Abstract
This article examines the potential risk of wrongful conviction for defendants facing historical sexual abuse charges where there is substantial delay. This risk arises from problems with truth-finding based on witness testimony, challenges posed by missing evidence and the increasing erosion of procedural safeguards. This article considers two recent proposals for reform, including first, whether the Court of Appeal should be more prepared to revisit the factual basis of decisions in historical sexual offence appeals; or second, whether there is a need to strengthen procedural safeguards at trial through the doctrine of abuse of process for delay. This article concludes that, whilst there would be advantages to broadening the grounds for appeal, the criminal courts should be more prepared to stay substantially delayed claims for abuse of process where there is missing evidence. The current approach has the potential to be unfair and fails to protect those defendants who are most disadvantaged by delayed claims.
Keywords
Introduction
Historical sexual offences present unique difficulties for the criminal justice system. The private context in which most sexual offences occur means there will often be no corroborating evidence outside of the complainant(s)'s testimony. In such a context, there is an ‘obvious evidential problem’ where accessing the truth is especially difficult because only the involved parties may have the potential to know whether the alleged crime occurred (Naughton, 2019: 462). This becomes more problematic where the reporting of the alleged offences has been significantly delayed. There is no statute of limitations for indictable offences in England and Wales, meaning that offences can be prosecuted decades after they allegedly occurred. In cases where there is substantial delay, there will often be a non-specificity to the allegations. This leads to defendants being charged for offences in specimen counts, which state that a certain offence happened at various times across multiple years with no specific details available around the dates or timing of the alleged incidents. This article is predominantly concerned with the danger of wrongful conviction in this context, where the substantial delay coupled with the vagueness of the allegations makes it extremely difficult for the defendant to mount a defence to the charge(s). Therefore, whilst there is no clear definition of what makes something a ‘historical’ offence or what constitutes ‘substantial’ delay, the concerns addressed in this paper are likely mostly relevant where the alleged offences occurred at least a decade before reporting.
This article will argue that problems with factfinding based on witness testimony leaves defendants vulnerable to wrongful conviction in this context. In response, it might be argued the systemic under-prosecution of sexual offences (including historical and otherwise) 1 (see for example, House of Commons Home Affairs Committee, 2022) justifies the current approach, as the proportion of cases brought to trial is so small that, within this, the chance of false allegations being prosecuted must be so limited as to be negligible. There is no certain way to assess the potential danger of wrongful conviction by debating whether (and to what extent) false allegations might be made. Research is limited due to difficulties with identifying and measuring what might count as a ‘false’ complaint (Rumney, 2006) and there are significant methodological limitations to many studies which make claims about the rate of false allegations (Rumney and McCartan, 2017). Therefore, debates around the prevalence (or lack thereof) of false complaints serve little purpose, as arguments on both sides are equally plagued by difficulties (Naughton, 2019). Furthermore, whilst wrongful convictions for all offences have significant consequences for those affected (see for example, Grounds, 2004 and Hoyle and Tilt, 2018), the serious nature of sexual offences and the stigma attached to them makes the impact of a false allegation particularly severe, as was demonstrated in empirical research by Burnett et al. (2017). Therefore, it is important that the criminal justice system has sufficient safeguards for defendants to protect them against wrongful conviction in this context. Crucially, whilst we cannot assess the potential extent of false allegations and wrongful convictions for historical sexual abuse, we can turn our attention to the process by which guilty verdicts are reached and upheld. If we cannot be confident that the criminal justice process can always reach a safe verdict in historical sexual abuse cases, then this should warrant consideration.
First, this article will outline why there is a potential danger of wrongful conviction in historical sexual offence cases. The key point is that jury verdicts based upon an assessment of witness credibility are vulnerable to error. Therefore, difficulties with truth-finding in historical sexual abuse cases make it imperative to have due process safeguards to protect defendants from wrongful conviction. Yet, as will be argued, it is debatable whether there is currently sufficient protection in cases involving substantial delay.
Second, this article will consider two recent proposals for reform in response to this perceived danger of wrongful conviction. First, whether there is an argument to broaden the right to appeal against convictions for historical sexual offences to encourage the Court of Appeal (Criminal Division) (hereafter CACD) to be more willing to revisit jury verdicts to examine the factual strength of the evidence underpinning the conviction (Henry and Gray, 2020). Or second, whether the courts should be more willing to stay prosecutions for historical sexual abuse where there is substantial delay on the grounds that the defendant cannot have a fair trial (Hickman, 2020). This paper will argue that, whilst there would be potential advantages to broadening the current grounds for criminal appeals, there may also be dangers with encouraging the CACD to revisit the factual basis of jury decisions. Furthermore, that this approach may simply defer issues that should have been resolved at trial to the jurisdiction of the appeal court. Therefore, it is argued that serious consideration should be given to whether the current approach of the courts to abuse of process for delay provides sufficient protection for defendants to ensure the right to a fair trial, or whether the courts ought to be using this power more readily in cases of substantial delay. Whilst this proposal is unlikely to find favour in the current criminal justice climate given the difficulties faced in prosecuting sexual offences, defendants have a right not to be wrongly convicted. Given the significant difficulties with accurate factfinding in this sphere, ensuring a robust approach to abuse of process for delay may be the only way to safeguard defendants from the danger of wrongful conviction.
Historical sexual abuse trials: A danger of wrongful conviction?
It is generally accepted that the innocent have the right not to be wrongly convicted and punished, which arguably incorporates a right to ensure the most accurate procedures for determining innocence and guilt (Roberts, 2019: 913–914). Yet it has previously been claimed that trials for historical sexual abuse ‘are perhaps the category of case in which the jury is the most likely to get it wrong’ (Zander, 2016: 215). As explained in the introduction, the danger of potential wrongful conviction lies in the fact that first, in sexual offence cases generally, and particularly in those involving substantial delay, there may be no other evidence available outside of the complainant's and defendant's testimony. Whilst this article is not specifically limited to considering single complainant cases only, the concerns flagged here are heightened where the defendant is facing allegations from one complainant. There are difficulties with truth-finding where the jury is required to base their decision on assessing witness credibility and which account (the complainant's or defendant's) they prefer. There does not seem to be a rational basis upon which a jury can be sure the complainant (or defendant) is telling the truth. First, cross-examination of witnesses in a historical case completely loses its (already debatable) (see for example, Roberts and Zuckerman, 2010: 298) evidential value, because delay in bringing the allegations to court means that inconsistencies in the accuser's or defendant's account cannot be taken as potential indicators of fabrication or a lack of reliability. Second, contrary to the emphasis in an adversarial trial upon the importance of a jury seeing a witness testify to observe their demeanour (i.e., by exclusion of hearsay evidence under the Criminal Justice Act 2003), research suggests that people are poor at judging whether a person is lying or being truthful, with a success rate little better than chance (Wellborn, 1991 and Bond and DePaulo, 2006). Mock jury studies have found that jurors may make false assumptions about the relationship between a witness's veracity and their body language (e.g., direction of their gaze and hand movements) (Chalmers et al., 2022). Furthermore, research in this field has also found that many of the cues which people depend upon to identify someone who is being untruthful or dishonest are unreliable indicators and often can be explained in other ways (Helm, 2023). Third, despite the above, research has also indicated that people have false confidence in their ability to detect when someone is lying (Chalmers et al., 2022 and Roberts and Zuckerman, 2010: 299). This adds further concern that jury verdicts in this context are potentially based upon false confidence in their own abilities to analyse the witnesses’ evidence.
Difficulties with truth-finding in this context has previously been balanced out by key procedural safeguards to provide protection against wrongful conviction. However, more recently, there has been a trend towards eroding various procedural safeguards with a view to ‘widening the goalposts’ to help facilitate the prosecution of sexual offences (Speechley, 2020: 92). This has been discussed more broadly in other literature (e.g., Burnett et al., 2017) but a key example would be the abolition of the corroboration warning, which required the judge to warn the jury of the need for caution before convicting a defendant solely on the testimony of a single witness (section 32 of the Criminal Justice and Public Order Act 1994). However, most important to the discussion in this article is the doctrine of abuse of process for delay. Historically, where the delay in bringing allegations was substantial, the court may have concluded this made it ‘impossible’ for the defendant to prepare or mount a defence, which led to the prosecution being stayed due to the difficulties with ensuring a fair trial on this basis. (e.g., R v Telford Justices Ex p. Badhan [1991] 2 Q.B. 78). However, this doctrine has gradually been restricted through case law with more recent decisions having been described as ‘tantamount to an abolition of the jurisdiction’ (Wood, 2019). It is now very rare for a prosecution to be stopped for abuse of process for delay. Instead, as will be discussed later, the criminal courts place faith in the criminal trial to resolve the issues, allowing for arguments about missing evidence and the potential prejudice this has caused to the defendant.
The erosion of other due process safeguards has left defendants more dependent upon the standard and burden of proof to protect them from wrongful conviction. The presumption of innocence requires the prosecution to prove the case so the jury can be ‘sure’ the defendant committed the offences. However, it has been argued this may leave defendants vulnerable to wrongful conviction by legitimising a passive defence, which focuses upon undermining the prosecution's case rather than advancing a positive defence of innocence (Naughton, 2013: 82). As will be discussed further later, this is particularly relevant for defendants in historical sexual offences where a lack of available evidence may leave them little choice but to focus on undermining the prosecution's case. Second, particularly where the sexual offences were allegedly committed against a child, it is questionable whether the standard of proof is adequate to protect defendants or whether ‘jurors are all too likely to be influenced by a climate of opinion’ (Zander, 2016: 215). The emergence of widespread sexual offending by Jimmy Saville and other high-profile examples, along with the growing international #MeToo movement, has led to what some describe as a ‘moral panic’ about the threat of child sexual abuse across England and Wales (Smith and Burnett, 2018). When combined with the stigma attached to sexual offences against children (Burnett et al., 2017), there is a danger jurors may be overly influenced by these contexts in making their decision. Given we are unable to observe jury decision making or carry out research with real jurors, our knowledge about the factors that might influence a jury's decision in this context is limited. Earlier mock jury studies had suggested that delay in making an allegation may make conviction less likely and lead to lower assessments of complainant credibility (Pozzulo et al., 2010: 952). However, recent mock jury research looking at delay in the context of sexual offences post #MeToo has been less clear around its impact. One study found mock jurors were more likely to find the defendant guilty where the delay in reporting was longer (i.e., 25 years) compared with a shorter delay (i.e., 15 years) (Fraser et al., 2022: 13). Another found that whilst a 10-year delay caused mock jurors to have less belief in the victim than a one-year delay, the believability in the victim increased again where the delay was 20 years (Thompson et al., 2021: 35). Whilst we must be cautious in drawing conclusions from a limited research base of mock jury studies, these findings indicate that longer delays may increase the chance of conviction. Arguably, if this might also apply in the context of real cases, this may be of concern given the potential prejudice caused to defendants by substantial delay, which will be discussed further later. Furthermore, this research suggests that public awareness of high-profile non-recent cases may have influenced the approach of jurors to historical cases. Therefore, the changing social context along with the potential passivity of the defence may bring into question whether the burden and standard of proof are effective as safeguards against wrongful conviction in historical sexual offence cases.
Concerns about the potential danger of wrongful conviction in the context of historical sexual abuse have recently been raised in other literature. First, in examining the approach of the CACD, Henry and Gray (2020: 326) claim the court's deference to upholding jury verdicts in the context of historical sexual abuse ‘shuts its eyes’ to the ‘uncomfortable truth’ that jury verdicts based on assessments of credibility are unreliable. Furthermore, that the approach of the CACD is ‘internally inconsistent with other constitutions of the Court of Appeal of England and Wales, and the judiciary of the UK’ where judges have acknowledged the fallibility of witness memory (Henry and Gray, 2020: 326). Second, in another article, the author (a retired judge) compared the approach taken by judges in criminal trials for historical sexual abuse with the approach of judges in the civil jurisdiction who are often ‘distrustful’ of claims based on memory of events many years ago (Hickman, 2020: 7). The approach of the criminal courts was criticised for appearing to ‘entail the view that while a professional judge cannot safely find a case proved to the civil standard, a lay jury—possibly “assisted” by a summing up…can safely find the same case proved to the criminal standard on the same evidence’ (Hickman, 2020: 10). Based on the suggestions made in these respective articles, two potential routes for reform will be considered. First, whether the CACD should be more prepared to set aside jury verdicts in historical sexual abuse cases where there are concerns about the strength of the evidence; and second, whether the criminal courts should be more willing to stay prosecutions for abuse of process for delay.
Responding to the danger of wrongful conviction: Proposals for reform
Examining grounds for appeal: The need for a broader approach to appeals against conviction for historical sexual abuse?
First, limitations to the criminal appeal framework in England and Wales may pose problems for those wrongly convicted for historic sexual abuse. Following a conviction in Crown Court there is no right to an appeal or a ‘re-hearing’ of the evidence in the case. Instead, a prospective appellant must apply for leave (or permission) to appeal to the CACD by establishing appropriate grounds. Section 2(1)(a) of the Criminal Appeal Act 1995 stipulates that the CACD shall allow an appeal against conviction if they think the conviction is ‘unsafe’. 2 However, prospective appellants must typically demonstrate either there has been a serious legal or procedural error during the investigation and trial that undermines the safety of their conviction, or that they have obtained significant fresh evidence which can do the same (section 23 Criminal Appeal Act 1968). Following an initial application for leave to appeal at the CACD, where this is rejected, or where an appeal is subsequently heard but refused, a convicted defendant must then apply to the Criminal Cases Review Commission (hereafter CCRC) to further challenge their conviction in the appeal process. The CCRC was established as an independent body to review potential miscarriages of justice but can only refer a case to the CACD where they think there is a ‘real possibility’ that the conviction would be overturned by the court (s. 13 Criminal Appeal Act 1995). 3 This has been criticised as a ‘statutory straitjacket’ which effectively confines the CCRC to only consider applicants’ cases on the same grounds as the CACD (Naughton and Tan, 2010: 342).
The approach of the CACD, and by association the CCRC, has been criticised for preventing the review of potential miscarriages of justice where there are concerns about the strength of evidence in the case, but where there is no fresh evidence and no identifiable legal or procedural errors (see, for example, the Westminster Commission Report on Miscarriages of Justice, 2021). This and other issues surrounding the legal tests applied for criminal appeals by the CACD and CCRC is currently under review by the Law Commission (2023). However, the CACD has long defined the limits of its role as a court of review and not re-hearing (R v Pendleton [2001] UKHL 66). The CACD defends its current approach on the basis it would ‘strike at the very root of trial by jury’ (R v McGrath [1949] 2 All ER 495) if the court was permitted to substitute its view for that of a properly legally directed jury, and furthermore, that it would be inappropriate because the jury has seen and heard the witnesses of fact, whilst the appeal court has not (see for example, R v Cooper and R v Melville [2005] EWCA Crim 1668). The limited framework for appeals against conviction in Crown Court therefore make the appeal system particularly difficult to navigate for defendants that might be wrongly convicted for historical sexual offences. The substantial delay in bringing the allegations to court inevitably means the potential to discover fresh evidence is likely to be limited and therefore many prospective appellants will be confined to pursuing potential legal and procedural errors. Where a defendant might be wrongly convicted for historical sexual abuse, they are therefore effectively dependent upon something having gone significantly wrong in the investigation or trial if they are to stand any chance at securing an appeal.
However, the perception the criminal appeal system is unduly difficult for appellants in historical sexual offence cases has recently been challenged based on original empirical research looking at the CCRC. Speechley (2020: 110) examined CCRC referral rates and outcomes in historical sexual offence cases (defined as where the offence occurred over a year prior to reporting) between April 1997 and April 2019 and concluded that the figures were largely in line with CCRC outcomes across all categories of cases. This was supported by findings that historical sexual offence cases accounted for 8% of all cases referred and that the conviction was quashed in 68% of these cases, which largely equated to the average of 67% across all categories of cases referred by the CCRC. Furthermore, Speechley (2020: 132) suggested that, as the majority of cases were overturned on ‘technicalities’, this showed ‘a favourable prejudice’ for historical sexual abuse cases at the CCRC/CACD. It is debatable whether we should attach much weight to this point however, as it is unsurprising that many appeals are based on legal arguments due to the difficulties with finding fresh evidence in historical sexual offence cases. Despite the conclusions above, it is notable that Speechley also found a decline in referrals of historic sexual offence cases over the last decade, with only two such cases being referred between 2013 and 2018, in which neither had their convictions quashed (2020: 127). This may suggest that convictions for historic sexual abuse are becoming increasingly difficult to appeal against. However, as Speechley (2020) acknowledges, because the research does not look at first instance appeal decisions, or the number of cases rejected by the CCRC, or total number of historic sexual offence convictions, this study can only provide limited insight into whether such cases are more difficult to navigate at appeal stage.
Due to the difficulties with truth-finding in historical sexual offences as explored above, Henry and Gray (2020) called for the CACD in England and Wales to take a broader approach to appeals in this context. The authors discussed the approach taken by the High Court of Australia in quashing the convictions of Cardinal George Pell for historical sexual abuse dating back 22 years (Pell v The Queen [2020] HCA 12). The High Court quashed the convictions following a review of all the evidence in the case, concluding that ‘the compounding improbabilities’ in the complainants’ accounts ‘nonetheless required the jury, acting rationally, to have entertained a doubt as to the applicant's guilt’ leaving ‘a significant possibility that an innocent person has been convicted’ (Pell v The Queen [2020] HCA 12 at paragraph 127). Henry and Gray (2020: 329) favour this approach for placing the safety of convictions as the highest priority in taking a ‘rigorously forensic approach to evidence heard at trial’. Contrastingly, however, the authors emphasise the reluctance of the CACD in England and Wales to review trial evidence in reference to a recent case where the CACD dismissed arguments undermining the safety of the conviction (Henry and Gray, 2020: 329). The case discussed was R v SJ and MM [2019] EWCA Crim 1570 where the CACD exhibited deference to the jury's decision concluding that ‘the critical issue’ was whether the jury were sure the complainants were telling the truth (paragraph 76). Henry and Gray (2020: 326) criticise this approach for resting upon ‘a fallacy that demeanour-based judgments are reliable’. The authors call for the CACD in England and Wales to take a robust approach to reviewing all the evidence in a case and to be prepared to overturn the jury's finding of the complainant's credibility ‘where this is incompatible with other evidence or there are any other factors which make the factual finding on which the conviction is based unsafe’ (Henry and Gray, 2020: 327). They emphasise the importance of appeal proceedings as a key stage in safeguarding the defendant's right to a fair trial with a need to ensure that convictions based on findings of credibility by the jury ‘are as safe and fair as possible’ (Henry and Gray, 2020: 329).
Whilst cases such as the Australian Pell case must be rare, it is arguably imperative that the jury's verdict can be reviewed if the evidence underpinning it appears weak or improbable. One such troubling case in England and Wales is that of R v BC [2018] EWCA Crim 2552, where the defendant was convicted of historical sexual offences against a single complainant 27 years previously. The case was appealed on grounds of inadequate defence, challenging the decision not to adduce evidence of the applicant's autism spectrum disorder. However, whilst the CACD were quick to dismiss the appeal on these grounds, they did highlight the overall weakness of the case at paragraph 13: 13. … It is also clear that there was an amount of material available to counsel in order to present the complainant as a thoroughly unreliable witness upon whose word the jury could place no credence: although that of course was ultimately a matter for the jury (emphasis added). One of the points, for example, made was that the complainant had said that the appellant had made him drink bleach on occasion, but there was no evidence whatsoever that there was any resulting injury, which one would have expected had the complainant indeed had to drink bleach. Further, so far as the matter relating to count 6 is concerned, there was no medical or other evidence available consistent with acid ever having been poured onto the exposed wounds in the way the complainant had asserted. There was, overall, no independent evidence of any assaults having occurred. It was, in effect, the complainant's word against that of the appellant. Moreover, the defence were in a position to point out that this complaint emerged very late in the day; and they were also in a position to deploy the complainant's criminal convictions, problematic lifestyle and other issues in a way in which defence counsel did.
There are arguments in favour of the CACD taking a broader approach to reviewing appeals in this context. As discussed, decisions based on credibility are inherently unreliable and therefore the CACD's deference to jury decisions in this context is difficult to justify. However, there may also be potential dangers with inviting the CACD to more readily make assessments on the factual strength of trial evidence, particularly in cases where the fundamental issue is complainant or defendant credibility. This is evidenced by the CACD's decision in R v SB [2019] EWCA Crim 565. The defendant in this case was convicted of sexual offences against his granddaughter. The complainant subsequently retracted her evidence, telling a family member she had fabricated the allegations. On appeal, the CACD rejected the complainant's retraction as ‘not capable of belief’ on the grounds that her trial evidence was compelling (R v SB [2019]). Whilst there were justifiable concerns about her retraction being made to family members, there was no direct evidence the complainant had been pressured to retract her evidence. Whilst it is important that retraction evidence is approached with caution, this decision arguably treads a difficult line when it comes to the CACD's role. This decision is reflective of the court's prevailing approach (Roberts, 2017) to considering fresh evidence appeals where the judges consider the safety of the conviction based on their own judgment of the impact of the fresh evidence (Stafford v DPP [1974] AC 878). However, in a case where the only evidence against the defendant was the complainant's testimony, upholding this as a legally safe conviction seems difficult to defend. Not only does this rest upon the judges own assessment of credibility, which is susceptible to the same flaws as credibility assessments made by the jury, but it seems to stray beyond the review function of the court to amount to a decision by the judges around whether the appellant was, in fact, guilty. 4 Perhaps R v SB [2019] is simply illustrative of what has been considered an increasingly restrictive approach of the CACD to fresh evidence appeals (Roberts, 2017). But it also highlights the tensions which may be involved with further invitation to the CACD to review the factual basis of convictions, particularly where it may allow the CACD to justify its decisions on the same flawed premises of assessments of credibility as that underpinning jury verdicts.
Notably, the CACD has previously shown willing to quash convictions in the absence of fresh evidence or new legal argument where some ‘lurking doubt’ existed in their mind that an injustice had been done (R v Cooper [1969] 1 Q.B. 267: 271). 5 The decision in R v Pope [2013] 1 Cr. App. R. 14 (para 14) is generally considered to have ‘killed off’ lurking doubt as a potential appeal ground (Zander, 2015: 481) but it had previously been used as a ground in certain historical sexual abuse cases. One such case was R v B [2003] EWCA Crim 319 where the appellant had his convictions quashed for alleged sexual offences committed against his stepdaughter around 35 years previously. On appeal, the CACD concluded there was ‘a residual discretion to set aside a conviction if we feel it is unsafe or unfair to allow it to stand’ even where ‘the trial process itself cannot be faulted’ (R v B [2003] paragraph 27). Lord Woolf emphasised the importance of ensuring that no injustice was done to the defendant and that whilst ‘it may mean that some guilty people go unpunished, it is more important that the innocent are not wrongly convicted’ (R v B [2003] paragraph 27). In particular, the court was concerned that the substantial delay left the defendant in ‘an impossible position to defend himself’ and they acknowledged the defence's argument that ‘when faced with allegations of the sort that were made here, “I have not done it” is virtually no defence at all’ (R v B [2003] paragraph 28). The CACD concluded that ‘having regard to the lapse of time, the very limited evidence that was available in this case, we have come to the conclusion that it is our duty to allow this appeal’ (R v B [2003] paragraph 29). The use of the lurking doubt doctrine in this case (and others) led to criticism for its uncertain and subjective basis (Leigh, 2006). However, in defending its use, Leigh (2006: 815) concluded that a principled basis could be identified and the ‘general impression’ of a case ‘need not be restricted to the question whether the jury's verdict or verdicts was or were defensible on the evidence’ but could consider ‘whether the defendant had an opportunity to defend himself, or whether the prosecution evidence at its highest is capable of sustaining the charge’, or if the case, ‘for whatever reason, can safely be left to the jury’. Therefore, if a principled basis for a broader residual discretion could be developed to encourage the CACD to review trial evidence more critically, this could be an important safety net for defendants convicted of very delayed allegations of historical sexual abuse.
There would be clear advantages to focusing potential reform on broadening the grounds for appeal at post-conviction stage. First, as indicated above, the grounds for appeal are currently under review by the Law Commission (2023) and so this area is ripe for reform. Second, given the difficulty in prosecuting such cases, broadening the grounds for appeal would not prevent cases from reaching trial. This may therefore be preferable to the second suggested reform considered in this paper, which looks at whether the court should be more prepared to stay proceedings for abuse of process for delay, because the latter has the effect of stopping the case from proceeding at trial. Focusing reform on appeal grounds would therefore allow cases to continue proceeding to trial as is currently the case but would provide potentially wrongly convicted defendants with a wider safety net. This may go further towards preventing the acquittal of factually guilty defendants because they may be prepared to accept the jury's verdict and not pursue further legal challenge post-trial. However, there are flaws to this proposed reform of broadening appeal grounds in this context if the effect of this is to simply allow the Court of Appeal a broader discretion to set aside the conviction on the basis that the evidence underpinning the conviction is weak, or witness testimony is unreliable, or the delay is too substantial, because these are all issues that should have been dealt with in advance of the trial. If there are arguable points in relation to these issues then the best outcome for all parties would be to stop the case from proceeding at trial stage, because otherwise it causes more distress to the victim(s) and is unfair to the defendant. Therefore, if there are justifiable concerns about the strength of the evidence in substantially delayed claims of historical sexual abuse as claimed here, then it is necessary to consider whether the current approach of the criminal courts to delayed claims is appropriate to protect defendants from wrongful conviction at trial.
Strengthening procedural safeguards: Reconsidering the approach to abuse of process for delay
Another suggestion for reform was put forward by Hickman (2020: 10) who argued that, where a case depends ‘solely on whether the complainant's evidence is reliable,’ then ‘the criminal courts should be far more ready…to stay stale cases as being incapable of being fairly tried, or even to withdraw such cases from the jury.’ This was considered preferable to any reform that would introduce an inflexible limitation period for delayed claims because there will sometimes be a strong evidential case to pursue, for example where the defendant may be willing to admit the truth of stale allegations, or there may be other independent evidence that provides support, such as other independent complainants, or perhaps the defendant may have previous convictions (Hickman, 2020: 10). This section will firstly consider how the abuse of process doctrine has evolved to its current position, before proceeding to explain why this current approach fails to protect those defendants who are most prejudiced by substantial delay.
The evolution of abuse of process for delay in the criminal courts
Historically, staying the prosecution for abuse of process due to delay was an important safeguard for defendants facing historical sexual offence charges, so that even ‘the most minor of potential evidential disadvantage to the defence’ could lead to a stay of proceedings on grounds of unfairness to the defendant (Wood, 2019). This was evident in cases such as R v Telford Justices Ex p. Badhan (1991) where the prosecution depended wholly upon oral testimony about an offence alleged to have been committed 15 to 16 years before. As cited by Knapman (1991: 528), the court stated: any investigation for the purpose of preparing a defence was impossible. There was no corroborative evidence. The case was one in which the Court should infer prejudice ……and conclude that a fair trial would not now be possible.
When deciding whether to stay the prosecution for abuse of process for delay, the court will consider the potential prejudice caused to the defendant where relevant evidence might have been lost or destroyed due to the passage of time. It has been made clear in case law that a stay would only be imposed where the missing evidence is so prejudicial to the defendant that no judicial direction could compensate. A key example of this is R v Sheikh (Anver Daud) [2006] EWCA Crim 2625, where the appellant's convictions for sexual offences allegedly committed 25 years previously were quashed. The CACD concluded the absence of a staff rota and personnel record was significant because these documents could have determined if the defendant had the opportunity to commit the offences as alleged. The importance of the missing evidence was heightened because the offences could only have been committed within a very narrow time frame of two days and that, whilst the complainant claimed the appellant was on duty these particular nights, the defence were able to show there was a real possibility the appellant would have been on leave at this time, but were unable to prove this without the missing records. Hooper LJ said at paragraph 47, ‘In these circumstances we have grave doubts whether a judge who properly analysed the consequences of the missing documents would conclude that the trial was fair.’ Similarly in R v Joynson (Frank) [2008] EWCA Crim 3049 the appellant's convictions for sexual offences allegedly committed over 40 years previously were quashed where it was concluded that missing documents could have settled key points in dispute, including whether the complainant had mistaken the defendant for another staff member who had been convicted for sexual offences. The CACD held that no warning could counter the prejudice caused to the defendant and therefore the convictions were found to be unsafe.
However, subsequent jurisprudence clarified these were not precedents which could be utilised to make broad arguments about the impact of missing evidence. In R v MacKreth (deceased) [2009] EWCA Crim 1849 at paragraph 39, Rix LJ explained that although ‘principles have not changed’, case law since the early 2000s had been ‘astute to pay real and not mere lip service to a concern to do justice in such cases’ and that stays for delay would have to turn on each case's special facts. It was emphasised that previous precedents were not authority for mere speculation about potential missing evidence and that such arguments would be insufficient to justify a stay on grounds of delay (R v MacKreth [2009] at paragraph 39). Similarly, in R v Dent [2014] EWCA Crim 457 at paragraph 47, the court again distinguished the appellant's argument from Sheikh/Joynson, emphasising that in those cases the convictions were unsafe because the missing evidence was capable of being ‘determinative’ about whether the incident could have occurred.
More recent jurisprudence has been considered to narrow the doctrine further with these decisions being described as ‘tantamount to an abolition of the jurisdiction of abuse of process on grounds of delay, save in the most exceptional of circumstances’ (Wood, 2019). The key issue in R v PR [2019] EWCA Crim 1225 was whether it was fair for the case to proceed when evidence pertaining to a previous police investigation into the same offences had been destroyed. The court concluded it was possible for the appellant to have had a fair trial due to other material, stating: 71. It is clear that imposing a stay in situations of missing records is not a step that will be taken lightly; it will only occur when the trial process, including the judge's directions, is unable adequately to deal with the prejudice caused to the defence by the absence of the materials that have been lost…
Examining the current Approach: Can defendants really still have a fair trial?
The current approach of the courts to abuse of process for delay therefore prioritises cases reaching trial. It is possible to understand why jurisprudence has evolved to its current position. Missing evidence due to delay is disadvantageous to both the prosecution and defence (or potentially advantageous to a party which may be fabricating their account). Therefore, whilst this causes deficiencies in factfinding, it is not necessarily unfair when both parties are unable to access the missing evidence. Furthermore, there are additional protections for defendants to help protect them from wrongful conviction, which may go some way towards mitigating potential prejudice from missing evidence due to delay.
First, as alluded to above, the judge is required to give a clear direction to the jury which should emphasise: how delay can place a defendant at a material disadvantage in challenging allegations; that memories may fade and evidence may become unavailable; that the jury should consider the prejudice caused by the delay when considering whether the prosecution has proved the defendant's guilt; and the main elements of prejudice identified during the trial should be highlighted (R v PS [2013] EWCA Crim 992: paragraph 35). However, it is debatable whether jury directions are sufficient to safeguard defendants’ interests. Empirical research has consistently demonstrated that jurors struggle to understand judicial instructions, which has been found in both mock jury studies and studies involving post-trial interviews with real jurors (Barnstein and Hamm, 2012: 49). Furthermore, despite the courts justifying the importance of the delay direction in safeguarding the defendant's rights in historical cases, the CACD has not been consistent in enforcing this on appeal. In R v H (Henry) [1998] 2 Cr App R 161, in reviewing previous authorities, Potter LJ emphasised that each case falls on its own particular facts, and that, whilst it is ‘desirable in cases of substantial delay’ to have ‘some direction’ around difficulties for the defence caused by the delay, ‘such a direction is not to be regarded as invariably required except in cases where some significant difficulty or aspect of prejudice is aired or otherwise becomes apparent to the judge in the course of the trial’ (R v H (Henry) [1998]: 168). This was reaffirmed in R v PS [2013] where the court found that the direction was insufficiently focused on the impact of the delay on the defendant, but nevertheless concluded that it did not amount to a misdirection, noting that the case against the appellant was ‘compelling’ when finding the conviction safe (paragraphs 37–40). Similarly in R v AT [2013] EWCA Crim 1850 the trial judge did not highlight the main elements of prejudice caused to the defence in the direction. The CACD accepted that ‘while the summing-up should…have alluded to the specific elements of prejudice said to have been occasioned to the applicant by reason of the delay, the failure to do so was not such as to render the convictions unsafe’ (R v AT [2013] paragraph 88). Therefore, whilst the complete absence of a delay direction in appropriate cases might be sufficient grounds for appeal, the CACD is less likely to allow an appeal where the delay direction was deficient in some way. But given the emphasis placed upon delay directions as a substitute for abuse of process, the limited grounds upon which the delay direction can be challenged on appeal is a further erosion of protection for defendants in this context.
The second principle safeguard that defendants have which may justify allowing cases of substantial delay to proceed to trial is the burden and standard of proof, which is on the prosecution to prove the case so that the jury can be ‘sure’ of the defendant's guilt. In theory, this should balance out the inequality of power between the state and the defendant in criminal cases to achieve something akin to equality of arms (Ashworth, 2000: 239). Recent jurisprudence from the CACD has implied that missing evidence should be seen as potentially advantageous to the defendant because it allows the argument that the jury should acquit because they cannot be sure of the defendant's guilt (Hungerford-Welch, 2020). But this feels like a weak argument. First, if we presume the defendant is factually innocent of the offences, it is difficult to see how missing evidence can ever be construed as an advantage. It will always be likely that, had the allegations been brought sooner and/or framed in more specific terms, the defendant might have presented evidence that may have disproved the allegations, or at least have helped to undermine them. Second, identifying examples of potential evidence that is now missing but which may have been previously available becomes very challenging when the allegations are framed in broad specimen counts across multiple years because the lack of specificity in the allegations means it is almost impossible to identify anything that could have proven this either way. Even if the defendant can identify examples of potentially important missing evidence that may have been available were the allegations brought sooner, this is unlikely to assist the jury in their role of factfinding because they know that, were it available, it might help either the prosecution or the defence (or neither). This effectively neutralises the evidential value of missing evidence making it difficult to see how it can effectively be deployed to weaken the prosecution's case. Third, whilst the jury is directed to consider the potential prejudice this causes to the defendant, ultimately the jury are factfinders, and it is not their job to assess the fairness of the process. Therefore, whilst it can be argued that allowing cases of substantial delay to proceed is conducive to a fair trial by highlighting the additional safeguards in place to protect against wrongful conviction, it is debatable whether these are sufficient to protect defendants from wrongful conviction when facing historical allegations with substantial delay.
Should the current approach to abuse of process for delay be reformed?
First, it is debatable whether the current approach protects those defendants who are most disadvantaged by substantial delay. As indicated above, currently, missing evidence is unlikely to be grounds for arguing that the defendant cannot have a fair trial unless shown it would potentially be determinative about a particular issue in the case i.e., it would have been significant in proving or disproving that the offence(s) could have taken place. This approach is summarised by Treacy LJ in paragraph 15 in R v RD [2013] EWCA Crim 1592, stating that: ‘In considering the question of prejudice to the defence…it is necessary to distinguish between mere speculation about what missing documents or witnesses might show, and missing evidence which represents a significant and demonstrable chance of amounting to decisive or strongly supportive evidence emerging on a specific issue in the case’. However, this approach excludes a vast majority of defendants in historical sexual offence cases and particularly those who are facing the most significant and extreme delays. In cases dating back decades, it is typical for the allegations to form specimen counts where the charges will be framed as having occurred at some point across multiple years (e.g., an allegation that the complainant was raped several times between a span of years). These defendants are those most disadvantaged by delay because the wide-ranging scope and non-specificity of the allegations, coupled with the delay, makes it impossible to even identify any potential evidence that might have been available nearer the time which could have been determinative or even helpful to examining whether the abuse occurred. This was at issue in the recent case of R v Pipe [2023] EWCA Crim 328 where the complaints dated back to the mid 1960s. The defence had tried to argue for a stay on the grounds there was important material missing, such as records from the school (e.g., including employment records, staff rotas, records about when particular rooms were occupied by certain children, locations and dates of trips and the names of pupils and staff members who would have been involved in those trips), along with now unavailable witnesses. The court rejected these arguments on the basis this was ‘not a case in which the missing material was determinative of any specific issue before the jury’ because the allegations were not date-specific but ‘couched in general terms of sexual abuse occurring on different occasions, in different locations, within broad periods identified in the indictment’ (paragraph 84 in R v Pipe [2023]).
However, it seems counterintuitive to conclude that the vaguer the delayed allegations, the less prejudice is presumed to be caused to the defendant. The court justifies this view on the basis that whilst ‘it is to be regretted if relevant records become unavailable’ the effect is ‘to put the defendant closer to the position of many accused whose trial turns on a decision by the jury as to whether they are sure of the oral evidence of the prosecution witness or witnesses, absent of other substantive information by which their testimony can be tested’ (R v PR [2019]). Whilst it is true there are contemporary claims that ultimately have no corroborative evidence or evidence outside of the testimony of the complainant and defendant, given all the difficulties with cases of this kind, it is debatable whether this should be used to justify disregarding the impact of missing evidence in a case of substantial delay. It is also not true to say that such defendants are in the same position. Defendants facing more contemporaneous claims at least have an opportunity to identify any potentially relevant evidence that might assist the jury, for example, witnesses who may have been around in close proximity to when the offence occurred, or in a case of denial, evidence about the defendant's movements, for example. However, a defendant facing historical charges dating back decades has lost all opportunity to access such potential evidence. Furthermore, in many of these cases, there would have been evidence available that, whilst it might not have been determinative of the allegations, it would at least provide important context for the jury in helping them make their decision. Therefore, under the current approach, the greater the delay and the vaguer the details of the allegations, the more difficult it is for the defendant to argue that they cannot have a fair trial. This seems to fly in the face of earlier judgments, such as R v B [2003] EWCA Crim 319, where the court recognised the extreme difficulties faced by defendants looking to argue simple denial.
A second issue arises around whether the current approach is fair to the defendant in protecting the presumption of innocence. One problematic issue with delay being considered as part of the doctrine of abuse of process is its link to fault. Case law has emphasised that it would only be in rare circumstances that the doctrine would be used to stay the prosecution in absence of some fault on behalf of the prosecution or complainant (Attorney General's Reference (No. 1 of 1990) [1992] QB 630). Delay in bringing the allegations will almost never be the fault of a genuine complainant. There are numerous reasons why victims of sexual offences may take a long time to report offending and, often, the delay in reporting is directly caused by threats or control from the perpetrator (Lewis, 2006: 6–10). This means that when assessing whether the delay is justified, the judge has to do this in a way which requires a presumption of the defendant's guilt (Lewis, 2006: 66). Arguably, the question of fault around the delay in bringing the allegations is largely unhelpful in the majority of historical sexual abuse cases. Instead, it would be better to focus purely on whether the defendant can still have a fair trial. Another factor in the court's approach which raises questions surrounding the presumption of innocence is how the court assesses the impact of the missing evidence upon the defendant. This is again evidenced in the court's recent judgment in R v Pipe [2023]. In rejecting the defence's arguments about the prejudice caused by missing evidence, the court criticised the submissions for being premised upon an ‘assumption that the missing evidence would necessarily have supported Pipe's case, when the reverse may well have been true, as the judge explained’ (paragraph 84 in R v Pipe [2023]). Whilst it will always be impossible to know whether the missing evidence would have supported the defendant's or complainant's case, it is arguable that fairness to the defendant requires assessing the impact of the missing evidence with the presumption of innocence in mind, i.e., what the impact of the missing evidence would be on the presumption that it would have supported the defendant's case.
Therefore, the court's current approach to abuse of process, which considers whether missing evidence would be determinative, is insufficient to safeguard defendants who are most disadvantaged by delay. Instead, the court should be more prepared to accept abuse of process arguments in cases where the defendant is facing substantially delayed allegations which are framed in broad, wide-ranging specimen counts. Here, the broad frame of the allegations means that it will be impossible for defendants to identify any evidence that would be determinative of whether the allegations could have occurred. However, it is possible that had the allegations been brought contemporaneously, not only would the allegations be framed in more specific terms, but there may have been a wealth of evidence that (whilst not necessarily being determinative) could have had contextual importance. It is therefore unfair to say that such defendants are in the same position as many other defendants facing contemporaneous claims. Where the defendant can point to missing evidence which might have assisted in examining the circumstances surrounding the allegations, with the presumption of innocence in mind, the court should give serious consideration to the potential impact of its loss. In assessing whether the defendant is still able to have a fair trial, the court should bear in mind the strength of the evidence against the defendant and to what extent there is other evidence available for the defendant to challenge the allegations. Factors which might be considered important in assessing the above is the number of complainants, any potential connection between them, the existence of previous complaints made against the defendant by the complainant(s) and any other evidence which might towards supporting the allegations. Where the overall case against the defendant is evidentially weak, the court ought to give serious consideration to whether the prosecution should be stayed for abuse of process for delay, if the only defence now available to the defendant is straightforward denial.
Conclusion
To conclude, the current approach of the criminal courts for dealing with historical sexual abuse arguably leaves defendants vulnerable to wrongful conviction. It was argued, in agreement with others (Henry and Gray, 2020 and Hickman, 2020), that the courts’ reliance upon the jury to reach a safe verdict based on their analysis of defendant and complainant credibility is based upon a fallacy, which ignores evidence from research that tells us this is impossible. Whilst this criticism can also be levelled at contemporary claims of sexual offences, substantial delay in many historical sexual offences compounds the problem. The delay in bringing the allegations weakens the probative value of witness testimony even further, and the lapse in time makes it more difficult for either party to identify relevant evidence that might help the jury reach a decision.
Based on these concerns, two potential routes for reform were considered. The first looked at the potential to broaden the grounds under which the CACD consider appeals so that appellants would not be confined to having to demonstrate fresh evidence or legal and procedural errors. It was recognised this would have advantages in providing defendants with an important safety net to protect against wrongful conviction, whilst still allowing cases to proceed to trial. This would potentially make this first reform a favourable option given concerns about the low prosecution rate of sexual offences and the acquittal of factually guilty defendants. However, concerns were raised around inviting the CACD into making factual determinations on trial evidence insofar as it might require the judges to make their own fallible judgements about the credibility of trial witnesses. Furthermore, if, as suggested, the broader appeal grounds are focused on re-examining the strength of evidence against the defendant, it is debatable whether this is simply deferring issues to the appeal court that should have been addressed at trial. Therefore, it was argued that consideration needed to be given to whether the court was currently taking a fair approach to allowing historical cases with substantial delay to proceed at trial, or whether there might be an argument to strengthen the current approach of the courts to abuse of process for delay. It was concluded that the current approach to abuse of process for delay, which focuses on when missing evidence may be determinative, is inadequate because it fails to protect those defendants who are most disadvantaged by delay. Instead, the courts should be more prepared to accept arguments around abuse of process in a broader range of cases, particularly where the delay is so significant that the allegations can only be presented as wide-ranging specimen counts covering multiple years. In such cases it is almost impossible for the defendant to identify any examples of missing evidence that could be significantly important, and in many cases, the defendant will be completely reliant upon simple denial as their only means for defence.
The suggestion in this paper to strengthen the doctrine of abuse of process for delay is likely to be controversial. If the courts become more prepared to accept abuse of process arguments in a wider range of cases, this would have the effect of preventing more historical sexual abuse cases from proceeding at trial. This is understandably not going to be a popular argument in the current climate where there remains significant concern that sexual offences are under-prosecuted. However, whilst these concerns are acknowledged, this should not be used to legitimise the current approach, which arguably puts defendants at risk of wrongful conviction. The concerns flagged in this paper (and elsewhere) about the unreliability of jury verdicts based on assessments of credibility are valid and should not be ignored. Given the difficulties with factfinding in cases of historical sexual abuse, there is a strong argument that we need to shift our attention from a focus on potential factual innocence or guilt to consider whether the trial process is fair and whether there are sufficient safeguards to protect defendants from a miscarriage of justice.
Footnotes
Acknowledgements
I would like to thank the two anonymous reviewers for taking the time to review this paper and for their helpful feedback. I would also like to extend huge gratitude to my colleagues who have read and provided comments on previous versions of this paper including, Professor Stewart Field, Professor Rachel Cahill-O'Callaghan, Professor Julie Price, Dr Dennis Eady and Abdallah Barakat. All errors are my own.
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
