England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM QUEEN'S BENCH DIVISION
MR JUSTICE FIELD
HQ08X02890

		Royal Courts of Justice Strand, London, WC2A 2LL
		2nd February 2011

Lady Justice Smith:

Introduction

1. This is an appeal from the order of Field J made on 22 January 2010 when he dismissed the claim for damages brought by Robert Lee Uren in respect of very severe personal injuries which he sustained while taking part in a 'Health and Fun Day' at RAF High Wycombe on 28 July 2005.
2. The Health and Fun Day at which Mr Uren was injured included a series of competitive games of the kind shown on the television programme 'It's a Knock Out'. The injury occurred during the last game of the day, a relay race in which the members of four teams, representing the 'Flights' present on the station had, in turn, to run up to and enter an inflatable rectangular pool, retrieve one of several objects which were floating in the shallow water and return with it to the starting point. Mr Uren entered the pool head first and, in a manner and for reasons which have never become entirely clear, struck his head on the base of the pool. His neck was broken and he is now tetraplegic. He was only 21 years old at the time. This was a tragedy for him and his family.
3. Mr Uren sued his employers the Ministry of Defence (MOD) and also Corporate Leisure (UK) Limited (CL), the company which supplied the equipment and personnel for the series of games. On a trial of liability only, the main allegations were that the defendants had failed to carry out suitable and adequate risk assessments. Had adequate risk assessments been carried out, it was said that the participants would have been forbidden to enter the pool head first. Such a method of entry gave rise to a foreseeable and unacceptable risk of injury. Had such instruction been given, the claimant's injury would have been avoided. The judge held that the risk assessments were indeed inadequate but that the game was reasonably safe even when participants entered the pool head first. There was no breach of the defendants' duty of care. Accordingly he dismissed the claim. Mr Uren appeals against the dismissal of his claim and the MoD cross appeals against the judge's holding that its risk assessment was inadequate.

Factual Background

4. It is necessary to describe the game in a little more detail. The inflatable pool had been installed by CL on a grassed playing field at the RAF station. When inflated, the sides or walls of the pool were approximately cylindrical, being 1.04 metres high and 0.98 metres wide. The internal dimensions were about 5 metres by 3 metres. The depth of water in the pool was just under half a metre or about 18 inches. The starting line for the game was about 14 to 15 feet from the pool. At the start, one member of each team had to run up to the long side of the pool and, in some way, get into the water. He would retrieve an item of plastic fruit from the water then retrace his steps to the starting point and place the fruit in a bucket. The next member of his team would then set off to repeat the exercise.
5. Staff employed by CL gave instructions to the participants and supervised the operation of the game. There was, at the trial, some dispute as to the nature of the instructions given. Mr Darren Berry, CL's General Manager, who gave the instructions, claimed that he told the participants that they were not to go into the pool head first but were to climb in. No other witness agreed with that evidence and the judge rejected it, accepting instead that no specific instructions were given as to the mode of entry and no safety instructions were given other than a general exhortation to take care and to use common sense.
6. There were eight flights taking part so the game had to be played in two heats of four teams each. During the first heat, the participants used a variety of means of entering the pool. Some scrambled over; some sat on the side and swung their legs over; others went in head first, sliding or slithering over the side. Although Mr Berry was watching the game, he did not suggest that this last means of entry should not be repeated.
7. Senior Aircraftman Plant was taking photographs of the event. One of his photographs showed a participant in the first heat entering the pool head first. The photograph shows a man with his arms outstretched in front of him, as if diving, but with his body in contact with the inflated side of the pool. Mr Plant told the judge that he thought that this practice would lead to an accident and said as much to another spectator.
8. There was a good deal of argument and confusion at the trial about the use of the words 'dive' or 'diving'. I can see how this arose. The kind of head-first entry shown in the photograph was in some ways similar to a dive but it was not a dive in the sense that the body was ever free in the air before entering the water. Evidence was given (and accepted by the judge) that it would have been virtually impossible to execute a free dive over the side of the pool; there was no spring board and the run up was only 15 feet. What people meant when they spoke of 'diving' was a head-first entry with the body in contact with the side of the pool. Some witnesses described this as 'sliding' or 'slithering' into the pool.
9. Mr Uren's flight took part in the second heat. The first member of the team was Sergeant Beeken. He went in head first and landed awkwardly. However, he was not injured, only slightly winded. He got up and carried on with the game. He did not tell anyone at the time that he had been winded.
10. Mr Uren intended to enter the pool head first. He knew that the water was shallow; he had been in the pool earlier that afternoon during a bit of horseplay. When his turn came, he ran up to the pool and went over the side head first. Those watching observed that, as he went over, something caused his legs to flip up in the air so that he entered at a steep angle. He must have struck his head on the base of the pool. He was knocked unconscious very briefly and was seen face down in the water. He was rescued but was found to have been gravely injured.
11. There was no clear evidence as to how or why this had happened. Mr Uren recalled that, as he approached, he realised that there were three people in the pool and he may have altered his direction of travel slightly to avoid a collision. But that would not explain what went wrong. Some witnesses said that it appeared that his legs seemed to catch on the side but I find it hard to see how that could happen. The judge rejected the suggestion that Mr Uren's legs were 'bounced' up by the effect of another participant getting into or out of the pool. He accepted that there was some bounce in the side wall but there was no evidence that anyone else was on the side at the time. The judge's finding as to how and why the accident happened was as follows:

"In my judgement, Mr Uren's legs were lifted up as a result of the flatness of his trajectory as he went over the side head first with his arms stretched out in from of him. He was trying to enter the pool by sliding in over the side as quickly as possible. In my judgement, he was not attempting a manoeuvre which he must or ought to have appreciated was dangerous."

I myself have some difficulty in understanding why or how the flatness of the trajectory should have caused Mr Uren's legs to be lifted up. It would seem to me that the most likely explanation is that, due to the position in which he landed on the side and the angle of his body at that moment, the bounce caused his legs to go upwards and tipped him forward so that he entered too steeply. However, I do not think that it matters for present purposes what was the precise mechanism of the accident. What matters is that, while doing the same thing as others had done before and which had not been forbidden, Mr Uren's entry went disastrously wrong for no very clear reason.

The allegations of breach of duty

12. The pleaded case included allegations of statutory breaches against both defendants but it was accepted before the judge that they did not add anything to the duty which both defendants owed at common law. The judge defined this as the duty to take reasonable care to ensure that Mr Uren was safe in taking part in the game.
13. It was common ground that section 1 of the Compensation Act 2006 did not add anything to the common law position. Section 1 requires the court, when considering a claim in negligence or breach of statutory duty and when determining whether the defendant should have taken particular steps to meet a standard of care, to have regard to whether a requirement to take those steps might prevent or discourage people from undertaking desirable activities. At least since Tomlinson v Congleton Borough Council [2004] 1A.C. 46, if not before, the common law has required the court to take such matters into account.

The risk assessments

14. A major plank of the claimant's case was that the risk assessment carried out by each defendant had been hopelessly inadequate. A great deal of time was spent at the trial in considering their adequacy. Mr Berry had been responsible for CL's risk assessment or at least for reviewing an assessment which had been carried out earlier by a former employee. CL's evidence was that this pool was not often hired out as it required a great deal of water. Indeed, there does not appear to have been any evidence that this game had ever been played before using this pool. The risk assessment, as modified (if it was) by Mr Berry made no reference at all to methods of entry and there was no evidence that the person responsible for the original assessment appreciated that participants might enter the pool head first. Mr Berry had not seen this game played and had not applied his mind to the question of how participants would enter and leave the pool. The judge held that this risk assessment was inadequate and that holding was not challenged.
15. The MOD's internal instructions also required the RAF to carry out a risk assessment. The person responsible for this, Mr Richard Cassford, a civilian employee, who was suitably qualified to carry out a risk assessment did nothing about it. However, two risk assessments were carried out by Flight Lieutenant Taylor of the Physical Education Flight. One related to the Fun Day as a whole and was in very general terms. The other related to the 'It's a Knockout' games and was rather more detailed. From that assessment, it was apparent that the RAF regarded CL as being responsible for the safety of the events. The judge found that Flight Lieutenant Taylor completed the risk assessment without knowing how the pool game was played, let alone how participants might enter the pool. He held that it was not open to the RAF to leave risk assessment to CL. The RAF's risk assessments were 'fatally flawed'. Those conclusions are challenged on the cross appeal.
16. It followed that, in the judge's view, both defendants were in breach of their duty to the claimant. However, at paragraph 39, the judge said:

"My findings as to the risk assessments prepared for and on behalf of the first and second defendants do not mean that Mr Uren succeeds on the issue of liability. The question for decision is not whether adequate risk assessments had been undertaken, but whether the defendants took reasonable measures to ensure that the game was safe."

17. Later in this judgment, I will return to the question of whether the judge was right to put aside, as he did, the breaches of duty in respect of the risk assessments.

Was the game as played reasonably safe?

18. The claimant's case was that, although he himself had not appreciated it at the time, a head-first entry into the pool carried with it a foreseeable risk of serious injury. Head-first entry should have been forbidden. In support of this contention, the claimant relied first on the impression of the eye witness SAC Plant, the photographer, who immediately thought that head-first entry was going to lead to an accident. The claimant relied also on the accident report completed by Mr Berry of CL which gave the following account of the accident:

"During a knock out game, a young man decided to dive into a paddling pool, all participants had been asked to only climb into pool as pool is very shallow. We had already played 1 game this was the second."

It was the claimant's submission that it must be inferred from this report that CL was of the view that a head-first entry was obviously dangerous. Therefore, if CL had carried out a proper risk assessment, head-first entry would in fact have been forbidden. It is pertinent to note, that in the initial stages when the claim had been intimated, CL's stance was that the accident had happened because the claimant had acted in a way which was different from any other competitor by going over the side head first. It was only after disclosure of the photograph taken by SAC Plant showing a competitor in the first heat making a head-first entry that that allegation was dropped and CL began to contend that the game as played was safe. I should mention that Mr Berry was also shown on that photograph and was watching and commentating on the event.

19. Finally, the claimant relied on the evidence of an expert witness, Mr Andrew Petherick. He had long experience in the organisation and management of sporting events, including 'It's a Knockout' events. In a nutshell, his view was that head-first entry into this pool was similar or analogous to diving from the side of a swimming pool. Diving into shallow water is always forbidden as the risk of serious injury is well known. He mentioned the regulations concerning swimming pools. However, he accepted that on this occasion the pool was not being used as a swimming pool and that this was obvious to participants. He also accepted that it is impossible to make sporting activities risk-free and that there was an acceptable degree of risk. But he maintained that a head-first entry created an unacceptable risk of injury.
20. The claimant also relied on the report of the expert instructed by the MOD, Dr Simon Jones, a consultant mechanical engineer. The MOD did not wish to rely on this report and therefore did not call him. The gist of his report was that a slow head-first entry into this pool would not be dangerous because the participant could control his landing using his hands and arms. However, a fast head-first entry might not be capable of being controlled. A fast head-first entry was similar to a dive and it was well known that it was dangerous to dive into shallow water. He said that because it was difficult to differentiate between slow and fast head-first entries, it would probably have been advisable to ban all head-first entries. It is clear from Dr Jones' participation in the discussion between the experts that it was Dr Jones' view that it would not be practicable to give an instruction that participants could enter head-first provided that they did not go too fast; therefore the better course would be to ban all head-first entries. The judge did not comment upon this important aspect of Dr Jones' opinion; he merely said that he had not found Dr Jones' report of much assistance.
21. Both defendants contended that the game as played, permitting head-first entry was reasonably safe. The participants had been warned to be careful; they knew that the water was shallow. The pool was sited on grass which was a much softer surface than the bottom of a swimming pool or the floor of a gymnasium. Such small risk as existed was acceptable for a sporting event. The game was fun and was intended to present a challenge for fit young servicemen. To forbid head-first entry would have removed much of the enjoyable challenge.
22. In support of these submissions, the defendants relied on the evidence and report of the expert instructed by CL, Professor David Ball. He is the Professor of Risk Management at Middlesex University and a consultant on health and safety issues. He has long experience in the field of risk assessment and management and has practical training as a risk assessor. He has acted as consultant for several government departments including the Health & Safety Executive, the Department of Health and has contributed to the work of the European Safety Association and the World Health Organisation.
23. The gist of Professor Ball's opinion was that a head-first entry into the pool with the body in contact with the inflated side gave rise to only a very low risk of serious injury. He acknowledged that, if it were possible to make a free dive over the side, that would be dangerous. But contact with the side of the pool would slow entry into the water sufficiently that it would be possible to control landing by using the hands and arms. The distance of descent was only about one metre. Moreover, the grass underneath the pool provided a more absorbent surface than would a floor. Given that the risk of serious injury was very low, and given that there was some social benefit to the activity, the risk was acceptable. To ban head-first entry would reduce the game to a boring and pointless activity.
24. The judge rejected the opinions of Mr Petherick and Dr Jones and accepted Professor Ball's opinion that the game as played gave rise to only a very small risk of injury. He noted that the participants had been warned to take care on entering the pool. It was obvious that they should not attempt to dive in without sliding over the side. In sliding over the side, they would be moving horizontally and friction would slow the pace of entry. At that point they would be only a metre above ground and by extending their arms they would be able to control the impact with the bottom of the pool, which rested on a grassed playing field. In short, he accepted Professor Ball's opinion in its entirety.
25. The judge then considered whether the existence of a very small risk of serious injury meant that the defendants were in breach of duty. He concluded that, bearing in mind the social benefit of the game, it did not. He said that enjoyable competitive activities are an important part of the life of many people young and fit enough to take part in them. This is especially so in the case of fit service personnel. A balance had to be struck between the level of risk and the benefits the activity conferred. The pool game was enjoyable in part because of the physical challenges it posed. The risk of serious injury was small. The judge concluded that neither CL nor the MOD was 'obliged to neuter the game of much of its enjoyable challenge' by prohibiting head-first entry. The claim failed.

The appeal to this court

26. At the outset of the appeal, Sir Geoffrey Nice QC for the appellant applied for permission to introduce and rely on three documents as fresh evidence, which he claimed had not been available at the hearing below. Two of these were British Standards relating to the safety requirements for children's playgrounds. These had been referred to by Professor Ball in the course of his evidence, although not produced. The respondents did not object to their introduction on the appeal.
27. The third document for which permission was sought was a report prepared by Professor Ball published by the HSE entitled 'Playgrounds – risks benefits and choices'. Sir Geoffrey contended that this document contained material which tended to undermine the validity of Professor Ball's evidence. Although the claimant's team had seen this document about 24 hours before the trial began, this had been too late for them to read and digest it; it is very long.
28. We rejected this application for two reasons. First, we were not satisfied that, with due diligence, the document could not have been obtained and read in good time before the hearing; it is in the public domain. Second, we consider its relevance to the issues in this appeal to be marginal. It certainly could not be said that it would have been likely, if admitted before the judge, to make a significant impact on the judge's thinking.
29. Sir Geoffrey's submission on the substantive issues was that the judge's decision was clearly wrong. In effect, that was the sole ground of appeal. The skeleton argument was extremely long and challenged almost every finding of fact and conclusion of law. I do not propose to deal with each and every argument but to focus on those which Sir Geoffrey dealt with in oral argument.
30. Sir Geoffrey began by impressing upon the court what he said would be the effect of the judge's decision. It would mean, he said, that CL and companies like it would be able to promote this game for use by people of all ages without any restriction as to way in which participants were to enter the pool, subject only to an exhortation that they should 'take care'. That, he submitted, was a recipe for further tragedies in future similar to this one. He submitted that there was no evidence that this game was safe if played as it had been here; in the past CL had played this game only once and then with a different pool which was round and had deeper water. On CL's website, a picture was shown of a pool not dissimilar to the one used in the instant game; there was a pictogram on the side which forbade diving. Moreover, there was no evidence that anyone from CL had ever tried out the head-first method of entry. Professor Ball had not tried it out or watched it being tried out before giving his opinion that the serious injury from head-first entry was very small.
31. Having found that the risk assessments carried out by CL and the MOD were inadequate, the judge should have considered and decided what a suitable and sufficient risk assessment would have concluded. He failed to do so; instead he simply moved on to the question of whether the game was safe.
32. In considering whether the game was safe, Sir Geoffrey submitted that the judge made several errors. First, it is said that he wrongly focussed on the risk of serious injury rather than the risk of any physical injury. It is said that had he considered that wider risk, he would have been bound to find that head-first entry had to be forbidden. Professor Ball had accepted that the risk of breaking an arm on head-first entry was greater than the risk of spinal injury. The judge had ignored that and had focussed only on the risk of serious injury by which he meant spinal injury. Further, when considering the risk of serious injury, (ie spinal injury) the judge should have paid attention not only to the likelihood of the risk materialising (which he held was very small) but also to the fact that the risk in question here was of catastrophic injury. The more serious the injury the greater the need for precautions. Also, the judge should have considered the ease and practicability of avoiding the injury. Here it would have been easy to give an instruction avoiding head-first entries. The social value of the game was not great and it could still have been fun for the participants even if head-first entry had been forbidden.
33. Further, the judge had erred in accepting Professor Ball's evidence in preference to the opinions of the other two experts who were in agreement on the crucial question of whether a head-first entry at speed, following a run up to the pool, gave rise to a significant risk of injury. The judge had accepted Professor Ball's evidence without grappling with the merits of this difference of view. He had been impressed by Professor Ball's reliance on statistics relating to spinal injuries in sporting activities; these showed that spinal injuries were rare occurrences. However, it was submitted, such statistics were wholly irrelevant to the degree of risk entailed in this particular game, for which, obviously, no statistics were available. Further, the judge had erred in that he had regarded the impression of the eye-witness SAC Plant as irrelevant.
34. The respondents were separately represented because there was a potential conflict of interest in the apportionment of liability in the event that the appeal were to be allowed. However, Mr Richard Lynagh QC for CL and Mr Derek Sweeting QC for the MOD made similar submissions in support of the judgment. They submitted that the judge's findings of fact were open to him on the evidence. In particular, he had been entitled to prefer the opinion of Professor Ball to those of the other experts. He had made no error of law. Further, having identified that a very small risk of serious injury did arise from head-first entry, he was right to balance that risk against the social benefit of the activity. His judgment on that balancing exercise should not be disturbed.
35. The respondents accepted that the judge had not specifically considered the causative effect of his holdings in respect of the risk assessments. He had merely pointed out that such holdings did not lead directly to the conclusion that the defendants were liable. However, by considering whether the game as played was safe, the judge was in fact considering what a suitable and sufficient risk assessment would have shown and it was clear that the judge was of the view that a suitable and sufficient risk assessment would have concluded that the game was reasonably safe. There was no error of law.
36. On the cross appeal, the MoD argued that the judge had been wrong to criticise its risk assessment as inadequate. Bearing in mind that the MoD was relying on a reputable contractor (CL) to provide the game, it was entitled to rely on CL to conduct the risk assessment. The MoD wished this point to be resolved by this court even though the judge's adverse holding had had no effect on the final outcome. It contended that authoritative guidance was needed as to the duty in respect of risk assessments of an organisation which used reputable contractors.
37. For Mr Uren it was argued that judge had been right to criticise the MoD's risk assessment. As a matter of law, it had a statutory duty as Mr Uren's employer to conduct a suitable and sufficient risk assessment. That duty, like its common law duty to take reasonable care for the employee's safety, was non-delegable. Moreover, on the facts, the MoD's risk assessment was manifestly inadequate.

Discussion

Risk assessment- issues in the appeal

38. I begin by considering the suggestion that the judge erred in not following through from his findings that both respondents' risk assessments were inadequate. I confess that when I read paragraph 39 of the judgment, quoted at paragraph 16 above, I was surprised that the consequences of the failure to carry out a suitable and sufficient risk assessment were dealt with so briefly.
39. It is obvious that the failure to carry out a proper risk assessment can never be the direct cause of an injury. There will, however, be some cases in which it can be shown that, on the facts, the failure to carry out a proper risk assessment has been indirectly causative of the injury. Where that is shown, liability will follow. Such a failure can only give rise to liability if a suitable and sufficient assessment would probably have resulted in a precaution being taken which would probably have avoided the injury. A decision of that kind will necessitate hypothetical consideration of what would have happened if there had been a proper assessment.
40. The evidence which the appellant wished to rely on to prove their point on this issue was that of Mr Berry, CL's General Manager. His evidence was that, although it did not form part of his written risk assessment, he had in fact prohibited head-first entry when giving oral instructions. From that, said the appellant, it should be inferred that, if he had carried out a proper assessment, that is what he would have put down and that is what would have been required. If Mr Berry had been a witness of integrity, that would have been a powerful point to make. But the judge rejected Mr Berry's evidence as to the instructions he gave because no one who heard them agreed that that prohibition had been imposed. Moreover, the photograph showed a competitor going in head-first while Mr Berry was watching him. If the judge had found that Mr Berry was honestly mistaken in thinking he had given such an instruction and had meant to do so, then the judge may, I think, have had to draw the inference that a proper risk assessment would have led to the banning of head-first entries. But Mr Berry's evidence was rejected as deliberately self-serving. He was either trying to avoid personal criticism or seeking to help CL avoid liability or possibly both. In any event, his evidence was to be distrusted and it was entirely reasonable for the judge to refuse to draw an inference that, if he had conducted a proper risk assessment, Mr Berry would have banned head-first entries.
41. It is perhaps unfortunate that the judge did not deal expressly with this point, although in the event the omission cannot have any effect on the outcome of this appeal. I say that it would have been better if he had dealt with it because there was some evidence potentially relevant to the issue. Also, I would not wish judges to have the idea that risk assessments are unimportant and can never affect the outcome of a claim. I would not wish them to think that all that matters is an objective assessment of safety by the court. Sometimes the failure to undertake a proper risk assessment can affect or even determine the outcome of a claim and judges must be alive to that and not sweep it aside.
42. In any event, risk assessments are an important feature of the health and safety landscape. At their best, they can provide an opportunity for intelligent and well-informed appraisal of risk and can form a blueprint for action leading to improved safety standards. It must, however, be admitted that they are not a panacea. There was, during this case, some discussion about the difficulty of conducting an effective risk assessment in connection with some kinds of activity. For example, as Professor Ball pointed out, there were in this game so many potential variables, depending on the decisions of individual participants, that it would be virtually impossible to give separate consideration to every possibly variable of what might be attempted. His recommendation was that risk assessors should 'keep it simple'. I see the force of that point. It seems to me that formal risk assessments are probably more effective in relation to static conditions or activities which are often repeated in a fairly routine way. They may well be a less effective tool where a lot of variables may come into play.
43. Professor Ball also drew attention to what I think he regarded as a degree of artificiality in the approved methodology of risk assessment. The approved method falls into two parts. First, it requires the assessment of the likelihood of the eventuation of a risk. A number of points is assigned to each degree of likelihood. In one of the assessment templates considered in the present case, the points ranged from 1 to 4: 4 points for a risk that is very likely, 3 for likely, 2 for possible but unlikely and 1 for very unlikely. Then the severity of the resultant injury is assessed: 4 points are given to a very serious injury, 3 for less serious, 2 for moderate and 1 for trivial injury. Finally, the overall risk assessment is made by multiplying the two numbers together. Someone has decided in advance what overall number is acceptable and what the threshold is for requiring further controls or precautions to be applied. A second template was also considered in the evidence in this case; this used a wider range of numbers and was therefore potentially a little more sensitive but the essence of the operation was exactly the same.
44. I am reluctant to criticise this methodology; it is advanced by reputable bodies and no doubt they have been anxious to give guidance which can be applied by a wide range of people to a wide range of circumstances. I say only that I can understand Professor Ball's reservations about it. It is rather a blunt instrument. For one thing it is heavily dependent on personal impression. Second, it seems to me to encourage a mechanistic or 'tick-box' approach, rather than a thoughtful appraisal of risk. Third, the setting of a fixed threshold for acceptability may fail to draw appropriate attention to a 'very unlikely' risk of catastrophic injury.
45. I would like to think that the careful use of such a template will produce a sensible result. However, I do see a danger that, by carrying out the template exercise, an employer or other responsible person may lose sight of the real objective of identifying dangers which should be avoided. I would not like it to be thought that the performance of such an exercise with a reasonable degree of competence and compliance with its results will necessarily provide a complete answer to liability. It seems to me that the use of a template can never fully replace the reasoning processes of an intelligent and well-informed mind. I hope that is not too much to hope for.

The safety or danger of this game as played

46. As I have said, the judge assessed this game as giving rise to only a very small risk of serious injury. The respondents sought to argue that the notice of appeal did not seek to challenge this finding and the appellant cannot now do so. I cannot accept that. In my judgment, it was always clear that the appellant was challenging every aspect of the judgment, factual and legal, which contributed to the conclusion. That clearly included the holding that the game gave rise to only a very small risk of serious injury.
47. I will deal first with Sir Geoffrey's criticism of the judge's approach to the law. It is said that the judge should have considered the risk of any physical injury and not limited himself to consideration of serious injury by which he meant spinal injury. I do not accept that the judge limited his consideration to spinal injury but, even if he did, this would not, in the circumstances of this case, have been an error of law. It was accepted on all sides that, arising from this game, there was some risk of moderate injury (for example a fracture of the upper limb) and no doubt a variety of trivial injuries too. It was not in issue whether those risks were such that the game should have been stopped or that head-first entry should have been forbidden on account of them. I would have thought that the balance of those risks and the benefit of the game would favour the continuance of the game. But those wider risks were irrelevant to the issue before the judge which was whether the risk of a fractured spine or other serious injury (eg to the head) was such that head-first entry should have been banned.
48. An important part of the judge's conclusion on this issue was his acceptance of the entirety of the opinion of Professor Ball and his rejection of the opinions of the other two experts. In general, a judge is entitled to accept the evidence of any witness in preference to that of others. Nowadays, one expects a judge to give reasons for such a preference even in the case of lay witnesses. When preferring the evidence of one expert to that of others, the judge's reasons are open to quite close scrutiny. It will be necessary for me to examine quite closely Field J's reasons for preferring Professor Ball.
49. The judge accepted Professor Ball's opinion because he was very experienced and because it appeared to the judge that his opinion was 'solidly scientific with due regard to common experience and common sense'. There can be no doubt that Professor Ball had very extensive experience in advising about safety, reduction of risk and acceptability of risk in relation to sporting activities. He had a particular expertise in relation to children's playgrounds but his expertise was not limited to that. It is not suggested that Professor Ball was not appropriately qualified to pass an opinion on the crucial issues in this case. Rather Sir Geoffrey sought to attack the judge's conclusion that Professor Ball's opinion was solidly scientific. He pointed out that the judge appeared to have been impressed by certain statistical information which Professor Ball provided and which the judge put at the forefront of his summary of Professor Ball's evidence. He submitted that, if that was the solid scientific base which the judge had in mind, he had erred because the statistics were quite irrelevant to the formation of an opinion on the safety of this game.
50. In my view, there is some force in this submission. At paragraph 48 of his judgment, the judge outlined Professor Ball's credentials as an expert witness. As I have indicated, these were impressive. At paragraph 49, he began his summary of the professor's evidence:

"49. In his report Professor Ball states that it is actual, rather than perceived, risk that has to be managed. Precise figures are not available, but it appears that there are roughly 800-1000 new cases of spinal injury each year in the UK. The main causes are road accidents (about 80%), falls and disease. In the region of 12% may be related to sports activities such as rugby, horse riding and winter sports. The top sports-related cause of spinal related injury is often cited as "diving" and most agencies suggest that "diving" is responsible for 3.5% of all cases. Therefore the population averaged risk of spinal injury, based on a UK population of 60 million, is in the region of 17 in a million per year, or one chance in 60,000. This, in Professor Ball's opinion, is a fairly modest risk compared with say the risk of death from injury (about 1 in 4000 overall). None of these statistics was challenged on behalf of Mr Uren.

50. Professor Ball's estimate of the average individual risk of serious spinal injury from swimming activities for participating young males (the highest risk group) is 1 in 130,000 per year or 8 in a million. He estimates the risk of paralysis to young males from each session of swimming and diving to be roughly 1 in a million per swimming/diving session. This compares with 1 in 320,000 rock climbs and 1 in 750,000 canoe outings. Again, these statistics were not challenged on behalf of Mr Uren.

51. In Professor Ball's opinion, even for the swimming/diving fraternity, the risk of serious injury when measured on an annual basis, or a per outing basis, is very small. There were, however, no statistics for the pool game played by Mr Uren."

51. Pausing there, Sir Geoffrey agreed that he had not challenged these statistics but said that there was no reason why he would wish to do so. They did nothing to undermine his contention that this pool game as played gave rise to a risk of spinal injury. I would accept that they do not undermine that contention. Their potential relevance is that they demonstrate that there is some risk of spinal injury (and death) in things that we all do and that that risk does not stop us from doing them. That I accept is an important consideration when it comes (as it will at a later stage of the process of decision-making) to deciding whether the risk is acceptable or whether it is such that the activity should be stopped. However, I would accept Sir Geoffrey's submission that those statistics cannot be used to assess the level of risk arising from this game.
52. Whether the judge relied on the statistics for that purpose is not clear. He did not say what he relied on them for. I do not understand the introductory words of paragraph 49 where the judge quotes Professor Ball's view that it is actual, rather than perceived, risk which must be managed. He then goes on to quote the statistics. It sounds as if he is relying on the statistics as evidence of the actual risk whereas other (undefined) sources of evidence might only relate to perceived risk. I think that the reference to the management of risk suggests that the statistics are being given as part of the assessment (and therefore management) of the actual risk.
53. However, it seems clear that those statistics could not be used to help in the assessment of the risks arising in a particular game. First, the number of spinal injuries in the UK each year from all causes does not help to assess the danger of this game. Nor is the number of spinal injuries related to sports activities as a whole. The only figure given which is of any relevance to that issue is that 3.5% of spinal injuries are caused by "diving". Taking 800 to 1000 as the annual figure for all causes, that would be 28 to 35 diving injuries per year. But having noted that potentially useful statistic, the judge goes back to the figures which give the risk faced by the whole of the UK population of spinal injuries from all causes, namely one in 60,000. It is self-evident that that is a fairly modest risk compared with the risk of death by injury at one in 4000.
54. This reference to comparative risks suggests that what the judge might have had in mind is that we all willingly accept some risks of serious injury every day when we travel by car and those of us who are a little more adventurous willingly take similar or possibly rather greater risks for the pleasure of taking part in outdoor activities. If the judge was only making comparisons between the general dangerousness of road travel, swimming, diving, rock climbing and canoeing, that is unexceptionable. If he was using the figures to support his view that the risk from this game is very small, then he is open to criticism. The final words of paragraph 51 could be regarded as some indication that he recognised that the figures are not useful for an assessment of the safety of the game. However, the opening words of paragraph 49 quite strongly suggest that he was relying on them as contributing to his conclusion that the risk from this game was very small.
55. The judge's approach then moved away from consideration of statistics. At paragraph 52, he summarised Professor Ball's approach to his assessment of the risks of a particular activity and his assessment in relation to a head-first entry to this pool:

"Professor Ball explained that when he assesses the risk of an activity he extrapolates from his technical knowledge, experience of life and comparable activities. In his view, an assessment of the risk ex ante of the type of accident that happened to Mr Uren would have forecast it to be very low indeed and he personally would have given it the lowest score on the likelihood scale. Diving into the pool without touching the side would be very difficult to accomplish and obviously dangerous. However, to run up and slide over would be reasonable. It was not necessary to discourage sliding over the side head first with arms outstretched because the friction with the inflatable side would slow entry into the water and the distance of descent would be about 1 metre which, with arms outstretched, would allow the strain to be taken by the arms before impact with the grassy ground below. Grass was a good deal more absorbent than a hard surface such as a floor. For these reasons, the risk of injury, especially serious injury, was very low."

56. In the following paragraph, the judge recorded Professor Ball's opinion that Mr Uren had been extremely unlucky to suffer the injury he did and that the probability of serious injury was 'remote'. He said that he found the Professor to be a most impressive witness.
57. The judge has not made it clear why he regarded the statistics quoted by Professor Ball as important or for what purpose he intended to rely on them. I am not completely persuaded that he has relied on them for any inappropriate purpose. I do, however, accept that the uncertainty as to the use which they were put gives rise to some concern about the judge's reasoning.
58. The judge then turned to deal with the other expert witnesses. He said that he found Mr Petherick a 'considerably less persuasive witness'. This was because Mr Petherick had placed too much reliance on the swimming pool regulations and because he had said that the inflatable was not constructed to be used with water. I would unhesitatingly agree with the judge that Mr Petherick's opinion about the unsuitability of the inflatable's construction was a thoroughly bad point for him to take. His objection could not sensibly be related to the happening of this accident. When an expert witness makes a bad point, it damages his authority and his opinion on other potentially better points is undermined. However, the judge's rejection of Mr Petherick's reliance on the swimming pool regulations and their embargo on diving into shallow swimming pools may be not rationally justified. The judge's reason for his rejection of this opinion was that the inflatable was not being used as a swimming pool and it was impossible for the contestants to dive into the water without sliding over the side. That was so, although it is fair to say that Mr Petherick had accepted that what was in issue was the safety of a fast head- first entry and not the safety of a true dive; that was accepted by all to be thoroughly dangerous. His opinion was that a fast head-first entry was similar or analogous to a dive and that therefore some regard should be had to the swimming pool regulations. The difference between him and Professor Ball on this crucial issue was that Mr Petherick was saying that a running head-first entry might be such that the competitor might not be able to control his entry with his hands; if he lost control there was a risk of serious injury. The Professor was saying that the competitor would be able to retain control. The judge also rejected Mr Petherick's opinion on the lack of absorbent quality of the flat plastic base of the pool lying on a grassy surface. I think that his position was that such a surface may be more absorbent of energy than a hard floor but it was still pretty hard. I can well understand why the judge might have preferred Professor Ball's opinion on that issue; the Professor had special expertise in the safety of children's playgrounds and this included expertise in the absorbency of different surfaces. However, it does not seem to me that that issue is anywhere near as important as the issue of whether the competitor could control his head-first entry or whether there was a risk of loss of control. The judge added that he could not 'escape the conclusion' that Mr Petherick had 'allowed hindsight to creep into his reasoning'. He did not say what aspect of Mr Petherick's evidence had led him to make this remark. In a perfect world, he might have made that remark good by providing an example but I do not think that his failure to do so is a serious omission.
59. It seems to me that, if Mr Petherick's evidence had stood alone against that of Professor Ball, it would be difficult to criticise the judge for preferring the Professor. Mr Petherick had damaged his authority by advancing an untenable argument. Even so, I would have liked to see the judge's reasoned explanation for accepting Professor Ball on the crucial issue of controlled entry. On a matter such as this, one does expect to see some discussion of the technical issues rather than a mere assertion that the judge prefers the view of one witness to that of another.
60. However, Mr Petherick's evidence did not stand alone. Dr Jones agreed with him on the crucial point of controlled entry and he did not damage his authority by associating himself with any of Mr Petherick's less sustainable arguments. Dr Jones was of the view that a fast (running) head-first entry was dangerous because it might not be possible to control the entry using the hands. A slow entry sliding or slithering over the side was reasonably safe. His point was that one could not forbid a fast head-first entry but allow a slow one; therefore both should be forbidden. It was relevant in this context to consider the likelihood that at least some of the competitors were fit young servicemen who would be expected to play this game in a robust and competitive way. The judge did not discuss the merits of Dr Jones' evidence at all. He merely said that he did not find it helpful. He observed that Dr Jones had only recently qualified; in fact he obtained a degree in 2001 and his PhD in 2005. He was clearly much less experienced than Professor Ball and, for that matter, Mr Petherick. Of course, the judge had not had the advantage of seeing Dr Jones and hearing him cross-examined; that might well affect the weight he felt able to attach to his evidence. He did, however, have the quite lengthy accounts of the joint experts meetings where Dr Jones' explained the reasons for his opinion. In my view Dr Jones' relative inexperience, even coupled with the lack of oral evidence, could not amount to a sufficient reason to reject his evidence without considering its merits. This was a claim relating to a very serious injury and the parties (particularly, one might say, the losing party) were entitled to have the evidence carefully considered.
61. I am not saying that, on the crucial issue of control in a fast head-first entry, the judge was wrong to prefer the opinion of Professor Ball to the combined opinions of Dr Jones and Mr Petherick. I am saying only that I do not know why the judge formed his view on this issue because he did not discuss the pros and cons. He simply accepted Professor Ball's view because he was impressive and very experienced and because his evidence was, in an undefined way, 'soundly scientific'. In my view, that is not satisfactory.
62. I have a further concern about the judge's holding that the risk of serious injury was very small. As noted, the judge accepted that SAC Plant who was watching the game immediately thought that the head-first entries which he observed were dangerous and would lead to an accident. The judge said that he thought that what spectators thought of the risks of the game was of very little relevance. I am afraid that I do not agree with him on that. This was a game which no one had seen played in this way before. There was no evidence that anyone from CL had seen it played in this way, permitting head- first entry. No one from the RAF had seen it played until the day of this accident. None of the experts ever saw it being played; they all had to envisage it. It seems to me that the impressions of those who actually saw it that day were potentially important.
63. SAC Plant gave evidence, which the judge accepted, that he immediately thought head-first entry was dangerous. The judge noted that three trained physical education instructors were also watching the game but had seen no reason to intervene and stop it. The judge did not examine their evidence, presumably because he thought that the impressions of witnesses were not relevant. Nor did he examine the extent of their training to see whether it gave them any special insight into the risk of this kind of activity. In any event, I note that of those three, only one, Sgt Thomas, was called to give evidence. He was called by Mr Sweeting and his witness statement and a statement made for the RAF internal inquiry were put in evidence. Neither said anything about his impressions of the safety of the game. He was then cross-examined by Mr Lynagh, for CL who was able to ask leading questions. The witness described his function on that day as that of trouble-shooter and said that he had gone over to the area of the game to see how things were going and to have a chat with the Station Commander. He said that he was watching the game while having a conversation; he was not sure whether he was there for both heats or only the second. He said that people were going into the pool in a variety of ways, some diving, some scrambling, some vaulting and some just climbing on to the top and sliding in on their bottoms. He was then led by counsel and agreed that those who went in head first were going on to the top of the side and were sliding in with their hands out in front of them. Counsel asked: "And can I take it that, from what you say, that did not look dangerous to you?" The witness agreed. Counsel then asked "And if it had, presumably you would have said something?" and the witness agreed that he would have spoken to the organisers. If the judge had thought that the impressions of eye witnesses were of some relevance, I am not sure how much weight he could properly have attached to those answers, extracted as they were by directly leading questions asked by counsel for a party who, on the issue of safety, was running in tandem with the MoD who had called the witness.
64. As for the other PE instructors, Corporal Gaze did not give evidence at the trial. He did give evidence at the internal inquiry. The only part of that evidence which related to events preceding the accident was his description of the briefing given by CL. He said that the organisers had told people to use common sense when getting into and out of the pool but that people were not told not to dive into the pool 'and I commented on this to my colleagues'. He was not asked then (or ever) what he meant by that remark. However, it certainly does not seem to me to amount to evidence that he thought that head- first entry would be safe.
65. I cannot find any evidence from Corporal Williams either given at the trial or to the internal inquiry. I do not think one can safely draw any inference as to what he thought.
66. To complete the picture, there was evidence from a Major Westwood. I am not sure exactly who he was but he was watching the game at the time of the accident and he gave evidence to the internal inquiry although not at the trial. At the internal inquiry he was asked about CL's briefing and about the supervision by CL during the game. He said that no instructions were given during the game, only encouragement. That would be from Mr Berry who was commentating. The words he remembered were "Nice diving, I'm liking that." Asked whether he was concerned over safety he said: "In hindsight maybe I should have approached the game organisers, but looking back it was only the last game that anything dangerous occurred. It was only in the second round that people started diving into the pool".
67. Accepting as I do that the impressions of eye-witnesses are of some relevance to the issue of the safety of head-first entries, I would say that, taking all that evidence together, there was modest support for the appellant's case. This would not be determinative but it does underline the need for a fuller and more critical examination of the views of the experts, including the basis for Professor Ball's opinion, and not merely an acceptance of Professor Ball on the grounds given by the judge.
68. My conclusion on this issue is that I cannot be satisfied that the judge reached a sound and tenable conclusion when he held that the game as played carried only a very small risk of serious injury. I say that for three reasons. First, I am not satisfied that the judge carried out a sufficient analysis of the conflicting opinions of the experts. In particular, he failed to deal with the obvious point that this game was to be played competitively by a group of people, most of whom were fit young servicemen, who might be expected to display a considerable degree of enthusiasm. Second, I think that the judge was wrong to disregard the impressions of eye-witnesses, especially SAC Plant. Third, I am concerned about the use to which the judge put the statistics which he quoted. Considered together, those three reasons give rise to very serious concerns that the conclusion was not sound. I do not say that the judge was wrong, let alone clearly wrong, to hold as he did on this crucial issue. However, I cannot say that his decision is sound. It follows that my view is that this judgment cannot stand.
69. In view of that conclusion, the appeal will have to be allowed and it is not strictly necessary for me to discuss the final aspect of the judge's decision. The judge held that there was a risk of serious injury in the game as played but because that risk was very small and because the game as played had some social value, the respondents were not in breach of duty in allowing it to be played. Balancing judgments of this kind are very much a matter for the trial judge and this court will not interfere with such a judgment unless the judge has made a recognisable error or unless his conclusion is clearly wrong. I wish to make it plain that, if I had been satisfied that the judge's conclusion as to the low level of risk entailed, I would not have interfered with the way in which he balanced that risk against the social benefits of the activity. I confess that I personally would not have assessed the social value of this game in quite such glowing terms as did the judge but I accept that he did not make any error of approach. However, if the judge's conclusion on the degree of risk is unsound, the balancing exercise is affected and the final conclusion must be set aside.

Cross appeal- MoD's risk assessment

70. The MoD cannot and does not suggest that it had no duty to carry out a sufficient and suitable risk assessment. It plainly did. It was subject to a statutory duty under regulation 3(1) of the Management of Health and Safety at Work Regulations 1999 to carry out a suitable and sufficient risk assessment of its activities. The existence of this duty was acknowledged in two 'in- house' booklets which were before the court in which its importance was stressed and guidance was given to commanding officers as to how it should be complied with.
71. It is trite law that the common law duty of an employer to an employee cannot be delegated: see Wilson's and Clyde Coal Co v English [1959] A.C. 604. It seems to me that the duty to undertake a risk assessment is so closely related to the common law duties of the employer that it would be remarkable if the duty to undertake a risk assessment were delegable and yet the general responsibility for safety were not. In my view, the judge was clearly right to hold that the risk assessment duty is non-delegable.
72. I do accept that what amounts to 'a suitable and sufficient' risk assessment may well vary according to circumstances. For example I can see that if an employer uses a contractor for some activity and satisfies himself that the contractor has carried out a thorough risk assessment in relation to that activity, that might well lead to the conclusion that the risk assessment carried out by the employer is suitable and sufficient even though it is not as detailed as would be otherwise required. That would be a question of fact in each individual case and it is impossible to generalise as to the standard of risk assessment which will be required of an employer. Here, on the facts, it is clear that CL did not carry out a suitable or sufficient risk assessment and it could not sensibly be argued that the MoD could properly rely on it. The two defendants did not even confer about risk assessments.

Conclusion

73. For the reasons I have given, I would allow the appeal and dismiss the cross appeal. I would remit the action for retrial by a different High Court Judge. I would direct that the issues in the action should be limited to the question of the degree of risk of serious injury entailed in the game as played and whether that degree of risk was acceptable in the light of the social value of the game. If the judge concludes that the risk entailed in this game was unacceptable, it will be necessary for the judge to apportion responsibility between CL and the MoD. That apportionment should be carried out on conventional lines, after an assessment of the culpability and causative effect of each defendant. I would also direct that, before the action is listed for retrial, the parties should attempt mediation.

Lord Justice Aikens:

74. This is an anxious case. I have concluded, after much thought, that the appeal must be allowed, for the reasons given by Smith LJ. I also agree with her conclusions concerning the cross-appeal of the MoD. As we are differing from the trial judge on his assessment of facts and expert evidence, I will shortly express why I have reached this view.
75. CL and the MoD were under a non-delegable duty to Mr Uren and other participants in the game to take reasonable care to ensure that they were not exposed to an unacceptable risk of serious injury, such as to the spine or the head, whilst taking part in it. (Everyone agreed that there was a risk of minor injury and that was acceptable). Sometimes the nature and extent of risks involved in activities can be considered in a "Risk Assessment". The judge held that CL and the MoD failed to carry out proper Risk Assessments in this case. That finding in itself did not mean that CL and MoD automatically became liable for Mr Uren's most serious injuries. The judge had to decide whether CL and MoD, in permitting the game to be played as it was, so that competitors could lunge into the inflatable pool head-first, failed to take reasonable care to ensure that the competitors were not exposed to an unacceptable risk of serious injury.
76. There can be no legal definition of an "unacceptable" risk in relation to any particular game. It must depend on the activity concerned, the way in which it is being undertaken, who is doing it and in what circumstances. But if in this case the judge had concluded that in all the circumstances the risk of serious injury from playing the game so that participants could lunge head first into the pool was an unacceptable risk, it would be a short step to deciding that a failure by CL and MoD to prevent participants being exposed to that risk constituted a failure to take reasonable care on their part.
77. The judge's first task was therefore to assess what the risks of serious injury were if the game was played as it was. Field J's conclusion that the risk of serious injury was very small was based principally on his view of the expert evidence. The judge had evidence from three experts on the nature of the risks involved in this game being played so that competitors could lunge into the pool head first. Like Smith LJ and for the reasons she gives, I have come to the conclusion that the judge's justification for preferring the expert views of Professor Ball is not satisfactory. With respect to the judge, it is not possible to see from the judgment, in particular [39], that Professor Ball's views were subjected to a sufficiently critical examination such as satisfactorily to explain why the judge preferred Professor Ball's views to those of Mr Petherick and Dr Jones. I have reached this conclusion with great reluctance, because the trial judge heard and saw two of the experts and we have not.
78. I am, like Smith LJ, also concerned with the use to which Field J put the statistics which were provided by Professor Ball and quoted at [49]-[50] of Field J's judgment. The overall statistics concerning spinal injury in the UK are of no assistance in assessing the nature and extent of the risk of spinal injury in playing the game as it was. Nor are the general statistics on serious spinal injury from swimming activities. They are too distant from the facts of the present case, which is unique because the game had not been played in that format before and, I understand, has not been so since this incident. However, the judge seems to have used them in reaching his conclusion that the risk of serious spinal injury from lunging head first into the pool in playing this game was small.
79. Field J also held that the opinions of spectators on the risks of the game as it was played were irrelevant to his assessment of the risks. With respect, I think the judge erred in this respect. In my view their opinions should not have been discounted unless obviously partisan or flippant.
80. Therefore, like Smith LJ, I am reluctantly driven to the view that the bases for the judge's conclusion that the risk of spinal injury was small and was therefore acceptable were not sound and cannot stand. I agree that the action must be remitted for retrial by a different High Court Judge on the issues identified at [73] of Smith LJ's judgment.

Lord Justice Pitchford:

81. I agree with both judgments and with the order proposed.