Dawkins v Carnival plc


Dawkins v Carnival plc


Dominic Lang a Partner in the Shipping Litigation team at Hextalls Limited provides a commentary on the recently reported case of Dawkins v Carnival plc



“When I use a word,” Humpty Dumpty said, in rather a scornful tone, “it means just what I choose it to mean, neither more nor less. “The question is,” said Alice, “whether you can make words mean so many different things.” “The question is,” said Humpty Dumpty, “which is to be master, that’s all.” (Through The Looking Glass)


Amid the twists and turns of the law, it is sometimes forgotten that most cases based in negligence turn first and foremost upon evidence, or the lack of it. Often linked to this is the question of where the material burden of proof lies. All this was recently highlighted by the Court of Appeal in Dawkins v Carnival plc, giving judgment for the Claimant and reversing the decision of the court below in favour of Carnival. However, the case illustrates the Herculean, even Sisyphean, effort often required of a Defendant.


 On 23 November 2006, Mrs Janet Dawkins was a cruise passenger on board the “Oriana”, a luxury cruise ship operated by Carnival under the P&O Cruises brand name. At the time, the ship was in international waters. That afternoon, she was walking through one the ship’s restaurants, the Conservatory, when she slipped and fell, suffering soft tissue injuries to her knees and wrists. She brought a claim against Carnival for damages for her injuries.


 Four years later on 3 December 2010, the matter went to trial before Mr Recorder Wright in the Southampton County Court. Mrs Dawkins had pleaded her case under the Occupiers’ Liability Act 1957 but it was common ground that the claim was governed by the Athens Convention 1974, incorporated into English law by s.183 of the Merchant Shipping Act 1995. Article 3(1) of the Athens Convention provides:


 “The carrier shall be liable for the damage suffered as a result of the death of or personal injury to a passenger ……. if the incident which caused the damage so suffered occurred in the course of the carriage and was due to the fault or neglect of the carrier or of his servants or agents acting within the scope of their employment.”


Article 3(3) provides that “….. the burden of proving fault or neglect shall lie with the claimant.” (emphasis added). It was accepted, however, that the tests applicable under the Athens Convention, including the burden of proof, were those applicable if the accident had happened on land premises inEngland, a significant and important point of principle in its own right.


 From the judgment of the Court of Appeal, delivered by Pill LJ with whom Moore-Bick and Aikens LJJ were in full agreement, the material findings, either as common ground and/or by the Recorder, were as follows.


  1. The accident happened at about 2.00pm. The Conservatory was one of the most popular venues on the ship. Given this and the timing, one might safely assume that it was also reasonably busy.


  1. When she slipped, Mrs Dawkins was close to a drinks dispensing station. The Recorder found that, on the balance of probabilities, Mrs Dawkins “slipped on some liquid, probably water, on the floor of the Conservatory”, which seems a wholly unexceptionable finding. It is not clear whether she was about to have lunch or had just done so, or in either case if she was carrying anything, or was simply passing through. No doubt from tacit chivalry, there is no reference to Mrs Dawkins’ age, nor any suggestion that she was anything other than fully fit and able-bodied. The judgment is also silent about her footwear but no matter how well-heeled financially she might have been, she was probably not in Christian Laboutins at the time. All in all, Mrs Dawkins was not open to any criticism.


  1. Equally, nothing is said about the sea conditions at the time and in particular whether there was any excessive rolling motion of the ship which might have increased the risk of a spillage and perhaps enhanced the duty of care owed by the defendant carrier.


  1. The Recorder found that a crew member should normally be “permanently present” at the station, although there was no evidence whether anyone was actually there when Mrs Dawkins fell. However, there was a “full contingent of staff in the Conservatory at the time of the accident”; the daily report recorded staff status as “full”; and the number of staff in the Conservatory passenger area at the time was between 15 and 24. Only two accidents had occurred there in six years.


  1. The only evidence as to how long the culpable liquid might have been on the floor came from two other passengers, a Mr and Mrs Thomas, who said that “it was not there when they got to the Conservatory, which may have been from 10 minutes to 30 minutes before the accident”. No doubt they were doing their best to be helpful but the vagueness of their recollection was unfortunate as the timing proved to be a crucial factor. The Recorder said that “in theory at least, the spillage could have occurred only moments before, or several minutes before the accident”. In this context, “several” was perhaps an unfortunately vague and flexible term to use, but clearly the Registrar was satisfied that the liquid had not lain on the floor undetected for an excessive time.


  1. He went on to say, however, that “Nevertheless, the possibility of slippage in the cafeteria, and of someone slipping on spilt liquid, are obviously foreseeable. Although a danger such as this could be expected or at least foreseen also by a passenger, nevertheless it obviously imposes on the Defendant a clear duty to operate a reasonably effective system for getting rid of spilt liquid and minimising the danger it may cause. The fact that in the ordinary course of things an accident such as this would not normally happen clearly raises the suggestion that the accident arose from want of care by the Defendant. It will be for the Defendants to discharge an evidential burden to the effect that the accident did not happen through want of care on their part. In effect it is for the Defendants to show that the accident was not the result of any failure by them”.


  1. With an admirable display of common sense not seen everywhere nowadays, however, he continued: “… one could not expect a system of continual inspection of every part of the floor. There are bound to be occasions when liquid is spilt, especially in this area ….. Thus there are bound to be times when there is unguarded spillage on the floor. A properly operated system must take this into account, but no system can completely prevent this”. Can anyone sensibly disagree?


So far, so good. None of the above found disfavour with the Court of Appeal and it was all entirely consistent with the approach laid down by the Court of Appeal in two leading slippage cases, Turner v Arding & Hobbs [1949] 2 All ER 911 and Ward v Tesco Stores [1976] 1 WLR 810. Both cases involved a spillage on the shop floor – of vegetable matter and yoghurt respectively – and also lacked evidence as to when the spillage had occurred. In Ward v Tesco Stores Lawton LJ said: “If an accident does happen because the floors are covered with spillage, then … some explanation should be forthcoming from the defendants to show that the accident did not arise from any want of care on their part; and in the absence of any explanation the judge may give judgment for the plaintiff. Such burden of proof as there is on defendants in such circumstances is evidential, not probative.” Or, as Megaw LJ put it in the same case, “It is for the plaintiff to show that there has occurred an event which is unusual and which, in the absence of explanation, is more consistent with fault on the part of the defendants than the absence of fault …”


In other words, although the initial burden (whether at common law and/or, as with Mrs Dawkins, under the Athens Convention) lies on the claimant to show fault/negligence of the defendant, the mere presence of a spillage on the premises floor may well (and, realistically, usually will) raise a prima facie but rebuttable case of negligence, thereby imposing on the defendant an evidential burden to show otherwise. As Megaw LJ said in Ward v Tesco Stores (albeit in slightly tortuous language): [the Defendants] “could escape from liability if they could show that the accident must have happened, or even on the balance of probability would have been likely to have happened, even if there had been an existence of a proper and adequate system, in relation to the circumstances, to provide for the safety of customers ….” but nevertheless continuing, “if the defendants wish to put forward such a case, it is for them to show that, on balance of probability, either by evidence or by inference from the evidence that is given or is not given, this accident would have been at least equally likely to have happened despite a proper system designed to give reasonable protection to customers.”


That was the nub of Mrs Dawkin’s case – could Carnival displace the prima facie case against them ?


At trial, Carnival presented evidence of their general inspection system. As well as formal weekly inspections, there was an informal daily inspection of food and drinks areas including “walking the floor monitoring cleaning and cleanliness generally”. As one might expect, a Carnival witness said that “spillages are inevitable but would be picked up almost instantaneously” and it is surely safe to accept that a spillage would have been cleared as soon as it was noticed by a crew member or brought to their attention by a passenger, with passengers diverted away from the offending area until it had been cleaned.


Carnival also gave evidence of training given to all crew and the Recorder accepted that “all staff are very safety conscious and are briefed to ensure the safety of themselves and the passengers”. However, and crucially, no evidence was given by any crew member present in the Conservatory restaurant on the day of the accident. According to the Recorder, this was because Carnival “could not establish who it was”, although just who “who” might have been is not clear. The Court of Appeal, however, regarded it as “a significant omission having regard to the number of staff claimed to have been in the area and the involvement very soon after the accident” and the absence of such as “remarkable”.


The Recorder found in favour of Carnival, concluding from the evidence that:


“…. the weight of the evidence shows that the Defendants did all that would be required of them in establishing a proper system for ensuring the safety of their passengers, and that the presence of the water does not on the facts of this case show that they failed in their duty that day. The proper inference in all the circumstances is that even with the best possible safeguards an accident such as this is bound to happen occasionally. I do not consider that it can be inferred that the accident occurred through want of care by the Defendants who have in my view discharged insofar as it possible to do so their evidential burden of showing no lack of care.”


In the Court of Appeal, Counsel for Mrs Dawkins submitted that, on the basis of Turner v Arding & Hobbs and Ward v Tesco Stores, this was not enough to allow Carnival to evade liability. They had to establish not merely that a reasonable system was in place but that they actually took all reasonable care on the day of the accident itself (emphasis added) – in other words, that the system was properly implemented at the time. Not a particularly controversial proposition and one with which the Court of Appeal agreed. The fly in that otherwise unsullied ointment, however, was: How were Carnival to meet that requirement?


According to Pill LJ, the facts imposed on Carnival “an evidential burden to show how long the water had been on the floor. On the balance of probability, it can be inferred that it was a passenger rather than a member of staff who was responsible for the spillage.” Pausing there, it seems harsh to attribute the cause of the spillage to a passenger rather than to a crew member merely on a balance of probability – such evidence as was adduced about system and training makes it reasonable to suppose that if a crew member had been responsible, the spillage would have been cleaned immediately and the accident would not have happened. Be that as it may, Pill LJ continued: “If, through another passenger’s lack of care, it had only been there for a very short time, the respondents may escape liability. If, on the other hand, it had been there for a significant period of time, they were negligent in failing to have taken remedial action.” (emphasis added). Pill LJ interpreted the Recorder’s conclusion as based on an inference “first that the system was operating effectively at the time of the accident and, secondly, that, since the system was operating effectively, the water could not have been on the floor for a significant time.”


Did the Court of Appeal not like that !! After referring to the “remarkable” (a term pregnant with hidden meaning?) absence of evidence from “one or more of the many members of staff claimed to be present in the Conservatory”, Pill LJ concluded as follows.


“In my judgment, in the absence of evidence from members of staff claimed to be implementing the system, the judge was not entitled to infer from the existence of a system that the spillage which led to the fall occurred only a few seconds, or a very short time, before the accident. The claim succeeds on the evidence in this case. There is nothing to suggest such closeness in time between the spillage and the accident as would, at a place where close observation was required, exclude liability. In the absence of evidence to the contrary, I can conclude only that on a balance of probabilities the water had been there for longer than the very brief period which, in this particular place, would have excused the defendants from taking remedial action before the accident.” (all emphasis added).


On the face of it, the obvious lesson to draw from this case is the critical importance of evidence. It was the complete absence of evidence from any crew member working in the Conservatory at the time of the accident which, for the Court of Appeal, was a decisive factor. We do not know why no such evidence was called, but even if a suitable witness could not be identified (perhaps the person who was closest to the drinks station at the time of the accident), one might have expected Carnival to call someone (perhaps the restaurant supervisor) to confirm that in principle the system was being observed at the time, or at least that, to the best recollection, there was no reason to believe that it was not. The Court of Appeal makes no reference to any on-board accident report from which it should perhaps be assumed that no such document featured in the Recorder’s judgment. In any event, Carnival seem implicitly to have asked the Court to assume that, in the absence of evidence to the contrary, the system was operating effectively. Any adverse inference which the Court of Appeal might tacitly have drawn from the lack of such evidence is perhaps understandable. But, on closer analysis, what does this say about the burden of proof which the Court of Appeal retrospectively imposed on Carnival and the evidence which was required to meet it?


Let us look at it another way. In a personal injury case, except where liability is strict on proof of the necessary facts, a fair follow-up to a decision for the Claimant is to ask what, with hindsight, the Defendant should or could have done, or not done, to avoid liability and, if that can be identified, whether this was truly practicable. Usually the answer is pretty obvious. However, in the case of Mrs Dawkins, it is as opaque as the depths of the ocean below the “Oriana” at the time of her accident.


According to the Court of Appeal, once Mrs Dawkins had made out a prima facie case – satisfied by the mere fact of her slipping on a spilt substance, liquid or otherwise – it was for Carnival to displace that case by showing not only that they had a proper system in place but that it was being properly implemented at the time. Fair enough – no-one can take issue with that proposition. But is that not effectively what the Recorder had found? Did he not effectively accept that, on a balance of probabilities, the liquid had not been on the restaurant floor long enough for him to decide that the monitoring system in place allowed carnival to avoid liability?


True, if the system required the permanent presence of a crew member at the drinks station and in fact there was no such presence at the material time, and there was no evidence specifically to explain this, then one might question the Recorder’s decision for this reason. But the Recorder – who saw and heard the witnesses – did not attach any material significance to this and he was satisfied on the evidence before him that in principle there was no reason to infer that the general system was not being implemented at the time of the accident. More to the point, it played no explicit part in the Court of Appeal’s reasoning, who did not see fit to criticise the evidence which Carnival did present or the Recorder’s acceptance of it. Equally, the Court of Appeal chose to ignore the evidence, presumably incontrovertible, that there had been only two accidents in the Conservatory over six years, surely not a wholly insignificant factor.


If the absence of evidence that a crew member was present at the drinks station or to explain why no-one was there at the critical time, with whatever adverse inference might be drawn from this, had been the reason for overruling the Recorder, then it would at least have been understandable, even if it would be tantamount to imposing strict liability. But, rightly concerned with what happened, or did not happen, on the day of the accident, the Court of Appeal attached the utmost importance to the time between the spillage and the accident. Carnival accordingly faced “an evidential burden to show how long the water had been on the floor.” But how were Carnival to discharge that burden? The Court of Appeal said that Carnival might avoid liability if it could show, by either direct evidence or by necessary inference from indirect evidence, that the accident happened within either “only a few seconds or a very short time” (the latter obviously more than a few seconds but otherwise as flexible as the Recorder’s “several minutes”) before the accident.


What does this mean? A few seconds is tantamount to instantaneous, requiring the system to ensure that a spillage was actually seen and cleaned as soon as it occurred. Such an obligation would be patently absurd – if such systems were possible, very few slipping accidents would happen, and the Court of Appeal did not go so far as to demand this. But despite the get-out offered, the Court of Appeal seemed to conflate the existence of a safe system and its implementation, such as to demand that the system will ensure remedial action taken quickly enough to prevent the risk of anyone slipping. In practice, that may amount to much the same thing as instantaneous action which we have just dismissed as absurd. Indeed would it not be close to strict liability upon proof of the accident?


That seems the logical conclusion from the Court of Appeal judgment. But the inherent fallibility of this is easily demonstrated. Which is the safer and better system – one which requires an inspection of a risk-area every say three minutes or one which has an inspection only every say half hour? The answer is obvious. But, even though one is a good system but the other patently inadequate probably sufficient, this would be immaterial if liability turns on the time between spillage and accident. If, under the latter system, a spillage occurred moments before an inspection, it would be cleared immediately and the risk of accident removed. However, under the first and better system, if a spillage occurs unseen only moments after an inspection of the area, the risk of an accident will exist for the remainder of the time until the next inspection is due.


If Carnival had been able to say when the spillage occurred, then their system – which the Recorder accepted as adequate – would necessarily have caused it to have been cleared immediately in which case it must follow that the accident would not have happened in the first place !


To try to clarify the confusion, we turn to Humpty Dumpty. What, we ask him, is meant by “significant period of time” or “a very short time” or “such closeness in time” or “a very brief period of time”, all terms used by the Court of Appeal as directly relevant to the finding of liability against the cruise company. The answer according to our egg-shaped friend is that they mean exactly what the Court of Appeal chooses them to mean, neither more nor less and without need for further explanation. They, after all, are the masters.


Dominic Lang



About Dominic Lang