Abstract
Facts
The present appeal stemmed from a deliberate collision between a car and four pedestrians that resulted in the death of one person (Flamson) and injuries to the other three. The appellant Ayre appealed against his convictions for manslaughter, inflicting grievous bodily harm (GBH) and assault occasioning actual bodily harm, while his co-accused (Henneberry and Tomlinson) renewed their applications for an extension of time to apply for leave to appeal against their convictions for the same offences.
It was the prosecution case that Flamson was a cannabis dealer who was owed money by one of Henneberry’s sons, which created tension. On the night of the incident, Flamson and three others (Corbett, Barton, and V, a minor) went to the road where Henneberry lived. Henneberry was inside his house with Ayre, Tomlinson, Donald, and Rollason. An exchange of words occurred between the two groups, during which Henneberry was carrying a wooden pole. Flamson threw something towards the house. Henneberry threatened Barton and V. Flamson and his group then walked away. Less than a minute later, Donald got into a car, followed by Tomlinson, Ayre (who had retrieved a baseball bat), Henneberry (still carrying his wooden pole), and Rollason (unarmed). Donald manoeuvred the car in the direction Flamson’s group had taken and set off slowly with the headlights on. A short distance from the house, he turned off the headlights, sped up, and drove into the group, hitting Flamson and Corbett. The damaged car was later abandoned in a car park.
At the conclusion of the prosecution case, submissions were made of ‘no case to answer’ on the basis that there was insufficient evidence from which the jury could properly infer that the defendants had provided encouragement for the attack. It was also submitted that the overwhelming supervening act (OSA) defence applied. The trial judge rejected the submissions. The appeal focused on the judge’s refusal to leave the OSA issue to the jury.
The concept of OSA arose when D2’s conduct ‘has faded to the point of mere background, or has been spent of all possible force by some overwhelming intervening occurrence by the time the offence was committed’ (at [21], citing Jogee at [12]). The court approved of the trial judge’s reasoning for rejecting the OSA defence. The judge had correctly found that the appellants’ conduct (being part of the group making threats, getting into the car armed with a baseball bat) ‘could in no way be said to have faded to the point of mere background; it was a current and contemporaneous encouragement’ (citing the trial judge’s determination, at [30]). The court considered that in the immediate run up to the collision ‘the joint intention to use unlawful violence against the other group was still continuing, and it was put into effect by Donald acting in a way which resulted in death’ (at [48]). It added that ‘even if [the Court of Appeal] had taken a different view of the application of the principles stated in Jogee to the evidence in this case, the judge would still have been correct to reject a submission of no case to answer’. This is because the court found it ‘difficult to think of any circumstances in which a judge could properly rule not only that the defence of OSA arose, but also that it was of such strength that no reasonable jury could properly reject it’ (at [49]).
The court agreed with the observations of the single judge in his written reasons for refusing leave to appeal and rejected the argument that the prosecution’s case had changed from a developing plan to a new plan requiring the jury to be sure of a specific intent to kill with the car. It endorsed the single judge’s reasoning that these were ‘narrowly linguistic points’ and an attempt to reintroduce the ‘discredited notion of “fundamental departure” by the back door’ (at [50]). The verdicts which were returned were properly open to the jury. It was not, in the court’s view, ‘possible to argue backwards from the verdicts in order to conclude … that if they had been permitted to consider OSA, the jury would have acquitted on all counts. Nor [were] the appellants assisted by seeking to rely on the jury’s decision to acquit Rollason, who was the last to join the group in the Mondeo and was not armed’ (at [52]).
The court declined to certify questions for the Supreme Court, as their decision in Ayre did not raise any new point of law but applied the principles stated in Jogee. The court did not see any inconsistency between the decisions in Jogee and Grant [2021] EWCA Crim 1243. But even if it did the decision of the Supreme Court was binding on the Court of Appeal. The court concluded by noting that Jogee made ‘clear the approach to be adopted, and there was no basis for inviting further consideration of a debate about causation and remoteness’ (at [57]).
Commentary
The outcome in Ayre is unsurprising, given the Court of Appeal’s consistently strict approach to the OSA principle in earlier cases. The appellants sought to argue that the driver’s actions – switching off the headlights, accelerating, and steering directly into the group – were his alone, forming a separate plan, rather than an escalation of an agreed plan, that broke any ‘material connection’ with their own conduct (at [39]). The court’s firm rejection of this claim reinforces how narrow and unforgiving the OSA doctrine has become: as the appellants astutely observed, the Court of Appeal’s jurisprudence has rendered OSA into a defence ‘which is largely theoretical and almost never exists in practice’ (at [36]). The appellants considered Grant to be the prime culprit of this development, as that decision had ‘wrongly narrowed that which the Supreme Court intended when giving judgment in Jogee’ (at [35]).
Although Jogee made clear that an accessory need not encourage a particular means of committing the offence ([98]), cases like Ayre (and Grant before it) push that principle to its limits. The use of a car as a lethal weapon, following an earlier confrontation involving only blunt wooden instruments, represents a major escalation. Prior to Jogee, secondary parties who had shared an intent that the victim be caused GBH but did not intend death could avoid liability if death resulted from the use of a fundamentally different (i.e. more dangerous) weapon or manner of attack. It is instructive to compare the decision in Ayre with the outcome in English [1999] 1 AC 1 (a pre-Jogee appeal) where a murder conviction was quashed because the fatal knife attack fundamentally differed from the contemplated use of wooden poles. In Ayre, the court reached the opposite conclusion, agreeing with the trial judge that the car attack was an ‘anticipated culmination’ (at [31]) based on an inference of a shared purpose that the victims be caused GBH. This leaves little room for defendants to claim a lack of intention regarding unforeseen modes of violence, even if they genuinely did not intend such a method.
If OSA works, as the court suggests, as a special defence to complicity (rather than a denial of its actus reus or mens rea elements), then Ayre (and Grant before it) leaves it with very little scope for practical application. Alternatively, if OSA is, as I have argued (Beatrice Krebs, ‘Overwhelming Supervening Acts, Fundamental Differences, and Back Again? (2022) 86 Journal of Criminal Law 420)’, a remoteness principle underpinned by ideas of fair attribution and designed to prevent overreach in constructive crimes like murder, the court’s approach renders it hollow: once GBH is jointly intended, the method of infliction becomes irrelevant. The logic appears to be that if the accessory was willing to support serious violence, he remains responsible whatever form that violence takes. Once there is a shared intent to cause GBH, the accessory is treated as supporting the act, even if the principal escalates the violence dramatically. But this approach can seem unfair. GBH covers a broad range of harm, and some methods of attack are far more violent and dangerous than others. And while the principal controls how the attack unfolds, the accessory is usually bound by what he originally intended. The focus should therefore be on the specific risk that the accessory intended to support, not simply on whether he intended GBH in general. The principal’s choice of weapon sets the level of risk, and it may be disproportionate to attribute that higher risk to an accessory who anticipated a less deadly form of violence. On the court’s strict interpretation of OSA, an accessory can seemingly escape liability if and only if he explicitly limits his assistance or encouragement to a specific means or method (e.g. by insisting that the victims be struck with blunt wooden instruments only).
Concerns about over-reach are compounded by the court’s admission in Ayre that it could hardly ‘think of any circumstances in which a judge could properly rule not only that the defence of OSA arose, but also that it was of such strength that no reasonable jury could properly reject it’ (at [49]). The court’s continued inability to identify even a hypothetical example where OSA could apply, let alone succeed, is deeply troubling. Surely, the Supreme Court would not have retained the OSA principle in Jogee if it had no realistic practical application? While Lords Hughes and Toulson in Jogee acknowledged that earlier uses of the principle were rather generous (see Jogee at [33]: ‘That there had been such an [overwhelming supervening] event in Anderson and Morris may have been a charitable view on the facts…’), the pendulum would now appear to have swung too far in the opposite direction.
The court’s reluctance to engage with the OSA principle is understandable, given the complex litigation that followed the ‘fundamental difference rule’ (FDR) that OSA replaced. The difficulties with the FDR lay less with the principle itself than with its inconsistent application and the proliferation of unmeritorious appeals it encouraged. This history may explain the court’s current hesitance to engage with OSA, but the steady stream of appeals suggests that it is a pressing issue that remains unresolved. What is needed now is guidance: examples of when OSA might apply, rather than repeated warnings about its demanding threshold. The court’s refusal to certify questions for the Supreme Court for clarification and advice demonstrates an unwillingness to decisively tackle this issue, despite ongoing uncertainty and debate amongst practitioners and academics alike.