In our March 2021 edition of the Criminal Appeals Bulletin, Katy Thorne QC asks: What is the Court of Appeal’s attitude when experts turn out to be not as expert- or as honest - as they claim?. Peta-Louise Bagott analyses the CACD’s decision on the admissibility of EncroChat messaging
Failings of prosecution trial experts as a ground of appeal
R v Byrne and others
 EWCA Crim 107
By Katy Thorne QC
What is the Court of Appeal’s attitude when experts turn out to be not as expert - or as honest - as they claim?
We all know the duties that experts giving evidence in criminal courts are supposed to comply with: to be unbiased and objective, not to stray outside of their expertise and to disclose material and information which may undermine their opinion or standing. These duties are currently set out in Part 19 Criminal Procedure Rules but have been in previous iterations of the Criminal Procedure Rules and set out many times the Courts and by Regulators. But what happens if experts do not comply? What happens when experts are not so expert? What happens when experts are not honest? Unfortunately, it appears, the Court of Appeal has recently been unwilling to interfere.
In the case of R v Byrne and others  EWCA Crim 107, the Court of Appeal gave judgment in seven conjoined appeals which followed the collapse of a major fraud trial (R v Sulley and others) due to the wholesale discrediting of the Prosecution’s carbon credit expert witness, Andrew Ager. In that earlier trial, at Southwark Crown Court, Ager had not just been shown to have questionable qualifications, but not to have complied with, or even understood his duties as an expert, and he had even attempted to improperly persuade a defence expert not to give evidence, in the fear that he, Ager would be exposed. The very senior trial judge found him not to be an expert of “suitable calibre” and the CPS indicated he would never be used as an expert again.
All the cases in the Byrne and Others appeal were previous trials where Mr Ager had given evidence about the carbon credit industry. The Court of Appeal accepted Mr Ager’s wholesale failure in Sulley to comply with the requirements of an expert witness and the trial judge’s conclusion that he was not suitable, but declined to quash any of the convictions. In their judgment, the Court sought to remind all parties of the need to ensure that inappropriate expert witnesses are not called in criminal trials, stating that the guidance issued by the CPS, and compliance with the Criminal Procedure Rules will ensure that rogue experts are weeded out. The Court said:
The Crown must take all necessary steps to ensure that inappropriate expert witnesses are not called in criminal trials in the future. Proper adherence to the two sets of Crown Prosecution Guidance set out in  above, together with the Criminal Procedure Rules and the Criminal Practice Directions, should ensure that this regrettable lapse will not be repeated. The failure to detect the underlying problems with Ager as an expert witness was a notable error on the part of those with conduct of these cases. [emphasis added]
Unfortunately, history tells us that such confidence may be misplaced, because time after time prosecution experts have overstepped their expertise or even turned out to be dishonest, but the same system has not exposed the failings until much later. There are many examples of such experts, from the infamous Sally Clarke case with Professor Roy Meadows and Dr Alan Williams, to the recent example, Saul Rowe in the LIBOR trial R v Pabon , where again the Court of Appeal refused to quash convictions despite finding Mr Rowe had, “signally failed to comply with his basic duties as an expert” indicating,
” there is no room for complacency and this case stands as a stark reminder of the need for those instructing expert witnesses to satisfy themselves as to the witness’ expertise and to engage (difficult though it sometimes may be) an expert of a suitable calibre.’
Mr Ager continued to be instructed by and give evidence for the Crown despite the fact that the Criminal Procedure Rules and CPS guidance had long been in existence, and after the judgment in Pabon, and that Court’s “stark reminder”, without any apparent interrogation of Ager by the Crown, or any change in the Crown’s approach to his evidence.
Practitioners should note that this ruling is yet more reason to try get it right first time: not to accept a prosecution witness’s credentials at face value, to instruct alternative experts and to ensure that they challenge the evidence, even when it appears difficult to do so, and at appeal, similarly, to ask for disclosure and instruct experts to truly examine the accuracy and honesty of the Crown’s case at trial.
However, the concern must be that if there is no effective sanction for the prosecution in cases where they have called an unqualified or unscrupulous expert, and have failed to properly supervise and interrogate them, there is simply no incentive for any prosecutor (public or private) to ensure their experts comply with the CPS guidance or the Criminal Procedure Rules or the Forensic Science Regulator’s latest Code of Conduct. Without convictions being quashed, the guidance, rules and codes are effectively toothless and, whatever is said by the Court, history will, inevitably, repeat itself.
If you would like to speak to Katy Thorne QC about this article, please click here.
About Katy Thorne QC
Katy Thorne QC has a long and varied appellate practice which she combines with her trial practice in homicide, fraud and sexual crime. She is an expert on challenging expert evidence and is also an Assistant Coroner and an editor of Mason’s Forensic Medicine for Lawyers. The combination of trial and appellate practice and her expertise in expert evidence provide unrivalled perspective and experience to successfully assist both privately funded and legally aided clients alike. She appeared in Byrne and Others.
To see Katy’s full profile, click here.
Think twice before sliding into EncroChat DMs
A, B, D, & C v R
 EWCA Crim 128
By Peta-Louise Bagott
EncroChat was a private messaging service that promised its users secure and encrypted communications. The use of the past tense is perhaps now more appropriate following the Court of Appeal’s judgment in A, B, D & C v R.
The Court was asked to consider whether material obtained from the EncroChat application was admissible at trial. The appeal turned on whether communications had been intercepted while ‘being transmitted’ by the device or while ‘stored’ on it. As the material had been extracted from the device itself and was unencrypted, the Court found that the evidence was admissible.
The public policy arguments in favour of allowing EncroChat material before courts are obvious. For that reason, many will find the decision unsurprising. But, on a practical level, where does that leave future defendants facing proceedings on the basis of EncroChat material?
The Court of Appeal’s ruling
‘Being transmitted’ vs ‘stored’
In this case, the material was intercepted by the French and Dutch Joint Investigatory Team (‘JIT’) through the EncroChat communications system. The French Gendarmerie sent an undercover implant to all EncroChat devices disguised as an ‘update’ for the system.
The implant worked in two stages. Stage 1: it caused the device to transmit all of the data that had not been erased to the French police digital crime unit server (C3N) and then to Europol. Stage 2: it continued to collect data that was created after the device had been ‘infected’ by the update.
The central issue before the Court was the construction of s.4(4) Investigatory Powers Act 2016 (‘IPA’). The distinction between material ‘being transmitted’ (s.4(4)(a)), and material being ‘stored’ (s.4(4)(b)) is significant. It was agreed by the parties that s.4(4) should not be construed as meaning that if a communication was ‘being transmitted’ it could not be ‘stored by or in the system’ at the same time .
Within the IPA itself there is a specific provision prohibiting the admission of material intercepted in the course of its transmission: s.56. As a result, if EncroChat material were found to have been intercepted while ‘being transmitted’ (s.4(4)(a)), s.56 would have operated to prevent the material from being admitted at trial.
The Appellants argued that the material had indeed been intercepted while ‘being transmitted’ so fell within s.4(4)(a) and not s.4(4)(b) [34 – 37]. There were a number of technical points raised by the Appellants, supported by expert evidence, in this regard.
The Crown responded, arguing that the material had been intercepted when it was ‘stored’ in accordance with s.4(4)(b): s.56 did not apply and the evidence was admissible, subject to any other challenge.
Despite numerous technical arguments being raised by the Appellants about storage, the Court found that the intercepted material had been ‘stored’ and was not intercepted while ‘being transmitted’ .
The Court’s interpretation of s.4(4)(b) is set out in paragraph 62:
As a matter of ordinary language, section 4(4)(b) is clear and unambiguous in its meaning. It extends to all communications which are stored on the system, whenever that might occur. That broad meaning coheres with the structure of the 2016 Act considered in overview, and importantly with the different types of warranty for which the Act provides. Part 5 warrants are required for the interception of stored material, and Part 2 warrants for material which is to be intercepted while being transmitted. It also advances the overall purpose of the legislation in preserving the legislative framework – and the distinction between the different types of intercept – to which we have referred. The statutory question for any court in determining section 4(4)(b) applies is this: was the communication stored in or by the system at the time when it was intercepted?
Particular attention was paid to the fact that the material was recovered in the form of unencrypted messages ‘stored’ on the device. Had a message been intercepted while ‘being transmitted’, it was found that it would had been encrypted [66, 14].
It concluded that:
the communications were stored on the handset;
the EncroChat material was not caught by the s.56 exclusion;
the material was lawfully intercepted while it was stored on the handsets; and
the material was admissible [66-67, 69].
For proceedings in the future seeking to rely on extracted material from a device, providing the Prosecution can show that the material is ‘stored’ somewhere on it, any arguments made on the basis of s.4(4)(a) are unlikely to get off the ground.
Supported by expert evidence, the Appellants argued that the communications were extracted while ‘being transmitted’ on the basis that transmission starts when the user presses ‘send’ and ends when the communication is accessible by a human recipient .
The arguments made about how a device stores data within Random Access Memory (RAM) and Read-only Memory (ROM, or Realm for EncroChat devices) were dismissed. The Court endorsed the Judge’s finding that the communications were extracted directly from the handset of the user and not while they were in transit to, through, or from any part of the system. It explained that ‘the material was recovered in the form of unencrypted messages stored in the RAM of the device in a form in which they existed before they were transmitted from the decided to the servers’ . For this reason, the intercepted material was ‘stored’ when it was intercepted and fell within s.4(4)(b).
Going forwards, raising these points again on the basis of expert evidence is likely to be met with the following rationale:
The 2016 Act does not use technical terms in this area. The experts have an important role in explaining how a system works, but no role whatever in construing an Act of Parliament. They appear to have assumed that because a communication appears in the RAM as an essential part of the process which results in the transmission it did so while “being transmitted”. That is an obvious error of language and analysis .
The judgment does not bode well for future technical arguments. If the communications can be ‘transmitted’ and ‘stored’ at the same time, then providing the unencrypted material is somewhere on the device a court is likely to fall back on s.4(4)(b).
Exclusion and abuse
The scope of the appeal did not include the Judge’s decision to reject the application to exclude the EncroChat material under s.78 Police and Criminal Evidence Act 1984 and stay the proceedings as an abuse of process [4-5]. The effect of the Court of Appeal’s reasoning is that any challenges to the admissibility of EncroChat material under the IPA are unlikely to get very far. However, these avenues remain open for issues separate and distinct from this.
The decision will be celebrated by the National Crime Agency and the Prosecution as the precedent set will make it harder to challenge the admissibility of EncroChat material intercepted by intelligence agencies. Unaffected by this judgment are the common evidentiary challenges in relation to attribution, continuity, and other deficiencies in the Prosecution’s case. Whether that is enough to overcome the impact of having EncroChat material before the jury will depend on the facts of the case.
One unencrypted message to screenshot from this judgment, is that communications arising from sliding into EncroChat DMs are likely to find their way before a jury one way or another.
If you would like to speak to Peta-Louise Bagott about this case, please click here.
About Peta-Louise Bagott
Peta-Louise is an experienced advocate in both domestic and international criminal law, and professional discipline. Within crime, Peta-Louise has particular experience dealing with complex, document-heavy cases involving telephone and cell-site evidence. She is equally adept at acting as a led junior or as a junior alone in cases ranging from complex frauds to organised crime and terrorism offences.
To see Peta-Louise’s full profile, click here.
This article was a part of our monthly Criminal Appeals Bulletin, edited by Paul Taylor QC and Farrhat Arshad, which aims to highlight recent changes in case law and procedure in England and Wales, Northern Ireland, the Caribbean and Hong Kong and to provide practical guidance to those advising on appellate matters. Our monthly case summaries illustrate when an appellate court is likely to interfere with conviction or sentence, as well as looking at the courts’ approach to procedural matters. The featured article provides an in depth commentary on a current appeal topic.
To receive our monthly appeals bulletin direct to your email, sign up here.
 Part 19 Criminal Procedure Rules , Criminal Practice Directions, Crown Prosecution Service Guidance for Experts on Disclosure, Unused Material and Case Management (updated 30 September 2019) and the Crown Prosecution Expert Evidence Legal Guidance (updated 9 October 2019).
 Eg R v Harris  EWCA Crim 1980,  1 Cr App R 5.
 Code of Practice and Conduct For Forensic Science Providers and Practitioners in the Criminal Justice System.