Hanna, Kirsten; Davies, Emma; Henderson, Emily; Crothers, Charles; Rotherham, Clare --- "Child witnesses in New Zealand criminal courts: a review of practice and implications for policy" [2010] NZLFRRp 1

	Home \| Databases \| WorldLII \| Search \| Feedback New Zealand Law Foundation Research Reports

NEW ZEALAND CRIMINAL COURTS: A REVIEW OF PRACTICE AND IMPLICATIONS FOR POLICY

Kirsten Hanna Emma Davies Emily Henderson Charles Crothers Clare Rotherham

A project funded by the New Zealand Law Foundation with contributions from the Ministry of Social Development, Ministry of Justice and the New Zealand Police.

The opinions expressed in this report are those of the authors and do not necessarily represent the views of any funding agency or any agency or individual who contributed data or any other information to the project. Nor does this necessarily represent the views of individual members of the Advisory Group or any other individuals who commented on earlier drafts of this report, whose comments were gratefully received by the authors.

This report is dedicated to the many practitioners, past and present, who from within and across sectors have worked tirelessly to improve conditions for children who testify in criminal proceedings.

Dr Kirsten Hanna (Principal Researcher) received her PhD in Linguistics from the University of Auckland in 2005 where, as a postgraduate, she lectured in Linguistics and tutored in both Linguistics and Latin language. Her PhD thesis examined aspects of Archaic Latin syntax. Since joining the Institute of Public Policy at AUT in 2005, her research interests have focused on the place of children in public policy. She has published on child-focused government structures, the rationale for government investment in children‘s early years and child poverty.

Dr Emma Davies is Programme Director – Social Development at AUT‘s Institute of Public Policy. Emma conducted her doctoral research on children‘s experiences of investigation and criminal justice processes in the late 1990s. The findings included a lack of coordination between agencies in child abuse investigations and a lack of education for court for child witnesses. Therefore, with support from the New Zealand Law Foundation, in association with senior practitioners in the field, she led the initial work on the establishment of Auckland‘s multi-agency centre for child abuse investigations – Puawaitahi, which opened in 2002. Emma also led a group of senior practitioners to develop the framework for New Zealand‘s Court Education for Young Witnesses programme in 1999; then led the research team that evaluated the pilot programme in 2004. In 2006, Emma was a Principal Analyst for the Taskforce for Action on Violence in Families and in 2007-8 was Principal Advisor to the Children‘s Commissioner, responsible for the Auckland office.

Dr Emily Henderson holds a BA-LLB(Hons) and a MJur from Auckland University and a PhD in Law from Cambridge University. She has conducted studies of NZ barristers‘ self-reported practices in and beliefs about child sexual abuse trials and the theory and practice of cross-examination in the criminal trial, utilising a qualitative study of barristers in NZ and the UK, inter alia. Her research interests range from the practice of family law to the reform of the criminal trial and the position of vulnerable witnesses. She also practices part time as a barrister and solicitor in civil and family litigation at Henderson Reeves in Whangarei where she lives with her husband and four children.

Charles Crothers is Professor of Sociology in the Department of Social Sciences at AUT, after previously serving as a Professor of Sociology at the University of Natal, Durban, South Africa. Prior to this position Charles lectured in the Departments of Sociology at the University of Auckland and Victoria University, and had been President of the New Zealand Sociological Association, co-editor of 'New Zealand Sociology' and Vice-President of the International Sociological Association's Research committee on the history of Sociology. His interests include social theory, social research methods, New Zealand and other settler societies and policy applications.

Clare Rotherham is a research officer in the Local Government Centre within AUT‘s Institute of Public Policy. Clare‘s professional background includes practising as a

solicitor for four years in Auckland and London in a variety of areas, including property law, corporate law, employment law and civil litigation. Clare has worked on research projects on corporate citizenship, as well being part of the team that designed an institutional strengthening project for the Ministry of Women Community and Social Development in Samoa.

Figure 2 : Types of offences alleged by the complainants in the full data sample (n=55) 23

Figure 3 : Average time taken to process child witness trials (Auckland, Manukau,

Figure 6 : Alternative modes of evidence used at court by the 71 child witnesses 39

Figure 8 : Duration of child‘s entire evidence at court (by role and type of evidence) 51

Figure 9 : Children‘s and defence lawyers‘ demeanour during questioning at trial 52

Table 4 : Time taken to process child witness trials (Auckland, Manukau, Christchurch & Wellington) 24

Table 7 : Associations between alternative modes of giving evidence and age/role 41

Table 11 : Associations between child having a trial support person and child's age and role 47

Table 12 : Associations between timing of testifying and child‘s role and age 53

Table 15 : Manner of initial statement: associations by age, role and nature of alleged

Table 16 : Applications for forensic interviews sought/not sought by region (n=98) 100

Table 18 : Applications for CCTV or screens by age, role and relationship with accused 101

Table 23 : Advantages and disadvantages of the two "specialist child examiner" systems 175

This report came about as a result of the support and assistance of a multitude of individuals and agencies.

We wish first to thank our funders, the New Zealand Law Foundation, the Ministry of Social Development, the Ministry of Justice, and the New Zealand Police for providing the financial resources needed to conduct this research. We also wish to thank the organisations which provided significant contributions in kind, namely, Meredith Connell; Luke, Cunningham & Clere; Raymond, Donnelly & Co (Offices of the Crown Solicitors in Auckland, Wellington and Christchurch respectively); the Office of the Children‘s Commissioner; and AUT University.

The authors gratefully acknowledge the enormous contribution of the project Advisory Group to this research. The members of the Advisory Group are (in alpha- betical order) Gill Basher (Forensic Interviewer), Dr Suzanne Blackwell (Clinical Psy- chologist), Alison Clelland (Senior Lecturer, Faculty of Law, University of Auckland), Dr Emma Davies (Advisory Group Chair, Programme Director Social Policy, Institute of Public Policy, AUT University), Peter Dean (Prosecutor, Meredith Connell), Richard Earwaker (Defence Lawyer), Phil Hamlin (Prosecutor, Meredith Connell), Dr Ian Hassall (Senior Researcher, AUT University), Judge Anne Kiernan (Auckland District Court), Detective Sergeant Phil Kirkham (New Zealand Police), Professor Fred Seymour (Head of Department, Psychology Department, University of Auckland), Karen Wilson (National Trainer, Forensic Interviewers) and Pat Worthington (Northern Regional Coordinator, Victims Advisers).

We are also grateful to Hon Justice Randerson, Chief High Court Judge, Hon Judge Johnson, Chief District Court Judge, and the New Zealand Police for permitting access to data. The process for collecting the data relied heavily on prosecutors, officers in charge (members of the regions‘ Child Abuse Teams in particular) and administrators in the three Offices of the Crown Solicitors. We gratefully acknowledge the enormous contribution of these individuals to the project. While the prosecutors and officers in charge are too numerous to name, we wish to thank per- sonally the administrators; they are (from south to north) Jenna Nielson, Lynette Birnie, Lynda Brown and Sarah Freeman of Raymond, Donnelly & Co. in Christchurch; Paula Rich of Luke, Cunningham & Clere in Wellington; and Lolita Lotu- Iiga, Alana Patterson, Maxine Morar and Julie Amphlett of Meredith Connell in Auckland and Manukau. We are indebted to the law firms and the Police District Commanders of Auckland Central, Waitemata, Counties Manukau, Christchurch and Wellington for allowing staff to be engaged in data collection.

Data collection also required ongoing liaison and assistance from court staff, includ- ing caseflow managers and trial schedulers. In this regard, we gratefully acknowl-

edge the assistance of Shelley Raskin, Rob Neems, Phil Clarke, Kevin Conroy, Phil Miles and Helen Vermeulen.

We would also like to thank those individuals in the funding and data-holding agen- cies with whom we liaised in negotiating funding agreements, research protocols and data collection, including Lynda Hagen, Joane Ciesiolka-Lord and Dianne Gallagher of the New Zealand Law Foundation and representatives of the Ministry of Justice, Ministry of Social Development and the New Zealand Police.

In Chapter 5, we survey innovative practices in overseas jurisdictions. The informa- tion in that chapter drew upon not only published material, but also information pro- vided by practitioners and academics in the relevant jurisdictions. The authors gratefully acknowledge the contributions of Rita Blumrick, Gert Jonker, Karen Hollely, Beverley Edwards, Irit Hershkowitz, Carmit Katz, Mel Pipe, Paul Roth, Phillippe Lemaire, Elisabeth Peden, Judy Cashmore, Chris Bleby, Philip Crowe, Paul Yovich, Sonia Kolodenski, Tracey Brandis, Hylton Quail, Luke Brasch, Liesl Chapman and Dale Ives.

Further, as part of this study, the first three authors spent two weeks in Britain examining innovative practice in England and Wales. We would like to express our deep gratitude to all of those who helped us in the course of our research in England, namely, DCS Simon Deacy, Superintendent of the Gwent Police Dept; Detective Constable Tracey Walker and Detective Inspector Michele Dillon of the Oxfordshire Child Abuse Investigation Unit of the Thames Valley Police; His Honour Lord Justice John Thomas, His Honour Lord Justice Anthony Hughes, His Honour Judge Peter Rooke QC, Her Honour Judge Susan Tapping; His Honour Judge Julian Hall and Her Honour Judge Mary Ann Mowatt together with a number of their colleagues at the Oxford Crown Court; Jan Jones; members of the Office for Criminal Justice Reform led by Julie Clouder and Gillian Harrison and also with Lucienne Edge and Gohil Rekha; Michael Bowes, QC, Janine Sheff QC; John Riley, David Wurztel, and Nadine Tilbury; Joyce Plotnikoff and Richard Woolfson of Lexicon Limited; Laura Hoyano, lecturer in law and Fellow of Wadham College, Oxford; Professor Di Birch of the Law Faculty of Nottingham University; Professor John R Spencer, Fellow of Selwyn College and Professor of the Cambridge Law Faculty; and Barbara Esam of NSPCC. We are grateful to Laura Hoyano, Joyce Plotnikoff and Richard Woolfson, Di Birch, Judge Mowat and Julie Clouder for their kind hospitality while the researchers were in England. We would also particularly like to thank Joyce Plotnikoff and Laura Hoyano for arranging many of the interviews.

We are also grateful to those practitioners in Western Australia who shared their views on the pre-recording of children‘s evidence; they included (in alphabetical order) Natalie Adams, Judge Ellis, Professor Judith Fordham, Judge Hal Jackson, Justice Lindy Jenkins, Nuala Keeting, Judge Antoinette Kennedy (Chief District Court Judge), Glenn Mace, Helen Prince, Hylton Quail, Judge Kevin Sleight, Justice Christine Wheeler, Christine White, Paul Yovich and Lyn Zinenko.

In addition, we wish to thank those individuals who reviewed and commented on earlier drafts of Chapter 5 pertaining to their jurisdictions: Trond Myklebust, Philippe Lemaire, Carmit Katz, Karen Hollely, Gert Jonker, Rita Blumrick and Beverley Edwards. We are especially grateful to Joyce Plotnikoff, Thomas Biss, Professor Spencer, Dr Hoyano, members of the Office for Criminal Justice Reform in the UK, His Honour Justice Thomas and His Honour Chief District Court Judge Johnson who read and commented on various drafts of Chapter 5 as a whole. We are also grateful to Beryl Jack for proof-reading and formatting an earlier version of this report, to Kieran Raftery (of Meredith Connell) and Elizabeth Rowe (of Elizabeth Rowe Consulting) for their incisive reviews of Chapter 6, Professor Jane Koziol-McLain for her review of Chapter 2 and to Catherine Harland of the Institute of Public Policy for her incisive review of the entire report.

Finally we wish to thank our colleagues at the Institute of Public Policy, AUT University and elsewhere, for their support and encouragement, as well as our families for their patience and forbearance.

Ensuring that all accused persons have a fair trial, and obtaining the most accurate and complete testimony from witnesses, are critical to the quality of justice delivered by the courts. Testifying can be difficult for adults, let alone children. Indeed concerns have been raised in New Zealand about the treatment of child witnesses since at least the 1960s. The purpose of this study is to see how far we have come today in addressing the concerns since then, and the extent to which measures adopted to ameliorate conditions for children—while protecting accused adults—have achieved their aims.

The research was funded by the New Zealand Law Foundation with further contributions from the Ministry of Social Development, Ministry of Justice and the New Zealand Police. It focuses on children aged 17 years and under who have given evidence as witnesses for the prosecution in criminal proceedings, whether as complainants or witnesses. Data was collected throughout 2008 and 2009 on children involved in trials set down for the District Courts at Auckland, Manukau, Wellington and Christchurch and in the High Courts at Auckland, Wellington and Christchurch.

Despite the support of the data-holding agencies (i.e. Ministry of Justice, the Ministry of Social Development and New Zealand Police), it was laborious to access basic information across agencies to build up a picture of how the criminal justice system treats its child witnesses as cases involving child witnesses were not easily identifiable within databases. It is much harder to monitor progress on issues not visible in tracking systems.

There are legislative and procedural measures available to assist children who testify in the criminal courts. For example, children can be forensically interviewed by people specially trained to effectively communicate with children and the courts, and these videorecorded interviews are permitted as evidence-in-chief in court.

Children may also be cross-examined via closed-circuit television or behind a screen. The Evidence Act 2006 extended eligibility to these measures to include all vulnerable witnesses and introduced a wider interpretation of ―alternative modes‖ to include ―any appropriate practical and technical means... to enable the Judge, the jury (if any), and any lawyers to see and hear the witness giving evidence‖ (s105(1)(b)). Child witnesses are also entitled to information about the legal process and their part in it through the Ministry of Justice‘s Court Education for Young Witnesses programme.

However, this study has found inconsistent implementation of this existing law and policy in relation to children and some regional variation in practices.

No children under 6 years gave evidence in court in the timeframes of the study.1 Younger children were more likely than older children to be forensically interviewed and hence to have the option of their videorecorded forensic interview being played at court as evidence-in-chief; they were more likely than older children to meet the prosecutor before trial, testify with a trial support person on hand and testify via CCTV. Applications for children to testify via CCTV were more common in Auckland and Manukau than in Wellington and Christchurch.

This research shows that child witnesses faced long delays while awaiting trial and inappropriate questioning in the courtroom. While child complainants in the 1990s waited around eight months (255-260 days) for their cases to be processed through the courts,2 children in the current sample waited nearly twice as long: 15 months (477 days) on average with adjournments extending the process. This is a long time in the life of a child.

The standard of forensic interviewing practice in the linguistic analysis of 8,154 questions posed to children was generally high (although we do not claim that this was a representative sample). However children‘s memory and ability to provide good-quality evidence in the criminal courts can be seriously undermined by these lengthy delays between their initial complaint and testifying at court, particularly if they are asked questions that they do not understand and if the act of testifying is unnecessarily stressful.

It is not only the language used (i.e. grammar and vocabulary) but also the question type and sequencing which can impact on children‘s comprehension and accuracy. The transcripts considered here showed that judges intervened to stop inappropriate questioning more often than their counterparts in the 1990s. Despite these interventions, this study also found that children are still frequently being cross- examined in ways that are forensically unsafe.

We have identified alternative processes that work well in other jurisdictions for facilitating children‘s participation in criminal proceedings which could be adapted to the New Zealand context to complement the work of agencies currently focusing on issues of delay. We found the practice in Western Australia, Norway and England to be particularly instructive and have discussed options to improve the system here.

The goal of such innovations would be to ensure that children‘s evidence can be heard and tested in a transparent and timely manner regarding matters of concern to all parties, using sequences of questions posed in accordance with best examination practice. We think there is most merit in a system that adds pre-recorded cross- and re-examination by a court-appointed specialist child examiner to our current forensic interviewing processes. The logistics of implementation in relation to the

criminal justice and child protection systems would need to be thought through. We also hope that this study will result in the systematic monitoring of the status of child witnesses in the system, not least to enable effective ongoing evaluation of any improved or new measures.

All child victims and witnesses to crime, as well as disabled and other vulnerable adults, could benefit from the recommended measures. This report also documents why the needs of children should remain a primary consideration in the discussions that we hope emerge upon its release.

In this chapter, we see how involvement in the criminal justice system unfolded for

71 children who testified in the criminal courts in the sample period. The offences alleged by complainants were predominantly in relation to sexual offending (69%), sexual and physical abuse (7%), sexual abuse and abduction (5%), physical abuse only (13%) and aggravated robbery or arson (5%).

The experiences of the children who testified depended sometimes on their age and sometimes on their role in the proceedings (i.e. complainant versus witness). Younger children tended to testify via CCTV while older children used screens or gave evidence unshielded in court, and more complainants than witnesses viewed their forensic interview before trial. Younger children were less likely than their older counterparts to be consulted on which alternative mode they would like to use to testify at court, but more likely to meet the prosecutor before trial. Younger children and complainants were more likely to be forensically interviewed or testify with a trial support person than older children and witnesses.

There were, however, some aspects of the experience which were common across the sample. For example, nearly all of the children indicated that they hadn‘t understood questions posed by the defence lawyer and there were significant delays in processing these cases. Table 1 shows that some children are still not benefiting from the full range of protective and preparatory measures available under current legislative and procedural frameworks—that is, measures intended to make testifying easier and safer for New Zealand‘s children.

55) of complainants and 60% (n=9 of 15) witnesses were not forensically interviewed).

The purpose of this chapter is to examine the questioning practices of forensic interviewers, prosecutors and defence lawyers to ascertain whether the questions they posed to children were in accordance with best practice, and the extent to which practice has improved since the 1990s and early 2000s when previous transcript analyses suggested problems in cross-examination (Davies & Seymour, 1998; Zajac, Gross, & Hayne, 2003). All these analyses were based on non- representative samples, including the current study which is based on the transcripts of 18 children who gave evidence as a complainant or witness for the prosecution in 2008. The children were testifying in the High Court in Auckland or the Auckland, Manukau or Christchurch District Courts.

Compared to forensic interviewers and prosecutors, and as one would expect, defence lawyers used higher proportions of closed, leading questions. However, research has shown that these question types are more likely to elicit inaccurate information and control the witness‘ testimony. Defence lawyers also used more double negatives than forensic interviewers or prosecutors. Forensic interviewers‘ questions, on the other hand, were more age appropriate in terms of grammar and vocabulary compared to lawyers‘. For example, 0.3% of forensic interviewers‘ utterances contained complex vocabulary, compared to 12% of prosecutors‘ utterances and 18% of defence lawyers‘. Forensic interviewers also tended to avoid complex grammatical constructions more than prosecutors and/or defence lawyers, such as verbs in the passive voice, sentences containing multiple embedded clauses and sentences containing multiple types of complexity.

The transcripts also revealed lawyers still using tactics which could confuse child witnesses. For example, in 75% of the cross-examinations analysed, lawyers used a technique where they switched rapidly from one topic to another unrelated one, without using the conventional means of indicating a topic shift; this tactic may confuse and disorientate the child but is recommended in leading advocacy texts. Another technique is to focus questions on peripheral details of the events, capitalising on the fact that such information tends to be less well remembered than information that is salient or central. Inconsistencies in relation to peripheral details are to be expected. They are, however, often seen as a marker of inaccuracy, which could cast doubt on the child‘s testimony as a whole.

The analysis shows that children are still being questioned in the courtroom in ways that are forensically unsafe and subjected to tactics in cross-examination that are likely to confuse them. Under these conditions, it is not surprising that inconsistencies arise, especially as children are often reporting on multiple events which occurred months if not years beforehand. Having highlighted changes in the child‘s testimony, lawyers can then use these inconsistencies as the basis for accusations of lying, which can add to the child‘s stress. The value of such tactics in testing the evidence is unclear. The nub of the matter is that there is now a substantive body of literature on ways to get the best possible evidence from children (see, for example, Lamb, Orbach, Hershkowitz, Esplin, & Horowitz, 2007; M.

E. Lamb, Y. Orbach, I. Hershkowitz, D. Horowitz, & C. B. Abbott, 2007; Poole & Lamb, 1998) that profoundly contradicts the general principles and practice of cross- examination (Henderson, 2000, 2002; Spencer & Flin, 1993).

Judges have a duty to control inappropriate questioning. In the 1990s, Davies & Seymour (1998) reported that judges rarely did so—there were six examples of judges intervening in twenty-six trials. In the current analysis, judges intervened 38 times within 10 of the 16 trials. Not only did judges in 2008 intervene more often than their counterparts in 1994, but they did so in relation to a wider range of issues, including pointing out that a question was too difficult for the child (or the

judge) to understand; pointing out that the lawyer was presupposing facts denied by the child; and preventing a lawyer from continuing his/her focus on an irrelevant, peripheral detail. The greater attention by the judges here to complex questions and inappropriate tactics is encouraging. However there were many complicated or inappropriate questions which were not challenged.

It is clear that at least in some regions children are still being subjected to suggestive questioning in the courtroom, to complex language and to dubious tactics such as abrupt changes in topic and intense questioning on irrelevant details. In terms of adherence to best practice, the superiority of the questioning skills of forensic interviewers compared to lawyers is apparent. It is time for New Zealand to seek ways of more fully incorporating best practice into the courtroom, in the interests of the function of the courts and in the interests of child witnesses.

The rationale for alternative modes of evidence is to reduce the stress of testifying, be it confrontation or courtroom stress. It is expected that this will, in turn, ―promote the rational ascertainment of facts by enhancing reliability and facilitating communication‖ (New Zealand Law Commission, 1996, p. 25). While screens may go some way towards reducing confrontation stress, CCTV and video testimony address both confrontation and courtroom stress.

There are three particularly pertinent findings in this chapter. First, the data suggests that, for this sample, there were significant regional variations in practice in relation to modes of evidence applications sought. The reasons for this variation are unknown. In particular, applications for CCTV were more common in the northern cities than in the southern cities. This variation in practice runs counter to the notion that the law provides protection for all vulnerable witnesses, no matter where in the country they testify.

Second, applications for CCTV were more common for younger than older children in some centres, yet there is no presumption in the legislation that CCTV is most suitable for young children.

Finally, 24% (n=31) of 131 applications were contested by defence (in three cases it is unknown if the application was contested or not), yet opposition to the child‘s use of an alternative mode was rarely sustained. Of the 125 applications for which the outcome is known (i.e. either defence consented or the judge made a ruling) and it is known whether the application was contested or not, defence consented at some stage to 114 (91%). A great deal of time and resources is being expended on

applications even though, at least in this sample, in the vast majority of cases the application was consented to by defence. Furthermore, although we only sighted a handful of judicial rulings, they did indicate that there is no basis for fears that a child‘s use of an alternative mode contravenes the accused‘s right to a fair trial or

―right‖ to confront the witness. Indeed, this was Parliament‘s view more than 20 years ago.

Unlike other jurisdictions, there is no presumption in favour of or against alternative modes in New Zealand. A presumption in favour of CCTV and videorecorded forensic interviews in particular is warranted in the New Zealand context, in the interests of facilitating all children‘s testimony (not just younger children‘s), in the interests of efficiency (reducing the time spent on applications as well as the possibility of taping CCTV evidence for use at retrials) and to ensure consistent standards of practice across the country. We note that allowing children to testify from a location away from the courthouse via live audiovisual link would have many of the benefits of CCTV in terms of courtroom and confrontation stress.

This chapter examines processes for dealing with child witnesses in six other jurisdictions: the aim is to consider whether these processes are superior to those currently used in New Zealand and to consider whether any might serve as a model for future development here.

The jurisdictions considered are South Africa, England, Israel, Western Australia, Norway and France, each of which has found innovative ways of facilitating child witnesses‘ participation in criminal proceedings. The systems which seem to hold most promise within the countries studied are intermediaries and pre-recording of a child‘s entire evidence. Four of the jurisdictions have intermediary systems in place (South Africa, England and Wales, Norway and Israel). South Africa has used intermediaries since the early 1990s, that is, adults who relay counsel‘s questions to the child in age-appropriate language, then relay the child‘s response to counsel. England and Wales have recently introduced an intermediary system following its successful pilot in 2001, but in this case intermediaries at court typically monitor counsels‘ questioning and indicate to the judge when counsel has posed an inappropriate question; they can also assist in police interviews early on in the process. Under the longstanding Norwegian intermediary system, children are questioned by specialist interviewers, with a judge and counsel both observing from an adjoining room (via videolink or one-way glass). The interviewer conducts the forensic interview according to his or her professional judgement. When the interview is complete, the interviewer consults with the judge and counsel and they are given the opportunity to suggest topics to be covered or contradictions to be explored and so on. The interviewer returns to question the child on the agreed topics. This process is repeated until all are satisfied. The entire interview is

videorecorded and the video is shown at trial, avoiding the need for children to testify at trial (although they can be re-interviewed using the same system if necessary).

Prerecording of a child‘s entire evidence is the norm in three jurisdictions (Norway, Israel and Western Australia). The Norwegian system combines intermediaries with prerecording within a mixed inquisitorial/adversarial system. In Western Australia, prerecording of a child‘s evidence is successfully achieved within a more typically adversarial system. Here, children are forensically interviewed and the recording can be played at trial, as we do in New Zealand. However cross-examination and re- examination are conducted at court (via CCTV) at a prerecording hearing with the judge, counsel and the defendant present, but no jury. The examinations are recorded and these too are played at trial after any necessary editing.

Pre-recording has been a great success in Australia, the only adversarial country to make much use of the system. The potential benefits include allowing children to testify earlier in the process when their memory is fresher, allowing children to participate in legal proceedings without the stress of trial, and allowing children to exit the system at an early stage.

Both the South African and English models demonstrate that intermediaries can offer significant advantages to the courts and to children. However, the expanded version of the intermediary (the specialist examiner) in Norway has more merit in addressing the communication issues raised in this report. Intermediaries in adversarial countries so far are quite restricted in their roles, leaving the essentials of cross- examination untouched, together with many of its problems: some issues with the syntax and vocabulary of individual questions may be addressed but children must still cope with the bulk of an examination designed by a lawyer with little or, more likely, no training in children‘s language and cognitive development or best practice in eliciting accurate evidence from young witnesses.

A further advantage of examination by specialist is that it is entirely congruent with pre-recording the child‘s evidence. In fact the countries which use such a system routinely pre-record the interviews for trial.

In this chapter we discuss the legislative, policy and practice implications of the information documented in the report. In particular, we focus on ways to reduce children‘s exposure to delays and improve court communication with children, these being two particularly intractable problems identified in the research that significantly impact on children‘s ability to give their best evidence in New Zealand courts.

For many readers, the best way to use this report will be to begin with the executive summary (which focuses on the main problems identified through this research) and the final chapter (which discusses the key legislative and policy implications). The Table of Contents and Lists of Figures and Tables then can guide the reader into the body of the report to read the detail of those aspects of particular interest.

The report is dense in information, reflecting the complex web of agencies and systems which are involved with these children over a significant period of time. Individual parts of the report will be particularly relevant to some agencies, but not others; some discussion points will be of more interest to some disciplines than others. In particular, readers interested in starting with an in-depth understanding of the key problems facing child witnesses and the administration of existing law that pertains to them, should start with Chapters 2-4:

Readers particularly interested in the development of the law as it relates to child witnesses, should focus on Chapters 4, 5, 6 and Appendices I and II:

Entering the criminal court system as an adult can be a bewildering experience—and it can be even more so for child witnesses. Concerns have been raised in New Zealand about the treatment of child witnesses since at least the 1960s (Pike, 1973). Legislative change in the 1980s (including the Evidence Amendment Act 1989) helped improve conditions for child witnesses; however a series of studies in the 1990s suggested that children still faced significant difficulties when participating in the criminal justice system (for example, Davies, Henderson, & Seymour, 1997; Davies & Seymour, 1997). Moreover complaints by parents of children who had testified in court prompted the Courts Consultative Committee to conduct a review of the treatment of child witnesses (see Working Party on Child Witnesses, 1996). The purpose of this study is to see how far we have come today in addressing the concerns raised in the 1990s and the extent to which measures adopted to ameliorate conditions for children have achieved their aims.

This report focuses on children aged 17 years and under who have given evidence as witnesses for the prosecution in trials held in District and High Courts, whether as complainants or witnesses.

There are two main strands to the kaupapa of the research design. Firstly, the research was designed to place no burden on child witnesses: there was no contact between the researchers and the child witnesses and their family/whanau/supporters. Rather, we grounded the current study in interview- based research from the 1990s (for example, Davies, 1998; Davies et al., 1997; Davies & Seymour, 1997; Pipe, Henaghan, Bidrose, & Egerton, 1996; Working Party on Child Witnesses, 1996) both in terms of identifying issues to follow up and, in the handful of cases where measurements were available, as a baseline against which to measure progress to date.

Secondly, the authors were clear that the research must be informed by the experience and wisdom of practitioners working in the criminal justice system. Indeed, the shape of the project was first determined in October 2006 as a result of a meeting with such people; they, along with several others, went on to become the project Advisory Group. This Group comprised senior practitioners from the judiciary, police, prosecution, a defence lawyer, forensic interviewers, a Victims Adviser and academics with expertise in the field (see Acknowledgements for full membership of the Advisory Group).

The Advisory Group not only guided the researchers in determining the scope of the project (which included suggesting avenues for investigation over and above those highlighted by the 1990s research), but provided practical assistance in sourcing and collecting data, enlisting the engagement of colleagues to supply data, providing the background information needed to interpret findings and understand agency systems, and reviewing drafts of the report. That said, any errors or inadequacies in this report remain the full responsibility of the authors.

In the 1990s, the Working Party on Child Witnesses (a subcommittee of the Courts Consultative Committee) reviewed the treatment of child witnesses in the New Zealand criminal courts, following complaints by parents of children who had given evidence in court. The committee identified a series of deficiencies in the way child witnesses were treated and drew up recommendations for change (see Working Party on Child Witnesses, 1996). The recommendations included:

Beyond these recommendations, the Working Party noted issues raised by practitioners and others during their review, such as children not having the opportunity to practise with the CCTV equipment before testifying and a tendency for age to be a factor in which alternative modes of evidence were sought (i.e. pre- recorded evidence-in-chief and CCTV tended to be sought for younger children and screens for older children).

The aim of this chapter is to see what progress has been made in the last decade or so towards resolving the problems identified by the Working Party. We also examine issues identified by the Advisory Group to this project as warranting investigation over and above those identified in the 1990s research. We do this by examining how involvement in the criminal justice system unfolded for a sample of children who gave evidence in Auckland, Manukau, Wellington and Christchurch courts. The discussion is organised chronologically, following children from the time of their first contact with police through to sentencing. Specifically, we explore (a) the duration of police and court processes; (b) children‘s and their caregivers‘ access to information about the process and access to key participants; (c) the way in which children gave their initial statements and gave evidence at court; (d) whether children had a trial support person with them; (e) testifying on the day; and (f) the trial outcome and debriefing.

In the literature, the term child witness is used to refer to children who give evidence irrespective of whether they are a complainant or a witness to crime. In this chapter, we adopt the same terminology; however the term witness on its own will always refer to children who are giving evidence as witnesses to crime, as distinct from complainants.

Our original intention was to investigate only how long it takes to process child wit- ness trials. We envisaged gathering the relevant data from police and/or Ministry of Justice databases for analysis. However it soon became evident that information could not easily be drawn from those databases on the basis of the age of a com- plainant or witness at the time of trial.6

We therefore designed a draft questionnaire to collect this information from those with the greatest knowledge of each case, namely, the officer in charge and Crown prosecutor. We took advantage of this change in tack by collecting information on a much wider range of administrative issues than originally anticipated, drawing heav- ily on the report of the Working Party on Child Witnesses in determining which issues to collect information on. The draft questionnaire and proposed process for collecting

the information were subsequently revised in consultation with the Advisory Group to ensure, amongst other things, that we covered key contemporary concerns identified by sector representatives on that Group.

Previous research has shown that children‘s age was a factor in relation to, for example, decisions surrounding which alternative mode of evidence children used when testifying in court (see, for example, Pipe et al., 1996; Working Party on Child Witnesses, 1996). Furthermore, the law itself has in the past made a distinction between complainants and witnesses in relation to, for example, access to alternative modes of giving evidence and whether children can testify with a support person by right or by permission. We therefore wished to test two key hypotheses:

The report of the Working Party on Child Witnesses also pointed to regional variations in relation to some practices, such as in relation to children viewing their forensic interview before trial or prosecutors meeting child witnesses prior to trial. We would have liked to examine associations by region, however we decided not to pursue this because of the different sampling methods (see next section).

We note that the sample size is relatively small and there may be any number of reasons for differences found between outcome variables and the child‘s age and role, including the possibility of the difference being due to chance. Where we have conducted tests for statistical significance, we note that the results are tentative.7

Permission to collect anonymised data on the administration of trials was granted by the New Zealand Police in April 2008 and by the Auckland Central, Waitemata and Counties Manukau Police District Commanders in May 2008. The agreement of Meredith Connell, the Auckland/Manukau Office of the Crown Solicitors, had been secured in 2007 and the study was approved by AUT‘s Ethics Committee in June 2008. In late September 2008 the researchers asked permission to extend the research to include Wellington and Christchurch courts. Raymond, Donnelly & Co and

variables, we have used Chi-square tests (―The Chi-Square Test procedure tabulates a variable into categories

and computes a chi-square statistic. This goodness-of-fit test compares the observed and expected frequencies in each category to test that all categories contain the same proportion of values or test that each category contains a user-specified proportion of values‖ (definition from SPSS)). Specifically, we used the Continuity Correction result, rather than a Pearson Chi-Square, as the former is a more stringent test of significance. Where one or more cells had an expected count less than 5, we used a Fisher‘s Exact Test. To compare means (such as age or duration of processes), we used T-tests when comparing two groups or Tukey‘s (a post-hoc multiple comparison test) to compare three or more.

Luke, Cunningham & Clere, the Offices of the Crown Solicitors in those regions, agreed to participate in the research in the same month; permission was granted by the New Zealand Police and the Wellington and Christchurch Police District Commanders in early October.

Data was collected via questionnaire; these were completed jointly by the Crown prosecutor and officer in charge of each case on the basis of informed consent.8 The questionnaire also invited respondents to make any other comments they wished on the case, such as what had worked well and what had not; some of these comments are quoted throughout this chapter. Although the information did not fall under the Court Search Rules, hence did not form part of the court record, AUT‘s Ethics Committee required that the presiding judge‘s consent be gained for the information to be collected.

Throughout the data collection period, administrators at the Offices of the Crown Solicitors alerted the Principal Researcher to upcoming child witness trials and gave the prosecutor a questionnaire to complete with the officer in charge; the adminis- trators also alerted the Criminal Caseflow Managers or Jury Trial Schedulers at each of the courts in the study, who placed a consent form on the presiding judge‘s trial file.

One questionnaire was completed per child witness (i.e. under the age of 18) for the prosecution who gave evidence at a trial in the Auckland, Manukau,9 Wellington and Christchurch District Courts and in the High Court at Auckland and Christchurch. There was no straightforward way of collecting data on child witnesses called by defence, hence questionnaires were not filled out for these children;10 however we asked questionnaire respondents to note how many child witnesses for the defence gave evidence in the trials under investigation and how those children gave evidence.

Ten months‘ data was collected in the Auckland and Manukau regions (28 July 2008 to 29 May 2009);11 six months‘ in Christchurch (1 December 2008 to 29 May 2009); and just under six months‘ in Wellington (8 December 2008 to 29 May 2009).

To the best of our knowledge, 69 trials were held in the courts surveyed during the data collection period in which a child under 18 gave evidence.12,13 We received com-

pleted questionnaires in relation to 46 of these trials across the regions, involving 71 children. There were 20 trials for which no questionnaire was completed but for which the courts provided basic data. This basic data was sufficient to establish the duration of the court processes and the outcome of the trial. Hence there are two samples: one in relation to the 46 trials for which we have the full information (full data sample) and one in relation to the 20 trials for which we have basic information (basic data sample). This allowed us to report on one cohort using data from two sources (the questionnaires and the court data). We have assumed that the two sources are equally accurate as the court data was limited to dates and trial outcomes.14

Two of the 69 trials were aborted or hung and then retried within the data collection period. In these cases, we only included the data on the later trial. In a further case, the questionnaire could not be deciphered accurately and was excluded.

At the beginning of each section in this chapter, we indicate which sample(s) that section relates to. Furthermore, as there was missing data for some of the questions on the questionnaire, we also indicate the proportion of questionnaires for which the relevant information was provided.

ascertain whether the children in this sample differ significantly from children in other periods. We assume that our findings are generalisable to New Zealand child witness populations beyond the data collection period in the regions studied.

The characteristics of the full data sample (46 trials, 71 children) are given in Table

3. The majority of the children (n=48, 68%) gave evidence in the Auckland and Manukau Courts. Most of the children were complainants (n=55, 77%) and female (n=53, 75%). The children‘s ages ranged from 6 to 17, with a mean of 13.7. Three (4%) of the children in the sample had intellectual disabilities15 and the majority (73%) lived within 50km of the courthouse.

There was a trend (though not statistically significant) for witnesses (n=16) to be marginally older (mean age 14.8 years) compared to complainants (n=55, mean age 13.4; p=.074). The proportion of witnesses (as distinct from complainants) varied by region: 50% (n=9 out of 18) in Christchurch, 24% (n=5 out of 21) in Auckland, 7% (n=2 out of 27) in Manukau, and 0% in Wellington.

Role
N(%) complainants	55 (77%)	16	25	5	9
N(%) witnesses	16 (22%)	5	2	0	9
Gender
N(%) female	53 (75%)	15	23	2	13
N(%) male	18 (25%)	6	4	3	5
Age
Mean age	13.73	14.1	12.89	11.8	15.11
Median age	14	14	13	12	15
Age range	6-17	6-17	7-17	9-15	11-17
Other
N(%) intellectual disability	3 (4%)	0	1	0	2
N(%) living <50km away from court	52 (73%)	12	21	1	18

Figure 2 shows the nature of the charges alleged by the 55 complainants in the full data sample. In the vast majority of cases (82%, n=45 out of 55), the offences alleged included sexual abuse.

Figure 2 : Types of offences alleged by the complainants in the full data sample (n=55)

Much of the research literature cited throughout this chapter relates to the experi- ences of child complainants in sexual assault cases. The reason for this is evident from Figure 2: that the majority of children who give evidence in courts do so in relation to these types of offences is also noted by American researchers (Quas & McAuliff, 2009). However there are reasons to believe that the experiences of these children can be generalised to children who give evidence in relation to other crimes:

―All of the factors related to adverse outcomes in [child] sexual abuse cases ... (legal knowledge, repeated interviews, testifying, and case length/outcome) are common to virtually all types of criminal cases, not just those involving sexual abuse.‖ (Quas & McAuliff, 2009, p. 81)

In this section we examine how long it took for child witness trials to be processed, from the time of the child‘s first involvement with the police through to the time of sentencing. We also examine whether children gave evidence at preliminary hearings and whether the case was put on stand-by or adjourned.

Figure 3 shows the average time taken to process child witness trials across all regions, from the time the complaint was received by police through to sentencing. For ease of comprehension, the data is presented as if the complaint were first received by police on 1 January 2008. As can be seen, the entire process took over 20 months for the children in the current samples.

Figure 3 : Average time taken to process child witness trials (Auckland, Manukau, Christchurch and Wellington)

Police process Court process Sentencing

Jan Apr

Police receive Information Jul _Sep

complaint laid at court Trial _Sentencing

Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

Jan 08 Dec 09

Descriptive statistics for each of these processes are given in Table 4 below. While the mean for the police process was 3.35 months, the median was 1.6, indicating that for half the sample the police investigations were completed within 1.6 months.16

Table 4 : Time taken to process child witness trials (Auckland, Manukau, Christchurch & Wellington)

		Police process (months)	Court process (months)	Sentencing (months)
Number	Valid	43	65	35
	Missing	3	1	2
Mean		3.35 (=102 days)	15.68 (=477 days)	1.74 (=53 days)
Median		1.6	14.45	1.51
Std. Deviation		5.32	6.11	.877
Minimum		0	5.65	.00
Maximum		28.62	39.75	4.83

Across all regions, children waited on average 16 days between making a complaint to police and giving their initial statement; the median wait was 6½ days.

However this data must be treated with caution. Firstly, the police investigation is not necessarily undertaken in the same region as the court site; hence these figures may not accurately represent practice in the four police regions. Secondly, there was a high proportion of missing data (13%).

Thirteen percent (n=9) of the 71 children in the full data sample gave evidence at a pre-trial hearing. Five were complainants and four were witnesses, all aged 14 and above at the time of trial. Only one of the nine had been forensically interviewed.

In June 2010 (that is, after the data for this chapter was collected), a new committal procedure came into effect; as a result of this, and amendments to other statutes, it would now be highly unlikely that a child gives evidence at a committal hearing.17

Thirty-five percent (n=16) of the 45 trials were put on stand-by and 34% (n=15) of the 44 trials were adjourned at some stage, including after the trial had commenced. The adjourned trials took significantly longer to process through the courts than trials that were not adjourned (mean 18 and 13 months respectively, p=.005).

Table 5 summarises the reasons given for trials being adjourned. Not all of the reasons were preventable (for example when a participant became ill) but some were (e.g. failing to book the CCTV equipment). In two cases, the trial was adjourned because the complainant refused, or was too distressed, to testify.

―The 17 month delay in getting this matter through the Court system I believe contributed significantly to the victim‘s confusion under cross-examination. The delay experienced in child sexual abuse cases is outrageous and needs to be urgently addressed to assist in helping these victims get the best chance of positive resolution which can only occur if these matters are tried at the earliest available opportunity.‖ (Auckland region police officer)

When the Working Party on Child Witnesses reported their findings in 1996, they noted that delays in processing cases through the courts were a major issue for child witnesses. The Working Party concluded that, ―...reform in this aspect of the court process would have the greatest overall beneficial impact on children and their supporters‖ (Working Party on Child Witnesses, 1996, p. 7).

However the time taken to dispose of child witness cases in the courts has increased, rather than decreased, since that time. In 1993 complainants under the age of 17 waited on average 255 to 260 days—or around eight months—for their cases to be processed through the court (Lash, 1995, p. 13).18 The average wait for the children in the current sample was 477 days (15 months), with adjournments extending the process. The increase in court processing time more generally reflects a consistent trend over the past decade or so.19 One of the key aims of the current Ministry of Justice/New Zealand Law Commission Criminal Procedure (Simplification) Project is to reduce delays in processing cases more generally.

There are at least two ways in which delays can negatively affect children. Firstly, there is the impact on children‘s (and their families‘) functioning in the period await- ing trial. This time can be highly stressful; lengthy delays no doubt prolong the strain (Spencer & Flin, 1993) and can impair children‘s mental health (Runyan, Everson, Edelsohn, Hunter, & Coulter, 1988). Other studies have reported a range of impacts on children‘s lives during the pre-trial wait, including social exclusion when the abuse became known to peers and serious disruptions to schooling and family life (Davies, 1998; Eastwood & Patton, 2002).

Secondly, there are the potential impacts on memory. These can include a decline in the ability to accurately recall information (Poole & White, 1993, cited Fivush, Peterson, & Schwarzmueller, 2000) (although memory for distinctive and salient events tends to be remembered well even over long periods of time (Fivush et al., 2000)), a diminished ability to distinguish between the details of different episodes of

Courts—the latter by 25%; sitting times for jury trials in both courts also increased (New Zealand Law Commission, 2004, pp. 356-358). Since 2004, the median time taken to dispose of a jury trial in the District and High Courts has increased by 5½ weeks (Bazley, 2009, p. 18).

abuse,20 and the possibility of recalling and disclosing different aspects of the events over time (Fivush, 2002; Fivush et al., 2000). All of these can diminish the quality of a child‘s evidence and/or give rise to inconsistencies (apparent or otherwise) which can in turn undermine children‘s credibility.

While the waiting can be stressful, changes to the scheduling of trials can also be unsettling for children and their caregivers (Davies, 1998) and may be an additional cause of stress (Whitcomb, Shapiro, & Stellwagen, 1985). More than a third of the trials in the full data sample were adjourned at some stage and more than a third were put on stand-by, contrary to the recommendations of the Working Party on Child Witnesses.21 In one case here, the child had to prepare for two stand-by trials, neither of which proceeded, before eventually getting to trial; there were then two adjourned sentencing dates.

While there are long delays in the court process, it appears that police processes are being carried out relatively promptly: half of the investigations in the full data sample were completed within 1½ months.22 For many children, there are minimal delays between making a complaint and making an initial statement to police. Fifty percent of the children gave their first statement within six days of the complaint being made.

The average duration of the entire process recorded in the current study is similar to that recorded in a study of child sexual assault trials held in New Zealand in the 1990s (Davies, 1998). We know that the court processing time has increased since 1993; hence the similarity between these two samples must arise from changes in the duration of the police investigation. That is, police investigations are perhaps being completed more quickly now than in the 1990s, thus counterbalancing the increase in court processing time.

―The complainant gave better evidence in the second trial. She was more confident as she knew what to expect in terms of being in court, giving evidence-in-chief and being cross-examined.‖ (Manukau prosecutor)

In this section, we report on processes that help prepare children for the task of giving evidence in court. Specifically, we examine children‘s access to the Court

to have their cases put on stand-by (Criminal Jury Manager, personal communication, 9 July 2008). This may be happening in other courts too.

Authority include consideration of the timeliness of police responses to child abuse cases. The focus on the inquiry was originally on delays in the Wairarapa, but was subsequently extended to include all police districts.

Education for Young Witnesses Service (a court preparation programme run by the Ministry of Justice and delivered by Victims Advisers),23 as well as whether children were familiarised with the courtroom and CCTV room before testifying. We also examine whether caregivers were briefed on how to support their children, pre-trial interactions with the prosecutor and whether the child met the judge prior to trial. All of the information in this section is based on the full data sample.

According to the questionnaire respondents, 64% (n=45) of the 70 children participated in the Education for Courts programme. Thirty-three percent (n=23) did not. In a further 3% of cases (n=2), the respondent did not know. Two questionnaire respondents, both police officers, noted that they had not heard of the programme.

The respondents were asked ―Did the child receive Education for Courts?‖. It should be noted that the formal title of the programme is ―Court Education for Young Witnesses‖ not ―Education for Courts‖.

Officers in charge are expected to accompany child witnesses during the Court Education for Young Witnesses sessions.24 In most cases (69%, n=31 out of 45) they

therein; publications explaining court processes; felt pens; and a pamphlet for caregivers on supporting their

children. The resources include explanations of the process as well as the participants (e.g. what a judge or jury is, and what they do). In addition, the paper resources are available in English for download from the Ministry of Justice website. Police alert Victims Advisers whenever a case involves a child witness via a CSV1 form and this form is sent to the Victims Advisers; arrangements are then made for the child to come to court to undertake the Court Education for Young Witnesses programme, including visiting the courtroom and, where appropriate, the CCTV room.

attended at least one session with the child; in 18% (n=8 out of 45) the respondent did not indicate whether this had happened or not.

Seventy-one percent (n=49) of the 69 children were shown around the courtroom prior to giving evidence; 20% (n=14) were not; for 9% (n=6) of the sample, the respondent did not know.

Across all regions, it was typically the officer in charge and/or Victims Advisor who showed the courtroom to the child, although in some cases the prosecutor was also involved.

Of the 31 children, 93% (n=29) were shown the CCTV room prior to testifying (including both witnesses who used this mode). Only six of the children were given the opportunity to practise with the equipment prior to testifying; in two cases the respondent did not know.

Ninety percent (n=64) of the 71 children met the Crown prosecutor before the trial, while 10% (n=7) did not. In two cases where the child did not meet the prosecutor before trial, the file had been assigned to the prosecutor within a week of trial (one was assigned the day before).25

Across all regions, children tended to meet the Crown prosecutor two or more days before the trial (n=36 out of 64, 56%). Nine percent (n=6 out of 64) of the children

met the Crown prosecutor for the first time on the day before trial. Nearly one in five (n=12 out of 64, 19%) met the prosecutor on the day of trial. In 16% of cases (n=10 out of 64), the timing was not indicated or unclear. On average, the meetings lasted just over an hour.

Prosecutors were asked the purpose of their meetings. The questionnaire asked specifically whether they met to (a) go over evidence;26 (b) build rapport; and (c) explain processes; respondents could choose more than one reason and provide

‗other‘ reasons. As can be seen from Figure 4, the most commonly cited reasons were to build rapport with the child (97%, n=62 out of 64) and to explain processes (91%, n=58 out of 64). Other reasons given included:

Figure 4 : Reasons for prosecutor meeting the child prior to trial (n=64)
Explain process
Build rapport
Go over evidence
0
0
0
3
2
10
9
30

Twenty percent (n=9 out of 46) of cases in the full data sample were assigned to the presenting prosecutor within a week of trial.

showing the child their forensic interview or statement prior to trial to refresh the child‘s memory.

The reasons for the late assignment of trials included the original prosecutor‘s unavailability due to family reasons (two cases) or the original prosecutor having to appear in another matter (two cases). In one instance, the trial was on stand-by but went ahead unexpectedly (hence presumably the original prosecutor was not avail- able) and in one case the respondent noted only that there was a shortage of prose- cutors.

None of the 71 child witnesses in the full data sample met the judge prior to trial.

Eighty-seven percent (n=61) of the 70 children‘s caregivers were briefed on how to support their child, including briefing on evidential issues.

In most cases the information was given verbally. In four cases, the briefing included written information. Information was most often provided by the officer in charge (45 cases), the officer in charge and the prosecutor (six cases), or by Victims Advisers or Victim Support with or without the officer in charge (five cases). In a further five cases the information was provided by the officer in charge as well as other professionals, such as counsellors.

There were no significant associations between children‘s age and role with children‘s access Court Education for Young Witnesses, having a courtroom tour, or whether children‘s caregivers were briefed. Younger children and complainants, however, were more likely to meet the Crown prosecutor before trial (see Table 6). It should be noted that overall complainants were marginally younger than witnesses and there were few witnesses in the sample, hence reiterate that this finding is tentative.

Children‘s understanding of criminal justice processes, the roles of the various par- ticipants and legal terminology is often sketchy at best (Crawford & Bull, 2006; Freshwater & Aldridge, 1994; Quas & McAuliff, 2009; Saywitz, Jaenicke, & Camparo, 1990) or wholly inaccurate at worst. For example, Sas (1991) (cited Quas, Goodman et al., 2005, p. 15) found that younger children‘s fears of testifying included fear of being sent to jail. Nor does participating in the processes automatically lead to improved understanding (Davies, 1998; Freshwater & Aldridge, 1994). When a child does not understand what is happening or what to expect, their anxiety may increase, both while awaiting trial and while testifying:

―Children‘s lack of understanding may also affect how distressed they are during the case (which can then affect their ability to communicate and susceptibility to false suggestion while answering interview questions; Goodman et al., 1992; Nathanson & Saywitz, 2003). Specifically, several lines of research indicate that poor understanding of an impending stressor is associated with increased distress. For instance, in medical contexts, children are significantly more anxious when they do not know what will happen during a procedure that they are about to undergo than when they are knowledgeable about the procedure ... Similar findings have emerged in legal contexts: Goodman et al. (1998) administered a legal knowledge questionnaire to 5- to 8-year-olds, some of whom later testified in mock trials. Greater legal knowledge was associated with lower levels of anticipated stress while testifying.‖ (Quas & McAuliff, 2009, p. 83)

Children‘s understanding may be improved (and anxieties alleviated) through, among other things, court preparation programmes, visiting the courtroom, seeing the CCTV room and practising with the equipment, if applicable, and meeting key figures beforehand, particularly the prosecutor and judge. The evaluation of New Zealand‘s pilot Court Education for Young Witnesses programme confirmed that the programme helped reduce the levels of anxiety experienced by children and their caregivers:

―Young witnesses and parents/support people said that attending the Court Education service helped reduce levels of worry, fear and anxiety as they had an opportunity to become familiar with the court and the role of the people in the courtroom. The Court Education took away a certain amount of fear and misunderstanding, making young witnesses feel more confident and relaxed about the experience.‖ (Davies et al., 2003, p. 116)

Children interviewed in other common law jurisdictions have also confirmed that visiting the courtroom in advance helped them know what to expect (Cashmore & Trimboli, 2005; Hamlyn, Phelps, Turtle, & Sattar, 2004) and that meeting the prosecutor and judge beforehand had helped them cope with the court process

(Plotnikoff & Woolfson, 2004). Here in New Zealand, the Working Party on Child Witnesses noted that, ―Children who have been given the opportunity [to practise with the equipment] have been found to be more competent in using the equipment and more comfortable giving evidence‖ (Working Party on Child Witnesses, 1996, p. 23).

Only two-thirds of the children in the current sample attended the Court Education for Young Witnesses programme. That two police officers did not know about the programme suggests that police briefing is not adequate; however, it is also possible that the officers knew the programme by another name. However police officers were often involved in Court Education for Young Witnesses sessions with the child, as well as briefing those children who did not attend.

Only 71% of children saw the courtroom before testifying, but most children who used CCTV did visit the room prior to giving evidence. Letting children practise with the equipment before testifying is not a standard procedure. However, as the CCTV equipment is hired on an as-needed basis in many courts, it may not be feasible for children to practise if their court visit takes place before the day of trial.

The majority of children met the prosecutor before trial; it may be that this is more likely to happen with younger children and complainants. However nearly one in five did not meet the prosecutor until the day of trial, a practice criticised by the Working Party on Child Witnesses.27 Across all regions, the purpose of the meetings included explaining processes and building rapport. In no case did a respondent indicate that the meeting was to gauge the child‘s level of language development in order to tailor their questions in court. Finally, in no case did a child meet the judge before giving evidence.28

While it is important that children understand the process that they are participating in, it is also important that briefing occurs in a timely manner if it is to ease anxiety. Children usually access the Court Education for Young Witnesses programme within a week of trial and may meet the prosecutor just before trial, yet children‘s need for information begins many, many months before this:

improve the quality of the children‘s evidence by putting the children at ease (Criminal Justice System, 2007, section 5.68), although care is taken to ensure this does not ―create an impression of bias in favour of the witness‖ (section 6.12). Because of the inevitable difficulties of scheduling such a meeting before the trial date, if this were to become standard practice in New Zealand, the meeting would no doubt have to take place on the day of trial and with the consent of all parties. In a recent review of British children‘s experiences of court, the comments of some young witnesses showed that being introduced to the judge was appreciated (Plotnikoff & Woolfson, 2009); in an earlier British study, children reported that meeting the judge helped them deal with the court process (Plotnikoff & Woolfson, 2004).

―The international reviewer [of the pilot] felt that [having the sessions close to trial] was of concern as she said the worry, fear and anxiety starts earlier for children, usually around the time they know they will have to attend court.‖ (Davies et al., 2003, p. 114)

The pre-trial period can be highly stressful for children and their families. Caregivers may have difficulty managing the tension between supporting their child and fear of contaminating the child‘s evidence29 and may themselves be uncertain about the police and court processes. As the Working Party on Child Witnesses noted, ―Care- givers are better able to support children if [the caregivers] are properly informed about the process‖ (Working Party on Child Witnesses, 1996, p. 18).

In most cases here, the children‘s caregivers were indeed briefed on how to support their child, including information on evidential issues. Caregivers tended to be given information orally rather than in writing; we did not seek to ascertain the quality, accuracy, appropriateness or timeliness of those oral briefings, although the range of sectors involved (police, Victims Advisers, prosecutors, counsellors) suggests that briefing may be occurring at different points throughout the process. In only four cases did the questionnaire respondent indicate that written information had been provided, despite the fact that a pamphlet on how to support child witnesses is included in the Court Education for Young Witnesses packs and is also available on the Ministry of Justice website.

All children should (amongst other things) meet the prosecutor before trial, attend the Court Education for Young Witnesses programme, be shown the courtroom and their caregivers should be briefed on how to support them. For 58 out of the 71 children in the full data sample, respondents indicated whether or not all four of these events had taken place:30 for two of the 58 children (3%), none of these events took place; for seven children (12%) only one or two occurred; for 16 (28%), three took place and for 33 children (57%) all took place.

In this section we examine the way in which children gave their initial statements to police and explore a range of issues in relation to forensic interviews. We also examine whether children were asked how they wished to testify and the modes of evidence they went on to use in court.

cases, these concerns also resulted in caregivers withholding support. Both these situations are undesirable and

unnecessary if adequate advice was available‖ (Working Party on Child Witnesses, 1996, p. 18).

Sixty-six percent (n=46) of the 70 children in this sample gave their initial statement via forensic interview; 6% (n=4 out of 70) gave a forensic interview as well as a signed statement. A sizeable minority gave a signed statement only (n=18, 26%) while two children‘s statements were taken on audio tape (3%).

Information on the child‘s age at the time of their initial statement was provided for 63 (89%) of the 71 children. All of the 30 children aged under 13 were forensically interviewed; 60% (n=20 out of 33) of children aged 13 and over were forensically interviewed.

Eighty percent (n=44) of the 55 complainants were forensically interviewed, compared to 40% (n=6) of the 15 witnesses.31

In 77% (n=37 out of 48) of cases, the Crown prosecutor viewed the forensic interview prior to trial. Twenty-three percent of tapes (n=11 out of 48) were not viewed. The 11 tapes comprised nine interviews with complainants and two with witnesses; all of the tapes were played in court as part of the children‘s evidence-in- chief.

In most cases (65%, n=30 out of 46) children viewed their forensic interview prior to trial. Sixteen of the 46 children (35%) did not. In 14 of these 16 cases, the tapes were shown in court as part of the child‘s evidence-in-chief.

In 16% (n=8) of the 49 cases, the methodology of the forensic interview was challenged by one of the parties. All of the tapes which were challenged were eventually shown in court as the child‘s evidence-in-chief.

Six forensic interviews were contested by defence on the grounds of leading questions. In one further case, the Crown requested that irrelevant information be edited out. In the final case, there was a request to edit out reference to the accused

Was the forensic interview admitted in court as the child’s evidence-in-chief?

Ninety-four percent (n=47) of the 50 forensic interviews were eventually shown in court as the child‘s evidence-in-chief. We asked respondents why interviews were not shown in court. The reasons given were not always enlightening, but in one case the respondent noted that this was the child‘s preference.

In 48% (n=23) of the 48 cases, a copy of the forensic interview was removed from the police station by a defence lawyer.

CAUTION: Questionnaire respondents were asked whether children were consulted over their preferred modes of giving evidence at court. We did not ask for details on how children were consulted; nor did we ask for details of the information provided to children to help them form a view. We cannot discount the possibility that the children‘s preferences were influenced by those consulting with them. Hence we warn against drawing firm conclusions on children‘s actual preferences from this data in this section.

Questionnaire respondents reported that 40% (n=28) of the 70 children were asked how they preferred to give their evidence-in-chief in court and 18% (n=13) were asked their preferences for cross-examination.

Sixteen percent (n=3 out of 19) of children under 13 were asked their preference for either evidence-in-chief and/or cross-examination; 51% (n=26 out of 51) of older

children were asked. Only one of the three children with intellectual disabilities in the sample was consulted.

In total, 29 of the 70 children (41%) children were asked how they wanted to testify at court. Twelve had been forensically interviewed (nine complainants and three witnesses); all but one of the 12 preferred to use the interview as part of their evidence-in-chief.32 They preferred to give the rest of their evidence via CCTV (n=7)33 or with a screen (n=5).34

Sixteen of the children had not had a forensic interview. This group preferred to testify via CCTV (n=2),35 with a screen (n=6),36 or with no alternative mode (n=8).37

Finally, there was one 9-year-old child who had undergone a forensic interview, but was not asked his preference in relation to evidence-in-chief. He was, however, asked his preference for cross-examination and elected to testify with a screen.

The children‘s wishes were granted in 79% (n=23 out of 29) of cases. However, in four cases, the child preferred CCTV but testified with a screen instead.38 In a further two cases, the child preferred to give evidence without an alternative mode and did so for evidence-in-chief, but used a screen for cross-examination.39 All children who preferred to have the forensic interview played at court were able to do so.

Eight of the 10 children who gave evidence in the ordinary way had been asked their preference and opted to testify in this way.

Figure 6 : Alternative modes of evidence used at court by the 71 child witnesses

Forensic interview Forensic interview CCTV only (n=1) Screen only (n=13) Ordinary way
+ CCTV (n=32) + screen (n=15) (n=10)

As shown in Figure 7, all 20 children under 13 gave their evidence-in-chief via forensic interview and the rest of their evidence via CCTV (n=16, 80%) or with a screen (n=4, 20%).

For the 51 older children (those 13 and over) the modes used were more varied: 31% (n=16) used a combination of forensic interview/CCTV; 22% (n=11) used a combination of forensic interview/screen; 25% (n=13) testified with a screen; 2% (n=1) gave all the evidence via CCTV; and 20% (n=10) gave all their evidence in the ordinary way.

Figure 7 : Alternative modes of evidence used (by child’s age) (n=71)
0-12 year olds (n=20) 13-17 year olds (n=51)
0
2
0 0 0
20

The 10 children who gave evidence in the ordinary way were all over 15 years old. Seven were 16- or 17-year-old witnesses in sexual assault trials (five males, two females); five of these had indicated that they preferred not to use an alternative mode. Three were 15- to 17-year-old complainants in aggravated robbery or sexual assault trials (one male, two female); these children had all opted to testify in the ordinary way.

Three 15-year-old child witnesses were called by defence; they were called in relation to two separate trials. None used an alternative mode for giving evidence.

Table 7 shows the associations between outcome variables examined in this section and the child‘s age and role. As can be seen, age was a significant factor in relation to children‘s initial statement to police, whether they were consulted over which modes they used to testify and the modes of evidence used at court. Role was a significant factor in relation to children‘s initial statement to police and viewing their forensic interview before trial, although once again we note the small number of witnesses hence these findings are tentative.

In relation to children‘s initial statement to police, there were significant patterns in terms of both role and age. Given that witnesses in this sample were marginally older than complainants, it is possible that the pattern in relation to role is mediated to some degree by age. However, as we will see in Chapter 4 where we examine applications sought for modes of evidence with a different and larger sample, the same pattern in relation to role emerged although witnesses and complainants did not differ significantly in age.

Table 7 : Associations between alternative modes of giving evidence and age/role

		N (%)
	Complainants	Witnesses	Role	total		Mean	age
Children‘s initial statement to police
Forensic interview	44 (80)	6 (40)	50**			11.36**
Other	11 (20)	9 (60)	20**			14.77**
Did the prosecutor view the forensic interview prior to trial
Yes	33 (79)	4 (66)	37		12.7
No	9 (21)	2 (33)	11		12.45
Did the child view the forensic interview prior to trial
Yes	30 (75)	0 (0)	30**		12.6
No	10 (25)	6 (100)	16**		13.25
Consultation with children over their preferred mode of giving evidence at court
Yes	20 (37)	9 (56)	29		14.86**
No	34 (63)	7 (44)	41		13.12**

In the New Zealand courts, there is a range of ways in which child witnesses can give their evidence. Firstly, New Zealand has developed a specialised forensic interviewing service whose members, police and statutory social workers, are jointly trained in communicating with children; they are expected to attend to evidential as well as care and protection issues. Each interview is videorecorded and the prosecution can apply to have that interview played as part of the child‘s evidence- in-chief at court. The rest of their evidence can be given via CCTV, in the courtroom with a screen between the witness and the accused, or in the ordinary way (that is, in the courtroom in full view of the accused).

One of the rationales for these special measures at court is to reduce the stress of testifying, thereby enhancing children‘s (and adults‘) ability to provide good quality evidence. While studies have shown that these measures do indeed reduce the stress of testifying (Davies & Noon, 1991; Davies, Wilson, Mitchell, & Milsom, 1995; Goodman et al., 1998; Murray, 1995), one study suggests that it was not so much the mode, but children‘s choice over the mode, which made the difference: children who gave evidence via CCTV in that study found testifying easier than those who

wanted to use CCTV, but did not have that opportunity (Cashmore & De Haas, 1992).

Children are often good judges of their own needs; however, it may be difficult for them to anticipate how they will cope with testifying in court. Hence children and their caregivers must be well informed if they are to make good decisions about how to testify and, in most cases, will be relying on knowledgeable adults to help them make the decision. Given that applications to use alternative modes are often filed many months before the trial, careful consultation needs to take place as soon as children know they will have to testify in court.

The data suggests tentatively that age may be a factor in the children‘s chances of being consulted about their preferred ways of testifying: younger children were consulted less often than older children.43 However age may also be a factor in relation to children‘s use of alternative modes of giving evidence: a higher proportion of younger children were forensically interviewed compared to older children. Hence the option of using a forensic interview as part of the child‘s evidence in court may be more often available to younger children. The data also suggests a pattern whereby CCTV was used for younger children and screens for older children.44 This is despite the fact that there is no presumption in the Evidence Act 2006 that forensic interviews and CCTV are appropriate for younger children only.45

There may also be patterns depending on the child‘s role: witnesses may be less likely to be forensically interviewed than complainants and no witness saw their forensic interview again before trial.

As noted above, a higher proportion of younger children underwent a forensic inter- view than older children. We did not ask questionnaire respondents to explain the basis for deciding whether or not to have a child forensically interviewed; however, it is possible that some decision-makers believe that these interviews are most appro- priate for younger children, particularly complainants. Yet there was at least one case in the current sample where a child in her late teens would have benefited if a forensic interview had been available. This child was testifying in relation to repeated and serious abuse. She hid her face while giving evidence and was unable to talk at trial about some of the offences she had disclosed previously; she withdrew part-way through giving evidence by concealing herself under the desk in the CCTV room. One of the prosecutors on the case noted that:

recorded video is more likely with younger children. (Indications are that an upper age level has developed in some areas: for example, up to 13 years in one area, and 10 years old in another). Similarly the older the child, the more likely that an order would be sought that the child would be screened and present in the courtroom. Indications are, however, that where possible the preference for many Crown Solicitors is to have children given their evidence in the courtroom‖ (Working Party on Child Witnesses, 1996, p. 21).

―It was unfortunate there was no evidential interview of the complainant because she had to give all of her evidence-in-chief. A video record would have shortened the process for us and guaran- teed we had proof of each count. She found it very difficult to give evidence.‖

The availability of a forensic interview, in this prosecutor‘s opinion, would have benefited the child and ensured the accused could be tried on all of the disclosed offences. Other benefits of having a forensic interview available include (a) that the prosecutor can view it before trial; and (b) children can also view the tape prior to trial which may serve to refresh their memories. In a surprisingly high number of cases in the current sample, the prosecutor and/or child did not view the forensic interview before trial.

Commentators have argued that pre-recorded testimony offers a range of benefits for the child as well as for the legal process more generally, irrespective of the age and role of the child (see, for example, Cashmore, 2002; Edelstein et al., 2002; Spencer & Flin, 1993; Whitcomb et al., 1985). Some of these benefits are summarised in Table 8 below.

Having said that, some questionnaire respondents highlighted deficiencies in the forensic interviews in the current sample, including a complaint that the forensic interviewer did not explore whether there were any witnesses to the offences (it came out under cross-examination that there was indeed a witness and the case was aborted) and another complaint that the forensic interviewers did not explore the disclosures sufficiently:

―Success of the evidential interview [was] limited by a lack of questions that explored in depth the child‘s disclosures. This meant that 19 months after the interview it was difficult for the child to explain answers given in their EVI.‖ (Manukau prosecutor)

As we have seen, a small percentage of the forensic interviews were challenged by defence on the basis of leading questions, although the interviews were all eventually shown in court.

effect of videotaping a child‘s early statement: it tends to prompt a guilty plea when viewed by defendants and their attorneys. Apparently the defense reasons that a child who performs well on videotape will perform equally well in court—an assumption that has not been empirically tested. This effect was reported to us in telephone interviews with prosecutors across the country‖ (Whitcomb et al., 1985, p. 60).

A number of objections have been raised to videorecorded evidence more generally,48 including the possibility of the tapes falling into the wrong hands (MacFarlane, 1985). Nearly half of the forensic interviews in the full data sample were removed from the police station by the defence lawyer. This is permitted under ss29-34 of the Evidence Regulations (2007) (as amended by the Evidence Amendment Regulations 2008), which allow for a copy of a complainant‘s or witness‘ forensic interview to be supplied to lawyers acting for a party to the proceedings. This lawyer‘s copy must be kept ―in safe custody‖ and can only be viewed for specified purposes and always in the presence of the lawyer. The copy must be returned to the police as soon as possible after the case is concluded.

Given the delays in processing these cases, this means that the copy may remain in the lawyer‘s possession for over a year. We do not know whether the tapes are indeed kept ―in safe custody‖ or whether they could be removed or copied and circulated without the lawyer‘s knowledge. Police have an exhibit-handling process and are looking at national monitoring of videorecorded forensic interviews as part of case management.49 Nonetheless, prior to the Evidence Regulations 2007, videorecorded interviews in relation to criminal proceedings could only be viewed in a police facility. This had the benefit of ensuring that the state‘s duty of confidentiality towards child witnesses remained the responsibility of a statutory agency, thereby ensuring that tapes were not copied or otherwise handled in ways contrary to the Evidence Regulations.

When the Working Party on Child Witnesses was reviewing practice in the 1990s, the law allowed complainants in sexual abuse cases to testify with someone near them to provide support. Under s79 of the Evidence Act 2006, which came into effect in 2007, this provision was extended to all complainants by right and to witnesses with the permission of the judge. In this section, we examine how many children in the current sample did in fact testify in the presence of a trial support person, who that trial support person was, how close to trial the support person was designated and whether the support person was briefed on how to conduct themselves in that role.

Seventy per cent (n=50) of the 71 children had a trial support person with them when they gave evidence.

In most cases where the child had a trial support person, that person was a member of the child‘s whanau or a friend (n=26 out of 50, 52%). Twelve of the 50 children (24%) had a counsellor, social worker or Child Youth and Family caregiver accompany them while they gave evidence; 10 (20%) were accompanied by court personnel.

Across all regions, the trial support person was most often decided upon before the day of trial (58%, n=22 out of 38); in a substantial minority of cases (39%, n=15 out of 38) this was decided on the first or second day of trial. In the final case, the date supplied was clearly incorrect. However the high proportion of missing data makes it difficult to reliably interpret this information.

In four of the 15 cases where the support person was designated on the day of trial, court personnel undertook the role.

A support person may be restricted from, amongst other things, touching or talking to a child witness while the child testifies; hence support persons need to be briefed on their conduct while supporting a child witness on the stand.

In 10 cases, the support person was a member of the court staff and presumably needed no briefing. In another 10 cases, the respondent did not know whether the support person had been briefed on their role. Twenty-four of the 27 caregivers were briefed and three were not.

Defence objected to the support person selected for three of the children who gave evidence. One of the children was a 16-year-old complainant; the other two children were witnesses, aged 8 and 11.

Table 11 suggests that complainants and younger children were more likely to testify with a trial support person. Given that witnesses were marginally older than complainants, it is possible that the pattern for complainants is mediated by age.

Table 11 : Associations between child having a trial support person and child's age and role

Underpinning the provisions of s79 of the Evidence Act is the rationale that having a loved one as a support person provides emotional support for children while they testify (New Zealand Law Commission, 1996); indeed, testifying in the presence of a non-offending parent or loved one was associated in one study with the child‘s being able to answer more of the prosecutor‘s questions, appearing less fearful of the defendant and being more consistent about peripheral details (Goodman et al., 1992).

As we have seen, 70% of the children in the full data sample had a trial support person with them while they gave evidence but older children and witnesses may be less likely to do so.

Most children had a family member or friend perform that role, although a sizeable minority were supported by a member of the court staff. Of the cases where the date on which the support person was designated is known, nearly 40% (n=15 out of 38) were designed on the first or second day of trial—in four of these cases a member of the court staff took the role. In two of the three cases where defence objected to the designated support person, court personnel eventually took the role.

When the Working Party on Child Witnesses reported their findings in 1996, they noted that:

―The Working Party has been told about situations where there have been problems associated with arranging support people. These situations include failure to arrange for an appropriate person to support the child, including the need to change the designated support person at a late stage through failure to appreciate that a person to give evidence later in the trial cannot support the child who has to give evidence first.‖ (Working Party on Child Witnesses, 1996, p. 36)

It is possible that the cases where the support person was designated close to trial in the current sample were due to similar problems. This may be unsettling, particularly if the child is expecting to have a significant adult or peer with them, only to find that someone with whom they have little rapport—e.g. court personnel—takes that role. In such instances the protective potential of the measure is presumably reduced.

The impact on the support person of hearing the child‘s evidence also needs to be taken into account if last-minute changes are to be avoided. The following quote clearly articulates this point:

―Dad (not a witness) had been lined up to be the support person for this witness and her sister. On the day, he decided he couldn‘t do it as he didn‘t want to know the details of what had happened to his girls. Court staff and Victim Support eventually sat in CCTV room with victim but difficult. Dad‘s decision should have been made earlier and with thorough discussion either with police, prosecutors or victim support.‖ (Auckland respondent)

The high proportion of cases where the support person was not briefed or the respondent did not know whether this had happened may indicate that briefing is unsystematic or that briefing was being undertaken by others but the questionnaire respondent was not aware of this (e.g. briefing of support people is part of the Victims Adviser role).50 We did not ask whether children had been briefed on what they can and cannot expect from their support persons while testifying.

In this section we explore children‘s experience of testifying at trial. Specifically, we examine the number of days children come to court, the time spent waiting in the courthouse before taking the stand, the time of day children began their evidence

and the time spent giving evidence. We also report on children‘s and defence lawyers‘ demeanour during questioning.

Most of the children in the full data sample completed their evidence on the first day that they came to court (58%; n=41 out of 71). A sizeable minority (35%, n=25 out of 71) came to court on two days and 6% (n=4 out of 71) came on three. One child came to court on four separate days.

However, the children did not necessarily give evidence on each of the days they came to court. Nine of the 71 children (13%) had the experience of coming to court but not giving evidence on that day; one child came to court over four days but only gave evidence on two.

The reasons given for the child not giving evidence on a given day included (a) the trial moving slowly (e.g. because of a problem with the jury or pre-trial taking longer than expected); (b) the prosecutor or defence lawyer was ill; or (c) no further evidence was required from the child. In some cases, witnesses came to court to support a complainant even if they were not testifying on that day.

Adding together the time spent waiting on each day children came to court, the children spent on average 2:42 hours in the courthouse waiting to give evidence; the shortest wait was 9 minutes and the longest 8:21 hours.51 For children who did not give evidence on the day, the average wait was 4:30 hours before they left the courthouse for the day.

On their first day of giving evidence, most children (64%, n=38 out of 59) took the stand at or after midday. The youngest child in the sample was aged 6 and began giving her evidence at 2:40pm, having arrived at court at 10:30am.

Of the 30 children who came to court on a second (or third) day to give evidence, all but one took the stand in the morning.

Children in this sample spent on average three hours giving evidence, from the time they first took the stand to the time they finished (i.e. including breaks and including the playing of a forensic interview, if any). The shortest time recorded was 16 minutes and the longest 11 hours.52

In this discussion, the ―forensic interview group‖ refers to those children whose forensic interview was admitted as part of their evidence-in-chief and the ―live evidence group‖ refers to those who gave the entirety of their evidence live in court.

Figure 8 shows for how long the children in each group gave evidence in court and indicates the average duration of each type of evidence (i.e. evidence-in-chief, cross- examination, re-examination).

Figure 8 : Duration of child’s entire evidence at court (by role and type of evidence)53
Witnesses (live evidence group)

0
11
17
31
20
31
0
85
51
40
35
28
60
Re-examination Cross-examination Forensic interview
Evidence-in-chief
7
3
80
0
Minutes (mean)
62
100
59
120
10
140
16
180

Accusing the child of lying: 69 (99%) of the 70 children who were cross-examined in the full data sample.

Respondents were asked to rate the children‘s and lawyers‘ demeanour during questioning in court. Specifically, they were asked whether the child wept while being questioned and whether they said they did not understand lawyers‘ questions. We also asked whether the defence lawyer accused the child of lying and raised his/her voice at the child.

The results are shown in Figure 9. As can be seen, both younger and older children wept while giving evidence, including complainants and witnesses, males and females. That is, 10% (n=2 out of 19) of younger children wept during evidence-in-

cross-examination x 9; re-examination x 4. Live evidence group witnesses—evidence-in-chief x 10; cross-

chief while 18% (n=9 out of 50) of older children did; 35% (n=7 out of 20) of younger children wept during cross-examination while 25% (n=12 out of 48) older children did.

Both younger and older children indicated that they had not understood a question posed by the prosecutor (42% (n=8 out of 19) and 29% (n=14 out of 49) respectively). Eighty percent (n=16 out of 20) of younger children told the defence lawyer they hadn‘t understood a question, as did 71% (n=34 out of 48) of their older counterparts.

Defence accused both younger children and older children of lying at similar rates (65% (n=13 out of 20) and 65% (n=32 out of 49) respectively). Defence lawyers also raised their voices at 21% (n=4 out of 19) of younger and 27% (n=13 out of

49) of older children. Likewise with witnesses: 27% (n=4 out of 15) of witnesses were accused of lying and 25% (n=4 out of 16) were spoken to with a raised voice.

Figure 9 : Children’s and defence lawyers’ demeanour during questioning at trial
Defence raised

Child wept Child wept (cross- Child did not
(evidence-in-chief) examination) understand
prosecutor

Table 12 shows that there were no significant patterns between the time at which children began their testimony on their first day of testifying and the child‘s age or role.

Table 12 : Associations between timing of testifying and child’s role and age

Children will often be apprehensive about testifying in court; hours spent waiting to take the stand on the day of trial can add to the strain (Goodman et al., 1992; Plotnikoff & Woolfson, 2004; Flin et al., 1988, cited Spencer & Flin, 1993; Wade, 2002). These stressors have the potential to impact negatively on the quality of evidence a child can give:

―It appears that waiting for a significant period of time before being called to the witness-box can have a disastrous effect on the quality of a child‘s evidence. We have been told by both prosecuting barristers and procurators fiscal of cases which have collapsed because the child was unable to give coherent evidence because she or he had become too tired or over-anxious while waiting several hours to be called.‖ (Spencer & Flin, 1993, p. 386)

Anecdotes from the current study showed that some children became overwhelmed while testifying. For example, a 12-year-old complainant arrived at court at 9.30am, was taken to the CCTV room at 12.18pm where she waited until 1.00pm. The court adjourned for lunch and she returned to the CCTV room at 14.15pm where she waited until 14.29 to begin her evidence. She collapsed in the CCTV room at 4.45pm while giving evidence (the child returned on the next day to complete her evidence and the case proceeded to sentencing).

Limiting the waiting time in court was a key recommendation of the Working Party on Child Witnesses, as well as ensuring that young children do not begin giving evidence late in the day (Working Party on Child Witnesses, 1996, p. 35). However, as we have seen, children waited in the courthouse for a total of nearly three hours

on average across all days to testify; many children began testifying in the afternoon on their first day of giving evidence. They then faced lengthy periods of questioning. Some children had to return to court over a series of days.

Testifying is itself a stressor for many children. While measuring that stress is difficult, overt displays such as weeping are indicative. Thirty percent of the children in the full data sample were reduced to tears while being questioned. In one case, a witness became distraught when cross-examined over inconsistencies between his initial statement and courtroom testimony:

―The defence counsel cross-examined the child on two small irrelevant inconsistencies between the statement taken shortly after the incident (approx 2 years earlier) and the child‘s evidence- in-chief. The inconsistencies were due to memory lapses, but left the child thinking he had messed up the whole case for the victim [...]. He broke down in tears at the end of his evidence and had to be consoled by family and by the O/C who all reinforced to him what a great and courageous thing he had done to give evidence.‖

In another case, although the witness did not break down on the stand, the questionnaire respondent believed that the full impact of the crime hit him while testifying:

―Witness very nervous when giving evidence. Mind went blank. Legal argument at end of day one about refreshing memory which was opposed. Witness extensively cross-examined by defence. Coped, but stress of cross-examination on him clearly showed—lunch break was well placed. Impression gained was that during the course of his evidence the full enormity of [the crime] was brought home.‖

Cross-examination, being accused of lying and being yelled at are often cited by children as particularly difficult aspects of the court experience (Cashmore & Trimboli, 2005; Dezwirek-Sas, 1992; Eastwood & Patton, 2002; Hamlyn et al., 2004). Yet in the current study, being accused of lying was the norm and the defence lawyer raised his/her voice at four of the 19 under-13-year-olds and 13 of the 49 older children. Furthermore, nearly all of the children in the sample said that they had not understood questions posed by the defence lawyer and sizeable minorities said that had not understood prosecutors‘ questions (see Chapter 3 for a detailed analysis of the language posed to children in court). The police officer in one case illustrated how a child‘s evidence faltered in the face of complex questions, defence using a raised voice, and after significant delays in the court process:

―Cross examination of the child was the biggest problem in this case. Defence counsel asked a series of ‗double negative‘ type questions and used words which I believe were not in her normal vocabulary. This in turn caused her to become somewhat confused about the answers she should be giving. At the beginning of cross examination she started out very clear in giving her answers, however as time wore on and she became more confused, her answers faltered and she started responding with answers which were clearly not correct.

It was also clear when defence counsel questioned her in a firm tone or slightly raised voice, she tended towards agreeing with the question. This suggests to me a typical childlike response to an adult reprimanding a child and the child giving the response she thinks will please that adult. ... The 17 month delay in getting this matter through the Court system I believe contributed significantly to the victim‘s confusion under cross-examination. ‖ (Auckland region police officer)

Clearly, testifying can be distressing not only for complainants, but for witnesses too; not only for younger children, but also for older children. Interwoven with the impact on the children themselves is the impact on the quality of their evidence: experimental and court observation studies have shown that stress can inhibit the completeness and accuracy of children‘s evidence (Cashmore, 2002, p. 208).

There were 55 complainants in the full data sample and 20 trials in the basic data sample. We received information on the outcome of the trial for 72 of these cases. The outcomes are summarised in Figure 10.

Eighty-two percent (n=56) of the 68 children were debriefed after giving evidence.55

Debriefing was most often done by the officer in charge (n=44). In the other cases, debriefing was undertaken by either the officer in charge and prosecutor (n=5), the prosecutor (n=5), the officer in charge, prosecutor and Victim Support (n=1) or a counsellor/social worker (n=1).

In the 1990s, the Working Party on Child Witnesses noted that debriefing of child witnesses after they had given evidence was haphazard:

―A number of sources referred to the haphazard nature of after court debriefing. While children and their caregivers will be advised of the outcome of the trial or hearing, there is often little other help for them to deal with the aftermath of the judicial process. This may be especially important in cases where the accused is ultimately found not guilty. The literature review, for instance, notes that research in the area suggests that when the accused is not convicted, some young child witnesses are leaving the court with the perception that the adults think they lied so they will be punished.‖ (Working Party on Child Witnesses, 1996, p. 41)

Where debriefing did not occur in this sample, the trial outcomes included two not- guilty verdicts, two discharges, and two where the accused was acquitted on some charges but there was a hung jury on others. These are all situations where debriefing is particularly important.

Throughout this chapter, we have seen how involvement in the criminal justice system unfolded for a group of children who gave evidence in six of New Zealand‘s courts over 2008-2009. One conclusion to be drawn from this chapter is that many of the recommendations proposed by the Working Party on Child Witnesses in the 1990s have yet to be fully realised. That is, while there have been improvements (such as the development of the court preparation programme, extension of eligibility to alternative modes of giving evidence and having a support person), the data tentatively suggests that children‘s access to a range of preparatory and protective measures may sometimes depend on their age and their role. Yet we have seen that younger as well as older children, witnesses as well as complainants, can find testifying distressing; these measures need to be available to all children, irrespective of age or role.

There was, however, one factor common to all cases: the substantial delays in the court process. When the Working Party on Child Witnesses examined the plight of child witnesses in the criminal courts in the 1990s, they identified delays as a priority for reform. But the time taken to process these cases has increased since that time, reflecting substantial increases in court caseloads, despite increases in productivity:

―The crime rate ... has remained relatively static for the past decade but that stability is not reflected in the caseload of the criminal courts. I am told that the increase in the list court workload arises partly from the higher apprehension rate by police, and by an albeit slight decline in the use of police diversion. ... An increase in the proportion of defended cases has also added to the workload of the courts. ... while an increasing number of cases are being dealt with given the increases in productivity, this is not keeping up with the increasing number of cases (and more complex cases sometimes involving one or more defendants) which are being filed.‖ (Smith, 2007, pp. 115-116)

The joint New Zealand Law Commission/Ministry of Justice Criminal Procedure (Simplification) Project, established in 2007, has as one of its key aims the reduction

of court delays; some measures have already been introduced which are expected to assist in achieving this (e.g. the change in committal processes which came into effect in June 2009, after the data reported on here was collected). However, pressure on the courts may also increase under the current government‘s policy of boosting police numbers.

There are two ways of reducing the delays in processing child witness trials: reducing demand on the courts or modifying the way in which these cases are processed. The first option is unlikely in the current climate. As for the second, there is a range of possibilities. One is to fast-track child witness cases. There have been Practice Notes from Chief Judges to this effect since at least 1992; however this has not been sufficient to reduce delays to an acceptable level.56 Another is to increase resources for these trials. This has already happened in some courts, such as the Auckland District Court, where a new courtroom designated solely for child witness trials opened on 5 October 2009; as a result, some trials set down for 2010 were brought forward by several months to 2009.57 A further courtroom opened in March 2010. However a more sustainable approach may be to explore innovative ways of capturing children‘s evidence much earlier in the process. We return to this point in Chapters 5 and 6.

Whichever approach is taken, we should be able to guarantee that children will be treated with dignity and that children will not be questioned inappropriately or humiliated on the stand. The following quote, from the police officer in an Auckland case, shows that we still cannot make these guarantees:

―This case has angered me like no other. The victim was ridiculed and called a liar by defence counsel despite the offender admitting to everything the victim said he did in a video interview I did with him. Defence counsel made the jury laugh when cross-examining the victim. [Defence made reference to the child‘s weight and clothing]. He was rude and aggressive.‖

In the next chapter, we examine transcripts of children‘s evidence to assess the appropriateness of the language used, as well as the tactics of defence lawyers when questioning child witnesses. As will be seen, many of the concerns raised about the language and tactics of lawyers in the 1990s (Davies et al., 1997) remain causes for concern today.

Over the past decade, a series of studies has been published examining the way in which child witnesses are questioned by forensic interviewers and/or lawyers in the New Zealand criminal justice system (Davies, 1998; Zajac & Cannan, 2009; Zajac et al., 2003). These studies showed that many of the questions posed to children—in particular those posed by lawyers—were in breach of best practice as outlined in the psychological and forensic interviewing literature; that is, children were often being questioned in ways that increased the likelihood of eliciting inaccurate evidence and/or of confusing the child. When children attempt to answer questions they cannot understand or which are coercive, it is not only the child‘s credibility and dignity which are potentially undermined, but also the fact-finding function of the criminal justice system.

The purpose of this chapter is to compare forensic interviewers‘, prosecutors‘ and defence lawyers‘ practices when questioning child witnesses with the practices described in Davies & Seymour (1998) to see whether we have come any closer to ensuring children are questioned in age-appropriate ways. The chapter comprises four separate analyses. The first examines the types of questions posed to children; the second examines the grammatical complexity of those questions. In the third section, we examine some of the tactics employed in cross-examination and in the fourth we examine the extent to which judges and prosecutors intervene to stop inappropriate questioning during cross-examination. The implications for practice and future directions are discussed in Chapter 6.

Throughout this chapter, the term examination is used to refer to any interview with, or questioning of, a child witness, whether conducted by a forensic interviewer, prosecutor or defence lawyer; the term adult questioners is used to refer to forensic interviewers, prosecutors and defence lawyers collectively.

This analysis is based on the transcripts of 18 children who gave evidence as complainant or witness for the prosecution in a criminal court during 2008. We do

not claim that this is a representative sample. The trials related to physical and sexual assault. Permission was granted by the Chief High Court Judge and the Chief District Court Judge for the researchers to access anonymised transcripts of those children‘s evidence; administrators at the Offices of the Crown Solicitors in Auckland and Christchurch sourced and anonymised the transcripts for the researchers.58 The study was approved by AUT‘s Ethics Committee in June 2008.

Three of the children gave their evidence in the High Court at Auckland; five in the District Court at Auckland; three in the District Court at Manukau; and five in the District Court at Christchurch. In a further two cases, only the forensic interviews were analysed. In five cases, the children gave their full evidence-in-chief live in court (hence the researchers did not receive a transcript of the forensic interview, if any); in 11 cases, the transcripts of both the forensic interview and courtroom questioning were analysed. This resulted in 45 examinations in total:

At the time of the forensic interview, the children ranged in age from 8 to 15 (mean 12.5); at the time of their court appearance, the age range was 9 to 17 years (mean 14).

The past three decades has seen a sharp increase in the number of studies investigating the reliability of children‘s memories and reports of past events. These studies have led to a growing understanding of the factors which may enhance or diminish children‘s ability to produce accurate reports of events they have participated in or witnessed. This research has established that the way in which children are questioned plays a key role in this regard; that is, the types of questions posed to children can have a dramatic impact on the accuracy of the testimony elicited (Brown, Lamb, Pipe, & Orbach, 2008; Poole & Lamb, 1998). Broadly speaking (but subject to qualification, as outlined in the discussion on page 66), open questions are more likely to elicit accurate information from children than closed questions.

In this section, we replicate Davies & Seymour (1998)‘s examination of the types of questions posed by forensic interviewers, prosecutors and defence lawyers to child complainants who gave evidence in sexual abuse trials held in New Zealand courts in 1994. In the current study, we adopt the same methods and statistical analyses used

by Davies & Seymour so that we can both ascertain the current state of questioning and determine whether practices have changed over the past 14 years.

The transcripts yielded 9,536 questions posed by forensic interviewers,59 prosecutors and defence counsel to child witnesses. The researchers then excluded from this sample all questions which did not relate to forensically relevant information,60 utterances that were interrupted in such a way that the type of question could not be ascertained, and sentence fragments. Following Davies & Seymour, the remaining 8,154 questions were then split depending on the age of the child at the time they were questioned: younger children were defined as those under 13 and older children as those 13 and over.61, 62 The questions were then classified into one of seven mutually exclusive categories, as described in Table 13. The categories mirror those used by Davies & Seymour. A further category was added, namely, ―multiple‖, being those utterances containing both open and closed questions. Two researchers tested the coding system on 211 questions, including questions posed by forensic interviewers, prosecutors and defence counsel. The agreement between the coders for the categories was 97% (K=.943).

The 8,154 questions comprised 4,048 questions posed by forensic interviewers, 1,752 questions posed by prosecutors and 2,354 by defence. To allow for these differences, frequencies are described in this section as the mean proportion of utterances containing the target features.

the child can see or hear properly); explanations as to what will happen next (e.g. when a prosecutor tells the child that the forensic interview is about to be played to the court); and instructions on how to handle court exhibits where those instructions are not forensically relevant. Responses by adult questioners to a child‘s question were also excluded unless the response was to rephrase a question.

with Davies & Seymour (1998). However as we saw in Chapter 2, this age split was implicated in other aspects of children‘s interactions with the criminal justice system, most notably in relation to whether the child was forensically interviewed and which modes of testifying in court children used. Furthermore, we refer later on in this chapter to Henderson (2003) which reports that lawyers in the 1990s believed that there was no need to adjust their language with children of normal intelligence who were over 10-12 years old.

was 11 at the time of the forensic interview was classified as ―younger‖ for the purposes of the analysis of that examination; if the same child were 13 at the time of trial, s/he would be classified as ―older‖ for the purposes of analysis of the courtroom examination.

Open A question which asks the child to supply missing information. These include wh-questions, imperatives and unrestricted alternative questions.63

answer to either yes or no or, in the case of restricted alternative questions, one of the options posed.

A closed question which indicates the expected answer (that is, declaratives with or without tags, but excluding Recaps—see below).64

questioner faithfully repeats back to the child information which the child had provided earlier in the same examination. The recap may or may not use the same words; it may be adjacent to or at a distance from the child‘s original speech.

Interviewer: Oh, OK, so when you were massaging his back where was he? Was he sitting or standing or--?

Echo Verbatim echo of the child‘s words, immediately after the child utters them.

Interviewer: Yeah so we are gonna talk about two different times that something happened.

content; the questioner provides neutral feedback to the child but does not take the conversational floor.

Child: The first time one time when dad went to Northanger he was touching me around there.

In the following, we describe the mean proportional use of these features across the three types of adult questioner.

Figure 11 shows the frequency with which the question types were posed to children under 13 years of age.65 To determine whether the differences between the examiners‘ use of these question types were significant or not, Tukey‘s tests were used. These tests revealed the following:

Figure 12 shows the profile of the questions posed to children aged 13 years and over. Once again, Tukey‘s tests were used to determine whether the differences in the types of questions posed by the adult questioners were significant. These tests revealed the same patterns of significance found for questions posed to younger children, with the following exceptions:

Given the small number of children under 13 who were questioned by lawyers, we also analysed the profile of questions across all witness ages. The results are shown in Figure 13. Tukey‘s tests were used once again to determine the significance of the differences between the three types of adult questioner. These tests revealed the same patterns of significance found for questions posed to younger children, with the following exceptions:

We also sought to determine whether the adult questioners changed the frequency with which they used each question type, depending on the age of the child being questioned (for example, do defence lawyers use the same proportion of closed leading questions with children under 13 as they do with children 13 and over?). An ANOVA test66 was used to determine differences within each group. Tests for linearity within an F-test showed no significant differences between the frequencies with which the types of questions were used by forensic interviewers depending on the child‘s age, although there was a marginal difference in relation to the use of facilitators (p=.051) (that is, forensic interviewers used facilitators marginally more frequently with younger children than with older children). However the Eta coefficients67 are suggestive of an additional difference in relation to closed questions (.509) posed by forensic interviewers (that is, they used closed questions more frequently with older children than with younger children). The differences for prosecutors and defence lawyers were not significant (for either the F-test or Eta coefficients), showing that these questioners did not use different question types for children under 13 compared to the questions used for children 13 and over.

In the final analysis, we compared the profile of questions in the current data with the profiles reported by Davies & Seymour based on data collected from trials held in 1994. We used unpaired t-tests to compare the mean percentages of each question

type by questioner across time. Those tests showed the following differences in relation to the question posed to younger children:

A considerable body of research has shown that children‘s (and adults‘) responses to open-ended questions are more likely to be accurate than their responses to other types of question:

―Regardless of the experimental procedures, the ages studied, the cognitive capacity of the subjects, or the length of the delay between events and the interview, open-ended questions are more likely to elicit accurate accounts.‖ (Poole & Lamb, 1998, p. 53)

For this reason, interviewers are urged to use open-ended questions as much as possible when questioning children. However children‘s responses to such questions do not always include all of the details required in a forensic context, such as time, place, duration, potential witnesses and so on. When open-ended questions fail to draw out such details, the literature recommends that interviewers use wh-questions (‗specific questions‘)69 before resorting to the riskier question types, namely closed questions, to elicit these details (Lamb, Orbach, Hershkowitz, Esplin, & Horowitz, 2007; Poole & Lamb, 1998). Leading questions should be avoided altogether (M. E. Lamb, Y. Orbach, I. Hershkowitz, P. W. Esplin et al., 2007; Walker, 1999).

Poole & Lamb (1998) have arranged these question types into a hierarchy, ranging from least to most suggestive, in the same order as appears above:

Free narrative/open-ended questions < Specific nonleading < Closed < Leading Least suggestive Most suggestive70

The UK Home Office manual, Achieving Best Evidence in Criminal Proceedings, describes the same question types as varying ―...in the amount of information they are likely to provide and their susceptibility to producing inaccurate responses from children‖ (Home Office, 2006, p. 43). However, Poole & Lamb note that ―there is no consensus in the interviewing literature about how to define these terms on a question-by-question basis‖ (Poole & Lamb, 1998, p. 137).

The main difficulty for our purposes relates to the distinction between open-ended and specific nonleading questions. Poole & Lamb (1998) describe these as follows:

―...open-ended question refers to questions that require a multiple-word response (e.g., initial free-recall questions such as, ‗Tell me about that,‘ or more focused questions such as, ‗Tell me about all the things you saw in the house‘). In contrast, specific questions ask about a particular detail or concept and often can be answered in a single word (e.g., ‗What color was her hair?‘) ... Open-ended questions can be quite focused, such as ‗Tell me about his eyes,‘ but in general open-ended questions deal with broader topics than do specific questions, and open-ended questions allow children more flexibility to choose which aspects of an event they will describe (e.g., ‗Tell me about his eyes‘ is more open than ‗What color were his eyes?‘).‖ (Poole & Lamb, 1998, p. 52)

The difference in ―openness‖ between Tell me about that and What colour was her hair is clear enough: the former gives the addressee wide scope while the latter asks for a very specific piece of information. However at the less extreme ends, the distinction between the two becomes blurred. For example, the sentence What clothes was he wearing? is classed as open-ended in the Home Office manual but Poole & Lamb (1998, p146) describe an almost identical question, What was he wearing when that happened?, as specific nonleading.71

It is perhaps more accurate to see ―open-ended‖ and ―specific‖ not as discrete categories, but as points along a continuum. They differ in the degree to which the questioner determines what information is to be supplied—or, put the other way, the degree of narrative freedom afforded the addressee. Open-ended questions give the addressee more leeway to decide which aspects of an event or entity to report while specific questions give less. With closed and leading questions, on the other hand,

the same category as open-ended questions (see, for example, Krahenbuhl & Blades, 2006; Saywitz & Camparo, 1998); others collapse them with yes/no questions into the category of ―specific questions‖ (see, for example, Powell & Guadagno, 2008); others still give them as examples of closed questions (see, for example, Kebbell, Deprez, & Wagstaff, 2003).

the questioner takes the narrative role and the witness is relegated to the task of agreeing or disagreeing (although they have the option of giving a more elaborate response). Therefore closed questions—and leading questions in particular—typically give the questioner more control over the responses than their open counterparts.

In summary, the literature suggests that open-ended and specific questions tend to be forensically safer than closed or leading questions (Lamb & Fauchier, 2001; Orbach & Lamb, 2001; Peterson & Biggs, 1997).72 Given the difficulties inherent in distinguishing between open-ended and specific questions, we follow Davies & Seymour (1998) in collapsing them into a single category of ―open questions‖ (excluding open loaded questions) which we contrast with closed and leading questions in terms of the accuracy of the response and the degree of control exerted by the questioner over the response.73

Turning now to the data analysed here, we saw that forensic interviewers and prosecutors used significantly more open questions than defence lawyers with both younger and older children. Open loaded questions were rarely used by any of the adult questioners. Forensic interviewers tended to use significantly fewer closed questions than prosecutors, who used significantly more than defence lawyers with older children and across all ages. While leading questions are generally not permitted in examination-in-chief, there is much greater freedom to use them in cross-examination.74 It is therefore not surprising to find that the defence lawyers used significantly more leading questions than anyone else.

These results show that forensic interviewers and prosecutors used higher proportions than defence lawyers of the types of questions that are more likely to elicit accurate information and in the witness‘ own words, namely, open questions. Defence lawyers used higher proportions of the types of question which exert the most control over the witness‘ testimony and are more likely to elicit inaccurate information, namely, closed leading questions.

was more likely to be confirmed by the perpetrator than information elicited via specific or closed questions.

However that same research found no significant effect of question type on the proportion of details contradicted by the perpetrator (M. E. Lamb, Y. Orbach, I. Hershkowitz, D. Horowitz, & C. B. Abbott, 2007).

questions form a natural class, being questions which require the respondent to supply missing information (e.g. in the sentence Who broke the pot the missing information is the identity of the person who broke the pot); closed and leading questions too form a natural class, being those whose unmarked response is yes or no. The division also seems to reflect the distinction made in the psychological literature between questions which involve recall (e.g. open) and those involving recognition (e.g. closed). While a thorough discussion about this distinction is beyond the scope of this chapter, the literature suggests that responses to recall questions are more accurate than responses to recognition ones (Pipe, Lamb, Orbach, & Esplin, 2004).

examination, ―but those which suggest the existence of unproved facts might well be disallowed, even in cross- examination. The Judge has a wide discretion in these matters. The most accurate way of stating the position is that leading questions will usually be disallowed in chief or in re-examination, but they will generally be permitted in cross-examination‖ (Mathieson, 2005, p. 245).

Having said that, there are times when closed, polar yes/no questions are perhaps forensically safer than open questions. For example, if a child states that an offence occurred at home, the questioner must establish whether there were any potential witnesses. The open question Who else was at home? would be suggestive, as it presupposes that there was in fact someone else at home; the closed question Was anyone else home?, on the other hand, is neutral in this respect.75 Many of the closed yes/no questions posed by forensic interviewers were used in situations such as this.

Forensic interviewers used a significantly higher proportion of facilitators than their lawyer counterparts, particularly with the younger children. These utterances have been described as a useful means of ―[expressing] attention and interest‖ and are considered a best practice device by Köhnken (1993) (cited Poole & Lamb, 1998, p. 93). In particular, they are recommended as a way of encouraging the child to continue talking (Home Office, 2006). Recaps also featured more commonly in forensic interviews than in the courtroom. These utterances involve repeating back to the child part of his/her testimony while faithfully preserving the meaning of the child‘s original speech. These appeared to serve two main functions: to check or reiterate the testimony and/or to cue further questions. There are examples from forensic interviewers, prosecutors and defence of a recap cueing an open-ended question, a practice recommended in the interviewing literature:

There were some instances where the information reported back was subtly—or not so subtly—different from the child‘s original testimony (such instances were not classified here as recaps). In the following example, the child responds no to a question. The lawyer then rephrases the response with you don’t think so. The rephrased version projects a lower degree of certainty than the child‘s original utterance, but the child acquiesces:

has already established that a home exists. Compared to the question What happened next, the question What

did he do next? carries a higher presuppositional load in that it assumes not only that something happened, but that the next most forensically relevant piece of information is whatever ―he‖ did (rather than what someone else did); as such, the question wrests some of the narrative control away from the child.

1	Defence	Okay well I think what actually happened is that Jordan went to sleep on the couch with you and Chris while you were watching Grease.
2	Child	No.
3	Defence	You don‘t think so?
4	Child	No.

Echoes too were more frequent in the forensic interviews than in the courtroom, although the differences in use were only significant when the questions were examined across all ages. Echoes appear to be used to perform a range of functions, sometimes operating like a facilitator and encouraging the child to continue talking, at other times used to ask the child to repeat something just said or to seek elaboration.

In short, forensic interviewers used more echoes, facilitators and recaps than their legal counterparts. What these devices have in common is that they can be used to encourage the child to testify in their own words, but cannot introduce new, and potentially leading or misleading, information into their evidence. However that is not to say that they are always neutral. On the contrary, with the appropriate intonation, all could be used to indicate irony or disbelief or ridicule. In the absence of intonation (and visual) cues, we cannot determine whether they were in fact used in this way in the transcripts examined.

A comparison of the use of open and closed questions over time showed neither a consistent improvement nor consistent deterioration in the practice of any of the adult questioner types between 1994 and 2008. There is evidence that some defence lawyers‘ practices have improved with older children, but not with younger ones; some of prosecutors‘ practices with younger children have improved marginally, but deteriorated with older children; forensic interviewers‘ practices with younger children have both improved and deteriorated. All three questioner types used facilitators with younger children significantly more often than their counterparts in 1994. In forensic interviews the use of recaps has changed little over time, but in the courtroom, the use of this device has tended to decrease. Forensic interviewers used echoes significantly less often in the 2008 transcripts compared to the 1994 ones with younger children; defence and prosecution used them significantly less often with older children.

It must be stressed that neither the examinations analysed in the current study nor those in Davies & Seymour (1998) are scientifically representative samples; indeed the results of this entire chapter must be understood in that light. Hence any conclusions as to improvements or otherwise in interviewing practices over time relate solely to the practices evidenced in these sets of transcripts. There are other reasons for treating the comparative figures with caution. Firstly, we were not able to ascertain from Davies & Seymour the exact number of examinations conducted

with children over 12; this may have impacted on the statistical results.76 Secondly, the current study involved only six courtroom examinations of children under 13, which may have masked significant differences. Thirdly, there may have been differences between the two studies in the way questions were classified, despite best attempts to maintain consistency. And finally in the current study there was a new category, multiple; although the number of questions classified as multiple was so small that it is unlikely that its inclusion had any real effect on the figures.

A number of language features has been identified as causing comprehension difficulties for children. These include double negatives, complex vocabulary, the presence of multiple subordinate (―embedded‖) clauses (particularly those which appear before the main verb (Anderson & Davison, 1988, cited Bailin & Grafstein, 2001)) and passives (Brennan & Brennan, 1988; Perry et al., 1995; Walker, 1999). Similarly, the literature points to certain concepts posing difficulties for children, such as estimating time, duration, frequency, other people‘s intentions, and distinguishing between before and after (Walker, 1999).

The purpose of this section is to compare the frequency with which these features occur in the language used by forensic interviewers, prosecutors and defence counsel when questioning children; we will also investigate whether there is any evidence that the adult questioners adjust their language for younger witnesses and the frequency with which adult questioners‘ utterances contain more than one complex feature.

The 8,154 questions extracted for the previous analysis were examined to establish how many contained the target features, namely:

expressions such as put it to you; suggest to you; give evidence; tell the Court; my learned friend; accused; is

it your evidence that... Also included in this category was unnecessarily formal language (such as make available to the Police (instead of give to the Police); How did that arise? How did it come about?) and figurative language such as the idioms in the spotlight; to hold water; hammer the point; come to the surface.

oOther people‘s intentions (e.g. why someone else did or did not do something)

Two researchers tested the coding system on 211 questions, including questions posed by forensic interviewers, prosecutors and defence counsel. The agreement between the coders for the categories ranged from 90% (K=.801) to 100%. To allow for differences in examination length, frequency is described as the mean percentage of utterances containing the target feature. As per the questioning analysis, the questions analysed were split depending on the age of the child at the time they were questioned: younger children were defined as those under 13 and older children as those 13 and over.

Differences are typically described as significant when they reach the 0.05 level. Given the small number of examinations involved here (14 involving children under 13; 31 involving children 13 and over), one would be justified in considering differences significant at less than .05. In the following, however, significance is reported at the 0.05 level unless stated otherwise.

Figure 14 shows the frequency with which these features are used in questions posed to children under 13 years of age. To determine whether the differences in mean proportional use were significant or not, Tukey‘s tests were used and the following significant differences were found:

≥2 embedded Passives Difficult Double Difficult clauses Vocabulary Negatives Concepts

Figure 15 shows the distribution of the features in questions posed to children 13 years and over. To determine whether the differences in mean proportional use were significant or not, Tukey‘s tests were used. These tests revealed the following:

≥2 embedded Passives Difficult Double Difficult clauses Vocabulary Negatives Concepts

To establish whether the small number of cases of children under 13 was masking significant patterns, the same data was analysed further but this time the questions were not split by the age of the witness. Tukey‘s tests revealed the following:

≥2 embedded Passives Difficult Double Difficult clauses Vocabulary Negatives Concepts

We also tested whether there was any evidence that the adult questioners adjust their language with younger children. An ANOVA test was used to probe differences in the frequency with which the adult questioners used the various types of complexity depending on the age of the child. The tests showed no significant differences between the groups.

Left-branching clauses and subordinate clauses containing verbs in the passive voice

A further analysis was conducted to determine whether there were any differences between adult questioners in the use of subordinate clauses which appear before the main clause (―left-branching clauses‖) and the use of verbs in the passive voice within subordinate clauses. An ANOVA test revealed no significant difference in the use of these features between adult questioners.79

The final analysis examined how often adult questioners produced utterances which contained more than one type of complexity (―multiply-complex sentences‖). Specifically, we examined all instances where the question contained two or more of the following: two or more embedded clauses, passives, difficult vocabulary or difficult concepts.80 To ensure a sizeable sample, the questions in this case were not split by age group.

A one-way ANOVA test revealed significant differences in the frequency of multiply- complex questions between adult questioners (p<.000). Tukey‘s tests were once again used to probe the differences between questioners and revealed the following:

When Henderson interviewed lawyers in the UK and New Zealand in the 1990s about questioning children, she found that, although most recognised the need to modify their speech for young witnesses, they had a modest understanding of how to do so. The consensus among these lawyers was that it was not necessary to adjust their language with children of normal intelligence over 10 or 12 years old (Henderson, 2003).

This view is consistent with the received wisdom of the 1960s and into the 1970s that children acquire most of their native tongue by the age of five or at least by late childhood. Although it was recognised that children‘s language continued to develop beyond that period, this development was thought to involve mainly refinements in grammar and the continued acquisition of vocabulary (Nippold, 2007). Recent scholarship in the field of first-language acquisition has overturned this view, demonstrating that significant growth occurs during the school years and into adulthood in a number of areas of language acquisition, including grammar,

semantics and pragmatics (Miller, 2002; Nippold, 2000, 2007; Perera, 1984; Walker, 1999). Even when a child uses words or grammatical structures in an apparently adult-like way, this does not necessarily mean that the child‘s understanding of them matches an adult‘s (Dezwirek-Sas, 1992; Nippold, 2007; Schuman, Bala, & Lee, 1999; Walker, 1999). As Walker (1999, p. 12) puts it:

―...adults and children do not speak the same language. Language and cognition do not mature simultaneously, and for that reason, children can appear to be using words and sentences in an adult way, but in reality be operating on an entirely different level. In particular, the fact that children often use words before they really understand them can deceive us as to what they are actually thinking.‖

For most adults, the need to adjust their language for a 6-year-old is obvious, even if they are not entirely sure how to do so. But it may be less obvious that we need to take care when communicating with older children and adolescents too. Even 15- and 17-year-olds‘ understandings of formal or high register words like assume or concede may be incomplete;81 children up to 15 may not fully comprehend common structures, like clauses beginning with although and before;82 clauses beginning with unless can confound adults (Perera, 1984). Bread-and-butter terms for lawyers or forensic interviewers may be poorly understood (or not at all) by some children, as seems to be the case in the following exchange. Here the 14-year-old appears to change her testimony in line 4; but it may have been that she didn‘t associate cutting oneself with physically abusing oneself as the defence lawyer did:83

1	Defence	Did you ever see her physically abuse herself in any way?
2	Child	No.
3	Defence	You never saw her scratch herself, cut her arm?
4	Child	At one point I did. It was when I was around 12. I walked in on her cutting herself.

There is ample potential for speakers to misjudge a young person‘s level of language development and for miscommunication to arise. When that miscommunication occurs in the criminal justice system, the consequences can be grave. Yet the results of the current study suggest that children are frequently being asked questions, particularly in court, that are likely to cause comprehension difficulties.

accurate in 59% of cases; 17-year-olds managed 71% accuracy and college students 92%. The accuracy of 11-

year-olds stood at 45% (Astington & Olson, 1990). As a further example, Nippold et al. 1992 (cited Nippold, 2007, p. 275) examined 120 young people‘s production and comprehension of the adverbial conjuncts consequently, contrastively, conversely, furthermore, however, moreover, nevertheless, rather, similarly, and therefore. In the comprehension test, the mean accuracy scores were 50% for 12-year-olds; 64% for 15-year- olds; 85% for 19-year-olds; 94% for 23-year-olds.

clauses and concluded that, ―nine is the earliest age at which a rudimentary understanding of although can be expected and that comprehension is not fully established by fifteen‖ (Perera, 1984, p. 144). Nippold notes that,

―...preschoolers often use the words because and before but they may not completely understand those words as intrasentential connectives until adolescence‖ (2007, p. 28).

Turning first to complex vocabulary, this was used in 0.3% of forensic interviewers‘ utterances, 12% of prosecutors‘ and 18% of defence lawyers‘. Forensic interviewers also checked children‘s definitions of terms more often than prosecutors and defence did. While prosecutors and defence lawyers were not always successful in avoiding legalistic or difficult language, on a number of occasions they appeared to catch themselves doing so, then repeated the idea using a more familiar term (the more familiar term is underlined):

There were also examples of the opposite, with defence lawyers using a familiar term then rephrasing the idea using a legalistic term:

The children interviewed here rarely asked for clarification when faced with difficult vocabulary.84 When they did so, the adults‘ responses did not always show an understanding of what might have been troubling the child. In example 7 below, the initial sentence contains both difficult phrases (put to you, allegation, prior to) and complex syntax. In the rephrased version, the legalistic phrase put to you remains and the sentence is still grammatically complex.85 In example 8, the complainant states that she does not know the meaning of the word allegation, yet the lawyer uses the word again later on in the examination. In the last example, the witness, who is not a native speaker of English, asks the lawyer to repeat the question; the lawyer rephrases the question using an even more complex sentence: 86

when children asked for clarification, a defence lawyer‘s next question was ―more likely than chance to be complex,‖ whereas a prosecutor‘s was not (p. S47).

1	Defence	So what I‘m putting to you is that you‘ve discussed this allegation with your mother prior to making it. What do you say?
2	Child	I don‘t get what you‘re trying to say.
3	Defence	Okay. What I‘m putting to you is that your mother has asked you to say bad things about your father.
4	Child	No she didn‘t.

1	Defence	Isn't the allegation of oral sex and having sex at Wickham that night just really a fantasy of yours?
2	Child	Isn't the what—allegation.
3	Defence	Pardon?
4	Child	What's an allegation?
5	Defence	Okay. Isn't the suggestion or that you had oral sex and full sex with Andy just really a fantasy of yours?
6	Child	No.
		94 utterances intervene
7	Defence	Is it true that your mum was pressuring you to come forward to these allegations about Andy?
8	Child	Ah, no.

1	Defence	But you were quite happy to go and be close to your mother in all those events including in the photo, weren‘t you?
2	Child	Can you just say the question again?
3	Defence	That you were quite happy to be content in those photos, and be quite happy, you didn‘t show any sign of being hostile to your mother or being upset at being close to her, were you?
4	Child	Just can‘t think.

While complex words and phrases are avoidable, it would be difficult to avoid embedded clauses altogether. Nonetheless forensic interviewers and prosecutors tended to avoid them more often than defence lawyers. We recognise that embedded clauses are not uniformly difficult to comprehend; we recognise further that the presence of multiple embedded clauses in a single sentence is not necessarily a clear indicator of difficulty.87 However there were some utterances in our data containing so many embedded clauses that it is probable they were difficult to comprehend. There were also many utterances containing multiple embedded clauses before the main clause (―left-branching structures‖); these structures are known to be more difficult to process than those in which the embedded clauses appear after the main clause (―right-branching structures‖) (Bailin & Grafstein, 2001; Scott & Stokes, 1995; Walker, 1999). In the examples below, the embedded clauses are marked

with square brackets and the main clause underlined (note that the child‘s request for clarification in Example 10, line 2 is followed by an even more complex question):

1	Defence	And [if I put it to you [that nobody drank coffee]], what would you say?
2	Child	I don‘t get you.
3	Defence	Well, [what I‘m trying [to say]] is [that it‘s odd [that there‘s coffee in the house [if it was against their religion at the time [to drink it]]]]. Do you understand [what I‘m asking you]?
4	Child	No.

1	Defence	[If he told the police [that that was [what he thought [you wanted [to do,]]]]] are
2	Child	you saying [that you don‘t think [he could have thought that?]] I don‘t know why he will have considered it anyway; doesn‘t have a wife.

Passives, in contrast, are avoidable: for every passive verb, there is an active counterpart available. Passives are infrequent in everyday speech and were similarly rare in the examinations analysed here, appearing in 0.3% of the forensic interviewers‘ utterances, significantly fewer than in prosecutors‘ (5%), and in 3% of defence lawyers‘. Passives are generally more difficult to comprehend than verbs in the active voice (Perera, 1984), however not all passives are equally problematic. Get-passives, such as My sister got cut on her finger by my scissors,88 are acquired relatively early by children (Walker, 1998); one prosecutor in particular appeared to be aware of this, as nine of eleven passives s/he used were of this type. The passive of some verbs, such as to be called, is also likely to be acquired early. For this reason, these two types of passive were excluded from the counts here. However, there are some indications that passives are particularly difficult to comprehend when they appear in embedded clauses (Charrow & Charrow, 1979). Embedded passives were rare in the utterances of all three questioners; there were no significant differences in the frequency with which they were used by adult questioners.

Double negatives were similarly infrequent in the transcripts examined. These comprise utterances containing two (or sometimes more) negative elements, including negative words such as no and not, as well as concealed negatives such as unless (which means if not) and words with negative affixes (unhappy or clueless):

Forensic interviewers and prosecutors used these in less than 1% of their utterances, while defence lawyers used them significantly more often—in just over 1% of utterances.89 Double negatives are known to cause comprehension problems for children as well as adults (Perry et al., 1995); indeed, even sentences containing single negatives are more difficult to process than their affirmative counterparts (Perera, 1984).

There were no significant differences between adult questioners in the mean proportion of utterances involving difficult concepts. It is, of course, part of the questioner‘s role to establish (or test the evidence concerning) the timing, frequency, sequence and duration of events. Adults as well as children find it difficult to specify dates for autobiographical events (Orbach & Lamb, 2007). Furthermore, children‘s ability to judge time accurately more generally increases with age but their ability to do so can be affected by, amongst other things, the time delay between the event and retelling it (Orbach & Lamb, 2007); for child complainants in the full data sample (see Chapter 2) the average time that elapsed between the first offence and appearing at trial was 31 months (two and a half years).90

There were clear differences in relation to the frequency with which adults posed questions containing multiple forms of complexity, with forensic interviewers avoiding this significantly more often than lawyers. By way of example, the following sentences contain complex vocabulary, complex concepts, passives and multiple embedded clauses:

1	Prosecutor	Right okay. Now it was put to you that after the accused broke up with you on
		Christmas Day and she started seeing Kerry that you wanted more than just the
		friendship that you had with her and, and your answer was half. What did you mean
2	Child	by that. Half. I‘m sorry, can you please repeat that?

1	Defence	But, when all this came to the surface and Dominique had told you what had
		happened, and you became aware of it, some of the touching that had happened
		before, did you start suddenly ask—were you asked to think about it in a different
2	Child	light? Kind of.

Children were asked 2,233 questions in total that contained at least one form of complexity; they sought clarification for 76 (3.4%) of these questions. By contrast, they were asked 10 questions involving complex vocabulary, plus complex concepts, plus passives, plus multiple embedded clauses; they sought clarification in three (30%).

As noted in the Results section, we found no evidence that any of the adult questioners adjusted their language depending on the age of the child.91 In relation to forensic interviewers, this finding is perhaps not surprising: they tended to use less complex language as a matter of course. In relation to lawyers, on the other hand, the tendency not to adjust their language supports Henderson‘s (2002) finding that lawyers‘ understanding of how to accommodate children‘s language competencies is not adequate or that that understanding is not being translated into practice. However this finding must be treated with some caution: the failure to find such a difference may reflect the low numbers of interviews involving children under 13 or it may also be that we needed to put the split at some age other than 13.

Past examinations of courtroom language have shown that the questions posed to children and adolescents are often complex and are likely to be difficult for them to comprehend (see, for example, Brennan & Brennan, 1988; Davies & Seymour, 1998; Walker, 1993; Zajac et al., 2003). We know from experimental research that when children are questioned about events using complex language, the accuracy of their responses can be diminished (Carter, Bottoms, & Levine, 1996; Perry et al., 1995). The current study shows that the use of age- or individual-inappropriate language continues to this day. Not only is it disrespectful to the young person to use language that they may struggle to comprehend, but the chasm between the language used in the criminal justice system and the language children can be expected to understand has the very real potential to thwart the fact-finding functions of that system.

―It may be that in more than one sense [cross-examination] takes the place in our system which torture occupied in the mediaeval system of the civilians.‖ (Wigmore, cited Spencer & Flin, 1993, p. 270)

The treatment of child witnesses in the adversarial system, and in cross-examination in particular, has long been criticised by commentators (Brennan & Brennan, 1988; Davies & Seymour, 1998; Eastwood & Patton, 2002; Spencer & Flin, 1993; Walker, 1993). In this section, we examine some of the tactics adopted in cross-examination

which have been criticised for limiting children‘s ability to tell their stories and for potentially undermining their credibility.

Advocacy in adversarial courts is described by Henderson (2002) as competitive story-telling: a competition between lawyers to persuade the judge or jury to accept their version of events. In this contest, the witness becomes ―the principal medium by which [the advocate] tells his story‖ (Stone, 1985, cited Henderson, 2002, p. 282). As one lawyer put it:

―In trial I have three speeches: my opening, my cross-examination, and my closing.‖ (cited Henderson, 2002, p. 282).

During examination-in-chief, prosecutors aim to portray a version of events that supports their case. The objective of cross-examination is ostensibly to test that evidence, although manuals on cross-examination, such as Salhany (1999), tend to describe the objectives differently:

―The first [objective] is to weaken the case for the other side by discrediting the witness‘ testimony in-chief. The second is to elicit evidence from the witness which will establish facts favourable to the cross-examiner‘s side.‖ (Salhany, 1999, p. 25)92

The author goes on to state that these objectives are met by controlling the witness‘ responses through careful framing of questions. He suggests that closed or leading questions, in which the lawyer asks the addressee to either agree or disagree with the proposition in the question, are ideal vehicles for controlling the evidence and building the storyline that best supports the lawyer‘s case, particularly with children:

―Children usually want to please adults. If the cross-examiner takes the time to frame his questions carefully, he will discover that the child witness will often be very helpful to him. ... Because the child will naturally find the courtroom setting somewhat frightening, the child‘s answers should be short and simple. Most importantly, they should be framed 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, 1999, p. 72)

Another way of controlling the witness‘ responses is simply to confuse them (Eichelbaum, 1989, cited Henderson, 2002). Using complex language which exceeds the child‘s comprehension is one way of doing so, as described by a New Zealand lawyer in Henderson‘s study:

―The difference [between adult and child witnesses] is in being able to consistently communicate in a public forum ... You‘re looking ... to make sure they make mistakes. [...] Some counsel ... give double negatives to kids. And the kids get it wrong ... But that is a valid technique that is used by very senior counsel and very successfully.‖ (Lawyer cited by Henderson, 2002, p. 286)

As we have already seen, defence lawyers in the transcripts examined here made ample use of closed and complex questions. One particular device was to present the

defence‘s version of events via a series of closed questions, asking if the child remembered those particular events. This strategy exploits the inherent ambiguity of a negative response to do you remember [X] or you don’t remember [X] questions: a no response could be intended to mean no, I don’t remember [X] or no, [X] didn’t happen.

In one cross-examination, over 20% of the lawyer‘s utterances posed to the 9-year- old witness were of the do you remember/you don’t remember variety. For example, in line 5 of the following exchange, the lawyer asks the child whether she remembers asking Chris to carry her to bed; she denies asking Chris to do this. Then in line 12, the lawyer states that the child doesn‘t remember doing this, contrary to the child‘s evidence. This time, the child simply answers ‗no‘. This response could be interpreted as the child acquiescing to the lawyer‘s insinuation that her memory is at fault, rather than that the event did not occur. The defence then moves to a new topic, and the child does not have a chance to clarify her meaning, if indeed she knew her response was ambiguous. The rhetorical repetition of you don’t remember in lines 8, 10, 12 arguably constructs an impression that the witness‘ memory is not reliable, thereby undermining the child‘s credibility.

1	Defence	So the first night you spoke about, Jordan went to sleep on the couch.
2	Child	Not that I remember.
3	Defence	You don‘t remember Chris carrying Jordan down to bed?
4	Child	No.
5	Defence	OK. Do you remember asking Chris if you could be carried to bed as well?
6	Child	No, I didn‘t ask him that.
7		[16 utterances intervene]
8	Defence	So you don‘t remember Jordan falling asleep on the couch.
9	Child	No, she was asleep with Charlie.
10	Defence	You don‘t remember Chris carrying Jordan to bed.
11	Child	No.
12	Defence	You don‘t remember you asking Chris if he would carry you to bed.
13	Child	No.
14	Defence	Okay. Can we talk about the second time?

The defence lawyer used the same technique in 30% of his questions to another child witness in the same case.

Davies & Seymour (1998) investigated the use of two other techniques common to cross-examination: the skiparound tactic and focusing on peripheral details. As for the former, this involves moving rapidly from one topic to an unrelated one without the use of conventional means of heralding a topic shift. Such a technique may confuse and disorientate the child (Brennan & Brennan, 1988) but is recommended in leading advocacy texts; for example, Levy (1991, cited Ellison, 2001, p. 361)

suggests this tactic is helpful if the lawyer suspects a child has memorised his or her testimony. Davies & Seymour (1998) reported that this tactic was used in 65% of the 26 cross-examinations analysed; in the current study, this tactic was used at least once in 12 out of 16 cross-examinations (75%). By way of example, in the following exchange the defence lawyer questions the child about a family disagreement, before switching abruptly to an aspect of the alleged assault in line 5:

1	Defence	Now you were aware that there were some arguments between your mother and your uncle. Is that correct?
2	Child	Yes.
3	Defence	Okay. There was a bit of argument over family things and over a funeral. Would that be correct?
4	Child	Yes.
5	Defence	You say he had ripped trousers—sorry ripped boxers—I put it to you that he didn‘t have ripped boxers at all.
6	Child	He did.

The second technique is to focus on peripheral details. It is well established in the memory literature that information that is salient or central to events is better remembered than information that is peripheral (Fivush et al., 2000; Reed, 1996; Toth & Valentino, 2008). Inconsistencies around peripheral details are to be expected. Furthermore, when children report on multiple events, as is often the case in sexual and physical assault trials, ―increasing error for less distinctive events is most often a matter of confusing details among similar experiences rather than reporting details that were never experienced‖ (Fivush et al., 2000). A recent experiment involving 4- to 8-year-old children showed that children who reported on similar repeated events were judged by adults to be less competent at remembering—and less consistent—than children who reported on single events, even though the actual accuracy of the children‘s reports did not differ (Connolly, Price, Lavoie, & Gordon, 2008).93 The relevance of this is that inconsistency is often seen as a marker of inaccuracy (Connolly et al., 2008). In contrast to the memory literature, manuals on cross-examination tend to assume that ―real memories are complete in every particular and peripheral detail and inconsistencies indicate untruthfulness‖ (Henderson, 2000, p. 89). Some advocate questioning the witness about ―fine peripheral details as a test of reliability‖ (Henderson, 2000, p. 91) (for an example, see Example 21 on page 90).

Davies & Seymour reported that lawyers ―asked in detail about peripheral events in 100% of the cross-examinations of children under 13 years and in 85% of cross- examinations of young people aged 13 years and over‖ (p. 58). The 2008 transcripts revealed this tactic in one of the three cross-examinations involving younger children

and 61% of those involving older children. Neither set of transcripts is representative; however, if these findings were replicated in a larger sample, it would represent a marked change for the better since the 1990s in the culture of cross-examination in New Zealand. Certainly, s92 of the Evidence Act 2006, which came into force after the Davies & Seymour study, states that cross-examination must be ―...on significant matters that are relevant and in issue and that contradict the evidence of the witness.‖ 94 The New Zealand Law Commission notes that, with the exception of the reference to matters that contradict the witness‘ evidence, this section ―largely codifies existing law and practice‖ (1999b, p. 217); hence the change in practice between 1994 and 2008 presumably cannot be attributed to the new Act.

Another technique in cross-examination is to accuse the child of lying. Children who have given evidence in the courts often cite this as one of the worst aspects of their entire court experience (Cashmore & Trimboli, 2005; Eastwood & Patton, 2002; Hamlyn et al., 2004). Allegations of lying, from the outright (That’s a lie) to the more subtle (Isn’t the real truth that you...), featured in five of the 16 cross-examinations analysed, including suggestions that the alleged offences were a fantasy or imagined (Is it the case that you have re-created this event in your mind). In one trial, the child was accused of lying five times in the space of 11 defence utterances. As noted by Spencer & Glaser, ―Calling a child a liar rapidly reduces most children to tears. The stress induced by this then makes it hard for the child to remember accurately and think clearly‖ (1990, p. 378).

There are other ways of casting doubt on the child‘s testimony that avoid outright accusations of lying. These include suggesting the child‘s memory is faulty (It’s quite hard to remember the details now, isn’t it?) or that they have become confused (Do you think you may have got a bit mixed up?) or questioning their certainty (You sure about that?) or insinuation (So now you’re saying...). In the following example, the use of the word selected in line 1 is an overt but subtle accusation of dishonesty:

1	Defence	The first question I want to ask you is about the dates which you have selected for
		these allegations. Was the first time that you said it occurred on [date] or [date] was
		that some two years later when you spoke to the CYFS or the video interviewer?
2	Child	[emphasis added] Yeah.
3	Judge	I think that question is a bit complicated somehow, Mr X.

Children were accused of lying, having faulty memories, being confused or uncertain in 11 of the 16 cross-examinations analysed here. The value of such techniques has been questioned by non-legal commentators:

―One particular aspect of cross-examination that children report finding very difficult is being accused of lying and making things up. While challenging the reliability and veracity of the evidence is part of the defence lawyer‘s job, especially in ‗word against word‘ allegations, directly and repeatedly accusing the child of lying or misrepresenting the child‘s response or insinuating that the child was at fault is very stressful for children and arguably goes beyond testing their evidence.‖ (Cashmore & Trimboli, 2005, p. 49)

If one of the objectives of cross-examination is to discredit prosecution witnesses, then the defence lawyer has many tools at his or her disposal: phrasing questions as closed and complex, skipping from one topic to another and back again, focusing on peripheral details. These techniques are likely to confuse child witnesses, putting them off-balance. It is therefore not surprising that inconsistencies arise—especially as children are often reporting on multiple events which occurred years beforehand. Having highlighted or, in the case of apparent changes in testimony,95 manufactured them, these inconsistencies become the basis for accusations (direct or otherwise) of lying, which can add to the child‘s stress. Indeed, as we saw in Chapter 2, by the time a child witness takes the stand, they have been waiting on average 19 months to testify, nearly 2¾ hours in the courthouse across all days waiting to give their evidence. This is unlikely to be ideal preparation for children to withstand coercive questioning techniques in the courtroom or bear attacks on their integrity.

Judges have a duty to control inappropriate questioning. Section 14 of the Evidence Act 1908 prohibited lawyers from posing scandalous or indecent questions to witnesses, as well as questions ―intended to insult or annoy, or needlessly offensive in form,‖96 while s23F(5) of the Evidence Act, inserted under the Evidence Amendment Act 1989, stated that judges can disallow questions which are intimidating or overbearing given a complainant‘s age.97 In the Evidence Act 2006, which came into effect on 1 August 2007, there is now reference to questions which are too complex for the witness to comprehend:

Indecent or scandalous, although such question may have some bearing on the case before the Court, unless the question relates to facts in issue, or to matters necessary to be known in order to determine whether or not the facts in issue existed; or (b) Intended to insult or annoy, or needlessly offensive in form, notwithstanding that such question may be proper in itself.‖

―In any proceeding, the Judge may disallow, or direct that a witness is not obliged to answer, any question that the Judge considers improper, unfair, misleading, needlessly repetitive, or expressed in language that is too complicated for the witness to understand.‖ Section 85(1) Evidence Act 2006 [emphasis added]

Section 85(2) notes that, amongst other things, the age or maturity of the witness is to be taken into account when applying this rule.

The transcripts which Davies & Seymour (1998) examined related to trials held under the Evidence Act 1908 while those analysed in the current study were held under the new Act. The purpose of this section is to compare the quantity of judicial interventions and prosecution objections during cross-examination between Davies & Seymour‘s data and the current data; we also examine whether or not judges and prosecutors are intervening on a wider range of issues, including complicated language.

In the transcripts analysed by Davies & Seymour (1998), prosecutors in two different trials made four objections in total, while judges intervened six times. The prosecution objections related to a defence statement that was inaccurate; defence asking the witness about events she had already indicated she didn‘t remember; a question where the witness was asked to comment on what someone else was thinking; and in relation to questioning on the child‘s inconsistency over a peripheral detail. As for the judicial interventions, three judges asked the child to clarify ambiguous answers; one asked defence to rephrase a question from He will also say... to If he were to say...; the final example was of a judge directing the child to answer the question.

In the 16 trial transcripts analysed here, prosecutors in four different trials made a total of five objections. Unfortunately, the trial transcripts only indicated that an objection had been made, but did not give the substance. However one followed a defence question asking the child to comment on what someone else might have been thinking and another followed a defence statement indicating what the accused will say in her evidence. The third followed a very complex defence question.

Judges intervened 38 times in 10 of the 16 trials.98 In seven cases, the intervention was in response to the child‘s evidence: the judge either asked questions to clarify the child‘s evidence or asked the child to clarify an ambiguous response. The following is an example of the latter. The question on line 5 requires a yes or a no response, but either one would be ambiguous.99 The judge therefore asks the child in line 7 to resolve that ambiguity:

response could mean, you are right, she never caused injury or no, you are wrong, she did cause injury.

1	Defence	She never hit your wrist or your arm with [instrument], did she?
2	Child	She did.
3	Defence	Never hit you?
4	Child	She did.
5	Defence	Never caused any injury to your skin, did she?
6	Child	No.
7	Judge	Bobby, did you mean no she never caused you any injury or did you mean no she did. I wasn‘t sure of your answer.
8	Child	Um, no she did cause injury.

In 10 cases, the intervention followed a complex defence question. In six of these, the judge indicated that the question was too difficult for the child (or the judge) to understand or asked the witness if s/he had understood. For example, in the following exchange the judge tells the defence lawyer on line 6 to ask the question more clearly, then steps in to show the lawyer how to do so in line 8:

1	Defence	Now do you remember what day you said that the later sexual violation occurred?
2	Child	Pardon?
3	Defence	Do you remember what date it was or when it happened?
4	Child	On the first one.
5	Defence	No the second.
6	Judge	Mr X, ask a question in a less confusing manner.
7	Defence	Yes, Your Honour. The—
8	Judge	Like this. When was the first time that you told anybody about sex abuse when you were a lot younger?
9	Child	I, I only—on—the first time I told anyone was—I don‘t know what was the date—I just told the nurse.

Two more of these involved complex concepts: dates or measurements. In the first, the judge reminded the defence lawyer that other witnesses will be called and in the second the judge reminded the 14-year-old witness that, if she doesn‘t know the answer, she should say so rather than guess. In the ninth case, the judge asked the defence to be more precise and in the tenth the judge indicated that the child did not have to answer a question about someone else‘s intentions.

A further intervention followed a defence question which presupposed facts previously denied by the 9-year-old witness. In line 4, the child denies playing PlayStation on the day in question; in line 5, the lawyer weakens the child‘s no response to a don’t remember response, then goes on to say after that (i.e. after playing on the PlayStation). The judge points out that the child has not accepted this and requires the lawyer to rephrase:

1	Defence	Well, my understanding is that you and Jordan went and played in Charlie's room with Chris.
2	Child	No.
3	Defence	On the PlayStation?
4	Child	No, we didn't play PlayStation.
5	Defence	You don't remember that.
6	Child	No.
7	Defence	After that, Chris had to go—
8	Judge	She hasn't accepted it, just rephrase.
9	Child	Well, in the morning, Bobby, do you remember going in the car with Chris?

The defence lawyer in one of the sexual abuse trials was reprimanded twice for his line of questioning. In the first, he was pulled up for confusing cross-examination with examination-in-chief after asking a series of open-ended questions that went over the evidence again. In the second, the lawyer asked the 15-year-old complainant a series of complex questions about a peripheral detail (whether she had picked up a baby or was forced to pick up the baby, which had no bearing on the sexual assaults), before the judge challenged that line of questioning (see lines 12 and 14). The defence lawyer argues on line 15 that the focus on the peripheral detail was to establish credibility (as advocated in manuals on cross-examination; see discussion on tactics):

1	Defence	And you picked the baby up and were holding it. Is that correct?
2	Child	He gave me the baby.
3	Defence	Okay now what if he were to say he didn‘t give you the baby?
4	Child	Pardon?
5	Defence	What is your comment if he will testify that he didn‘t give you the baby to hold?
6	Child	He did give me the baby.
7	Defence	And his wife will also testify that you picked the baby up yourself and held the baby?
8	Child	No he gave me the baby before he went to move the car.
9	Defence	So I put it to you that no one forced you to pick up the baby or hold the baby. Can you comment on that?
10	Child	He—he gave me the baby.
11	Defence	Well did he give you the baby or he forced you to hold the baby?
12	Judge	Relevance, Mr Y.
13	Defence	Relevance, Your Honour, she said she was forced to hold the baby in her statement.
14	Judge	Is that an ingredient of sexual violation, remind me.
15	Defence	Credibility, Your Honour.
16	Judge	Why don‘t you move on? I think you have explored it as far as you can go.

In two more cases, the interventions were aimed at the child witness‘ behaviour, namely, the need to answer the questions posed; in a further two, the judge intervened to reframe the defence lawyers‘ closed, declarative questions as open- ended questions or yes/no questions. In a further seven, the judge asked either defence or the child to repeat what they had said or asked defence questions in relation to exhibits. Finally, there were seven cases where the transcript showed only that the judge addressed counsel or the reasons for the intervention were not clear.

When we compare the judicial interventions with those from 1994, we see that judges in 2008 intervened far more often than 14 years ago and in relation to a wider range of issues, including the witness‘ ability to comprehend the questions, presupposing facts denied by the child, and a focus on irrelevant details. The greater attention by some judges to complex questions is encouraging and may be as a direct result of the broader definition of ―unacceptable questions‖ under s85 of the Evidence Act 2006. However, there were many, many more complex questions which were not challenged. It is not clear whether they were allowed because the judges did not consider those questions difficult or because judges are wary of the repercussions of intervening too often or for some other reason altogether.

The number of prosecutor objections, on the other hand, has remained relatively constant between 1994 and 2008, suggesting that prosecutors are still cautious about objecting to defence questions on the grounds of their complexity, misleading nature and so on. The preferred strategy seems to be to address any damage done retrospectively, during re-examination.

―...it takes no account of the distorting effect of suggestive questions, which get worse as the case is more bitterly contested ...

Wouldn‘t the wretched witness have to be made of marble to stay calm and unruffled under the cross-fire of interrogation and counter-interrogation, examination, cross-examination, and re- examination which he must endure at the hands of the two adversarial opponents? Just think of a frightened witness, a weak one, or a child having to give evidence under such conditions!‖ (cited Spencer & Flin, 1993, p. 271)

As we have seen, the children whose transcripts we analysed were frequently subjected to suggestive questioning by defence lawyers,100 to complex language, to dubious tactics such as abrupt changes in topic and intense questioning on irrelevant details (in the name of establishing credibility), as well as accusations of dishonesty and deceit. Neither the transcripts used here, nor those used in the Davies & Seymour study were scientifically representative samples. However these separate studies, along with others conducted in New Zealand (e.g., Zajac & Cannan, 2009;

Zajac et al., 2003), have reached a similar conclusion on the basis of different samples over time: there are problems with the way children are questioned in court. In Chapter 2 we saw the traumatic effect the courtroom experience had on some young witnesses.

Since 1989, New Zealand legislation has permitted children‘s videotaped forensic interviews to be admitted as evidence-in-chief. Forensic interviewers are specially trained in communicating with children and the national interviewing model is updated yearly to incorporate the latest developments in best practice regarding the questioning of children (Wilson, 2007). The benefits of this provision are evidenced here by the superiority of those interviewers‘ practices compared to those tolerated in the courtroom. It is time for New Zealand to seek ways to more fully incorporate best practice in the courtroom too, both in the interests of truth-seeking and in the interests of delivering on the pledges New Zealand made when it ratified the UN Convention on the Rights of the Child nearly 20 years ago.

While inappropriate questioning may result in a decrease in the accuracy of children‘s testimony, some studies have shown that accuracy increases when children are permitted to testify using modes that put some distance between the child and the courtroom and/or courtroom participants. It is to this that we turn in the next chapter.

When the Evidence Amendment Act 1989 came into force on 1 January 1990, alternative modes of giving evidence became available upon application to a select group of witnesses, namely, complainants under 17 years of age in sexual abuse cases. These alternative modes included admitting videorecorded forensic interviews as the whole or part of the child‘s evidence-in-chief at trial and allowing children to testify via two-way CCTV or in the courtroom with a screen placed between the child and the accused. These measures were introduced in recognition of the fact that the ordinary way of giving evidence (i.e. in court, unshielded) may cause unnecessary stress to some witnesses and impair their ability to provide best evidence; in addition, these modes were seen as going some way towards overcoming barriers to prosecuting cases of child sexual abuse (New Zealand Law Commission, 1996, p. 21).101

By the mid-1990s, these alternative modes were in common use, to the point of being ―the norm rather than the exception‖ (New Zealand Law Commission, 1996, p.

25) and judges were using their discretion to extend eligibility to other classes of child witnesses (i.e., children who did not fall within the statutory categories, such as non-complainant witnesses and complainants in non-sexual cases) (New Zealand Law Commission, 1996; Pipe et al., 1996). There was general consensus among practitioners at the time that these modes did indeed reduce the trauma of testifying, compared to giving evidence unshielded in court, and enhanced the truthfulness and quality of children‘s evidence (Pipe et al., 1996). The New Zealand Law Commission concluded in 1996 that these new measures were working well102 and proposed in a discussion document that eligibility be extended to all witnesses, including adults and witnesses in non-sexual cases.

the manner in which it is done and in making its import accessible to the jury. Recent statutory measures have

been designed to achieve both objectives. As aids to the ascertainment of the truth they are important contributions to the course of justice.‖ (R v S, 1991)

the disadvantages are minimal, and the benefits for people to whom the provisions presently apply are clear‖ (New Zealand Law Commission, 1996, p. 31).

By this time some practitioners had already formed a view that videorecorded forensic interviews and CCTV were most appropriate for younger children, while older children should give evidence in the ordinary way in court (Pipe et al., 1996, p. 22). Indeed, the view that CCTV was most appropriate for younger children was already evident before the Evidence Amendment Act 1989 came into effect (Whitney & Cook, 1990);103 the Working Party on Child Witnesses similarly reported in 1996 that there was an age cap on the use of videorecorded forensic interviews (13 years in one area and 10 in another) and that ―the older the child, the more likely that an order would be sought that the child would be screened and present in the courtroom‖ (Working Party on Child Witnesses, 1996, p. 21). A 1998 study of child sexual abuse cases held in Auckland courts confirmed these patterns: that study found that children aged under 12 used videorecorded forensic interviews and CCTV while those aged 12 and over used screens (Davies, 1998, p. 138).

Since that time, eligibility to alternative modes has been extended to witnesses in any proceedings, irrespective of age; such applications are made under s103 of the Evidence Act 2006.104 There is no presumption in favour or against the use of alternative modes and no age cap. Section 107 obliges prosecutors to apply to the court for directions regarding mode of evidence whenever there is a child complainant.105 By extending eligibility to adults under s103, it is arguable that

(h) the relationship of the witness to any party to the proceeding: / (i) the absence or likely absence of the witness from New Zealand: / (j) any other ground likely to promote the purpose of the Act.

evidence would be appropriate. If the defence does not object, the application is usually granted. If the defence

does object, the matter will be debated in closed court; in anticipation of this, the prosecution usually engages a psychologist to interview the child and the resulting report is made available to all parties and the judge. Defence may consent to the application at this point. If not, the matter will be decided by the judge. In reaching his or her decision, judges are guided by sections 103 and 107 of the Evidence Act. Section 103 sets

Parliament intended to reinforce the argument made under the earlier legislation that:

―There is nothing in [the legislation] which requires that the nearer a person gets to the age of 17 the less the Court should consider exercising the jurisdiction [to permit alternative modes of receiving evidence].‖ R v Raj [2007] NZCA10

In Chapter 2 we found that the patterns evident in the 1990s are still evident today: younger children used CCTV and videorecorded forensic interviews while older children used screens or gave evidence in the ordinary way. In this chapter, we explore these patterns with a different and larger sample (n=134). We begin with an overview of the sample before examining factors which may have a bearing on whether children were forensically interviewed and which modes prosecutors sought, namely, the child‘s age, their role in the proceedings, demographic factors, the child‘s relationship with the accused, the nature of the charges and region.106 There follows an examination of the reasons why applications were contested, based on a small number of cases, before we go on to consider the implications for current and future practice.

We note that the sample size is relatively small and there may be any number of reasons for differences found between outcome variables and the factors listed above, including the possibility of the difference being due to chance. Where we have conducted tests for statistical significance, we note that the results are tentative.107

out a non-exhaustive list of the grounds upon which a direction regarding mode of evidence can be made. Section 107 makes particular reference to directions when the witness is a child. The full text of s107 is as follows:

testing checks the extent to which the patterns observed have arisen by chance or evidence a robust pattern. To test categorical data, we have used Chi-square tests (―The Chi-Square Test procedure tabulates a variable into categories and computes a chi-square statistic. This goodness-of-fit test compares the observed and expected frequencies in each category to test that all categories contain the same proportion of values or test that each category contains a user-specified proportion of values‖ (definition from SPSS)). Specifically, we used the Continuity Correction result, rather than a Pearson Chi-Square, as the former is a more stringent test of significance. Where one or more cells had an expected count less than 5, we used a Fisher‘s Exact Test. To compare means (such as age), we used T-tests when comparing two groups or Tukey‘s (a post-hoc multiple comparison test) to compare three or more.

This study is based on anonymised applications for alternative modes filed in Auckland, Manukau, Wellington and Christchurch courts on behalf of 134 children. Permission to access this data was granted by the Chief High Court Judge and the Chief District Court Judge, and the research methodology was approved by AUT‘s Ethics Committee in June 2008.

Administrators at the Offices of the Crown Solicitors in Auckland, Manukau, Wellington and Christchurch filled out a questionnaire for every application filed in those courts in relation to a child complainant or witness for the prosecution over a 12-month period, from 1 March 2008 to 27 February 2009;108 hence data collection took place seven months after the Evidence Act 2006 came into effect in August 2007.

If an application was not contested, the administrators completed the questionnaire. If the application was contested, the administrators collected and anonymised the relevant documentation (e.g. all applications, psychologists‘ reports, judicial rulings) as well as completing the questionnaire. Information on the outcome of each application was provided by the administrators and the courts.

The characteristics of the 134 children in the sample are shown in Table 14.109 As can be seen, the majority of children were involved in trials in Auckland and Manukau (67%, n=90).

The children in the four centres did not differ significantly in age at the time of the application; however the intellectually disabled children were significantly older than children without such disabilities (p=.032). There were no significant differences

between the regional samples‘ composition in terms of the children‘s role (complainant versus witness).

For all of the cases in Wellington, the offences alleged by the complainant included sexual offences, significantly more than in Manukau (p=.009) or Christchurch (p=.032). Finally, more Auckland children were testifying against a family member than in Manukau (p=.005), Christchurch (p=.027) or Wellington (p=.002).110

Age	at application
▪	Mean	13.7	13.45	13.37	14.38	14
▪	Median	14	14	14	15	14
▪	Age range	5-17	5-17	8-17	8-17	5-17
▪	No. of children for whom data is missing	4	2	1	0	1
Role
▪	Complainants	114 (85%)	54 (87%)	22 (79%)	21 (81%)	17 (94%)
▪	Witnesses	20 (15%)	8 (13%)	6 (21%)	5 (19%)	1 (6%)
Gender
Male		23 (17%)	13 (21%)	6 (21%)	3 (12%)	1 (6%)
Female		110 (82%)	49 (79%)	21 (75%)	23 (88%)	17 (94%)
No. of children for whom data is missing		1 (1%)	0	1 (4%)	0	0
Intellectual disability
Yes		8 (6%)	2 (3%)	2 (7%)	4 (15%)	0
No		126 (94%)	60 (97%)	26 (93%)	22 (85%)	18 (100%)
Charges alleged by complainants
Sexual		96 (84%)	47 (87%)	15 (68%)	21 (100%)	13 (76%)
Non-sexual		18 (16%)	7 (13%)	7 (32%)	0	4 (24%)
Child’s relationship with accused
Intrafamilial		68 (51%)	40 (65%)	9 (32%)	11 (42%)	8 (44%)
Extrafamilial		45 (34%)	10 (16%)	12 (43%)	15 (58%)	8 (44%)
Unknown		21 (15%)	12 (19%)	7 (25%)	0	2 (11%)

The applications sought and contested, as well as the outcomes of the contested applications, are shown in Figure 18. That figure shows that, of the 134 applications, 23% (n=31) were contested by defence (in three cases it is unknown whether the application was contested or not). Of these 31 contested applications, 14 (10% of the full sample) were eventually consented to by defence, eight (6%) were granted by the judge and three (2%) were disallowed. In the remaining six cases, the trial did not proceed (2%), the outcome was pending at the time of writing (1.5%) or the outcome was unknown (1%).

Ninety-eight (73%) of the 134 children gave their initial statement to police by way of forensic interview; 36 (27%) did so by way of written statement.

Table 15 shows the significant associations found between the manner of children‘s initial statement and role, age and the nature of offences alleged by complainants. The figures tentatively suggest that complainants generally and complainants alleging sexual offences (rather than other types of offences) were more likely to be forensically interviewed.111 It also shows that the children who were forensically interviewed were younger than those who were not.112

Table 15 : Manner of initial statement: associations by age, role and nature of alleged offences

These patterns in relation to role and age mirror those found for the full data sample described in Chapter 2. However the witnesses in that sample were marginally older than the complainants. It was suggested in Chapter 2 that age may have mediated the tendency for more complainants to be forensically interviewed than witnesses. The findings of the current sample support a view that this was not the case, as complainants and witnesses here did not differ in age.

There was only one significant regional variation in relation to children‘s initial statement: while 88% (n=23 out of 26) of children in Wellington were forensically interviewed, only 56% (n=10 out of 18) of the children in the Christchurch sample were (p=.031); the children in these samples did not differ significantly in age or role.

Figure 19 shows how these combinations were distributed across the regions. In this section, we examine patterns surrounding the modes sought, including the regional variations evident in Figure 19.

Figure 19 : Modes of evidence sought by prosecutors (by region) (n=134)
Manukau Christchurch Wellington

(n=62)
0
0
0
4
2
6
6
8
10
11
20
28
31
29
29
30
Forensic interview/CCTV (n=57) Forensic interview/screen (n=30) Screen only (n=45)
CCTV only (n=2)

Applications for forensic interview to be admitted as child’s evidence-in-chief

Ninety-eight of the children in the sample had been forensically interviewed; in 87 of these cases (89%), an application was sought to admit the videorecorded forensic interview as the child‘s evidence-in-chief.

Table 16 shows how many applications for forensic interviews were sought across the four regions. The figures tentatively suggest that applications for the use of forensic interviews were more common in Auckland and Manukau than in Wellington and Christchurch, although the difference between Manukau and Wellington was marginal.

Table 16 : Applications for forensic interviews sought/not sought by region (n=98)

Auckland v Wellington p<.05 Manukau v Wellington p=.051 Auckland v Christchurch p<.01 Manukau v Christchurch p<.01

For children across all four regions without intellectual disabilities, there was no significant age difference between those children for whom an application was made and those for whom no application was made. Applications were made for six of the

eight intellectually disabled children; the two for whom no application was made were both 17-year-old witnesses.

Nor were there significant patterns to forensic interview applications depending solely on the child‘s role or gender, the presence or absence of intellectual disability (although we note the small numbers involved), the nature of the charges (sexual versus non-sexual) or the child‘s relationship with the accused (intra- versus extra- familial).

Table 17 shows how many applications were made for CCTV and screens in the four regions. The figures suggest that applications for CCTV were more common in Auckland and Manukau than in Wellington and Christchurch.

Auckland v Wellington p<.01 Manukau v Wellington p<.01 Auckland v Christchurch p<.01 Manukau v Christchurch p<.01

Table 18 shows the significant associations found in relation to age, role and the child‘s relationship with the accused. The figures suggest that, in Auckland and Manukau, applications for CCTV were more often sought for younger children; and that, in Auckland, applications for CCTV were more common for complainants than witnesses, although we note the small number of witnesses involved. The figures also suggest that CCTV was more often sought when the child was testifying against a family member.113

Table 18 : Applications for CCTV or screens by age, role and relationship with accused

There were no other significant patterns to the choice between CCTV or screens depending on the child‘s role in the proceedings, the presence or absence of intellectual disabilities (although we note the small numbers involved), the child‘s gender or the nature of the charges (sexual versus non-sexual).

Information on whether applications were contested was given for 131 of the 134 applications. Of the 131, 24% (n=31) of the children‘s applications were contested. The actual type of modes contested is shown in Table 19.

Our original intention was to systematically analyse the grounds upon which defence contested applications. However the grounds are often argued in closed court and records of those discussions were not readily available. Furthermore, in cases where defence initially opposes an application but consents before any discussion in court, the reasons for opposing may not be set out at all.

When a judge rules on an application, however, the defence grounds are usually summarised in that ruling. We received written rulings on nine of the 11 contested applications; the discussion below draws from information contained in those rulings. Due to the small number, we make no claim that these are a representative sample.

In four of the nine cases, defence objected to the admission of the forensic interview as the child‘s evidence-in-chief. Two objections rested on technical grounds. In the first, the defence objected on the grounds that the forensic interview had been recorded more than two years previously and the transcript contained gaps where the transcriber could not decipher some of the child‘s words. As a consequence, the child‘s testimony would not be fully available to the jury. The judge ruled against the

use of the child‘s forensic interview. In considering the fairness of the proceedings, the judge took into account both the incompleteness of the transcript and the time lapse. The relevance of the latter is entirely unclear to the authors given the delays in processing all trials (see Chapter 2). The judge also noted that s/he had no clear evidence of the child‘s views. However s/he gave weight to the views of the Victims Adviser, who had already prepped the child for court (and whom the judge consulted under s107(3)(b) of the Evidence Act) and who believed that giving evidence live would not be a particular problem for the child; in fact, she had already been prepped to do so.

In the second case, defence objected on the grounds that the forensic interviewer had not correctly administered the oath/affirmation to the complainant, now aged

16. This objection capitalised on a contradiction in the earlier Regulations that children over 12 make an oath or affirmation,116 despite the fact that forensic interviewers were not empowered under the Oaths and Declarations Act to do so. The judge noted that in the meantime the Regulations had been amended and that the new regulation, by indicating Parliament‘s intention, acted as a guide on how to reconcile the contradiction. The judge argued that the forensic interviewer had acted appropriately under the circumstances and ruled in favour of the prosecution.

In the third example, it was not clear from the judge‘s ruling what grounds were given by defence for contesting the use of the forensic interview. However the judge ruled in favour of the prosecution, having regard to the child‘s age (16), the nature of the proceedings and the evidence (i.e. sexual), the child‘s relationship with the accused, the child‘s personality (i.e. lack of self-confidence and fear of public speaking), and the need to minimise stress on the child.

In the fourth case, which also related to the child‘s use of CCTV, defence wished to have their own clinical psychologist assess the child before the court made its decision as to how the child would testify (the child, however, refused to be assessed by the defence expert). In deciding in favour of the prosecution, the judge took into account the child‘s age (17 years), her psychological condition, the nature of the proceedings and the nature of the evidence (e.g. sexual violation), the child‘s relationship with the accused (extra-familial) and the need to reduce stress on the child and to promote her recovery.

There were four cases opposing the child‘s use of CCTV and one opposing screens. In each of the CCTV cases, the underlying presumption was that evidence should be given in the courtroom either with or without a screen. For example, in one instance

oath appears to have been an oversight in the drafting of the Regulations; this oversight was subsequently

involving a 16-year-old complainant in a sexual abuse trial, the judge described the defence opposition as being:

―...primarily on the grounds that any stress that the complainant may suffer is the natural and ordinary anxiety, or stress, that attaches to all Court proceedings and she should be required in the circumstances to give her evidence, certainly further evidence in chief, and be cross- examined in open Court.‖

The defence argument here assumes that alternative modes are intended to address unnatural and extraordinary stress. The judge ruled in favour of the prosecution on the basis of the need to ensure fairness of the proceedings (―ensuring ... that the complainant is best placed to give an account of what happened in a neutral and objective environment, thereby giving the jury the fairest opportunity to evaluate her account‖) while still allowing the defence an opportunity to cross-examine. The judge noted that giving evidence in open court would be stressful for the child and would not promote her recovery.

In another sexual abuse case, this time involving a 12-year-old complainant, the judge described the defence grounds for opposing the CCTV application as:

―[Defence lawyer] informed the Court that he had been instructed by the accused to oppose the application but was not able to offer any grounds in opposition other than the broad proposition that evidence should be given in the courtroom and that screening is an appropriate alternative.‖

In this instance, opposition to CCTV appears to have been driven by the client, rather than the defence lawyer. The judge did not directly address the assumption that evidence should be given in court, but on the basis of the child‘s age and a written report from the child‘s school counsellor, the judge ruled in favour of the prosecution, noting further that the jury would not be disadvantaged by seeing the evidence on a screen rather than from a witness box.

In a further case involving a 14-year-old complainant, the defence lawyer‘s favouring of courtroom evidence over CCTV was explicitly linked to the right to confrontation in the judge‘s summary:

―[Defence lawyer]‘s concern is that there is no real basis for her to be relieved of her presence in Court citing the confrontation right. He argues also the principles of open justice.‖

In his ruling, the judge noted that the accused does not have a right to ―eyeball‖ the witness in court; rather, the accused has the right to know the evidence against him or her and to test that evidence and, ―...in so doing to confront as it were the witness with an alternative proposition‖, which CCTV testimony still permits. The judge noted that being a teenager would not mean that the nature of the evidence would be any less stressful or traumatic; he further noted the suggestion in the psychologist‘s report that the stress of courtroom testimony in front of the accused (a family member) may impact on the child‘s ability to testify. The judge ruled in favour of the prosecution.

Confronting the witness ―on equal terms‖ was cited in the one instance where the use of screens was opposed. This sexual abuse case involved a young complainant who was 17 years old at the time of the application but would be 18 at trial:

―The main nub of [defence lawyer]‘s opposition is that the main thrust of the defence will be based on consent, the parties being of similar age and attending a [venue] where alcohol was consumed and the accused would like the opportunity to confront the complainant on equal terms. Secondly, [defence] argues that the jury would wonder why it is necessary for an adult witness to be screened and (despite the usual warning from the Judge) may draw an adverse inference against her client.‖

As per the previous ruling, the judge noted that the law does not in fact recognise ―a

‗right‘ for an accused to confront or ‗eyeball‘ a witness‖ although the Bill of Rights Act 1990 does confer the right to be present at one‘s own trial and to cross-examine prosecution witnesses. On the basis of the child‘s own wishes, the trauma suffered, the nature of the evidence and fairness to the proceedings, the judge ruled in favour of the prosecution. The judge did not address directly the comment about why an adult witness would need such protection.

The final case involved an 8-year-old witness; in this case the defence submissions were sighted by the researchers. The defence lawyer objected to the witness giving evidence via CCTV, even though the two (older) complainants were using that mode. The grounds were that none of sections 103(3)(b)-(g) of the Evidence Act 2006 applied (see footnote 104 on page 94) and that, further, the child was a witness and not a complainant. Defence further argued that there was no evidence of threats or intimidation towards the witness; that the witness saw very little in any event; that

―an unfamiliar environment is an insufficient basis for the witness to avoid giving evidence in open court;‖ and that the child will be more accurate if he testifies in open court before the accused. The submission concludes that the child‘s use of CCTV would abrogate the respondent‘s right to a fair trial and that, ―...the need to minimise any perceived rather than real stress for the witness can be met by the provision of a screen ... without placing at risk the fairness of the trial for the respondent.‖

In the ruling, the judge noted that any 8-year-old would find testifying in a courtroom ―highly stressful to say the least‖, that the child‘s role in the proceedings is irrelevant. S/he ruled that although the evidence may indeed be unremarkable, the child‘s age, immaturity, the unfamiliar environment of the court and the child‘s fear of the respondent may all ―hinder his ability to give evidence‖; furthermore, the judge noted that the jury might find it odd if the youngest child testified in court while the two older complainants testified via CCTV. The judge ruled in favour of the prosecution.

In summary, while the reasons for contesting forensic interviews were varied, the reasons for contesting CCTV and screens included references to three broad assumptions: (a) that testimony should be given in court; (b) that the accused‘s right to a fair trial may be contravened by these measures; and (c) that the accused‘s ―right‖ to confrontation is also contravened. While no judge directly

addressed the first assumption, the rulings were clear that there was no basis for the second or third. Furthermore, in a number of cases the child was in his or her mid- to-late teens; in no case was this put forward by the judge as a reason to refuse access to an alternative mode.

The data presented in this chapter tentatively suggests that age was a significant contributor to the patterns surrounding modes of evidence applications; hence the patterns between age and mode found in the 1990s remain today. This is despite the Evidence Act 2006 extending eligibility to all complainants and witnesses, adult and child alike, subject to a non-exhaustive set of grounds. The current study also suggests that there are regional variations in practice, particularly in relation to CCTV applications. This emphasis on age is inconsistent with the spirit of the new Evidence Act while regional differences in practice suggest that the law is not being applied consistently across the country.

When Pipe et al. (1996) surveyed legal practitioners in the 1990s, lawyers suggested that children‘s testimony had less impact on the jury when it was relayed via CCTV or videotape compared to in-court testimony:

―Individual comments [from practitioners surveyed] indicated that if an older child is able to give his or her evidence in Court it will have most impact on the jury. However social workers and police members of sexual abuse teams, in particular, commented that facing an abuser may be too difficult, even for older children. Conversely, the main disadvantages noted in relation to closed circuit television and, to a lesser extent to videotaped interviews, were that they distanced the child from the Judge and the jury and lessened the impact of the child‘s evidence.‖ (Pipe et

There were also concerns that children giving evidence via CCTV may be so composed that their testimony became unconvincing:

―Indeed, closed circuit television was seen by some as clearly being very effective in reducing the stress for the child, so much so that it was this, rather than depersonalisation, that might affect the impact of their evidence on the jury. For example, one High Court Judge commented: ‗Juries tend to be suspicious of evidence given in this way ... the more relaxed attitude of child witnesses

contact (18%). Forty-four per cent of the 50 judges surveyed agreed that CCTV evidence resulted in a loss of

impact on the jury compared to in-court testimony (screened or otherwise) while 34% disagreed; 18% were undecided (Davies & Noon, 1991, p. 108). However not all barristers felt that CCTV reduced the impact on the jury; three suggested that CCTV ―gave the jury the advantage of seeing the child ‗close-up‘ which made it easier to ‗read the witness‘‖ (Davies & Noon, 1991, p. 79). This point about the visibility of witnesses on CCTV was also made in an Australian study on two-way CCTV: ―There was, however, some concern that the impact of that evidence might be changed as a result of being viewed via a television screen, although magistrates, in particular, were impressed with the ability to see the witness more clearly than when the child is in court‖ (Cashmore & De Haas, 1992, p. xi).

way CCTV (―live link‖): ―...there was concern among some prosecutors that if a child appeared very relaxed

These concerns were not raised by practitioners in relation to screens; indeed, this measure was not seen as particularly suited to one group of children over another. However comments from those surveyed suggested a view that screens were advantageous, ―perhaps especially for older children‖ because they did not have the distancing effect of CCTV (Pipe et al., 1996, p. 22). Some defence lawyers viewed screens negatively, perceiving them to ―convey the implication of guilt‖ despite judicial warnings to the jury to draw no such negative inference (Whitney & Cook, 1990, p. 22).

In the current study, we did not ask prosecutors to explain their choice of mode, nor did we receive as much information from defence submissions on the reasons for contesting applications as we had hoped. However it is possible that the concerns raised by lawyers in the 1990s about the impact of out-of-court testimony on jurors and the trial process are still alive and well today. If so, the patterns surrounding applications sought may be a pragmatic response to those perceptions. That is, prosecutors may be balancing the child‘s needs, on the one hand, with what is best for their case, on the other, by seeking CCTV for those they perceive to be most vulnerable (i.e. younger children, complainants, those testifying against a family member) and screens for those they perceive to be less vulnerable (i.e. older children, witnesses and those testifying against a non-family-member). It is also possible that prosecutors opt for those modes which are least likely to be contested by defence to avoid raising children‘s expectations.

Defence lawyers in the current study contested 22% (n=19 out of 87) of applications for forensic interviews to be used as part of children‘s evidence-in-chief; they also contested 22% (n=13 out of 59) of CCTV applications and 19% (n=14 out of 75) of screen applications. This was presumably based on a perception that these modes were prejudicial to their case or because, irrespective of the defence lawyer‘s views, their clients require them to do so.119 In our non-representative sample of judicial rulings outlining defence grounds for opposition, concerns were expressed that CCTV or screens are not fair to the defendant and contravene confrontation ―rights‖. In addition, one defence lawyer expressed a belief that the child would be more accurate if he testified screened in court than if he testified via CCTV. Throughout there was an assumption that testifying in court is the default mode.

The upshot is that some prosecutors may have a preference for older children and witnesses to testify in the courtroom (with a screen) and some defence lawyers may similarly prefer children to testify in the courtroom (with or without a screen).120 The

while recounting horrific events, the jury might have difficulty believing the story. They felt that a frightened child in the court room can be a more effective witness‖ (Murray, 1995, p. 113).

question is, then, who is right: does in-court testimony favour the prosecution and/or the defence or neither? To answer this question, we must return to the rationale underlying the provision of alternative modes of giving evidence.

Testifying at a trial is a daunting prospect for adult and child witnesses alike. Children may find the prospect—and the reality—of testifying in the formal, rule- bound and unfamiliar environment of the courtroom stressful; in addition, confronting the accused is one of children‘s most salient worries about the whole trial process (Dezwirek-Sas, 1992; Goodman et al., 1992). These stresses may, in turn, inhibit children‘s ability to communicate effectively on the stand:

―If children‘s testimony is compromised by confrontation or courtroom stress, then jurors‘ duty as factfinders may be impeded. In his dissenting opinion in Coy v Iowa (1988), Justice Blackmun argued that the fear and trauma associated with testifying in front of the defendant may traumatize the child and undermine the truth-finding function of the trial by inhibiting effective testimony. Indeed, confronting the accused is a major stress for child witnesses (Goodman et al., 1992; Murray, 1995; Sas, 1991; Spencer & Flin, 1990; Whitcomb, Shapiro, & Stellwagen, 1985) and the completeness and accuracy of children‘s testimony can be seriously hampered by intimidation and/or heightened emotion.‖ (Goodman et al., 1998, p. 167)

There is certainly evidence that alternative modes of evidence reduce the stress of testifying (while giving evidence and/or in anticipation) (Davies & Noon, 1991; Davies et al., 1995; Goodman et al., 1998; Murray, 1995),121 that less stressed children provide better quality evidence (Davies & Noon, 1991; Goodman et al., 1998; Murray, 1995) and that alternative modes may allow children to testify who might not have been able to otherwise (Cashmore & De Haas, 1992; Cashmore & Trimboli, 2006; Goodman et al., 1998; Murray, 1995).122 There is, furthermore, evidence from a court observation study that, even when stressed, children who testified via two-way CCTV were more consistent in relation to central events and the timing of events than stressed children who testified in court (Murray, 1995). Finally, an experimental study showed that children who testified in a mock courtroom showed memory impairment compared to children who testified in a separate,

court as were possible under the circumstances. It was felt there was a greater likelihood of ascertaining whether or not the evidence given was the truth‖ (Working Party on Child Witnesses, 1996, pp. 1, Appendix V). 121 In the Davies et al. (1995) study, children‘s anxiety was rated by observers; the ratings compared children‘s

anxiety while testifying via videotaped forensic interview with the anxiety of children who were examined live in court. In the remaining three studies, researchers compared the demeanour and anxiety of children who testified via CCTV with those who testified in court. All bar the Goodman study involved observation of actual trials.

private room although there were no significant differences in error rates between interview conditions; the children who testified in the mock courtroom also showed greater heart rate variability, a factor associated with stress (Nathanson & Saywitz, 2003).

There is no evidence that testifying via alternative modes reduces children‘s accuracy or compromises adults‘ ability to detect lies. Experimental studies have shown that some children performed equally or better on measures of accuracy and/or consistency when testifying via CCTV compared to in-court or face-to-face testimony (Doherty-Sneddon & McAuley, 2000; Goodman et al., 2006; Goodman et al., 1998).123 Experimental studies have also shown that, compared to in-court testimony, CCTV and videorecorded evidence do not impede jurors‘ ability to distinguish between children‘s intentionally untruthful and truthful testimony (Goodman et al., 2006; Orcutt, Goodman, Tobey, Batterman-Faunce, & Thomas, 2001, p. 366).124 Finally, we have found no research proving that children‘s ability to lie is assisted by shielding them from the accused; rather:

‖...the current research on whether shielding procedures facilitate lying indicates that children do not lie better when they are protected from face-to-face confrontation with the alleged defendant.‖ (Marsil, Montoya, Ross, & Graham, 2002, p. 223)

This same point was also made by the New Zealand Law Commission more than 10 years ago:

―There is ... simply no empirical evidence that people are less likely to lie when faced with the person they are accusing and, from the point of view of promoting reliability, there is no basis to conclude that alternative ways of giving evidence detract from either the rational ascertainment of facts or procedural fairness.‖ (New Zealand Law Commission, 1996, p. 28)

While putting some distance between the child and the accused and/or courtroom might benefit the child and result in more accurate and/or consistent testimony (all without interfering with the jury‘s ability to detect deception), it is critical that this introduces no element of unfairness to the proceedings. Few studies exist which examine actual jurors‘ perceptions of the fairness of special measures. Notable exceptions to this are two recent studies from Australia (Cashmore & Trimboli, 2006) and New Zealand (Blackwell, 2007). In the Cashmore & Trimboli study, 277 jurors from 25 juries were surveyed by questionnaire. All of the trials involved child witnesses testifying via two-way CCTV. The overwhelming majority of jurors believed CCTV was fair to both the accused and the child:

children‘s accuracy was described as follows: ―...[children in the video condition] were less likely to answer

incorrectly in response to closed questions (probably because they were less inclined to guess) and were more resistant to misleading questions. The impact of visibility [i.e. CCTV versus face-to-face] on resistance to misleading questions occurred in the younger children, raising their level of resistance to that of the older children. Furthermore, the interviewers spent more time managing younger children in the video condition, again suggesting that the children were more relaxed‖ (Doherty-Sneddon & McAuley, 2000, pp. 387-388).

―The vast majority of jurors saw [CCTV] as either ‗quite fair‘ or ‗very fair‘ both to the child complainants (250, 90.3%) and to the defendants (243, 88%).‖ (Cashmore & Trimboli, 2006, p. 6)

Furthermore, the overwhelming majority found the trial as a whole very fair or quite fair to the defendant. In the Blackwell study, the jurors interviewed (n=60 from nine trials) saw no prejudicial impact of CCTV and screens on the accused:

―Eleven (18.3%) participants spontaneously mentioned that there had been no impact on the accused‘s case as a result of the use of CCTV and screens and no participant made any mention of any negative impact that could be prejudicial to the case of the accused ... Most participants ... indicated that the use of CCTV and screens had a positive impact on their understanding because they would have felt uncomfortable about seeing the child in open court in person.‖ (Blackwell, 2007, p. 219)

As for the views of the accused, Murray (1995) interviewed eight men who had been defendants in Scottish trials where children had testified against them using two-way CCTV; five of the accused had been convicted and three acquitted. None believed that the use of CCTV had prejudiced their case. In addition:

―Two prisoners who thought the whole process had been unfair said the injustice they felt had nothing to do with the provision [of CCTV]. Indeed all of the men thought it was very appropriate that children should use the live television link since they believed it was harmful and frightening

for children to appear in a large court room in front of a jury.‖ (Murray, 1995, p. 148)125

These findings from actual jurors add weight to the findings of experimental studies which have reached the same conclusions:

―Thus, the [experimental] evidence is consistent across a number of different studies. When protective measures are used, the trial is generally perceived to be as fair as when a child testifies in open court. This is especially true when the judge provides a warning to the jurors not to allow the use of the shield or closed-circuit television to imply or suggest that the defendant is

While the empirical evidence shows that the accused is not prejudiced when alternative modes are used, the possibility still remains that jurors take a less positive view of the child‘s testimony. Experimental research investigating this possibility has produced mixed results. For example, in two separate—and elaborate—mock trials, jurors rated in-court testimony as more credible than video testimony (Goodman et al., 2006) and more believable than CCTV testimony (Goodman et al., 1998) despite the fact that, ―if anything, children who testified via CCTV were more accurate‖ (Goodman et al., 1998, p. 199). However, where the children‘s CCTV testimony was indeed more accurate, children were perceived as

going to say he would have had ‗serious talks‘ with his lawyer‖ (Murray, 1995, p. 149).

al. (1993) and Lindsay et al. (1995). What‘s Fair When a Child Testifies? Journal of Applied Social Psychology, 25.

more accurate (Goodman et al., 1998).127 Another mock trial study reported that children who testified in court were rated as more accurate, believable and honest than their counterparts who testified via CCTV (Tobey, Goodman, Batterman-Faunce, Orcutt & Sachsenmaier, 1995, cited Davies, 1999). Other studies found that credibility ratings were the same for in-court and video testimony, but both were higher than CCTV testimony (Eaton, Ball, & O'Callaghan, 2001); the same for in- court and video testimony (Swim, Borgida, & McCoy, 1993); the same for in-court unscreened, in-court screened and one-way CCTV testimony (Ross et al., 1994). However these latter three studies suffered from the same methodological flaw. The children testified using one of the various modes (in court, videotape, CCTV) and that testimony was videotaped; the videotaped testimony was then shown to the observers, hence the distinction between videorecorded and live evidence was compromised.

In a Swedish study, observers rated children‘s live evidence as more convincing than the same testimony relayed by video (Landström, Granhag, & Hartwig, 2007); yet, in another study, the same authors do not report any significant pattern between mode of evidence (live, the New Zealand-style two-way CCTV, video) and observers‘ perceptions of how convincing the child‘s story was (Landström & Granhag, In press). One of the previously-mentioned elaborate mock trial studies found that jurors empathised more with children who testified in court than those who testified on video (Goodman et al., 2006) while another mock trial study found no impact on jury empathy for the child or the defendant depending on whether the testimony was presented in court or via one-way CCVT (Orcutt et al., 2001). Finally, observers have rated children who testified in court to be more honest, attractive, intelligent and accurate than children who testified via CCTV (Orcutt et al., 2001) or more confident, pleasant, natural and forthcoming than children whose testimony was relayed by video (Landström & Granhag, In press).

Although the evidence is mixed, prosecutors may have been right all along: removing children from the courtroom may have a negative impact on how children‘s testimony is received by jurors. Furthermore, defence lawyers may be doing themselves a disservice by contesting such applications.

The key question becomes whether any bias results in fewer convictions. Several of the studies which reported a bias against children who used distal modes (CCTV and/or video) also reported no significant impact on convictions (after deliberations), for example, between:

Similarly, when we examined patterns for the full data sample described in Chapter 2, Chi-square tests tentatively suggest (bearing in mind the small numbers) no significant pattern between verdict (not guilty versus guilty on at least one charge) and:

That is, while testimony relayed over distal modes may impact negatively on jurors‘ perceptions of children and their testimony, these negative reactions do not appear to translate into significant patterns in relation to convictions. Indeed, it is likely that children‘s evidence—and their demeanour while giving it—are only two of many factors that jurors weigh up in reaching a verdict.131 For example, Blackwell found

than after watching CCTV testimony. However this relationship was mediated by jurors‘ perceptions of the

child‘s accuracy; that is, ―trial condition [e.g. in-court v CCTV] no longer had a significant effect on pre- deliberation verdict once jurors‘ ratings of Accuracy were controlled‖ (Orcutt et al., 2001, p. 363). Jurors were surveyed again after deliberations. At the post-deliberation stage, there were no significant main effects for the relationship between the mode of the child‘s evidence and verdicts.

different modes of evidence (in-court and videotaped) affected jurors‘ perceptions of the children‘s credibility; this then contributed to jurors‘ perceptions of the defendant‘s guilt. Jurors were surveyed again after deliberations; in the post-deliberation phase, there were no significant effects of mode of evidence on jurors‘ confidence in the defendant‘s guilt or perceptions of trial fairness. Having said that, the authors acknowledge that, ―...given the small number of juries included in the analyses, a lack of statistical power may have precluded detection of significant effects‖ (Goodman et al., 2006, p. 388).

viewing the child‘s testimony only, they were more likely to convict if the child testified in-court unscreened than when they testified in-court screened or via one-way CCTV.

CCTV images and videorecorded interviews: ―In an unpublished exercise, the resident judge of Liverpool Crown Court monitored the outcomes of vulnerable witness trials at his court between January 2005 and November 2006: 49 per cent of young witness trials resulted in convictions where the jury watched children‘s testimony

that the presence of at least three of nine trial factors was significantly predictive of guilty convictions, none of which related to the mode of testifying.132 The model also showed that the absence of any three or more of these factors predicted a not guilty verdict (Blackwell, 2007, p. 253).

In summary, the evidence for a negative impact of alternative modes on perceptions of children and their testimony is mixed; evidence for a negative impact on convictions is even weaker. Yet it is possible that when alternative modes—as well as other pre-trial and trial measures such as court preparation—achieve their aims, the end result can be a relatively composed child, whose demeanour is at odds with jurors‘ expectations.

Some have suggested that, in light of the experimental findings on jurors‘ perceptions of children and their testimony when it is delivered using an alternative mode, we need to reassess the emphasis on alternative modes for child witnesses.133 We would argue this is not an option: the courts have a duty to attend to the well- being of child witnesses and allowing children to testify via alternative modes is one way of doing so. Indeed, any system that relies on exposing children to confrontation or courtroom stress in order to get a conviction is not a system that is working well:

―...to exploit a witness‘s vulnerability and stress for the benefit of achieving a conviction seems a somewhat dubious procedure. ... the oral tradition of English Law is based on the objective evaluation of factual evidence and the emotional exploitation of witnesses by barristers is clearly incompatible with such an objective.‖ (Davies & Noon, 1991, p. 135)

An alternative is to ensure that jurors are instructed in ways that allow them to more objectively assess children‘s evidence and that address myths about child witnesses and child sexual abuse head-on (for a discussion of jurors' responses to common myths about child sexual abuse, see Blackwell, 2007; Cossins, Goodman-Delahunty, & O'Brien, 2009; Quas, Thompson, & Clarke-Stewart, 2005). For example, some of the witness behaviours which jurors and others associate with credibility, such as

on large plasma screens, as compared with 36 per cent where the jury saw it on out-of-date, small TV screens (the overall conviction rate at the court for this period was 22 per cent)‖ (Plotnikoff & Woolfson, 2009, p. 91). 132 This study was based on a sample of 137 sexual assault trials held across New Zealand in the mid-2000s; of

these, 98 (72%) involved complainants under the age of 17 at the time of trial and 18% involved adults testifying in relation to offences that had occurred when they were children. The nine variables were (1) complainant was 12 years or younger at the time of trial; (2) similar fact evidence given; (3) recent complaint evidence given; (4) assault included penile penetration; (5) multiple complainants (2 or more); (6) multiple charges (4 or more); (7) the accused admitted to some of the alleged offences but ―not to the point of a guilty plea‖ (Blackwell, 2007, p. 231); (8) positive medical or DNA evidence; (9) a witness to the offending. The model was predictive for the whole sample, the child witness sample and the older witness sample (Blackwell, 2007, p. 250).

using protective measures such as shielding or hearsay may not be as effective as more time-consuming and expensive methods like court preparation techniques and court schools, which have the potential to empower some child witnesses‖ (Marsil et al., 2002, p. 240).

emotional distress (Blackwell, 2007; Golding, Fryman, Marsil, & Yozwiak, 2003),134 are exactly the behaviours which alternative modes of evidence were introduced to prevent. At trial, the point could be made that effort goes into ensuring children are as unstressed as possible while testifying and that children may therefore appear more calm and composed than lay persons might expect.135 Jurors are warned not to draw adverse inferences in relation to the accused because a child testifies via an alternative mode; jurors should also be warned not to draw adverse inferences in relation to the child if those measures achieve their aims.

Such a warning would be consistent with the advice given in the judges‘ Bench Book (2006) that judges caution juries against any emphasis on the demeanour of key witnesses,136 given the body of research which points to the unreliability of demeanour and behavioural cues for assessing veracity:137

―Recent reviews of the research (Memon et al., 1998; Vrij, 2000) indicate that it is difficult to discriminate between truthful and deceptive behaviour, and that lay persons and professionals alike are not very good at detecting lies ... even in actual high-stake forensic contexts ... The poor performance in lie catching may partly reflect the fact that reliable behavioural signs of deceptive behaviour are difficult to detect, as in the case of so-called micro-expressions (Ekman et al., 1999), and partly that many of the signs we believe to be associated with deceptive behaviour are in fact not associated with deception...‖ (Kaufmann, Drevland, Wessel, Overskeid, & Magnussen, 2003, pp. 21-22)

―Recent investigations into the extent to which these methods assist witnesses and increase the amount of reliable evidence available to fact-finders, have all resulted in recommendations for greater use of alternative ways of giving evidence, in particular closed-circuit television and videolinks.‖ (New Zealand Law Commission, 1999a, p. 119)

Perhaps the most pertinent and concerning finding of this chapter is that there were significant regional variations in practice in relation to modes of evidence applications sought in this sample. In particular, applications for CCTV were more common in the northern cities than in the southern cities. The differences in the regional samples do not support any conclusion that children in Christchurch or Wellington were less vulnerable than their northern counterparts in terms of age or role or, for the Wellington children, the nature of the offences; rather the regional patterns suggest that practitioners may be taking different approaches to the legislation. As a consequence, fewer children in the Christchurch and Wellington samples were

witnesses‘ face during the trial. The victim was rated more credible when she was described as teary than when

permitted to testify using modes that address both confrontation and courtroom stress. This variation in practice runs counter to the notion that the law provides protection for all vulnerable witnesses, no matter where in the country they testify. It should be stressed, however, that we did not ask practitioners to explain their choices over modes sought or contested; if there are barriers to the use of CCTV in some centres, these barriers should be identified and appropriate measures taken to remove them.

The second most pertinent finding is that opposition to the child‘s use of an alternative mode was rarely sustained. Of the 125 applications for which the outcome is known (i.e. either defence consented or the judge made a ruling) and it is known whether the application was contested, defence consented at some stage to 114 (91%).

Children‘s use of alternative modes has been part of the New Zealand courtroom landscape for 20 years and there is every reason for this to continue. There is no evidence that such provision prejudices the case against the defendant; there is no evidence that such provision results in less accurate or more untruthful testimony; there is no evidence that such provision inhibits the jurors‘ ability to detect lies; there is no evidence that such provision interferes with the rational ascertainment of facts; there is no evidence that such provision impacts on conviction rates. Finally, none of these alternative modes interferes with the accused‘s rights under the Bill of Rights Act 1990.

There is, however, evidence that such provision improves children‘s access to justice, their levels of anxiety prior to and while testifying, and the accuracy of that testimony. But there is also evidence that a great deal of time and resources are spent on these applications by prosecutors, defence lawyers, psychologists, judges, administrators and the children themselves, particularly when the applications are contested and even though defence consents to most applications in the end.

In New Zealand, there is no presumption for or against the use of alternative modes. That is, there is no basis to the argument put forward by some defence lawyers that, in general, evidence should be given orally in court.138 In other jurisdictions, such as England and Wales, a stronger line is taken and there is a presumption in favour of the use of alternative modes for children (see s21 of the Youth Justice and Criminal Evidence Act 1999). The Queensland Law Commission has taken a similar line, recommending the presumption in favour of the use of CCTV in particular:

relevant circumstances‖. If there is no presumption in favour of the ―ordinary‖ way, then the ―ordinary way‖ of

testifying unscreened in court cannot be interpreted as meaning ―the expected‖, ―the normal‖ or even the default mode, despite the nomenclature. Rather, giving evidence unscreened in court is simply one of a set of equally valid alternatives available. By this analysis, any notion that, as a matter of course, evidence should be given in court is untenable.

―Such a presumption would reverse the current situation under which it is necessary for the prosecution to apply for an order authorising the use of closed-circuit television and to persuade the court that closed-circuit television should be used. Rather, it would be necessary for the party opposing the use of closed-circuit television to show why it should not be used...‖ (Queensland Law Reform Commission, 2000, p. 218)

Where CCTV is not available or where the child does not wish to use that measure, the Queensland Law Commission recommended that there should then be a presumption in favour of the use of screens, although the Commission notes that screens are an unsatisfactory substitute for CCTV (which have the added advantage that the CCTV evidence can be videorecorded and used in retrials).

We would argue that a presumption in favour of CCTV and videorecorded forensic interviews in particular is warranted in the New Zealand context, in the interests of facilitating all children‘s testimony (not just younger children‘s), in the interests of efficiency (reducing the time spent on applications as well as the possibility of taping CCTV evidence for use at retrials) and to ensure consistent standards of practice across the country. We note that allowing children to testify from a location away from the courthouse via live audiovisual link would have many of the benefits of CCTV in terms of courtroom and confrontation stress. Puawaitahi, the multi-agency centre in Central Auckland, and the centres planned for Wellington and Manukau could be suitable remote locations. In addition, this would prevent children from coming into contact with the accused and his/her supporters in the courthouse itself, which is a significant source of stress for young witnesses (Davies, 1998; Davies, 1992; Eastwood & Patton, 2002; Goodman et al., 1992; Hamlyn et al., 2004). Indeed, the Courts (Remote Participation) Bill currently before Parliament intends to pave the way for greater use of audiovisual links in the courts, including for the taking of evidence in criminal proceedings.

While we support the continued and more extensive use of pre-recorded evidence-in- chief (especially given the superiority of forensic interviewers‘ questioning compared to lawyers; see Chapter 3), as well as CCTV and other audiovisual systems, these are only some of the special measures that could be used to facilitate children‘s evidence. That is, there are other ways that children‘s evidence can be facilitated that hold promise for addressing some of the other issues identified in this report, such as long delays and inappropriate questioning techniques. It is to this that we turn in the next chapter, where we examine innovative measures for assisting child witnesses that have been adopted in overseas jurisdictions.

The purpose of this chapter is to consider whether there are processes for dealing with child witnesses in other jurisdictions outside New Zealand which surpass our current processes and which might serve as models for future development.

The chapter considers six jurisdictions: Australia, England, South Africa, Israel, Norway and France, which have found interesting and worthwhile ways of facilitating child witnesses‘ participation in criminal proceedings.

The main two innovations under consideration are first, the pre-recording of children‘s entire evidence (direct and cross) before trial and second, the use of skilled intermediaries to facilitate children‘s communication during examination. The chapter also considers the use of remote CCTV suites to take children‘s evidence from a location away from the court building and the provision of separate independent legal representation for complainants.

The information for this chapter was drawn from published works and through discussions with practitioners in each of the six jurisdictions.

In addition, two study tours were conducted. In November 2008, Peter Dean (prosecutor with Meredith Connell in Auckland), received a Winston Churchill Memorial Trust Fellowship to travel to Western Australia to canvass the views of senior practitioners in that state on the practice of pre-recording children‘s evidence. In January 2009, the author, Dr Emma Davies and Dr Kirsten Hanna travelled to the UK to canvass the views of senior practitioners, Home Office officials and academics about the intermediary system and pre-recording.

South Africa is of great interest because it has a long-standing and, where properly resourced, successful intermediary programme for child witnesses operating within a completely adversarial trial process.

South Africa introduced intermediaries in 1993 in s170A of the Criminal Procedure Act 1977, together with a relatively limited package of other special measures. Intermediaries operate at trial only, sitting with the child in separate CCTV rooms, where they paraphrase and relay every question and answer.

South African commentators claim that intermediaries markedly reduce children‘s stress, increase the quality of their evidence and may have some impact on conviction rates (Klink 1996; Centre for Child Law (University of Pretoria) & Childline, 2007, para. 129, 140-147; Coughlin & Jarman, 2002, 541; Jonker & Swanson, 2007,

South Africa is a constitutional democracy with a highly adversarial court system, but aspects of its practice are distinct from other Commonwealth jurisdictions. There is no jury, although expert lay assessors may sit with the judge in limited instances. South African commentators discussing child witnesses make very similar complaints about aggressive trial advocacy to those made in other adversarial jurisdictions.139 However, it appears that their advocacy is more belligerent than that of New Zealand, England and Wales and Australia and that even the cross-examination of children is aggressive (Klink 1996; S v Mokoena 2008; Hoyano & Keenan, 2007, p. 665).140

The overall situation of child witnesses in South Africa is poor, reflecting a state struggling to meet the needs of a huge, diverse and largely impoverished community. South Africa lacks many protections considered routine elsewhere. Before trial, children are usually interviewed multiple times by multiple people,141 wait in the general waiting area and are regularly threatened by the defendant and supporters.142 There is no provision for a support person or to admit pre-recorded video statements (Centre for Child Law (University of Pretoria) & Childline, 2007;

enforce limits, but B. Edwards (personal communication, 28 July 2009) confirmed from her observation advocacy is much less aggressive and less procedural here.

Schutte, 2005; South African Law Commission, 1997; South African Law Reform Commission, 2000).

Although there are dedicated sexual offence courts (but no judicial training)143 and there is provision for CCTV, screens and for closing the court in cases involving sexual offences or child complainants (s153(3) and s153(5) Criminal Procedure Act), the vast majority of child witnesses testify in court without any screen or CCTV. There are simply not the resources to provide those protections in most cases (S v Mokoena 2008; Centre for Child Law (University of Pretoria) & Childline, 2007, para.63).

Children also face delays of up to two years before trial (Jonker & Swanson, 2007,

18) and there is a lack of appropriately trained police, prosecution and court staff and, as will be discussed, intermediaries.

However, unlike in other jurisdictions under discussion, s28(2) of the South African Constitution makes ―[a] child‘s best interests . . . of paramount importance in every matter affecting the child.‖ Consequently, the higher courts have taken quite radical positions regarding children in court, even questioning the suitability of traditional adversarial processes, including cross-examination (Klink 1996; S v Mokoena 2008; Muller & Hollely, 2009).144 In 2009 the paramount Constitutional Court has also taken a harder line regarding the judge‘s responsibility to consider appointing an intermediary for a child (DPP Transvaal 2009).145

Under s170A(1) of the Criminal Procedure Act 1977, an intermediary may be appointed for a child under 18146 who ―would be‖ exposed to ―undue mental stress or suffering‖ testifying by another method. Under this test children with communication difficulties found not to suffer undue stress do not qualify for assistance.147

questionable and may not be in the best interests of the child.‖ Several courts have emphasized the courts‘

responsibility for facilitating the child‘s communication both in order to improve the quality of the evidence and to facilitate children‘s participation in the justice system.

while the Constitutional Court in DPP Transvaal held that the judge must consider making an appointment and give reasons for any refusal.

measures to facilitate communication: under s161(2)A a child under 18 may give his or her evidence ―by demonstration, gesture or any other form of non-verbal expression‖ which the intermediary shall relay to the court verbally.

How much stress is ―undue‖ has been debated extensively by the courts. Until recently, the child had to suffer stress beyond the norm for child witnesses (S v Stefaans 1999; Muller & Hollely, 2009, pp., 16-17; Schutte, 2005, p. 6), so that intermediary assistance was exceptional (Schutte, 2005, p. 6). However, in 2009 the Constitutional Court held that a child testifying in court about abusive acts will almost inevitably suffer unduly (DPP Transvaal 2009, para.108; Muller & Hollely, 2009).

The prosecutor must apply both for the intermediary and for CCTV (a child cannot access CCTV unless an intermediary is appointed) (Centre for Child Law (University of Pretoria) & Childline, 2007, para.137; Schutte, 2005, p. 6). However, in 2009 the Constitutional Court held that if the prosecution does not act the court must consider appointing an intermediary of its own discretion (DPP Transvaal 2009, para.114; Muller & Hollely, 2009, p. 21), and must give reasons for any refusal.148

If applications are opposed, expert evidence should be called to discuss the likelihood of trauma. Opinions vary as to whether many applications are opposed.149

The criterion for appointing intermediaries is very broad: under s170A(4)(a) the court can appoint anyone qualified as a social worker, pediatrician, psychiatrist, family counsellor, child care worker, teacher or psychologist. There is no accreditation process, qualification or training except on an ad hoc basis through various private institutions and charities.

To preserve intermediaries‘ neutrality, they do not meet the children prior to the day of hearing (Australian Federal Police & Prosecutions, 2005, p. 154; Schutte, 2005, p. 13). This causes some controversy within South Africa (e.g. Schutte, 2005, p. 17).

Children testify either via one-way CCTV from an ―informally arranged‖ room (s170A(3) Criminal Procedure Act 1977) (Muller & Hollely, 2009, p. 13; Schutte, 2005, p. 3) in the courthouse (occasionally an adjoining room with a one-way glass window),150 or very occasionally from outside the courthouse (Muller & Hollely, 2009, p. 14).

Within the room the child cannot see or hear anyone besides the intermediary, who wears headphones to communicate with the courtroom (although the judge can insist on speaking to the child directly) (s170A(2)(a) Criminal Proceedure Act 1977). The child is never directly questioned by lawyers, the intermediary even

administering the oath (s165 Criminal Procedure Act 1977). The court, the defendant and counsel, however, can hear and see the child (and the intermediary) via CCTV (Klink 1996; Australian Federal Police & Prosecutions, 2005, p. 154).

The South African intermediary is purely ―an interpreter‖ (Klink 1996, 411I) who communicates individual questions from lawyers in simplified, age-appropriate language ,(Klink 1996, 411I-J; Schutte, 2005, p. 13) and then relates the child‘s answers, including non-verbal statements (s161(2)a Criminal Procedure Act 1977). The judge retains the power to direct that a question be put verbatim (s170A(2)(b) Criminal Procedure Act 1977) (Klink 1996, 411I-J; Jonker & Swanson, 2007, p. 5;

Muller & Hollely, 2009, pp. 39-41; Schutte, 2005, pp. 11, 13). Intermediaries cannot comment on the intelligibility of questions or question-sequences or on whether the child understands what is being asked (Muller & Hollely, 2009, p. 43). Some commentators even believe that intermediaries cannot advise the court if the child needs a break or is tired or stressed (Muller & Hollely, 2009, pp. 43-44), although others disagree (Jonker & Swanson, 2007, p. 6).151 Practice may vary across the country.

The intermediary‘s ability to ―buffer . . . aggression and intimidation‖ (Jonker & Swanson, 2007, pp. 5-17) is considered as significant as the rephrasing of questions.

Although there is little published comment or empirical research (Jonker & Swanson, 2007, p. 7), it appears universally accepted in South Africa that intermediaries significantly reduce witness stress both by distancing them from aggressive cross- examination and shielding them from the defendant‘s presence (e.g. Centre for Child Law (University of Pretoria) & Childline, 2007, para 147; Jonker & Swanson, 2007; Muller & Hollely, 2009, p. 28; Schutte, 2005, p. 4). Apparently some South African children testify at extraordinarily young ages, even as young as 3 years old (Jonker & Swanson, 2007; Muller, 2002; Muller & Hollely, 2003)152 because ―the intermediary

The courts and commentators agree that the intermediary system improves the quality and extent of children‘s evidence (S v Mokoena 2008, 21; Centre for Child Law (University of Pretoria) & Childline, 2007, para.147; Schutte, 2005, p. 9). Some courts and commentators also argue that intermediaries can affect conviction rates (S v Mokoena 2008, 21; Schutte, 2005, p. 9).

In 2000 a national moratorium on the release of conviction and acquittal rates was imposed (Jonker & Swanson, 2007, 7) but there is some evidence that intermediaries working within a professional structure and with appropriate training can make a difference. Jonker and Swanzen studied 384 cases using intermediaries. Not guilty pleas accounted for 7% of the sample, 56% pleaded guilty, and 37% were withdrawn. There were no mistrials. Police statistics available from before 2000 show that for the same age of witness and type of crime without intermediaries 58% of offenders pleaded not guilty, only 9% pleaded guilty and 18% were withdrawn.

The higher courts are very supportive of the intermediary system (e.g., Klink 1996; S v Mokoena 2008), as are prosecutors (e.g., Blumrick, 2006; Centre for Child Law (University of Pretoria) & Childline, 2007, para.180), academics and other involved professionals (e.g., Coughlin & Jarman, 2002; Jonker & Swanson, 2007; Schutte, 2005). However, both the latter are highly critical of shortcomings in its implementation and of limitations in the legislation (e.g., Jonker & Swanson, 2007; Open Society Foundation for South Africa, 2006).

There are also suggestions that a proportion of counsel and judges dislike intermediaries. There are reports that defence counsel routinely oppose applications for intermediaries,154 that lower courts routinely refuse such applications, and that prosecutors often fail to apply (Hoyano & Keenan, 2007).

Several academics (Coughlin & Jarman, 2002; Muller & Hollely, 2009) and many practitioners (Jonker & Swanson, 2007, p. 18; Open Society Foundation for South Africa, 2006) complain that the system has serious flaws. The remainder of this section deals with these reported problems.

First, the threshold for eligibility is widely criticised as too vague, too stringent and too narrow, excluding many who might benefit and leading to controversy in the courts and a tendency to set the bar too high (S v Mokoena 2008, 29-30; Centre for Child Law (University of Pretoria) & Childline, 2007, para.130; Muller & Hollely, 2009, 24; Schutte, 2005, 6, 9). Although the Constitutional Court in S v Mokoena 2009 has alleviated some pressure by holding that complainant children testifying about abuse will almost inevitably be unduly stressed (DPP Transvaal 2009, para.108; Muller & Hollely, 2009, 20), the test still excludes non-complainant witnesses (Muller & Hollely, 2009, 25) and complainants with little stress but serious communication difficulties (S v Mokoena 2008, 29; Muller & Tait, 1999, 247-48; Schutte, 2005,

There is also widespread criticism, supported by one of the few published studies (Jonker & Swanson, 2007, 8), that there is a de facto limitation to victims of sexual violence (S v Mokoena 2008, para.30; Jonker & Swanson, 2007, 10; Schutte, 2005), and to children between 9 and 13 (Australian Federal Police & Prosecutions, 2005, 154; Centre for Child Law (University of Pretoria) & Childline, 2007, para.134).

Commentators call for a complete overhaul of the eligibility test (e.g.: Schutte, 2005, 8-9) and the new Sexual Offences Bill would allow wider access for ―vulnerable witnesses‖, a group which will automatically include children under 18 years old and victims of sexual offences without further test.

Second, the intermediary‘s role is widely regarded as too limited. Many, including the court in Mokoena, argue that intermediaries need greater powers: they should be able to alert the court if a child does not understand questions, is tired, stressed or requires a break (S v Mokoena 2008, 34), and some suggest that intermediaries should comment on the child‘s understanding of questions and on the suitability of questions and question sequencing (Muller & Hollely, 2009, 48).

Third, there is widespread criticism of the thin legislative criterion for the appointment of intermediaries (Open Society Foundation for South Africa, 2006, para.3.2.1; Schutte, 2005, 16).

Fourth, there are issues created by South Africa‘s financial and logistical problems. Technological support is seriously lacking, including CCTV equipment and suitable rooms (S v Mokoena 2008, para.27)155 and the lack of training and support is also heavily criticised (Coughlin & Jarman, 2002; Jonker & Swanson, 2007, 21-22). There is no central authority for intermediary services, no accreditation process, register (Centre for Child Law (University of Pretoria) & Childline, 2007, para.130) or national training programme (Schutte, 2005, 16). While a few non-governmental agencies in particular locations manage and train intermediaries (e.g., Jonker & Swanson, 2007), generally intermediaries appear to be appointed piecemeal on a needs-must basis (Open Society Foundation for South Africa, 2006, para.3.2). The result is stressed intermediaries and variable performance (Muller & Hollely, 2009, 46). There is also a serious shortage of intermediaries. The National Prosecuting Authority estimated that intermediaries are only available in 14% of South Africa‘s courtrooms (S v Mokoena 2008, 32; Centre for Child Law (University of Pretoria) & Childline, 2007, para.130). It seems that intermediaries are only available in major city centres (Coughlin & Jarman, 2002; Jonker & Swanson, 2007, 17; Open Society Foundation for South Africa, 2006, 3). Finally, there are criticisms of judicial management and the understanding of the legislation and of children‘s needs amongst the legal profession (OSF-SA Meeting Minutes 2006, para. (Jonker & Swanson, 2007, 21-22; Open Society Foundation for South Africa, 2006, para.3.1.1, 3.2).

The overall impression is one of patchy and ad hoc service delivery of widely variable quality. If more demand is put upon the system, as seems possible given DPP Transvaal 2009, the situation may require serious government investment.

At its best, the South African intermediary system lowers children‘s stress, improves the quality of their evidence and enables those who might otherwise be excluded to participate in the justice system, including even very young children.

The literature suggests that the system is nearly universally approved even while critics lament its shortcomings. There may however be opposition in practice from the highly adversarial legal profession.

However, the legislation and regulations are threadbare and there are criticisms of the definitions of the role, criteria for appointments and eligibility for assistance. There are also problems with the lack of centralised control, regulation and training of intermediaries and legal professionals alike. There is an obvious need for greater professional regulation and support.

Moreover, the system still leaves children facing a normal cross-examination whose structure, content and tactical manoeuvrings remain untouched.

England and Wales are of interest in this study for two innovative provisions for children: the intermediary service and provision for pre-recording children‘s testimony (in conjunction with an intermediary if necessary) although the latter is still not implemented.

This section draws upon the extensive published literature on the English156 situation and also on interviews with senior judges, police, counsel, an intermediary, academics, researchers and others undertaken by the author, Dr Hanna, and Dr Davies in early 2009.

The origins of the current situation lie in the mid to late 1980s, a period of remarkable academic research and advocacy in that country spearheaded by a loose group of high-profile and prolific academics, independent researchers and children‘s charities.157

Graeme Davies, Helen Dent, Rhona Flin, Dana Glaser, David Jones, Joyce Plotnikoff and Helen Westcott; the NSPCC and ChildLine in particular.

In 1986 the government proposed allowing CCTV to children in criminal cases, but was defeated, partly on the basis that the proposal was too narrow and in particular should include video interviews as evidence-in-chief. Meanwhile, in 1987 the English criminal courts allowed screens for children (something they had first done in 1919 but had subsequently restricted to adult witnesses in terrorism cases). The following year, the government managed to pass legislation allowing CCTV to some (s32 Criminal Justice Act 1988), but to deflect those seeking greater reforms, it established the "Advisory Group on Video-Recorded Evidence" or ―Pigot Committee‖ to investigate video interviews158 (Spencer & Flin, 1993, 103). In 1989 the Committee recommended a series of radical innovations reminiscent of Libai (Libai, 1969), including intermediaries, informal round-table hearings to examine children without the jury and the pre-recording of all the child‘s testimony, as well as video- interviews and CCTV. The report has set the perimeters of debate in the UK ever since.

The complete set of Pigot reform proposals became known by Professor J R Spencer‘s term, the ―full Pigot.‖ The measures the government was prepared to introduce became the ―half-Pigot‖. The half consisted of increasing the availability of CCTV slightly and introducing video interviews for some child witnesses to sexual or violent crimes (Spencer & Flin, 1993, 103-5). Videos were only obtainable if the child was available for cross-examination (Spencer & Flin, 1993, 176-85). The criminal courts later introduced communication aids, increased their use of the power to close the court whilst children testified and allowed counsel to remove wigs and gowns (Hoyano, 2007; Spencer & Flin, 1993).

The ―half-Pigot‖ was widely perceived as insufficient (Hoyano & Keenan, 2007; Spencer & Flin, 1993), and continued pressure from reformists finally found an outlet when in 1997 the incoming New Labour government introduced a new focus on witness rights and, ultimately, the Youth Justice and Criminal Evidence Act of 1999 (―the YJCE Act‖). The Act codified the six special measures already permitted159 and introduced via s28 the possibility of pre-recording cross-examination and/or re- examination before trial (―pre-recording‖) and via s29 examination through intermediary.

However, child witnesses in the English courts still face a range of difficulties including significant delays (Plotnikoff & Woolfson, 2009, 32-43), inadequate waiting facilities (Plotnikoff & Woolfson, 2009, 72-84), insufficient breaks during testimony (Plotnikoff & Woolfson, 2007b, 46; 2009, 126), broken technology (Plotnikoff & Woolfson, 2009, 96-98) and problems with cross-examination and levels of judicial intervention (Hamlyn et al., 2004, 55-56; Kebbell, Hatton, & Johnson, 2004; O'Kelly,

Kebbell, Hatton, & Johnson, 2003; Plotnikoff & Woolfson, 2006, 48-50, 73-74; 2009, 107-32) although the Coroners‘ and Justice Act 2009 (―CAJ Act 2009‖) introduces support persons for witnesses using CCTV (amending s27(5)(b) YJCE Act).

The eligibility provisions for the use of special measures under the YJCE Act are exceptionally difficult. This is in keeping with the drafting of the Act as a whole,160 which has been heavily criticized from the outset (Auld, 2001, p. 570). However, the Act was improved to an extent by amendments under the Coroners‘ and Justice Act 2009 (―CAJ Act 2009‖). This section takes those amendments into account although they have not yet been implemented.

While children under 18161 are automatically eligible for special measures, a series of presumptions and graduated provisos apply to the selection of the special measures, resulting from Parliament‘s apparent desire to limit the court‘s ability to avoid special measures (Hoyano & Keenan, 2007, pp. 633, 637-638; Spencer, 2008, p.

The eligibility provisions involve a series of stages. The first is simple eligibility to apply. Child witnesses and complainants under 18 are automatically eligible (s16(1)(a) YJCE Act).163 Defendants under 18 are also eligible

to use CCTV (s33A-C YJCE Act 1999 as amended by s47 Police and Justice Act 2006) and intermediaries (s33BA YJCE Act as amended by CAJ Act 2009). However, adult witnesses must prove a particular need under complicated provisions,164 and defendants under 18 are only eligible where the measure is necessary to ensure

a fair trial and their ―ability to participate effectively in the proceedings as a witness giving oral evidence in court is compromised by the accused‘s level of intellectual ability or social functioning‖.165

Once eligibility is established, the court must decide ―whether‖ any special measure(s) would be useful (―likely to improve the quality of evidence‖) and if so, determine ―which‖ measure or measures would best ―maximise .

. .the quality of such evidence‖ (s19(2) YJCE Act). Adult witnesses face a different set of provisions again.

fear and distress related to testifying. The court must then also find the likelihood of actual improvement in

measure to ensure a fair trial and suffer ―a mental disorder (within the meaning of the Mental Health Act 1983) or otherwise has a significant impairment of intelligence and social function‖ such that they cannot function

There are two provisos to the usefulness test. First, a video cannot be admitted against the interests of justice (s27(2) YJCE Act).166 Second, the child must also be available for cross-examination, unless the statutory hearsay exception for unavailable witnesses applies.167

However, the legislation then goes on to restrict the courts‘ discretion as to which special measure can be used for children. The ―primary rule‖ (s21(3) YJCE Act) dictates that where a child has made a forensic video it shall be used as their evidence-in-chief, as if the test of usefulness had already been passed (s21(2) YJCE Act), and the child shall then be cross-examined using CCTV.

The primary rule is subject to three exceptions which restore some of the courts‘ discretion. The exceptions are

(i) s21(4)(a): if the technology is unavailable; (ii) s21(4)(b): if it is not in the interests of justice to admit the video or.168

Further, following outcry over the way the primary rule restricted children and the prosecution, forcing the use of even substandard videos, another exception was introduced to allow children to opt out of the primary rule if doing so ―would not diminish the quality of the witness‘s evidence‖.169 However, such children are now deemed to require a screen, although they may opt out of this also.170

Originally, a child witness to sexual or violent offences was deemed in need of ―special protection‖, making exception (ii) unavailable to him or her and requiring the use (were it available) of pre-recording.. This has now been removed.171 Children could, were it available, veto pre-recording in favour of live CCTV, but if the video is

not used as evidence-in-chief, pre-recording of the rest of the testimony would be disallowed (Hoyano & Keenan 2007, 634).

Theoretically, s28 of the YJCE Act makes pre-recording available for all child witnesses (other than child defendants) and mandatory for children in sexual and violence cases. There are as yet no regulations or practice directions for the use of pre-recording, but the legislation envisages that it would take place under conditions virtually identical to a normal trial, although without a jury. The child would give evidence-in-chief via their videotaped interview and then be cross-examined using CCTV. The judge (whom the legislation assumes will also preside over the trial (Hoyano & Keenan, 2007, p. 657)), counsel and the defendant would be in the courtroom, thus recognising the defendant‘s right of confrontation. Children could be recalled for further cross-examination on discovery of new evidence (s28(6)(a)172 or

vital loophole in a scheme which otherwise hampers badly the prosecutor‘s ability to direct the case.

(2)(b) mentally or physically unfit to testify or who (2)(e) in fear. This section does not however seem much used in relation to children.

168 One of the anomalies of s21 is that it also makes an exception where the special measure is not likely to

maximize quality of evidence, but this directly contradicts s21(2) which deems the special measures of video and CCTV to maximize the quality.

complainants can now dispense with the video if it does not maximise the quality of their evidence: s22A YJCE Act 1999, inserted by s101 Coroner‘s and Justice Act 2009.

opts out of the primary rule (under s21 4)(ba)) or screens (under s21(4)(B)(a)) are in s21(4C) YJCE Act.

Pre-recording has been ―the most controversial of all‖ the YJCE Act‘s innovations174 despite the fact the UK has had analogous legislative provisions for over 100 years (Spencer & Flin, 1993, 90).175 Despite strenuous advocacy during the 1980s and early 1990s (Spencer & Flin, 1993, 83-86),176 including two Bills to revive the earlier legislation (Spencer & Flin, 1993, 89-90),177 and a brief high point when the Pigot Committee recommended pre-recording in 1989 (Home Office, 1989),178 the government remained opposed (Cosgriff, 1988; Spencer & Flin, 1993, 89-90).179

Even after the YJCE Act the unease persisted. The government deferred piloting pre- recording numerous times, until finally in mid-2004, following the Birch Report 2003 (which concluded pre-recording as originally envisaged in the Act was unnecessary (but see Hoyano, 2007, pp. 852-854; Hoyano & Keenan, 2007, pp. 661-663)) and opposition to the idea from some of the legal profession180 but without public consultation (Hoyano, 2007, p. 852), it announced that s28 would not be implemented. However, following a review of the law (Hoyano, 2007; Office of Criminal Justice Reform, 2007), in February 2009 the Government finally opted to retain pre-recording but not to run a major pilot because it is unlikely to become common practice, and effective evaluation would be impossible (Ministry of Justice UK, 2009). Thus, without renewed advocacy s28 appears unlikely to take root in the criminal court,181 although there is a strong possibility that without it the current processes for children violate EU law (Pupino, 2005; Spencer, In press (b)).

taken. However, the unease with pre-recording is equally long-standing: the provision seems never to have

onwards partly in rejection of the government‘s 1986 proposal to introduce video interviews only. Pre- recording was also discussed at the 1988 Cambridge International Conference in 1989. For the earlier history of pre-recording legislation see Spencer & Flin 1993, 83-86.

video and the video would be shown to the defendant and judge. If the video was admissible, the child, a support person, lawyers and judge would meet informally in chambers while the defendant observed through one way glass and communicated with counsel via earphone. The hearing would be videoed. The judge would have a discretion to allow further cross-examination in the same circumstances as before. At trial all videos would be played to the jury. See also Spencer & Flin 1993, 88.

and the Home Office were against allowing intermediaries or pre-recording and ministers had deep reservations also. However, later, the Law Society publicly supported the two Bills to update the 1894 pre-recording provisions and debate in the House of Commons disclosed support from the Criminal Bar Association and the Council of Her Majesty‘s Circuit Judges. Spencer & Flin believed opposition came from the influence of a few

Association apparently argued that the aim of keeping children out of court is largely achieved by the admission of forensic videos and CCTV, ignoring the stresses of pre-trial delay and of waiting at court.

However, our research found that professionals from all levels and disciplines liked the idea of pre-recording and agreed there were no philosophical objections (Hoyano & Keenan, 2007, p. 660). The main concerns with pre-recording both then and now amongst the practitioners and officials we consulted were the purely practical issues of developing effective procedures and guidance, arranging prosecutorial and especially third party disclosure with the necessary speed, especially given the slow pace of disclosure allowed by the Criminal Procedure and Investigations Act. However, these are practical problems for which practical solutions could be found (Hoyano & Keenan, 2007, p. 660; Spencer, 2008; Spencer, In press): as one very senior judge commented to us, if speed were required the Bar would soon adjust.182 Conversely, Spencer argues that s28 has been rightly abandoned because it allows pre-recording to take place too late in the process to be of much benefit, and what is needed is a better provision (Spencer, 2008; Spencer, In press).

Interestingly, the Supreme Court recently explicitly suggested that the Family Court make use of ―the full Pigot‖ including pre-recording, in the unusual event of a child testifying in that court (Re W, 2010).

An intermediary under the YJCE Act is a specialist in communication who assesses and reports on the witness‘ special language needs and facilitates communication during police interviews and trial examinations. They monitor questioning for developmentally inappropriate language. The role may also involve relaying answers that would not otherwise be understood or interpreting non-verbal methods of communication.

An intermediary system in conjunction with pre-recording was promoted in the UK from the mid-1980s (David, 1990; Spencer, 1987a 1987b 1987c 1987d; Williams, 1987a, 1987b, 1987c). In 1989 intermediaries became the only measure on which the Pigot Committee was divided, resulting in the recommendation that they be used only for very young children with whom it was ―absolutely impossible‖ for counsel to communicate. However, the idea appears to have been disliked by the legal profession and by the government (Cosgriff, 1988).

With the new government in 1997 came a change of heart. The Home Office began work on a pilot for six areas in 2001. An inter-agency steering committee was created to develop, inter alia, guidelines for police, counsel and courts, a national selection and training programme, a register and a central referrals service (Plotnikoff & Woolfson, 2007a, 6-7). The first accredited intermediaries began work in 2004 (Plotnikoff & Woolfson, 2007a, 5).

Several positive evaluations followed, the main one being Plotnikoff and Woolfson‘s 2007 The Go-Between. A phased roll-out began in mid-2007 and was completed in September 2008, making it available to all police forces and Crown Prosecution Services nationwide. During 2009 some aspects of management were devolved so that while overall responsibility for policy and strategy and for selection, training and referral remain with the Office for Criminal Justice Reform (―OCJR‖) (which took over from the Home Office in 2004), referring agencies now pay directly for intermediaries‘ services183 and the matching service184 is now run by the National Police Improvement Agency.185 Recent information suggests that referral rates remain healthy: by late 2007 there had been nearly 800 referrals nationwide (Office for Criminal Justice Reform, 2007) and at the time of writing there had been 3,500 referrals, at a rate of about 100 a month.186

Under s29(6) of the Act intermediaries are neutral officers of the court, under an oath similar to that of interpreters. As such, intermediaries are not ordinarily liable to testify (Plotnikoff & Woolfson, 2007a, 55-60).

There were originally around 150 government-registered intermediaries in England and Wales. This number has now fallen somewhat (Hoyano, 2010).187 Although the court may appoint an unregistered intermediary, only three unregistered intermediaries were appointed during the pilot (Plotnikoff & Woolfson, 2007a, 33) and according to those we interviewed, none or very few thereafter.188 The police interview guidelines strongly suggest that unregistered intermediaries should be a last resort (Birch, 2007, para.2.100).

Registered intermediaries are recruited from a range of professions. Currently, the vast majority are speech and language therapists,189 followed by psychologists, occupational therapists, social workers and others (Plotnikoff & Woolfson, 2007a, 13).190 The range of skills encompassed by the group is broad, including expertise in

defence solicitor) to a Registered Intermediary identified as being available to undertake the work, suitable in terms of the skill-sets and experience required and prepared to work in the geographic area concerned.

189 SALT training comprises a four-year degree and ongoing professional development, supervision and membership of appropriate professional bodies. The academic education includes extensive study of both normal and abnormal language development. Thus, SALTs are often considerably better qualified on communication issues than are the clinical psychologists previously considered for the role (e.g.: see Ellison (2001), 130, before the pilot).

children‘s language development and in a range of learning and physical disabilities, mental health problems and in various communication aids and systems (Plotnikoff & Woolfson, 2007a, 14). Intermediaries need both tact and robustness in order to negotiate an unfamiliar forum with senior members of another profession.191

Intermediaries are not employed full-time, instead fitting referrals around other employment (Plotnikoff & Woolfson, 2007a, 20-21). Despite concerns over lack of support, job satisfaction is high and there did seem to be a very low attrition rate (Plotnikoff & Woolfson, 2007a, 17).192 Recent figures however show intermediary numbers down from 150 in early 2009 to 99 and 33 new appointees awaiting final accreditation (Hoyano, 2010).

There are four phases in the intermediaries‘ role: (1) assessing witnesses‘ communication needs, (2) assisting at the police interview if requested (3) preparing a report from the assessment and (4) assisting at trial. The following is an overview of the normal process where an intermediary is involved from the police investigation through into the trial, beginning with the referral.

A request and description of the witness is submitted to the matching service, which identifies an intermediary with the requisite expertise (Office for Criminal Justice Reform, 2005, para.3.6.3-4; Office of Criminal Justice Law Reform, 2005). The fit between witness and intermediary is reportedly excellent.193

Referrals can come from the police, Witness Care Units (a witness liaison service), the Crown Prosecution Service (―CPS‖), defence solicitors and the court. Eighty-nine percent of referrals during the pilot were from the police (Plotnikoff & Woolfson, 2007a, 24). The police should consult the witness before making a referral (Office for Criminal Justice Reform, 2005, para.1.6, 3.2.1; Office of Criminal Justice Law Reform, 2005, 6) and witnesses may reject particular intermediaries (Office for Criminal Justice Reform, 2005, para.3.9.4; Office of Criminal Justice Law Reform, 2005, 6).

Practice guidelines emphasise that the decision to refer is based on a broad consideration of whether the child needs assistance to give ―their best evidence‖ (Office for Criminal Justice Reform, 2005, para.3.2.1, 3.5.1, 3.6.8; Office of Criminal Justice Law Reform, 2005). The police interview manual states:

―...as a general rule of thumb, an intermediary may be able to help improve the quality of evidence of any child who is unable to detect and cope with misunderstanding, particularly in the court context, i.e., if a child seems unlikely to be able to recognise a problematic question or tell

the questioner that they have not understood, assessment by an intermediary should be considered‖ (Birch, 2007, 7).

Theoretically the court must authorise intermediaries‘ involvement in the police investigation, but in practice consent is retrospective as charges will not usually have been laid at referral. The court first considers the appointment when an application is made for the intermediary to appear at trial (Hoyano & Keenan, 2007, p. 668). Anecdotally, very few appointments have been rejected, but some are opposed.

The intermediary first assesses the witness‘ communication abilities and needs. Assessments are highly specialised, going beyond the expertise of a New Zealand forensic child interviewer.194 They cover the witness‘ intelligence, developmental age and language competence, relevant medical conditions and any need for communication aids. They also consider issues for the interview including attention span, understanding of time, and comfort objects.

Intermediaries are not told about the alleged offence, unless it is relevant (i.e., the police need to know the witness‘ vocabulary for genitalia) (Plotnikoff & Woolfson, 2007a, 36). Assessments never cover the witness‘ credibility (Plotnikoff & Woolfson, 2007a, 22).

Interviews are usually undertaken at the witness‘ home. The guidelines allow more than one interview if necessary (Birch, 2007, para.2.99; Office of Criminal Justice Law Reform, 2005, 6) but in 80% of pilot cases there was only one (Plotnikoff & Woolfson, 2007a, 37). Interviews should be supervised by a police officer or other professional third party to forestall accusations of contamination (Birch, 2007, para.2.99; Office of Criminal Justice Law Reform, 2005, 6). Most intermediaries also take notes of or record meetings, although this is not mandatory (Plotnikoff & Woolfson, 2007a, 38-39).

Following the assessment, the intermediary reports to the Officer in Charge (―OC‖) in person and later writes a formal report for the court. This report is discussed separately below.

When reporting to the OC, intermediaries first advise whether the witness is capable of an interview. These assessments also contribute to the CPS decision to prosecute.195 If a forensic interview occurs, the intermediary can, if requested, help to plan and conduct it. The police have found this very helpful (Plotnikoff & Woolfson, 2007a, 40-41).

During interviews intermediaries may be required to relay answers that would not otherwise be understood or to interpret non-verbal methods of communication but generally they play more of a monitoring role, intervening to rephrase questions or indicate the witness needs rest as necessary. In pilot cases intermediaries intervened very seldom (Plotnikoff & Woolfson, 2007a, 40-41).

The judicial training material includes footage of a real interview in which a young woman witness used a book of symbols to communicate. The intermediary created a supplementary folder of relevant diagrams, including floor plans of her house and drawings of furniture. She rephrased various questions or repeated them more slowly. She assisted the witness to turn pages and relayed her replies. The witness clearly understood and demonstrated her approval or disapproval of the translations of her replies.

Eight out of ten police officers in the pilot found the intermediary‘s assessments and interview assistance helpful (Plotnikoff & Woolfson, 2007a, 40-41). Similarly, the two Oxford police officers we interviewed greatly valued intermediaries‘ involvement in investigations involving children under 7 or with obvious communication difficulties. While they saw little need for intermediary assistance when interviewing other children themselves, they did highlight the importance of intermediaries in the court setting. The judiciary and counsel we consulted frequently commented that interviews with intermediaries were better structured and performed than most police forensic interviews.

The written report of the assessment is regarded as one of the most useful aspects of intermediaries‘ work. There is no standard template, but the guidelines advise that reports should cover the witness‘ physical and mental state, communication style and system and its implications for court layout and equipment, the witness‘ likely concentration span in court, best approaches to questioning and suggestions for communicating with the intermediary while the witness testifies (Plotnikoff & Woolfson, 2007a, 73, 99).

Reports we sighted were clear, concise documents, around three to ten pages. Beginning with basic information (the intermediary‘s qualifications and the times and dates of assessment interviews) they described briefly the child‘s developmental age and capabilities and any special conditions. Next they gave a detailed account of specific language issues and the modifications necessary. Some included reference to a table with examples of problem syntax and a corrected version. Some included the witness‘ likely suggestibility and likely response to typical cross-examination tactics. Reports then suggested procedures for the trial (e.g., break intervals, or that the

court allow the child a toy or drawing materials) (Plotnikoff & Woolfson, 2008, 103).196

The reports are highly prized by counsel and the judiciary (Plotnikoff & Woolfson, 2007a, 44; 2008, 93-94).197 Counsel found them extremely helpful for planning examinations (Plotnikoff & Woolfson, 2008, 94), and the language examples were particularly appreciated (Plotnikoff & Woolfson, 2008, 95).198 Similarly, judges we interviewed found reports invaluable because they alerted them to previously unsuspected communication issues and to best practice (see also Plotnikoff & Woolfson, 2008, 94). One senior judge described how an intermediary recommended that a young autistic man be allowed a special toy while testifying. With the toy, the man testified very competently. The judge stated that without the explanation in the report he would never have allowed the toy and the man would probably have been unable to speak.199 Others stated that the reports validated their prior understandings and provided authority for interventions: judges used reports to monitor questioning and referred counsel back to them when they fell short. As a judge in the pilot said, the report ―tied both sides down to the propositions and I could refer to it in my summing up‖ (Plotnikoff & Woolfson, 2008, 94).

Applications for any special measure at trial can be made either by a party or by the court of its own motion (s19(1) YJCE Act). Anecdotally, very few have ever been declined,200 although some are opposed.201

Although not always followed (Plotnikoff & Woolfson, 2007a, 43; 2008, 99-101), the English experience is that the best way to ensure a successful trial is for the judge to issue pre-trial directions regarding procedures for the intermediary, commonly known as the ―ground-rules meeting‖, rather than leaving it to the day of trial (Jones, 2007; Plotnikoff & Woolfson, 2007a, 49). Accordingly, the OCJR advises that judges for intermediary trials be assigned early (Office of Criminal Justice Law Reform, 2005, 11).

The directions should cover a range of practical issues including where the intermediary is placed, and how the intermediary communicates with the witness and the court and counsel (Office for Criminal Justice Reform, 2005, para.3.10.6; Office of Criminal Justice Law Reform, 2005, 11). For example, where witnesses do not require full translation, His Honour Judge Hall typically directs that: (1) the intermediary acts only through the judge; (2) counsel conducts questioning as usual;

(3) if a problem arises the intermediary alerts the judge; (4) counsel then has one opportunity to rephrase the question and; (5) if the intermediary still has concerns then he or she may rephrase the question (Plotnikoff & Woolfson, 2007a, 45).202

Best practice appears to be for the intermediary to attend the pre-trial directions hearing so that the court and counsel can address any concerns with the report‘s recommendations directly (Plotnikoff & Woolfson, 2007a, 45).203

The same intermediary who was involved pre-trial usually attends court (Birch, 2007, para.2.99; Office of Criminal Justice Law Reform, 2005, 6). The intermediary sits or stands beside the witness in court or in the CCTV suite.204 The intermediary is always visible on the CCTV screen (s29(3) YJCE Act). There is no standard jury instruction regarding intermediaries, although it has been suggested one is needed (Plotnikoff & Woolfson, 2007a, 52; 2008, 100).

The Act states that the role is to communicate the questions to the witness and the witness‘ reply to the questioner, ―and to explain such questions or answers so far as necessary to enable them to be understood by the witness or person in question‖ (s29(2)) (see also Office for Criminal Justice Reform, 2005, para.3.11..3). The intermediary‘s usual role is to monitor the examination for problems (inappropriate language, flagging attention, need for rest), intervening as necessary. Depending on the situation, intermediaries may use communication devices or systems or simplify questions into more comprehensible language, without changing their meaning (s30 YJCE Act).

The intermediary is restricted to communicating individual questions and replies (Office for Criminal Justice Reform, 2005, para.3.11.3), suggesting that they are unable to intervene in relation to the pattern or sequence of questioning. When relaying answers, intermediaries ―communicate the witness‘s reply as given, however irrelevant or illogical‖ (Plotnikoff & Woolfson, 2008, 96) and should not flag up possible misunderstandings: ―It is up to the court to seek clarification from the intermediary.‖ Intermediaries are not responsible for witnesses‘ welfare and cannot intervene purely to protect them (Office for Criminal Justice Reform, 2005, para.3.11.3).

for example, in North London, Surrey. Judges in more remote areas are apparently particularly supportive of remote links.

During the pilot, most intermediaries conformed to Judge Hall‘s suggested guidelines, and addressed their concerns to the judge. Judges almost always accepted intermediaries‘ offers to rephrase questions (Plotnikoff & Woolfson, 2007a, 53; 2008, 94). Most intervened seldom and intervened for answers much less than for questions, unless the answer was unintelligible (Plotnikoff & Woolfson, 2007a, 54- 55). Intermediaries rarely suggested that answers showed misunderstanding (Plotnikoff & Woolfson, 2007a, 55).

The intermediary we spoke to said that her practice is to interrupt before the witness replies to a question to forestall concerns that she is intervening because of the substance of the answer. She refers back to her report to validate interventions. In her experience most counsel and judges are anxious to use appropriate language, although some had great difficulty understanding how to do so and had difficulty slowing down their questions (Plotnikoff & Woolfson, 2008, 95).205

Practitioners appear very satisfied with intermediaries‘ conduct at trial. During the pilot only three barristers and one judge criticised any intervention, and even they were generally satisfied (Plotnikoff & Woolfson, 2007a, 53). None of our interviewees described intermediaries intervening inappropriately or overreaching their role. This was attributed to good pre-trial communications and clear judicial directions (c.f. Plotnikoff & Woolfson, 2007a, 44-50, who found these lacking). Barristers we spoke to reported that their initial fears of losing control or rapport with the witness had not been realised.

It is important to note that the intermediary‘s presence does not negate the judge‘s pre-existing responsibility to control the questioning. However, in the pilot some judges and prosecutors had abdicated their responsibilities: even where they believed the intermediary had not intervened enough, some did not themselves step in (Plotnikoff & Woolfson, 2008, 95). Judicial interventionism is supported by the higher courts. Another senior Crown Court Judge in interview declared unequivocally that the intermediary is responsible for intelligibility while the judge remains responsible for overseeing the subject matter, repetition and must also control any harshness as the witness‘ welfare is the judge‘s responsibility.206 Lord Justice Hughes and Lord Justice Thomas stated firmly in interview that the appeal courts supported judges intervening to control examination and they endeavored to communicate this to the trial bench.207

The national training programme for prospective intermediaries admits around ten candidates per intake. The week-long programme outlines the criminal court system, interviewing procedures, report-writing, courtroom skills and the shift in professional responsibility from client welfare to the court.208 The emphasis is on practical learning, using role-plays of every aspect of an intermediary‘s role (Plotnikoff & Woolfson, 2007a, 55),209 including cross-examination at a real CCTV facility.210 Intermediaries must then pass two tests: one written and one role-play, assessed by a retired judge. One-day refresher courses are sometimes held and accredited intermediaries are subject to bi-annual re-approval (Plotnikoff & Woolfson, 2007a, 16).211

Professional and collegial support for intermediaries is difficult because they work independently and are scattered nationwide. Support is provided by experienced intermediaries acting as mentors and via an internet smart-site site and through regional support groups set up by intermediaries themselves.212 There is also an annual conference.

It appears that the intermediary system in the UK delivers significant benefits both for child witnesses and for the courts. The system attracts strong support from professionals.

The police, the Bar, the judiciary and most academics agree strongly that the service results in major improvements in the quality and completeness of children‘s evidence and that it increases access to justice for children formerly considered incapable of testifying competently and those who had previously struggled. In particular, younger children and those with special issues (e.g., behavioural difficulties and autisitic spectrum disorders) who had been virtually excluded from court are now able to testify. This is a remarkable outcome compared to the previous situation.

Many witnesses and their carers in the pilot believed that without the intermediary their testimony would have been impossible. Similarly, practitioners in the majority of the first trials believed that prosecution would have been impossible without the intermediary (Plotnikoff & Woolfson, 2007a, 57-8; 2008, 92).213 In another case an

intermediary‘s involvement resulted in a wrongfully accused suspect‘s release when the witness was enabled to communicate that the police had the wrong man (Plotnikoff & Woolfson, 2008, 96).

Even when cases did not proceed to court the police commented that they had noticed that witnesses and carers felt they had been able to communicate their concerns fully (Plotnikoff & Woolfson, 2007a, 59).

Second, there was agreement amongst practitioners, academics and all of the witnesses‘ carers in the pilot that intermediary assistance markedly reduced children‘s stress when testifying (Plotnikoff & Woolfson, 2007a, 60).

Third, intermediaries have increased professionals‘ understanding of the difficulties children face in court and of best interviewing practice. Remarkably, every professional with whom we spoke described working with intermediaries as educative, if not ―revelatory‖. Judges and police have become more ready to modify standard court practice to accommodate children‘s needs (Plotnikoff & Woolfson, 2008, 93-94).214 Some police interview practice has become more child-friendly and rigid adherence to the interview guidelines has sometimes been relaxed, although there are still problems (Stern, 2010, pp. 68-69). In court, intermediaries have, for example, facilitated decisions to hear children‘s evidence in the morning only, to use a puppet to pose questions when a young child ―froze‖ during cross-examination215 and to allow witnesses comfort objects and drawing materials in court.216 The police we interviewed believed that such modifications would have been impossible without the intermediary because the courts react poorly to similar suggestions from police.

Significantly, intermediaries did not impede counsels‘ or the judiciary‘s work and nor did they adversely affect the rights of the defendant to a fair trial. Judges we interviewed had not found that intermediaries interfered with their control of their courtrooms. Barristers had not found intermediaries interfered with their control of or rapport with the witness. There were no fears amongst either group that intermediaries impeded the right of the defendant to a fair trial. Instead, intermediaries‘ neutrality, methods, and performance were universally praised by those we interviewed,217 a finding which replicates that in the pilot evaluation (Plotnikoff & Woolfson, 2007a, 59-60; 2008, 95).

defendant was convicted and given a 10-year sentence.‖ Another rape case ended in a guilty plea on the eve of trial. The defendant received an eight year sentence. The police officer said that without the intermediary scheme the victim—who had severe learning difficulties—would not even have been interviewed. The pilot did not find intermediaries had any measurable effect on conviction rates.

Nor did intermediaries increase practitioners‘ workloads. The police reported only very minor additional paperwork and said that the benefits of the intermediary‘s involvement far outweighed any disadvantage. Neither lawyers nor judges complained of their workload, despite the additional reading and issues before and during the trial. Instead, they were grateful for the assistance.

Most professionals interviewed noted that they had, or that they had colleagues who had, not initially seen the usefulness of the service but they stressed that this view changes completely after one has experienced the scheme.

However, some police and barristers we interviewed would restrict intermediaries to children under 7 or those with manifest communication difficulties, as they believe that developmentally normal children over 7 are generally capable of testifying unaided.

Although the intermediary scheme is highly praised, there are problems. First, as described above, the YJCE Act is quite remarkably intricate ―linguistic linguini,‖218 and in its original form unduly restricted the child and the party calling him or her in the selection of the appropriate special measures (Birch, 2000; Hoyano & Keenan, 2007, pp. 630, 633-638). The exclusion of defendants from most of the special measures was also unfortunate and practitioners reported that a surge in judicial creativity ensued, using their inherent jurisdiction to appoint intermediaries for vulnerable defendants. The criticism eventually resulted in the amendments introduced in the CAJ Act 2009, referred to above.219 The YJCE Act remains, however, a most complex document.

Moreover, there have been difficulties with the piecemeal introduction not only of pre-recording or intermediaries but also of measures such as video interviews (Spencer, 2008).220

There are also major practical problems identifying eligible witnesses (Burton, Evans, & Sanders, 2006; Plotnikoff & Woolfson, 2007a, 11, 62; 2009, 21-27).221 The pilot attracted far fewer child referrals than was anticipated (Plotnikoff & Woolfson, 2007a,

62) and many were not made until trial (Plotnikoff & Woolfson, 2008, 11). Recent

impaired communication abilities and (5) allows witnesses a support person in court.

intermediary. One newspaper reported that, whilst remaining clear and consistent, the child became anxious and tired and was confused by the lawyers‘ language: Anthony 2009. However the jury convicted and the conviction was upheld on appeal, the Court of Appeal resoundingly rejecting submissions that such a young child could not be a competent witness (R v Barker 2010; Spencer 2010a in press; D Wurtzel, personal communication 6 April 2010).

research (Plotnikoff & Woolfson, 2009) shows that take-up rates for all special measures remain low (around 50%) and many eligible witnesses remain unidentified (see also Burton et al., 2006; Cooper & Roberts, 2005; Hoyano, 2007; Plotnikoff & Woolfson, 2007a). CCTV and FVI use are hampered by poor-quality technology and lack of facilities, causing delays and forcing some children to testify in court (Hoyano, 2007; Plotnikoff & Woolfson, 2007a, 47-49; 2008, 101; 2009, 90-94).

The major reason for poor referrals appears to be that although children are automatically eligible, practitioners in the pilot and in our interviews tended to apply an ―age-plus‖ requirement: believing that normal children can cope with questioning from around 7 years old, they only refer the very youngest or older children suffering some clear communication disability (Plotnikoff & Woolfson, 2008, 97-99), echoing the eligibility criterion for adults. In the pilot, under half the children were referred for age alone,222 and almost all of those were extremely young (i.e., 3 to 5) where age alone already functions as a clear impediment to communication (Plotnikoff & Woolfson, 2007a, 25, 27). It would be unfortunate if only the youngest children are offered intermediaries as studies new and old show that communication difficulties are more widespread and harder to detect than ordinarily thought, and that even developmentally-normal children over 7 have serious difficulties with lawyers‘ questions and in recognising their own misunderstandings (e.g., Brennan & Brennan, 1988; Davies, 1998; Ministry of Justice UK, 2009; Plotnikoff & Woolfson, 2008, 98;

Third, arguably it would be preferable if intermediaries took a more active role in questioning. In England, as in South Africa, the restriction to individual questions and answers limits intermediaries‘ ability to address wider issues of misleading questions, including suggestive questions and question sequences223 (although some intermediaries are beginning to address these in reports) (Plotnikoff & Woolfson, 2008, 103). A wider role for intermediaries might increase the safety of cross-examination and quality of the resultant evidence.

Plotnikoff and Woolfson raise concerns that many intermediaries in the pilot were too diffident to intervene sufficiently (Plotnikoff & Woolfson, 2007a, 54-55). It is difficult,

―as the questioning [becomes] faster and more probing, . . . to monitor whether [a witness is] . . . being suggestible and processing the questions before answering, and therefore whether [to] . . . intervene‖ (Jones, 2007).

99 English and New Zealand barristers almost invariably nominated 5 or 6 years as the normal cut-off point for

a witness in the criminal court (Henderson, 2001). However in R v Barker 2010 the Court of Appeal came out strongly in support of the capacity of even a 4 year old to be a competent and compelling witness.

Hoyano however points out that more interventionist intermediaries might alienate the Bar and judiciary,224 who were, as we discovered, much relieved to find intermediaries usually kept a low profile. Hoyano also makes a case for leading questions as a legitimate testing device.225 However, the alternative viewpoint is that there is not much relevance in proving what is known to be true of most adults as well as children: that leading questions can mislead witnesses. Further, most lawyers use leading questions not to demonstrate suggestibility but to lead the witness to some contradictory point, which they then imply to be an admission or retraction (Henderson, 2002; Hobbs, 2003). It is hard to see this as legitimate testing rather than gamesmanship and possible misrepresentation of evidence. There seems no valid reason intermediaries should not intervene to point out the danger of miscommunication in a leading question.

According to the discussions we had, intermediaries at trial seem to be becoming more advisers and monitors than communicators: optional ―go-tos‖ rather than ―go- betweens‖.226 Even the preferred pre-trial directions described above, with their emphasis on the intermediary seeking permission from the judge before intervening, seem likely to discourage interventions except in the most pressing instances.

Given the research as to rates of miscommunication in trials (Hamlyn et al., 2004, 55-56; Kebbell et al., 2004; O'Kelly et al., 2003; Plotnikoff & Woolfson, 2006, 48-50, 73-74; 2009, 107-32) it would seem unfortunate if the intermediaries‘ role is restricted. If, as Plotnikoff and Woolfson comment, the availability of intermediaries means a higher percentage of suggestible witnesses will come before the courts (Plotnikoff & Woolfson, 2008, 103), there may even be an increasing need for intervention.

Nothwithstanding these concerns, it seems that there has been a reasonably significant shift in legal attitudes to child witnesses in England in recent years and that this has been crucial to the success of the intermediary system. It is often said that legal reform founders without significant change in the culture of the legal profession, especially as regards cross-examination (Birch, 2000; Ellison, 2001; Plotnikoff & Woolfson, 2008, 102; Spencer & Flin, 1993; Temkin, 2000). Qualitative research in the late 1990‘s indicated both a lack of awareness of children‘s language needs and developmental difficulties and a willingness to use suggestive and inappropriate language to discredit them (Davies et al., 1997; Henderson, 2001, 2002, 2003). Now however, barristers and judges we interviewed appeared more aware of children‘s needs and less ready to take advantage.227 This is not to say that trial practice is now child-friendly, as the research cited above regarding miscommunication shows. Trial observers report multiple instances of poor

examination practice228 and our interviews demonstrated a continuing underlying conservatism amongst legal practitioners. Plotnikoff and Woolfson (2009) and very recently Baronness Stern (2010) in her report on rape complaints note multiple failures to implement policy throughout the investigative and trial process. However, overall there are some improvements in practitioners‘ awareness of children‘s developmental needs.229

Amongst the possible causes of this change are long-term advocacy by the child witness reform movement and strong and proactive leadership from Lord Judge, Lord Chief Justice, and from the higher courts generally.230

Moreover, only accredited judges may preside over sexual offence trials (known, colloquially, and unfortunately, as ―having a sex ticket‖). These judges undergo a three-day training course (covering, inter alia, the legislation and the use of the special measures, debunking prevailing myths regarding sexual assault, patterns of sexual abuse disclosure and of the difficulties of cross-examination) with refresher courses every few years thereafter.231

The YJCE Act‘s pre-recording provisions remain unused, apparently because of fears over the practical issues of ensuring quick disclosure. However it is interesting to see that neither academics nor practitioners nor the judiciary saw any real philosophic difficulties in the scheme.

Conversely, the intermediary service has been very well-received. Despite some problems, it is felt that the English intermediary service offers significant advantages to child witnesses and to the courts in terms of evidence quality, access to justice and stress reduction, and that it does so without disrupting the rights of the defence or the smooth running of the trial.

However, it should be remembered that intermediaries only address specific language problems in individual questions and that children are still required to undergo a normal cross-examination unaltered in most respects. The service of course leaves wider issues such as delay untouched.

Nonetheless, in the pilot evaluation every witness or carer interviewed was extremely positive about the intermediary‘s involvement, many stating that the witness could not have coped in court without the intermediary (Plotnikoff & Woolfson, 2007a, 59). Perhaps it is this finding which deserves most notice.

Israel reformed procedures for child witnesses in 1955, the first country to do so by around 20 years. Israel created specialist youth interrogators who have sole jurisdiction over certain child witnesses, conducting interviews, determining how the child participates in proceedings and, if the interrogator vetoes the child testifying, even presenting that child‘s evidence in their stead. In many ways the system became the model and inspiration for reform in other adversarial countries.

This system was introduced to protect children from the trauma of testifying against family members and especially from the trauma of prolonged, aggressive cross- examination (Harnon, 1990, 81; Sternberg, Lamb, & Hershkowitz, 1996, 63) and to address concerns regarding under-reporting of abuse due to fear of the court process (Sternberg et al., 1996, 63).

Although Israel‘s private law is based on European Civil Law, its criminal procedure and evidence law are based on the English model (although without juries). Israeli advocates are known for aggressive advocacy and robust cross-examination (Harnon, 1990, 81).

All children, regardless of age, are able to testify subject to the court‘s determination of the child‘s ability to give an accurate account (United Nations Committee on the Rights of the Child, 2001, 181). However, no one can be convicted on the uncorroborated evidence of a child under 12.

Under s9 of the Law of Evidence Revision (Protection of Children) Law 1955 children under 14 who are witnesses to certain types of offence (sex, violence, prostitution or vice offences and parental abuse or neglect) and, since 2008, child defendants under

12 must be investigated by a youth interrogator. Interrogators also interview vulnerable adult witnesses, thus extending protection to children over 14.232

Interrogators are not purely concerned with the criminal trial but also work closely with and collect evidence for child protection officers. Parallel child protection actions are apparently common (David, 1990, 101).

The interrogator should interview the child within 72 hours of the reporting of the offence.233 Interviews take place in a variety of settings, including the child‘s home

(David 1989, 100), school or nursery as well as at the interrogator‘s office.234 Only the interrogator may interview children, excluding all the professionals usually relied upon. In most cases, the courts will not have any first-hand contact with the child (David 1989, 103). The child is usually only interviewed once: any supplementary interview must be approved by the interrogator and the interrogator conducts the re- interview.

The interrogator first assesses the child‘s reliability as a witness and then determines the child‘s role in the investigation and the trial:235 no one can overrule an interrogator‘s decision. The interrogator decides whether the child undergoes a medical examination, identity parades or reconstructions or testifies in court (Sternberg et al., 1996, 64). He or she accompanies the child throughout.236

The sole basis for all decisions during the process is whether the child will suffer

―mental anguish‖. Interrogators give two main reasons for refusing to allow children to testify: first, that the child is likely to suffer trauma due to testifying and second, that delays of months or years before trial make it likely that testifying would reopen the trauma of the events (David, 1990, 106-7; United Nations Committee on the Rights of the Child, 2001).237 After the interview the interrogator compiles a written report and sends it and the video recorded interview to the police who forward it to the prosecutor.238

Only a small percentage of child witnesses ever testify. Statistics are lacking but in 1984 28% of children were given permission to testify; only 6.8% could testify in 1993 (Sternberg et al., 1996, 65) rising to 15% in 1998 (United Nations Committee on the Rights of the Child, 2001, 1439).

If the child testifies, and the defendant is a parent or is accused of sex offences, the child may use CCTV or the defendant may be removed or screened from view (United Nations Committee on the Rights of the Child, 2001, 1437-38)239 on the basis of the likelihood of trauma or diminished quality of evidence (United Nations Committee on the Rights of the Child, 2001, 185, 1437-38). Interrogators can be appointed to act as intermediaries during questioning. The court can also take a child witness‘ live evidence early in the process at either party‘s request and with the youth interrogator‘s authorisation.

Before a child testifies the interrogator will read the testimony the child gave during the interview. The interrogator then stands with the child while he or she testifies. The interrogator cannot interfere with questioning until asked but can ask the court to discontinue the hearing if the child is too distressed (David, 1990, 107-8). The court can under s10 of the Law of Evidence Revision (Protection of Children) 1955 also initiate a discontinuance for a distressed child and may instead ask the interrogator to re-interview the child (David, 1990, 107). However, the interrogator can refuse to put any or all requested questions if he or she believes the questions are distressing (David, 1990; United Nations Committee on the Rights of the Child, 2001, 107).

Conversely, if the child is deemed unavailable to testify, the interrogator recounts the child‘s disclosures and presents an assessment of the child‘s credibility, having first explained the reasons for vetoing the child‘s appearance. However, the interrogator‘s account of the testimony must be corroborated, even where the child is between 12 and 14 and not bound by the corroboration requirement for all under 12s (s11 Law of Evidence Revision (Protection of Children) 1955). The corroboration requirement is designed to meet objections against convictions based on hearsay and on the lack of opportunity for the defence to test the witness. Counsel can also request the child be re-interviewed if new material emerges after the first interview, but the interrogator can refuse (Spencer & Flin, 1993, 398).

If the youth interrogator refuses to allow the child to testify, counsel‘s only opportunity to test the child‘s evidence is to challenge the interrogator‘s assessment of the child‘s credibility (David, 1990, 107). The only protections for defence are the strict requirement of corroboration and the request to re-interview (Harnon, 1990, 84, 92).

The youth interrogator system was reformed in the late 1980s following widespread concerns. First, the vast majority of cases involving child witnesses were never prosecuted, mainly for lack of corroboration. In 1988 it was estimated that 40-60% of cases investigated folded for that reason. Instead, most child abuse cases were resolved through plea bargaining (Sternberg et al., 1996, 65).

Second, the police and the judiciary complained about the poor quality of youth interrogator interviewing, which the police also blamed for the poor conviction rates. Judges complained that youth interrogators‘ questioning was unsystematic, leading, biased and that there were long delays before interrogators conducted interviews (Sternberg et al., 1996, 66-7).

Simultaneously the youth interrogators themselves were unhappy and disaffected (Sternberg et al., 1996, 70). While the original aim had been to create a ―unique

professional creature‖, training and infrastructure had been neglected (David, 1990, 99, 109; Sternberg et al., 1996, 67-69).

The resultant government inquiry, the Melamed Commission, recommended no reform beyond the audiotaping of interviews. There is a suggestion that the Commission feared prompting any parliamentary debate, believing that the youth interrogator system might well be abolished (Sternberg et al., 1996, 67). Audiotaping became mandatory in 1991, changing in 2001 to videotaping.240

Regardless of the Commission, shortly afterwards tremendous efforts were made to improve and standardise youth interrogator interviewing practice. The National Institute of Child Health and Human Development (―NICHD‖) protocol became mandatory in 1998 and is continually updated in line with ongoing research (M. Lamb, Y. Orbach, I. Hershkowitz, D. Horowitz, & C. B. Abbott, 2007, 1119-20). Professor Michael Lamb and colleagues began a series of detailed long-term studies of youth interrogators, resulting in an extensive published literature on the system, informing interview development internationally.

However, although there are no statistics available, prosecution and conviction rates apparently continue to be low.241 In 1996 it was estimated that only 10% of allegations investigated were tried, although the majority were credible (Sternberg et al., 1996, 65). In 2000 one article suggested that Israel was considering reverting to children testifying live because of the continuing low conviction rate (Cordon, Goodman, & Anderson, 2003, 185). However, this has not eventuated, although the difficulties in prosecuting continue and cases are generally still resolved by plea- bargaining.242

Regardless, the current system apparently satisfies the Israeli public. In 2002 the UN argued that a significant increase in child abuse reporting showed an increased confidence in the youth interrogators (United Nations Committee on the Rights of the Child, 2001). One contention is that the forensic interview quality is so improved that there are no longer any concerns.243

The Israeli system provides an excellent shield from potentially traumatic aspects of the criminal investigations: interrogators have robust powers to protect children and their interview practice is evidence-based and standardised using the NICHD protocol. However, this does not translate into prosecutions and convictions, partly

because of the strict corroboration requirements and possibly also a tendency to regard familial child abuse as a care and protection matter. The failure to prosecute alleged offenders is disadvantageous to children, allowing offenders to escape (Harnon, 1990, 92), potentially endangering the complainant and other children and denying the complainant the right of access to the courts.

The youth interrogator system would be very difficult to implement in other jurisdictions—inquisitorial or adversarial—because the interrogators‘ total control of witnesses impedes the parties‘ and judges‘ ability to pursue investigations. In particular, the inability of the defence to contribute to the examination of the child is a serious issue under most conceptions of natural justice.

Discussion now moves from intermediaries to the pre-recording of a witness‘ entire evidence before trial (or ―full pre-recording‖ as opposed to pre-recording the evidence-in-chief only). Although the USA has experimented with pre-recording since the 1970s (Anon, 1985; Armstrong, 1976; Humphrey, 1986; Stiver, 1974; Whitcomb, 1986), it has failed to make much use of the provision for child witnesses due to constitutional concerns (Spencer & Flin, 1993, 166-9; Whitcomb, 1986). Western Australia, however, began pre-recording the entirety of a child witness‘ evidence following a 1992 amendment to the Evidence Act 1906, and has continued to do so with considerable success. The system is seen as improving the quality of children‘s evidence, reducing children‘s stress levels, and increasing guilty pleas and withdrawn prosecutions.

Western Australia introduced pre-recording in addition to the standard raft of special measures then being introduced in many adversarial jurisdictions,244 although forensic interview videos were not admitted into court until 2004. When evidence is pre-recorded most children give their evidence-in-chief via a forensic interview video with a specialist interviewer. Cross-examination is then recorded during a pre-trial hearing, usually via CCTV. There are limited opportunities to recall the witness for further examination using the same process. All recordings are then replayed at trial. Pre-recording hearings must take place within six months of committal but interviews with practitioners suggest that most now take place within three.

Pursuant to s106I of its Evidence Act 1906, Western Australian children under 16 who witness sexual, prostitution or familial violence offences are automatically eligible to fully pre-record their evidence. Other vulnerable witnesses can also apply

to pre-record under s106R(3).245 Under s106O the court takes the child‘s wishes into account and can dispense with pre-recording or CCTV during pre-recording.

Applications for pre-recording are made by the prosecutor, usually at the first appearance after committal. Under s106K(1)(a) directions for procedures at the pre- recording hearing are given and a date is set. If, exceptionally, the application is opposed, a directions hearing is held. The court rarely rejects applications unless it can fix the trial date within a shorter period of time than it can fix a pre-recording hearing date.

Pursuant to s106K the judge may also direct other special measures be used in conjunction with pre-recording (e.g., CCTV).

The child sees his or her forensic video a day or so prior to the pre-recording hearing. At the hearing the child is usually examined via CCTV from a room in the courthouse.246 The judge, counsel (all robed) and the defendant are present in court. Under s106K the defendant must be able to see the child and communicate with counsel but always remains out of the child‘s sight (Australian Federal Police & Prosecutions, 2005, 144). The court itself is usually open to the public.

In the CCTV room the child is supported by a worker from the Child Witness Service (a court preparation and witness liaison service). There are two screens in the child‘s room: one showing the judge and the other the counsel speaking. The judge has two screens also, showing the child‘s head and shoulders and the whole CCTV room. Counsel see the child‘s head and shoulders both on a large screen and individual monitors. Later the jury watch the child on a large screen (Australian Federal Police & Prosecutions, 2005, 145). The camera must be able to show the child‘s body as well as face.

The child is taped walking into the room in order to give the jury an impression of the child‘s age and size. Children under 12 must undergo a competency test. If successful, the child is sworn in (although children may also testify unsworn) (s106C Evidence Act 1906 (WA)) (Harding v R 1989).

The prosecutor begins with introductory questions and then screens an excerpt from the forensic video.247 The child is asked to accept the video and the prosecutor may

the pressure of confronting the parties, especially defence counsel, was too great: ACT Report 2005 145.

then ask some supplementary questions. The defence then cross-examines and the prosecutor re-examines as usual.

The recording is then edited. Although theoretically the court must pre-approve edits, in practice the prosecution edits the tape in consultation with the defence and obtains retrospective permission (Australian Federal Police & Prosecutions, 2005, 144). None of the barristers interviewed had any concerns with the editing and it appears that nearly all conflicts regarding admissibility are settled by consent.

At trial the judge gives the jury a standard warning (1) not to infer guilt from the use of CCTV and pre-recording as both are normal; and (2) that counsel and judge may differ from those used at pre-recording.

The pre-recording is then played, stopping at the point where the child adopts the forensic video to play it in full. The pre-recorded cross-examination is then played. The pre-recording is used for any subsequent retrial or appeal hearing.

Children do not attend trial (s106I) and although theoretically they can be recalled, in practice applications are rarely made ―and even more rarely granted‖ (Australian Federal Police & Prosecutions, 2005, 144; Jackson, 2003) and only if there is a significant reason. If recalled, the witness‘ evidence is again pre-recorded.

The pre-trial hearing is supposed to take place within six months of committal. Prior to late 2008 this was of real advantage as there were significant delays between committal and the actual trial. In 2005 one study found 18-month delays between committal and trial (Australian Federal Police & Prosecutions, 2005, 144), similar to the 2002 average of 17.5 months (Eastwood & Patton, 2002, 49, 114). The situation in 2002 and 2005 had worsened since 1998 despite various initiatives to improve case management and prioritise cases involving children (Eastwood & Patton, 2002, 115).

However, in 2008, Chief District Court Judge Kennedy imposed a new regime where all trials are now allocated very quickly. Whereas previously there were on average 70 weeks between committal and trial, the delay is now 35 weeks, and less for cases involving child witnesses. The aim is to reduce delay to 20 weeks. Consequently, by late 2008 far fewer pre-recording applications were being granted as the trial could be convened as quickly as a pre-recording hearing. Judge Kennedy has stated that if the trial date can be fixed within three months of committal she will dispense with full pre-recording, except where the child is very young or intellectually disabled.248

Some counsel we spoke to did complain, however, that disclosure is causing delays before committal rather than afterwards, a problem which may be masked in the foregoing figures.

Barristers and judges we interviewed suggested that most or all child complainants in sexual cases use pre-recording but that few applications are made for child witnesses (including complainants) to other offences. Their discussions suggest child sex abuse complainants and younger children are regarded as more in need than other witnesses, with adult vulnerable witnesses rated lowest. It does seem that most children use CCTV, whether in conjunction with pre-recording or not. Similarly in 2002 Eastwood and Patton found that 30% of Western Australian child witnesses used pre-recording while 70% used CCTV alone.

Pre-recording is highly regarded by judges, counsel, other practitioners and commentators, all of whom cite numerous advantages for both children and the trial process.

First, pre-recording is believed by academics, commentators and by those we interviewed to be far less stressful for children than normal testimony (Eastwood & Patton, 2002, 49; Jackson, 2003, 205). Children who testified in sexual abuse trials in Western Australia in 2002 (when pre-recording was used only in Western Australia) were far more positive about the experience than children elsewhere in Australia. ―When asked if they would ever report sexual abuse again, only 44% in Queensland, 33% in NSW and 64% in Western Australia indicated they would‖ (Eastwood & Patton, 2002, 111). Thirty nine percent of Queensland child witnesses and 56% of those in NSW said they would not testify again as opposed to 17% of Western Australia child witnesses, regardless of the outcome of trial. Tellingly, in 2002 only 18% of Queensland barristers and 33% of NSW barristers would have allowed their own children to testify compared to 46% of Western Australian barristers (Eastwood & Patton, 2002, 111-12).

Barristers and judges we interviewed stated that the speed of pre-recording enables children to conclude their involvement in the investigation early and ―get on with life‖ and that it removes both the stress of waiting for trial and of the delays and interruptions at trial. However, Judge Kennedy‘s proviso that young or intellectually disabled children should be pre-recorded even if the trial can be convened quickly underlines that pre-recording has further advantages.

Professionals interviewed stated that pre-recording results in fewer objections to the child‘s testimony since in the absence of the jury, issues of admissibility can be resolved afterwards. The jury‘s absence also means courts are more willing to allow

the child breaks, including allowing support staff to prepare them to continue to testify. Ultimately, however, practitioners we spoke to thought that pre-recording resulted in savings of jury time as the edited video is quicker to view than live testimony.

Counsel and judges interviewed cited an improved quality of evidence due to its early capture. They praised the ability to edit prejudicial inadmissible evidence before the jury sees it.

Although there is no empirical evidence, practitioners we interviewed believed that foreknowledge of the strength of the child‘s evidence increased both guilty pleas and prosecution withdrawals (see also Jackson, 2003, 205). The prosecution also uses the recording to reduce or change the charges to avoid an aborted trial. Many barristers mentioned the advantage of knowing the substance of the evidence for trial preparation.

However, some were concerned that a separate pre-recording hearing doubles up on some resources for the trial, preferring trial alone if one could be arranged equally speedily. These comments, however, were made in the context of the recent remarkable reductions in delays, so that equidistant fixtures were a realistic possibility. Defence counsel also complained that legal aid does not cover the extra preparation required.

Nonetheless, crucially, neither counsel nor judges saw pre-recording as endangering the defendant‘s right to a fair trial.

The essential attitude amongst barristers appears to be that pre-recording is nothing to be concerned about. As Judge Jackson commented:

―[w]hatever the original preconceptions . . . the provisions have been accepted without difficulty by all [T]here is no pressure to go back to the old regime‖ (Jackson, 2003, 205).

Another interesting but not overly successful project is the 2003 pilot specialist court for sex offences which was established in West Sydney at Parramatta containing a CCTV suite capable of linking with other off-site CCTV facilities, enabling witnesses to testify long-distance. The court is staffed by specially-trained judges and prosecutors.249

While NSW‘s legislature still prefers the specialist court to other reform options, Cashmore and Trimboli‘s evaluation in 2005 indicates that the pilot was

unsuccessful. Technological difficulties have dogged the court. Many courts in NSW apparently lack CCTV facilities or screens. Meanwhile the performance of barristers and the judiciary has been variable to poor. Case management has largely failed except where a few particular judges were enthusiastic about it. Notably, specialist training has failed to improve standards of cross-examination or to increase the extent of judicial control. Cashmore and Trimboli found that regardless of specialist training, lawyers and judges generally continued to underestimate the difficulties that testifying poses for children and overestimate their own competence as communicators (Cashmore & Trimboli, 2005).

Finally, the extension of the right to use the remote facility to vulnerable adult witnesses has resulted in a logjam in which children have great difficulty accessing the facilities (Cashmore & Trimboli, 2005; Eastwood & Patton, 2002, 120).

Interestingly however, NSW apparently has a reasonably robust case management system and shorter delays than comparable systems (e.g., New Zealand or England and Wales).250

While the NSW specialist courts have not achieved well, and in particular specialist training for lawyers and judges has not succeeded, the Western Australia system of pre-recording has been extremely successful for both children and professionals (Australian Federal Police & Prosecutions, 2005, 145; Cashmore & Trimboli, 2005; Eastwood & Patton, 2002). Pre-recording is credited with having reduced children‘s stress, improved the quality of their evidence and streamlined the trial, whilst also reducing aborted or unnecessary trials. Further it presents no particular practical difficulties for counsel or judiciary and does not harm the interests of the defence.

One of the clearest signs of its success is the adoption of the system in several other Australian states. Queensland adopted it in 2003, Northern Territory in 2004 and South Australia will follow suit shortly.251

However, investment in technology and staff training are crucial for a successful pre- recording system (Australian Federal Police & Prosecutions, 2005, 131, 148; Eastwood & Patton, 2002). Moreover, the Queensland experience also appears to suggest that the best acceptance of the system occurs where the legislation strongly encourages its use. When Queensland first introduced provision for pre-recording it was largely ignored until the provision was changed so as to be the usual option.

Critics of the way in which adversarial courts treat vulnerable witnesses often look towards the inquisitorial courts as a model for reform (e.g.: McKillop, 1997; The Royal Commission on Criminal Procedure, 1981, 3; Tominson, 1983). They point to the way in which inquisitorial courts rely upon a neutral investigator to investigate the witness‘ evidence, eschewing party-led examination and to the fact that complainants are afforded separate legal representation. However, inquisitorial systems have problems of their own. Commentators from inquisitorial systems also complain of delays and unfair processes. There is also the question of the extent to which inquisitorial procedures could be adopted into an adversarial system given their very different organisation and philosophy.

This section describes the French criminal trial and procedures for taking children‘s evidence as an example of an inquisitorial legal system.252

There are three levels of criminal court in France. The police tribunal deals with minor offences; the Tribunaux Correctionels covers crimes with a penalty of up to 10 years and the Cours d’Assises deals with cases where the penalty is 10 years or more. Serious sexual assaults such as rape are matters for the Cours d’Assises (McKillop, 1997, 758). 253

Serious criminal investigations are conducted by neutral investigating judges in camera. Child witnesses are questioned during the pre-trial phase in chambers by specialist interviewers and the investigating juge, but rarely testify and are never cross-examined. The investigating judge produces a dossier of evidence for both sides which then forms the bulk of the evidence at the trial. A different judge conducts the trial, leading the inquiry and calling and questioning witnesses. While lower courts do not use juries, the highest criminal court, the Cours d’Assises, uses a jury of nine sitting with three judges.

The defendant will be questioned and the complainant can become a party to the proceedings with full representation.

The remainder of this section considers the crucial elements of the pre-trial process and the trial itself and the position of the defendant and the child witness.

In the adversarial system the trial is the point at which information crystallises into evidence but in an inquisitorial system the pre-trial investigation is the ―crucial determinative phase‖ (McKillop, 1997, 565).

When a serious crime is committed, the specialist police judiciare undertakes an initial investigation and once they have sufficient information for a charge, the file goes to the local public prosecutor, who elects whether to have the police continue the investigation under the prosecutor‘s supervision or, in the most serious cases, to notify a juge d‘instruction (McKillop, 1997, 534). Only a tiny percentage of cases are led by a juge (e.g., in 2009 Lille‘s public prosecutor reported 4%)254 (Spencer in Delmas-Marty & Spencer, 2005, 19-20). Most are led by the prosecutor, whose role is wider than that in adversarial systems (Lemonde in Henderson, 2001).255

The juge d‘instruction and the public prosecutor are neither police nor advocates but professional judges and as such have a neutral role. French judges are a separate profession to advocates, selected from law graduates and specially trained. They act both as trial judges and investigators at different points in their careers (although never in the same trial).

The juge undertakes a thorough investigation, generally longer and more meticulous than those in adversarial jurisdictions (Delmas-Marty & Spencer, 2005, 22-3, 30-1; Dervieux, 2005, 241-2), using the police as a workforce (McKillop, 1997, 567). The juge conducts ―numerous interrogations of the accused, commissioning the various experts to report and arranging . . . the reenactment and confrontation . . . and inquir[ing] into . . . the accused‘s personnalite‖ (Dervieux, 2005, 252-3; McKillop, 1997, 567).

Following such an investigation, cases which come to court are generally very strong. Whereas the high level of convictions in inquisitorial trials (McKillop, 1997, 565)256 is often seen as evidence of unfairness to the defence, in fact weak cases are likely to have been weeded out during the investigation (Delmas-Marty & Spencer, 2005, 22-3).

The investigation is closed to the public but the defence and prosecution are entitled to all reports and documentation and can request additional inquiries, such as searches of premises, interviews or wiretapping. The juge d‘instruction can refuse such requests but must give reasons and the decision can be appealed, although this is rare (Spencer and Lemonde in Henderson, 2000, 36; McKillop, 1997, 546).

Witnesses give their evidence to the juge first. Interviews take place in chambers attended by the juge d‘instruction and both counsel. Witnesses‘ procès-verbaux (evidence statements) are recorded in writing (McKillop, 1997, 394-5; Spencer & Flin, 1993) but most interviews, especially children‘s interviews, are now recorded on DVD (Dervieux, 2005, 252-3, 258).257

All evidence including witness statements, expert reports and so on are then entered into the evidence dossier which is the final product of the investigation (Spencer in Henderson, 2001, 42). The dossier goes to the prosecutor and thence to the court.

The dossier is one of the most crucial elements of the French trial (Delmas-Marty & Spencer, 2005, 22; McKillop, 1997, 582). It holds the interview notes, expert reports, photographs of the scene etc and records of the mechanics of the investigation including correspondence orders or directions (McKillop, 1997, 565).

Whereas in the adversarial trial nothing is evidence until it is stated in court (Delmas-Marty & Spencer, 2005, 22; Henderson, 2001),258 information in the dossier exists as evidence in and of itself, to the extent that if a witness fails to appear the court can usually rely on their written evidence alone (Henderson, 2000, 44; McKillop, 1997, 566).

As a rule, only evidence in the dossier is adduced at trial. Witnesses are essentially called to confirm their procès-verbaux (McKillop, 1997, 566). Although the judge can stop the trial for further investigations, additional material is rarely accepted (Lemonde in Henderson, 2001, 35).259

Hearsay evidence (Lemonde in Henderson, 2001, 44), previous convictions260 and acquittals, similar fact evidence and evidence of previous good character are freely available to the court (Henderson, 2001, 46-47; McKillop, 1997, 579-82).

The dossier is available to the defence during the investigation and before trial (Dervieux, 2005, 265-6; McKillop, 1997, 546; Spencer & Flin, 1993). However, there is no obligation to disclose unused material, particularly if gathered by the police (Lemonde in Henderson, 2001, 32). If relevant material was undisclosed, it would not invalidate the trial but would lead to punishment of the non-disclosing party (Lemonde and Van den Wyngaret in Henderson, 2001, 32-34).261

At trial, the judge actively directs the inquiry according to the judge‘s statutory duty to search for the truth (Delmas-Marty & Spencer, 2005, 26) and the duty to search for material exculpating the defendant (Cordon et al., 2003, 184; Lemonde in Henderson, 2000, 32; Spencer, 1998). Accordingly, a guilty plea does not obviate the need for investigation or trial.262

At trial the witness first accepts their procès-verbaux after which the judge leads the examination (including of the defendant). After the judge has finished, he or she asks counsel, the parties and the jury whether they have questions (McKillop, 1997, 551).263 Counsel are generally restricted to submissions and suggesting lines of questioning for the judge to put to the witness but are increasingly able to put questions directly (McKillop, 1997, 551-56). Both the prosecutor and defendant can require extra witnesses be called. The defendant has a right to require an unlimited number of extra witnesses be called (but only the first five are automatically at the state‘s expense) (Lemonde in Henderson, 2001, 44).

The defendant continues to play a central role during the trial. The judge questions the defendant repeatedly as the evidence is adduced. If they elect to answer, defendants must do so personally and cannot answer via counsel.

Defendants have a right to silence, as of 2000, and must be cautioned before police questioning (Lemonde in Henderson, 2001, 37), but defendants come under serious pressure to answer questions before and at trial and the court will draw adverse inferences from silence (McKillop, 1997, 577). However, there is a strong presumption of innocence; defendants are represented and have adequate opportunities for challenge and confrontation (Spencer, 2005).

During the investigation the juge d‘instruction, usually although not mandatorily, refers child witnesses to specialist interviewers (usually specialist police officers).264 French courts frequently refer witnesses for psychological evaluation (McKillop, 1997), and doubts about a child‘s credibility are usually resolved by an expert (Spencer & Flin 1993, 394; Spencer 1998; Condon, Goodman and Anderson, 185). If the case proceeds, the juge or prosecutor also interviews the child.265 In larger centres, this person may be a specialist in cases involving children (Dervieux, 2005, 259, n.30; Spencer & Flin, 1993, 394).266

At court there is no competency test or corroboration requirement for children, following the general assumption in continental Europe that everyone is competent (Spencer & Flin, 1993, 402). Children under 16 testify unsworn, but so do many crucial and competent adult witnesses (McKillop, 1997, 556) and this has no impact on the weight given to their evidence (Art.108 Code of Criminal Procedure) (Spencer & Flin, 1993, 401).

However, court is usually regarded as inappropriate for children. Although some children over 7 occasionally appear at the Cours d‘Assises, generally their evidence is given via their procès-verbaux and occasionally via the interview DVD267 (Spencer & Flin, 1993, 394). Judges can dispense with children‘s testimony if there is concern for their welfare.268

Few supplementary questions and no cross-examination are allowed. Lower courts send children back to the juge for re-examination while Cours d‘Assises judges will question children directly.269 If a child testifies, they may use CCTV if the court believes it necessary for the child‘s welfare but this is unusual.270

Complainants can become parties to the proceedings (or partie civile) either by initiating proceedings or by joining the public prosecution and may apply for compensation (Lemonde in Dervieux, 2005, 226; Henderson, 2001, 49). Such complainants will have separate legal representation, for which legal aid is available (Lemonde in Henderson, 2001, 48). The complainant‘s counsel protects the complainant from harassment (seen by the French as a great advantage in sexual assault cases) (Lemonde in Henderson, 2001, 48) and has similar powers to other counsel regarding witness examination and submissions (Spencer & Flin, 1993, 404) except for the power to make submissions on sentence (Lemonde in Henderson, 2001, 48).

The French system offers significant advantages to child witnesses: their evidence is taken early by a neutral specialist and they are then largely spared further involvement. If they appear at trial there is no hostile cross-examination. Complainants also have the right to become parties and have state-funded representation to protect their interests. Nonetheless, defendants can challenge the evidence and participate in the investigation in accord with the right to a fair trial. The system‘s advantages for children are reflected in the fact that the French are far

less concerned for child witnesses than we, as demonstrated by the fact that they have far fewer protections for children in court (CCTV only being introduced in 2004 and still rarely used).

However, some prosecutors fear that the French public is skeptical of child witnesses in familial abuse cases and so minimise the charges in order to avoid a lay jury. Thus although French procedures are less traumatic, children‘s access to justice may still be restricted in practice.

Commentators, most recently and persuasively Spencer et al., make the point that the adversarial and inquisitorial systems share the same underlying principles to the point that it is very hard to define exactly where their differences lie (Delmas-Marty & Spencer, 2005). However, before any borrowing could be contemplated, it needs to be recognised that there are still significant practical and theoretic differences between the systems. A judge-led inquiry would require a fundamental shift from the current dialectic system of party control over the case and the evidence. Moreover, inquisitorial procedures are underpinned by an infrastructure which we do not have. For example, the French judiciary is a professional bureaucracy. Judges apply for the judiciary out of law school and are trained separately from advocates (Damaska, 1986; McKillop, 1997, 582). Without such a shift in the balance of responsibility and power at trial and the necessary practical resources and organisation, grafting inquisitorial procedures onto adversarial stock is impractical at least.

There are also problems associated with the lengthy pre-trial investigation, which contributes to long delays before trial, made more serious by the fact that defendants are often kept on remand (Delmas-Marty & Spencer, 2005, 3). Arguably, it is preferable to take longer and be thorough—within reason—than to risk miscarriage of justice which, in adversarial systems, is often attributable to inadequate pre-trial investigation (Delmas-Marty & Spencer, 2005, 33-4). However, those within inquisitorial systems are not always convinced. The juge d’instruction is a controversial position and some inquisitorial systems (i.e., Germany) have abolished it (Delmas-Marty & Spencer, 2005, 11-12). French commentators also make spasmodic calls for the juge’s removal (Delmas-Marty & Spencer, 2005, 11-12) and even for adversarial examination—although not as much as some adversarial commentators have claimed (Spencer & Flin, 1993). Italy switched completely to an adversarial system in 1988, although they recoiled from allowing children to be cross-examined and retained judicial examination for minors (Art.498(a) Italian Code of Criminal Procedure 1988) (Spencer & Flin, 1993, 400).

The inquisitorial model of justice is therefore of interest as a model for future reform in the adversarial system only so long as its differences are borne in mind.

Norway is of interest to this study for three reasons. First, it provides for children‘s evidence to be taken before the trial by specialist interviewers. Second, it provides state-funded counsel to complainants in alleging sexual and violent offences. Third, it is an example of a legal system which mixes elements of the inquisitorial and adversarial, in particular still providing protections for the defence and opportunities for challenge which are recognisable within an adversarial framework.

Whilst Norway‘s relatively informal criminal justice system is quite distinct from most strongly inquisitorial systems, its trial process in particular contains significant inquisitorial elements.

The trial is led by professional judges (Brienen & Hoegan, 2000, 724)271 who are not drawn from amongst the advocates but become part of a parallel judicial profession on graduation (Brienen & Hoegan, 2000, 725).272 The Code of Criminal Procedure (―CCP‖) lays a positive duty on judges to discover the truth and to ―ensur[e] the case is fully clarified‖ (s63 and s294 CCP) (Brienen & Hoegan, 2000, 723; Cordon et al., 2003, 184). To this end, judges have the power to question witnesses freely and to determine who examines them and how, to adjourn proceedings for further investigations, and to call new evidence. Further, like an inquisitorial system, Norway allows evidence into the court relatively unrestrainedly (Bygrave, 1997; Melinder, Goodman, Elertsen, & Magnussen, 2004, 350).

Judicial control is such that although there are methods of expediting cases where the defendant pleads guilty (Brienen & Hoegan, 2000, 724), the parties do not control the charge or closure of the case and there is no plea-bargaining. Judges can ignore guilty pleas and reject the prosecutor‘s decisions as to the scope of the indictment (Brienen & Hoegan, 2000, 722).

However, Norway‘s criminal justice system is similar to the adversarial model in the pre-trial stages in that judges cannot initiate proceedings and they have no substantive role in the investigation. Norway does not have a juge d’instruction or examining magistrate, instead the police are in charge of the investigation and prosecutors make the decision to prosecute (Brienen & Hoegan, 2000, 722-3).

Moreover, as in all adversarial systems, Norway has a strong preference for oral evidence and examination, although many inquisitorial systems—for example, Germany—share this too (Delmas-Marty & Spencer, 2005). Most witnesses testify

live and on oath and are examined by counsel as well as by the judge. Although the trial judge has a wide discretion as to how the witness examination is conducted, both counsel have rights to attend pre-trial and trial questioning and frequently examine witnesses. However, apparently there is no tradition of aggressive, adversarial cross-examination in Norway and even the judge-led direct examination is relatively informal in terms of the evidence rules and procedure, with judges able to ask witnesses to relate events in their own words (Brienen & Hoegan, 2000, 746).

Unlike in fully inquisitorial systems, complainants in Norway do not ordinarily become parties to proceedings unless they bring a claim for compensation or a private prosecution (Brienen & Hoegan, 2000, 731). They generally have the same status as other witnesses, testifying under oath (Brienen & Hoegan, 2000, 731, 734). Adult witnesses must testify live and that means at both the first instance trial and at any appeal and re-hearing (Brienen & Hoegan, 2000, 735), although the trial can be closed to the public and the defendant can be required to leave the court while the complainant testifies (Brienen & Hoegan, 2000, 746-7).

However, although they cannot become parties, since 1981 complainants of sexual assault and some violent offences are entitled to separate representation on legal aid (Chapter 9a CCP) (Brienen & Hoegan, 2000, 72, 730, 747). The complainant‘s lawyer has ―the task of . . . guard[ing] his client‘s interests in connection with the investigation and the main proceedings‖ (Brienen & Hoegan, 2000, 735). The lawyer must be ―notified of and has the right to be present during‖ the complainant‘s police interview and at both the ―preliminary investigations‖ and ―main proceedings‖ stages of the trial (Brienen & Hoegan, 2000, 735). The lawyer has several roles. The main role appears to be to support and guide the complainant through the pre-trial stages and to prepare him or her for court, including, inter alia, accompanying the complainant at the police interview, notifying the complainant of progress in the case, contacting appropriate support agencies and conducting court visits, introducing the complainant to the other counsel and briefing the complainant as to the likely questions (Brienen & Hoegan, 2000, 732, 736-37). At trial, the lawyer‘s role is to protect the complainant during examination. Accordingly the lawyer may opt to leave after the complainant has testified although able to remain for all the evidence (Brienen & Hoegan, 2000, 732, 737). During the complainant‘s examination the lawyer may object to irrelevant (Brienen & Hoegan, 2000, 735), ―abusive or unnecessary questions‖ (Brienen & Hoegan, 2000, 737), or to those ―posed in an improper way‖ (Brienen & Hoegan, 2000, 735) and may put additional questions to the complainant at the conclusion of the other participants‘ questioning. The lawyer may also make submissions on any procedural question concerning the complainant (22 European Criminal Justice Systems, 735). However, there have been concerns raised that the lawyers assigned to complainants tend to be inexperienced and possibly insufficiently confident to protect the complainant adequately (Brienen & Hoegan, 2000, 737).

The Norwegian system for child witnesses dates back to 1926 after a campaign by a women‘s organisation concerned by the possibility of trauma to child witnesses (Myklebust, 2005, 142-72; Spencer & Flin, 1993, 399).273

In Norway all children under 16 years in child sexual abuse trials and some children in other cases, especially cases of domestic violence274 and mentally disabled adult witnesses, testify by a pre-trial deposition (Spencer & Flin, 1993, 398) in the Field Investigative Interview of Children (or ―FIIC‖), an ―investigative interview under judicial supervision.‖275 The child‘s examination can take place quite astonishingly early, potentially even before the defendant has been named. If so a lawyer will still be present to represent his or her interests.276

Children‘s evidence is taken before a magistrate with both counsel attending. Originally the magistrate would have undertaken the questioning personally in chambers with counsel (but not the defendant) present (Cordon et al., 2003, 184-5). In the late 1990s interviewing was generally delegated to clinical psychologists (magistrates can obtain help from a range of people including psychologists or social workers) (Spencer & Flin, 1993, 399). This was however deemed unsuccessful (Brienen & Hoegan, 2000, 746) and nowadays the responsibility is delegated to a police officer specialising in the forensic interviewing of children (Brienen & Hoegan, 2000, 746).277 Following public outrage over the notorious Bjugn trial in 1992 in which police and forensic interviewers were castigated for poor interview practice, Norway has put considerable effort into improving its interviewing processes, although efforts to improve it further continue.278

The local police prosecutor contacts the local court to advise an interview is required and the court then allocates a judge to supervise. The interviewer prepares a plan for the interview and forwards that plan to the judge. The interviewer then arranges the interview time. Interviews usually take place in specialist video interview rooms (Spencer & Flin, 1993, 399). The judge and counsel observe from an adjoining room, although the defendant has no right to be present.279 For example:

―[T]he Oslo police station has a special room with one-way mirrors where the children can sit and play during questioning. The session is observed from another room though the mirrors by, among others, the defendant‘s lawyer, the prosecutor, and the victim‘s lawyer.‖ (Brienen & Hoegan, 2000, 746)

Many rooms now use video link rather than one-way glass.280 The child is not told that there will be observers.281

There is scope for defence and prosecutorial contribution to questioning: interviewers must consult both lawyers and the judge during the interview.282 While counsel cannot question or cross-examine the witness directly, they can challenge the evidence and put lines of inquiry through the interviewer (Cordon et al., 2003, 189).283

The interviewer first conducts an interview according to his or her professional judgment and when he or she considers it is complete, the interviewer takes a break to consult counsel and the judge, leaving the camera running.

The judge gives both parties the opportunity to suggest topics or any contradictions in the evidence and so on they want investigated. The interviewer returns to the interview room to address these and then consults the observers when he or she considers the task complete. This process continues until the judge and counsel are satisfied.284

The style of examination is significantly different to cross-examination, as is suggested by comments by one Nordic commentator who was ―appalled‖ by the treatment of children in adversarial trials and found cross-examination

―disrespectful‖ of the child and of the court also, making ―a mockery out of something much too serious‖ to be treated thus:

―Such a line of questioning [in cross-examination] would never be allowed in Nordic countries. Certainly not with a child witness but not with any witness for that matter. This has to do with our principle about trying to find the truth. Intimidating a witness this way is assumed to be counter- productive to that search.‖ (Smith cited in Cordon et al., 2003, 189)

It is notable that this commentator explains the appearance of cross-examination in adversarial trials as the result of differences in the aim of the investigation: their system searches for the truth. Cross-examination he regards as antithetical to that aim.

Interviews are videotaped and a written transcript is made. The record of the interview becomes both the child‘s evidence in place of live testimony (Spencer & Flin, 1993, 399) and it is also a record for the purpose of the police investigation.286 The child does not appear at the trial: instead the video is played and the members of the court are given copies of the transcript.287

The FIIC must by law be convened within two weeks of the report to the police and apparently, although difficult to achieve, the Norwegians are largely able to keep to this deadline. In order to facilitate the necessary speed, the government is currently piloting ―safe-houses‖ which are inter-agency centres with suites for the FIIC, staffed by police, medical and social services so that all the child‘s legal duties and welfare needs can be met under the one roof.288

Norway‘s process for interviewing children is impressive. Tellingly, there do not appear to be concerns about child witnesses‘ welfare during criminal investigations, although in the early 1990s after a notorious trial over sexual abuse allegations Norway went through a similar period of concern about the competence and credibility of children as witnesses and the potential for contamination from poor interviewing as occurred in many adversarial jurisdictions at around the same time.289

Norway manages to circumvent most of the problems which have been felt in mainly adversarial countries, namely, avoiding delay by taking children‘s evidence early and avoiding the problems associated with cross-examination by using highly-trained specialist interviewers and restricting counsel‘s involvement. Further, holding the examination in a special suite and having the legal professionals observe from outside avoids the problems of the unfamiliar, inappropriate courtroom environment and some of the embarrassment of the presence of strangers. Child complainants in some cases also have access to state-funded legal representation to guard their interests.

Simultaneously the Norwegian process manages to avoid at least some of the objections raised by adversarial advocates to inquisitorial processes, especially the unfairness of preventing the defence from challenging the evidence fully. In Norway the defence counsel is present together with judge (and prosecutor) to oversee the safety of the interview. The judge is not a juge d’instruction; he or she has no part in the investigation generally but is called in to run the examination of a child witness in order to generate safe evidence for trial. Further, although the examination is

delegated to a specialist, the interviewer consults with the defence during the interview to ensure the issues the defence wants covered are investigated. Finally, although some complainants have rights to representation during their testimony, they cannot become parties. Norway retains the essential two-party nature of the adversarial proceeding.

However, it does seem unfortunate for the child‘s rights to privacy and to self- determination that the child is unaware of the fact that the FIIC is observed by others.

The balance between the adversarial and inquisitorial in Norway‘s process makes it particularly interesting to adversarial legal systems. It has to be remembered that Norway, although closer to the adversarial process than other, fully inquisitorial systems, is still very different. Aspects such as the judicial control of the trial and the free admission of evidence challenge the adversarial system, as would the holding of an examination as early as in Norway. It is hard, as a practical issue, to imagine an adversarial system being able to convene the full coterie of professionals for an examination as early as they manage in Norway. More seriously, such a hearing would raise serious issues as to whether defence counsel could be sufficiently prepared, particularly with regard to the current emphasis on the need to obtain disclosure before constructing one‘s case theory and even more so where the defendant is not yet identified. Counsel and court would have to come to seriously different understandings as to what is sufficient to test the evidence.

However, it may be worth considering whether there is scope to accommodate a similar procedure at the pre-trial stages. This is still a process which might potentially be imagined as part of an adversarial investigation.

This section now considers the systems in use in the above six jurisdictions thematically rather than country-by-country. The main reforms under discussion are pre-recording and the intermediary system but the discussion also considers separate legal representation for complainants.

Pre-recording is the practice of recording a witness‘ entire evidence, including defence challenges to it, in advance of trial. Its major advantages, as seen by users, are summarised in Table 21.

Pre-recording is a truly cross-over reform, successful in inquisitorial and adversarial systems. Pre-recording is the norm within inquisitorial jurisdictions such as France and is also practised in the semi-inquisitorial state of Norway and in two adversarial jurisdictions, namely Australia and Israel, while England and Wales have technical provision for it. However, whereas Australia (and England) uses lawyers to cross- examine children but specialist interviewers for the direct evidence, Israel, France and Norway employ specialists to conduct the entire examination as they see fit, subject to judicial and party supervision. The advantage of the latter is that children are never exposed to hostile cross-examination and the risk of inappropriate language is much reduced.

All of the systems which pre-record children‘s evidence experience it as a major benefit to the children themselves and (with the exception of Israel) to the court also.

There seems however nothing antithetical to pre-recording in the adversarial process, as is demonstrated by its successful use in Western Australia. In fact, pre- recording would seem to be the best and most efficient method for reducing delays and expediting children‘s exit from the system seen so far.

In adversarial jurisdictions, intermediaries are specialists in communicating with children (or with witnesses with specific communication difficulties) in an adversarial court and assist the examiner to speak with and understand the child witness. They are essentially communication facilitators or, in extreme cases, translators. They must not alter the substance or sequence of questions or alter or comment on the substance of replies. They have no role in assessing the witness‘ competence.290

Intermediaries are used in South Africa and England and, to a limited extent, in the USA. There is also an unused legislative provision for them in Western Australia.

Other jurisdictions, including most European inquisitorial countries, France, Israel and Norway, use specialists to interview children under judicial and counsel supervision rather than merely to assist counsel‘s examination.

Users believe that intermediaries are of great assistance to both child witnesses and to the court, as summarised in Table 22.

Two other possibilities would be to offer witnesses state-funded legal representation to protect their welfare during trial or to hear cases involving child complainants in specialist courts. Separate legal representation is attractive since neither the party calling the witness nor the judge at the present time can be fully relied upon to control opposing counsel from pursuing unfair lines of cross-examination, both because they are unlikely to notice unfairness and because they feel constrained by concerns that they might be suspected of covering up weaknesses in the witness‘ evidence. Counsel for the witness need have no such qualms.

France, as inquisitorial systems generally do, allows complainants separate representation as does Norway in sexual and violent cases. Israel‘s youth interrogator, not a lawyer, is also responsible for protecting the child witness‘ welfare at trial (David, 1990, 106).

Adversarial systems would see separate counsel for witnesses as problematic and many would argue that counsel for a witness would unbalance the contest and lend support to the party calling the witness. However, even within the adversarial context there are already occasions in which the court will appoint additional non- party counsel. The court can allow interested third parties to comment upon the case as Amicus Curiae (as in South Africa‘s S v Mokoena) and the New Zealand Family Court not only has a separate lawyer for child but can appoint a ―Lawyer to Assist‖, to pursue issues the parties are ignoring but which the court wishes to see represented.

Nonetheless, there are limits to what counsel could achieve. Counsel could not shorten the delays before or at trial although they could make submissions calling for an early hearing. At trial the process of objecting in cross-examination is blunt and it is difficult to imagine counsel intervening as often as research suggests problems would arise. Most lawyers are unlikely to fully appreciate that many problems had arisen. More impact could be had by judges taking greater control of case management and of cross-examination.

Specialist courts are in operation in a number of adversarial jurisdictions and this study has touched some of them: the New South Wales pilot specialist court has been evaluated and found wanting. Of particular interest is the finding that training the advocates and judiciary had not had much impact. However, training the judiciary does appear to have had an effect in England and Wales where cases involving children and sexual assault are restricted to accredited, trained judges.

This review of jurisdictions would suggest that the systems which hold most promise are pre-recording and intermediaries. The review has also highlighted the different models of these functions across the six jurisdictions surveyed.

Western Australia has successfully integrated pre-recording into an adversarial regime but has not combined this feature with an intermediary system. Britain and South Africa do not use pre-recording but now have established intermediary systems; however, these intermediaries are somewhat restricted in their ability to ensure that children are questioned according to best practice. The Israeli system of youth interrogators is robust, but contains elements that restrict both the defendant‘s rights and children‘s access to justice. The French system, while offering significant advantages to child witnesses, is deeply embedded within an inquisitorial system that differs fundamentally from our own.

While introducing significant elements of the inquisitorial process into an adversarial process is impractical, there is scope for importing those elements which are not necessarily tied to that regime. The Norwegians have managed to integrate both pre- recording and specialist child examiners, who have wide scope to interview children according to best practice, within a trial process which contains elements very similar to ours. Indeed, integration of these two functions would offer significant advantages to children and to the inquiry, while simultaneously addressing delay and inappropriate cross-examination. It is therefore to Western Australian and Norway that New Zealand might look to ascertain what more could be done to facilitate children‘s participation in the criminal justice system. The potential of these systems for the New Zealand context is considered in more detail in the following chapter.

New Zealand has come a long way in accommodating the needs of child witnesses in the criminal courts since concerns about their treatment were documented in the 1960s. Following the legislative and procedural changes of the 1980s and subsequently, there are now special measures available aimed at making it easier for children to testify, thereby enhancing the quality of their evidence (see Chapter 4). New Zealand‘s specialised forensic interviewing service, comprising police and statutory social workers, are jointly trained in best practice for communicating with children; interviews are expected to cover both evidential and care and protection issues. The frameworks necessary for enabling children to give evidence in court are largely in place.

As evidenced over time, law and policy-making do not guarantee effective and consistent implementation on the ground. This is borne out in the current study. The criminal justice system is still finding it difficult to accommodate child witnesses to the extent possible under current legislative and procedural frameworks (see Chapters 2, 3 and 4). For example, one third of eligible child witnesses did not access the Ministry of Justice‘s Court Education for Young Witnesses programme (see Chapter 2). Fourteen percent of the child witnesses in the 2008-2009 sample testified without using any alternative modes of evidence; in only 53% of cases involving 13 to 17 year old witnesses was a videotaped forensic interview used at court as the child‘s evidence-in-chief (see Chapter 2).

Significant regional differences in implementing existing law have also emerged, in particular in relation to children‘s access to alternative modes of evidence (see Chapter 4). This study did not set out to determine why practitioners did—or did not—facilitate children‘s access to these modes. Hence the reasons underlying the regional differences in practice are not known. The extent of such variation suggests that it would be worth investigating these issues so that we might determine how to more effectively implement existing legislation and address any barriers in some parts of the country. For example, some prosecutors may avoid CCTV in the belief that if children are too composed, their testimony might be seen as unconvincing in the jury‘s eyes. A barrier such as this might be addressed by a judicial warning to the jury to not draw an adverse inference if alternative modes of evidence achieve their aims of reducing the stress of testifying.

But the point remains that, in enacting the Evidence Act 2006, Parliament intended that alternative modes of evidence be available to all child witnesses. Changing legislation to a presumption in favour of alternative modes of evidence (as in England and Wales) would likely improve children‘s access to them. Further, a presumption in favour of using videorecorded forensic interviews and CCTV or live audiovisual links in particular is warranted to facilitate all children‘s testimony, not just younger children (see Chapter 4). This could increase the efficiency of the system by reducing the amount of time spent on the application process (as well as the possibility of recording CCTV evidence for use at retrials).

A factor which may have contributed to the current situation facing child witnesses is the lack of systematic monitoring. At the time of writing, child complainants and witnesses are still not visible in the data monitoring processes of New Zealand‘s criminal court system, although progress in this regard has been made in at least some courts, such as the Auckland District Court. Children‘s invisibility in the national data system makes it harder to see the challenges and patterns over time and harder for Parliament to monitor progress towards fulfilling the intent of the law.

The Ministry of Justice could consistently track and monitor child witnesses through the criminal justice system in all regions, and ascertain the reasons for the problems outlined in this report (including regional disparities). This would put a spotlight on some of the scheduling difficulties outlined in Chapter 2—a key first step towards resolving them. The research team‘s interactions with practitioners suggest that compliance with an auditing system would be high as there is significant interest in improving the way the system treats children. The establishment of an appropriately designed data system, including Police and Justice data, prior to enacting any further legislative change could also give a baseline from which any additional measures introduced could be evaluated.

We would add that security concerns also warrant the careful tracking of videorecorded forensic interviews that are removed from police premises as a required part of case management.291 The State could be more confident that it had protected these recorded interviews if s106(a) of the Evidence Act 2006 was amended so that videorecorded forensic interviews were not allowed to be removed from police premises (save for court purposes alone).

Moreover, this research has identified two particularly persistent and intractable problems that impact on children‘s ability to give their best evidence: firstly, children are still being asked confusing age-inappropriate questions, particularly in cross- examination (see Chapter 3); and, secondly, despite judicial directions to prioritise child witness cases since 1992, cases involving children are still subject to long delays (see Chapter 2). Both substantive issues were named in research in the

1990s (Davies, 1998) and the report of the Courts Consultative Committee Working Party on Child Witnesses in 1996, and are discussed in more detail in the following sections.

In 1993, child complainants under 17 years of age waited about eight months for their cases to be processed through the courts (Lash, 1995). Children in the current sample, some 15 years later, waited almost twice as long. These children‘s cases took on average 15 months to process, with adjournments extending the process. Pressure on the courts overall has increased during this time, partly because of a higher police apprehension rate and more defended and complex cases (New Zealand Law Commission, 2004; Smith, 2007). The joint New Zealand Law Commission/Ministry of Justice Criminal Procedure (Simplification) Project, established in 2007, has as one of its key aims the reduction of court delays: this is an opportune time to consider innovative practices which might work in concert to reduce delays in processing child witness cases.

One clear possibility is the pre-recording of a child‘s entire evidence, using the Western Australian two-tape method: that is, the pre-recorded forensic interview on one tape and cross- and re-examination on another (see Chapter 5 and also Appendix II for relevant Western Australian legislation on pre-recording). The pre- recording of cross- and re-examination could be conducted within a short time of committal via CCTV. This would expedite the process and largely keep children out of the trial itself, unless the court deemed it necessary to communicate with a child on significant new evidence that emerged after the evidence was pre-recorded.

Some will argue that pre-recording will be unfair to the party cross-examining the child as there will be insufficient time to obtain disclosure and to construct a case theory. These are essentially logistical problems rather than principled objections (see Appendix I). Counsel might have initial difficulties getting used to a new timetable but it should be noted that Western Australia had little difficulty adjusting (see Chapter 5). The improved quality of evidence should prove a significant advantage to the justice system and the savings in time and money of being able to weed out weak accusations would benefit everyone.

Logistical issues can be overcome. Pre-recording evidence could become the routine way for children to give evidence, whether witnesses or complainants, regardless of the nature of the offence. This is arguable under s105 of the Evidence Act 2006 and consistent with the Courts (Remote Participation) Bill currently before Parliament which aims to increase the use of audiovisual links, including for the taking of evidence in criminal proceedings. The multi-agency centres for child abuse investigations (e.g., Puawaitahi in Auckland, and emerging centres in Wellington and Manukau) add to the opportunities to take children‘s evidence from appropriate locations away from the court building.

If the Government wishes to see the pre-recording of children‘s entire evidence become routine practice, it might be preferable to amend the provisions of ss102-

107 of the Evidence Act to include pre-recording of a witness‘ entire evidence explicitly, thereby avoiding a narrow interpretation of the current provision. While this could be achieved by introducing an Evidence Amendment Bill, it may be simpler and quicker to add a further clause to this effect to the Courts (Remote Participation) Bill or other relevant legislation currently going through, or about to go through, the House.

Nearly all of the children in this sample said that they had not understood questions posed by the defence lawyer and sizeable minorities said that they had not understood prosecutors‘ questions. The analysis of transcripts found no evidence that lawyers adjusted their language to accommodate younger children‘s linguistic and cognitive competence (see Chapter 3).

Parliament has already recognised this. Section 85 of Evidence Act 2006 extends restrictions on inappropriate questions to those ―expressed in language that is too complicated for the witness to understand,‖ enjoining judges to consider, amongst other things, the witness‘ ―age and maturity . . . any physical, intellectual, psychological, or psychiatric impairment . . . and linguistic or cultural background.‖ The court therefore now has comprehensive powers to control inappropriate language, provided of course the court is equipped to recognise it as such (see Chapter 3). Herein lies part of the problem.

Assessing children‘s communicative competence requires specialist knowledge and skills (see Chapter 3).292 If the courts are to get the best evidence from children, ways can be found to increase the questioning competence of agents of the court. A range of measures, like judicial education and legal training could be implemented, to develop knowledge and skill in the legal profession. Nevertheless it is probably unreasonable to expect busy legal practitioners to become competent child questioners and keep up to date with international research and best practice in the field. It is too easy to become acculturated to courtroom jargon. Training may increase the visibility of this jargon, and lawyers‘ awareness of what they do not know about children‘s communicative competence, but it is unlikely to be significant enough to change the way that children are questioned in court and it certainly won‘t profoundly impact on the ways that children‘s testimony is challenged.

familiarity with one‘s own children means miscommunication is probably easier to detect than with a child

whom one barely knows. Nor are the consequences of miscommunication in the everyday family context anywhere near as grave as ―conversations‖ in court.

Children were frequently subjected to suggestive questioning in the courtroom (see Chapter 3). Some will argue that this is an inevitable part of cross-examination in an adversarial system. The nub of the matter is that there is now a substantive body of literature on ways to get the best possible evidence from children (see, for example,

M. E. Lamb, Y. Orbach, I. Hershkowitz, P. W. Esplin et al., 2007; M. E. Lamb, Y. Orbach, I. Hershkowitz, D. Horowitz et al., 2007; Poole & Lamb, 1998) that profoundly contradicts the general principles and practice of cross-examination (Henderson, 2000, 2002; Spencer & Flin, 1993). Therefore it is worth considering how to ensure that children‘s evidence can be transparently tested regarding matters of concern to all parties using sequences of questions posed in accordance with best interviewing practice.

Options implemented in other jurisdictions to achieve this goal are outlined in Chapter 5. They are all congruent with pre-recording testimony to reduce delays in the system. The balance between the adversarial and inquisitorial processes in Norway makes it particularly helpful for New Zealand. Norway uses a specialist forensic interviewer to access all children‘s evidence under the directions of the judge, with scope for defence and prosecution lawyers to contribute to the themes of the questions asked. The notion of specialist interviewers conducting cross- examination and re-examination, with the judge, defence and prosecution lawyers instructing the interviewer on which areas of the evidence to test or clarify, warrants further consideration. It is, after all, a variation on an existing practice, namely, the way in which children‘s evidence is currently elicited by specialist forensic interviewers then played at court as the child‘s evidence-in-chief. It is worthy of note here that when these current measures were introduced in the late 1980s, initial adverse reactions quickly dissipated. Nevertheless, this would require a shift in thinking about the nature and practice of cross-examination of children in New Zealand‘s criminal justice system.

Successive analyses of transcripts of children‘s evidence (Davies & Seymour, 1998; Chapter 3) suggest that New Zealand‘s forensic interviewing service would have the knowledge and skill to conduct interviews that include themes of concern to all parties.293 Hence, one possibility is to build on the existing forensic interviewing service and to extend its operations to include the conduct of cross-examination and re-examination as well. This may be the most cost-effective option for improving the quality of children‘s evidence in the court setting. However there would be some barriers to overcome: forensic interviewers might struggle with conceptualising a role in challenging children‘s evidence, at least initially. Any communication service would also need the trust of all stakeholders and it is unlikely in the short term that all counsel would trust the neutrality of the current forensic interviewing service to allow extension into the realms of challenging evidence.

One way of addressing the issue of trust might be to house the service within the Ministry of Justice. However, the purposes of a forensic interviewing service go beyond the realms of the criminal courts, and into child protection (see Chapter 2). Hence housing the service within the court system would run the risk of losing this important child protection lens built into the current system. Nonetheless, if the system could be designed in a way that mitigated this risk, it could be an efficient and effective use of existing knowledge and skills. Further training and guidelines would also be necessary.

We can see no reason, in theory, why the same specialist group could not provide interviewing services for both the prosecution (including the care and protection lens) and defence. However, in the New Zealand context, children might need to be forensically interviewed by different individuals than those who challenge the evidence. It may even be necessary to ensure that forensic interviews are conducted by a different specialist group altogether to that used to challenge the evidence.

Therefore another option would be the formation of a new and separate independent service to conduct cross-examination and re-examination of child witnesses. Such a service could be housed within the Ministry of Justice, possibly as part of the court interpreters service. These ―specialist child examiners‖ could be sworn as officers of the court on the same terms as interpreters and would not ordinarily be liable for examination themselves, underlining their separate neutral status. The service could comprise appropriately qualified professionals in private practice who have received extra training as specialist child examiners in court and who are contracted to the Ministry—in other words, a similar administrative arrangement to that used in the UK (see Chapter 5). The benefit of this would be neutrality which would help to develop stakeholders‘ trust of the service. However, a barrier to this scheme could be the cost of developing a new service, although the numbers of children currently involved (see Chapter 2) mean the costs should not be prohibitive. Implementing such a system would not obviate the need for training judges on children‘s language.

Table 23 outlines potential advantages and disadvantages of the two systems outlined above that could reach the goal of ensuring that all child witnesses are questioned and challenged by adults skilled at communicating with children. There could be other ways to reach this goal. Appropriate consultation with practitioners would be essential to reduce the risks of unintended negative consequences of change.

Table 23 : Advantages and disadvantages of the two "specialist child examiner" systems

Current Forensic Interviewing service (police and CYF) and establishing a separate

Trust could be developed in a new service without reference to the existing forensic interviewing service.

Evidential and care and protection concerns dealt with in the forensic interview.

Duplication of infrastructure, skills and knowledge of the forensic interviewing and ‗specialist child examiners‘ services.

Forensic Interviewing service combined with a ‗specialist child examiners‘ service within the Ministry of Justice

Potential to strengthen the current forensic interviewing service alongside the ‗specialist child

Forensic interviewers may not feel comfortable challenging children‘s evidence.

Adequate opportunity for defence counsel to observe and to direct the ‗specialist child examiner‘ to question the child further or on further issues would of course need to be a cornerstone of any such system. The questions and answers could be recorded and, together with the videorecorded forensic interview, would comprise the child's evidence, subject to rights to recall the child (see Appendix II for examples of relevant legislation from other jurisdictions which achieve this).

Legal objections to such a system would likely focus around the adequacy of testing conducted by any intermediary. Lawyers prize the traditional techniques of cross- examination for their investigative power. Some would be concerned by the severance of their personal connection to the witness, and loss of control and the element of surprise that goes with it. However, the extensive empirical evidence gathered over the last quarter century strongly suggests that lawyers are mistaken in their beliefs about cross-examination‘s efficacy as a forensic test with children (see Chapters 3 and 5). Greater evidential safety can be achieved via a questioning method that is empirically proven. If there is proper opportunity for the defence to ensure all the relevant doubts are explored in such an interview and the defence has a proper opportunity to argue the case in submission there should be no valid concerns over the adequacy of the protections for the defence. We note that the European Court of Human Rights, which hears cases from both inquisitorial and adversarial systems, has suggested that proper questioning by a specialist examiner does not violate the defendant‘s rights (see Appendix I).294

294 S.N. v Sweden (2002), Application no. 34209/96); 39 EHRR 13, [2002] CrimLR 831.

The data outlined in this report supports a goal of ensuring that all child witnesses' cross- and re-examination are routinely conducted by adults skilled at communicating with children, in an age-appropriate forum, observed by both counsel, defendant and the judge and on topics agreed by all parties involved. This would have the added advantage of ensuring that defence lawyers and prosecutors planned clear lines of questioning child witnesses ahead of time. It might also increase earlier disposal of cases either by the prosecution not pursuing charges or by the entry of more guilty pleas, thereby reducing court time.

The logistics of implementation of a system that is fairer for children would need to be worked through if such an arrangement were Parliament‘s will. While s80 of the Evidence Act 2006 already allows for communication assistance for witnesses, the law could be clarified to indicate that, given the developmental and linguistic chasm that can separate children and adults, age alone is grounds for allowing access to this measure; it could also be strengthened by a presumption that the use of specialist examiners become routine practice.

A system that integrates recorded forensic interviewing with pre-recorded cross- and re-examination by a court-appointed ‗specialist child examiner‘ is an ambitious goal. It is also the natural next goal for adversarial systems looking to further enable child witnesses to provide accurate evidence with minimal trauma. Other adversarial jurisdictions have tested and adopted elements of this system, but none have yet used them in combination, although they do discuss such a move. Integration would offer significant advantages to fact-finders and to children, simultaneously reducing delays and unsafe questioning, particularly in cross-examination.

New Zealand has a real opportunity to seize the initiative and become a world leader in integrating the best of the inquisitorial system‘s treatment of children into our adversarial system. Establishing an effective baseline set of indicators on the status of child witnesses would increase the visibility of children in criminal justice sector reforms and enable effective ongoing evaluation of any measures introduced. New Zealand and its vulnerable child witnesses deserve nothing less.

This section expands upon the discussion of the arguments for and against the two major proposals in chapter 6, namely full pre-recording and intermediaries. It is based upon a review of the published literature and on interviews with legal practitioners in Western Australia, South Africa and England referred to in Chapter 5.

The major advantage of full pre-recording is that it removes one of the main causes of stress for child witnesses, namely the long delay before trial. Pre-recording enables children to ―get on with life‖ much earlier (Australian Federal Police & Prosecutions, 2005, 133) and there is no risk of records of therapy being disclosed in court as therapy would take place effectively after trial (Eastwood & Patton, 2002, 115; Hoyano & Keenan, 2007, 640).

Pre-recording also lessens the stress of attending court. As the hearing only concerns the child there should be less inconvenience and delay on the day (Australian Federal Police & Prosecutions, 2005, 133) and less intimidation as fewer (or none) of the defendant‘s witnesses and supporters will be required (Hoyano & Keenan, 2007, 641).295 Testifying at a pretrial hearing might also be less stressful than at a trial—even via CCTV—knowing a full court and jury was listening (Australian Federal Police & Prosecutions, 2005, 133).

Further, pre-recorded evidence is available for any re-hearing, so that children would be unlikely to have to testify again, which is a cause of collapse in some re-trials (Australian Federal Police & Prosecutions, 2005, 133; Hoyano & Keenan, 2007, 641-

295 Conversely Plotnikoff and Woolfson in 2009 indicated that defendant/witness contact and intimidation is still a regular occurrence at UK trials conducted without pre-recording (Measuring Up).

The second major reason in favour of pre-recording is the benefit to the trial and to society generally of an improved quality of evidence.

Memory deteriorates over time and young children‘s memories are subject to greater deterioration than adults‘ memories (Malloy & Quas, 2009). Presently the delays between the complaint and the trial are so great that the child‘s memory is likely to have suffered significantly. The effects of time (e.g.: on peripheral detail) are a common theme in cross-examination (Davies et al., 1997; Henderson, 2003).

Pre-recording enables witnesses‘ testimony to be taken while their recollections are still relatively clear, diminishing the likelihood of inconsistencies from natural attrition (Australian Federal Police & Prosecutions, 2005, 132; Hoyano & Keenan, 2007, 638- 9; Spencer, In press (b); Spencer & Flin, 1993).

Another common concern in sexual abuse trials is that the child‘s account has been contaminated by suggestive interviewing (Davies et al., 1997; Henderson, 2003). Suggestive cross-examination questions are also the target of complaints (Spencer & Flin, 1993, 270-73). As suggestibility also appears to increase over time (Malloy & Quas, 2009), taking a child‘s evidence early should reduce the risk of contamination from either source.

The third group of arguments regarding pre-recording concern its likely effect on the jury.

Currently, the combination of long delays between video and trial and the rapid pace of child development mean the child in the video and the child testifying live often appear and act very differently. The discrepancy may mean juries find it hard to relate one to the other or evaluate their evidence. Pre-recording means that the child on the video is the same or nearer in age, developmental stage and appearance to the child they see being cross-examined (Australian Federal Police & Prosecutions, 2005, 132; Hoyano & Keenan, 2007, 641).

It is obviously beneficial to all concerned that with pre-recording inadmissible and prejudicial material is edited out of the video so that the jury are not exposed to its effects (Hoyano & Keenan, 2007, 640).

As pre-recorded evidence is also edited of the pauses for the child‘s breaks and legal argument juries have perhaps a better chance of grasping the chronology and digesting the evidence than when they are subjected to the stop-start of the live trial experience (Hoyano & Keenan, 2007, 640).

There are also some savings to be made from pre-recording. Counsel we interviewed in Western Australia found that knowing how a key witness will perform and what evidence they will give enables both sides to make better pre-trial decisions. The prosecution can withdraw or refine the charges and the defence counsel can better advise the defendant to change plea if necessary. Any withdrawal or guilty plea represents a saving for the criminal justice system and Legal Aid (Australian Federal Police & Prosecutions, 2005, 132; Hoyano & Keenan, 2007, 640).

Further, editing the video not only results in slightly shorter trials (Australian Federal Police & Prosecutions, 2005, 132; Hoyano & Keenan, 2007, 640) but trials become easier to schedule as none of the specialist equipment nor personnel are required and cases no longer require priority (Hoyano & Keenan, 2007, 640). Some may object that the expense is merely re-shuffled to the pre-trial hearing but the fact that a jury and other witnesses and officials are neither required nor kept waiting during that pre-trial is still a saving.

The major controversy surrounding all uses of video for children‘s evidence is whether it lessens the emotional impact of the testimony. The mainstream view amongst many researchers (Hoyano & Keenan, 2007, 641) and barristers (Eastwood & Patton, 2002, 120-1) is that video/CCTV monitors do blunt the effect.296

Some barristers claim that pre-recording has an even greater distancing effect than CCTV: with CCTV the jury knows the child is testifying live whereas pre-recorded evidence is ―just another TV programme‖ (c.f. Hoyano & Keenan, 2007, 644).

Conversely, the US Supreme Court argues that seeing the child on video gives greater weight and credibility to his or her evidence (Hoyano & Keenan, 2007, 643), implying the truth of the witness‘s allegations because special protection has been necessary. That this argument is accepted in many jurisdictions is suggested by the fairly universal practice of warning juries not to read anything into the use of a special measure (Hoyano & Keenan, 2007, 643). Others argue that pre-recording reduces prejudice because the jury will not realize that the child was separated from defendant whilst testifying (Hoyano & Keenan, 2007, 639).

Whatever the impact, as some witnesses can only testify via monitors we must weather the perceived disadvantage (Hoyano & Keenan, 2007, 641). However, research has not disclosed any particular impact of monitors on juries. Studies in America, Australia, England and Scotland of cases with and without CCTV have found no difference in conviction rates,297 although one found that children using CCTV

296 Personal communications with barristers in WA but c.f. Eastwood and Patton 121 regarding WA.

297 See Plotnikoff and Woolfson‘s review (2009) citing the American Bar Association, 2002; Joudo & Taylor, 2005 (adults); Davies and Noon, 1991; Murray, 1995. An unpublished study over Jan 2005- Nov 2006 by a judge at Liverpool Crown Court does suggest that the type of screen affects the jury: In a court where the

were less tearful than those testifying live (Davies & Noon, 1991), which, while excellent for the children, might support contentions that monitors benefit the defence as, in the words of one barrister ―if you‘re acting for the bloke, the worst thing to see in front of a jury is to have an eight year old girl crying‖ (Eastwood & Patton, 2002, 121; see also Henderson, 2003).

There are also objections to pre-recording. The first is that it detracts from the witness‘s appreciation of the solemnity of the criminal trial and the witness may take testifying too lightly (Hoyano & Keenan, 2007, 641). This argument applies to all special measures which soften the child‘s experience of the trial, whether it is removing wigs or using CCTV. However, while witnesses should understand the seriousness of what they are doing, the question is not merely what is necessary to render a witness suitably sober but also what is sufficient. Rather than inducing a serious and truthful frame of mind the full solemnity of a jury trial may be more likely to overwhelm a child‘s ability to think coherently at all (Spencer & Flin, 1993, 118-9).

The most serious arguments against pre-recording concern possible adverse effects on the defendant.

There are concerns that defence counsel cannot prepare cases sufficiently to be able to cross-examine the most important witness at a pre-trial hearing as all the issues cannot be known until the trial begins. These concerns are said to have discouraged the UK government from implementing the relevant section of their legislation (Spencer, In press (b); Spencer, In press). English judges and counsel we interviewed worried about delays in obtaining disclosure from prosecution and third parties (especially as disclosure rules in Britain are so extensive (Birch & Powell, 2003) and allow prosecutorial disclosure to occur so late).298 Further, some argue that witnesses in child sexual abuse cases take longer to emerge and defendants take longer to come to terms with the allegations. Therefore, they argue, it is impossible to stage a pre-trial examination until so close to the actual trial as to render the attempt pointless (Birch & Powell, 2003; c.f. Hoyano & Keenan, 2007, 642, 660; but see Spencer, In press (b)).

The argument is unconvincing: child witnesses are usually called first so that the defence always has to cross-examine them before hearing other witnesses. Further, in child sex cases the defence also generally have already seen the child‘s evidence– in-chief via the forensic interview video rather than merely a written statement and so have an advantage in preparing for cross-examination (Hoyano & Keenan, 2007, 642).

overall rate of convictions was 20%, 49% of juries watching the child on a large plasma screen convicted as opposed to 36% of those watching a small TV screen (Plotnikoff & Woolfson, 2009, 87).

As regards tardy disclosure all that is required to solve the issue is sufficient will to do so (Spencer, 2008; Spencer, In press). Both Western Australia and the USA have managed to achieve efficient disclosure and it should therefore be achievable elsewhere (Hoyano & Keenan, 2007, 639) as the English judges to whom we spoke agreed.299

In any case, although there will be exceptions, the problem is unlikely to occur very often:

―[I]n many child abuse cases the issues are very simple, the questions the defence need to put to the child are obvious from the outset and [questioning at an early pre-recording hearing] . . . would give the defence everything they need.‖ (Spencer, In press).

The speed of disclosure in New Zealand is likely to increase in the near future anyway. In 2008 the government codified the common law on disclosure, requiring the prosecution to disclose ―all relevant information, unless there is good reason to withhold it‖ (McIntosh, 2009) within 21 days of committal (ss9 & 12 Criminal Disclosure Act 2008) and the defence must disclose details of its aibi witnesses and expert evidence 14 days before trial (s22-23 Criminal Disclosure Act 2008). In a further indication of a will to improve disclosure in September 2009 the Attorney General called for judges to take a stronger stand to enforce expeditious discovery (Armstrong, 2009).300

Most of all, ―the fact that the reform would not solve every problem is no reason for rejecting a reform that would solve many of them‖ (Spencer, In press).

Another argument is that splitting the hearing into two would lead to practical difficulties arranging to have the same counsel act in both ―and the defence would not want to entrust the examination to someone else‖.301 Several English practitioners and academics raised this objection. Conversely Western Australian judges and counsel did not, although their standard jury directions include a warning that the counsel and judge may differ. It may be this is more about convention than real practical difficulty. Where counsel are accustomed to sharing cases between a team of advocates (as in New Zealand law firms) there is less likely to be any perception of risk.

299 This discounts any effect of the cultural differences between the allegedly cooperative Western Australian bar (personal communication, J. Cashmore and Western Australia barristers, and see Eastwood & Patton 2002, 61-62, 121; Hoyano & Keenan 2007, 669) and its more adversarial cousins. However, pre-recording was achieved in the US in earlier times and no one could fault their adversarial credentials.

300 See also the Law Commission‘s recommendations for reform : Striking the Balance NZLC PP51 (Wellington,

2002); Seeking Solutions: Options for Change to the New Zealand Court System: Have your Say: Part 2 NZLC PP52 (Wellington, 2002); Delivering Justice for All NZLC R85 (Wellington, 2004).

However, a similar issue of being bound by earlier counsel‘s mistakes occurs in appeals based on inadequacies in the performance of previous trial counsel. The court usually deals with such questions very robustly.

It is reassuring on one level that essentially all of these are practical problems rather than principled objections and should be able to be overcome by better case management (Spencer, 2009). However, it is important not to underestimate the difficulty of overcoming the problems created by New Zealand‘s overloaded court system. It might still be difficult to find time even for a pre-trial hearing at present.

While no barrister or judge in Western Australia or England we interviewed objected to pre-recording as a violation of the right to a fair trial, some have argued that pre- recording violates the defence right of non-disclosure, by forcing the defence to reveal their case theory, names of witnesses and lines of cross-examination (Hoyano & Keenan, 2007, 642).302

In the UK House of Lords in the early 1990s this argument stymied the introduction of video pre-recording (Spencer & Flin, 1993, 89-90). As Spencer and Flin retorted

―[t]he answer . . . should have been that the only defendant who stands to lose from disclosing his defence ahead of trial is one whose defence will not stand investigation‖ (Spencer & Flin, 1993, 90).

Moreover, the right of non-disclosure is not absolute: the defence already have to disclose expert and alibi witnesses in New Zealand under s22 & 23 of the new Criminal Disclosure Act 2008.303

Mandatory pre-recording by child defence witnesses would violate the principle that the prosecution must call its evidence before the defence (Hoyano & Keenan, 2007, 659). However, whether this would have undesirable consequences or mean the defendant might be caught unawares of the prosecution‘s case is unlikely provided the prosecution had carried out a reasonably expeditious and full disclosure – as will happen in New Zealand under the 2008 Criminal Disclosure Act. In any case, the issue is unlikely to occur very often (Spencer, In press).

Most of all, although cases may arise whose special circumstances might create, say, a need for re-examination, to reiterate, ―the fact that the reform would not solve every problem is no reason for rejecting a reform that would solve many of them‖ (Spencer, In press).

302 In the UK Parliamentary debates at the introduction of the YJCE Act the government apparently accepted this argument as it assured the opposition that lines of cross-examination need not be disclosed.

303 In the UK Hoyano and Keenan argue that the right of non-disclosure has been so reduced by legislation that

this is a complete non-issue. The defence has had to disclose the identity of alibi witnesses since 1968 (CJA 1967 s. 11, now part of the CPIA 1996) and while there have been isolated complaints, there does not appear to have been widespread abuse of the requirement. What is more, all of the main professional bodies are on record as agreeing to defence disclosure in principle, although there is still argument about how much (Hoyano and Keenan 642).

A very practical issue raised by Western Australian counsel was that Legal Aid Services will not pay for the extra preparation time required for pre-recording as well as trial (Australian Federal Police & Prosecutions, 2005, 133; Hoyano, 2000).304 As a matter of winning hearts and minds amongst barristers this issue is significant. Legal aid rates of pay and the accompanying rigid and formulaic assessment methods are already major bones of contention in many jurisdictions including New Zealand. However, again this is a practical issue and not a principled one and as such is capable of resolution if the will is there. It may even be that the costs of increased preparation are offset by the financial savings from pre-recording described above.

Another argument is that pre-recording is unnecessary given all the other special measures now available, especially fast-tracking: that, in short, testifying has been rendered so much less stressful that it is not necessary to introduce anything as drastic as pre-recording (Birch & Powell, 2003; Hoyano & Keenan, 2007, 660-1).

Hoyano & Keenan (2007, 662) argue that it is still worth pre-recording even were fast tracking effective because of its other advantages. This is the position of Western Australia‘s Chief Judge, who has stated that pre-recording should still be used where young or impaired children are involved (see Chapter 5).

However, the advantages referred to are largely extraneous to pre-recoding per se, being the use of CCTV and more private nature of a pretrial hearing. If those can be achieved at trial the case for pre-recording that can only be achieved late is weakened. There is little point in pre-recording unless it can be conducted early (Spencer, In press (b)).

Time will tell whether there is any realistic prospect of fast-tracking succeeding in NZ but it has to be said that on past performance speed appears unlikely.

In conclusion, pre-recording is one of the few realistic prospects for reducing the impact of delay, one of the most serious problems facing child witnesses both in terms of stress and the quality of evidence. The arguments against it are in the main practical rather than principled and the principled arguments are relatively minor.

The potential benefits of the use of intermediaries to take children‘s evidence are agreed by virtually all to whom we spoke whether in South Africa, Norway or the UK, despite the differences in approach in the systems.

An intermediary system reduces the stress to the witness of testifying in court by reducing the fear of miscommunication and by insulating the witness from aggression or sarcasm in the cross-examiner‘s tone (Centre for Child Law (University of Pretoria) & Childline, 2007, para.147; Jonker & Swanson, 2007, 5-17; Muller & Hollely, 2009, 28; Plotnikoff & Woolfson, 2007a, 60; Schutte, 2005, 4). All of the English interviewees agreed that witnesses using intermediaries were less stressed. In other systems such as France and Norway, where specialist examiners are routinely used to conduct children‘s examinations, secondary trauma to child witnesses is virtually unknown (see Chapter 5).

Hoyano and Keenan have raised a concern that intermediaries might undermine the witness‘s confidence in his or her ability to testify, presumably because the witness might interpret the intermediary as a signal of the court‘s lack of confidence in him or her (Hoyano & Keenan, 2007, p. 664). Studies of witnesses‘ opinions, however, show that not one wished they had not had an intermediary (Plotnikoff & Woolfson, 2007a, 59).

The second advantage of intermediaries is that they improve the quality and quantity of vulnerable witnesses‘ evidence, enhancing the accuracy of the fact-finding process (S v Mokoena 2008, 21; Australian Federal Police & Prosecutions, 2005, 158; Centre for Child Law (University of Pretoria) & Childline, 2007, para.147; Plotnikoff & Woolfson, 2008; Schutte, 2005, 9).305 To have the best evidence benefits not only society but also innocent defendants: in one UK case an intermediary enabled the complainant to communicate that the police had the wrong person imprisoned (Plotnikoff & Woolfson, 2008, 96).

Intermediaries both facilitate the witness‘s communication and reduce witness stress, which is known to inhibit memory (Malloy & Quas, 2009), and can also assist the court to make appropriate modifications to the trial process (Plotnikoff & Woolfson, 2008, 93-94).306

―Confusing language in the court prevents children giving the best possible evidence. If a communicator or intermediary reduces the stress of cross-examination and elicits clearer evidence – evidence that is truly responsive to questions the child can understand – the trial process would be fairer, without derogating from the defence‘s right to cross-examine.‖. (Australian Federal Police & Prosecutions, 2005, 158)

The English experience also suggests that contact with intermediaries raises awareness of the needs of child witnesses and of best practice amongst other practitioners in the criminal justice system (see Chapter 5).

Our English and South African contacts were unanimous in their opinion that their respective intermediary services have enabled significant numbers of children and (in England) vulnerable adults to come to court who would previously have been excluded from court because it could not meet their communication needs. There are reports from South Africa that intermediaries enable very young children to testify (Jonker & Swanson, 2007; Muller & Hollely, 2003).307 The English practitioners we interviewed stated that intermediaries led to a significant increase in access to justice for the very young, children with disabilities and children with Asperger‘s in particular.

In Plotnikoff and Woolfson‘s pilot evaluation participants estimated at least half of the 12 trial cases would not have reached trial without the intermediary‘s assistance (Plotnikoff & Woolfson, 2007a, 57). In their small sample, every witnesses, family member or carer interviewed emphasized the intermediary‘s crucial role in enabling the witness to testify (Plotnikoff & Woolfson, 2007a, 60; 2008, 92).

Better quality evidence should also lead to some cost savings. The ability to obtain better evidence before trial helps police and prosecutors to recognise and drop poor cases (Plotnikoff & Woolfson, 2007a, 59) and at trial improves the chances of the child coming up to proof (Plotnikoff & Woolfson, 2007a, 58). Some English interviewees believed that the increased quality of the child‘s statement encourages offenders to plead guilty.

Criticism of intermediary systems are less prevalent in countries which have adopted such a service than the author expected. In discussions with practitioners and judges in the UK we encountered almost no criticism of the service.

The first objection is that intermediaries undermine cross-examination, impeding the conduct of questioning and counsel‘s rapport with the witness and thus the defense‘s right to test the evidence. Some suggest that the connection between counsel and witness is already under strain from the artificiality and technical difficulties created by the imposition of CCTV.308

307 K Hollely personal communication December 10 2008, c.f.: B Edward‘s skepticism, personal communication August 28 2009.

Most lawyers firmly believe that cross-examination is a powerful investigative tool, even as ―the greatest legal engine ever invented for the discovery of truth‖ (Wigmore, 1979, para.1367). The key to its success is usually seen as the psychological control exerted over the witness. Issues of timing, tone, stance and the order of questioning are seen as absolutely critical to this (Henderson, 2001, 2002; Hobbs, 2003).

It is therefore obvious that an interlocutor would concern many advocates and that that person might rephrase or re-order questions would raise even greater fears.

The first point to make is that in England practitioners we interviewed generally said that intermediaries had not interfered with their conduct of questioning or rapport with witnesses. However, in England the norm is for intermediaries to intervene relatively little.

Second, the South African courts‘ position is that it is a matter of balance: any detriment to the defence is made up for by the protection of the child and the benefit to the court and society of better quality evidence (Klink 1996, 443; Australian Federal Police & Prosecutions, 2005, 155; Centre for Child Law (University of Pretoria) & Childline, 2007, para.144).

Some might argue that the South African courts‘ example is not relevant as they are following their constitutional duty to protect children. In fact the common law limits cross-examination on the basis of the same competing rights of witnesses and the court itself.

The right to cross-examine has never been absolute: it is already limited in numerous ways. Under the common law the court has always been entitled to disallow questions which, as now codified in s85 Evidence Act 2006, are ―improper, unfair, misleading, needlessly repetitive or expressed in language that is too complicated for the witness to understand‖ (Phipson, 1990, para.12.30; Spencer, In press (b)).

Arguably if the defendant can be prevented from examining a witness because of the witness‘s likely distress then distress should also prevent counsel cross-examining.309 Also, that a judge can substitute for the defendant suggests that it is not impossible that another suitably qualified person be appointed to the task.

Opponents of intermediaries respond that problem questioning can be overcome by educating lawyers and judges (Australian Law Reform Commission, 1997; Plotnikoff & Woolfson, 2007a, 67; Queensland Law Reform Commission, 2000). Unfortunately

previous attempts to educate barristers have had little impact (Australian Federal Police & Prosecutions, 2005, 157; Cashmore & Trimboli, 2005; Eastwood & Patton, 2002; Plotnikoff & Woolfson, 2007a, 68). Similarly the New Zealand judiciary first issued practice directions calling for greater case management in 1992 to no particular avail.

Recent studies show that standards of cross-examination of children are still too low to allay fears that children‘s evidence is not being misrepresented (Hamlyn et al., 2004, 55-6; Hoyano & Keenan, 2007, 669; Kebbell et al., 2004; Ministry of Justice

UK, 2009; O'Kelly et al., 2003; Plotnikoff & Woolfson, 2006, 48-50, 67, 73-73; 2009; Zajac et al., 2003; Zajac & Hayne, 2006). It appears that rates of judicial intervention are also too low for safety (Kebbell et al., 2004; O'Kelly et al., 2003; Plotnikoff & Woolfson, 2007a, 67-68).

The difficulty is twofold: first, many of the problem techniques result from a deep- seated conviction that they are legitimate. Most lawyers hold that obstructive, obfuscatory tactics are integral to the practice and to the defendant‘s right to a fair trial (Henderson, 2002, 2003; Hobbs, 2003). Education must not merely raise awareness of children‘s needs but change the philosophy of advocacy (Birch, 2000; Ellison, 2001; Henderson, 2002; Plotnikoff & Woolfson, 2007a, 68-9; Temkin, 2000). Such a change is likely to be very slow in coming. Second, proponents of education underestimate the complexity of the task they set themselves. Evidential interviewing is extremely skilled work. It is highly unlikely advocates will be able to undertake the necessary training (Australian Federal Police & Prosecutions, 2005, 157; Plotnikoff & Woolfson, 2007a, 68).

So far this section has focused on the English and South African models. It now considers whether cross-examination by counsel might be fully replaced by specialist examination.

Many lawyers will be appalled by this idea on the basis cross-examination is fundamental to a fair trial. However, surely the real heart of all of the above argument is not whether cross-examination occurs but whether the purpose behind it is met.

There is a parallel between the use of specialists and the so-called right of cross- examination and the debate over the defendant‘s right of confrontation and the use of hearsay evidence or screening devices. Opponents of special measures, particularly in the US, have argued that they violate the right which, it is argued, requires a personal meeting between accused and accuser. This argument has particular traction in the US whose constitution guarantees a right of confrontation. However, it is generally accepted outside the US (and in principle now within it (Crawford v Washington 2004) (Spencer, In press (b))) that confrontation means the

right to have the witness‘s evidence tested, not that witness and defendant come face to face. So long as there are proper opportunities for the witness to be cross- examined the purpose is satisfied.

The fundamental purpose of cross-examination is to test the witnesses‘ evidence. This is this purpose for which it was introduced to the criminal court in the mid 1700s (Cairns, 1998; Langbein, 1983); the purpose to which lawyers continue to refer today (DuCann, 1993, 62; Eichelbaum, 1989 ; Henderson, 2002; Munkman, 1991, 10; Spencer & Flin, 1993, 274) and the purpose, together with the protection of the witness, which underpins the various limitations imposed by the courts described above.

Applying the argument above, if the purpose of cross-examination is met by another, better method there is nothing to suggest it might not be replaced. As has been discussed widely in this report there is considerable empirical evidence that cross- examination is a poor test of a witness‘s accuracy or veracity. As Spencer & Flin said in 1993, comparing cross-examination with investigative interviewing:

―The characteristics of a typical interview conducted during cross-examination appear to violate all the principles of best practice, with the predicted outcome of maximising the risk of contaminating the evidence.‖ (Spencer & Flin, 1993, 307)

However, there are other effective and safe methods of testing of the evidence. Inquisitorial jurisdictions enable the defence to test the evidence by providing opportunities to have input into the child‘s interview and by allowing the defence an uninterrupted view of the child throughout (Bottoms & Goodman, 1996; Centre for Child Law (University of Pretoria) & Childline, 2007, para 144; Cordon et al., 2003; Delmas-Marty & Spencer, 2005; Spencer, In press (b); Spencer, In press; Spencer & Flin, 1993, chapter 14; see also Chapter 5). The European Court has explored the issue of what is adequate testing of witness‘s evidence – protected under Article 6 of the European Convention on Human Rights – in both adversarial and inquisitorial trials. It holds that the testing in inquisitorial courts meets the constitutional requirements (Hoyano & Keenan, 2007, 658; Spencer, In press (b); Spencer, In press).

In conclusion, intermediaries of all varieties can significantly reduce the secondary trauma to witnesses, improve the quality of their evidence and improve their access to justice. Opposition focuses around concerns that they will damage the defendant‘s right to a fair trial by impeding cross-examination. These arguments overlook the fact that cross-examination is already limited by competing interests and that, as empirical studies show, it has little to recommend it as an investigative method.

Section 106I .Visual recording of child’s evidence, application for directions

(b)that the whole of the affected child‘s evidence (including cross-examination and re- examination) be —

(i)taken at a special hearing and recorded on a visual recording; and (ii)presented to the court in the form of that visual recording,

[Section 106I inserted by No. 36 of 1992 s. 8; amended by No. 71 of 2000 s. 20; No. 46 of 2004

and, without limiting section 106M but subject to section 106HB(3), may include directions and conditions as to the giving up of possession and as to the playing, copying or erasure of the recording.

(i)is not to be in the same room as the affected child when the child‘s evidence is being taken; but (ii)is to be capable of observing the proceedings by means of a closed circuit television system and is at all times to have the means of communicating with his or her counsel;

(5) Where circumstances so require, more than one special hearing may be held under this section for the purpose of taking the evidence of the affected child, and section 106I and this section are to be read with all changes necessary to give effect to any such requirement.

[Section 106K inserted by No. 36 of 1992 s. 8; amended by No. 53 of 1992 s. 39; No. 71 of 2000

judge means the judge who presided at the special hearing or a judge who has jurisdiction co- extensive with that judge.

[Section 106M inserted by No. 36 of 1992 s. 8; amended by No. 71 of 2000 s. 24 and 29; No. 46

(3a)Where arrangements are made under subsection (2)(a) or (b) the affected child‘s evidence is to be recorded on a visual recording.

[Section 106N inserted by No. 36 of 1992 s. 8; amended by No. 48 of 1998 s. 8; No. 71 of 2000

[Section 106O inserted by No. 36 of 1992 s. 8; amended by No. 2 of 2008 s. 52.]

Where in any proceeding on indictment evidence of an affected child is given in a manner described in section 106N(2) or (4), the judge is to instruct the jury that the procedure is a routine practice of the court and that they should not draw any inference as to the accused‘s guilt from the use of the procedure.

[Section 106P inserted by No. 36 of 1992 s. 8; amended by No. 84 of 2004 s. 82.]

by reason of age, cultural background, relationship to any party to the proceeding, the nature of the subject-matter of the evidence, or any other factor that the court considers relevant. (3a)Despite subsection (3), in any proceeding for a serious sexual offence an order must be made under subsection (1) in respect of the person upon or in respect of whom it is alleged that the offence was committed, attempted or proposed unless the court is satisfied —

(4a)Where an arrangement under subsection (4)(c) is directed to be made, section 106N applies, with any necessary changes, as if the special witness were an affected child.

(4b) Where an arrangement under subsection (4)(b) is directed to be made, section 106F applies, with any necessary changes, as if the special witness were an affected child.

[Section 106RA inserted by No. 84 of 2004 s. 38; amended by No. 2 of 2008 s. 54.]

Section 17. Witnesses eligible for assistance on grounds of fear or distress about testifying.

(c)such of the following matters as appear to the court to be relevant, namely—

in relation to the accused or (if there is more than one) in relation to each of the accused.

(b)refusing an application for, or for the variation or discharge of, or (c)discharging,

a special measures direction and, if it is a magistrates‘ court, must cause them to be entered in the register of its proceedings.

and for the purposes of section 19(2), as it then applies to the witness, any special measures required to be applied in relation to him by virtue of this section shall be treated as if they were measures determined by the court, pursuant to section 19(2)(a) and (b)(i), to be ones that (whether on their own or with any other special measures) would be likely to maximise, so far as practicable, the quality of his evidence.

the direction shall cease to have effect at the time when the witness attains the age of 17.

then, so far as it provides as mentioned in paragraph (a)(i) or (ii) above, the direction shall continue to have effect in accordance with section 20(1) even though the witness subsequently attains that age.

(c)any interpreter or other person appointed (in pursuance of the direction or otherwise) to assist the witness.

(b)if a previous application has been made under that subsection, the time when the application (or last application) was made.

the court may sit for the purposes of the whole or any part of those proceedings at a place where such facilities are available and which has been appointed for the purposes of this subsection by the justices acting for the petty sessions area for which the court acts.

A special measures direction may provide for the wearing of wigs or gowns to be dispensed with during the giving of the witness‘s evidence.

(b)if a previous application has been made under that subsection, the time when the application (or last application) was made.

(b)to any person asking such questions, the answers given by the witness in reply to them,

and to explain such questions or answers so far as necessary to enable them to be understood by the witness or person in question.

A special measures direction may provide for the witness, while giving evidence (whether by testimony in court or otherwise), to be provided with such device as the court considers appropriate with a view to enabling questions or answers to be communicated to or by the witness despite any disability or disorder or other impairment which the witness has or suffers from.

whom a court has appointed an intermediary under subsection (1), except examination by the court, shall take place in any manner other than through that intermediary.

(b) The said intermediary may, unless the court directs otherwise, convey the general purport of any question to the relevant witness.

(b) An intermediary who is not in the full-time employment of the State shall be paid such travelling and subsistence and other allowances in respect of the services rendered by him or her as the Minister, with the concurrence of the Minister of Finance, may determine.

through an intermediary shall be inadmissible solely on account of the fact that such intermediary was not competent to be appointed as an intermediary in terms of a regulation referred to in subsection

(4) (a), at the time when such oath, affirmation or admonition was administered or such evidence was presented.

(b) If in any proceedings it appears to a court that an oath, affirmation or admonition was administered or that evidence has been presented through an intermediary who was appointed in good faith but, at the time of such appointment, was not qualified to be appointed as an

intermediary in terms of a regulation referred to in subsection (4) (a), the court must make a finding as to the validity of that oath, affirmation or admonition or the admissibility of that evidence, as the case may be, with due regard to—

(6) (a) Subsection (5) does not prevent the prosecution from presenting anew any evidence which was presented through an intermediary referred to in that subsection.

(b) The provisions of subsection (5) shall also be applicable in respect of all cases where an intermediary referred to in that subsection has been appointed, and in respect of which, at the time of the commencement of that subsection—

DPP Transvaal 2009] The Director of Public Prosecutions, Transvaal v the Minister of Justice and Constitutional Development and Others (the Centre for Child Law; Childline South Africa, Resources aimed at the Prevention of Child Abuse and Neglect, Operation Bobbi Bear, Children First, People Opposing Women Abuse and The Cape Mental Health Society as amici curiae) CCT 36/08. Constitutional Court. 2009: 8

Pupino, 2005] Judgment of the Court of Justice, Pupino, Case C_105/03 (16 June 2005), Court of Justice of the European Communities

S v Mokoena 2008] S v Mokoena, S v Phaswane CC7/07, CC192/07, High Court (Transvaal Provincial Division). 288 (2): 216.

Anon. (1985). The Testimony of Child Victims in Sex Abuse Prosecutions: Two Legislative Innovations. Harvard Law Review, 98, 806.

Armstrong, J. (1976). The Criminal Videotape Trial: Serious Constitutional Questions.

Astington, J. W., & Olson, D. R. (1990). Metacognitive and Metalinguistic Language: Learning to Talk about Thought. Applied Psychology: An International Review, 39(1), 77-87.

Auld, R. E. (2001). Review of the Criminal Courts of England and Wales. London: Lord Chancellor's Department.

Australian Federal Police, & Prosecutions, A. C. T. O. o. t. D. o. P. (2005). Responding to Sexual Assault: The Challenge of Change Canberra, A.C.T.: Australian Federal Police,

Australian Law Reform Commission. (1997). Seen and Heard: Priority for Children in the Legal Process. Sydney: Australian Law Reform Commission.

Bailin, A., & Grafstein, A. (2001). The Linguistic Assumptions Underlying Readability Formulae: a Critique. Language and Communication, 21, 285–301.

Birch, D. (2000). A Better Deal for Vulnerable Witnesses. Criminal Law Review.

Birch, D. (2007). Improving the Criminal Trial Process for Young Witnesses: A Consultation Paper. London: Office for Criminal Justice Reform, Home Office

Birch, D., & Powell, R. (2003). Meeting the Challenges of Pigot: Pre-trial Cross- examination of Vulnerable Witnesses Under s.28 of the Youth Justice and Criminal Evidence Act 1999. London: Home Office UK.

Blackwell, S. (2007). Child Sexual Abuse on Trial. Unpublished PhD, University of Auckland, Auckland.

Blumrick, R. (2006). Intermediary System to Minimise Stress in Child Witnesses. Bottoms, B., & Goodman, G. S. (1996). International Perspectives on Child Abuse and

Children's Testimony: Psychological Research and Law. Newbury Park, CA: Sage Publications.

Brennan, M., & Brennan, R. (1988). Strange Language: Child Victims under Cross Examination. Wagga Wagga: CSU Literacy Studies Network.

Brienen, M. E., & Hoegan, E. H. (2000). Victims of Crime in 22 European Criminal Justice Systems. Nijmegen, The Netherlands: Wolf Legal Productions.

Brown, D., Lamb, M. E., Pipe, M.-E., & Orbach, Y. (2008). Pursuing 'the Truth, the Whole Truth, and Nothing but the Truth': Forensic Interviews with Child Victims or Witnesses of Abuse. In M. L. Howe, G. S. Goodman & D. Cicchetti (Eds.), Stress, Trauma, and Children's Memory Development: Neurobiological, Cognitive, Clinical, and Legal Perspectives. Oxford: Oxford University Press.

Burton, M., Evans, R., & Sanders, A. (2006). Are Special Measures for Vulnerable and Intimidated Witnesses Working? Evidence from the Criminal Justice Agencies. London: Home Office.

Bygrave, L. (1997). Norway. In W. MacDonald (Ed.), The World Factbook of Criminal Justice Systems. Albany, New York: Bureau of Justice Statistics.

Cairns, D. (1998). Advocacy and the Making of the Adversarial Criminal Trial 1880-1865

Carter, C., Bottoms, B. L., & Levine, M. (1996). Linguistic and Socioemotional Influences on the Accuracy of Children's Reports. Law and Human Behaviour, 20(3), 335- 358.

Cashmore, J. (2002). Innovative Procedures for Child Witnesses. In H. L. Westcott, G. M. Davies & R. H. C. Bull (Eds.), Children's Testimony. Chichester: John Wiley & Sons, Ltd.

Cashmore, J., & De Haas, N. (1992). Children's Evidence Research Paper 1: The Use of Closed-Circuit Television for Child Witnesses in the ACT: Australian Law Reform Commission.

Cashmore, J., & Trimboli, L. (2006). Child Sexual Assault Trials: A Survey of Juror Perceptions. Crime and Justice(102).

Centre for Child Law (University of Pretoria), & Childline. (2007). Written Submissions to the High Court in the case of Phaswane and Mokoena 2007. Pretoria: The Centre For Child Law, Univeristy of Pretoria.

Charrow, R. P., & Charrow, V. R. (1979). Making Legal Language Understandable: A Psycholinguistic Study of Jury Instructions. Colombia Law Review, 79, 1306-1374. Connolly, D. A., Price, H. L., Lavoie, J. A. A., & Gordon, H. M. (2008). Perceptions and Predictions of Children's Credibility of a Unique Event and an Instance of a

Cooper, D., & Roberts, P. (2005). Special Measures for Vulnerable and Intimidated Witnesses: An Analysis of Crown Prosecution Service Monitoring Data London: Crown Prosecution Service UK.

Cordon, I., Goodman, G., & Anderson, S. (2003). Children in court. In P. v. Koppen & S. Penrod (Eds.), Adversarial versus Inquisitorial Justice: Psychological Perspectives on Criminal Justice Systems (pp. 167-189). New York: Kluwer Academic.

Cosgriff, C. (1988). Memorandum to New Zealand Courts Division. Unpublished Memorandum. Department of Justice.

Cossins, A., Goodman-Delahunty, J., & O'Brien, K. (2009). Uncertainty and Misconceptions About Child Sexual Abuse: Implications for the Criminal Justice System. Psychiatry, Psychology and Law, 16(3), 435-452.

Coughlin, S., & Jarman, R. (2002). Can the intermediary system work for child victims of sexual abuse? Families in Society, 83(5/6).

Crawford, E., & Bull, R. (2006). Teenagers' Difficulties with Key Words Regarding the Criminal Court Process. Psychology, Crime & Law, 12(6), 653-667.

Criminal Justice System (2007). Achieving Best Evidence in Criminal Proceedings: Guidance on Interviewing Victims and Witnesses, and Using Special Measures. Retrieved from

Damaska, M. (1986). The Faces of Justice and State Authority: A Comparative Approach to the Legal Process. New Haven: Yale University Press.

David, H. (1990). The Role of the Youth Interrogator. In J. R. Spencer, G. Nicholson, R. Flin & R. Bull (Eds.), Children's Evidence in Legal Proceedings: An International Perspective. Cambridge: Cambridge University Law Faculty.

Davies, E. (1998). Sexual Abuse Investigation and the Criminal Court Processes: Doing Justice to the Child? Unpublished PhD, University of Auckland, Auckland.

Davies, E., Fortune, C., Verbitsky, J., Seymour, F., Devere, H., & Smith, L. (2003). Evaluation of the Court Education for Young Witnesses Pilot. Wellington: Department for Courts.

Davies, E., Henderson, E., & Seymour, F. (1997). In the Interests of Justice? The Cross- examination of Child Complainants of Sexual Abuse in Criminal Proceedings. Psychiatry, Psychology and Law, 4(2), 217-229.

Davies, E., & Seymour, F. (1997). Child witnesses in the criminal courts: furthering New Zealand's commitment to the United Nations Convention on the Rights of the Child. Psychiatry, Psychology and Law, 4(1), 13-24.

Davies, E., & Seymour, F. W. (1998). Questioning Child Complainants of Sexual Abuse: Analysis of Criminal Court Transcripts in New Zealand. Psychiatry, Psychology and Law, 5(1).

Davies, G. (1992). Protecting the Child Witness in the Courtroom. Child Abuse Review, 1, 33-41.

Davies, G. (1999). The Impact of Television on the Presentation and Reception of Children's Testimony. International Journal of Law and Psychiatry, 22, 241-256.

Davies, G., & Noon, E. (1991). An Evaluation of the Live Link for Child Witnesses.

Davies, G., Wilson, C., Mitchell, R., & Milsom, J. (1995). Videotaping Children's Evidence: An Evaluation (Home Office Research and Statistics Department Research Findings No. 20). London: Home Office.

Delmas-Marty, M., & Spencer, J. R. (Eds.). (2005). European Criminal Procedures (2nd ed.). Cambridge: Cambridge University Press.

Dervieux, V. (2005). The French System. In M. Delmas-Marty & J. R. Spencer (Eds.),

Dezwirek-Sas, L. (1992). Empowering Child Witnesses for Sexual Abuse Prosecution. In

H. Dent & R. Flin (Eds.), Children as Witnesses. Chichester: John Wiley & Sons. Doherty-Sneddon, D., & McAuley, S. (2000). Influence of Video-Mediation on Adult-Child

DuCann, R. (1993). The Art of the Advocate (Second ed.). London: Penguin Books. Eastwood, C., & Patton, W. (2002). The Experiences of Child Complainants of Sexual

Eaton, T. E., Ball, P. J., & O'Callaghan, M. G. (2001). Child-Witness and Defendant Credibility: Child Evidence Presentation Mode and Judicial Instructions. Journal of Applied Social Psychology, 31(9), 1845-1858.

Edelstein, R. S., Goodman, G. S., Ghetti, S., Weede-Alexander, K., Quas, J. A., Redlich,

A. D., et al. (2002). Child Witnesses' Experiences Post-Court: Effects of Legal Involvement. In H. L. Westcott, G. M. Dazvies & R. H. C. Bull (Eds.), Children's Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John WIley & Sons, Ltd.

Eichelbaum, T. (1989 ). Mauet’s Fundamentals of Trial Techniques. Auckland: Oxford University Press.

Ellison, L. (2001). The mosaic art? Cross-examination and the vulnerable witness. Legal Studies, 21(3), 353-375.

Fivush, R. (2002). The Development of Autobiographical Memory. In H. L. Westcott, G.

M. Davies & R. H. C. Bull (Eds.), Children's Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John Wiley & Sons, Ltd.

Fivush, R., Peterson, C., & Schwarzmueller, A. (2000). Questions and Answers: The Credibility of Child Witnesses in the Context of Specific Questioning Techniques. In

M. L. Eisen, J. A. Quas & G. S. Goodman (Eds.), Memory and Suggestibility in the Forensic Interview (pp. 331-354). Mahwah: Lawrence Erlbaum Associates.

Freshwater, K., & Aldridge, J. (1994). The Knowledge and Fears about Court of Child Witnesses, Schoolchildren and Adults. Child Abuse Review, 3, 183-195.

Golding, J. M., Fryman, H. M., Marsil, D. F., & Yozwiak, J. A. (2003). Big Girls Don't Cry: The Effect of Child Witness Demeanour on Juror Decisions in a Child Sexual Abuse Trial. Child Abuse & Neglect, 27, 1311-1321.

Goodman, G. S., Myers, J. E. B., Qin, J., Quas, J. A., Castelli, P., Redlich, A. D., et al. (2006). Hearsay Versus Children's Testimony: Effects of Truthful and Deceptive Statements on Jurors' Decisions. Law and Human Behavior, 30(3), 363-401.

Goodman, G. S., Taub, E. P., Jones, D. P. H., England, P., Port, L. K., Rudy, L., et al. (1992). Testifying in Criminal Court: Emotional Effects on Child Sexual Assault Victims. Monographs of the Society for Research in Child Development, 57(5), 1- 159.

Goodman, G. S., Tobey, A. E., Batterman-France, J. M., Orcutt, H., Thomas, S., Shapiro, C., et al. (1998). Face-to-Face Confrontation: Effects of Closed-Circuit Technology on Children's Eyewitness Testimony and Jurors' Decisions. Law and Human Behavior, 22(2), 165-203.

Hamlyn, B., Phelps, A., Turtle, J., & Sattar, G. (2004). Are Special Measures Working? Evidence from Surveys of Vulnerable and Intimidated Witnesses. Retrieved from http://www.homeoffice.gov.uk/rds/pdfs04/hors283.pdf

Harnon, E. (1990). Children‘s Evidence in the Israeli Criminal Justice System with Special Emphasis on Sexual Offences. In J. R. Spencer, G. Nicholson, R. Flin & R. Bull (Eds.), Children’s Evidence in Legal Proceedings: An International Perspective Cambridge Cambridge University Press.

Henderson, E. (2000). Cross-Examination: A Critical Examination. Unpublished PhD thesis, Cambridge University.

Henderson, E. (2001). Cross-examination: A Critical Examination. University of Cambridge, Cambridge.

Henderson, E. (2002). Persuading and Controlling: The Theory of Cross-examination in Relation to Children. In H. L. Westcott, G. M. Davies & R. H. C. Bull (Eds.), Children's Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John Wiley & Sons.

Henderson, E. (2003). Psychological Research and Lawyers' Perceptions of Child Witnesses in Sexual Abuse Trials. In D. Carson & R. Bull (Eds.), Handbook of Psychology in Legal Contexts (2 ed.). Chichester: Wiley.

Hobbs, P. (2003). "You Must Say It For Him": Reformulating a Witness's Testimony on Cross-examination at Trial. Text, 23(4), 34.

Home Office. (1989). Report of the Advisory Group on Video Evidence. London: Home Office.

Hoyano, L. (2007). The Child Witness Review: Much Ado About Too Little. Criminal Law Review, 2007.

Hoyano, L. (2010). Coroners and Justice Act 2009 -- (3) Special Measures Directions Take Two: Entrenching Unequal Access to Justice? Criminal Law Review, 2010, 345-368.

Hoyano, L., & Keenan, C. (2007). Child Abuse: Law and Policy Across Boundaries.

Humphrey, J. (1986). Preparing the World for the Child: California's New Child Sexual Abuse Law. California Western Law Review, 23(52-71).

Jackson, H. (2003, 1-2 May 2003). Child Witnesses in the Western Australian Criminal Courts. Paper presented at the Child Sexual Abuse : Justice Response or Alternative Resolution Conference, Adelaide. from

Jones, J. (2007, 12 June 2007). Working with Defendants. Paper presented at the National Roll-Out and Beyond: Intermediaries Shaping the Future Fourth National Registered Intermediary Conference London. from

Jonker, G., & Swanson, R. (2007). Intermediary services for child witnesses testifying in South African Criminal Courts. SUR International Journal of Human Rights, 3.

Kaufmann, G., Drevland, G. C. B., Wessel, E., Overskeid, G., & Magnussen, S. (2003). The Importance of Being Earnest: Displayed Emotions and Witness Credibility. APPLIED COGNITIVE PSYCHOLOGY, 17, 21-34.

Kebbell, M., Hatton, C., & Johnson, S. (2004). Witnesses with intellectual disabilities in court: What questions are asked and what influence do they have? Legal and Criminological Psychology, 9(1), 13.

Kebbell, M. R., Deprez, S., & Wagstaff, G. F. (2003). The direct and cross-examination of complainants and defendants in rape trials: a quantitative analysis of question type. Psychology, Crime & Law, 9, 44-59.

Krahenbuhl, S., & Blades, M. (2006). The Effect of Interviewing Techniques on Young Children's Responses to Questions. Child: Care, Health & Development, 32(2), 321-331.

Lamb, M., Orbach, Y., Hershkowitz, I., Horowitz, D., & Abbott, C. B. (2007). Does the type of prompt affect the accuracy of information provided by alleged victims of abuse in forensic interviews? . APPLIED COGNITIVE PSYCHOLOGY, 21, 13.

Lamb, M. E., & Fauchier, A. (2001). The Effects of Question Type on Self-contradictions by Children in the Course of Forensic Interviews. APPLIED COGNITIVE PSYCHOLOGY, 15, 483-491.

Lamb, M. E., Orbach, Y., Hershkowitz, I., Esplin, P. W., & Horowitz, D. (2007). A Structured Forensic Interview Protocol Improves the Quality and Informativeness of Investigative Interviews with Children: A Review of the Research using the NICHD Investigative Interview Protocol. Child Abuse & Neglect, 31, 1201-1231.

Lamb, M. E., Orbach, Y., Hershkowitz, I., Horowitz, D., & Abbott, C. B. (2007). Does the Type of Prompt Affect the Accuracy of Information Provided by Alleged Victims of Abuse in Forensic Interviews. APPLIED COGNITIVE PSYCHOLOGY, 21, 1117-1130.

Landström, S., & Granhag, P. A. (In press). In-Court Versus Out-of-Court Testimonies: Children's Experiences and Adults' Assessments. APPLIED COGNITIVE PSYCHOLOGY.

Landström, S., Granhag, P. A., & Hartwig, M. (2007). Children's Live and Videotaped Testimonies: How Presentation Mode Affects Observers' Perception, Assessment and Memory. Legal and Criminological Psychology, 12, 333-347.

Langbein, J. (1983). Shaping the Eighteenth Century Criminal Trial: A View From the Ryder Sources. University of Chicago Law Review, 50.

Lash, B. (1995). Time Taken to Process Sexual Offence Cases Through the Courts.

Libai, D. (1969). The Protection of the Child Victim of a Sexual Offence in the Criminal Justice System. Wayne Law Review, 15, 977-1032.

MacFarlane, K. (1985). Diagnostic Evaluations and the Use of Videotapes in Child Sexual Abuse Cases. University of Miami Law Review, 40, 135-165.

Malloy, L., & Quas, J. (2009). Children's Suggestibility: Areas of Consensus and Controversy. In K. Kuehnle & M. Connell (Eds.), The Evaluation of Child Sexual Abuse Allegations: A Comprehensive Guide to Assessment and Testimony. Hoboken, N.J: Wiley & Sons, Inc.

Marsil, D. F., Montoya, J., Ross, D., & Graham, L. (2002). Child Witness Policy: Law Interfacing with Social Science. Law and Comtemporary Problems, 65(1), 209- 241.

Mathieson, D. L. (2005). Cross on Evidence (8th New Zealand ed.). Wellington: LexisNexis New Zealand Limited.

McKillop, B. (1997). Anatomy of a French Murder Case. American Journal of Comparative Law, 45.

Melinder, A., Goodman, G., Elertsen, D., & Magnussen, M. (2004). Beliefs About Child Witnesses: A survey of Professionals. Psychology, Crime and the Law, 10(4).

Miller, J. (2002). Questions about Constructions. Journal of Child Language, 29, 470-474. Ministry of Health. (2005). Living with Disability in New Zealand: Summary. Wellington:

Ministry of Justice UK. (2009). Government Response to the Improving the Criminal Trial for Young Witnesses Consultation. Retrieved. from.

Muller, K. (2002). Baseline Success Indicators for Training of Social Workers involved with Children in the Criminal Justice System: UNESCO.

Muller, K., & Hollely, K. (2003). The Hidden Benefits of Court Preparation for Young Witnesses.

Muller, K., & Hollely, K. (2009). Introducing the Child Witness (Second ed.). Port Elizabeth: Printrite.

Muller, K., & Tait, A. (1999). Little Witnesses: A Suggestion for Improving the Lot of Children in Court‖ THRHR, 62, 17.

Murray, K. (1995). Live Television Link: An Examination of its use by Child Witnesses in Scottish Criminal Trials. Edinburgh: The Scottish Office.

Myklebust, T. (2005). Dommeravhør av barn i Norge—intensjon og fakta [Investigative interviews of children in Norway—intention and facts] In M. Egge & J. Strype (Eds.), Politirollen gjennom 100 år. Tradisjon og endring [The role of the police in 100 years. Tradition and changes] (pp. 142-172). Forskning.: PHS.

Nathanson, R., & Saywitz, K. J. (2003). The Effects of the Courtroom Context on Children's Memory and Anxiety. The Journal of Psychiatry & Law, 31, 67-98.

New Zealand Law Commission. (1996). The evidence of children and other vulnerable witnesses: A discussion paper [PR26]. Wellington: New Zealand Law Commission.

Pragmatics, Syntax, and Semantics. Topics in Language Disorders, 20(2), 15-28. Nippold, M. A. (2007). Later Language Development: School-Age Children, Adolescents,

O'Kelly, C. M., Kebbell, M., Hatton, C., & Johnson, S. (2003). Judicial intervention in court cases involving witnesses with and without learning disabilities. Legal and Criminological Psychology, 8(2), 12.

Office for Criminal Justice Reform. (2005). Procedural Guidelines for Practitioners on the Intermediary Provision (s.29) of the Youth Justice and Criminal Evidence Act 1999 Retrieved. from.

Office for Criminal Justice Reform. (2007). Intermediaries: A Voice for Vulnerable Witnesses - Frequently Asked Questions. Retrieved. from.

Office of Criminal Justice Law Reform. (2005). Intermediary Procedural Guidance Manual.

Office of Criminal Justice Reform. (2007). Improving the Criminal Trial Process for Young Witnesses. London: Office of Criminal Justice Reform.

Open Society Foundation for South Africa. (2006). Minutes of Meeting on Models for the Management of Sexual Offences. Johannesburg, SA.

Orbach, Y., & Lamb, M. E. (2001). The Relationship between Within-interview Contradictions and Eliciting Interviewer Utterances. Child Abuse and Neglect, 25, 323-333.

Orbach, Y., & Lamb, M. E. (2007). Young Children's References to Temporal Attributes of Allegedly Experienced Events in the Course of Forensic Interviews. Child Development, 78(4), 1100-1120.

Orcutt, H. K., Goodman, G. S., Tobey, A. E., Batterman-Faunce, J. M., & Thomas, S. (2001). Detecting Deception in Children's Testimony: Factfinders' Abilities to Reach the Truth in Open Court and Closed-Circuit Trials. Law and Human Behavior, 25(4), 339-372.

Perera, K. (1984). Children's Writing and Reading: Analysing Classroom Language.

Perry, N. W., McAuliff, B. D., Tam, P., Claycomb, L., Dostal, C., & Flanagan, C. (1995). When Lawyers Question Children: Is Justice Served? Law and Human Behavior, 19(6), 609-629.

Peterson, C., & Biggs, M. (1997). Interviewing Children about Trauma: Problems with 'Specific' Questions. Journal of Traumatic Stress, 10(2).

Pike, J. (1973). Ought there be an Alternative Procedure to Calling Child Victims of Sexual Offences as Witnesses in the Trial of Accused Persons? Unpublished Report to Minister. Ministry of Justice.

Pipe, M.-E., Henaghan, M., Bidrose, S., & Egerton, J. (1996). Perceptions of the Legal Provisions for Child Witnesses in New Zealand. New Zealand Law Journal, January, 18-26.

Pipe, M.-E., Lamb, M. E., Orbach, Y., & Esplin, P. W. (2004). Recent Research on Children's Testimony about Experienced and Witnessed Events. Developmental Review, 24, 440-468.

Plotnikoff, J., & Woolfson, R. (2006). In Their Own Words: The Experiences of 50 Young Witnesses in Criminal Proceedings NSPCC & Victim Support

Plotnikoff, J., & Woolfson, R. (2008). Making Best use of the Intermediary Special Measure at Trial Criminal Law Review, 2008.

Poole, D. A., & Lamb, M. E. (1998). Investigative Interviews of Children; A Guide for Helping Professionals. Washington, DC: American Psychological Association.

Powell, M., & Thomson, D. (2002). Children's Memories for Repeated Events. In H. L. Westcott, G. M. Davies & R. H. C. Bull (Eds.), Children's Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John Wiley & Sons, Inc.

Powell, M. B., & Guadagno, B. (2008). An Examination of the Limitations in Investigative Interviewers' Use of Open-Ended Questions. Psychiatry, Psychology and Law, 15(3), 382-395.

Quas, J. A., Goodman, G. S., Ghetti, S., Alexander, K. W., Edelstein, R., Redlich, A. D., et al. (2005). Childhood Sexual Assault Victims: Long-Term Outcomes After Testifying in Criminal Court (Vol. 70(2)). Boston: Blackwell Publishing.

Quas, J. A., & McAuliff, B. D. (2009). Accommodating Child Witnesses in the Criminal Justice System: Implications for Death Penalty Cases. In R. F. Schopp, R. L. Wiener, B. H. Bornstein & S. L. Willborn (Eds.), Mental Disorder and Criminal Law: Responsibility, Punishment and Competence New York: Springer.

Quas, J. A., Thompson, W. C., & Clarke-Stewart, K. A. (2005). Do Jurors 'Know' What Isn't So About Child Witnesses? Law and Human Behavior, 29(4), 425-454.

Queensland Law Reform Commission. (2000). The Receipt Of Evidence By Queensland Courts: The Evidence Of Children (Report No 55): Queensland Law Reform Commission. Retrieved from http://www.qlrc.qld.gov.au/reports/r55pt2.pdf

Reed, L. D. (1996). Findings from Research on Children's Suggestibility and Implications for Conducting Child Interviews. Child Maltreatment, 1(2), 105-120.

Ross, D. F., Hopkins, S., Hanson, E., Lindsay, R. C. L., Hazen, K., & Eslinger, T. (1994). The Impact of Protective Shields on Videotape Testimony on Conviction Rates in a Simulated Trial of Child Sexual Abuse. Law and Human Behavior, 18(5), 553-566.

Runyan, D. K., Everson, M. D., Edelsohn, G. A., Hunter, W. M., & Coulter, M. L. (1988). Impact of Legal Intervention on Sexually Abused Children. Journal of Pediatrics, 113(4), 647-653.

Salhany, R. E. (1999). Cross-Examination: The Art of the Advocate (2nd ed.). Toronto: Butterworths.

Saywitz, K., & Camparo, L. (1998). Interviewing Child Witnesses: A Developmental Perspective. Child Abuse & Neglect, 22(8), 825-843.

Saywitz, K., Jaenicke, C., & Camparo, L. (1990). Children's Knowledge of Legal Terminology. Law and Human Behavior, 14, 523-535.

Schuman, J. P., Bala, N., & Lee, K. (1999). Developmentally Appropriate Questions for Child Witnesses. Queen's Law Journal, 25, 251-304.

Schutte, F. (2005, 20-23 March). Child Witnesses in the Criminal Justice System in South Africa: An Overview of Proposals for Reform. Paper presented at the 4th World Congress on Family Law and Children's Rights, Cape Town, South Africa. from

Scott, C. M., & Stokes, S. L. (1995). Measures of Syntax in School-Age Children and Adolescents. Language, Speech, and Hearing Services in Schools, 26, 309-319.

Smith, M. (2007). A Report of Mel Smith, Ombudsman Following a Reference by the Prime Minister under Section 13(5) of the Ombudsmen Act 1975 for an Investigation into Issues Involving the Criminal Justice Sector. Wellington: Office of the Ombudsmen. Retrieved from http://www.ombudsmen.parliament.nz/imagelibrary/100258.pdf

South African Law Commission. (1997). Sexual Offences Against Children Issue Paper 10, Project 108 South African Law Commission.

South African Law Reform Commission. (2000). Sexual Offence Courts 155 Crime Index vol.4, no.2 South African Law Reform Commission.

Spencer, J. (2005). Introduction. In M. Delmas-Marty. & J. R. Spencer (Eds.), European Criminal Procedures. Cambridge Cambridge University Press.

Spencer, J. (2008). Special Measures and Unusual Muddles. Archbold News, 2008(6), 1. Spencer, J. (In press (b)). Evidence and Cross-Examination. In L. D. L. Rooy, L. C.

Malloy, C. Katz & M. E. Lamb (Eds.), Children's Testimony (Second ed.). Chichester: Wiley.

Spencer, J. R. (1987a ). Child witnesses, video technology and the law of evidence.

Spencer, J. R. (1987b ). Child Witnesses and the Criminal Justice Bill. New Law Journal, 137

Spencer, J. R. (1987c ). Child Witnesses—a Further Skirmish. New Law Journal, 137.

Spencer, J. R. (1987d). Child Witnesses and Video technology: Thoughts for the Home Office. Journal of Criminal Law, 51.

Spencer, J. R. (1998). The role of experts in the common law and the civil law: A comparison. Expert witnesses in child abuse cases: What can and should be said in court. In S. J. Ceci & H. Hembrooke (Eds.), Expert witnesses in child abuse cases: What can and should be said in court (pp. 29-81). Washington, DC, US: American Psychological Association. doi:10.1037/10272-002

Spencer, J. R. (In press). Children's Evidence: The Barker Case, and the Case for Pigot.

Spencer, J. R., & Flin, R. (1993). The Evidence of Children: The Law and the Psychology

Spencer, J. R., & Glaser, D. (1990). Sentencing, Children's Evidence and Children's Trauma. Criminal Law Review, June, 371-382.

Stern, V. (2010). The Stern Review: A Report by Baroness Vivien Stern CBE of an Independent Review into How Rape Complainants are Handled by Public Authorities in England and Wales. London: Home Office.

Sternberg, K. J., Lamb, M. E., & Hershkowitz, I. (1996). Child Sexual Abuse Investigation in Israel In B. L. Bottoms & G. S. Goodman (Eds.), International Perspectives on Child Abuse and Children's Testimony: Psychological Research and Law. Newbury Park, CA: Sage Publications.

Stiver, C. (1974). Video-Tape Trials: A Practical Evaluation and a Legal Analysis. Stanford Law Review, 26.

Swim, J. K., Borgida, E., & McCoy, K. (1993). Videotaped Versus In-Court Witness Testimony: Does Protecting the Child Witness Jeopardize Due Process? Journal of Applied Social Psychology, 23(8), 603-631.

Temkin, J. (2000). Prosecuting and Defending Rape: Perspectives from the Bar. Journal of Law and Science, 27.

The Australasian Institute of Judicial Administration Incorporated. (2009). Bench Book for Children Giving Evidence in Australian Courts. Melbourne: The Australasian Institute of Judicial Administration Incorporated.

The Royal Commission on Criminal Procedure. (1981). The Royal Commission on Criminal Procedure (No. 13 (Cmnd 8092, 1981)).

Tominson, E. (1983). Non-adversarial Justice: The French Experience. Maryland Law Review, 42.

Toth, S. L., & Valentino, K. (2008). Translating Research on Children's Memory and Trauma into Practrice: Clinical and Forensic Implications. In M. L. Howe, G. S. Goodman & D. Cicchetti (Eds.), Stress, Trauma, and Children's Memory Development: Neurobiological, Clinical, and Legal Perspectives. Oxford: Oxford University Press.

United Nations Committee on the Rights of the Child. (2001). Consideration of Reports Submitted by States parties Under Article 44 of the Convention: Periodic reports of States parties due in 1993: United Nations Committee on the Rights of the Child

Wade, A. (2002). New Measures and New Challenges: Children's Experiences of the Court Process. In H. L. Westcott, G. M. Davies & R. H. C. Bull (Eds.), Children's Testimony: A Handbook of Psychological Research and Forensic Practice. Chichester: John Wiley & Sons, Ltd.

Walker, A. G. (1993). Questioning Young Children in Court: A Linguistic Case Study. Law and Human Behaviour, 17(1), 59-81.

Walker, A. G. (1998). A Few Facts about Children's Language Skills. Retrieved 8 August, 2008, from

Whitcomb, D. (1986). Prosecuting Child Sexual Abuse: New Approaches. National Insitute of Justice Report, 2(6).

Whitcomb, D., Shapiro, E. R., & Stellwagen, L. D. (1985). When the Victim is a Child: Issues for Judges and Prosecutors. Washington D.C: National Institute of Justice. Retrieved from

Whitney, L., & Cook, A. (1990). The Use of Closed-Circuit Television in New Zealand Courts: The First Six Trials. Wellington: Department of Justice.

Wigmore, J. H. (1979). Evidence in Trials at Common Law (Vol. 5). Boston: Little, Brown & Co.

Williams, G. (1987a). Videotaping Children‘s Evidence New Law Journal, 137 Williams, G. (1987b). More About Videotaping Children. New Law Journal, 137 Williams, G. (1987c). The Corroboration Question. New Law Journal, 137.

Wilson, K. (2007). Forensic Interviewing in New Zealand. In M.-E. Pipe, M. E. Lamb, Y. Orbach & A.-C. Cederborg (Eds.), Child Sexual Abuse: Disclosure, Delay, and Denial. New York: Routledge.

Working Party on Child Witnesses. (1996). Child Witnesses in the Court Process: A Review of Practice and Recommendations for Change. Wellington: Ministry of Justice.

Zajac, R., & Cannan, P. (2009). Cross-Examination of Sexual Assault Complainants: A Developmental Comparison. Psychiatry, Psychology and Law, 16, S36-S54.

Zajac, R., Gross, J., & Hayne, H. (2003). Asked and Answered: Questioning Children in the Courtroom. Psychiatry, Psychology and Law, 10(1), 199-209.

Zajac, R., & Hayne, H. (2006). The Negative Effect of Cross-Examination Style Questioning on Children's Accuracy: Older Children are Not Immune. Applied Cognitive Psychology, 20, 3-16.

▪	Yes	52 (95)	12 (75)	64 (90)*	13.52* (n=64)
▪	No	3 (5)	4 (25)	7 (10)*	15.71* (n=7)
* p<.05

	N (%)
Whanau or friend	26 (52)
Counsellor, social worker or CYF caregiver	12 (24)
Court personnel	10 (20)
Other	2 (4)
Total	50

	N (%)
Support person briefed	24 (51)
Support person not briefed	3 (6)
Unknown	10 (21)
N/A (court staff acted as support person)	10 (21)
Total	47
Objections by defence

Yes	44 (80)	6 (38)	50**	13.04**
No	11 (20)	10 (63)	21**	15.38**
** p<.01

Before midday	17 (37)	4 (31)	21 (36)	13.19
At or after midday	29 (63)	9 (69)	38 (64)	13.89
* p<.05

All regions				10 months‘ data collected Auckland Manukau		6 months‘ data collected Wellington Christchurch
N	children	(total)	71	21	27	5	18
N	trials		46	13	21	4	8

1	Defence	After going to the Church, I‘ll put it to you that he at no time told you not to say
2	Child	anything to anyone. He told me to say that.

All regions		Auckland	Manukau	Wellington	Christchurch
N children	134 (100%)	62 (46%)	28 (21%)	26 (19%)	18 (13%)

	N (%)				Offences alleged by complainants
Complainant Witness Total			Mean age		Sexual Other Total

Manner of initial statement
Forensic interview	88 (77)	10 (50)	98 (73)*	12.15 (n=95)**	79 (82) 9 (50)	88**
Written statement	26 (23)	10 (50)	36 (27)*	14.87 (n=31)**	17 (18) 9 (50)	26**

N (%)
Auckland			Manukau	Wellington		Christchurch		Total
Application for forensic interview
Yes	43 (96)	20 (100)		18 (78)	6 (60)		87 (89)
No	2 (4)	0		5 (22)	4 (40)		11 (11)

Auckland/Manukau (mean age) 114		Auckland only N (%)		Child‘s relationship with accused N (%)
Auckland Manukau		Complainant Witness Total		Family Non-family Total

CCTV	12.22**	12.29**	37 (69)	2 (25) 39 (63)*	40 (59)	14 (31)	54**
Screens	15.43**	15.2**	17 (31)	6 (75) 23 (37)*	28 (41)	31 (69)	59**

New Zealand Law Foundation Research Reports

Hanna, Kirsten; Davies, Emma; Henderson, Emily; Crothers, Charles; Rotherham, Clare --- "Child witnesses in New Zealand criminal courts: a review of practice and implications for policy" [2010] NZLFRRp 1