England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT,
QUEEN'S BENCH DIVISION
MANCHESTER DISTRICT REGISTRY
His Honour Judge Platts
(Sitting as a Judge of the High Court)
Claim No: F90MA137

		Royal Courts of Justice Strand, London, WC2A 2LL
		19/11/2021

Lord Justice Dingemans:

Introduction

1. On 3 November 2015 the claimant, Colin Cunningham, the Assistant Head Teacher at Brownhill Learning Community ("the school") in Rochdale was assaulted by a pupil ("the pupil") who punched Mr Cunningham in the face. As a result Mr Cunningham suffered a fractured cheekbone and consequential psychiatric injuries. Mr Cunningham did not recover and has retired from teaching. The pupil was excluded and transferred to another school.
2. The school provided education and support to children between the ages of 4 to 16 years who exhibited challenging emotional and behavioural difficulties and who were not in mainstream schooling. There were about 220 pupils at the school which was spread over three sites. The evidence showed that Mr Cunningham was a respected and experienced Assistant Head Teacher at the site for the most senior pupils at the school. Mr Cunningham was referred to in the judgment below as a "devoted teacher". At the material time, the pupil was still being educated at a site for younger pupils at the start of the academic year. This appears to have been in the light of the pupil's deteriorating behaviour over the course of 2015. The pupil, who lived with his mother following his parents' separation, had joined the school in 2012. His attendance at the start had been good and he had then built positive relationships with staff members at the school. In earlier reports the pupil had been described as kind, caring and helpful.
3. The pupil had suffered bereavements in 2015. The first was the death of his grandfather in the early part of 2015. Then the pupil's father, with whom the pupil was still in regular contact, developed a serious illness and died in the summer of 2015. Notwithstanding the deterioration in the pupil's behaviour, the records show that there was still some hope that the pupil would be in a position to leave the school and re-join mainstream schooling. In 2015 the pupil had become involved in other altercations and incidents outside school. In addition the pupil had attacked Mr Cunningham on 22 September 2015 following which he had been excluded from school for three and a half days. On 5 October 2015 the pupil had attacked another teacher, and had been excluded for one day.
4. There was a multi-agency system for supporting the pupil which escalated to Team Around the Child ("TAC") meetings. There were a number of Child in Need ("CIN") and then TAC meetings in 2015 about the pupil. The pupil had also had bereavement counselling and had taken part in a Strengthening Families Course in October 2015. This last course was reported to have gone well.
5. The 3 November 2015 was the first day back in school after the half term break. There had been a minor incident with other students about a set of keys which had led to the pupil becoming agitated. The pupil had been kept back after school as part of a disciplinary process. The pupil had damaged a panel on the entrance to the school and there had then been an incident which lasted for about half an hour from the time that the pupil had left the classroom until the serious assault on Mr Cunningham. During that time the pupil had taken out his frustration on school property. At times the pupil had been less aggressive and he had stood against a wall while other students collected their belongings. The pupil became angrier and more frustrated as he was denied access to his belongings. He was banging a door which Mr Cunningham, when he arrived on the scene after the start of the incident, properly prevented. The pupil had then suddenly and without warning struck Mr Cunningham.

The claim in this action

6. Mr Cunningham brought a claim against the defendant, Rochdale Metropolitan Borough Council who ran the school and employed Mr Cunningham. The claim was pleaded to be for negligence and breach of statutory duty.
7. The claim was heard by His Honour Judge Platts sitting as a Judge of the High Court ("the judge") in the Manchester District Registry over four days from 2 to 5 November 2020. By a judgment dated 30 November 2020 the judge dismissed Mr Cunningham's claim.
8. Two main parts of the claim made against the school were first that the pupil should have been excluded from the school before the assault on 3 November 2015, and second that the incident on 3 November 2015 should have been handled in a different manner. In the light of the expert and other lay evidence at trial, the case that the pupil should have been excluded from the school before 3 November 2015 was not pursued in final submissions at the trial. The judge rejected both of these ways in which the claim was put, and there is no appeal against those findings. This appeal focusses on the claims made relating to a failure to produce risk assessments, and a failure to follow policies and arrange a return to school interview and a restorative justice meeting between the pupil and Mr Cunningham at any time after the pupil's assault on Mr Cunningham on 22 September 2015 before the assault on 3 November 2015.

The judgment below

9. The judge set out a summary of the case for Mr Cunningham and the council in paragraphs 2 and 3 of the judgment. Mr Valentine submitted that the judge had not fairly reflected all the parts of Mr Cunningham's case. This was because the judge had omitted to specify the reliance placed on the failure by the school to have a return to school interview with the pupil and a restorative justice interview with the pupil and Mr Cunningham. It is right that the judge had adopted a general description of part of Mr Cunningham's case as proposed by Mr Vaughan, who appeared below on behalf of the school, namely "the defendant failed in any event to respond adequately to the deterioration in [the pupil's] behaviour by providing adequate support or referring to outside agencies including mental health services", but the judge was alert to the absence of an interview and restorative justice meeting and identified at paragraph 51 of the judgment "what is of more concern, however, is the lack of any evidence to suggest that the incident [of 22 September 2015] was followed up either by way of restorative meeting or discussion with [the pupil]".
10. The judge then turned to the details of the school, and the level of support provided to children. In the judgment it is said that these levels of support were all provided by the school, but it was common ground at the hearing of the appeal that some of those levels of support were provided by the council's children services with the involvement of the school. It was clarified by Mr Valentine, in answer to a question from Lord Justice Arnold, that nothing turns on this point of who provided the services to the pupil.
11. The judge then referred to Mr Cunningham and the pupil before turning to the documentary evidence. This included a written behaviour policy which was dated April 2015. It had not been updated to reflect the Department of Education's "Mental Health and Behaviour in Schools" written guidance, but the judge found that was not relevant in this case. There were various pro forma documents and a proforma generic risk assessment dated December 2019 and an Aggression policy dated 2018. The judge accepted that these probably mirrored the risk assessments and policy in place at the material time.
12. It was common ground that there was no risk assessment relating to the pupil. Witnesses gave evidence about what were called "dynamic risk assessments" which the judge said "entailed staff using their experience and knowledge of the individual pupils and making their own assessments of their behaviour and acting accordingly". Mr Valentine complained that this was really little more than saying the teachers reacted to information which came to their notice, and I will return to the issue of risk assessments. The judge accepted that there was a behaviour plan for the pupil but it had not been located for the trial. It was suggested on appeal that this was probably because the file had gone with the pupil when he was removed from the school after the attack on 3 November 2015.
13. The judge identified the lay witness and expert evidence that he had heard before identifying the pupil's progress before the incident. The judge reviewed the CIN and TAC meeting notes from paragraphs 19 to 37 of the judgment. The judge then set out his findings about the attack in paragraph 38 before turning to the allegations made on behalf of Mr Cunningham. So far as is material to this appeal the judge recorded that Mr Cunningham had to prove causation in the sense that he had to prove that the breach of duty caused the injury. The judge identified that it was self-evident that if the pupil had been moved before the attack it would not have taken place. The judge said "it is more difficult to say that different interventions whilst the [pupil] remained at [the school] would have made a difference". The judge recorded the reliance placed by Mr Valentine on the decision in Vaile v London Borough of Havering [2011] EWCA Civ 246; [2011] ELR 274.
14. The judge then addressed the lack of recorded risk assessments. The judge found that the school had failed to carry out or record any formal risk assessment. The judge went on to state that it was clear that "the senior staff at the school were aware of the [pupil's] deterioration generally and the events that manifested it". This was on the basis of the notes from the TAC meetings. The judge said "I am not persuaded that if there had been any formal written risk assessment or proper written behavioural plan that it would have altered the defendant's approach to [the pupil] and his difficulties". This was because the judge found, in paragraph 44 of the judgment, that the school was a small community and it had not been shown that the incident arose because of a lack of awareness of either the deterioration in his behaviour or the risk he posed. At the end of paragraph 44 the judge did find that "the defendant can properly be criticised for failing to have and retain those documents" but did not relate that finding to the duty of care or a breach of it.
15. The judge summarised in paragraphs 45 to 54 the response to the pupil's deterioration, relying again on the notes made at the CIN and TAC meetings. The judge considered the incident of 22 September 2015 in paragraphs 50 and 51 and the sanction of exclusion for three and a half days. The judge noted the referral to Outreach Intervention on 28 September 2015, but recorded that he did not quite know what that involved. At paragraph 51 the judge said "What is of more concern, however, is the lack of restorative meeting or discussion with [the pupil]. However, that concern has to be set against the reality that [the pupil] did not return to school when he should have done. He therefore was not available for such a meeting or discussion or work until the afternoon of 5 October 2015."
16. The judge recorded concern about the incident on 5 October 2015 and the lack of any record of it, in paragraph 52 of the judgment. The judge found that despite the absence of documentation he was satisfied by the witness evidence that this incident was dealt with informally by way of a successful restorative meeting between the teacher and pupil.
17. The judge set out his conclusions from paragraphs 55 to 59 of the judgment. The judge said "As has been seen, I am critical of the lack of relevant documentation". The judge looked at the totality of the evidence and concluded "I am not satisfied that the defendant was in breach of its duty either to [the pupil] or to the claimant before the 3^rd November 2015". The judge recorded that senior staff were aware of the pupil's difficulties, and the deterioration in his behaviour, and they were aware of the incidents of 22 September and 5 October 2015. There were references to various bodies including: CAMHS, Early Help and Family Support, Resolve, Hype, The Youth Offending Team, Crisis Intervention, the school counsellor, Outreach Intervention, one to one youth work, Early Break and Strengthening Families. There was some evidence of improvement before the 3 November 2015 and the judge described the meeting on 22 October 2015 as "relatively encouraging".
18. Having found that it was reasonable not to exclude the pupil before 3 November 2015 and dismissed criticism of the handling of the incident on 3 November 2015 the judge found that Mr Cunningham "has failed to persuade me that part of his serious injury was foreseeable or that it was as a result of any breach of duty on the part of his employers". The claim was therefore dismissed.

The appeal

19. Mr Cunningham appeals on a number of grounds relating to: (1) the judge's treatment of what were called "dynamic risk assessments"; (2) the school's failure to comply with its own behaviour policy and have a return to school interview and a restorative justice meeting between Mr Cunningham and the pupil; (3) the school's failure to comply with guidance set out in the publication "Mental Health and Behaviour in Schools"; (4) the judge's finding on reasonable foreseeability; (5) the judge's failure to provide reasons for rejecting expert evidence called on behalf of Mr Cunningham.
20. The issues became more refined in the course of argument. It became clear that this was not a case where the expert evidence was relevant to the critical issues on the appeal. It also was common ground that although the school's policies had not been updated to reflect the latest published guidance in "Mental Health and Behaviour in Schools", those failures were not material to the events in this case and on the appeal.
21. In the course of submissions made by Mr Valentine it became clear that the essence of Mr Cunningham's case on the appeal was that the judge should have found that there was a breach of duty by failing to have a return to school interview and a restorative justice meeting with Mr Cunningham after the pupil's earlier attack on Mr Cunningham, and the judge, following the approach in Vaile v Havering LBC, should have found that such an interview and meeting would have prevented the assault on 3 November 2015. It is apparent that this way of putting the case was sufficiently pleaded in the Particulars of Claim, but it is also fair to record that it was not the primary focus of Mr Cunningham's case on causation, whereas it had been specifically pleaded in the Particulars of Claim that if the pupil had been excluded before the relevant attack on Mr Cunningham, the attack could not have taken place.
22. The council resists the appeal and contends that the judge made findings of fact which were properly based on the evidence and that the claim was properly dismissed. The council submitted that the judge's findings about dynamic risk assessments were reasonable, and although the school did not complete written risk assessments in relation to the pupil the failures to do so could not have made any difference, given the circumstances of the assault on Mr Cunningham. There was no breach of duty, and if there was any breach of duty it did not cause the assault on Mr Cunningham. The judge had provided clear reasons for his findings in relation to the expert evidence, and nothing said by the expert would have led to a different outcome for Mr Cunningham. As to the critical issue on the appeal, Mr Blakesley QC, who appeared on behalf of the council with Mr Vaughan, submitted that the judge had been entitled to find that there was no breach of duty in not having the return to school interview and restorative justice meeting after the assault on 22 September 2015 for the reasons given by the judge. He further submitted that the judge was right to find that it was speculative to suggest that the interview or meeting would have had any effect on the actions of the pupil on the actual day so causation could not be established, and that the appeal should therefore be dismissed.

Duty of care and reasonable foreseeability

23. I record that neither party suggested that the analysis of duty of care, breach, causation and damage in this case should be modified to pick up the six questions identified by the Supreme Court in the appeals of Manchester Building Society v Grant Thornton [2021] UKSC 20; [2021] 3 WLR 81 and Meadows v Khan [2021] UKSC 21; [2021] 3 WLR 147, and for the purposes of this appeal it is not necessary to do so.
24. It was common ground that the council owed a duty to take reasonable care to provide Mr Cunningham with a safe system of work. The relevant standard of care to be applied is that of a reasonable, prudent and competent school. In the course of submissions it became clear that although the pleaded claim included a claim for breach of statutory duty, the claim was for negligence only and that the Management of Health and Safety at Work Regulations 1999 were relied on as evidence of the standards of care applicable to a reasonable, prudent and competent school. This was because the incident post-dated the entry into force of section 69 of the Enterprise and Regulatory Reform Act 2013.
25. A breach of duty is "something which a reasonable man would blame himself as falling beneath the standard of conduct for himself" and required of a person in a similar position, see Smith v Littlewoods [1987] AC 241 at page 270 referring to Bolton v Stone [1951] AC 850 at 868-869. The school had formulated policies to protect both pupils and teachers. It was again common ground at the hearing that the school's own policies were evidence of the standard to be expected of a reasonable, prudent and competent school.
26. I turn now to the judge's finding that Mr Cunningham's serious injury was not foreseeable in paragraph 59 of the judgment. In order to bring a claim it is necessary to show that the attack was "reasonably foreseeable" to the reasonable, prudent and competent school, see Overseas Tankship (UK) Limited v Miller Steamship Co Pty (The Wagon Mound (No2)) [1967] 1 AC 617. This is more than bare foreseeability, but I infer that the judge was intending to refer to whether the attack was "reasonably foreseeable" when referring to "foreseeable" in the judgment.
27. As far as the judge's finding on reasonable foreseeability is concerned, it might be noted that an attack by a pupil was specifically identified as a risk and indeed was known to have occurred before with an attack on Mr Cunningham and another teacher. In my judgment it was therefore reasonably foreseeable to the school and council that Mr Cunningham might be attacked by the pupil. It is established that it is not necessary to show the exact nature of the attack which took place could be foreseen. In these circumstances in order for Mr Cunningham to succeed on the appeal he will in addition need to show that there was a relevant breach of duty, and that the relevant breach of duty caused loss in the sense that if there had not been a breach of duty the attack would not have occurred.

The risk assessments

28. Although the judge did record the absence of completed risk assessments and did state that "the defendant can properly be criticised for failing to have and retain those documents" in paragraph 44 of his judgment, he found that there had been no breach of duty on the totality of the evidence. In my judgment this was not a sufficient analysis of the case for breach of duty of care made against the council. An employer is generally required to carry out a suitable and sufficient risk assessment for the purposes of finding out what reasonable steps should be taken to provide a safe system of work, compare Allison v London Underground Ltd [2008] EWCA Civ 71; [2008] ICR 719. This is particularly the case if the school has set out in its own written policies the requirement to carry out a risk assessment. It is also clear that a risk assessment should not be a tick box exercise but should be a competent attempt to identify risks to safety so that reasonable steps to reduce risks can be taken.
29. The school's own policies and the evidence at the trial prove that the school acted in breach of the standard of care owed to Mr Cunningham by failing to complete risk assessments. In the event I understood Mr Blakesley in his submissions to accept this point on behalf of the school. Therefore a breach of duty in this respect was established. However this leaves the issue of causation to which I will return.

The return to school interview and the restorative justice meeting

30. The school's written policies did provide for a life space interview with a pupil which was to be completed after the incident. The school's behaviour policy provided that "restorative justice" was an "alternative approach to behaviour and relationship management in schools and is gradually being adopted" by the school. Its intended use was to reduce offending, victimisation, bullying and truancy from school. The ethos was said to be about, among other matters, encouraging accountability, building and nurturing relationships and repairing harm done to relationships through inappropriate behaviours. The evidence at the trial established that the school's policies provided for a return to school interview after an exclusion, and for a restorative justice meeting after the attack on Mr Cunningham on 22 September 2015 and the pupil's subsequent exclusion for three and a half days.
31. It was apparent that neither the return to school interview nor the restorative justice meeting took place. The judge recorded that this was "of concern" but continued "that concern has to be set against the reality that [the pupil] did not return to school when he should have done". The judge then summarised the pupil's return on 5 October, the attack on that day, the failure to make a record of that assault, the restorative justice meeting with the teacher involved in that assault, the continuing efforts to support the pupil including the completion of the Strengthening Families course and a note that the pupil had been "a bit more forward thinking".
32. The school had set out policies to have a return to school interview and a restorative justice meeting. Although it is right to show that there were difficulties in organising those meetings because the pupil was not regularly attending school, there was no evidence at the trial below to show that those meetings could not take place. In these circumstances the unexplained failure by the school to comply with its own policies was a breach of duty, because it fell below the standards of care that the school had set for itself. This again leaves the issue of causation.

Causation

33. So far as causation is concerned in order for Mr Cunningham to succeed on the appeal he will need to show that there was a relevant breach of duty, which caused loss. As this is a case where the breach of duty is an omission to act (or a failure to make things better, compare Robinson v Chief Constable of West Yorkshire [2018] AC 736) it might be more accurate to say that in this case Mr Cunningham needs to satisfy the Court on the balance of probabilities that the failure to complete the risk assessment, or the failures to have the return to school interview or restorative justice meeting, caused the attack in the sense that if the action had taken place, the assault would not have taken place. In this respect it is well-known that some measures might prevent an attack from a third party but it is sometimes very difficult to say that they would be more likely than not to prevent such an attack, compare Al-Najar and others v Cumberland Hotel [2019] EWHC 1593 (QB); [2019] 1 WLR 5953 at paragraph 235.
34. In Vaile v Havering LBC a pupil with Autistic Spectrum Disorder ("ASD") had attacked a teacher. The trial judge found that there should have been a system for revealing whether pupils had ASD and that the teachers should have been informed of the fact. The teacher had not been trained in the implementation of the relevant procedures for pupils with ASD, but the trial judge dismissed the claim. The Court of Appeal accepted the trial judge's findings of primary fact which included the fact that the teacher had not been informed that the relevant pupil had ASD and had not been trained in how to deal with pupils with ASD. The Court of Appeal accepted that had the school taken these steps the attack would have been prevented. Longmore LJ referred to Drake v Harbour [2008] EWCA Civ 25; [2008] NPC 11, a case involving the causes of a fire in an unoccupied house, where Toulson LJ had said "where a claimant proves both that a defendant was negligent and that loss ensued which was of a kind likely to have resulted from such negligence, this will ordinarily be enough to enable a court to infer that it was probably so caused, even if the claimant is unable to prove positively the precise mechanism". At paragraph 32 of his judgment, Longmore LJ concluded that "it may be difficult for Mrs Vaile to show precisely what she or the school could have done to avoid the incident if she had been appropriately instructed in suitable techniques for dealing with ASD children but the probability is that, if proper care had been taken over the relevant three year period, she would not have met the injury she did".
35. Etherton LJ agreed with the judgment of Longmore LJ. He recorded the expert evidence to the effect that if Havering LBC had done what it should have done "it was probable that the second attack would have been prevented". He held that the evidence of both experts was consistent with "an affirmative answer" to the question whether a strategy or combination of strategies would have been likely to avoid the second assault. Sir David Keene agreed with both judgments.
36. Mr Valentine submitted that the decision in Vaile v Havering LBC showed that causation could be established in a situation such as this because that Court approved the approach that "where a claimant proves both that a defendant was negligent and that loss ensued which was of a kind likely to have resulted from such negligence, this will ordinarily be enough to enable a court to infer that it was probably so caused, even if the claimant is unable to prove positively the precise mechanism". Mr Blakesley accepted that general statement of the law but submitted that Vaile v Havering LBC did not have the effect of altering the conventional rules on causation. Mr Blakesley submitted in that case causation was established because if the teacher had known about the pupil's condition and had training it was more likely than not that she could have taken steps to avoid the attack, so the loss was of a kind likely to have resulted from the negligence in failing to warn the teacher about the pupil's condition and to train the teacher on how to deal with it.
37. In my judgment Vaile v Havering LBC did not establish any new principles of law in relation to the issue of causation in general, or causation in particular relating to attacks on teachers by pupils. It was a case where the Court of Appeal considered that if a teacher had been warned about a pupil's ASD and had been trained in how to manage a pupil with ASD, the attack would, on the balance of probabilities have been avoided, even though the mechanism by which that would have occurred could not be shown. By contrast, in this case, the judge found, on the basis of evidence of records of TAC meetings and the witness evidence, that "the senior staff at the school were aware of the [pupil's] deterioration generally and the events that manifested it". The evidence also established that Mr Cunningham was experienced and trained. The situation in this appeal is different from that in Vaile v Havering LBC, and the issue of causation requires a careful analysis of the relevant factual situation.
38. As to the breach of duty in failing to complete risk assessments in this case the judge said "I am not persuaded that if there had been any formal written risk assessment or proper written behavioural plan that it would have altered the defendant's approach to [the pupil] and his difficulties". This was because the judge found, in paragraph 44 of the judgment, that the school was a small community and it had not been shown that the incident arose because of a lack of awareness of either the deterioration in his behaviour or the risk he posed. Nothing has been identified on behalf of Mr Cunningham, even with the benefit of hindsight and the passage of time, which might have been raised by a written risk assessment which would have prevented the assault on Mr Cunningham. Therefore this breach of duty did not cause the attack and Mr Cunningham's loss.
39. The most difficult issue is whether it has been shown that the failure to hold the return to school interview and the restorative justice interview between the pupil and Mr Cunningham would, on the balance of probabilities, have prevented the attack on Mr Cunningham by the pupil. Although, for the reasons given above, the issue was sufficiently pleaded and raised before the judge at the trial, it can be fairly said that this was not at the forefront of the case advanced on behalf of Mr Cunningham and the specific pleading of causation was in relation to the failure to exclude the pupil before the attack. The causative effect of either the pupil being excluded from school or the situation on 3 November being managed differently would have been obvious, but as the judge said at paragraph 40 of his judgment "it is more difficult to say that if there had been different interventions whilst the claimant remained at [the school] it would have made a difference".
40. The original lack of emphasis in Mr Cunningham's case on the causative effect of the breaches of duty by not having the return to school interview and by not having the restorative justice meeting is, in my judgment, not surprising. This is because the incident itself was a sustained incident, lasting well in excess of 30 minutes. It is apparent that the pupil's behaviour fluctuated during the incident. The situation on the day was, as the judge found, appropriately handled by the school.
41. The prospect that the pupil would, in the final event, have not assaulted Mr Cunningham because he had had a return to school interview and a restorative justice interview with Mr Cunningham is possible, but it is not probable and more likely than not to have prevented the attack. This is because the pupil had had the benefit of extensive interventions over the course of the year as his behaviour deteriorated coinciding with the time of his grandfather's death, his father's illness and subsequent death. As already recorded, the judge found that the school had been involved in referring the pupil to various bodies including: CAMHS, Early Help and Family Support, Resolve, Hype, The Youth Offending Team, Crisis Intervention, the school counsellor, Outreach Intervention, one to one youth work, Early Break and Strengthening Families. The pupil had had contact with the school counsellor, although he had refused external counselling. The pupil had been referred to bereavement services. The pupil and his mother and sibling had undertaken a strengthening family's course, which had been described in the evidence as a step forward.
42. In all of these circumstances the attack in this case was not of a kind likely to have resulted from the failure to have the return to school interview and the restorative justice meeting. This appears from the sustained nature of the incident, the circumstances of the assault, and the fact that all of the other interventions did not prevent the assault. In my judgement, therefore, the appellant is unable to show on the balance of probabilities that a return to school interview or a restorative justice interview would have prevented the pupil's serious assault on Mr Cunningham. This means that Mr Cunningham is unable to show that if there had not been any breaches of duty on the part of the school, the attack and Mr Cunningham's loss would have been avoided, and therefore causation is not established.

Conclusion

43. For the detailed reasons set out above I would find that Mr Cunningham has proved breaches of the duty of care owed by the school to him in that the school failed to carry out risk assessments, and failed to arrange a return to school interview and failed to arrange a restorative justice meeting between the pupil and Mr Cunningham. In my judgement, however, Mr Cunningham is unable to show that if the risk assessments had been carried out, or if the return to school interview and restorative justice meeting had taken place, the attack on 3 November 2015 would not have taken place. I would therefore dismiss the appeal.

Lady Justice Andrews

44. I agree.

Lord Justice Arnold

45. I also agree.