Almost 30 years have passed since Parliament enacted the Criminal Procedure and Investigations Act 1996 (CPIA 1996), but the challenge of securing a fair and proportionate system for the handling of unused material in criminal cases has remained undimmed. There is broad agreement that the structural foundation set out in the legislation is sound, but there are severe problems in its application, especially where digital material is involved.

The legislation in 1996 was promoted by a concern that defendants had been wrongly convicted in several high-profile trials after the prosecution had failed to disclose critical information which undermined the prosecution case. One high watermark of injustice arose in R v Ward (Judith) [1993] 1 WLR 619 when in 1992, after serving 17 years’ imprisonment for multiple murders in an IRA bomb attack, Ward’s convictions were quashed after it emerged that government forensic scientists had deliberately withheld information at her trial in 1974 which strongly indicated her innocence.

CPIA 1996 was preceded by a Royal Commission on Criminal Justice (CM 2263, HMSO, 2 June 1993) established to consider changes in the way in which the prosecution managed material generated in a criminal investigation. Chaired by Viscount Runciman, the Commission recommended the codification of the prosecution obligation to disclose unused material in primary legislation, with detailed procedures to be governed by subordinate legislation (report, paras 53, 55). In this way, Runciman hoped that future injustices caused by non-disclosure of unused material would be avoided.

Unfortunately, Runciman hoped in vain, as the spectre of injustice brought about by failures in the regime for disclosure of unused material continue to scar the criminal justice system. At the present time, the ramifications of the Post Office scandal involving the use of unreliable evidence produced by the Horizon programme is the subject of multiple investigations. Non-disclosure of unused material features as a significant factor in the wrongful conviction of hundreds of sub-postmasters.

To date, the Court of Appeal (Criminal Division) has overturned 101 convictions. In the first of these cases, Hamilton & Ors v Post Office Ltd [2021] EWCA Crim 577, the court concluded that the investigative and disclosure failings of Post Office Ltd were so egregious as to make the prosecution of any of the “Horizon cases” an affront to the conscience of the court’ (at para 137).

Notoriously, non-disclosure also played a part in Malkinson (Andrew) v R [2023] EWCA Crim 954 when the Court of Appeal quashed a conviction for rape after the defendant had served 17 years of his life sentence.

Unused material in the digital age

To compound worries brought about by injustices flowing from the non-disclosure of unused material, there are additional concerns about the efficacy of CPIA 1996 as it is applied in Magistrates’ Courts and Crown Courts alike.

One of these concerns focuses on the ability (or otherwise) of investigating and prosecuting authorities to manage large volumes of unused material in a digital age.

It is against this background that in October 2023, the Home Secretary appointed me to lead an independent review into how the disclosure regime is working. To quote directly from my terms of reference in relation to the first part of the review, I am tasked to ‘investigate the application of the disclosure regime and the challenges arising for the investigation of all crime types including fraud that handle large volumes of digital material’.

As a second part of the review, I am asked to ‘assess whether the nature of current fraud offences meet the challenges of modern fraud, including whether penalties fit the crime … [and] … if certain fraud offences should be summary only rather than triable either way’. I shall return to this part of the review in the autumn.

Interestingly, the problems presented by the management of digital evidence are not as new as might be first thought. The CPIA 1996 statutory drafters must have been aware of problems presented by the management of substantial amounts of digital evidence because Runciman had foreshadowed it in his report. Runciman noted a representation from the police that where they set up large databases in the course of their investigation, it was not always possible to ensure the listing of all material generated in an investigation. This would be the case even where there were computerised logs of all information collected during an investigation (report, para 41). In other words, even at this early stage in the electronic resolution, the police signalled concerns that they might not be able to cope with a large amount of unused material.

In the years since Runciman reported, the amount of digital material generated in a case has grown exponentially, especially in the prosecution of serious crime cases involving fraud. In written evidence to the House of Commons Home Affairs Committee in October 2023, the Serious Fraud Office (SFO) recently explained that its case teams deal with significant volumes of digital material. The average SFO case opened in 2010 would involve around 2 million documents (350 GB), whereas by 2017 the average size had risen to 6 million documents (850 GB). To date, the largest case on the SFO system has 48 million documents (6.5 TB or 6,500 GB).

The challenges encountered when dealing with unused material are not limited to SFO cases. The handling of unused material in rape and serious sexual offences cases (RASSO) as well as other serious criminal cases present similar challenges which must be met.

As the House of Commons Justice Committee (Disclosure of Evidence in Criminal Cases, 11th Report of Session 2017-19, HC 859, 20 July 2018, para 52) noted a few years ago, ‘police say that the average UK home contains 7.4 digital devices’, and ‘there are also the devices we interact with – bank cash machine ATMs, shop sale systems, restaurants, transport payment systems, when we use public wifi […] when we get caught on CCTV’.

One barrister told the Committee, ‘it is not a digital footprint; it is a digital crater’, explaining in detail that a single phone can tell you ‘what time [the user] woke up because they have an alarm app […] what they had for breakfast because they have a health app […] what they put in their satnav, where they went, what time they got there, potentially how fast they drove, where they parked and what they had for lunch. If they go to a bar […] a taxi app might show what time they left.’ In this sense, since the Runciman Commission, much has changed at the coalface of criminal investigation.

The digital experience has foreshadowed a brave new world, and with the advent of artificial intelligence programmes, the management of unused material in criminal cases is at the start of a new chapter. AI is not a silver bullet which will resolve all issues, but it is not forlorn to hope that the application of new electronic search and advanced review platforms will assist in resolving some of the challenges which the handling of high-volume digital material has brought in its wake.

I intend to publish my provisional findings and a direction of travel ahead of delivering my recommendations to the Home Secretary in the summer of 2024.