Aziz v R: Refusing Adjournment for Further Encrochat Expert Work and Managing Prosecution Closing Submissions Where an Accused Is Unrepresented

1. Introduction

Aziz v R ([2026] EWCA Crim 127, 20 February 2026) concerned an appeal against conviction by Danyal Aziz (“the appellant”) following a large-scale prosecution alleging conspiracies to supply Class A drugs and possession of prohibited firearms and ammunition, with significant reliance on Encrochat/Operation Venetic material.

The appellant was convicted at Birmingham Crown Court (1 June 2023) of multiple counts including: conspiracy to supply heroin and cocaine, possession of firearms with intent to endanger life, possession of prohibited firearms, possession of ammunition without certificate, and concealing criminal property. On 13 October 2023 he received a total sentence of 33 years’ imprisonment.

Two grounds were permitted to proceed:

  • Adjournment/refusal: whether HHJ Rochford’s refusal on 17 April 2023 to adjourn the trial (to allow further defence expert work on Encrochat reliability) rendered the trial unfair and convictions unsafe.
  • Closing speech: whether the prosecution improperly addressed the jury about the case against the appellant (a litigant in person), without leave under CPR 25.9(2)(j), and in terms said to be unfair.

2. Summary of the Judgment

The Court of Appeal dismissed the appeal. It held:

  • The refusal to adjourn did not render the trial unfair or undermine safety. The appellant already had the benefit of existing defence expert reports; any further expert work was, on the material before the court, largely speculative as to what it would add. The appellant could also have cross-examined the prosecution expert and, in practice, obtained the benefit of relevant cross-examination conducted for a co-accused.
  • Although the prosecution should have sought the judge’s leave regarding references to the unrepresented appellant in closing, the references made did not unfairly “give prominence” to the appellant beyond what was necessary in a multi-handed trial where he was alleged to be central. The lapse did not render convictions unsafe.

3. Analysis

3.1 Precedents Cited

(a) R v A [2021] EWCA Crim 128

This authority was used to situate the Encrochat issue historically and legally. In R v A [2021] EWCA Crim 128, the court held that because the Encrochat material was being stored (rather than transmitted) when intercepted, the statutory exclusion under the Investigatory Powers Act 2016 did not bar admissibility. In Aziz, the Court of Appeal treated Encrochat admissibility as “now well established” and therefore not the live battleground; the dispute shifted to reliability and attribution.

(b) R v Atkinson [2021] EWCA Crim 1447

Cited alongside R v A [2021] EWCA Crim 128 for the proposition that admissibility of Encrochat material is settled at appellate level. This context mattered because it narrowed what the defence expert evidence could realistically achieve: not exclusion, but at most evidential weight (reliability/attribution), assessed with other evidence.

(c) Tahir & Simpkins [1997] Crim L R 837

This case was invoked on the closing-speech ground. The appellant relied on it to argue that where one accused is unrepresented, the prosecution’s closing should focus on the case against the represented accused. The Court of Appeal accepted the underlying concern—especially the need for the prosecution to raise the issue with the judge—but rejected the contention that the references in this particular multi-handed trial were unfair in extent or effect.

(d) Michael Earp’s appeal: [2025] EWCA Crim 546 and the appellant’s renewed leave decision: [2025] EWCA Crim 605

These decisions were not substantive precedents on Encrochat, but they framed the procedural posture: leave was limited to two trial-management/fairness issues, and sentence was not reopened.

3.2 Legal Reasoning

(A) Refusal to adjourn: fairness, expert evidence, and speculation

The appellant’s case was that he was deprived of expert support because his expert (Duncan Campbell) had not completed “counterpart” analysis on ten additional Encrochat packages. The Court of Appeal’s reasoning can be distilled into several linked propositions:

  1. Define the real defence: The Defence Case Statement was largely an admissibility challenge. Beyond a bare denial, the defence did not particularise what messages were disputed or how other non-Encrochat evidence (CCTV/ANPR, seizures, DNA) was to be answered. This mattered because the claimed necessity for further expert work was not anchored to a concrete, particularised factual contest.
  2. Existing expert material already provided the “principle”: By April 2023 the defence had Campbell’s report (30 June 2022) and joint expert material. Campbell had already identified large “missing counterpart” rates (including the striking “more than 97%” non-correlation between Lushmace and KneeTown) and had articulated the reliability concern he wished to explore. The trial judge’s view— upheld on appeal—was that the foundational point (that counterpart gaps existed) was already demonstrable without examining ten further packages.
  3. “Missing counterpart” does not equal unreliability: The Court emphasised that even in joint expert material (included in Campbell’s report), it had been accepted that “the lack of a counterpart message does not make the message unreliable.” That substantially weakened the claim that more counterpart testing would likely yield materially exculpatory evidence.
  4. Attribution evidence was strong and largely independent: The judgment recited a detailed attribution case linking “Lushmace” to the appellant (personal events, travel records, cell site overlap with a phone tied to him, the handle giving his number ending 4310, instruction to Google “Danyal Aziz”, birthday message, and the 2016 firearms article “about me”). The Court held that the defence expert’s concerns did not “significantly undermine” attribution.
  5. Opportunity to test reliability at trial existed: The appellant could have (i) called Campbell (whose evidence existed in reports and who had indicated he would attend), and/or (ii) cross-examined the prosecution expert (Luke Shrimpton). In practice, cross-examination by counsel for a co-accused took place and the judge directed that it applied equally to the appellant.
  6. No post-trial expert “fresh evidence” was pursued: The Court considered it telling that, despite private funding and continued availability of the data packages, no further expert work was produced on appeal to demonstrate what the adjournment would have achieved. The Court treated the asserted benefit of an adjournment as speculative absent such material.

The result is a practical principle: an adjournment is not required for fairness merely because further expert work is desired; where the defence already has material establishing the “underlying principle” and cannot show (beyond speculation) what additional, case-changing evidence will follow, refusal to adjourn will not ordinarily render convictions unsafe—particularly where other evidence strongly supports attribution and guilt.

(B) Prosecution closing speech where the appellant is unrepresented

The Court drew a clear distinction between (i) procedural propriety, and (ii) appellate safety:

  • Procedural propriety: The Court accepted that “clearly” the prosecution should have raised with the judge the extent to which it proposed to refer to the appellant once he became unrepresented, and (as conceded) that leave ought to have been sought for the relevant passages.
  • Safety assessment in context: The trial was multi-handed; the appellant was alleged to have a leading role; and the prosecution faced real difficulty explaining the evidence against represented co-accused without referring to him. The Court also considered the appellant’s late self-representation to bear “hallmarks” of attempted manipulation, a factor that would have informed any trial judge ruling.
  • Extent and effect: Having reviewed the transcript, the Court held that although “less could have been said” in places, the references did not unfairly tilt the case against the appellant and did not undermine safety.

3.3 Impact

(1) Trial management in Encrochat “reliability/attribution” disputes

The judgment reinforces a management approach in post-R v A [2021] EWCA Crim 128 litigation: once admissibility is settled, reliability challenges must be framed with specificity and anchored to demonstrable forensic issues that could realistically affect attribution or weight. Courts are likely to be sceptical of open-ended requests for more Encrochat packages where the defence cannot articulate what the additional material is expected to prove and why existing material is insufficient.

(2) Adjournments for expert evidence: emphasis on materiality and non-speculation

Aziz illustrates that appellate courts may uphold refusal of an adjournment where: (i) the defence has had substantial time, (ii) some expert evidence is already available, (iii) the incremental value of further work is unclear, and (iv) the surrounding evidence is strong. The decision also signals that if a defendant claims “we needed more expert work,” the credibility of that claim may be tested by whether the defendant later produces the work as potential “fresh evidence.”

(3) Prosecutorial submissions in multi-handed trials with an unrepresented accused

The judgment does not dilute the expectation that the prosecution should seek directions/leave when addressing the position of an unrepresented accused. However, it indicates that failure to do so will not automatically lead to a successful appeal: the appellate focus remains on whether the closing speech, viewed in the whole trial context, created unfairness sufficient to render convictions unsafe.

4. Complex Concepts Simplified

  • Encrochat / Operation Venetic: Encrochat was an encrypted communications system used by criminals. Operation Venetic was the UK law-enforcement operation that obtained and exploited Encrochat data.
  • “Handle” attribution: A “handle” (e.g., “Lushmace”) is a code-name within Encrochat. “Attribution” means proving which real person used that handle. In this case attribution relied on personal-life references, travel evidence, phone number linkage, cell site correlation, and other confirmatory material.
  • Counterpart messaging analysis: A technique that checks whether a message shown on one device appears correspondingly on another participant’s device. A missing counterpart might reflect incomplete capture rather than fabrication or misattribution.
  • Section 8 CPIA 1996 applications: Defence applications seeking disclosure of unused prosecution material that may assist the defence or undermine the prosecution.
  • Adjournment: Postponing a trial date. The legal question on appeal is whether refusing an adjournment made the trial unfair and convictions unsafe, not whether an adjournment might have been convenient or preferable.
  • Litigant in person: A defendant who represents themselves. The court must ensure fairness, but self-representation does not automatically curtail the prosecution’s ability to present its case in a multi-defendant trial.
  • CPR 25.9(2)(j): A Criminal Procedure Rules provision referenced as requiring the court’s leave for certain steps; here, it was relied on to argue the prosecution should have obtained permission before addressing the appellant’s case once he was unrepresented.
  • “Unsafe convictions”: The appellate standard in criminal appeals. Even if there was an error, the question is whether it undermines the safety of the verdict.

5. Conclusion

Aziz v R [2026] EWCA Crim 127 is a notable Encrochat-era case on practical trial fairness rather than doctrinal admissibility. The Court of Appeal confirmed that:

  • A refusal to adjourn to obtain further expert evidence will not render a trial unfair where the defence already has expert material, the additional work is not shown to be materially outcome-changing (and remains speculative), and attribution is strongly supported by independent evidence.
  • In a multi-handed trial, the prosecution may properly refer to an unrepresented accused where necessary to explain the case against represented co-accused, though the prosecution should raise the issue with the judge and seek appropriate leave; failure to do so is not automatically fatal if the overall trial remained fair.

The decision therefore strengthens the message that Encrochat reliability challenges must be concretely developed and that appellate courts will look for demonstrable, not hypothetical, prejudice when evaluating adjournment refusals and closing-speech complaints.