Relevance-Linked Admissibility of Autism Evidence and Limits on Mid‑Trial Adjournments: R v Smilginis [2025] EWCA Crim 1261

Introduction

This commentary examines the Court of Appeal’s decision in R v Smilginis [2025] EWCA Crim 1261, a multi‑handed kidnapping and serious violence case arising from events in December 2022 in Staffordshire. The central appellate issue was whether the trial judge’s refusal to adjourn—so the defence could obtain and adduce expert evidence about the appellant’s autism—rendered the convictions unsafe. The case provides important guidance on three interlocking questions: when neurodiversity evidence (here, autism) is relevant and admissible; how courts evaluate late, mid‑trial applications for adjournment to secure expert evidence; and the approach to fresh evidence applications under section 23 of the Criminal Appeal Act 1968 in this context.

At trial, the prosecution alleged that the appellant, Evaldas Smilginis, travelled from London to participate in orchestrated kidnappings and assaults on two men, Struiga and Gulbinas, with the events filmed on mobile phones. The defence case was a denial of participation or knowledge: the appellant said he remained in his car, saw nothing, and later forwarded videos without viewing them. During the appellant’s evidence, defence solicitors formed concerns about autism and later sought an adjournment to obtain an expert report from Professor Baron‑Cohen. The judge refused. The jury convicted the appellant on all counts. On appeal, the appellant argued that the judge’s refusal unfairly prevented him from adducing relevant expert evidence supportive of his account.

Summary of the Judgment

The Court of Appeal (Criminal Division), per the Vice President, dismissed the appeal. The Court:

Accepted that a diagnosis of autism had been obtained, and, assuming in the appellant’s favour that certain aspects of the expert opinion might in principle be admissible, held that the evidence was at best marginal to the real issues in the case.
Upheld the trial judge’s refusal to adjourn. Adducing the expert evidence would have required a counter‑expert for the Crown and risked substantial disruption to a multi‑handed trial, without materially advancing a real issue (because the defence case did not assert manipulation or naivety about a plan).
Refused to admit the expert report as fresh evidence under section 23 of the Criminal Appeal Act 1968; in any event, the convictions were safe given the strength of the prosecution case, including video evidence strongly implicating the appellant as the principal filmer and participant.

The Court articulated a clear principle: relevance (and therefore admissibility) of neurodiversity evidence must be tied to the actual, litigated issues as defined by the defence case. Expert evidence cannot be used to invite the jury to hypothesise an alternative version of events that the defendant himself did not advance.

Background and Procedural Narrative

Factual matrix

In December 2022, arrangements were made by a co‑accused, Stoskus, to meet victim Struiga at a McDonald’s in Chesterton. The prosecution case was that this was a pretext to lure the victims to Meadow Street, a dead‑end industrial area, for a coordinated attack by a group that included the appellant. Messages showed the appellant required money to travel (a £100 transfer was made), and that he and another man travelled from London to Staffordshire. Several others converged on the scene from elsewhere.

At Meadow Street, both victims were forcibly removed from a car, pepper‑sprayed, tied up, and assaulted; one victim (Struiga) suffered severe injuries, including multiple stab wounds and a fractured leg, and was dumped still bound. Parts of the attacks and interrogation were filmed on mobile phones. The prosecution argued that, save for one clip, the appellant recorded the videos—his clothing on the footage matched earlier CCTV. The Crown also relied on post‑offence communications, including the appellant sending videos to co‑accused, and efforts to avoid number plate identification in shared footage. The appellant refused to provide his phone access code.

Defence case at trial

The appellant, a Lithuanian national, gave evidence with an interpreter. He denied participation, saying he travelled to reconcile with an old friend, waited in his car at Meadow Street without witnessing any assault, then went to Birmingham to dine with co‑accused and later drove a passenger to Stansted. He said he did not make any recordings; rather, he forwarded videos sent to him from a foreign number on the instruction of a co‑accused, without viewing them. There was no supporting phone download (the device was locked).

The autism issue emerges mid‑trial

During the appellant’s evidence (Thursday/Friday), his solicitor, herself a Lithuanian speaker, emailed the court raising serious concerns that he might be autistic, seeking permission to instruct an expert during an already scheduled non‑sitting week. No adjournment was sought at that point. The appellant completed his evidence. After the break, the defence served a report from Professor Baron‑Cohen (based on remote interviews with the appellant and discussions with family). The report concluded the appellant was autistic, and that autism might make him more trusting, less questioning, poor at “seeing the big picture,” and vulnerable to exploitation—potentially bearing on why he might have been manipulated into waiting near the scene and forwarding incriminating videos.

The Crown disputed the diagnosis and/or its significance, and indicated the need to obtain responsive expert evidence. The trial judge refused the defence application for an adjournment to accommodate that.

Verdicts and appeal

The appellant and one co‑accused (Saviokas) were convicted on all counts; another co‑accused (Seputis) was acquitted on all counts. The appellant appealed, seeking to introduce the autism report as fresh evidence (Criminal Appeal Act 1968, s.23) and arguing the judge’s refusal to adjourn was a wrongful exercise of discretion that rendered the convictions unsafe.

The Court’s Analysis and Decision

Key holdings

Relevance must be anchored to real, live issues: The Court emphasised that admissibility of expert evidence turns on its relevance to issues truly in dispute. Here, the defence case was categorical: the appellant had no knowledge of any plan and took no part; he did not even see events. It was not his case that he knew of a plan but, due to autistic traits, was manipulated or naïve about it. Autism evidence could not be deployed to infer an alternative, unpleaded narrative that he was misled into participation.
Marginal probative value does not justify disruptive adjournment: Even assuming diagnosis and some general traits (trusting, less questioning, difficulty with “big picture”) could be adduced, their probative value on the real issues was “at best marginal.” A mid‑trial adjournment in a multi‑handed case—particularly where the Crown would need to commission a counter‑expert—was properly refused.
Fresh evidence and safety of convictions: For appeal purposes, the Court assumed in the appellant’s favour that the report would not be excluded for reasons other than relevance, but nevertheless found it would not have altered the outcome. The prosecution case was strong, including clothing‑matched video evidence said to show the appellant both filming and physically interacting with a bound victim. The convictions were safe.

Why the autism evidence was treated as marginal

The defence theory never posited manipulation or exploitation. The report could not retro‑fit a different case on appeal or in mid‑trial to change the factual issues for the jury.
The credibility challenge the appellant faced was not just that his story sounded naïve; it was that, on the Crown’s case, he was an active participant, including recording and engaging physically with the victim, and circulating videos while discussing concealing the car registration. Autism traits do not bear on whether the jury could be sure the appellant was the filmer.
The Court found it implausible that an organiser intent on kidnap and violence would bring non‑participants to a secluded scene simply to sit in a car—undermining the defence’s central narrative. Autism traits could not address that implausibility.

The trial judge’s discretion to refuse an adjournment

The chronology mattered. Concerns were raised during evidence, but no adjournment was sought then. The report arrived only after the evidential phase, and by then the Crown would fairly require time for a responsive expert. In a three‑defendant trial, any adjournment risked substantial disruption and potential severance (which the Court considered unrealistic). Against that procedural backdrop, and given the marginal relevance of the proposed evidence, refusal of an adjournment was both open to the judge and correct.

Fresh evidence under section 23 Criminal Appeal Act 1968

The Court acknowledged the defence’s attempt to regularise matters via a “Gogana‑type” explanation (a solicitor’s statement accounting for late emergence of the issue). Even taking the report into account, however, the Court held it would not have changed the jury’s verdict in light of the strength of the Crown case. The safety of the convictions was not undermined.

Precedents and Framework

The judgment does not cite earlier authorities expressly. Nonetheless, the reasoning coheres with established principles that:

Expert evidence is admissible only where it is relevant to facts in issue and likely to assist the jury beyond their ordinary experience. It cannot be adduced merely to bolster credibility or invite speculation about unpleaded factual scenarios.
Case management decisions (including adjournments) lie within the trial judge’s discretion. The appellate question is whether the discretion was exercised properly, balancing fairness, relevance, probative value, prejudice, and the practical realities of a multi‑handed trial.
On appeal, fresh evidence will be admitted if it is credible, could not with reasonable diligence have been obtained at trial, and might reasonably have affected the jury’s verdict; even then, the ultimate question remains whether the conviction is unsafe.

Although not cited in terms, the Court’s approach aligns with long‑standing limits on psychiatric/psychological evidence being used simply to support a witness’s truthfulness and the emphasis on connecting such evidence to a live, material issue that lies beyond the jury’s common knowledge. The decision also resonates with appellate deference to trial management choices where late evidence would cause significant disruption and require responsive expert instruction by the opposing party.

Legal Reasoning Explained

1) Relevance is defined by the live issues, not by hypothetical alternatives

The Court insisted on a disciplined sequence:

Identify the real issues as defined by the parties’ cases.
Ask whether the expert evidence logically bears on those issues.
If relevance is weak or contingent on an unadvanced theory, exclude it.

The defence case was an outright denial of knowledge and participation. Autism evidence about suggestibility, trust, or poor holistic processing might be relevant if the defence had been that the appellant knew something unusual was happening but, because of autism, misinterpreted, minimized, or failed to challenge it. But that was not the case run. The expert report could not legitimately be used to rehabilitate credibility by positing an alternative factual scenario that the appellant himself disavowed.

2) Proportionality and fairness in adjournment decisions

The judge had to weigh:

Marginal probative value of the proposed evidence.
Inevitable need for a prosecution counter‑expert, and the time required.
Disruption to a multi‑handed trial and co‑defendants.
The timing: the application followed completion of the evidential phase.

Given these factors, and the tenuous connection to the issues actually joined, the refusal to adjourn was a measured exercise of discretion.

3) Safety of conviction assessment

Even if one assumed arguendo that the expert evidence should have been before the jury, the Court concluded the convictions remained safe. The prosecution case included strong circumstantial and direct evidence: travel arrangements and coordination; the contemporaneous sending of videos; instructions to obscure number plates; refusal to unlock the phone; and, critically, video imagery said to show the appellant as the filmer and physically interacting with a bound victim (placing and kicking a foot on the victim’s head). Against that backdrop, the autism evidence would not reasonably have affected the verdict.

Impact and Significance

Doctrinal impact

Clarifies that neurodiversity evidence (autism, ADHD, etc.) is not automatically admissible to explain “why a defendant might do unusual things.” It must be tied to the actual issues raised by the defence and cannot serve as a vehicle for speculative alternative narratives.
Strengthens the principle that late, mid‑trial adjournments to obtain psychological or psychiatric evidence will be refused where probative value is marginal and where the disruption to a multi‑handed trial would be substantial.
Reaffirms the appellate approach to fresh evidence: even credible late‑obtained reports will not disturb convictions if they do not materially affect the overall strength of the case and the issues litigated at trial.

Practical guidance for future cases

Early identification: Defence teams should proactively screen for neurodiversity and, where relevant, obtain expert input before trial. Delay risks exclusion or refusal of adjournment.
Align case theory and expert thesis: If a defendant intends to rely on autism to explain conduct (e.g., susceptibility to manipulation, compliance, literal thinking), the case theory should explicitly put those matters in issue. A pure denial case risks rendering such expert evidence irrelevant.
Scope of expert opinion: Experts should focus on traits that illuminate behaviour genuinely beyond a jury’s common experience and directly connected to facts in issue, avoiding commentary that strays into assessing credibility.
Multi‑handed trials: The disruption of late expert evidence is magnified. Courts will be slow to sever or adjourn where the additional evidence carries only marginal probative value.
Phones and digital evidence: Where video/metadata is central, a defendant’s refusal to unlock a device can have significant forensic and case‑management consequences. If an innocent explanation hinges on digital provenance (e.g., “videos were forwarded from abroad”), technical corroboration becomes critical.

Complex Concepts Simplified

Relevance: Evidence is relevant if it makes a fact in issue more or less probable. If it does not touch a live issue, it is generally inadmissible, however interesting or sympathetic it may be.
Admissibility of expert evidence: Experts may assist the jury on topics beyond ordinary experience (e.g., clinical features of autism and how they might affect behaviour). But experts cannot tell the jury who is telling the truth or speculate about facts not advanced in the parties’ cases.
Adjournment: A pause or postponement in proceedings. Judges balance fairness, necessity, and the impact on co‑defendants, jurors, and court resources. Late requests with marginal benefits are commonly refused.
Fresh evidence (s.23 Criminal Appeal Act 1968): On appeal, the court may receive evidence not adduced at trial if it is credible, could not with reasonable diligence have been obtained earlier, and might reasonably have affected the jury’s verdict. Even then, the appellate court asks whether the conviction is unsafe.
“Gogana” statement: A solicitor’s explanatory statement accounting for late steps (e.g., why certain evidence emerged late). It may assist contextually but does not guarantee admission or adjournment.

Conclusion

R v Smilginis draws a clear boundary around the use of autism evidence in criminal trials: it must be anchored to the live, pleaded issues and genuinely assist the jury on matters beyond ordinary understanding. Where such evidence is only tangentially relevant, courts will be slow to disrupt multi‑handed trials with adjournments, especially when counter‑expert evidence would be required. On appeal, even credible neurodiversity reports will not unsettle robust convictions unless they could realistically have altered the jury’s decision.

The decision is a cautionary tale for defence practitioners: identify potential neurodiversity issues early; ensure that expert opinion is closely fitted to the defence narrative; and appreciate that late attempts to reshape the case via expert psychology are unlikely to succeed. For trial judges and prosecutors, the case reaffirms principled case management and a relevance‑first approach to expert evidence. The overarching message is one of analytic discipline: expert opinion is a tool to resolve real issues—not to create new ones mid‑stream.