England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
TECHNOLOGY AND CONSTRUCTION COURT
His Honour Judge Bowsher QC

		Royal Courts of Justice Strand, London, WC2A 2LL
		19 December 2003

Lord Justice Clarke:

Introduction

1. This is the judgment of the court. On the night of 4 January 1998 a fire destroyed most of a food factory in Brent Road in Southall. At the time of the fire Sahib Foods Ltd ("Sahib") was the lessee of the building and Co-operative Insurance Co Ltd ("Co-op") was the owner of the freehold. The fire started in oil in a gas bratt pan in room G49. His Honour Judge Bowsher QC ("the judge") held that the fire was caused by the fault of Sahib. He further held that the spread of the fire outside room G49 was due to breach of contract and duty on the part of Paskin Kyriades Sands ("PKS"), who are a firm of architects. The total loss as a result of the fire was said to be some £27 million inclusive of interest.
2. The trial which has given rise to the appeal was concerned only with liability. Before the judge there were issues as to whether there was a contract between Sahib and PKS and as to whether PKS owed Sahib a duty to exercise reasonable care and skill as architects. The judge resolved both those issues in favour of Sahib and PKS do not challenge that part of his decision in this appeal.
3. The judge held that PKS were in breach of contract and duty in that they failed to exercise reasonable care and skill in order to prevent the spread of any fire in room G49 outside it. Although Sahib conceded at the trial that it was guilty of contributory negligence of the order of 50 per cent, the judge held that there was no relevant contributory negligence, essentially because, although Sahib was responsible for the fire, PKS was solely responsible for the spread of the fire. In this appeal PKS say that the judge was wrong to hold that they were negligent or in breach of contract as alleged or at all. Alternatively, they say that, if they are in principle liable to Sahib, the judge erred in two respects with regard to contributory negligence. First, they say that he was wrong to refuse to reflect Sahib's admitted fault in causing the fire and secondly, they say that he ought to have held that Sahib was also partly to blame for the spread of the fire and to have reflected both types of fault in an apportionment of damage under section 1 of the Law Reform (Contributory Negligence) Act 1945 ("the 1945 Act").

The Fire

4. There is no issue as to the cause of the fire. It started in room G49, which was in what was known as the veg prep cook area of the factory. There were three bratt pans in room G49, two of which were heated by steam and one by gas. The bratt pans in question were large, open, rectangular pans with sides far deeper than an ordinary frying pan, installed at waist height, each being on a sub-structure standing clear of the walls. The gas bratt pan was heated by gas jets within the sub-structure beneath the pan. Its dimensions were 650 mm x 730 mm x 250 mm deep. On the night of 4 January 1998 a Mr Singh was frying aubergines in the gas bratt pan. The thermostat was broken. When Mr Singh had finished frying the aubergines he left the pan on the heat, empty except for the remaining oil, without turning the gas off. The oil in the pan continued to heat up until it reached its flashpoint of about 320ºC and caught fire.
5. The causes of the fire and its subsequent spread were agreed by the fire experts instructed by the parties. They were Dr Neil Sanders of Burgoynes, who was called by Sahib, and Dr John Bland, who was called by PSK. The following points of agreement and disagreement between them are relevant to the issues in the appeal:

i) It is unlikely that ignition occurred within less than about one hour from the end of cooking at a nominal temperature of 180ºC.

ii) In any of the following events the fire would not have occurred:

a) if the bratt pan had been switched off immediately after the appliance was last used or at the end of the shift;

b) if the bratt pan thermostat had been functioning correctly and had been correctly located at the time of the fire; and

c) if the bratt pan had been fitted with an overheat limit thermostat which was functioning correctly and correctly located.

iii) The bratt pan was about 600 mm from the side wall, which was constructed of EPS panels, which were not fire resistant.

iv) It was Dr Sanders' view that the panels would probably have ignited in about 3.5 minutes and ignition may have occurred more quickly. Dr Bland believed that ignition would probably not have occurred in less than 3.5 minutes, whereafter there would have been an increasing risk of ignition. This difference between the experts is not significant in the context of the issues between the parties.

v) At an agreed burn rate of 1.75 mm per minute, an oil depth of about 6 mm would take about 3.5 minutes to burn.

vi) If the veg prep cook enclosure in room G49 had been constructed as a 'fire compartment' capable of containing a fire for one hour, the fire would not have spread outside room G49.

vii) G49 was not so constructed. The main features of such an enclosure would have been walls and ceilings constructed from non-combustible panels (such as mineral wool), fire resisting doors or automatic shutters of the appropriate specification and automatic fire shutters in the ventilation ducts where they crossed to other areas of the factory.

viii) There was no evidence that there was a breach of the relevant Building Regulations. The premises had a Fire Certificate, although both the relevant parts of the Building Regulations and the Fire Certificate were intended to prevent danger to persons, not damage to property.

6. The judge held that it followed from the evidence of the fire experts that if oil of a depth of 1 or 2 mm in the gas bratt pan had caught fire, the wall panels of the veg prep area would not have ignited but that, if a minimum of 6 mm or possibly 8 mm of oil had caught fire, the wall panels of the veg prep area would have ignited. Those conclusions are not challenged. Nor is the judge's further conclusion, based on the expert evidence, that at the time of the fire the level of oil in the bratt pan was about 100 mm.
7. The judge also held that it followed from the expert evidence that, if the veg prep area had been protected by non-combustible panels, then despite the concurrence of a number of negligent acts and omissions on the part of Sahib which caused the fire, the factory would not have burnt down. Thus, if the panels which were ultimately used in the deep frying room had been used, the fire would not have spread to the factory and burned it down but would have been contained in room G49.
8. It was not in dispute that Sahib was seriously at fault for causing the fire. The judge expressed his conclusions in this regard as follows in the relevant parts of paragraph 22 of his judgment:

"(a) …

(b) Sahib employed as the operator of the gas fired bratt pan Mr. Gurcharan Singh, an asylum seeker who could not speak English. On the day in question, he worked from 6 a.m. until 8 p.m. At the end of the day, he was responsible for turning off the gas supply to the bratt pan. There may be many women who work hours like that in their own homes and kitchens, but Mr. Reynolds [see below] agreed that it was not sensible to employ a man to work those hours and rely on him to perform an important task like turning off the gas at the end of the day. Mr. Singh did not turn off the gas supply or the pilot light. The gas burners were left on in the maximum position. In that failure, Mr. Singh was negligent, though in the circumstances one could only sympathise with him if he were here to accept that sympathy. His employers, Sahib, were not only vicariously liable for his negligence, they were independently negligent for relying on him to perform a responsible (though simple) task after working such long hours. It is true that his supervisor ought to have checked, and Sahib are vicariously liable for his failure also.

(c) Sahib were also negligent in allowing the gas fired bratt pan to be used with a depth of oil far in excess of that recommended by the manufacturers. Mr. Singh told Dr. Sanders of Dr. J.H.Burgoyne and Partners, through an interpreter, that he left cooking oil in the pan to a depth of about 50 to 100 mm. Dr. Sanders and another fire expert, Dr. Bland, agree that burn marks on the inside of the pan show that the pan had contained oil to a depth of 100 mm when the oil ignited. The evidence of Mr. Reynolds is that the night shift should have come on at 10.00 pm to clean the pan and its surroundings. However, before they arrived, the fire had started.

(d) The gas fired bratt pan was designed to be fitted with a thermostat controlled by a revolving switch to control the temperature of oil in the pan. On the day in question, and probably for some time earlier, the thermostat was broken, and if the gas was left on (as it was) there was nothing to stop the temperature of the oil rising until combustion took place. In the opinion of Dr. Sanders, that is what happened. There was no maintenance contract for the gas fired bratt pan. I have no evidence whether the broken thermostat was reported and not dealt with or not reported. In either event, Sahib were negligent.

(e) Later models of this bratt pan were fitted with an override temperature cut-out switch set at a fixed maximum temperature. This was an old model that did not have such a switch and there is no evidence that such a switch was required when the pan was brought back into service. However, the fire experts agree that it would have been preferable for an override thermostat to have been fitted.

(f) Mr. Reynolds said that operators at the bratt pan would use a paddle to turn over the vegetables while cooking and in doing so they would cause oil to go over the sides and onto the underside of the pan. Also, when emptying the pan after cooking, they would use a large scoop like a shovel and would tend to overload the scoop and spill more oil and food onto the floor and onto the gas flame where it would congeal and form a flammable solid that would be difficult to remove. That evidence is inconsistent with his evidence about shallow fat."

9. In our view the key negligence of Sahib which caused the fire is that identified in sub-paragraphs 22(b), (c) and (d) above. As appears below, sub-paragraph (a) is not concerned with the cause of the fire. Sub-paragraph (e) does not seem to us to amount to a finding of negligence and, as to sub-paragraph (f), it was not, as we see it, causative of the fire because the cause of the fire was the ignition of the oil in the pan when it reached its flashpoint and not any of the events identified in sub-paragraph (f). We should perhaps add that there is some reference in the judge's judgment to naked flame. However, again as we see it, such flame is irrelevant to the cause of the fire. No-one suggests either that the gas flame under the pan caused the fire or that the small pilot flame did so.

Liability of PKS

10. It is common ground in this appeal that PKS owed Sahib a duty to act as reasonable and skilful architects in connection with the refurbishment of the factory in 1994. It is also common ground that the refurbishment included refurbishment of the veg prep area in room G49, of the deep frying room G32e and of the main kitchen, where there were also bratt pans. Sahib acted as its own main contractor.
11. The judge held that PKS were in breach of duty in that they failed to advise Sahib to fit fire resistant panels in room G49, that if they had Sahib would have fitted such panels and that in that event the factory would not have burned down. As we understand it, PKS accept both that if Sahib had been given such advice it would have fitted appropriate panels and that, if such panels had been fitted, the fire would not have spread and the factory would not have burned down. The issue in this appeal is whether PKS should have given Sahib that advice.
12. The resolution of that issue depends to a large extent, if not entirely, upon what PKS knew or should have known about the cooking operations likely to be carried out in room G49. It is common ground that there was no significant risk of the spread of fire if the amount of oil used in the gas bratt pan was 2 mm or less, whereas there was such a risk if it was, say, 6 mm or more.
13. Sahib's case is that the operations carried out in the veg prep room in 1994 involved, at any rate from time to time, using the gas bratt pan for frying in over 6 mm of oil, whereas it is and was PKS' case that the pan was only used for frying in 2 mm of oil or less. Alternatively it is PKS' case that, if more oil than that was used, they were not told about it and were not at fault in failing to discover the true position.
14. The judge resolved this central issue in favour of Sahib. He did so in large measure by preferring the evidence of Mr Sahib Ahluwalia to that of Mr Barry Reynolds. Mr Ahluwalia was one of the founders of Sahib and gave one of his names to it. He was not, however, a production man but essentially a sales and marketing man. By the time of the refurbishment in 1994 and 1995 he had been to some extent sidelined by new shareholders but he had been the Managing Director between July and December 1989 and remained with Sahib thereafter and the judge accepted that he knew what processes were carried out by the company, including those carried out in room G49. Mr Reynolds was the Production Director at the time of the refurbishment but had left long before the fire.
15. The judge preferred the evidence of Mr Ahluwalia to that of Mr Reynolds as to the processes which were carried out in room G49 and one of the central issues in this appeal is whether he was entitled to do so. Mr Justin Fenwick QC correctly accepts that the authorities show that in order to persuade this court to take a different view from that of the judge on an issue of this kind he must show that the judge was plainly wrong: see eg Assicurazioni Generali SpA v Arab Insurance Group [2002] EWCA Civ 1642, [2003] 1 WLR 577, and the cases there cited.
16. Before considering that it is appropriate to put the evidence in context. In their capacity as architects PKS were required to and did prepare designs and specifications for the refurbished factory. There were three particular areas drawn to our attention in which cooking was carried out. They were those referred to above, namely the deep frying room G32e, the veg prep room G49 and the kitchen. The factory produced prepared foods for sale to large supermarkets such as Tescos, who have high standards. We will focus particularly on room G49. It was common ground at the trial and in this appeal that PKS had a duty to consider, in relation to each of the rooms in which cooking was carried out, the risk that a fire might break out, the risk that the wall panelling might catch fire and, if it did, the risk of the spread of fire throughout the factory. It was apparent to everyone that the spread of fire throughout the factory could be catastrophic.
17. In these circumstances it was, as we see it, the duty of PKS to ascertain what cooking activities were carried out in room G49 (and elsewhere) and what, if any, risks of fire there were as a result. It was apparent that oil in a bratt pan could catch fire if it reached its flashpoint and could spread to the polystyrene core of the EPS panels which were highly combustible. In these circumstances, we would accept the submission of Mr Stuart-Smith QC on behalf of Sahib that it was incumbent upon PKS, before deciding whether to advise Sahib to use EPS non-fire rated panelling, to satisfy themselves that the cooking operations in the room concerned were such that there was no significant risk of spread of fire.
18. The principal witness called from PKS was Mr Vassilas. He recognised the potential problems. He did so in part at least because PKS consulted LR Insulations Limited ("LRI"), who were specialists in insulation including fire proof insulation. On 25 September 1994 PKS produced a specification for the factory which did not include any fire rated insulation but only EPS panelling. At some stage they asked LRI to produce a specification and costing for the insulation to be used in the project and LRI replied by letter dated 6 October 1994, which was quoted by the judge and included the following:

"We have great concern that polystyrene core panels have been selected for use in areas where cooking or frying will take place. On page 7 of your specification, you refer to polystyrene having a core with "fire resisting additives." There is no such thing as a non-combustible polystyrene. All polystyrene and polyurethanes are combustible regardless of grade. We would strongly recommend that our Flameguard range of non-combustible panels are used in areas where there is a risk of fire. The fryers should be in self-contained units and here the enclosure should be constructed from our Flamelock panels. This system has been installed at a very large food factory in Shropshire. Two recent fires in frying enclosures were successfully contained for over two hours. There is, of course, a cost implication for these panels. May I suggest that one of my fire engineers visits you and puts his suggestions forward? I feel sure that your clients' insurance company would object to the scheme as it is proposed at present.

If I can be of any further assistance please do not hesitate call me."

19. The judge recited the evidence of the author of the letter, Mr Bailey, to the effect that letters in similar terms, tailored to the facts of individual cases, had been sent out since about 1992. Mr Bailey also said that, if there was gas in room G49, it was included in their warning, which expressly referred to places where there was "cooking or frying". The judge added that those words would include a place where there were bratt pans, whether they were referred to as cookers or fryers.
20. The judge also referred to the evidence of Mr Openshaw of LRI, who had retired by the time of the trial but had been the Technical Director at LRI. His evidence included the following. A letter in similar terms to that quoted above had been sent out for about 10 to 15 years before 1994. There had been some serious fires, which was why LRI had developed particular fire resistant panels in about 1986. In his view steam heated areas were fairly safe, although if there was naked flame there would be a danger. The judge particularly relied upon Mr Openshaw's evidence of a specific hygiene visit to the veg prep area with Mr Reynolds on an unspecified date. He asked Mr Reynolds whether the area should be enclosed and Mr Reynolds told him that the area was only used for steam cooking. He said that if he had been told that there were frying facilities, "I would have pushed the issue a lot sterner than I did. We were told it was steam cooking. It did not bother me".
21. In a later letter of 6 November 1994, which was also relied upon by the judge, Mr Bailey of LRI wrote to PKS, under the heading "½ Hour Fire Rated Doors":

"All polystyrene and polyurethanes irrespective of their grades are combustible. It is not possible to install fire rated doors into these panels. See letter attached regarding our concern at the use of polystyrene panels in certain areas."

The "letter attached" was the letter of 6 October quoted above. It is true that later letters did not include that warning but we agree with the judge that the warnings given to PKS were explicit, as we think was accepted by Mr Vassilas and indeed by Mr Fenwick on their behalf.

22. The judge referred in particular to the evidence of both Mr Kyriakides and Mr Vassilas of PKS that their knowledge was not simply knowledge that EPS will burn. The judge described it in this way in paragraph 39 of his judgment:

"It was knowledge that steel faced EPS panels will delaminate in a big fire, that the EPS core will then rapidly degrade, that the melted cores will transmit fire one to another very rapidly, that exactly this had happened in numerous recent fires, including at Sun Valley in 1993 and Noons' (a competitor – close to home) in 1994, and that there was available a reasonably priced fire-resistant alternative which specialist sub-contractors were currently recommending for all cooking areas."

The fire at Noons occurred on 15 November 1994.

23. Mr Vassilas' evidence was thus, not that he did not understand the potential risks, but that, so far as he was aware, the use of oil in the gas bratt pan was limited to some 2 mm or less, which posed no threat. PKS also relied upon the evidence of Mr Reynolds to the effect that oil was only used in that way and that he told or would have told Mr Vassilas that that was the case. Mr Reynolds was called to give evidence on behalf of PKS. However, before he gave evidence, Mr Ahluwalia gave evidence on behalf of Sahib. The sequence in which they gave evidence is or is potentially significant because one of the bases of PKS' appeal is that Mr Moxon Browne QC, who appeared for Sahib at the trial, did not cross-examine Mr Reynolds on the crucial point in the case, namely the depth of oil used in room G49 and that in those circumstances the judge should not have preferred Mr Ahluwalia's evidence to that of Mr Reynolds.
24. We turn to Mr Ahluwalia's evidence. The judge quoted the essence of his evidence as follows:

"2 mm of cooking oil hardly -- could even not be enough to do shallow frying in my view, with my experience. 2 mm is really nothing. I mean, it might be enough to do some flash frying, where you are just throwing things in and bringing them out, but not for shallow frying. For shallow frying, you do need, in my view, more oil in whatever vessel you are using than 2 mm. In my view, for the sort of process that those bratt pans were being used for, they would need at least 10 to 12 mm at least. … You heat up the oil first, and then you throw the ingredients in. Some oil would be absorbed, yes. So you top it up again before you put the next batch of ingredients in."

25. In the course of his evidence Mr Ahluwalia was asked about some recipes which had been obtained from a different factory but which were said to be very similar to those prepared in the veg prep room. The recipes were relied upon by PKS because they provided for quantities of oil to a depth of under 2 mm and, in particular, included a recipe for caramelising onions. In paragraph 19 of his judgment the judge observed that Mr Ahluwalia then went on to explain that the recipes, with their carefully measured ingredients, were not used in the veg prep area and quoted this further passage from Mr Ahluwalia's evidence:

"The process you have just described is actually making the product. The process for which this particular bratt pan was used, and is also used in other places, is a prepping process. In the prepping process, you do not continue to add different ingredients into it, and you do not add any spices or anything, it is just preparing -- for example, in this particular case, it was prepping the onions, they were caramelised.

So all you are doing is caramelising the onions, taking them out. Then those onions go to the main kitchen -- this is in the prep kitchen, that you are prepping the product. Then you go to the main kitchen, where you actually do what you have just described actually then make the product, a curry or whatever, into which these prepped onions or aubergines or potatoes or whatever is going. So you do not do in the prepping bratt pan what you would do in the main recipe cooking bratt pan."

26. The judge essentially accepted that evidence. As we see it, the key aspect of Mr Ahluwalia's evidence which the judge accepted was that the depth of oil used in the gas bratt pan in room G49 was not limited in the way being suggested to him but involved a process in which "they would need 10 to 12 mm at least". If the judge was entitled to accept that evidence, this appeal is very difficult, if not impossible to sustain. Was he entitled to do so?
27. It was suggested in argument that there was an inconsistency between Mr Ahwulalia's witness statement and his oral evidence because he said in his witness statement that 50 mm of oil would be used. However, we do not accept that there was any such inconsistency because in evidence he said that at least 10 to 12 mm would be needed. We see no inconsistency between those two statements. It is a reasonable inference from his conclusions that the judge did not do so either.
28. By contrast with the evidence of Mr Ahluwalia, the judge set out part of the evidence of Mr Reynolds including the following, which was taken from his witness statement:

"The bratt pan is used for cooking products such as rice and sauces and, if for frying at all, shallow frying. It should not be used for deep frying foods. I would describe shallow frying as using less than an overall depth of 2 mm of cooking oil and deep frying is using more than 2 mm. The deep frying of food would take place in the continuous fryers located in the fryer enclosure elsewhere in the factory.

Bratt pans are designed to be attended to by an operator at all times. They are not designed to be left to cook food unattended, unlike continuous fryers. Once a batch of products has been cooked and the working shift completed, any residual cooking oil or other cooking matter should be drained off and the bratt pan switched off by the operator. I recall that at the factory, we relied on the operators to switch off their equipment. The supervisors on site were also supposed to check that the equipment had been switched off before leaving the shift."

29. It can be seen from those extracts that the two witnesses described shallow frying as meaning different things. This is not we think surprising because a number of different definitions of shallow and deep frying were used in the evidence. For example Mr Turner, who is a self-employed general health and safety consultant, described shallow frying as where the product is turned over within the oil and deep frying as where the oil covers the product. However, not everyone so described it. Thus for example, Mr Highton, who was PKS' architectural expert, understood that shallow frying was frying in about 2 mm of oil, whereas deep fat frying was frying in any greater depth of oil. Like the judge, we do not accept the submission that shallow and deep fat frying are terms of art understood in one particular way.
30. In the course of argument we were referred to extensive extracts from the evidence, including in particular the evidence of Mr Ahwulalia, Mr Reynolds and Mr Vassilas. Mr Ahwulalia was cross-examined on the basis that he was not a cook or indeed, unlike Mr Reynolds, the Production Manager, but an accountant or a businessman. He accepted that but said that, as the founder of Sahib's business, who had continued to be closely involved, he had been engaged in this type of business for the last 20 years and noticed what processes were going on. He was adamant that he had seen what he called prepping in room G49, which included frying vegetables such as onions or aubergines in oil significantly deeper than 2 mm. He said that that was a different process from that described by Mr Reynolds. For example, he said that they would fry aubergines in the veg prep room bratt pan and then put them away in the chiller for later use as an ingredient in the main dish.
31. Subject to Mr Fenwick's submissions based on Mr Moxon Browne's failure to cross-examine Mr Reynolds by reference to Mr Ahluwalia's evidence, we see no reason why the judge should not have preferred the evidence of Mr Ahluwalia to that of Mr Reynolds. They were describing different processes. Mr Stuart-Smith suggested that what Mr Reynolds saw was being done in the kitchen and not the prep room, where the frying described by Mr Ahluwalia was being done. We would not go so far as that. It seems to us that one possible conclusion is that from time to time both the frying described by Mr Ahluwalia and the cooking described by Mr Reynolds were carried on in the prep room but, however that may be, we see no basis upon which (subject to the cross-examination point) we could properly interfere with the judge's acceptance of Mr Ahwulalia's evidence that the frying he described was carried out in the veg prep room.
32. There was a suggestion in the course of argument that Mr Ahluwalia may have been describing, not the method of cooking or frying in 1994 and 1995, which was the relevant time, but a later time. There was a certain amount of evidence which related to a period after 1995 but before the fire that oil of depths much greater than 2 mm was being used for frying in room R49. For example, a Mr Chattu, who was employed by Sahib and was Mr Singh's supervisor on the day of the fire, told Dr Sanders that the normal depth of oil was 3 to 4 inches, which is of course 75 to 100 mm and thus consistent with the level of oil in the bratt pan when the fire occurred. There was also evidence from Mr Turner, who said that oil was poured into the bratt pan from 25 litre drums. He described it as a rough and ready process, which is not consistent with the use of only 2 mm of oil.
33. The case for PKS is that, if such practices were carried on, that occurred after Mr Reynolds left and not while he was still at the factory. There was a suggestion during the appeal that Mr Ahluwalia's evidence can be explained on the basis that he was mistaken as to when the process he described was being carried out. It was suggested that the system may have changed after Mr Reynolds left when changes were introduced by a Mr Arnold. Mistake is of course possible but Mr Ahluwalia was cross-examined by reference to the evidence of Mr Reynolds and in our view a fair reading of the transcript leaves no doubt that he must have understood that he was being asked about the time that Mr Reynolds was still at the factory.
34. In short, there was a direct conflict of evidence between Mr Ahluwalia and Mr Reynolds as to what processes were carried out when they were both at the factory. It was a matter for the judge to decide whether the processes carried out in the veg prep room at least included those described by Mr Ahluwalia. The judge was well aware that Mr Reynolds was the Production Manager and that Mr Ahluwalia was not a technical man, but he was in our opinion entitled to hold that Mr Ahluwalia was in a position to observe what actually happened at the time that Mr Reynolds was still there and to prefer the evidence of Mr Ahluwalia.
35. The only point that has given us some concern is the submission made by Mr Fenwick that Mr Reynolds was not cross-examined on the basis that deep (or at least deeper) fat frying was being carried out and that in these circumstances the judge should not have rejected Mr Reynolds evidence that the depth of oil used in the bratt pan was limited to about 2 mm. He relies upon the following statement of principle in paragraph 49 of the judgement of Latham LJ (with which Brooke LJ and Hart J agreed) in Deepak Fertilizers & Petrochemical Ltd v Davy McKee (United Kingdom) London Ltd, [2002] EWCA Civ 1006:

"49. The general rule in adversarial proceedings, as between the parties, is that one party should not be entitled to impugn the evidence of another party's witness if he has not asked appropriate questions enabling the witness to deal with the criticisms that are being made. This general rule is stated in Phipson on Evidence 15^th Edition at paragraph 11-26 in the following terms:

'As a rule a party should put to each of his opponent's witnesses in turn so much of his own case as concerns that particular witness, or in which he had a share, eg if the witness has deposed a conversation, the opposing counsel should put to the witnesses any significant differences from his own case. If he asks no questions he will generally be taken to accept the witness's account and will not be permitted to attack it in his final speech. ... Failure to cross-examine will not, however, always amount to acceptance of the witness's testimony, if for example the witness has had notice to the contrary beforehand, or the story itself is of an incredible or romancing character.'"

36. Mr Fenwick submits that in the circumstances Sahib should not have been permitted to submit to the judge that more than about 2 mm of oil was used in the bratt pan in the veg prep room and that the judge should have accepted Mr Reynolds' evidence to that effect. We are bound to say that, in our judgment, Mr Moxon Browne should have put the point expressly to Mr Reynolds in cross-examination and we have carefully considered whether his failure to do so leads to the conclusions identified by Mr Fenwick. However, we have reached the conclusion that the judge was not bound to hold that only 2 mm of oil were used in the gas bratt pan.
37. The last part of paragraph 49 just quoted shows that failure to cross-examine will not always lead to such a conclusion and one example of such a case given by Latham LJ is where the witness has had notice to the contrary. Latham LJ continued as follows:

"50. The caveat in the last sentence that I have quoted is important, particularly in the context of the Civil Procure Rules in which, by Part 32 r. 1(3) the court is given a power to limit cross-examination. Nonetheless, the general rule remains a valid rule of good practice and fairness. The judge of fact is, however, in a different position from the protagonists. So long as a matter remains clearly in issue, it is the judge's task to determine the facts on which the issue is to be decided. However it seems to me that where, as in the present case, an issue has been identified, but then counsel asks no questions, the judge should be slow to conclude that it remains an issue which has to be determined on the basis of an assessment of reliability or credibility without enquiry of the parties as to their position. The judge should be particularly cautious of doing so if he or she has not given any indication of concern about the evidence so as to alert the witness or counsel acting on the side calling the witness, to the fact that it may be that further explanation should be given in relation to the issue in question.

51. At the end of the day each case will depend upon the way in which the issue arose, and was dealt with in evidence. …"

38. In the instant case we do not think that there can have been any doubt in the minds of counsel or solicitors who represented PKS at the trial, any more than in the mind of the judge, that the issue what depth of oil was used for cooking in the gas bratt pan in the veg prep room remained in issue at the trial. It was indeed one of the central issues between the parties. Mr Ahluwalia's evidence was put to other witnesses, notably the fire experts, Mr Turner and Mr Highton. Moreover, it seems to us to be most unlikely that Mr Reynolds was in any doubt that that fact remained in issue. If Mr Wilmot-Smith QC, who represented PKS at the trial, had wished to ask Mr Reynolds any further questions he could of course have done so, although we do not suggest that he should have done so. It seems to us that, in spite of Mr Moxon Browne's failure to cross-examine Mr Reynolds on the point, the judge really had no alternative but to address the issue. As Latham LJ put it, "so long as the matter remains clearly in issue, it is the judge's task to determine the facts."
39. In our judgment, this is such a case and there was no injustice to PKS in the judge addressing the issue. Having done so, as already indicated, there was in our opinion ample material upon which he could reach the conclusion he did. Although Mr Fenwick points to a number of aspects of the case which he says undermine the judge's conclusion in this regard, we do not think that, either individually or collectively, they have any such effect. We mention some of them briefly.
40. It was suggested that the judge misunderstood the term 'deep fat frying'. However, we have already expressed our conclusion that it is not a term of art, so that there is no substance in this criticism. The judge did mistakenly say that he had been told that the pan was being used for caramelising onions at the time of the fire, whereas it was in fact being used for frying aubergines, but we cannot see how that error is relevant to any of the issues in the appeal.
41. There was some debate as to the evidence as to dosing points in room G49. PKS' case was that there were dosing points in the room and complaint is made that the judge held that, although the plans showed dosing points, they did not show an indication of such points near the gas fired bratt. There is some force in that submission, although the evidence is far from crystal clear. However, the answer to the question whether or not there were dosing points, which were designed to be used to supply oil in correct quantities for whatever frying or cooking operation was required, does not answer either of the further questions whether any such dosing points were used or what depth of oil was used for cooking in the veg prep room.
42. It was further submitted that the judge reached a wrong conclusion as to the decision to replace the proposed EPS panels in the deep frying room G32e with Rockwool fire resistant panels. In paragraph 51 of his judgment the judge said this:

"Because of their knowledge of the risks, Sahib, not the defendants, took the decision to enclose the continuous deep fat fryer in Rockwool panels, though Mr. Vassilas denied that it was the decision of Sahib."

Mr Fenwick submits that that conclusion is wrong and that the decision was that of PKS. However, let it be assumed that that is correct; we cannot see what relevance that has either to the question what depth of oil was used in the veg prep room or to the question whether PKS were negligent with regard to the use of EPS panels in room G49.

43. We turn to the question whether PKS were negligent. Given the conclusion that oil was used for cooking or frying in room G49 at greater depths than 2 mm, the question arises whether PKS should have become aware of that fact and done something about it. There is no issue but that, if they had known the true position, they would have recommended fire resistant panels and that Sahib would have fitted such panels. The question is therefore whether they should have ascertained the true position. PKS' case was that Mr Vassilas and Mr Reynolds carried out a risk assessment and that Mr Vassilas did not learn the true position.
44. The judge held that the evidence of the risk assessment was vague and that no proper risk assessment could have been carried out. In our judgment the judge was fully entitled to reach that conclusion because, as he put it in paragraph 57, neither Mr Vassilas not Mr Reynolds had any clear or reliable memory of what was done or considered. If any proper risk assessment had been carried out, Mr Vassilas would have discovered what cooking processes were being carried out in room G49.
45. The judge's conclusions in this regard can be seen from paragraph 58 of his judgment as follows:

"In his written statement, Mr. Reynolds said that he and Mr. Vassilas conducted a risk assessment involving at various stages Mr. Becket and Mr. Openshaw and other contractors. He said, "I cannot now recall the details of the risk assessment but I know it involved going through the room loading sheets for the factory to assess what items of equipment would be contained in each room; the staffing levels in each room; and the services such as water and gas or electricity which would need to be supplied to each room." I have seen some of the room loading sheets to which he refers. In particular, there is a room loading sheet for Room G49, described as a Vegetable Cooking Area. Mr. Reynolds had fobbed off the concern of Mr. Openshaw by telling him that the area was only used for steam cooking, but if he said the same thing to Mr. Vassilas (and there is no evidence that he did) he ought not to have been believed. The Room Loading sheet clearly stated that power supply was required for 2 bratt pans and a gas bratt pan. There were two steam powered bratt pans. A gas bratt pan could not be thought to be a steam cooker. The description of the equipment was, "Kettles, fryer, bratt pans". It was established in evidence that the "fryer" must have been the gas bratt pan. As Miss Tooth indicated, one could not expect an architect to know all the details given by the fire experts about how much oil on fire would be required to set light to the EPS panels. But any reasonable architect, like any reasonable householder, can be expected to know that frying, even in a household frying pan, may produce a fire: hence the provision in Room G49 of a fire blanket. A frying pan on fire can be very frightening for a household cook, and as Miss Tooth said, a fire in a much larger bratt pan would be all the more frightening and one cannot rely on the operator (probably ill trained in fire precautions) to have the courage and presence of mind to throw a fire blanket over a bratt pan fire. A fire starting in Room G49 was quite on the cards. The architect knew that if the fire spread to the EPS panels it was likely to spread rapidly through the factory. He would not have known what circumstances were required to cause the fire to ignite the EPS panels, but in view of the gravity of the potential consequences, he should have found out and eliminated the possibility of those circumstances before deciding against the modest cost of using flameproof panels."

46. Mr Fenwick criticised that passage in some respects. For example he submitted that the judge should not have held that Mr Reynolds fobbed off the concern of Mr Openshaw or that if he said the same thing to Mr Vassilas he ought not to have been believed. If the judge meant that Mr Reynolds deliberately put Mr Vassilas off the scent we would agree that he should not have so held, but we do think that that is what he meant. It seems to us that the judge simply meant that Mr Reynolds ought to have given Mr Vassilas more complete details of what processes were carried out in the veg prep room. In our opinion the judge was justified in concluding (as we think he did) that no sufficient consideration was given to the problem by either Mr Vassilas or Mr Reynolds. He was entitled to have regard to the fact that the description of the equipment in room G49 as being "Kettles, fryer, bratt pans", which indicated that frying would take place in the room.
47. The judge was also entitled to have regard to the opinion of Miss Tooth , who gave expert evidence as an architect to this effect:

"… the important point, in my view, in that respect, was not to accept that that pan would never have more than 2 mm of oil in it. I mean, as a matter of common sense, any architect would suspect that at times, it would have more oil than that in it. … But 2 mm is hardly anything. I would expect an architect to strongly suspect that at times, or even all the time, it would have more oil in it than that, considering how big it was."

48. In our judgment the judge was entitled to conclude that PKS did not investigate the problem with sufficient care. By the time of the trial it is clear that Mr Reynolds did not recall very clearly what the veg prep room was used for: see for example the passage in his witness statement quoted above in which he said "if for frying at all" it was used for shallow frying. Whatever his recollection by the time of the trial, it seems to us to be more likely than not that he would have known what in fact went on at the time. On that footing, if Mr Vassilas had discussed the matter with Mr Reynolds in any detail, as it was his duty to do, we can see no reason why he should not have been given full details of what processes were carried out. Mr Reynolds had no reason whatever to mislead Mr Vassilas. However, if (contrary to the view just expressed) Mr Reynolds was not aware of what processes were carried on in the veg prep room, it seems to us that any detailed discussion of the potential problems would have led Mr Vassilas to realise that that was the case, in which case he should (and no doubt would) have made further enquiries. In short, in the light of the evidence of Mr Ahluwalia, which the judge was entitled to accept, any reasonable examination of the processes in fact being carried out would have led to Mr Vassilas discovering the true position.
49. It is clear on the findings of the judge that no proper investigation was carried out, despite the warning in LRI's letter of 6 October 1994. It was submitted that Sahib was sufficiently warned of the potential problems and, in particular that the judge should have held that Mr Vassilas gave a copy of LRI's letter of 6 October 1994 to Sahib. The judge did not so hold and it is we think fair to say that he expressed some doubt as to whether anyone at Sahib saw the letter. It is true that Mr Vassilas said in his statement that he believed that he had given a copy to Mr Reynolds, although Mr Reynolds had no recollection of it and no reference was made to it in any of the contemporary correspondence which survived. In these circumstances we accept Mr Stuart-Smith's submission that the judge did not have to accept that Mr Vassilas gave a copy of it to Mr Reynolds.
50. However, whether that is correct or not, the judge was in our opinion entitled to hold that the letter should have been passed to the highest authority in Sahib with a specific recommendation that fire resistant panels should be fitted in room G49. Whether the letter was passed on or not, the judge's findings justify his conclusions that PKS should have taken all reasonable steps to ascertain the nature of the cooking processes carried out in room G49 and that if they had done so they would have ascertained the true position and advised Sahib to fit fire resistant panels. In our judgment, the appellants have failed to show that the judge was wrong, let alone plainly wrong to hold that PKS were in breach of their duty of care owed to Sahib.
51. Given the evidence that, if PKS had given that advice, Sahib would have accepted it and fitted fire resistant panels, it follows that the judge was correct to hold that that breach caused the spread of fire and that PKS were liable to Sahib. We therefore dismiss the appeal on liability.

Contributory Negligence

52. Section 1(1) of the Law Reform (Contributory Negligence) Act 1945 ("the 1945 Act") provides, so far as relevant, as follows:

"Where any person suffers damage as the result partly of his own fault and partly of the fault of any other person or persons, a claim in respect of that damage shall not be defeated by reason of the fault of the person suffering the damage, but the damages recoverable in respect thereof shall be reduced to such extent as the court thinks just and equitable having regard to the claimant's share in the responsibility for the damage: … "

53. PKS' case is that the spread of the fire was caused by Sahib's fault in two independent respects. The first is that Sahib was solely at fault for the fire breaking out and the judge so held. As stated in paragraphs 8 and 9 above, it was not in dispute that Sahib was seriously at fault for causing the fire, especially in the respects identified in sub-paragraphs 22(b), (c) and (d) of the judge's judgment which we quoted above. Those conclusions are not challenged by Sahib in this appeal. PKS' case is that those faults contributed to the damage and loss which was also caused by the fault of both parties in contributing to the spread of the fire.
54. The second respect in which it is said that Sahib was at fault was thus with regard to the spread of the fire outside the veg prep room. In this regard the judge made this important finding in paragraph 22(a) of is judgment:

"22(a) Mr. Openshaw, then a director of LRI, visited the factory before the refurbishment and he advised Mr. Reynolds specifically in relation to the veg prep area that non-combustible panels should be used wherever cooking was taking place. Mr. Reynolds replied that the veg prep area was used only for steam cooking and that satisfied Mr. Openshaw that as result non-combustible panels would not be required in that area. Mr. Reynolds was wrong and negligent in giving that statement. One of the bratt pans was fired by a naked gas flame and if Mr. Openshaw had known that, he would have pressed his advice that non-combustible panels should be used: if that advice had been heeded, the spread of the fire would have been limited."

55. Sahib does not (as we understand it) submit in this appeal that the judge was not entitled to reach those conclusions. In any event, it appears to us that the judge was entitled to hold that Mr Reynolds told Mr Openshaw that the veg prep room was used only for steam cooking when it was not, that he was at fault for failing to tell him that one of the pans was gas fired and that, if he had told him the true position, Mr Openshaw would have pressed his advice that fire resistant panels be used with the result that the fire would not have spread out of the room.
56. Notwithstanding those conclusions of fact, the judge refused to hold that the damages recoverable by Sahib should be reduced under section 1 of the 1945 Act. He did so in spite of the fact that at the conclusion of the trial it was submitted on behalf of PKS that the contributory negligence of Sahib was so serious that the damages should be reduced by 90 per cent and it was accepted on behalf of Sahib that its damages should be reduced by 50 per cent. After reaching the provisional conclusion that he should not accept either submission and that the damages should not be reduced at all, the judge gave the parties the opportunity of addressing further submissions to him, which they did orally. However he adhered to his provisional view.
57. Somewhat different considerations apply to the two respects in which Sahib is said to have been at fault. Before considering them in turn, we should note in passing that it is common ground that no distinction is to be drawn for the purposes of considering contributory negligence between PKS' liability in contract and in tort. This is an example of the third category of case identified by Hobhouse J in Forsikringsaktieselskapet Vesta v Butcher [1986] 2 All ER 488 at page 508, namely a case where the defendant's liability in contract is the same as his liability in the tort of negligence independently of the existence of any contract.
58. It is convenient to consider first the question whether Sahib was partly at fault with regard to the failure to fit fire resistant panels, which would have avoided the spread of the fire and the loss of the factory. We will then consider the relevance of Sahib's fault in causing the fire.

Spread of the Fire

59. It was not pleaded by PKS that Sahib was partly at fault for the failure to replace the proposed EPS panels with fire resistant panels. However, the point was taken before the end of the trial and the judge considered the position on the facts in some detail, notably in paragraph 22(a) of his judgment quoted above. In spite of those findings he held that Sahib was not partly at fault in this respect. He said in paragraph 71 that Sahib "was in no way responsible form the spread of the fire, as opposed to its initiation".
60. His reasoning can be seen in paragraph 64. The approach to which he refers in the first sentence is the approach which he ultimately adopted, as explained below. He said:

"64. A complicating factor in that approach would be to consider whether Sahib had in any way contributed to the failure of the defendants to make a design that would have contained the fire within Room G49. Contributory negligence of that nature is not pleaded. However, in written closing submissions, counsel for the defendants submit (in paragraph 99) that "Sahib knew that if there was a fire in the veg prep cook area then it could spread to the panels and throughout the building." As to that latter submission, there is no pleading to support it, but there is support from the evidence of Mr. Openshaw to which I have referred in paragraph 22 of this judgment. There was also my finding in paragraph 51 of this judgment that both Mr. Ahluwahlia and Mr. Reynolds accepted in their oral evidence that that they knew that polystyrene panels were combustible and that if polystyrene panels caught fire there could be rapid spread of fire. Against that knowledge, it has to be remembered that it was not their responsibility to design the building. The fact of their knowledge did not make them negligent in any respect. They had no duty either to their employers and still less to the defendants to use that knowledge in any way."

61. Although the judge referred to the absence of a pleading, we do not think that he rejected this part of PKS' case on that ground. He expressly referred both to paragraph 22 of his judgment and to his conclusion that both Mr Ahluwalia and Mr Reynolds knew that polystyrene panels were combustible. The ground on which he rejected PKS' case that Sahib was partly at fault because Mr Reynolds' misled Mr Openshaw or failed to give full information to PKS about the cooking processes in the veg prep room was that it was not the responsibility of Mr Reynolds or Mr Ahluwalia to design the building and that they were not negligent because they had no duty either to Sahib or to PKS to use their knowledge.
62. We have reached the clear conclusion that the judge was wrong to reach those conclusions. It is common ground between the parties that, in order to show contributory negligence, it is not necessary to show that the claimant is in breach of any duty owed to the defendant. The question is simply whether the claimant has suffered damage partly as a result of its own fault. It was thus irrelevant whether Sahib owed PKS a legal duty to use its knowledge in any way.
63. In our judgment, the conclusions of the judge in paragraph 22(a) of his judgment lead inevitably to the conclusion that Sahib was partly at fault for the spread of the fire. Mr Reynolds was asked a question in his capacity as Production Manager. He should have taken care before answering Mr Openshaw's question. In our opinion he owed a duty to his employer to take that care and, in any event, negligently gave inaccurate information in the course of his employment. But for Sahib's fault, through Mr Reynolds in giving wrong information to Mr Openshaw, on the judge's findings, EPS panels would not have been used and the fire would not have spread. In these circumstances Sahib was, in our judgment, guilty of contributory negligence in this respect and the judge should have so held. It follows that we allow PKS' appeal against the judge's conclusion that Sahib was not guilty of contributory negligence.

The Fire

64. The above conclusion is independent of the question whether the judge's approach to the cause of the fire was correct. We turn to that question. The judge approached the problem as one of causation. He concluded that PKS should, by their design, have guarded against the consequences of negligence on the part of Sahib in starting a fire and said in paragraph 62 of his judgment that, if PKS had done their duty, room G49 would largely have been destroyed but the rest of the factory would have survived. His conclusions under this head can be seen from paragraph 63 as follows:

"63. One approach to the apportionment of loss is not on the basis of percentage contribution but on the basis of causation. On that basis one might say: (1) Sahib is wholly responsible for the destruction of Room G49 and any consequential loss that would have flowed if that had been the only fire damage: (2) the defendants are wholly responsible for any physical and consequential loss to Sahib resulting from the failure to contain the fire within Room G49. On that approach, the assessment of damages, which is still to take place, could be made on the basis that the judge first of all assesses the total loss to Sahib (both physical and consequential in terms of trading loss), and then deducts from that figure the total of what would have been lost (both physical and consequential) if the fire had been contained in Room G49 as it ought to have been."

We can well understand the logic of that approach.

65. The judge found support for it in the following passage from the judgment of Sedley LJ in Pride Valley Foods v Hall & Partners [2001] EWCA Civ 1001, (2001) 76 ConLR 1 at page 59:

"69. Contribution starts from a point at which two or more defendants have been held to have contributed by their own fault to the claimant's injury. The remaining task is then to measure their contributions by gauging the relative causative potency of their respective faults and their comparative blameworthiness. Contributory negligence, by contrast, starts from a point at which the defendant alone has been held to have caused the claimant's injury by his fault. Not one but three questions then arise. The first is whether the claimant too was materially at fault. The second, if he was, is whether his fault lay within the very risk which it was the defendant's duty to guard him against. It is only if his fault was not, or not wholly, within the causative reach of the defendant's own neglect that the question of relative culpability enters into the picture."

66. On the facts of that case, which were in some respects similar to this, the trial judge had held that the claimants were 50 per cent at fault and, indeed, Dyson LJ (with whom both Brooke LJ and Sedley LJ agreed) said (albeit without argument to the contrary) that the claimants correctly accepted that they were contributorily negligent. This court refused to disturb the apportionment. We entirely agree with Sedley LJ that, in considering an allegation of contributory negligence in a case like this, where the defendant's duty is to guard against the effects of a fire which might be caused by the fault of the claimant, that fact is relevant to the questions which the court must decide. We would not, however, put the questions quite as starkly as he has done in the passage quoted above. It seems to us that the scope of the defendant's duty and the extent to which that duty is to guard against the claimant's negligence are relevant both to the question whether the claimant's fault was causative of the damage and, if it was, what the relative blameworthiness and causative potency of the parties' respective faults were.
67. It also seems to us that, when the passage from Sedley LJ's judgment is read in the light of what he said in paragraph 71, he too looked at the problem in that way. He said:

"71. … On one view the risk of a conflagration caused by overheating of ovens was a single-skinned flue was exactly the risk to which the defendants' advice should have been directed. If it were so, the defendants might not have got past the second stage of the contributory negligence inquiry. Evidently the judge did not consider that it was so, and Mr Davidson has not sought to challenge this conclusion. But this was also a case in which, at the third stage of the inquiry, it remained arguable that the claimants' neglect had its catastrophic consequence only because of the defendants' breach of duty to guard the claimants against precisely such consequences. Left to myself, I would have apportioned a third rather than a half of the liability to the claimants."

68. That passage seems to us to show that it would only be a case in which, at what Sedley LJ identifies as the second stage of the inquiry, it could be seen that the whole of the responsibility for the damage was the defendants' failure to protect the claimants' against their own negligence, that it would be appropriate to hold that the claimant was not guilty of contributory negligence. As was stressed by Lord Hobhouse and Lord Millett in Platform Home Loans v Oyston [2002] 2 AC 190 at pages 211 and 214 respectively, a reduction for contributory negligence should only be applied to damage for which the claimants are partly responsible. See also Rahman v Arearose [2001] QB 351, where Laws LJ said at page 367 that "the real question is what is the damage in question should be held responsible" (Laws LJ's italics). We accept the submission that that approach is also relevant to contributory negligence.
69. The correct question in each case is identified by the words of section 1(1) of the 1945 Act and is thus whether the case is one in which the claimant suffered damage "as the result partly of his own fault and partly of the fault" of the defendant. If the answer to that question is yes, the damages must be reduced "to such extent as the court thinks just and equitable having regard to the claimant's share in the responsibility for the damage". It is well understood that the concept of fault includes considerations of both blameworthiness and causative potency. It is not in dispute that it is open to the court to conclude that the share of a claimant's responsibility is so small by reference to that of the defendant that it would not be just and equitable to reduce the damages at all.
70. The above conclusions are in part informed and seem to us to be supported by the approach of the majority of the High Court of Australia in Astley v Austrust Limited [1999] HCA 6, (1999) 161 ALR 155. That approach can be seen from paragraphs 29 and 30 of the judgment of the majority (Gleeson CJ, Murray, Gummow and Hayne JJ), which was delivered by Gleeson CJ:

"29. … There is no rule that apportionment legislation does not operate in respect of the contributory negligence of a plaintiff where the defendant, in breach of its duty, has failed to protect the plaintiff from damage in respect of the very event which gave rise to the defendant's employment. A plaintiff may be guilty of contributory negligence, therefore, even if the "very purpose" of the duty owed by the defendant is to protect the plaintiff's property. Thus a plaintiff who carelessly leaves valuables lying about may be guilty of contributory negligence, calling for apportionment of loss, even if the defendant was employed to protect the plaintiff's valuables.

30. A finding of contributory negligence turns on a factual investigation of whether the plaintiff contributed to his or her own loss by failing to take reasonable care of his or her person or property. What is reasonable care depends on the circumstances of the case. In many cases, it may be reasonable for the plaintiff to rely on the defendant to perform its duty. But there is no absolute rule. The duties and responsibilities of the defendant are a variable factor in determining whether contributory negligence exists and, if so, to what degree. In some cases, the nature of the duty may exculpate the plaintiff from a claim of contributory negligence; in other cases the nature of the duty may reduce the plaintiff's share of responsibility for the damage suffered; and in yet other casers the nature of the duty may not prevent a finding that the plaintiff failed to take reasonable care for the safety of his or her person or property. Contributory negligence focuses on the conduct of the plaintiff. The duty of the defendant, although relevant, is only one of the many factors that must be weighed in determining whether the plaintiff has so conducted itself that it failed to take reasonable care for the safety of its person or property."

We respectfully agree with those propositions. We do not detect any difference between them and the dicta of Sedley LJ quoted above. However, if (contrary to that view) there is a difference, we prefer the views of the High Court of Australia, which seem to us to reflect the words of the statute with which we are concerned.

71. We have reached the clear conclusion that, if that approach is adopted, the damage sustained by Sahib as a result of the spread of the fire throughout the factory was the result partly of its own fault in causing the fire. It is true that it was PKS' responsibility to take reasonable care to protect the premises from the spread of fire and, indeed, that it was reasonably foreseeable that a fire might begin as a result of negligence on the part of Sahib or its employees. On the other hand it was also Sahib's responsibility to take reasonable care to avoid a fire breaking out. It could certainly not be said that, in deciding what precautions to take to avoid a fire, Sahib either could or did rely upon the work done by PKS.
72. There is no doubt that the spread of the fire beyond room G49 was caused in part by the fact that the fire began. One approach would be to say that that was no more than what used to be called a causa sine qua non and not a causa causans. However, we do not accept that that fairly reflects the true position. It seems to us that there were two effective causes of the spread of the fire, the fault of Sahib in causing the fire and the fault of both PKS and Sahib (see above) in failing to install fire resistant panels which would have prevented its spread.
73. Equally, if the matter is approached by asking more generally whether the responsibility of PKS to take reasonable steps to design against the spread of fire was such that the spread should be treated as wholly its fault, or indeed whether that fact leads to the conclusion that it would not be just and equitable to make any reduction in the damages to reflect the fact that Sahib was responsible for the fire itself, the result is the same. It seems to us that, if regard is had to all the circumstances of the case, the following conclusions are justified:

i) one of the effective causes of the spread of the fire was Sahib's fault in causing the fire;

ii) while such a fire was foreseeable, PKS could reasonably have expected Sahib to take steps to avoid the kind of serious negligence which caused it;

iii) Sahib's fault in this respect was particularly blameworthy;

iv) both parties were thus responsible for the damage caused by the spread of the fire, which was partly the result of Sahib's own fault in starting the fire and partly due to the fault of both PKS and Sahib which caused the failure to fit fire resistant panels as discussed above; and

v) it would be just and equitable to reduce the damages under section 1(1) of the 1945 Act to reflect those conclusions.

74. The judge did not approach the problem in this way. His approach is set out in paragraph 63 quoted above. For the reasons which we have given, in our opinion he took too narrow a view of the problem.

Conclusions

75. For the above reasons we have concluded that the judge was correct to hold that PKS was liable for the damage caused by the spread of fire to the factory as a whole but that he should also have held that Sahib was guilty of contributory negligence in both the respects discussed above. It was solely responsible for the fact that the fire broke out, which was a cause of the damage to the factory as a whole and it was partly responsible for the failure to fit fire resistant panels. In these circumstances it is for us to decide to what extent it would be just and equitable to reduce the damages which would otherwise be recoverable from PKS.
76. It is never easy to resolve issues of apportionment of this kind. It is necessary to have regard to all the circumstances of the case, which include the following. The fire was solely caused by serious negligence on the part of Sahib, although we must have regard to the fact that one of the purposes of employing a firm like PKS was to ensure so far as possible that, if there was a fire, including a fire caused by the fault of an employee of Sahib, it would not spread out of the veg prep room. It seems to us that that fact must be set against the serious nature of Sahib's negligence. As to the failure to fit fire resistant panels, the principal responsibility in that regard was that of PKS, whose responsibility it was to ascertain the cooking processes which would be carried out in the room and guard against the risk of fire which they caused. On the other hand, Sahib contributed to the failure to fit non-EPS panels by its fault in not giving accurate information as to those processes.
77. In all these circumstances, although Sahib was in our judgment right to concede at the trial that it was guilty of contributory negligence, we do not think that its concession that 50 per cent would be appropriate reflects the relative measure of responsibility of the parties. We have reached the conclusion that, taking account of all the circumstances of the case, the damages recoverable by Sahib from PKS should be reduced by two-thirds and that it should be entitled to recover one-third of its damages attributable to the spread of the fire. We therefore allow the appeal to that extent.