, children, as a group, are disadvantaged by a criminal justice process that is not underpinned by trauma-informed principles (as discussed in Chapter 11), and, therefore, ‘does not allow their full and equal participation’ (Scottish Executive Central Research Unit, 2002: i) even though their evidence will be central to the prosecution’s case in a CSA trial.
Historical and recent analyses show that trials involving sex offences are governed by specific rules of evidence, warnings and methods of cross-examination that were originally based on cultural beliefs about women’s and children’s propensity for promiscuity and lying (Boniface, 1994; Cossins, 2000; NSW Department for Women, 1996; Smart, 1990). These beliefs are well entrenched not least because of the pronouncements by judges throughout the centuries1 of Lord Hale’s belief that rape is an accusation easy to make yet hard to prove (Hale, 1736/1971: 635). This view that has been used as the basis for justifying protections for the accused (Coyle, Field, Wilson, Cuthbert, & Miller, 2009; Criminal Justice and Sexual Offences Taskforce, 2006; Eastwood, Kift, & Grace, 2006) without concomitant consideration of the fair treatment of child witnesses.
Nonetheless, few reforms have recognised or addressed the inaccuracies produced during cross-examination, partly because of the way these prejudicial beliefs play out in the adversarial trial, particularly during cross-examination. While the last decade has seen major reforms in E&W and all Australian jurisdictions to address children’s vulnerabilities through the use of CCTV, remote rooms and pre-recorded evidence (Cossins, 2010a; Davies et al., 2010; Spencer & Lamb, 2012), the history of sexual assault law reform shows that the process of cross-examination process remains the most unregulated part of the adversarial trial (Caruso, 2012; Cossins, 2010a).
The centrality of cross-examination to the adversarial trial demonstrates the oft-repeated belief that it is the most effective method for testing a witness’s truthfulness and the accuracy of their testimony, though its actual effectiveness has been in dispute for many years (Davies et al., 2010). Yet the focus of the adversarial trial on two lawyers (prosecution and defence counsel) locked in verbal battle where one wins and the other loses is a verbal representation of a physical battle (Langbein, 2003) and is likely to have a considerable impact on the well-being of a child: ‘the contest between a child … and a defence advocate [will be] far from equal’ (Scottish Executive Central Research Unit, 2002: 16), with the child’s vulnerability and the adversarial nature of cross-examination combining to produce a relationship of power easily exploited by defence counsel.
As discussed in Chapter 11, this power relationship will be heightened where the traumatic effects of cross-examination trigger the feelings of powerlessness experienced by a child as a result of the original sexual abuse (Dulcan & Wiener, 2006; Westcott & Page, 2002). Indeed, the adversarial trial has no mechanisms for recognising the prevalence of PTSD suffered by sexually abused children (and adults) and, therefore, no ability to apply trauma-informed care principles to the adversarial trial (see Chapter 11).
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’ (Ellison, 1999: 35). However, the answers given by a child in response to cross-examination within the power relationship in an adversarial trial 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, as Ward v R [2017] VSCA 37 demonstrates below. While a child’s evidence during cross-examination may be the best evidence for the defendant, based on a review of relevant studies, the best evidence given by a child will be evidence that is not subject to standard cross-examination techniques.
Arguably, ‘the extent to which truth is prejudiced by the use of these tactics’ (Brennan, 1995: 73) is a question that must be addressed on the grounds 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.
3 The Impact of Cross-Examination on Children’s Evidence
cross-examination is one of the worst parts of testifying for children since it is stressful and anxiety-inducing (Eastwood & Patton, 2002; Plotnikoff & Woolfson, 2009, 2012; Zajac, Jury, & O’Neill, 2009) and may re-trigger the original emotions associated with being abused in children suffering from post-traumatic stress disorder (PTSD) and other trauma symptoms2;
children are often subject to aggression, humiliation, harassment and accusations of lying from defence counsel (Cashmore & Trimboli, 2005; Eastwood & Patton, 2002; Plotnikoff & Woolfson, 2009, 2012; Powell, Westera, Goodman-Delahunty, & Pichler, 2016; Victorian Law Reform Commission [VLRC], 2004);
the most damaging part of cross-examination for children is being accused of lying (Eastwood & Patton, 2002). In a 2002 report into the prosecution of CSA in NSW, the NSW Parliament’s Standing Committee on Law and Justice (NSWSCLJ) reported the views of many that the cross-examination of children was an abusive process, a fact also reported 15 years later by the Royal Commission into Institutional Responses to Child Sexual Abuse (2017b) in its Criminal Justice Report.
children are subject to complex, developmentally inappropriate and repetitive questioning and questions that are deliberately designed to confuse and create inconsistencies (Cashmore & Trimboli, 2005; Davies et al., 2010; Eastwood & Patton, 2002; Hanna et al., 2012);
the linguistic style of defence lawyers is least likely to match that of the child compared to other investigative professionals (Cashmore & Trimboli, 2005; Davies, Henderson, & Seymour, 1997; Davies et al., 2010; Plotnikoff & Woolfson, 2009; Zajac & Hayne, 2003; Zajac et al., 2003);
when comparing the professional manner of lawyers and interviewers towards children, only defence lawyers were rated by researchers as being aggressive, sarcastic or accusatory towards a child (Cashmore & Trimboli, 2005);
when comparing the fairness of all court participants towards them, defence lawyers were rated as the least fair by the 43 child complainants in Cashmore and Trimboli’s study (2005); and
compared to prosecutors, defence lawyers are more likely to use closed and leading questions resulting in the cross-examination effect, as discussed below, which has a dramatic impact on the accuracy of children’s evidence, including changes in testimony away from the truth.
[m]any survivors have told us how daunting they found the criminal justice system [and] … that they felt that they were the ones on trial. Some survivors told us that the cross-examination process was as bad as the child sexual abuse they suffered. Many survivors told us that they found the process re-traumatising and offensive.
heard from the families of young victims and victims with disability about the particular difficulties these victims face in giving evidence. Police and prosecutors have given us examples of complainants, especially children, breaking down during cross-examination, in some cases with the result that the prosecution has failed.
Overall, consultations and research by the Royal Commission (2017a: 75) concluded that, ‘at least in some cases, the way in which complainants are questioned by police, prosecutors and defence counsel has itself compromised their evidence’.
A recent development in relation to children’s evidence is that the inconsistencies produced during the cross-examination of children are being used by defendants to find appeals against convictions on the grounds that the ‘inconsistencies so damaged [the complainant’s] credibility and reliability that none of the convictions can stand’ (Ward [2017] VSCA 37, [3], Maxwell P and Redlich JA; see also R v W and M [2011] EWCA Crim 1926).
the cross-examination of a child complainant … is a task of considerable complexity, requiring great care and sensitivity. … The first challenge is to formulate questions which are age-appropriate … as a matter of basic fairness to the child witness … . Unless the questions are age-appropriate, answers which appear to create inconsistencies … are unlikely to serve the desired purpose of creating a doubt in the minds of the jury or … of persuading this Court that the jury ‘must have a doubt’ about the evidence. … [T]he child witness must be given the opportunity to say whether something they have said, and which the accused disputes, is true. This is the obligation of fairness … [under] the rule in Browne v Dunn. (Ward [2017] VSCA 37, [10]–[12])
The problem that few jurisdictions have addressed in terms of reform is that challenging a child witness using typical suggestive statements (such as ‘I put it to you that you XYZ did not happen’ or ‘I put it to you that your evidence is a pack of lies’) is bound to create ‘a significant risk’ that the child will accept such suggestions without understanding their meaning or the implications of agreeing with the cross-examiner (Ward [2017] VSCA 37, [12]). Thus, the rule in Browne v Dunn (1897) 6 R 67 can create unfairness in relation to child witnesses as discussed further below.
Most of the questions which produced the answers which were chiefly relied upon … constituted the putting of direct suggestions with an indication of the answer: ‘This happened, didn’t it?’ Or: ‘This didn’t happen, did it?’ The consequence of that is … that it can be very difficult to tell whether the child is truly changing her account or simply taking the line of least resistance.5
simply because a child gives evidence about a particular sexual episode … and the child is unable to confirm that [evidence] … that fact alone can’t result in the conviction being necessarily unsafe and unsatisfactory. One needs to look to the explanations. One needs look to the precision of the cross examination and whether or not it was recanted. If it was recanted, then clearly … the ground is made good. If it’s short of recantation and it can be explained by memory lapse … and it was not a witness who was not telling the truth at any time, then the conviction isn’t necessarily unsafe.6
[the complainant’s] memory lapse was explicable by reference to the rigours of cross examination, which the complainant was self-evidently finding stressful, as well as the difficulties and confusion inherent in the interpretation process. … [I]t was scarcely surprising that the complainant might struggle to recall events that occurred so many years before, having given an account of those matters in 2007, in the [recorded police interview], when his memory would have been fresher than it was by the time of the special hearing.8
was of such a questionable form as to leave us with a real doubt as to whether the appellant’s guilt has been properly established in relation to those matters. Having considered the evidence as a whole, … we are of the view that the verdicts … are, relevantly, unsafe and unsatisfactory, and cannot be permitted to stand.9
This case highlights the problem that arises where a child merely agrees with defence counsel’s suggestive, tagged statements. In other words, a child’s agreement is more likely to be the result of this subtle form of coercion than a statement of the truth, as discussed below. As such, the criminal justice system may be perpetuating injustice in the name of fairness to the accused ‘by taking unfair advantage of children’s emotional and intellectual stage of development’ (NSWSCLJ, 2002: 78). In fact, as the Ward case demonstrates below, because of age-inappropriate cross-examination, a child’s evidence is easily distorted and her/his credibility easily undermined since we know that jurors’ verdicts are influenced by the consistency of evidence and the credibility of the complainant (Cashmore & Trimboli, 2006).
for both types of attorneys, … the increased use of suggestive questions seemed to slightly increase the likelihood of [acquittals], particularly for defense attorneys. … Cases that ended in acquittals were characterized by lower rates of option-posing questions and higher rates of suggestive questions by defense attorneys out of all the questions posed to the child. (Klemfuss et al., 2014: 783–784; 786)
While children’s responses were neither directly nor indirectly related to case outcome, attorneys’ questions were … [which] suggest[s] that juries may be more sensitive to information provided by attorneys (both in statements and in questions to children) [than children’s responses]. (ibid.: 785–786)
Such a finding confirms the views that the purpose of cross-examination is to use the witness to make a series of declarations and statements (Caruso, 2012; Henderson, 2014) which conveys the defence case to the jury but is untested by the prosecution. Indeed, such a ‘finding is disturbing given that suggestive questions le[a]d to the most minimal responding and have been consistently shown to cause inaccurate responding, particularly in children’ (Klemfuss et al., 2014: 786).
In a later study, Mugno, Klemfuss, and Lyon (2016: 191) found that mock jurors rated the child more favourably during direct examination than cross-examination, with the ‘greater proportion of suggestive and leading questions’ during cross-examination probably explaining less favourable views of the child. Indeed, ‘the effect of questioning style is apparent even when responses from child witnesses are minimal’ which is likely due to the content of those questions, with jurors then ‘extrapolating this information to estimate the child’s credibility’ (ibid.: 193).
Some might consider these problems to be a feature of CSA trials before technological changes were made to the way children give evidence in court. Yet even with the advent of specially designed rooms from which evidence is given via CCTV, cross-examination is still the most traumatising aspect of giving evidence for children with the treatment of children by defence counsel and the styles and types of cross-examination questions remaining unchanged, at least in Australian jurisdictions (Cashmore & Trimboli, 2005, 2006; Eastwood & Patton, 2002; Royal Commission into Institutional Responses to Child Sexual Abuse, 2017a; Ward [2017] VSCA 37).
The damage that cross-examination can do to a child’s testimony, as well as their mental health, lies in the fact that, after disclosure, children may be involved with a range of professionals, including police, doctors, counsellors and child welfare agencies, who ‘convey … 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’ (NSWSCLJ, 2002: 61). By contrast, cross-examination ‘go[es] against everything else that a child has been told’ because s/he may be accused of lying, making up stories, wanting to cause trouble for the defendant or being coached by their mother or other adults (ibid.).
While there is a common belief (discussed below) that greater regulation by judicial officers will solve the problems associated with the cross-examination of children, ‘there is a wide variation as to what judges allow during cross-examination’ (Eastwood & Patton, 2002: 125), since some do not recognise age-inappropriate questioning. It will also be much more difficult for a judge to know when to intervene in a line of questioning that is intended to confuse or create inconsistencies in the evidence, rather than one that is aggressive or intimidating (ibid.; see also Davies et al., 2010).
This variation is likely to be even greater now in NSW since, with the appointment of only two specially trained judges who sit on CSA cases (see, further, Chapter 9, other untrained judicial officers still preside over the bulk of CSA cases in NSW.
There is also a conflict for judges between their role in controlling cross-examination and their duty to protect the rights of the accused which may produce judicial resistance to intervening in cross-examination. In 2002, the NSWSCLJ considered that the answer lay in imposing a duty on trial judges to protect witnesses under the age of 18 years and recommended that s 41 of the NSW Evidence Act 1995 be amended to impose a positive duty on judicial officers to intervene in relation to improper questions (for all witnesses). However, this reform has not had the desired effect. Anecdotal evidence (Cossins, 2009), as well as evidence from a study by Cashmore and Trimboli (2006) suggests that NSW judicial officers infrequently utilise s 41 to protect children from improper questions. As discussed later, Hopkins and Boyd (2010) made similar findings in interviews with practitioners.
While the Judicial Commission of NSW (2017 update) recommends that judges set ground rules for the developmentally appropriate cross-examination of child witnesses, there is no mandatory, legislative requirement to do so in NSW. By contrast, in E&W, ground rules hearings are now conducted in CSA cases where an intermediary has been appointed to advise the court on the communication abilities of the child complainant (see Chapter 9).
Indeed, the most common response about 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 (see Hopkins & Boyd, 2010; Caruso, 2012; Criminal Justice Sexual Offences Taskforce, 2006; Davies et al., 2010; NSWSCLJ, 2002). However, Cashmore and Trimboli (2006) found that educational packages for judges had no discernible effect on the cross-examination process in the CSA trials they evaluated. In fact, the study confirmed anecdotal evidence that judges are reluctant to intervene to protect a child witness from badgering during cross-examination.10
In the seventeen trials observed (from March to December 2004), judicial intervention was found to be least common ‘to protect the child witness from badgering or oppressive questioning, [or] to support and encourage the child’, with only 20.6% of judicial interventions being made to control cross-examination (Cashmore & Trimboli, 2005: 52; 54), compared to interventions for clarifying questions or answers.
Although judicial intervention varied considerably according to the trial judge, it did not vary according to the complainant’s age or the style of questioning by the defence. In fact, those judges who had received education packages about inappropriate questioning of children intervened less frequently to protect the child from badgering, compared to judges who had not received education packages (20 versus 40 interventions), or to support the child (4 versus 46 interventions) (Cashmore & Trimboli, 2006: 52), suggesting that the education received by judges had little or no impact on their control of proceedings.
A similar judicial reluctance to intervene to protect child witnesses was found by Davies et al. (2010) in a study of NZ cross-examination transcripts.
In a study of 120 transcripts of complainant’s evidence of CSA, Powell et al. (2016: 236) found that although judges intervened more often than prosecutors, the number of interventions by judges and prosecutors, in either evidence-in-chief or cross-examination did not differ according to the complainant’s age. As well, ‘the reasons for judges’ and prosecutors’ interventions did not differ according to complainant age’ while ‘judges and prosecutors intervened more about question form, followed by giving directions or asking questions, matters of law, complainant care and question manner’ (ibid.: 237).
the actual number of interventions for question form (172) was low compared to the total number of complex questions defence lawyers asked (432). This could be due to a range of factors, such as the judge not wanting to appear as though they were favouring the prosecution, or … [because] judges and prosecutors need up-skilling in how to identify the types of questions that are likely to decrease the accuracy of complainants’ responses. (ibid.: 239)
The study also found that despite the fact that ‘defence lawyers are frequently using questions that are based on stereotypes about complainant behaviour’, less than 1% of all interventions were related to question substance (ibid.: 239).
judges are more likely to intervene to prevent questions in cross-examination that are misleading or confusing than those that are annoying, harassing, intimidating, offensive, oppressive or insulting. (Royal Commission, 2017b: 85)
He also noted that court personnel ‘have a much higher threshold when assessing if a question is harassing or offensive than witnesses’ (ibid.).
to what extent is the right to cross-examine by the defence an absolute right?
what is the real purpose of cross-examination in the CSA trial?
how does the ‘strange language’ of the courtroom produce unreliable evidence?
what do studies about the impact of cross-examination on children’s evidence tell us?
why have reforms so far been inadequate?
what reforms could make demonstrable differences to improve the experiences and evidence of children (and sexual assault complainants generally)?
4 Attempts to Improve the Cross-Examination Process for Children
In 2015, the NSW Department of Justice (2015) released a consultation paper entitled, Children’s Champions & Pre-Recording of Evidence. However, the consultation questions were not unique, in that they had been considered by various government and non-government committees and researchers in both Australia and E&W. Thus, the consultation questions need to be understood within a framework of previous research.
For example, in 2010, the 492-page report of the National Child Sexual Assault Reform Committee (Cossins, 2010a) contained 34 recommendations for reform of the CSA trial. Three of these recommendations concerned the use of intermediaries and were based on the need to elicit the best evidence from a child complainant and to reduce the distress or trauma experienced by a child during the trial process.
- (i)
preventing the complainant from appearing at trial by pre-recording her/his evidence-in-chief and/or cross-examination (now available in NSW and E&W);
- (ii)
informalising the surroundings in which complainants give evidence (for example, the use of remote witness rooms now used in NSW and Western Australia);
- (iii)
preventing the defence from putting cross-examination questions directly to the child;
- (iv)
informalising the cross-examination process by allowing it to take place pre-trial under the control of a third-party examiner, such as an intermediary (see Chapter 9);
- (v)
introducing specialist child sex offences courts with trained specialist judges and prosecutors and 15 other features to reduce delays and produce cultural change.
Even with some or all of these reforms in place, without a cultural change in the form of a trauma-informed specialist court system (see Chapter 11), the key barrier that prevents children from giving their best evidence is cross-examination which involves the use of leading, confusing and suggestive questions designed to produce inconsistencies in a witness’s evidence. It also has the effect of establishing an unequal power relationship between child and examiner and the potential for eliciting inaccurate evidence on which jury decision-making is based.
5 Defence Cross-Examination Versus Forensic Interviews
a story to the jury by manipulating the witness into giving the desired evidence and by the advocate … convey[ing] certain messages to the fact finders during questioning. Traditionally, the advocate’s focus … is not really on the witness’s evidence but rather on what perception of it can be created. The witness is the medium but not necessarily the message. (Henderson, 2014: 97; footnotes omitted; emphasis added)
Leading and suggestive questions are known to be the most unreliable method for eliciting information from children (Davies et al., 2010; Lamb, Hershkowitz, Orbach, & Esplin, 2008; Snow & Powell, 2007; Powell, 2013). Since the basis of cross-examination is to put suggestions to witnesses that contradict their testimony, often using convoluted language and grammatical styles, it is surprising that the adversarial system places so much faith in cross-examination as a reliable way of testing the veracity of a witness, particularly child witnesses. Cross-examination of children highlights one of the great inconsistencies in the adversarial system since the type of suggestive and leading questions that would make a child’s forensic interview inadmissible on the grounds of unreliability is the very type of questioning adopted by the cross-examiner.
a leading question is not a question at all. It is a statement or suggestion, with an inquisitorial sentiment tacked to the end. … Leading is a euphemism for story-telling. It permits ‘editorial’ comments to be put in the form of questions. That style generally means the witness cannot properly respond to the comments. (Caruso, 2012: 217–218)
Because leading questions are the defence’s way of ‘communicating with the jury’, this results in questions ‘which confuse and unsettle the witness, since questioning is not designed with their understanding or even participation in mind’ (Caruso, 2012: 218). Even if a witness disagrees with a cross-examiner’s statement (‘You were drunk, weren’t you?’), the idea has been conveyed to the fact-finder without the fact-finder realising the statement may be nothing more than supposition based on rumour or stereotype. Such tactics are particularly destructive with a child or other vulnerable witness, as discussed further below.
The more unsettled the witness, the more likely the fact-finder will make a negative judgement of her/his credibility. Even Wigmore, who is often quoted for his famous phrase about the truth-seeking qualities of cross-examination, acknowledged that ‘[a] lawyer can do anything with a cross-examination … He may … do more than he ought to do … [he] may make the truth appear like falsehood’ (Wigmore, 1974: 32; cited in Bowden et al., 2014: 560).
In a study of transcripts of CSA trials in New Zealand, Davies et al. (2010: 352) found that, on average, 84% of cross-examination questions put to children were closed or leading. Also of concern was question complexity—‘nearly one in five questions … contained complex vocabulary (18%) and more than one in four contained two or more embedded clauses (27%)’, leading to clear incidences of miscommunication between child and lawyer, although sometimes children do not realise ‘they have misunderstood a complex question’ (ibid.).
Defence cross-examiners are, generally, not trained in the methods to elicit a child’s best evidence (compared to the training received by forensic interviewers). Indeed, defence counsel are unlikely to be interested in such methods since obtaining a child’s recantation of their allegations assists the defence case, as does obtaining inconsistencies and confusion. For example, cross-examiners are advised to frame questions ‘in such a way that they elicit either a “yes” or a “no”. A child will probably answer “yes” to a question that suggests a yes answer and “no” to one that suggests a no answer’ (Salhany, 2006: 103).
the truth being expressed is that believed … by the cross-examiner, rather than inquiry into the truth believed by the witness. … Leading questions advance the defence case through a technique which involves subterfuge and psychologically deleterious effects on a witness’s recall. (Caruso, 2012: 224)
A number of untested assumptions underlie the practice and role of cross-examination since there is no evidence to show that particular styles of cross-examination are effective at flushing out the dishonest (as opposed to the confused) child or adult witness. When it comes to children, ‘there is no simple “Pinocchio” test’ for determining whether a child is lying or telling the truth (Goodman & Melinder, 2007: 9). In fact, there is ‘no widely accepted criteria to differentiate true from false memory’ in the psychological literature (ibid.) while inconsistencies in details are not a reliable predictor of lack of veracity (Fisher, Vrij, & Leins, 2013).
6 The Right to Cross-Examine: An Absolute Right?
6.1 The Position in E&W
Because cross-examination is considered to be ‘the primary evidentiary safeguard of the adversary trial process’ (Ellison, 1999: 35), it is widely accepted that a defendant’s inability to effectively test the prosecution’s case will infringe the fair trial principle (Hoyano, 2014).11 Yet commentators are rarely able to articulate how that right would be affected if cross-examination was more tightly regulated.
to examine or have examined witnesses against him and to obtain the attendance and examination of witnesses on his behalf under the same conditions as witnesses against him.
The right to a fair trial embedded in the common law and the … [ECHR] art.6 is pre-eminent because it provides the platform for the vindication of all other legal rights. It is often described in absolutist terms, as ‘the birthright of every British citizen’, ‘fundamental and absolute’, and ‘axiomatic’. … Over the past decade, British and Strasbourg courts alike have lost sight of this premise. We have reached the point where a criminal trial held to have been unfair as a whole nonetheless can generate a ‘safe’ verdict under English law.
[t]he Court of Appeal has stated in the plainest of terms that the approach to cross-examination of children and vulnerable witnesses should be markedly different from the approach to non-vulnerable adults. … [T]he advocate must now adapt their cross-examination to the witness … to obtain reliable evidence whilst remaining fair to the defendant.
wasted time, repetition and prolixity. It is no part of the duty of counsel to put every point of the defendant’s case … to a witness or to embark on lengthy cross-examination on matters which are not really in issue. It is the duty of counsel to discriminate between important and relevant features of a defence case … and minor and/or unnecessary matters … We are not saying that it should become a routine feature … that judges should impose time limits for evidence in chief or cross-examination of witnesses. … But where … counsel indulges in prolix and repetitious questioning, judges are fully entitled, and … obliged, to impose reasonable time limits. (R v Butt (2005) EWCA Crim 805, [10], Dyson LJ)
[t]he entitlement to a fair trial is not inconsistent with proper judicial control over the use of time … [since] every trial which takes longer than it reasonably should is wasteful of limited resources. It also results in delays to justice in cases still waiting to be tried, adding to the tension and distress of victims, defendants, particularly those in custody awaiting trial, and witnesses. ([2003] EWCA Crim 1012, [37])
When the issue is whether the child is lying or mistaken …, it should not be over-problematic for the advocate to formulate short, simple questions which put the … the defendant’s case to the witness … . Aspects of evidence which … are believed to undermine the child’s credibility must, of course, be revealed to the jury, but it is not necessarily appropriate for them to form the subject matter of detailed cross-examination of the child and the advocate may have to forego … cross-examination which consists of no more than comment on matters which will be before the jury … from different sources. … [W]hen … the witness … is a child, … it should not take very lengthy cross-examination to demonstrate, when it is the case, that the child may indeed be fabricating, or fantasising, or imagining, or reciting a well rehearsed untruthful script, learned by rote, or simply just suggestible, or contaminated by or in collusion with others to make false allegations, or making assertions in language which is beyond his or her level of comprehension, and therefore likely to be derived from another source.12
This means that in E&W, wide-ranging, cross-examination has been limited in cases involving child complainants so that one senior advocate has advised colleagues that trial practices have changed to such an extent that it is ‘[n]o good resorting to old fashioned “rights,” [arguments]’ because judges are now trained in the matters affecting children’s evidence so that it is ‘best to focus on [the] most effective forensic approach in difficult circumstances’ (Dein, 2015, n.p.).
All witnesses … should be enabled to give the best evidence they can. In relation to young and/or vulnerable people, this may mean departing radically from traditional cross-examination. … [T]he court may … impose restrictions on the advocate ‘putting his case’ where there is a risk of a young or otherwise vulnerable witness failing to understand, becoming distressed or acquiescing to leading questions. … If the advocate fails to comply with the limitations, the judge should give relevant directions to the jury when that occurs and prevent further questioning that does not comply with the ground rules settled upon in advance.16
In R v JP [2014] EWCA Crim 2064, for example, the trial judge took unusual steps to prevent the cross-examination of a child complainant by defence counsel, steps that would be unprecedented in Australian jurisdictions and were not subsequently endorsed by the EWCA.
After visiting the nine-year-old complainant in a child witness waiting room, the judge announced to the court that ‘the child could not participate in cross-examination, however sensitively done’17 even though he had not undertaken a competency inquiry. Both counsel, who had also visited the child, protested, with defence counsel reminding the judge of the defendant’s right to a fair trial under Article 6 of the ECHR and ‘urg[ing] the judge to let him at least try and ask a few questions to see whether the witness could communicate’.18 The judge refused this request but allowed defence counsel ‘to prepare a document to be placed before the jury, containing any points the defence would have wished to have made to the witness’.19
Thus, the only evidence presented to the jury from the complainant was her video-recorded Achieving Best Evidence (ABE) interview. The defendant appealed on the grounds ‘that the judge erred in preventing cross-examination of the complainant so that the applicant did not have a fair trial and his convictions are unsafe’.20 In particular, he argued that it had been unjust for the defence to have been prevented from cross-examining and testing the Crown’s chief witness, particularly since the complainant showed no vulnerability apart from her age.
The prosecution conceded that the trial judge had made various procedural errors, including the requirement under s 27 of the Youth Justice and Criminal Evidence Act 1999 (YJCE) for a witness to be called for cross-examination where the recording of his/her ABE evidence is admitted as evidence-in-chief unless a special measures direction allowed cross-examination to be given ‘otherwise than by testimony in court’. The trial judge also failed to consider ss 53 and 54 of the YJCE Act which required him to test the complainant’s competency as a witness and to consider whether she would benefit from the appointment of an intermediary.21
It is now generally accepted that if justice is to be done to the vulnerable witness and also to the accused, a radical departure from the traditional style of advocacy will be necessary. Advocates must adapt to the witness, not the other way round. They cannot insist upon any supposed right ‘to put one’s case’ or previous inconsistent statements to a vulnerable witness. If there is a right to ‘put one’s case’ (about which we have our doubts) it must be modified for young or vulnerable witnesses. It is perfectly possible to ensure the jury are made aware of the defence case and of significant inconsistencies without intimidation or distressing a witness.22
Despite this radically new approach to the cross-examination of vulnerable witnesses, the EWCA in JP held that the trial judge had been wrong to disallow cross-examination of the complainant without undertaking a proper competency assessment conducted in the presence of both parties and without considering whether an intermediary would have assisted the complainant to give further evidence.23
reasonable time limits imposed by a trial judge do not infringe a defendant’s right to a fair trial;
there is no obligation on a trial judge to allow the defence to put their case in full to a vulnerable witness; and
a judge is duty-bound to intervene when questions are inappropriate.24
There must be fairness to all sides. In a criminal case this requires the court to consider a triangulation of interests. It involves taking into account the position of the accused, the victim and his or her family, and the public.25
Overall, there has been ‘a powerful push by successive Chief Justices to reform the management of child witnesses before and during the trial’ which has included a ‘series of judgments imposing judicial controls on cross-examination’ (Hoyano, 2015: 108–109) and ‘redefin[ing] the conventional understanding of the rules for cross-examining vulnerable people in criminal trials and deliver[ing] an exceptionally strong direction to advocates to improve their practice’ (Henderson, 2014: 94).
The decisions in the above EWCA cases reflect a number of policies regarding the treatment of vulnerable witnesses which have found expression in the Equal Treatment Benchbook 2013, the Criminal Procedure Rules 2015 and Criminal Practice Directions.26 In fact, Henderson (2014: 95) describes the EWCA as an ‘unlikely champion of reform’ which ‘challenges any pessimism about the willingness of the courts to tackle the unfairness latent in the heart of our fair trial protections’.
- (i)
‘miscommunication through the use of developmentally inappropriate language’;
- (ii)
the use of suggestive questions resulting in unreliable answers; and
- (iii)
confronting the witness with the defence’s version of events.
- (iv)
‘cross-examination on the same topic by multiple advocates’.
The effect of this level of control by courts in E&W ‘is a real challenge to the conventional [cross-examination] approach … since [i]t repositions the witness as a valued participant in a dialogue rather than a puppet in the lawyers’ presentation’ (Henderson, 2014: 97). Overall, the case law together with judicial training programmes have:
achieved improvements in an area so long resistant to change that some … had all but given up on the idea that reform was achievable whilst lawyers remained in the picture:
Barker has been a turning point of the whole thing. … Then Lord Justice Thomas and Lady Justice Hallet have come to training and said “get in there.” The sea change isn’t in the last ten years but in the last five years and Barker is the reason because that gave permission to everyone to behave differently. (Henderson, 2015: 201; citing a judge)
probably the most successful attempt to reform the culture of advocacy ever—but passive or even active resistance from an unconvinced frontline still poses a lingering threat to its long-term success.
Trial proceedings must ensure that a defendant’s Article 6 rights are not unacceptably restricted … [T]he Court has consistently assessed the impact that the defendant’s inability to examine a witness has had on the overall fairness of his trial … in order to determine whether the defendant’s rights have been unacceptably restricted … by having regard to such factors as the way in which statutory safeguards have been applied, the extent to which procedural opportunities were afforded to the defence to counter handicaps that it laboured under and the manner in which the proceedings as a whole have been conducted by the trial judge.27
when reviewing questions of fairness, … the traditional way in which the Court approaches the issue of the overall fairness of the proceedings … [is] to weigh in the balance the competing interests of the defence, the victim, and witnesses, and the public interest in the effective administration of justice. (Al-Khawaja [2011] ECHR 212, [146])
Greater control over suggestive questioning lessens the control that the cross-examiner has over the witness and hence the message they wish to send to the jury (Henderson, 2014: 98). Since manipulation of a witness is the goal of the cross-examiner, that alone justifies questioning the exact extent of the ‘right’ to cross-examine a witness for the sole purpose of changing a witness’s evidence to suit the defence case. Because of the suggestive nature of cross-examination, that right ought to be curtailed in relation to vulnerable witnesses in order to avoid unreliable evidence being given and to promote justice. Although ‘conventional cross-examination is not conceived of as a forensic examination’ for eliciting the truth, ‘[t]he startling aspect … is that it begins to look as if the Court of Appeal thinks that it should be’ (Henderson, 2014: 99).
Thus, changes endorsed by the EWCA to cross-examination mean that if cross-examination prompts changes to a complainant’s evidence, it must give a ‘guarantee of … reliability’ so as to produce reliable answers by vulnerable witnesses (Henderson, 2014: 101).
It is also important to consider criticisms of this greater level of control by the EWCA over cross-examination. In relation to the idea that defence advocates should be excluded from cross-examination, Hoyano (2015: 111) has observed that there is an assumption that ‘replacement cross-examiners … [are] capable of performing a more competent and fair cross-examination than the most highly trained barrister under close judicial supervision’. However, this chapter shows that cross-examination is far from a fair process for children and there is little competence associated with asking children confusing, suggestive and age-inappropriate questions.
has the cognition, and the independence of will, to reject questions which contain a suggestion or assertion of the existence (or non-existence) of a particular fact … [s]uch assumptions cannot be made in the case of a child witness. … [A] child’s vulnerability and the likelihood that they will assent to suggestive questions increase the risk that the child will give answers that are contradictory [to their previous evidence] … [F]airness dictates that counsel should only ask questions in a form that is appropriate to the age of the child.29
All in all, the EWCA has effectively abrogated the rule in Browne v Dunn (1897) 6 R 67 since, as discussed above, in Barker and Lubemba the EWCA accepted that a court can impose restrictions on the defence putting its case to a vulnerable witness in order to avoid misunderstandings, acquiescence to leading questions or distress to the witness. For example, in R v E [2011] EWCA Crim 3028, the EWCA stated that even though counsel had not been permitted to directly put the defence case to the child witness (such as asking, ‘You were never punched, hit, kicked as you have suggested, were you?’), this did not infringe the defendant’s right to a fair trial.30
The alternative to putting the defence case directly to a vulnerable witness is the practice of defence counsel making a statement to the jury about the matters the defence ‘would have put if the witness were not vulnerable’, coupled with a judicial explanation to the jury about the reasons for restrictions on cross-examination (Rook & Ward, 2016: 1562). Thus, alternative methods for presenting the defence’s case have been developed.
Overall, it appears that the fair trial principle has not been undermined by preventing cross-examiners from manipulating a child witness using unreliable methods such as leading and suggestive questions for the sole purpose of producing evidence that is compatible with the defence case. By focusing on the methods and reliability of cross-examination, the EWCA has raised the bar not only in terms of the quality and reliability of the evidence that is elicited from child witnesses but also the reliability of those methods.
At the same time, the EWCA has been at pains to emphasise the inviolability of a defendant’s rights under the fair trial principle.31 This means that the traditional, age-old methods of defence cross-examination have now been put under the microscope in terms of their reliability and have been found wanting.
6.2 The Position in Australia
Compared to E&W, Australia is a federal system, comprised of states and territories, with a Federal Government elected under the Australian Constitution to deal with Federal issues. Unlike E&W, Australia, does not have a Federal Bill of Rights so that Article 6 of the ECHR is not part of Federal law. Only three jurisdictions, Victoria, the Australian Capital Territory (ACT) and Queensland, have enacted Bills of Rights (Charter of Human Rights and Responsibilities Act 2006 [Vic]; Human Rights Act 2004 [ACT]; Human Rights Act 2019 [Qld]). All three contain wording that is similar to Article 6 of the ECHR.
For example, under s 25(2)(g) of the Victorian Act, a defendant in criminal proceedings has the right ‘to examine, or have examined, witnesses against him or her, unless otherwise provided for by law’, which amounts to a qualified right to examine prosecution witnesses but does not stipulate how that examination can be conducted.
Similarly, under s 22(2)(g) of the ACT Act, a defendant has a right ‘to examine prosecution witnesses, or have them examined’, although this right is not similarly qualified. While this provision also does not stipulate how that examination can be conducted, unrepresented defendants are not permitted to personally cross-examine a sexual assault complainant in the ACT, as discussed further below.
Despite the lack of a Federal Bill of Rights, it is clear from Australian common law that the rights of an accused under the fair trial principle are not absolute and are subject to ‘the interests of the Crown acting on behalf of the community’.32 Indeed, the concept of fairness is not fixed and immutable and ‘may vary with changing social standards and circumstances’,33 such that it is inextricably ‘bound up with prevailing social values’.34 The concept of fairness can take into account the interests of the victim,35 including the desirable goal of encouraging victims to report sexual offences to the police, as well as minimising the re-traumatisation experienced by sexual assault complainants during the trial process and achieving best evidence. Thus, protecting complainants from re-traumatisation does not necessarily result in a reduction in the rights of an accused, rather the introduction of a fairer system of justice for all concerned, which is ‘more conducive to eliciting the truth’ (Hoyano, 2015: 107).
In recent times, parliaments and other bodies have begun to accept that the right to cross-examination by the defendant is not an absolute right. For example, since September 2000 in E&W, s 34 of the YJCE Act has prohibited a defendant charged with a sexual offence from personally cross-examining a complainant.36 Similarly, s 35 prohibits a defendant charged with certain sexual offences, kidnapping, false imprisonment, abduction, slavery, human trafficking or child cruelty from personally cross-examining a child witness. In fact, the inroads into personal cross-examination by a defendant can be extended by a court in relation to other witnesses who might be affected by personal cross-examination under s 36 of the YJCE Act. Instead of personal cross-examination, s 38 allows the court to appoint a representative of an unrepresented defendant to undertake the cross-examination (see also Criminal Procedure Rules 2015, Part 23).
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 … potential witnesses are not bullied into giving untrue or inaccurate evidence, … [which would] undermine public confidence in the administration of justice. Without these protections …, the court would be an instrument of injustice … [T]o accommodate the accused’s wish to cross-examine the complainant personally is to confer an inappropriate advantage on the accused … 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 … [T]he benefit to complainants and to the community in general outweighs any perceived detriment to accused persons. (NSWLRC, 2003: 45–48; emphasis in original)
[t]he difficulties encountered by complainants in sexual assault cases … has 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.39
A compound question simultaneously poses more than one inquiry and calls for more than one answer. Such a question presents two problems. First, the question may be ambiguous because of its multiple facets and complexity. Secondly, any answer may be confusing because of uncertainty as to which part of the compound question the witness intended to address. But compound questions have additional vices. It is unfair to force a witness into the position of having to choose which questions in a compound question to answer and in which order. … To place witnesses in the position of having to reformulate a compound question and answer its component parts bit by bit is unfair to them in the sense that it prevents them from doing justice to themselves.41
There is, however, no way of knowing the extent to which judges in lower courts enforce this ‘general prohibition’ on compound questions. Indeed, the NSWCCA has said that ‘an occasional objectionable question in the course of a lengthy cross-examination rarely, if ever, gives rise to a miscarriage of justice’.42
Recently, the Victorian Court of Appeal has said that trial judges have ‘an obligation that the cross-examination is fair to the witness’43 in light of judges’ powers under s 41 of the Uniform Evidence Laws which operate in six Australian jurisdictions including NSW.44 It can be expected, however, unless specific legislation is enacted, the balance between inappropriate cross-examination of sexual assault complainants and the rights of the accused will favour the accused, as the Ward case, below, demonstrates.
Before considering whether or not the right to cross-examine ought to be subject to specific controls in child and adult sexual assault trials, it is necessary to examine the extent to which cross-examination is used as an oppressive tool for intimidation and confusion, rather than a forensic tool for exposing a dishonest witness. In other words, what is the extent of the problem and are reforms really necessary?
7 Juror Beliefs and Biases: The Impact of Rape Myths and Stereotypes During Cross-Examination
While each sexual assault trial will be unique, because of the particular facts of the case, the nature of the evidence, the complainant’s age and sex, the relationship between the complainant and the accused, and the complainant’s performance as a witness in the ‘theatre of the trial’, a fact-finder’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 complainant as an unreliable witness by tapping into common myths and stereotypes about the reliability of women and children, this may mean that a jury is more likely to acquit the accused.
[j]uror 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. (Taylor, 2007: 2)
[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. (Goodman & Melinder, 2007: 1)
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 creating ‘a confirmation bias that may result in faulty perception and incorrect interpretation’ (Goodman & Melinder, 2007: 2) on the part of the jury. Unfortunately, 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’ (ibid.). Yet the literature reviews in Chapters 4 and 5 suggest that jury decision-making is affected by ‘the tendency to deal with information [such as evidence presented in court] by incorporating it within an existing schematic framework’ (ibid.: 5).
A useful example of the ways in which defence counsel construct particular stories containing elements of the myths that are commonly associated with sexual assault is the published cross-examination of Tegan Wagner who was gang-raped by three brothers (known as MSK, MAK and MMK) at the age of fourteen. Although this case is more than 10 years old, there is still no legislation in NSW or any other Australian jurisdiction that restricts cross-examination based on commonly held myths about sexual assault.
When she was seventeen, Tegan faced almost 2000 questions over three days during her cross-examination by the three defendants’ counsel who were described as ‘three barristers operating as a hunting pack’ (Sheehan, 2006: 218–219) with ‘little intervention by the judge or the Crown’ (ibid.: 216).
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 didn’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 crotch area?
He put his hand on your hand and removed it from his crotch 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, it appeared that it was broken … .
You went from passionate, willingness, wanting, to hysterical crying because you thought, ‘My God, I could get pregnant!’ Correct?
For reasons of space it is not possible to include the cross-examination by the other two barristers (see ibid.: 231–234). At no stage did Tegan veer from her version of events. Despite the attempts to portray her 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 presented in court and the bizarre and disruptive behaviour of the three accused during the trial (Sheehan, 2006).
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 her testimony or retracts her complaint, or where issues, such as drunkenness and flirtatiousness, are not able to be denied.
8 The ‘Strange Language’ of the Courtroom: Are Leading Questions and Confusion the Way to the Truth?
Tagging: a statement with a question tagged on at the end which encourages the witness’s agreement with the tag, such as ‘it wasn’t him, was it?’ The tag encourages the answer ‘no’ but in order to disagree, the child would need to have the cognitive ability to understand it would be necessary to reply, ‘It was him’.
negative tagging such as, ‘Now this happened on a Friday, was it not?’ and the negative rhetorical question, such as ‘Now you had a bruise, did you not, near one of your breasts, do you remember this?’ (Brennan, 1995: 88; 75) are statements with a negative which creates a question. Both types are suggestive, although they ‘create the illusion of choice’ (Brennan, 1995: 75) by asking the witness to agree. The child must have the cognitive ability to dissect the question to be able to disagree with it.
the multi-faceted or compound question contains more than one proposition so that it is not clear which question should be answered. For example, ‘Did he take hold of you and make you do anything? Did he grab hold of your hand or do anything with your hand?’ (Brennan, 1995: 76). Where a child assents, ‘the control to interpret which sub question it most easily serves remains with the cross-examiner’ (Brennan, 1995: 76).
a question that lacks a grammatical and/or semantic connection. For example, ‘At any stage whilst you were in the bathroom did he ever enter the bathroom that previous week?’ The form of delivery implies the question makes sense, although ‘[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’ (Brennan, 1995: 77).
juxtaposition of topics: ‘[t]opics of unequal significance with no obvious sequential ties are placed alongside one another’ (Brennan, 1995: 78). For example: ‘Q: On that occasion when Mum went to, being that night that Mum [went] to Youth Group, you were at Clareville? . Q: You did not see the defendant at any time when he put his penis in your bottom, did you?’ (Brennan, 1995: 78). Although meanings between topics are not connected, they appear to be ‘of equal importance’ and serve ‘to create a “linguistic fog”’ (Brennan, 1995: 79), creating confusion and the risk that the child merely agrees with the suggestions put to them.
If these types of questions describe 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’ (Brennan, 1995: 71), the more so if they are re-experiencing any of the symptoms of trauma discussed in Chapter 11.
Although Brennan’s (1995) work is now more than 20 years old, subsequent studies and case law have confirmed the lack of linguistic match between cross-examiner and child complainants, as well as between cross-examiner and adult complainants (Andrews, Lamb, & Lyon, 2014; Bettenay, Ridley, Henry, & Crane, 2014; Cashmore & Trimboli, 2005; Hanna et al., 2012; Kebbell, Deprez, & Wagstaff, 2003; O’Neill & Zajac, 2013a, 2013b; Plotnikoff & Woolfson, 2012; Powell et al., 2016; Righarts, Jack, Zajac, & Hayne, 2014; Valentine & Mara, 2011; Westera et al., 2017; Zajac & Cannan, 2009; Zajac & Hayne, 2006; Zajac, O’Neill, & Hayne, 2012; Zajac et al., 2003).
For example, Davies et al. (2010) found that, on average, 84% of cross-examination questions in actual CSA trials in NZ were closed or leading, almost one-fifth of questions contained complex vocabulary, over a quarter contained two or more embedded clauses and defence counsel used significantly more double negatives than prosecutors.
defence lawyers did not adapt the number of questions asked to compensate for differences in age group (they used an average of more than 300 questions for each complainant);
prosecutors and police varied the number of questions according to the complainant’s age (an average of less than 100 for children/adolescents, compared to 300 for adults);
the types of questions varied as a function of the questioner;
prosecutors tended to ask closed questions and used leading questions as much as they used open questions. They used complex language in about a third of all questions, but they seldom used complex sense and repeat questions;
while prosecutors asked children and adolescents fewer open questions and more complex questions than they asked adults, they asked children and adolescents fewer repeated questions and gave more instructions than they did to adults;
more than 60% of defence lawyers’ questions were leading, less than 10% were open while 44% involved complex language;
judges tended to give instructions with closed questions, with leading questions being used in nearly one-fifth of all questions, while open questions were seldom used.
Judges used complex language in just over one-third of all questions, although they varied the types of questions they asked according to age. They asked children fewer leading questions than they asked adults, and more open questions than they asked adolescents and adults and gave fewer instructions than they did to adults.
Recent research also shows that young children have particular difficulty with indirect speech questions, beginning with ‘Do you know?’ which is a type of yes/no question. Children tend to answer yes or no but fail to elaborate on their answers, and, as a result, may be viewed as ‘an uncooperative conversational partner’ (Evans, Stolzenberg, Lee, & Lyon, 2014).
The most devastating cross-examinations of children are by clever barristers who do the ‘I’m your friend’ routine … [that is, a barrister] who is not overtly attacking the child, who is asking questions in a befriending manner but still using forms of questions that are apt to mislead or trick the child … In these cases, the questions are destructive but there is nothing ostensibly improper about them.
Furthermore, 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 effects of trauma and the instinctive compliance of trauma survivors with authority figures, especially when being questioned in a strange environment 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 (Goodman & Melinder, 2007). For example, an assent to a negative tag or negative rhetorical question does not necessarily mean the child agrees with the statement but may mean s/he does not have the developmental understanding to refute it. Yet the adversarial system has no method for ascertaining whether a child understood the question to which s/he has agreed and, therefore, no way of ascertaining the reliability of answers given in cross-examination.
It is also assumed that the dishonest witness will betray him or herself by their demeanour (Cashmore & Trimboli, 2006), especially under cross-examination. 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 laypersons at predicting veracity through observing a person’s demeanour such that both groups misinterpret behavioural cues at or below chance levels, that is, at or below 50% accuracy (Ekman, O’Sullivan, & Frank, 1999; Porter, Woodworth, & Birt, 2000; Vrij, Granhag, & Porter, 2010). Although Ekman et al. (1999: 265) found that some professionals, such as law enforcement officials and judges, have high levels of accuracy in lie detection, even in these groups there was ‘a substantial number who performed at or below chance’.
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’ (ALRC and the HREOC, 1997: 307). 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 (Saywitz, 1995; Spencer & Flin, 1993).
Leading and suggestive questions are known to be the most unreliable method for eliciting information from children (Home Office, 2011; Kuehnle & Connell, 2009; Lamb, La Rooy, Malloy, & Katz, 2011; Lamb et al., 2008; Snow & Powell, 2007). As a result, ‘researchers … have identified ways of interviewing children that maximize their accuracy and decrease their levels of stress while testifying’ (Zajac & Hayne, 2003: 187).
Phrase questions positively rather than negatively.
Use an active voice rather than a passive voice.
Separate questions on separate topics. Do not mix topics, or switch back and forth between topics.
Children’s conceptualisation of time, frequency and ordering of events is gradually acquired. … [P]rovide concrete anchor points, using times or events that are relevant to the child.
[U]se the child’s words to describe people, actions and objects.
Avoid ‘front loading’ questions that use a number of qualifying phrases before asking the crucial part of the question.
Use signposting. … Example: ‘I want to ask you some questions about your father’.
Discuss events in logical sequence.
Ask questions with the child’s point of view in mind.
Include only one query in each question.
Avoid questions that may be taken too literally.
identify vulnerabilities and witness limitations at the earliest possible stage;
carefully analyse the material that needs to be put to the witness/defendant;
use short questions and simple language, allowing adequate time for comprehension;
use a normal, non-aggressive tone;
avoid tag questions, double negatives and confusing or complex phrasing;
be aware of limitations in the witness’s/defendant’s grasp of abstract concepts such as time and the sequence of events;
identify areas that can be better put to other witnesses; and
identify areas of evidence that may be best dealt with not with the witness but by the parties putting before the jury agreed evidence from third parties (such as Social Services’ records).
While these guidelines are designed to achieve best evidence, they conflict with the practices of cross-examination. Although researchers warn against the use of multiple interviews of children, leading questions, tag questions and other complex questions because they increase the likelihood that a child will agree with an examiner’s suggestions or change their answers to please the interviewer (Goodman & Melinder, 2007; Lamb et al., 2008; Saywitz, Camparo, & Romanoff, 2010; Snow & Powell, 2007), in the adversarial trial, repetitive cross-examination and multiple cross-examination (where there is more than one defendant) is permitted, and leading questions are the order of the day in most Australian jurisdictions.
In Australia, there has been a move towards using a child’s investigative interview as his/her evidence-in-chief46 in order to improve the direct examination process but little or no real change has been made to counter the effect of cross-examination on children’s evidence, compared to the more radical, interventionist approach of judges in E&W, discussed above.
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. … [I]f I am defending a bloke I want to make life difficult for witnesses. … I’m not there to find the truth … no-one’s there to find the truth. (Eastwood & Patton, 2002: 76; quoting an unnamed defence lawyer)
In other words, with children, the technique is to extend cross-examination to make ‘them sweat a bit’ and ‘to put them under as much pressure as possible’ to make them ‘crack’ (Henderson, 2002: 279).
In 2018, a Melbourne defence lawyer admitted on Australian national television with tears in his eyes that he would not advise his own daughter to give evidence as a complainant in a sexual assault trial because of aggressive defence lawyers like himself.47 But aggression is not even necessary in order to change the evidence of a child as the Ward case [2017] VSCA 37 illustrates below.
8.1 A Case Study of the Cross-Examination Effect: What Price Fairness?
At the age of 68 years, Mr Ward (a pseudonym) was convicted of six counts of sexual assault against J who was between six and seven years of age at the time of the offences. These offences included showing J pornographic videos, exposure, inciting a child to commit indecent acts and various acts of sexual penetration. Ward had befriended J’s family after they emigrated to Australia in 2013 and leased two rooms in his house to the family. Ward was frequently J’s baby-sitter while her parents worked on Sundays.
On appeal, Ward argued that each verdict of guilty was ‘unsafe and unsatisfactory’ because of the inconsistencies between J’s recorded police interview (known in Victoria as Visual and Audio Recording of Evidence or VARE) and her answers during a special hearing for cross-examination. In a detailed and well-researched judgment, the Victorian Court of Appeal (VSCA) concluded that ‘none of the apparent inconsistencies were such as to preclude the jury from being satisfied beyond reasonable doubt that J was a witness of truth’.48
J’s VARE interview, which was clear and concise in response to open-ended questions by the interviewer, described ‘sexing’ videos in which Ward ‘showed me a lot of girls eating his penis’ and ‘he always want me to do whatever the girls in the [video]’.49 During her cross-examination, which took place over two days and eight months after her VARE interview, J gave some of her answers in English and some through an interpreter. She had viewed her VARE before being cross-examined.
It is possible that having English as a second language exacerbated the cross-examination effect since J appeared to contradict herself when it was put to her in ‘a series of propositions’ in the form of ‘[RW] never touched you’ since each time she answered ‘No’.50
a danger of a child witness wishing simply to please. There is undoubtedly a danger of a child witness seeing that to assent to what is put may bring the questioning process to a speedier conclusion than to disagree.51
[RW] never showed you rude things on his phone, [did he]?
Yes.
Okay. What did he show you on his phone?
Naked women and naked men.
And how long did you look at the video for?
Just one hour.
An hour, okay?
Yes.
Where were your mum and dad?
I don’t remember.
Okay. Was your sister at home?
No.
What was happening on the video?
I don’t remember.
Can you remember if people were doing anything on the video?
No.
Did [RW] ever touch you?
Only hug me.
Only cuddling, okay. So [RW] never touched you on your pee pee? [negative rhetorical]
No.
You never touched [RW’s] pee pee, did you? [negative tagging]
No.
Did [RW] ever touch you on the bum?
No.
Did you ever see [his] penis?
Yes.
And was that when he was outside having a pee?
No, in his bedroom.
Okay. And did [RW] ever put his penis near your pee pee?
No.
…
Did you ever touch his penis?
No.
Are you’re sure about that?
Yes.
the nature of the questions and their context can affect the significance to be attached to the answer. … As the Crown submitted, it is entirely plausible that a child of eight years old, being questioned through … an interpreter, did not think that, when RW placed her hands on his penis, she was ‘touching’ him. It was, afterall, his action, not hers. He was forcing her to do something she did not want to do.54
However, the answer is probably more complex since the cross-examination effect, discussed below has been shown in several studies to change the accuracy of children’s evidence. This phenomenon would explain J’s impaired ‘ability to comprehend the questions asked over a lengthy period of cross-examination’.55
Did [RW] ever touch you on the bum?
No.
[Did RW ever put his pee pee near your bottom57]?
Yes.
Can you tell me where you were when this happened?
In his bedroom.
Where was he?
On the bed.
Okay. What were you wearing?
Pyjamas.
Okay. And what did he do with his penis?
He took my pyjamas off and put it inside.
Put it inside where?
My bum.
[H]is penis didn’t go in your bum, did it? [negative tagging]
I don’t remember.
…
So he never put his penis in your bum again, you say? [negative rhetorical]
No.
…
And … granddad has never put his penis in your bum? [negative rhetorical]
Only once.
So was that the only time you say that [RW] put his penis in your vagina?
Yes.
Are you sure he put his penis in your vagina … ?
Yes.
So if I said to you he didn’t put his penis in your vagina, would I be wrong? [negative rhetorical]
I don’t remember.
Okay. Can you ever remember a time where [RW] touched your bum?
No.
So he never put his finger in your bum? [negative rhetorical]
I don’t think so.
As the VSCA recognised,60 these questions jump from topics about J’s vagina to questions about RW touching J’s bottom, with most questions containing either negatives or double negatives, with inherent comprehension difficulties for an eight-year-old child (see Sas, 2002).
[G]randdad has never asked you to eat his pee pee, has he?
No.
And [RW] or granddad has never put his penis in your bum?
Only once.
…
And granddad has never put his penis in your mouth?
INTERPRETER: I think she said ‘only once, ‘no only once’.
Okay. When did he put his penis in your mouth?
When I was in my room.
In your room. And what happened?
He came in my room.
Yes.
INTERPRETER: I am just asking her to tell me in [language], sorry.
Yes?
I don’t remember.
Okay. So you don’t remember him putting his penis in your mouth?
I don’t remember.
The VSCA noted that ‘[a]t this stage of the cross-examination it had been some time since J had had a break. It appeared that she was relatively restless’61 which suggests J had lost concentration. During the trial, it also appears that there was little or no intervention from the trial judge regarding the inappropriateness of the questions put to J and no recognition of the effect of the cross-examination on her emotional state.
Although J had initially answered that RW had put his penis in her mouth, it appears that J changed her answers due to the repetitive nature of the questions about the topic, a well-documented phenomenon in the literature (as discussed below) since her VARE interview described in detail how she was shown a pornographic video of women ‘eating’ men’s penises and that RW had asked her to do the same to him. Indeed, how else would J know about the video and why would someone show a child such a video?
[O]n the day that you got shown the video … he didn’t put his penis in your mouth, did he?
No.
He’s never put his penis in your mouth?
No.
this specific incident to a general question. … [I]t is unrealistic to suggest that she should have appreciated that counsel had moved away from [the video] occasion and was now putting to her that RW had never, on any occasion, placed his penis in his mouth. … [With submissions from the prosecutor], the jury were alive to the view that some answers were no more than inaccurate contradictions, resulting from misunderstanding or a willingness to agree with suggestions. Only non-leading questions … would have eliminated the risk that apparent inconsistencies would be seen by the jury as inconsequential or inaccurate.62
While the VSCA dismissed Ward’s appeal, the Court questioned the application of the rule in Browne v Dunn (1897) 6 R 67 when the witness is a child, and concluded that the application of the rule in the cross-examination of a child would create unfairness because of the suggestive nature of Browne v Dunn challenges.
- (i)
abolition of the requirement for children to give evidence in committal proceedings;
- (ii)
use of the initial police interview as the child’s evidence-in-chief in court (to avoid repeated interviews of children);
- (iii)
pre-trial pre-recording of the child’s cross-examination at a special hearing;
- (iv)
presence of a support person; and
- (v)
allowing the child to view their police interview before being cross-examined.
It is clear, however, that these reforms, as important as they are, have not addressed the central problem in CSA trials—the cross-examination effect which is discussed below.
The mismatch between the ‘strange language’ of the courtroom (Brennan, 1995) and the cognitive abilities of children reveal that defence counsel deliberately use strategies to take advantage of a child’s lack of comprehension in order to manufacture inaccuracies and inconsistencies in a child’s evidence. The inequality between cross-examiner and child due to this mismatch gives rise to profound unfairness within the CSA trial.
The fact that barristers and judges are aware that the adversarial system is not about the truth and is designed to make life difficult for witnesses, no matter how psychologically vulnerable, should ring alarm bells for the administration of justice and the welfare of children. Nonetheless, it is important to engage with the research literature to better inform the criminal justice reform agenda, in terms of measures that might reduce or eliminate the cross-examination effect on sexual assault complainants’ evidence.
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. (ibid.: 9)
Overall, this study provides empirical evidence to confirm what has long been suspected—that there is a positive correlation between jurors’ perceptions of complainant consistency, credibility and verdict. More recently, a study by Pichler et al. (2019) confirmed that inconsistencies in evidence are associated with acquittals and case attrition in CSA trials.
There is, however, a distinct mismatch between what lawyers and jurors believe children’s inconsistency indicates (unreliability and lack of veracity) and 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’ (Cashmore & Trimboli, 2006: 14), particularly in a context where CSA trials permit the very types of questions that forensic interviewers are trained to avoid.
While jurors may believe that consistency and reliability are intertwined they will have little understanding of the ways in which inconsistencies are deliberately created by skilful cross-examiners who exploit children’s vulnerability using questions that are deliberately crafted to create confusion and inconsistent answers.
While courtroom questioning styles in sexual assault trials are premised on the historical view that the evidence of children, and sexual assault complainants generally, is inherently unreliable (ALRC & HREOC, 1997), appeal courts in Australia and E&W have begun to accept that children’s evidence is no more unreliable than that of adults (Judicial College, 2018).63
Nonetheless, the majority of CSA trials in Australia are conducted without intermediaries (discussed in the next chapter) to bring about cultural change in the form of monitoring the age- and developmental appropriateness of cross-examination by defence counsel. As well, adversarial trials are not trauma-informed, as discussed in Chapter 11, such that the cross-examination of children would be categorised as emotionally abusive in any other context.
More importantly, the lack of regulation of cross-examination leads to an inequality of comprehension and resources between child and cross-examiner whose aim is to create inconsistencies in the child’s evidence. Defence counsel then argue that any inconsistencies in a child’s evidence indicate the unreliability of her/his evidence. More likely, however, these inconsistencies arise from the use of the strange language of the courtroom and an environment of intimidation which creates ‘confusion; lack of understanding …; [an] inability to process what is being put; acquiescence; and an overwillingness to agree with suggestions’.64 Overall, CSA trials are inherently unfair to child witnesses (Judicial College, 2018).
Even though children are more likely to give evidence using special measures to protect them from the rigours of adversarial processes, the adversarial trial is rigorous because of cross-examination which may go on for hours or days and may be ‘searching, aggressive or intimidating’,65 or overtly abusive, with defence counsel using whatever methods are permitted by the trial judge.
9 Studies on the Types of Questions Used in the Cross-Examination of Children
Ward’s case illustrates why it is important to understand the cross-examination effect on children’s evidence and the effects of trauma on a child’s ability to give his/her best evidence.
Several empirical studies from the last two decades are relevant to informing the reform agenda in relation to the cross-examination of children. While cross-examination is an integral aspect of adversarial trial practices, it is based on the logically flawed assumption that its methods will uncover an untruthful witness and will have no affect on the evidence of a truthful witness. For example, various studies of question type have found that cross-examiners use questions that are known to restrict a witness’s answers, interfere with their concentration and memory and to confuse their understanding of what is being asked. Such questions include closed, leading and yes/no questions, as well as those containing negatives and multiple questions, as seen in the Ward case above.
In a study of six rape trials, Kebbell et al. (2003: 54) found that open questions were rare, ‘closed [and leading] questions were more frequent, and yes/no questions were most frequent of all’. Since most questions were yes/no questions, witnesses had no choice but to comply with the constraints imposed by these questions. Their suggestive nature meant that there was no way of knowing whether a witness’s answers to those questions were the truth or the result of suggestion.
Indeed, if a witness’s particular recollection of events ‘does not further the lawyer’s cause … [then] lawyers are not necessarily interested in eliciting complete, accurate recollections’ (Kebbell et al., 2003: 56). They are more interested in promoting their client’s case, with little or no incentive to reduce the adversarial nature of their cross-examination unless mandatory controls are introduced. In other words, ‘[b]y controlling the evidence [produced] …, the lawyer becomes the narrator with the witness relegated to agreeing or disagreeing … [so that] cross-examination is a speech, just like an opening or closing’ (Hanna et al., 2012: 53).
In order to compare the questioning styles of defence lawyers and prosecutors, Zajac et al. (2003) made a surprising finding when they undertook an analysis of 600 examination-in-chief and re-examination questions and 2935 cross-examination questions involving child complainants in NZ between the ages of 5–13 years old (Zajac et al., 2003: 201).
When assessing whether the styles of questioning used by prosecutors and defence lawyers ‘differed as a function of the child’s age’, they found that defence lawyers asked a higher proportion of complex, grammatically confusing, credibility-challenging, leading and closed questions compared to prosecutors (Zajac et al., 2003: 202; 203).
76% 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. (ibid.: 204)
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. (ibid.: 188)
Overall, the results reported by Zajac et al. (2003) confirm that there is ‘a clear mismatch between courtroom language and the language capabilities of the children being questioned’. This was particularly so for children under the age of 10 years who were less likely than older children to seek clarification to complex, confusing or ambiguous questions, possibly because ‘children below the age of 8 years cannot reliably detect [their own] non-comprehension’ (ibid.: 207). 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’ (ibid.: 206).
Children’s responses depended on the questioning style of the lawyers and the types of questions asked which was encouraging because ‘it suggests that cross-examination can be made more appropriate for children merely by changing the types of questions deployed’ (ibid.: 207–208).
A further study by Zajac and Hayne (2006) showed that older children were also not immune from the negative consequences of cross-examination style questions. The authors reported that ‘the cross-examination interview significantly decreased the accuracy of 9- and 10-year-old witnesses who were initially highly accurate’, with over 40% changing their initial correct responses about a staged event.
If he told the Police that that was what he thought you wanted to do, are you saying that you don’t think he could have thought that? [to a 16-year-old witness]
I’ll put it to you that he at no time told you not to say anything to anyone. [age not specified]. (ibid.: 535–538)
[c]hildren were more likely to give elaborate responses (more than merely assenting, dissenting, or picking an option to a forced choice question) to prosecutors, and more likely to give unelaborated yes-no responses to the defense.
the complainant’s credibility, such as accusations of lying, false allegations and previous dishonesty;
the complainant’s reliability, such as accuracy of evidence;
the plausibility of the allegations, often based on myths around sexual assault such as absence of resistance, or delayed complaint; and
the complainant’s consistency by raising evidence of inconsistencies (often about minor details).
The authors also reported that the raising of inconsistencies with complainants was frequent but that defence counsel raised complainant inconsistencies with peripheral matters more frequently than inconsistencies regarding issues central to proving the offence (98.4% vs 72.4%) (Powell et al., 2016: 226). For complainants under the age of 18 years, ‘as the number of questions asked by police in the pre-recorded interview increased, so did the lines of questioning on inconsistencies in cross-examination’ (ibid.: 230). Better quality police interviews resulted in fewer inconsistencies being raised at trial (ibid.: 231).
Frequency of leading and closed questions during cross-examination by different examiners
Question type | Prosecutors (%) | Defence counsel (%) | Forensic interviewers (%) | Study |
|---|---|---|---|---|
Leading/closed | n/a | av. 78.5 | n/a | Davies and Seymour (1998) (NZ) |
n/a | 65 | n/a | Zajac et al. (2003) (NZ) | |
Closed Yes/no leading Other leading | n/a | 10.3 81.8 41.2 | n/a | Kebbell et al. (2003) (England) |
Leading/closed | n/a | 89 | n/a | Zajac and Cannan (2009) (NZ) |
Leading/closed | 52 | 84 | 26 | Hanna et al. (2012) (NZ) |
Closed Leading | 52 15 | 44 39 | Stolzenberg and Lyon (2014) (Calif, USA) | |
Closed Leading | 44 24 | 28 >60 | Powell et al. (2016) (Aust) | |
1950s trials Closed Leading | n/a | 30a 65a | n/a | Westera et al. (2017) (Aust) |
Twenty-first-century trials Closed Leading | n/a | 25a 70a | n/a | Westera et al. (2017) (NZ) |
- (i)
declaratory and suggestive questions that require a yes or no answer;
- (ii)
questions that are beyond children’s developmental abilities; and
- (iii)
questions that prevent the ability to put alternative versions of events.
10 Studies on the Impact of the Cross-Examination Effect on Children’s Evidence: Inconsistent Evidence or False Inaccuracies?
The fact that 76% of children changed their testimony during cross-examination in the study by Zajac et al. (2003), discussed above, led the authors to ask whether or not the changes made by children constituted a change towards the truth or away from the truth. To answer this, they examined the effect of cross-examination on five- and six-year-olds’ reports of a contrived event in which the children actually participated.66
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 (Zajac & Hayne, 2003: 188). All 46 children were interviewed six weeks after the event using open-ended questions and then more specific questions about two actual things that had happened (photo taken, police car) and questions about two things that did not happen (trying on handcuffs, report of a stolen bike). Eight months67 after this direct examination, they were interviewed by a different interviewer using cross-examination style questions and language in order to persuade the children to change their accounts of what happened at the police station.
Zajac and Hayne (2003: 190) reported that 85% of children ‘changed at least one of their original responses during cross-examination and one-third … changed all of their original responses’, with no significant differences in the number of changes made between the control group and the misled group.
When comparing their accuracy, the misled group was significantly less accurate during direct examination than the control group yet ‘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’ (ibid.: 190). In other words, the children in the misled group were as accurate as children in the control group when asked, during direct examination, about events that had actually happened (Zajac & Hayne, 2003: 191).
[f]or children in the misled group, their accuracy after cross-examination was not significantly different from the low level that was seen during direct examination [about false events] … . For children in the control group, on the other hand, their accuracy actually declined after the cross-examination interview. (ibid.: 190; emphasis added)
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’ (ibid.: 191). As well, the misled group’s prior exposure to misinformation did not make them more susceptible to probing cross-examination. This means that children who had given accurate answers were persuaded during cross-examination to change their previous accurate accounts of what had happened at the police station. This may have been due to either a weakened memory of the event and greater susceptibility to suggestion (since cross-examination occurred 9.5 months later) and/or because of ‘compliance with suggestions’ that the children knew to be incorrect (ibid.: 192). Thus, ‘cross-examination not only proved to be unsuccessful in discrediting … misled children questioned about false events …, it also decreased the accuracy of children who were initially correct’ (ibid.: 191).
the same characteristics that may serve as risk factors for being [sexually] abused would be the same characteristics that are associated with poor courtroom performance. (ibid.: 931)
Because of limitations in the studies by Zajac and colleagues, such as children’s participation in events that had no consequences for them or anyone else if reported, Fogliati and Bussey (2014) undertook a study that tested children’s accuracy in reporting a minor transgressive event to determine the ability of cross-examination to uncover children’s attempts to conceal the transgression. Nonetheless, unlike CSA, the transgression did not involve any wrongdoing committed against the children.
After participation in a healthy eating lesson in which the teacher accidentally tore a poster, two groups of children (aged 5–6 and 7–8 years) were interviewed twice, with the second interview being either another direct examination or a cross-examination. There was no delay between the first and second interviews. The authors found that both younger and older cross-examined children were significantly less accurate and made ‘a higher proportion of incorrect changes’ than those who underwent a second examination since cross-examined children changed more of their initial responses (Fogliati & Bussey, 2014: 306; 307; see also O’Neill & Zajac, 2013a, 2013b). The authors concluded that children’s changes were due to the types of questions asked, rather than ‘the effects of a repeated interview’ (ibid.: 309).
In relation to the witnessed transgression, ‘an equal number of children disclosed the transgression’ during both cross-examination and direct examination (ibid.: 310). While cross-examined children were more likely to ‘change from an initial non-disclosure to a truthful disclosure … they were also more likely to change from a truthful disclosure to a non-disclosure’ (ibid.).
However, the transgression used in this study was quite different to a positive report of CSA since the study focused on whether children would disclose a transgression which differs from the aim of cross-examination in a CSA trial which is to persuade a child to retract their allegations. Nonetheless, the study highlights the possibility that children who retract their CSA allegations under cross-examination may do so because of the suggestibility of cross-examination rather than because of the truth, as demonstrated in the Ward case above. Thus, ‘[s]trategies that adequately test children’s evidence, without compromising their accuracy and truthfulness, need to be identified to increase opportunities for justice’ in CSA cases (ibid.: 312).
The cross-examination effect may be amplified in relation to children with intellectual disabilities. With poorer cognitive abilities, children with an intellectual disability are more susceptible than other children to the suggestibility inherent in cross-examination (Bettenay et al., 2014 and references therein) and may also be more susceptible to the power dynamics between cross-examiner and witness.
After participation in an interactive magic show, Bettenay et al. (2014) found that 100% of 98 children aged 4–11 years with mild to serious intellectual disabilities changed at least one of their responses from their initial interview as a result of cross-examination. However, 96.7% of a control group of typically developing children also changed at least one of their initial responses under cross-examination. Overall, ‘[c]hildren with … poorer memory for event details … were more likely to cede during cross-examination challenges’ (Bettenay et al., 2014: 210). This meant that there was no significant difference between children as a function of cognitive ability, even though all children viewed the ‘videotape of their initial interview’ before being cross-examined which is standard practice in E&W and in some Australian jurisdictions.
The cross-examination effect: degree of changes to evidence by children subject to cross-examination-type questions
Children’s cognitive abilities | Age (years) | One or more changes under x-examination (%) | Event | Delay between interview & x-examination | Study |
|---|---|---|---|---|---|
TD | 5–13 | 76 | Court transcripts | N/A | Zajac et al. (2003) |
TD | 5–6 | 85 (33% changed all original evidence) | Participation in staged, neutral event | 8 months | Zajac and Hayne (2003) |
TD | 9–10 | 70 (43% changed correct responses) | Participation in staged, neutral event | 8 months | Zajac and Hayne (2006) |
TD | 5–12 | 93 | Court transcripts | N/A | Zajac et al. (2009) |
TD | 5–6 9–10 | 66a 42b | Participation in staged, neutral event | 1 week 6 months | |
TD | 4–11 | 96.7 | Staged, neutral event | 10 months | Bettenay et al. (2014) |
ID & BID | 4–11 | 100 | Participation in staged, neutral event | 10 months | Bettenay et al. (2014) |
TD | 9–11 14–16 | Mean changes in evidence#: 3.33 2.98 | Observed film clip: non-violent crime | 8 months | Jack and Zajac (2014) |
TD | 6 & 8 | Mean changes in evidence#: 1.44 | Participation in staged, transgressive event | No delay | Fogliati and Bussey (2014) |
TD | 5 & 6 | 78 | Participation in staged, neutral event | Short delay: 1–3 days Long delay: 8 months | Righarts et al. (2014) |
Responses to questions about the true events were just as likely to be changed (33%) during cross-examination as responses about the false events (42%) … [while] [a]ttempts to talk a child out of an event (36%) during cross-examination were just as successful as attempts to talk them into an event (38%).
Instead, cross-examination achieves what has long been thought to be its main aim in CSA trials—producing inaccurate inconsistencies or retractions. Cross-examination is not able to discriminate between true and false testimony and produces 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 above studies were based on the questioning styles used by defence counsel in CSA trials, and because children in court are cross-examined under much more intimidating conditions than those used in the above studies in a public forum by people unknown to them and for considerably longer periods of time, it is highly likely that cross-examination will produce similar inaccuracies during CSA trials, as illustrated by the cross-examination of the eight-year-old child in Ward, above.
Indeed, Bettenay et al. (2014: 211) reported ‘higher rates of changed responses than those reported in previous research … possibly due to the use of trained legal professionals rather than researchers to conduct interviews’. Thus, children of various age groups do not appear to give their best evidence when faced with cross-examination style questions, suggesting that children should be exempted from this type of questioning. What then would replace it?
An inherent assumption about the truth-finding effectiveness of cross-examination is that any change to a witness’s testimony is a change from a lie or mistake to the truth. However, in an actual trial, lawyers and juries have no way of knowing whether that is the case.
For example, Fogliati and Bussey (2014: 309) investigated ‘the claim that cross-examination promotes true reporting of neutral and transgressive events’ in a study with 119 male children aged 6 and 8 years. During a healthy eating lesson, they witnessed their instructor commit a staged, minor transgression and heard her say that she hoped her mistake would not be found out, although she did not ask the children to keep her secret. The children were divided into two groups and subject to two direct examinations or to a direct examination followed by a cross-examination.
During their initial open-ended narratives, only 10 children (8.40%) ‘spontaneously reported the transgression’ (ibid.: 305) after they were given two prompts, ‘Tell me everything that happened during the healthy eating lesson’ and ‘Tell me more about what happened’ (ibid.: 301). While there was no difference in the accuracy of children’s reports after the initial interview (with six-year-olds being ‘as accurate as older children, when asked non-misleading direct questions’: ibid.: 309), in line with other studies, children who underwent cross-examination were significantly less accurate than those who underwent a second direct examination since they changed more of their original responses during cross-examination (ibid.: 306). Irrespective of age, children subject to cross-examination ‘made a higher proportion of incorrect changes’ compared to those subject to a second examination (ibid.: 307).
While there was no difference in the number of children who disclosed the transgression after a second examination for both six- and eight-year-olds (ibid.: 308), it might have been expected that cross-examination would elicit more disclosures of the teacher’s transgression. However, ‘the number of kindergarten children who disclosed the transgression did not differ’ between their first and second interviews while fewer eight-year-olds disclosed the transgression after cross-examination compared to direct examination (ibid.: 308). Thus, ‘cross-examination did not promote truthful disclosure for kindergarten children, and actually reduced the number of older children who provided truthful disclosures’ (ibid.: 309).
Although children subject to cross-examination were more likely ‘to change from an initial non-disclosure to a truthful disclosure … they were also more likely to change from a truthful disclosure to a non-disclosure than … to make no disclosure change’. This means that cross-examination encouraged some children to disclose the transgression but also persuaded others to retract their initial report of the teacher’s transgression (ibid.: 310). While some would argue that cross-examination is, therefore, essential to reveal the child who is lying or mistaken about being sexually abused, cross-examination in court is not about getting children to disclose abuse but to retract allegations of abuse. In other words, cross-examination is used by defence barristers to elicit retractions not to elicit disclosures.
Overall, this study confirms the findings from other studies that have shown that ‘the detrimental effect of cross-examination [is] due to the types of questions asked during cross-examination, not to the effects of a repeated interview’ (ibid.). It also confirms that children should not be subject to cross-examination style questions.
Whether children should be subject to a second direct examination is also questionable since Fogliati and Bussey (2014: 310) found that six-year-olds made a higher proportion of incorrect changes when subject to a second direct examination, compared to eight-year-olds. This finding is consistent with previous research that ‘demonstrated that younger children had a greater tendency than their older counterparts to change their initial accurate response towards an inaccurate one across repeated interview’. It is possible that the incorrect changes by younger children were due to social pressure in that they assumed that their first answers were incorrect and made adjustments to please adult interviewers.
A series of studies have investigated the effect of coaching on children’s ability to give truthful evidence and the effect of cross-examination on persuading coached children to change their evidence. The first study was conducted by Zajac et al. (2003), discussed above, who found that misled or coached five- and six-year-old children were no more susceptible to the cross-examination effect than a control group of non-coached children. Thus, cross-examination proved unsuccessful in discriminating between coached and non-coached children.
Zajac and Hayne (2006: 12) found that when nine- and 10-year-old children were given misinformation before being cross-examined they were ‘significantly more accurate’ compared to five- and six-year-olds and did not differ from non-coached nine- and 10-year-olds after direct examination.
More recently, Fogliati and Bussey (2015) conducted a study in which children were coached to lie, resulting in three experimental conditions: (i) those who were coached to say that a transgression had occurred during a healthy eating lesson; (ii) those who were coached to deny that an actual transgression had occurred and (iii) those who were not coached.
Cross-examination of the two groups of coached children was found to be more effective than neutral, examination-style questions at eliciting accurate accounts of what the children had witnessed. However, cross-examination of the non-coached children produced more inaccuracies compared to neutral questioning, in line with previous research. Nonetheless, the limitations of this study must be noted, since the children were coached by a stranger about events that did not happen to them personally. Although the coached children readily fabricated events when examined, they did so about neutral events that had not happened to them and that were non-abusive in nature. Thus, the question arises, are children just as easily coached in relation to sexual acts which are beyond their actual lived experiences?
Indeed, in a CSA trial, there is rarely actual evidence to prove that a child complainant has been coached so that lawyers and fact-finders have no measure for determining whether a child’s change to her/his evidence, in the form of a retraction, is a change towards, or away from, the truth.
In order to determine how well cross-examination can discriminate between truthful and untruthful children, Zajac, Irvine, Ingram, and Jack (2016) conducted a study to determine the effectiveness of cross-examination on children aged 6–11 years who had been coached by a parent (who acted as a confederate), rather than a stranger, during a staged computer game, compared with a group of non-coached children. They found that 32 out of 33 out of the latter group ‘gave highly accurate’ evidence about two events while all 32 coached children ‘told at least one lie’ (Zajac et al., 2016: 168). After cross-examination, the coached children significantly increased their accuracy compared to the non-coached children whose accuracy decreased significantly. However, this study does not prove that cross-examined children in CSA trials necessarily change their evidence towards the truth since a ‘changed answer in response to a cross-examination-style challenge was not diagnostic of accuracy beyond the level of chance’ (Zajac et al., 2016: 172). Both neutral and cross-examination questions were found to be ‘poor discriminants’ in eliciting the truth (ibid.).
The above study was limited by the fact that parental coaching was designed to prevent a child from revealing that the parent had disobeyed researchers’ instructions in relation to a computer game, rather than coaching children about sexually abusive matters that had not occurred. Thus, the likely rationale for lying by the coached children was to protect their parents from the consequences of their parents’ disobedience, something that is emotionally and socially quite different to lying about sexual abuse in the face of police and prosecutorial questioning, and in the authoritarian and intimidating setting of a courtroom where children are generally required to take an oath before giving evidence.
- (i)
cross-examination questions significantly affect children’s accuracy about events they have witnessed or participated in (Bettenay et al., 2014; Fogliati & Bussey, 2015; O’Neill & Zajac, 2013a, 2013b; Righarts et al., 2014; Zajac & Hayne, 2003; Zajac, Jury, & O’Neill, 2009);
- (ii)
the accuracy of both younger (5 and 6-year olds: Zajac & Hayne, 2003) and older children is similarly affected (9 and 10 years olds: Zajac & Hayne, 2006);
- (iii)
participation in an event is no safeguard against the manipulative effects of cross-examination (Zajac & Hayne, 2003, 2006; Zajac et al., 2003);
- (iv)
the cross-examination effect has been seen when children are cross-examined soon after an event (1 week: Righarts et al., 2014) and several months after an event (8 months: Zajac & Hayne, 2003; 6 months: O’Neill & Zajac, 2013a, 2013b);
- (v)
particular characteristics of children increase the cross-examination effect, including low self-esteem, low self-confidence and low assertiveness (Zajac et al., 2009);
- (vi)
the cross-examination effect for younger children (5–6 years) is compounded by long delays between the initial interview and cross-examination compared to older children (9–10 years) (O’Neill & Zajac, 2013a, 2013b);
- (vii)
accuracy changes appear to be due to social factors (such as pressure from authority figures) rather than memory loss since the type of interview influences accuracy changes—re-examination soon after cross-examination restores accuracy (Righarts et al., 2014);
- (viii)
while children make changes when subject to any type of second interview, children make more changes and more inaccurate changes when they are subject to a second interview in the form of cross-examination compared to a re-examination (O’Neill & Zajac, 2013a, 2013b);
- (ix)
practice at answering cross-examination questions reduces the cross-examination effect but only after a one-week delay before cross-examination compared to a one-month delay (O’Neill & Zajac, 2013a, 2013b; Righarts, O’Neill, & Zajac, 2013).
We also know that mental abuse, intimidation and humiliation by defence lawyers are relatively common occurrences in CSA trials and that defence counsel deliberately use complex and leading questions to produce inaccuracies and inconsistencies in children’s testimony. In fact, the research shows that ‘narrowing children’s response options [through the use of leading yes/no questions] makes them more susceptible to errors’, compared to open questions which broaden their range of response options since they are more likely to guess the answer to yes/no questions (Sharman & Powell, 2012: 48).
11 The Effect of Cross-Examination on the Evidence of Adults
Most of the research on the effect of cross-examination has focused on children because of the mismatch, in terms of power and language, between a cognitively undeveloped child and a trained adult lawyer.
What do we know about the effect of cross-examination on the evidence of adult complainants?
When Westera et al. (2017: 28) undertook a comparison of courtroom questioning in adult sexual assault cases in the 1950s and the 2000s they found that complainants in contemporary cases were asked more questions overall compared to complainants in the 1950s, including ‘34% more questions during cross-examination’. In both time periods, lawyers spoke ‘considerably more words than complainants’ during cross-examination although more words were uttered by lawyers during contemporary cross-examination (ibid.: 23). In fact, ‘the length of cross-examination was independently driven by the number of words uttered by the [prosecutor]’ (ibid.: 27).
There was evidence that questioning during evidence-in-chief had improved over the years with ‘proportionately more open questions, … fewer closed questions, … and marginally fewer leading questions’ in contemporary cases, although most prosecutors’ questions were still closed and leading (ibid.: 24; 27). However, ‘the proportions of open, closed, and leading questions’ remained the same during cross-examination in both time periods (ibid.), with most being leading.
Do adults, like children, have difficulties in understanding the types of complex questions used in cross-examination? Westera et al. (2017: 28) found that in the longer cross-examinations in the 2000s cases, contemporary complainants used various ‘adaptive strategies’ such as clarification-seeking compared to historical complainants. When complainants’ responses were examined for levels of uncertainty, ‘complainant uncertainty was more common during cross-examination than during evidence-in-chief’ in both time periods, as were inconsistencies in evidence (ibid.: 26; 28).
The vast majority of complainants also made at least one change to their earlier testimony during cross-examination (100%, historical; 95%, contemporary). While the authors had no way of knowing whether these were changes towards or away from the truth, they found ‘a high rate of compliance with leading questions’. Critically, however, the number of changes to earlier testimony ‘was positively associated with the number of questions asked’ (ibid.: 26) which suggests that if a questioner repeats a particular question they will eventually get the answer they want.
Overall, the research shows that adults struggle to understand linguistically complex questions and are susceptible to leading questions and social pressures to conform (see review in Zajac & Cannan, 2009), with adults’ accuracy decreasing significantly in the face of cross-examination style questions.
To determine the effect of cross-examination as a function of age, Zajac and Cannan (2009) analysed 30 transcripts of sexual assault proceedings involving children aged 5–12 years and adults aged 18–35 years. As expected, defence lawyers asked more leading questions of both children and adults than prosecutors. They also asked a higher proportion of questions with complex vocabulary or language structure, and questions that suggested poor memory, confusion or lying, than did prosecutors. However, adults were more likely than children to be accused of lying by defence counsel (Zajac & Cannan, 2009: S44).
In terms of the cross-examination effect, ‘complainants were more likely to change their story when questioned by defence lawyers’ so much so that 100% of the adults and 93% of the children made at least one change to their original evidence during cross-examination (ibid.), although the authors could not determine if these were changes towards or away from the truth. In addition, ‘[t]he more cross-examination questions that complainants were asked, the more changes they made’ with these changes occurring at higher than chance levels (ibid.: S45; S46). The questions that challenged a complainant’s credibility were ‘predictive of changes’ made by complainants and when there was a change in story, ‘lawyers were more likely than chance to follow up … with a credibility challenging question’, revealing the specific strategies that lawyers employ (ibid.: S47). These results suggest that social pressure, rather than lack of veracity, explains some of the changes made under cross-examination.
Lawyers used specific strategies when complainants resisted a question, in that ‘resistance was [also] predictive of a subsequent credibility-challenging question … [and] … predictive of a subsequent closed question for children’ (ibid.: S47). Remarkably, when a complainant sought clarification, the ‘subsequent question was more likely than chance to be complex’ for both adults and children, although prosecutors were also more likely to do this for adults (ibid.). However, clarification seeking was low for both adults and children while ‘clear misunderstandings’ of questions was evident (ibid.: S49).
Overall, defence lawyers ‘were more likely to ask credibility-challenging questions suggesting that the complainant was purposefully untruthful … than … to ask questions that merely suggested’ confusion or mistake (ibid.: S48). Such a strategy aims to create reasonable doubt in the minds of the fact-finders since suggestions of lying are probably more likely to trigger jurors’ own RMA and gender biases than suggestions that the complainant has made a mistake.
There were also differences in responses between children and adults undergoing cross-examination in that children were more likely to comply, and less likely to disagree with lawyers’ suggestions, as well as less likely to clarify their responses with additional information (ibid.: S49). Surprisingly, adults were more likely to display clear misunderstandings of lawyers’ questions than children (ibid.: S45).
Valentine and Maras (2011) also investigated the effect of cross-examination on the evidence of adult eyewitnesses who watched a simulated crime event. After cross-examination, almost three-quarters (73%) changed one answer from their original statement, with more than a third (37%) changing two or more answers, although the witnesses’ memories ‘did not become more accurate as a result of cross-examination’. This study also illustrates the manipulative effect of cross-examination, with 84% of witnesses accepting ‘at least once that they may be mistaken’ (ibid.: 559), a finding that was not surprising since cross-examiners ‘made extensive use of social pressure (e.g. negative feedback …) and leading questions’ (ibid.: 559).
- (i)
closed (‘Did Eric have a company logo on his van?’);
- (ii)
closed specific (‘Did Eric have a company logo in large black letters on his van?’);
- (iii)
closed presumptive or suggestive (‘Eric had a company logo on his van, didn’t he?’); and
- (iv)
open presumptive or suggestive (‘Tell me about the company logo on Eric’s van?’),
They found that participants (aged 18–60 years) were more likely to be misled when asked closed specific and open presumptive questions: ‘[t]hey incorrectly agreed with more misleading suggestions when these suggestions were delivered in closed specific and open presumptive questions (37.8%) than closed and closed presumptive questions (24.2%)’ (ibid.).
were more misled when previously exposed to misleading information in closed specific and open presumptive questions than when exposed to this information in closed and closed presumptive interview questions. Not only did the closed specific and open presumptive questions encourage people to agree with the misleading questions during the initial interview but also these questions decreased the accuracy of participants’ memories during the later recognition test. … These findings fit with earlier research demonstrating that the way in which interview questions are worded affects witnesses’ memories. (Sharman & Powell, 2012: 52)
In a study of three age groups (45 children aged 9–11 years, 41 adolescents aged 14–16 years and 42 adults aged 25–60 years), Jack and Zajac (2014) found that participants in all age groups were more likely to change responses in their initial interview under cross-examination compared with a second examination. However, children and adolescents were more vulnerable to the cross-examination effect than adults while ‘the common practice of playing children a recording of their evidential interview prior to appearing in court provides no buffer against this effect’ (Jack & Zajac, 2014: 5). There was also a decrease in accuracy between the first and second interview which was ‘significantly greater for peripheral details than for central details’ (ibid.: 4).
Most recently, Westera et al. (2017) conducted a historical comparison of the questioning of sexual assault complainants in the 1950s and the 2000s, in order to determine whether questioning practices had evolved over the past 50 years, with the two time periods representing conservative and more liberal sexual and social views about women, respectively.
After examining 42 transcripts of the evidence of complainants aged 16 years or older (of which 50% were historical), Westera et al. (2017: 22) found that complainants were asked more questions in cross-examination than in evidence-in-chief and also more questions in contemporary compared to historical cases. Matching this latter finding was the fact that defence lawyers spoke considerably more words than complainants.
In terms of the cross-examination effect, more complainants made at least one change during cross-examination than during evidence-in-chief; that is, 100% (historical) and 95% (contemporary) compared with 29% (historical) and 43% (contemporary) during evidence-in-chief.
Possibly reflecting current modern approaches, prosecutors in contemporary cases ‘asked complainants a higher proportion of open-ended questions … than in historical cases’ but ‘the fact remains that the bulk of prosecutors’ questions are still closed and leading’ (ibid.: 26; 27). In contrast, ‘the format of cross-examination questions has remained remarkably consistent over time’, with leading questions still making up the bulk of questions, although contemporary cross-examinations ‘are significantly longer’ than in the 1950s (ibid.: 28; 29). As a result, contemporary complainants spent more time giving and seeking clarification than historical complainants. What had not changed was ‘a high rate of compliance with leading questions’, as well as uncertainty and inconsistency during cross-examination compared to evidence-in-chief (ibid.).
In a follow-up study to that of Westera et al. (2017) and Zydervelt, Zajac, Kaladelfos, and Westera (2017) explored the content of cross-examiners’ questions to determine the strategies used to discredit complainants in the 1950s compared with the 2000s. Because of the rules of evidence that applied in the 1950s, ‘[c]omplainants were routinely asked humiliating questions about their sexual morals and respectability, including … sexual history, sexual reputation, sexual desires and modes of dress’ (ibid.: 552). But with the introduction of rape shield laws which restrict the scope of cross-examination on sexual history and reputation, is the content of questions still the same?68
Zydervelt et al. (2017: 560) found that ‘[l]awyers conducting contemporary cross-examinations used a significantly larger number of specific tactics’ with complainants than in historical cross-examinations. In both time periods, strategies targeted ‘reliability, consistency with other evidence, credibility and the plausibility of the allegation’ (ibid.), although defence counsel mostly challenged plausibility. Like the Lazarus case discussed in Chapter 7, contemporary lawyers commonly challenged plausibility by referring to the complainant’s behaviour ‘immediately before the offence … or immediately afterwards’ (ibid.: 561). Delays in reporting, lack of resistance and continued relationship with the defendant were less common tactics.
The most common credibility tactics were about ‘the complainant’s personal traits or potential ulterior motives for making the allegation’ (ibid.: 563). Despite rape shield laws, questions about the complainant’s sexual history with other people was more common than in historical cases (ibid.: 563–564). Questions about reliability were more likely in contemporary cases and included issues around intoxication, memory and perception fallibility.
Overall, the tactics of defence lawyers have changed little over 50 years—they are ‘pursuing the same goals in slightly different ways’ with tactics based on rape myths being common (ibid.: 565). Complainant behaviour was used in more than 85% of cases to infer consent (and hence the falsity of the allegation) while a complainant’s personal attributes were used to make her appear less credible in about 60% of cases (ibid.).
As discussed in Chapters 4 and 5, one of the aims of defence cross-examination is to paint the adult complainant as someone who is morally questionable by focusing on rape myths about her behaviour and habits, thereby raising doubts about the veracity of her complaint. The studies above not only confirm this time-honoured strategy but highlight the fact that adult complainants’ accuracy in actual cases also decreases with changes to testimony under cross-examination and that longer cross-examinations will produce more inaccurate evidence.
As discussed in Chapter 11, the lack of a trauma-informed court environment means that the adversarial trial tends to ignore the effect of trauma on adult rape complainants’ ability to give evidence and the cross-examination effect on adults’ testimony. The above study by Westera et al. (2017) suggests that the length of cross-examination, the style of cross-examination questions, as well as who the questioner is, are key reform issues for the sexual assault trial. These reform issues are discussed in the next chapters.