A1/2003/1825 (A)
ON APPEAL FROM TECHNOLOGY AND CONSTRUCTION COURT
BIRMINGHAM DISTRICT REGISTRY
(HER HONOUR JUDGE KIRKHAM)
Royal Courts of Justice
Strand
London, WC2
B E F O R E:
THE VICE-CHANCELLOR
Sir Andrew Morritt
LORD JUSTICE BUXTON
LORD JUSTICE LAWS
SIX CONTINENTS RETAIL LTD
Appellants/Claimants
-v-
CARFORD CATERING LTD
First defendants and Part 20 defendants/First Respondents
R BRISTOLL LTD
Second Defendants and Part 20 Claimants/Second Respondents
(Computer-Aided Transcript of the Stenograph Notes of
Smith Bernal Wordwave Limited
190 Fleet Street, London EC4A 2AG
Tel No: 020 7404 1400 Fax No: 020 7831 8838
Official Shorthand Writers to the Court)
MR STEPHEN BOWLES QC (instructed by Eaton Ryan Taylor of Birmingham) appeared on behalf of the Appellants
MR G SAMPSON (instructed by Beachcroft Wansbroughs of Birmingham) appeared on behalf of the First Respondents
The Second Respondent was not represented and did not attend
J U D G M E N T
LORD JUSTICE LAWS: This is an appeal brought with permission granted by Lord Justice Potter on 16 September 2003 against the decision of Her Honour Judge Kirkham, sitting in the Technology and Construction Court at Birmingham on 29 July 2003, when she dismissed the appellants' claim for damages against the respondent.
The appellants were owners and operators of the Timberdine Restaurant in Broomhall, Worcestershire. It was what was called a Harvester restaurant. On 9 March 1997 there was a fire at the restaurant. It began behind a gas fire chicken rotisserie in a timber stud partition wall to which the rotisserie had been fixed. The fire caused property damage reflected in the sum of £54,974.39 in due course agreed between the parties as the appropriate quantum of damage subject to interest. The trial before Her Honour Judge Kirkham and this appeal have been concerned only with the issue of liability arising in the action.
In 1996 the appellants undertook works of refurbishment in the restaurant to turn it into a Harvester restaurant. They engaged architects, and a contractor, Stan Randell & Co. The appellants had adopted templates which had been produced for the design of kitchens intended to be used in Harvester restaurants. These had been designed by a firm called Advance Catering Equipment and had been adapted to suit the Timberdine building by the appellants' architects Symms Revill.
The respondents and first defendants in the action, Carford Catering Equipment Ltd, were engaged by the appellants as project managers for the design and installation of the kitchen catering equipment. I must deal in due course with the relevant provisions of the contract between the appellants and the respondents. The appellants specified the kitchen equipment they wanted and identified the supplier. The respondents prepared the necessary documents to order the equipment. It included a Libra spit roast rotisserie to be supplied by R Bristoll Designs Ltd who were to become the second defendants in the action. The rotisserie was delivered to the restaurant and the respondents arranged for its installation, in which activity Bristoll were not involved. It could be table mounted or hung on wall brackets. It was decided that it should be hung. It was delivered in July 1996. The respondents thereafter arranged for it to be hung on wall brackets which were fixed to a wall designed by the architects. That was a timber stud wall faced with plywood; the plywood was then tiled.
Practical completion took place on 12 August 1996 and the restaurant opened for business on 15 September 1996. This type of rotisserie had been installed in a number of Harvester restaurants. During 1996 Mr Bristoll of R Bristoll Designs Ltd became aware of a recurring problem with the rotisserie burners. Apparently they were failing to generate sufficient distribution of heat. This could be dealt with by means of a modest modification. At the beginning of January 1997 Mr Bristoll received a call from the respondents requiring Bristoll under the terms of the contractual warranty issued by them to investigate a complaint from the Timberdine Restaurant relating to the rotisserie there. It was concerned with the problem with which Mr Bristoll was by then familiar, that of insufficient heat distribution.
Mr Bristoll himself was not qualified or registered to undertake work on the rotisserie. He went to the Timberdine Restaurant on 7 January 1997 accompanied by Mr Soley, a gas engineer from Acorn Catering Equipment who were registered gas engineers recommended by the respondents. At the premises Mr Bristoll and Mr Soley together lifted the rotisserie off the wall. It was apparent to them that three tiles on the wall behind the rotisserie had fallen and become wedged behind the unit. These three tiles were left in the kitchen. Mr Soley carried out the necessary modification to the burners and then tested the rotisserie. He and Mr Bristoll lifted it back on to the wall brackets. The whole job only took about an hour. At the trial there was to be a contested issue of fact as to whether there were visible signs of burning when Mr Bristoll and Mr Soley took the unit off the wall on 7 January 1997. Mr Bristoll said there were not. Mr Soley had said in a witness statement that there were. The judge found that there were not (paragraph 35 of her judgment). There is no challenge to that finding.
On the same day, 7 January 1997, Mr Bristoll sent a fax message to Mr Richard Weller who at the time was the respondents' contract manager. That message included the following:
"The spit roast unit is mounted on a partition wall (not solid) and 3 tiles between and just above the 2 wall brackets had come away from the wall - we do not know whether this was due to a) deflection of the wall due to the weight of the unit - or b) heat effect from the burners.
If it was heat we recommend fitting a stainless sheet at the back of the unit to prevent a fire risk. Note - We did not replace these tiles."
On 21 January 1997 Mr Weller wrote to Mr White who was a project manager at the appellants' construction department. The letter included this:
"Please find enclosed a copy of the engineer's report after a recent service call at the above house on the spit roaster."
I interpolate, that is reference to the fax of 7 January.
"Could you please advise us what action, if any, you wish us/the builders to take."
Mr White's evidence was that he did not see this letter but I understand it to be undisputed that the appellants received it. They did not however reply to it or give any instructions whether to the respondents or anyone else in response to what it said.
There are one or two further events I should describe before coming to the fire. There was a three-months defects review meeting held on 5 February 1997 and attended by Mr White, Mr Humber of the respondents and representatives of Stan Randell and Symms Revill. An action list was drawn up which included a requirement to re-fix loose tiles behind the rotisserie. At some point, as the judge recorded, someone placed kitchen foil at the back of the rotisserie. The judge found (at paragraph 30) that the three tiles were replaced and the foil added at some time between the defects meeting on 5 February 1997 and the fire.
The fire took place on 9 March 1997, beginning in the timber stud wall on which the rotisserie was hung. The judge went into the issue of the cause of the fire in some detail. She noted (at paragraph 15) the undisputed fact that Bristoll had -
"15 ..... supplied installation and servicing instructions [which] provided that -
'under no circumstances must the unit be fixed directly on to a combustible or heat sensitive surface."
She referred also (paragraph 16) to guidance contained in what was called Approved Document J issued pursuant to building regulations and concerning the fitting of heat producing appliances. Paragraph 3-20 of that document was potentially material to the facts of this case. It provided that the back of the appliance should be separated from any combustible surface by (a) a shield of non-combustible material at least 25mm thick or (b) an air space of at least 75mm. The rotisserie at the Timberline was fixed to the wall with a gap of only 25mm and without any shield of non-combustible material.
The judge referred to evidence as to the possibility that there had been a minor gas leak in the pipework which had burned a hole in the wall. She recounted testimony given by the single joint expert Mr Anderson of the well known firm of Burgoynes. Her conclusion as to what I would call factual causation was as follows (at paragraph 32):
"32 Mr Anderson's starting point is that the heat from the rotisserie caused the fire. It appears that the heat was causing tiles to become detached before 7 January 1997. After that date, the heat output increased, because the burners had been adjusted in order to increase the heat. At some point thereafter, foil was added. That further increased the heat from the unit. And at some point a gas leak had also caused the temperature to increase. But Mr Anderson cannot quantify the effect of the additional factors. I conclude that, whilst these additional factors were significant, there was a risk of fire by reason of the way in which the rotisserie had been fixed to a combustible wall and without the necessary clearance or heat shield. Had the unit been mounted with the required clearance, there would have been a significant reduction in the amount of heat applied to the wall. It seems to me likely that the heat generated by the rotisserie was the cause of the fire."
The appellants brought proceedings against the respondents in contract and in negligence and, by amendment, added Bristoll as second defendants. The case against Bristoll was to the effect that they failed to take various proper or appropriate steps after Mr Soley had, on 7 January 1997, noted that the wall behind the rotisserie was showing signs of burning. The judge found that there were no such signs. Bristolls were accordingly absolved and have taken no part in this appeal.
By their contract with the appellants which concluded in October 1995 the respondents undertook to provide "project management services for the design of installation of kitchen catering equipment". The scope of their duties was prescribed in a document setting out the responsibilities of the project manager. It included the following obligations quoted by the judge at paragraph 38:
"'1 Produce a design/layout of relevant concept in line with architect's drawing .....
2 Attend development sites, and take all necessary site measurements and designs as required for progression of design layouts;
3 Review contractors' drawings and check against design/catering services requirements;
4 Check and re-affirm on site actual construction details and services for catering equipment and identify any areas of concern/variance to issued requirements immediately ..... "
The judge referred to the architects' drawing 106 which made it clear that the wall to which the rotisserie was to be fixed was a timber stud wall. She accepted however (at paragraph 42) that Mr Humber of the respondents did not know that the wall in question was of that kind. She accepted also (paragraph 46) that the respondents were not obliged to have regard to the building regulations. But she made these findings (paragraph 44):
"44 In my judgment, Carford should have taken steps to look at drawings relevant to the installation of equipment. That was an obligation which arose under the terms of the contract between the claimant and Carford. It included an obligation to look at the drawings available on site. By their failure to check the drawings, Carford were in breach of their contractual obligations to the claimant. Had Mr Humber checked the drawings, he would have seen drawing 106 and thus would have seen that the wall in question was a timber stud wall. He would have recognised that there was a potential fire risk. He acknowledged that, had he known that this was a timber wall, he would have acted differently in relation to the fixing of the rotisserie."
At paragraph 47 she said:
"47 The installation instructions for the Libra rotisserie note that under no circumstances must the unit be fitted directly on to a combustible or heat sensitive surface. The claimant's case is that Carford should have been aware of and had regard to those instructions. As I have found, Carford did not know that the wall was combustible. But their duties under the contract required them to check. Had they done so, they would have appreciated that the wall was combustible. They should then have paid heed to the installation instructions and taken steps to ensure that the rotisserie could be safely fitted to the wall. While their failure to comply with their contractual obligation to check on site actual construction details, Carford were in my judgment in breach of their contract. It follows that Carford were in breach of their contractual obligations by not installing the rotisserie in accordance with the manufacturer's instructions."
The judge then turned to the appellants' case in negligence against the respondents. That case was that the respondents owed a continuing duty to see that the rotisserie was installed safely. The judge held at paragraph 54 that the respondents had discharged their duty of care owed to the appellants by sending with the letter of 21 January 1997 a copy of Bristoll's faxed warning of 7 January 1997. The judge then held further (paragraph 58) that the appellants' failure to respond to the respondents' letter of 21 January 1997, and to take steps themselves to ascertain whether the rotisserie was safe, amounted to a break in causation as between the respondents' breach of contract and the fire. She said (paragraph 58):
"58 Had the claimant taken steps to investigate the matter, they would have been able easily to have ascertained that the rotisserie had been mounted on a combustible wall, and they would have been able easily to procure that it be operated safely. Although Carford's breach of contract set the scene, the cause of the fire was not any breach on the part of Carford, but the claimant's failure to act on the warning. Accordingly, the chain of causation is broken."
Further, the judge held in the alternative (paragraph 61) -
"61 ..... that there was overwhelming contributory negligence on the part of the [appellants]."
At counsel's request after judgment, that was clarified so as to amount to a finding of 100 per cent contributory negligence.
On this appeal the appellants complain of all three adverse findings, that is to say: (1) as to the respondents' discharge of his duty, (2) the breaking of the chain of causation and, (3) contributory negligence.
I find it convenient first to consider the judge's view that the chain of causation flowing from the respondents' breach of contract was broken by the appellants' failure to take any action on receipt of the letter of 21 January 1997.
A break in the chain of causation for the purpose of claims in contract and tort is commonly said to occur when unforeseeable extraneous event occurs after the breach of contract or of duty so as to have the effect of preventing any loss from flowing from the breach. But a formulation of that type merely begs the question what will count as stopping loss from flowing. In Galoo v Bright Grahame Murray [1994] 1 WLR 1360, cited in the appellants' counsel's skeleton argument, Lord Justice Glidewell held, after citing learning from Australia, that the question how the court decides whether a proven breach of duty was the cause of the loss claimed or merely the occasion for the loss was to be answered by the application of the court's common sense (page 1375 A). But that would not appear to take us very far and, with respect, an appeal to common sense is sometimes apt to be little more than an alibi for want of principle.
There is an underlying difficulty in the very concept of a break in the chain of causation. Whether there has or has not been such a break looks very much like a question of fact. Indeed in this very case Mr Sampson, for the respondents, has been at pains to submit that the judge's finding on the point, being a finding of fact, ought not to be interfered with by this court. But a finding as to a break in the chain of causation is not, or at least is not purely, a finding of fact because it is by no means value-free. The real question is whether on the proved or admitted facts the respondents should or should not be held responsible for the appellants' loss. If one looks at it that way one sweeps aside the metaphysics of causation.
Such an approach is, it seems to me, commended by material in their Lordships' opinions in Fairchild [2003] 1 AC 32. In that case their Lordships' House was concerned with a personal injury action in which the claimants had suffered asbestos dust disease, and there were multiple defendants. I will cite two paragraphs from the opinion of Lord Bingham and from Lord Hoffmann. Lord Bingham said at paragraph 12:
"My noble and learned friend Lord Hoffmann has, on more than one occasion, discouraged a mechanical approach to the issue of causation. In Environment Agency (formerly National Rivers Authority v Empress Kahn Co Ltd [1999] 2 AC 22 at 29, he said:
'The first point to emphasise is that common sense answers to questions of causation will differ according to the purpose for which the question is asked. Questions of causation often arise for the purpose of attributing responsibility to someone, for example, so as to blame him for something which has happened or to make him guilty of an offence or liable in damages. In such cases, the answer will depend upon the rule by which the responsibility is being attributed.'
More recently, in Kuwait Airways Corporation v Iraqi Airways Co (Nos 4 and 5) [[2002] 2 AC 883, 1106] paragraph 128 he said:
'There is therefore no uniform causal requirement for liability in tort. Instead, there are varying causal requirements, depending upon the basis and purpose of liability. One cannot separate questions of liability from questions of causation. They are inextricably connected. One is never simply liable; one is always liable for something and the rules which determine what one is liable for are as much part of the substantive law as the rules which determine which acts give rise to liability."
Lord Bingham then proceeded to set out a passage from a judgment of mine which I hope I may be forgiven for repeating. His Lordship said:
"Laws LJ was reflecting this approach when he said in Rahman v Arearose Ltd [2001] QB 351 at 367-368 [paragraph 33]:
'So in all these cases the real question is, what is the damage for which the defendant under consideration should be held responsible. The nature of his duty (here, in the common law duty of care) is relevant; causation, certainly, will be relevant - but it will fall to be viewed, and in truth can only be understood, in light of the answer to the question: from what kind of harm was it the defendant's duty to guard the claimant? ..... Novus actus interveniens, the eggshell skull, and (in the case of multiple torts) the concept of concurrent tortfeasors are all no more and no less than tools or mechanisms which the law has developed to articulate in practice the extent of any liable defendant's responsibility for the loss and damage which the claimant has suffered.'
13 I do not therefore consider that the House is acting contrary to principle in reviewing the applicability of the conventional test of causation in cases such as the present. Indeed, it would seem to me contrary to principle to insist on application of a rule which appeared, if it did, to yield unfair results. And I think it salutary to bear in mind Lord Mansfield's aphorism in Blatch v Archer [1774] 1 Cowp 63 at 65, quoted with approval by the Supreme Court of Canada in Snell v Farrell (above) [[1990] 2 Supreme Court Reports 311/328]:
'It is certainly a maxim that all evidence is to be weighed according to proof which it was in the power of one side to have produced, and in the power of the other to have contradicted.'"
Lord Hoffmann said at paragraphs 51 and 52 in the same case:
"51 First, in what sense is causation a question of fact? In order to describe something as a question of fact it is necessary to be able to identify the question? For example, whether someone was negligent or not is a question of fact. What is the question? Is it whether he failed to take reasonable care to avoid such damage as a reasonable man would have foreseen might result from his conduct. That question is formulated by the law. It is the law which says that failure to take reasonable care gives rise to liability. And the question is then answered by applying the standard of conduct prescribed by the law to the facts.
52 The same is true of causation. A question of fact is whether the causal requirements which the law lays down for that particular liability have been satisfied. But those requirements exist by virtue of rules of law. Before one can answer the question of fact, one must first formulate the question. This involves deciding what, in the circumstances of the particular case, the law's requirements are. Unless one pays attention to the need to determine this preliminary question, the proposition that causation is a question of fact may be misleading. It may suggest that one somehow knows instinctively what the question is or that the question is always the same. As we shall see, this is not the case. The causal requirements for liability often vary, sometimes quite subtly, from case to case. And since the causal requirements for liability are always a matter of law, these variations represent legal differences driven by the recognition that the just solution to different kinds of case may require different causal requirement rules."
That a finding of a break in the chain of causation is an evaluative exercise, not merely a finding of fact simpliciter, is I think starkly illustrated in the present case when one calls to mind that the judge's factual finding on causation at paragraph 32, which I have set out, shows that as a matter of fact the respondents' breach of contract, essentially fixing the rotisserie too close to a combustible wall, was certainly instrumental as a cause of the fire.
The true question here is whether the appellants' failure to respond to the letter of 21 January 1997 ought to absolve the respondents of what would plainly otherwise be their responsibility for the fire. For my part, I think it plain that the risk of a fire of this kind was, on the facts, well within the scope of outcomes which the respondents' contractual duties were intended to avoid. So much appears, I think, from the catalogue of obligations cited by the judge at paragraph 38 which I have set out. Accordingly, even if the letter of 31 January 1997 and the enclosed fax did constitute a warning of a risk of fire of the kind which occurred on 9 January 1997, a question to which I will come in a moment, it was a warning of an outcome which the respondents themselves should have prevented from happening. I find it very difficult to see how the giving of such a warning ought to transpose the burden of avoiding that very outcome from the respondents, who owed a duty in effect to prevent it, to the appellants who were the beneficiaries of that duty.
But in any event I do not consider that the letter was a warning - certainly not a sufficient warning - that there was a risk of fire happening as this fire happened. The wording of the letter did not in terms amount to a warning at all. Indeed the expression "please advise us what action, if any, you wish us/the builders to take", suggesting that action was optional rather than necessary, is all but inconsistent with the notion of a warning; and the enclosed fax is, to say the least, indefinite as to fire risk. To constitute a proper warning the respondents must have drawn attention - in terms, or at least very plainly - to the fact that the unit was fixed directly on to a combustible surface; and that, as I have foreshadowed, would have been a warning of the respondents' own breach of contract.
In my judgment for these reasons the judge below was wrong to find that the chain of causation as between the respondents' breach of contract and the fire was broken. Her errors were errors of principle, and so Mr Sampson, despite his energetic efforts this morning, can take no comfort as I see it from any suggestion that the judge was merely expressing a particular view of the facts.
These same considerations really conclude, in the appellants' favour, a further question - whether the judge was right to hold that the respondents' duty in tort was discharged by the letter of 21 January 1997. That duty cannot have been discharged by notifying the appellants that a risk of fire had arisen from the respondents' own want of care.
There is, lastly, in my judgment, nothing in the circumstances here to give space to a finding of contributory negligence. Mr Sampson merely seeks to uphold the judge's view that a finding of 100 per cent contributory negligence would be appropriate. That finding must fall with the judge's earlier conclusion. Mr Sampson has not in terms contended for some more moderate finding or conclusion in negligence, as to which possibility therefore it is not necessary to say any more.
For all the reasons I have given, I would allow this appeal.
LORD JUSTICE BUXTON: In his sustained submissions before us this morning Mr Sampson urged that the judge's conclusions as to causation involved questions of fact or, if not purely fact, fact and judgement of a sort with which this court should not interfere. In order to assess that submission it is necessary to go back to the guidance given by the House of Lords in Fairchild, to which my Lord has referred. I will venture to repeat - because I think it is of particular importance in the context of this case - a short passage from the opinion of Lord Hoffmann to be found in paragraph 52 of that report:
"The question of fact is whether the causal requirements which the law lays down for that particular liability have been satisfied. But those requirements exist by virtue of rules of law. Before one can answer the question of fact, one must first formulate the question. That involves deciding what, in the circumstances of the particular case, the law's requirements are."
The judge did not direct herself in such terms with the result that she did not articulate the particular rule or question to which the causal issue was relevant. Had she approached the matter in that way she would have inevitably been referred back to her own reasons for finding breach of contract and breach of duty on the part of Carford, as set out in particular in paragraphs 43 and 47 of her judgment which my Lord has already mentioned. In the light of that conclusion - that Carford were in breach of their contractual obligations by not installing the rotisserie in accordance with the manufacturer's instructions - the issue before her was therefore in what circumstances, if any, would it be possible for a notification or warning by Carford of its own breach discharge its continuing liability for damage caused by that breach.
For that to be achieved, any warning would as a matter of law have to be overwhelming and plainly effective before it could excuse Carford. Carford would in effect have had to make it its own business to ensure that the breach was nullified, and have been frustrated in that attempt by a lack of response by the building owner. The judge did not look at the question in that way. She seems to have assumed that the question was simply one of a breach of chain of causation looked at in general terms and that any warning that could be sufficiently described as such would suffice for that purpose.
That approach did not link the causal question with the question of liability in the way urged by Lord Hoffmann. Had the judge approached the matter with the nature of Carford's negligence in mind, she could not have reached the conclusion in paragraph 53 of her judgment that I will quote:
"I am unimpressed by the argument run by the claimant that Carford should have made it clearer in their letter that Mr Bristoll had identified a risk. A copy of Mr Bristoll's fax was attached. It explained the problem and possible risk in clear wording. That wording did not need any clarification or explanation by Carford to make it comprehensible. The claimant was as able as Carford to read and understand Mr Bristoll's words. It was for the claimant to decide what action was needed, and to take that action."
In truth, however, it was not for the claimant to decide what action was needed and to take that action; it was for Carford. It was for Carford to see that action was taken, either by itself or by someone else. It is not correct to say that the wording did not need any clarification or explanation by Carford. Certainly it did not need such explanation in order to make it comprehensible as a piece of English, but it needed that explanation in order to make it relevant in the circumstances, in particular by revealing, which never was revealed, that the rotisserie had been installed contrary to the manufacturer's instructions and in a way that the manufacturer accepted raised a serious fire risk.
Somewhat similarly, looking at the separate question (indeed if it is a separate question) of the breach of chain of causation, similar considerations necessarily impose themselves. The nature of the risk and the gravity of the breach are the important questions. Those the judge never addressed.
For those reasons therefore, which are merely a footnote to my Lord's judgment with which I agree respectfully, I would dispose of this matter in the way he proposes.
THE VICE-CHANCELLOR: I agree. This appeal will be allowed for the reasons given by each of my Lords.
Order: Appeal allowed