Unintended Consequences of the Expanded Use of Section 28 Pre-Recorded Cross- Examinations

As discussed in The Knowledge newsletter five years ago¹, special measures for witnesses in criminal proceedings have existed in England and Wales in various forms for over a century.

The Youth Justice and Criminal Evidence Act 1999 first introduced a range of special measures for eligible categories of adult witnesses, and also consolidated many of the existing protections for child witnesses, ushering in the modern era of special measures.

The most recently introduced special measure is pre-recorded cross-examination and re-examination of eligible witnesses under Section 28 of the Act. Taken together with s.27, which allows pre-recorded evidence-in-chief through video-recorded police interviews, s.28 allows a witness to record the entirety of their evidence in advance, without the need to attend trial. Section 28 is intended to improve the witness experience by reducing the stress and anxiety of waiting for an uncertain trial date and enhance evidence quality by allowing witnesses to give evidence closer in time to relevant events. It is also intended to increase participation by witnesses who might otherwise decline to attend a trial. Witnesses are spared the prospect of giving evidence again in the event of a re-trial, as their pre-recorded evidence can be re-used at any subsequent trial. These goals have become increasingly salient in light of the eyewatering trial delays caused by the current backlog in the criminal courts.

First piloted in 2014 for vulnerable witnesses in a small number of early adopter courts, the use of s.28 has dramatically increased since 2021, when it was made available for vulnerable witnesses in all Crown Courts and rolled out to intimidated witnesses in a larger number of courts. As of September 2022, s.28 has been made available nationwide to both vulnerable and intimidated witnesses, and prerecorded cross-examinations have now been relied on in thousands of criminal trials.

However, questions remain as to whether s.28 has been rolled out without sufficient consideration of the impact on criminal proceedings overall. Whilst the programme has been praised by many victims’ rights advocates, and touted by the Government in response to public outrage over the inadequate handling of rape cases, practitioners who work on s.28 cases have long raised concerns about the impact on listings practices and the court backlog. Concerns have also been raised about under-funding of the necessary staffing and technology and the potential undermining of defendants’ rights to a fair trial.

There is now evidence indicating that the use of s.28 special measures may also have a significant negative impact on conviction rates. One recent study found that there were 10 per cent fewer convictions overall in jury trials that used s.28 prerecorded evidence, and that conviction rates were 20 per cent lower in most types of rape trials.² A recent MoJ impact assessment called that conclusion into question, finding that s.28 had effectively no impact on conviction rates.³ Further data is clearly needed before any definitive conclusions can be drawn, but these preliminary findings raise difficult questions about how to balance the desire to improve the experience of vulnerable and intimidated witnesses against the possibility of distorting the outcomes of criminal trials.

Eligible Witnesses

The Section 28 procedure is now available in all Crown Courts to both vulnerable and intimidated witnesses. A vulnerable witness is defined as any child under 18 and any witness whose quality of evidence is likely to be diminished due to a mental or physical disorder or disability. An intimidated witness is defined as any witness whose quality of evidence is likely to be diminished by reason of fear or distress, as well as any complainant in a sexual offence, domestic abuse allegation, or specified human trafficking offence and any witness in specified categories of serious violent offences.

A court will not automatically grant a s.28 direction to all eligible witnesses. The court must first satisfy itself that the special measure is likely to maximise the quality of the witness’s evidence, and then consider any views expressed by the witness. Finally, the court must consider whether the measure might tend to inhibit the evidence from being effectively tested by a party to the proceedings. Certain groups, including the Law Commission⁴ and the Victims Commissioner⁵, have called for all complainants in sexual offences cases to be automatically entitled to pre-record the entirety of their evidence before trial, should they wish to do so. For the time being however, courts must still engage in a balancing exercise which includes consideration of whether a s.28 direction would infringe on a defendant’s ability to challenge the evidence against them fairly and effectively.

Restrictions on Cross-Examination

Where a s.28 direction is made, a judge has the discretion to order a ground rules hearing (GRH) and make directions, which may include restrictions on the form or content of questions that can be asked of the witness. The use of pre-approved limited crossexamination questions for child witnesses has been approved since the landmark case of R v Baker [2010] EWCA Crim 4. Subsequent case law has held that a defendant’s right to a fair trial is not compromised by judicial limitations on cross-examination of a child or vulnerable adult witness.⁶ It has also held that any potential unfairness can be dealt with by appropriate directions to the jury, explaining the limitations on cross-examination and the reasons behind them.⁷

Although s.28 is no longer limited to vulnerable witnesses, it has become standard practice in many s.28 cases to require a list of cross-examination questions for pre-approval. This approach was affirmed in R v Dinc [2017] EWCA 1206, in which the complainant was an autistic 14-year-old child. The trial judge reviewed and pre-approved questions to be asked in cross-examination, and the complainant was cross-examined using the s.28 procedure. The conviction was appealed partly on the basis that the restriction on cross-examination questions had undermined the fairness of the trial.

The Court of Appeal noted that the appellant had not been able to identify a proper question he had been prevented from asking, or an issue that was not aired, stating that, “There is nothing inherently unfair in restricting the scope, structure and nature of cross-examination or in requiring questions to be submitted in advance, in any case involving a child witness or a witness who suffers from a mental disability or disorder. This applies to all cases, not just those in the section 28 pilot scheme.” The Court clarified that previous case law “simply reminded trial judges of the potential for prejudice if a defence advocate is wrongly prevented from pursuing a legitimate line of questioning.”

In R v A [2022] EWCA Crim 988, the Court of Appeal considered the restriction of cross-examination questions asked of a complainant who was over 18 at the time of the s.28 hearing and not suffering from a mental disability or disorder. The Court found that the amendments to the questions made by the judge did not undermine the conviction, as they were “unobjectionable and proportionate” and “did not alter the substance or nature of the questions.” The Court noted that although the complainant was “no longer a child” this was “a sensitive case in which an allegation of sexual assault was made by one young family member against another – and where the complainant was barely out of her teens.”

With the extension of the s.28 procedure to intimidated adult witnesses, some judges have, in those cases, adopted the practice of pre-approving a list of topics, rather than specific questions to be asked in cross-examination.

Whilst this is a lighter touch approach than pre-approving every question, it remains entirely a matter of judicial discretion, and the practice has yet to be considered by the Court of Appeal. It is not clear that there is any justification for routinely constraining the form or substance of counsel’s questioning of adult witnesses with normal communication capabilities.

Further Questioning of Witnesses

Once a s.28 recording has been made, a witness may only be re-called for further questioning if the party wishing to re-call the witness has become aware, since the time when the original recording was made, of a matter which that party could not, with reasonable diligence, have ascertained by that time, or where it is otherwise in the interests of justice.

The timing of disclosure and formulation of charges therefore becomes particularly important in s.28 cases. The defence is effectively required to act with extra diligence in pursuing disclosure issues, given the high bar for re-calling a witness – even where new information or evidence becomes available after the cross-examination has taken place. As a practical matter, judges are often very reluctant to delay the s.28 timetable once it has been set, even where disclosure issues arise.

This issue was addressed in R v Benn-Landale [2017] EWCA Crim 132, a case in which the indictment was amended at trial to add additional charges. The defendant was initially charged with two counts of rape and the defence to the charges was consent. The cross-examination of the complainant was carried out accordingly, using the s.28 procedure. At trial, the Crown made a successful application to add two additional counts of sexual activity with a child under 16. The defence to those charges was a belief in the complainant being at least 16. Consequently, an issue arose which had not been relevant to the original charges, and therefore had not been explored in the cross-examination – namely when and how the complainant had communicated her age to the defendant.

The conviction was appealed on the ground that the judge should not have granted leave to add the alternative counts, because the complainant was not asked during the s.28 cross-examination about when she told the defendant her age. The Court of Appeal dismissed the appeal, finding that the safety of the conviction was not undermined because the complainant’s age and communication of her age to the defendant was not suddenly raised midtrial. It does not appear that any application was made by the defence to re-call the witness at trial. Nonetheless, the Court noted that “s28 aims, among other things, to protect a vulnerable witness from the ordeal of giving evidence during the trial in a Crown Court. Necessarily, the aims include attention to the undesirability of the process being undermined by subsequent recall of the vulnerable witness for further questions.”

This decision suggests that the bar will be very high for re-calling a witness in a s.28 case, even where the fundamental nature of the charges and applicable defences have altered by the time of trial.

Impact on the Trial Process

In April 2023, the Ministry of Justice published a process evaluation⁸ intended to assess whether the expansion of s.28 for intimidated witnesses was working as intended. The research consisted
of interviews with 11 intimidated witnesses, 3 trial advocates, and 26 other practitioners including court staff, police officers, Independent Sexual Violence Advisors and CPS reviewing lawyers.

The MoJ process evaluation found that both practitioners and witnesses noted an improved experience for witnesses giving evidence via s.28 as opposed to live cross-examination.

Key factors cited were physical separation from the defendant and being able to seek support and therapy at an earlier point in time. Practitioners raised concerns about whether police were unduly influencing witness decisions to seek special measures. Advocates and court staff both believed s.28 to have a negative impact on court listings, due to the additional hearings and the need for the same judge and advocates to be available at all hearings.

On the basis of this somewhat limited data, then-Justice Secretary Dominic Raab announced that “prerecorded evidence improves rape victims’ experience of court”.⁹ He asserted that the successful rollout of pre-recorded evidence to intimidated witnesses was a key indication that the Government was close to achieving its targets of driving up prosecutions and convictions for adult rape cases.

In December 2023, however, Professor Cheryl Thomas of UCL Law submitted written evidence on the use of s.28 to the Parliament Justice Select Committee enquiry on s.28.¹⁰ Professor Thomas provided a detailed analysis of all s.28 cases from June 2016 to June 2023, which included 4,392 cases involving 28,793 charges. Among the most striking findings of this analysis are that there were 10 per cent fewer convictions overall in trials that used s.28 prerecorded evidence and convictions were 20 per cent lower in rape trials (with the exception of rape of a male under 16 as a historic offence). The difference in conviction rates for s.28 was also greater with adult witnesses than children, and the conviction rate was markedly lower when s.28 was used in cases involving non-sexual offences – ranging from 8 to 23 per cent lower.

Professor Thomas cautioned that there was no reliable evidence to indicate which of the possible explanations might account for the stark results of her analysis, and called for further research and improved data collection.

She recommended that witnesses considering special measures be informed of the impact of s.28 on conviction rates, and that the Government pause any further changes to s.28 until the factors leading to lower conviction rates could be identified. Other researchers have raised reservations about the strength of claims that this preliminary data shows a strong correlation between the use of s.28 and lower conviction rates.¹¹ They have noted that the timing of the data collection meant that the national rollout of s.28 to intimidated witness had not been fully taken into account.

On 10 March 2025 the MoJ issued its second impact evaluation of pre-recorded cross examination for vulnerable and intimidated witnesses, which effectively concluded that s.28 has “no impact on conviction rates in either direction.”¹² The MoJ evaluation looked at cases between January 2019 and 2022 and found that “s.28 has reduced the length of time to cross-examination with no discernible impacts on case outcomes or court functioning.” The report noted that the analysis was constrained by “operational decisions and data collection at the time increased s.28 eligibility was rolled out.”

Professor Thomas criticised the report’s methodology, stating that, “They have designed a study that excludes more than 80% of verdicts.”¹³ She went on to state that the report is “making a pretty big statement in saying that s.28 does not affect a jury’s willingness to convict. I don’t think their analysis is strong enough to say that.” One Crown Court judge interviewed by the Law Society Gazette stated that “Professor Thomas’s research accurately reflects the experience of judges up and down the country that “juries just do not view [pre-recorded] evidence in the same way” as live evidence from a witness in court.¹⁴

The desire to relieve vulnerable and intimidated witnesses of the stress and anxiety of criminal proceedings is a worthy goal, and the last 25 years of modern special measures has been a success story in many respects.

However, the widespread use of entirely pre-recorded evidence for key witnesses may simply strike too closely at the heart of the adversarial process.

As ever, more funding for the criminal justice system, more transparent and reliable data collection and a genuinely evidencebased approach from the Government would be the most effective way to ensure that the s.28 initiative serves its intended purpose, without further unintended consequences for the fairness and integrity of jury trials.

¹Pre-recorded cross examination and the expansion of “special measures” | The Knowledge (12 November 2020)
²Written evidence submitted by Professor Cheryl Thomas (PRE0029) to Justice Select Committee (December 2023)
³Impact Evaluation of Pre-Recorded Cross Examination for Vulnerable & Intimidated Witnesses |Ministry of Justice (10 March 2025)
⁴Evidence in Sexual Offences Prosecutions: Consultation Summary | The Law Commission (23 May 2023)
⁵Next steps for special measures: A review of the provision of special measures to vulnerable and intimidated witnesses | Victims Commissioner (May 2021)
⁶R v Edwards [2011] EWCA Crim 3028 (a defendant’s right to a fair trial was not compromised because his advocate could not crossexamine in the “traditional” way).
⁷R v Wills [2011] EWCA Crim 1938 (judges should explain the limitations on crossexamination to a jury and the reasons for them).
⁸Process evaluation of Section 28: Evaluating the use of prerecorded cross-examination (Section 28) for intimidated witnesses | Ministry of Justice (3 April 2023)
⁹Pre-recorded evidence improves rape victims’ experience of court: Press release | Ministry of Justice and The Rt Hon Dominic Raab (3 April 2023)
¹⁰Written evidence submitted by Professor Cheryl Thomas (PRE0029) to Justice Select Committee (December 2023)
¹¹Supplementary written evidence submitted by Dr Natalie Kyneswood (PRE0037) to Justice Select Committee: Response to Submission PRE0029: YJCEA 1999 Section 28 and Conviction Rates (February 2024)
¹²Impact Evaluation of Pre-Recorded Cross Examination for Vulnerable & Intimidated Witnesses |Ministry of Justice (10 March 2025)
¹³Pre-recorded evidence has no impact on rape conviction rates, MoJ claims | Catherine Baksi, The Law Society Gazette (11 March 2025)
¹⁴Ibid.

This article was published in The Knowledge Newsletter April 2025 Spring Edition