R v Elias Almallah

On a submission of no case to answer under the Galbraith test, a case should ordinarily be left to the jury to assess the quality and reliability of the evidence. Withdrawal is exceptional where the evidence is so weak that a properly directed jury could not convict.

In a circumstantial case, the court must assess the evidence as a whole. A combination of strands may justify a guilty verdict although no individual strand would suffice. A jury direction accurately stating that a defendant has adduced no contradictory evidence does not remove defence challenges made through cross-examination. Directions on expert evidence must be read in the context of the summing up as a whole.

The applicant was convicted by a jury in the Crown Court at Reading of murder and sentenced to life imprisonment with a minimum term of 22 years and nine months. The prosecution alleged that he was the fourth armed attacker in a joint enterprise killing.

The case against him was circumstantial. It relied principally on CCTV comparison evidence, telephone and cell-site evidence, association with the other participants, a fingerprint in the vehicle used, and events after the killing including meetings and departure from the jurisdiction.

He renewed an application for leave to appeal against conviction after refusal by the single judge. He contended that the trial judge should have upheld a submission of no case to answer and, alternatively, that the summing up misdirected the jury about contradictory evidence and expert evidence.

1. The renewed application for leave to appeal against conviction was refused. Neither proposed ground had arguable merit.
2. The trial judge correctly left the case to the jury. Applying the Galbraith test, it is relatively rare for a court to withdraw a case because of the quality of the prosecution evidence. Questions of reliability normally fall for the jury. In a circumstantial case, the correct halfway-stage question is whether the evidence in its totality is capable of justifying a guilty verdict. Multiple evidential strands may safely support conviction even though none would be sufficient alone.
3. There was abundant evidence capable of connecting the applicant to the other known participants and to the vehicle. The proposed appeal concentrated on asserted weaknesses in individual strands, but did not confront the evidence of the meeting after the murder, the subsequent departures from the jurisdiction, and the striking pattern of calls and associations. The fingerprint point was also properly a matter for the jury. The judge was therefore not arguably wrong to reject the submission of no case to answer.
4. The summing up contained no material misdirection. The direction that neither defendant had adduced evidence contradicting, undermining or explaining the prosecution evidence was accurate. It did not deny that the defence had challenged prosecution evidence through cross-examination and by reference to prosecution material. The summing up had fairly reminded the jury of those challenges.
5. The direction on expert evidence was also proper when read in context. The jury were correctly told that they could assess reliability and reject expert opinion for good reason, while resolving any challenge in the ordinary way. The reference to evidence being largely unchallenged created no reasonable basis for complaint or for concluding that the defence case had been removed from the jury’s consideration.

• Court of Appeal (Criminal Division): The renewed application for leave to appeal against conviction was refused: [2025] EWCA Crim 433.
• Crown Court at Reading: The applicant was convicted of murder by a jury on 26 March 2024 and later sentenced to life imprisonment with a minimum term of 22 years and nine months.