Yaqoob & Anor v Royal Insurance (UK) Ltd

LORD JUSTICE CHADWICK: This is an appeal from an order made on 30 September 2005 by HHJ Mitchell sitting in the Telford County Court in proceedings brought by Mohammed Yaqoob against insurers, Royal Insurance (UK) Limited. After the proceedings were commenced the claimant, Mohammed Yaqoob, was adjudged bankrupt. The claim then vested in his trustee in bankruptcy, who has assigned it to the second claimant, Shazad Yaqoob, the eldest son of Mohammed Yaqoob. Strictly, it is Shazad Yaqoob rather than Mohammed Yaqoob who is now the effective claimant in the action, but nothing turns on that. When I refer to Mr Yaqoob, I refer to the father, Mr Mohammed Yaqoob.

The following facts are not in dispute. The insured property is a detached two-storey building. It was acquired by Mr Yaqoob in June 1994 with the aid of loan finance. Works of renovation and refurbishment were carried out by a builder, Mr Dave Clarkson, between June and September 1994 so as to convert the premises for use as a fast food take-away on the ground floor and a restaurant on the first floor. The premises were open for trade in September 1994; and they were open and trading on the evening of Tuesday 24 September 1996. Mr Yaqoob served his last customer – a taxi driver who was known to him as a regular – with a take-away shortly after 1.00am. By that time, on his evidence, there were no other members of staff on the premises. Mr Yaqoob’s evidence was that he locked up and went home at about 1.10am. There was no other evidence to corroborate that.

At some time after 4.00am a witness, Mr Hoye, who was leaving the Grand Hotel (a building immediately across the road from the insured premises), heard the burglar alarm at those premises sound for about 30 seconds. It then stopped. Some five minutes later two white youths were seen by Mr Hoye running from the side of the premises. He also saw smoke emerging from the roof of the premises. He took steps to call the emergency services. Telephone calls to the fire service, from a passing taxi driver and from the hotel, were logged at 4.38 and 4.42am respectively.

The first fire tender arrived at the premises at 4.41am. By then the premises were well alight, with flames emerging from the fire escape door at first floor level. It was later found that that door had been propped open by means of a fire extinguisher. All other external doors and windows, with one exception, were locked and secure. The exception was the fire door on the ground floor immediately beneath the fire escape, which was also open. An internal, self-closing fire door at the bottom of the staircase which led from the ground to the first floor had been tied open with a plastic bag.

The fire service found that there were two separate seats of fire. One was at the bottom of the staircase and the other was in the restaurant on the first floor of the premises. There was a heavy aroma of accelerants which, on a sample being taken for analysis, proved to be a petrol and diesel mixture. That mixture had been brought to the premises in two 25 litre plastic containers, the remains of which were found on the premises: one, at least, on the first floor. The release bolts on the ground floor fire door, to which I have referred, had been damaged in such a way as to make it appear that the door had been forced to obtain entry; but on close examination it was found that the damage to that door was such that it could only have been inflicted with the door open. That is to say, access was not gained by forcing that door: the door had been opened before the damage was done. That suggested that the damage had been contrived, in order to make it appear that there had been a forced entry to the premises.

The intruder alarm control switch was located in a locked cupboard in the customer area in front of the take-away counter or servery. The intruder alarm was fitted with sensors to the exterior doors and windows. The control switch itself, in the cupboard, was key operated. Mr Yaqoob had a set of keys for the cupboard and the control switch. A duplicate set of keys was kept on a hook underneath the take-away servery and out of sight of customers. The fire officers found that the door of the cupboard appeared to have been forced but the control switch itself had not been violated. The control was in the off position, suggesting that it had been turned from ‘on’ to ‘off’ by a key. The keys which should have been under the counter were missing.

Mr Yaqoob was at his home when the police called there at about 5.30am. He told the police that, before leaving the premises at 1.10am, he had switched off the gas and all electrical appliances. He had set the intruder alarm to ‘on’ using his own key, and had seen that the control panel was duly registering and that all doors were secure. He had finally locked the outer door using his own key on his way out and had gone home to bed. There was no corroboration of that evidence. In particular, there was no corroboration of the account that he had spent the night at home; notwithstanding that his wife and five children were in the house that night.

The insurers instructed an expert, Mr Bailey of Dr J H Burgoyne & Partners Limited, to investigate the cause of the fire. His conclusions appear at section 10 of his report dated 9 September 2005. He wrote:

“1.

The fire which occurred at Kashmir Balti Restaurant during the early hours of 25 September 1996 was started deliberately in at least two separate locations using a flammable liquid as a fire accelerant.

It is clear, therefore, that the fire was not started “accidentally”, in the sense in which that word is colloquially used. It was started deliberately. It was started by a person or persons who were able to gain access to the premises by means of a key, and who had taken some steps to make it appear that access had been gained not by a key but by a break-in.

The policy of insurance provided indemnity against “damage”. Damage in that context was defined as “accidental loss destruction or damage”. The judge held, as a matter of construction of the contract and on the law, that the effect of a policy in that form was that it was for the insured to prove that the fire was accidental “in the sense that it was not occasioned by Mr Yaqoob or with his connivance”. There is no challenge to that finding, whether by respondents’ notice or otherwise.

The judge went on to say this at paragraph 18 of his judgment:

“The standard of proof which the claimants must discharge is of course to show that it is more likely than not that Mr Yaqoob did not so cause the fire. This inevitably involves their having to prove a negative. However, it must be borne in mind that if Mr Yaqoob did indeed set fire to his property in the circumstances alleged, he would have been guilty of serious criminal offences. That has to be taken into account when weighing the balance of probabilities.”

There is no challenge to that direction. But it is important to have in mind that the judge approached the matter thereafter on the basis that it was not for the insurers to prove that the fire was caused by Mr Yaqoob or at his direction; although if they were to prove that they would be entitled to void the policy under condition 5, a condition enabling the insurers to avoid liability on the grounds of fraud. If after hearing the evidence the judge had been left in the position that he could not be satisfied, on the balance of probabilities, that the fire was not started by Mr Yaqoob or with his connivance, then the claim would fail. It was not necessary that he should be satisfied that the insurers had made out their pleaded and positive case that Mr Yaqoob had started the fire.

It is agreed on this appeal that the following facts were established by the evidence at the trial:

The appellant points also to the unchallenged evidence of the experienced fire officer, Assistant Divisional Officer David Bott, who attended the premises on 25 September 1996 shortly after 6.00am to carry out a fire investigation in accordance with his responsibilities. Mr Bott had said this at paragraphs 6 and 7 of his witness statement made on 11 April 2005:

“6.

In the last eight to nine years, I have probably attended 25 fires each year in which the fire has been deemed to have been caused by arsonists. …

7.

In those other incidents, I can say that the efforts made to ensure the spread of the fire did not extend to the lengths which were evident in this case. Whoever caused this fire seems to have been in the premises for quite a while and to have known the premises well.“

As I say, that evidence was not challenged.

For my part I would add to the four agreed facts the following facts, the first of which seems to me to follow as a matter of inescapable inference, and the second of which is supported by Mr Bott’s unchallenged evidence to which I have just referred.

The facts point also, as it seems to me, to the conclusion that the flammable accelerant was brought onto the premises and was distributed by being spread over the staircase and restaurant before the alarm sounded and was de-activated. Given the timing, as found by the judge in paragraph 8 of his judgment, between the alarm first sounding and the smoke being seen to emerge from the premises – a period of some five minutes – that seems to me a necessary conclusion. 25 litre containers of petrol/diesel mixture are heavy and unwieldy. I find it impossible to accept that the perpetrator could have brought those containers onto the premises having activated the alarm and then de-activated the alarm and distributed the contents within the period of five minutes. The facts point, as it seems to me, to the conclusion which I will number conclusion (7); that the alarm was activated, whether by accident or by design, and then de-activated sometime after the perpetrator had gained access to the premises and brought the containers onto the premises and had distributed the contents. Or, to put it another way, the alarm was not activated by the original access onto the premises with the containers.

Given those seven facts, who – other than Mr Yaqoob himself, or someone acting with his connivance – could meet the profile of the perpetrator which they disclose?

The facts point also to the conclusion that the fire was premeditated. It was not opportunistic. The diesel/petrol mixture had to be obtained. The two containers had to be filled and brought to the premises, no doubt by a car or van. The door had to be damaged so as to make it appear forced. Those were matters which would have taken time, planning and equipment. It is right, therefore, to look for evidence that a person who might otherwise fit the insider profile to which I have referred had planned or foreseen that the premises would be destroyed by fire on the night of 24/25 September 1996.

In that context, the evidence of Assistant Divisional Officer Bott and Police Constable Rollison is of importance. It was their evidence that there was little or no stock on the premises when they carried out their investigation on the morning of 25 September 1996. Mr Bott said this at paragraph 3 of the witness statement to which I have already referred:

“I can still remember the restaurant premises very well. I remember walking through the premises. … I saw very little food evident on the premises. I would have expected to see far more stock in a working restaurant/take-away.”

He was asked about that in the course of giving evidence in chief:

“Question: Do you have any recollection of any stock that you found on the premises?

“Answer: Not at that time when I first arrived, but as I was going round the property having a look during the investigation, I was quite surprised of the cleanliness of the property, considering it had been open until the early hours of the morning.

“Question: What about the amount of the stock there?

“Answer: Well, again that was brought to my attention by a fireman stating he was surprised that there were no remnants of food lying about, at which time one of the crews was having a look through the freezers to make sure we hadn’t - if we could, we would put the electrics back on to salvage some of the freezers to keep them working if there was stock, and it became apparent that there was no stock in the fridges.”

And he went on:

“… part of the fire service’s job is to mitigate damage and, if possible, when we’ve knocked electricity off or it has been tripped, if we can isolate certain circuits that have been affected by fire and keep circuits on such as freezers, that sort of thing, we will do that at the time.”

Mr Bott confirmed, also, that he found no alcohol on the premises, only an empty beer barrel; and that there were some optics behind the bar but they had no bottles in them, which he thought was unusual for an operating restaurant which, as this one was, was fully licensed. He was asked whether he had any reason for his recollection and he produced some notes which, as he said, had been made contemporaneously. The notes recorded:

“During the investigation it was apparent there were very little contents in the storage areas and fridge/freezers which would normally be expected in a restaurant/take-away, i.e. food and drink.”

And he confirmed in answer to a question in cross-examination that he had looked for food and drink with the scenes of crime officer who was with him at the time, PC Rollison, and that the only thing he could add to Mr Rollison’s evidence was that within the take-away area on the ground floor there was a glass-fronted cooler which was typical of that sort of establishment, but there was nothing in the cooler and no remnants on the counter either.

The scenes of crime officer, Mr Rollison, gave evidence to much the same effect. In chief, he was asked whether he could describe what he had found by way of stock when he walked around the premises. His answer was this:

“At some point in my examination, I think it came to my notice via a fire fighter coming to me and saying, ‘There isn’t much food here,’ and I thought, ‘Well, that’s odd’. So I think myself and another officer searched the premises for food. We found that there were two or three chest freezers, which I think were near the back door and, in those, there were possibly half a dozen bags of frozen chips, just enough to cover the bottom of the chest freezer, and there was a doorway between the kitchen and the take-away area and by that doorway - I think that’s where it was - there was a tall fridge/freezer or a tall freezer, and in that all the shelves were empty except the top one, which had a cardboard box with some hamburgers in. I think other than that we found a tray of onions and a few bits and bobs, but nothing substantive. We didn’t find a supply of drinks. We didn’t find any meat, which surprised me.”

Mr Yaqoob’s evidence was to the contrary. In the course of his evidence under cross-examination he said that he had on the premises a float of about one week’s requirements. He agreed that that would be the best part of £2,000 worth of food or stocks with it. He was then asked this question:

“Question: When the fireman and police went to the fire, they found, I think, about six bags of frozen chips and one cardboard box of burgers.

“Answer: I don’t think so, no. No, that is total lie. I don’t think so.

“Question: What else was there then?

“Answer: There was a lot of food there. There was a lot of frozen food. Six boxes is a joke.

“Question: What else was there then, Mr Yaqoob?

“Answer: Well, I cannot remember every item that was there, but there was – everything was everywhere. So it’s hard for me to say what was there. I was never told there was going to be fire, so I make a list of it, but it was a fair bit. It was a week supply.”

And he confirmed that there were lots of cans of soft drink; 4 boxes, containing at least 100 cans.

The stock level to be expected at the premises was the subject of evidence from Mr Yaqoob’s accountant. He was asked in the course of cross-examination, “What sort of stock would you expect to find in the freezers?” and he answered:

“I would have thought – I don’t think there was a lot of alcohol, was there, on the premises, by the look of it? You would have thought probably £2,000, £3,000, and I just wonder, as I say, whether the figure in the accounts is more an estimate that the accountants have put in the accounts rather than a true stock-take, which is not unusual in all the circumstances.

If the evidence of Mr Bott and Mr Rollison as to what they actually found on the morning of 25 September were to be accepted, and were to be compared with what Mr Yaqoob and his accountant said should have been there on an ordinary stock basis, then it is impossible to avoid the conclusion that stock was deliberately run down, reduced or removed in preparation for the existing fire. The two accounts cannot be reconciled. The account of the officers, if accepted, leads inevitably to the conclusion that there was much less stock there than there should have been. But there is, of course, a direct conflict of evidence as to what was actually there on the morning of 25 September 1996. Mr Yaqoob’s evidence is that the stock was maintained at a normal level and that a normal level of stock could have been expected to be there on the morning of the 25 September.

The judge set out the conflict between those two accounts at paragraph 26 of his judgment:

“The defendants submit that after the fire there was little food to be found on the premises and no bottles of alcoholic drink. The inference is that Mr Yaqoob had removed the stock or run it down prior to the fire. The evidence in this connection comes from fire officers and a police officer, principally Divisional Fire Officer David Bott and the police Scenes of Crime Officer, Peter Rollison, both of whom have given oral evidence. Mr Bott’s contemporary note simply records very little food or drink. Mr Rollison’s original manuscript record makes no reference to the matter at all. When following the fire Mr Yaqoob was interviewed under caution by Detective Sergeant Goodwin, he was asked nothing about this. In evidence to me, he explained that the base for his dishes was principally either chicken or lamb. Fresh meat was brought in and cooked daily. The contents of the freezers would principally comprise frozen chips, burgers and baps, together with a few other things such as prawns and scampi. The frozen supplies were delivered on Monday and Thursday of each week. Accordingly, he never carried a great deal of food stock, but would have had a week’s supply in hand at the time of the fire. So far as drinks were concerned, he carried little stock. He preferred customers to bring their own alcoholic beverages. He would bring in some cans of lager and also stocked Cokes and soft drinks.”

Unfortunately, although in that passage the judge identified the conflict of evidence, he did not resolve it. In my view he was required to do so. First, the evidence of Mr Bott and Mr Rollison, taken with the evidence of Mr Yaqoob and Mr Dixon as to the normal stocking levels, pointed to Mr Yaqoob having removed or reduced stock in anticipation of the fire. Second, if their evidence was accepted rather than that of Mr Yaqoob his credibility was necessarily put into question, as indeed Mr Yaqoob acknowledged when he remarked that their evidence was a total lie. That is not to say that Mr Yaqoob could not be believed on other matters. But the judge was bound to say why, if he did not believe him on this matter, he was able to accept the evidence that he had nothing to do with the fire.

As I have said, the judge made no finding on this point. But, as it seems to me, there was only one finding which he could have made upon the evidence. There was no reason to disbelieve Mr Bott or to doubt the integrity of his contemporaneous written notes. It was never suggested to him that the notes were not an accurate contemporary record of what he had found as investigating fire officer on a matter which he was bound to investigate, for the reasons which he gave. The judge suggested, at paragraph 33 of his judgment, that Mr Bott and the police officers had decided at an early stage that there was really only one candidate for suspicion: Mr Yaqoob. But that was in the context of an explanation as to why further enquiries seeking to identify other fire raisers or potential fire raisers had not been made. The judge did not say that he did not accept the evidence of Mr Bott and Mr Rollinson on this point. Indeed, to reach that conclusion would, in the case of Mr Bott, have involved finding that his contemporary written evidence had been deliberately fabricated. That allegation was never put to Mr Bott, and there was no basis for such a finding.

As I have said, all the facts point to the conclusion that the perpetrator was an insider: in the sense that he had keys, knew the premises well, was able to gain access to the premises without activating the intruder alarm, knew where the keys to the cupboard and the intruder alarm were to be found, and knew how to de-activate the alarm. That is the profile of the perpetrator which emerges from evidence which is substantially unchallenged. Add to that the fact that, as I would hold, the judge ought to have found that the premises were deliberately de-stocked of consumables in advance of the fire. Taking those matters together, the evidence points very strongly indeed to the fire having been started by Mr Yaqoob or, with his connivance, by somebody in his family or in his staff.

The first question, then, is: how did the judge deal with that point? At paragraph 32 of his judgment he said this:

“At the end of the day, I must assess the totality of the evidence, including the oral evidence of Mr Yaqoob. There is no doubt that he would have had the opportunity of lighting this fire, or of causing it to be lit by others. There are, however, others who could have committed the offence with intent to injure him and who might also have wished to cover their tracks, for example disaffected past members of his staff or contractors with whom he had been in dispute and would either have had access to keys or might have secured duplicates. Mr Yaqoob, without wishing to point the finger at anyone in particular, gave the police details of a number of persons who might have held a grudge against him. There is the evidence of the two white boys having been seen running away. Significantly, no rational motive is apparent for Mr Yaqoob having done it. So far as the various aspects of his conduct to which I have referred, when they are examined. None of these appears to take the matter any further.”

Whom did the judge have in mind when he said that there were others who could have committed the offence with intent to injure Mr Yaqoob? Those others had to be persons in relation to whom it could be said (i) that they fitted the profile of the perpetrator which emerged from the evidence and (ii) that it was at least equally probably that they had started the fire.

The position as to the keys was this. From the outset there had been three sets of keys. They had come into existence at the time of the refurbishment and renovation. Two of those sets were handed to Mr Yaqoob and the third was retained by the builder, Mr Clarkson, so that he could attend to necessary snagging works from time to time.

As to the two sets of keys handed to Mr Yaqoob: on his evidence one was always in his possession and the other would be given to whichever of his employees needed them from time to time to open up and lock the premises. There was no other employee who needed them for that purpose on the evening and night of the 24/25 September, because it was Mr Yaqoob himself who had locked up. But, in answer to a request for further information pursuant to CPR part 18, Mr Yaqoob had volunteered that the other keyholder on that night was an employee, Mr Haque. But that there was no reason to think that Mr Haque might deliberately set fire to the premises or otherwise have a grudge against him. The trial proceeded on the basis that those two sets of keys had not fallen into the hands of a wrongdoer outside the family and staff.

The third set of keys had been retained by Mr Clarkson. In 1995 his offices had been burgled and all the keys, including those to the Kashmir, had been stolen. The judge accepted that the keys to the Kashmir had a label on them with the description “Kashmir”. There were, it seems, at least two premises in the area with that name. But the profile which I have described eliminates the casual stealer of keys from Mr Clarkson’s premises. The perpetrator of the fire, for the reasons which I have set out, had to be someone who had a sufficient knowledge of those premises to know where the alarm was and where the keys to the alarm switch were. The alarm keys were not amongst the keys stolen from Mr Clarkson’s premises. The only possible candidates who could satisfy those requirements – apart from Mr Yaqoob, his family and staff – were Mr Clarkson and an electrician, Mr John Stride, who had installed the security alarms. Both Mr Clarkson and Mr Stride gave evidence. It was never put to them either that they had been concerned in the starting of the fire; or even that they had the opportunity to be concerned in the starting of the fire. As the claimant with the burden of establishing that on the balance of probabilities he was not the perpetrator of the fire, it might least have been expected that Mr Yaqoob would put to Mr Clarkson or Mr Stride that they had had an opportunity to start the fire. He did not do so. In those circumstances the judge was left with no evidence which pointed to anyone else at all.

The judge, as I have said, accepted that Mr Yaqoob had the opportunity to start the fire. He could not have held otherwise, given that there was no evidence of alibi. The judge did not accept that Mr Yaqoob had a motive; and he did not make a finding on the critical question whether Mr Yaquoob had anticipated the fire by the removal of the stock. Nor did he make any finding on another conflict of evidence which would have gone to an assessment of Mr Yaqoob’s credibility.

That conflict of evidence related to the time at which certain works had been carried out to the premises after the fire. It was the insurer’s case that, in the course of bringing the take-away portion of the premises – that is to say the ground floor – back into use, there had been an alteration of the lobby which gave access to the restaurant, so as to enlarge the take-away portion and shut off any possibility of gaining access to the restaurant upstairs by the staircase. It was said that that was an indication that Mr Yaqoob had no interest in carrying on the restaurant business and had a motive for seeking to abandon it. The judge did not accept that there was an intention to abandon the restaurant. But he was faced with a conflict between (i) the evidence of Mr Yaqoob’s expert, a Mr Chamberlain, whom Mr Yaqoob had told that the alterations had taken place at the time of the conversion of the take-away in 1996 following the fire, (ii) the evidence of Mr Yaqoob that those alterations had not taken place until 2002, and (iii) some rather inconclusive evidence from Mr Yaqoob’s eldest son, Shazad, that he thought they had probably taken place in 1999. The judge did not deal with that evidence at all. In particular he did not deal with the fact that Mr Yaqoob contradicted his own expert as to what he, Mr Yaqoob, had told that expert at the time when he instructed him. That was a matter which must have been relevant to Mr Yaqoob’s credibility.

It is important to have in mind that the judge decided this case on credibility. At paragraph 34 he said this:

“I have observed the demeanour of Mr Yaqoob closely over part of two days whilst he was rightly subjected to a searching cross-examination by Mr Lord [counsel for the insurers]. I detected no indication that he was being mendacious, nor was his evidence materially undermined. Placing that in the scales together with all the other evidence is sufficient to tip the balance in his favour. I find on the balance of probabilities that this fire was not occasioned by him or with his connivance.”

If that were a finding made after a proper balancing of the evidence – placing proper weight on the evidence as should be put in each scale – this court would be slow indeed to interfere with it. But the finding is flawed in three respects. First, because the judge fails to deal with real issues of credibility affecting Mr Yaqoob’s evidence: in particular in the two respects that I have mentioned – (i) his assertion that the evidence given by Mr Bott and Mr Rollinson was a total lie, and (ii) the direct conflict between his evidence and that of his expert Mr Chamberlain. Second, because, in looking at the evidence on the other side, the judge made no attempt to identify the profile which would identify the perpetrator of this fire. And, third, because the judge failed to deal at all with the critical question whether or not there had been a reduction of stock immediately before the fire. The judge’s failure to take those points into account vitiates the balancing exercise that he was conducting in paragraph 34 of his judgment. In my view, this appeal must be allowed on that ground alone. I regret to say that the judge did not deal with the case in the way that was required.

We were taken to a number of authorities which emphasised the proposition that an appellate court should be slow to reverse a finding of fact made by a judge who has had the advantage of seeing and hearing the witnesses. In particular, we were taken to the opinion of the Privy Council in Akerhielm v De Mare [1959] AC 789 at 806 where, after reference to the well-known authorities of the The Hontestroom,Watt v Thomas and Benmax v Austin Motors, it was said that:

“Their Lordships can hardly imagine a case in which the credibility of a witness could be more vital than a case like the present where the claim is based on deceit, and the witness in question is one of the defendants charged with deceit. Their Lordships would add that they accept, and would apply in the present case, the principle that where a defendant has been acquitted of fraud in a court of first instance the decision in his favour should not be displaced on appeal except on the clearest grounds.”

That principle, of course, is compelling; but its force depends upon the premise that the judge at trial has taken advantage of the benefits which seeing and hearing the witness give to him. If, as I have sought to explain in the present case, the judge has not taken proper advantage of that opportunity – by failing to make findings of fact which were essential, by failing to address the question of credibility and by failing to analyse and given proper weight to the necessary conclusions to be drawn from the forensic evidence as to the profile of the perpetrator – then it cannot be enough for this court simply to say, “Oh well, the judge believed the witness and so must we”.

The question, then, is whether this court should send the matter back for a re-trial or whether it is in a position to reach a conclusion on its own. In my view this court is entitled to ask itself the question whether a judge, properly directing himself on the evidence which was before him, could reach the conclusion that Mr Yaqoob had succeeded in showing, on the balance of probabilities, that he was not the perpetrator of this fire. In my view the answer to that question is that the burden was not discharged. I take the view that this court should say so; and that, rather than remitting the matter for a further trial, it should dismiss the claim in the action.

LORD JUSTICE LLOYD: I agree. Although we are differing from the view of the learned judge below, on a matter which does involve a decision as to the credibility of the claimant who was cross-examined at length before the judge, I find it necessary to add almost nothing to what my Lord has said. My Lord has cited all the passages from the learned judge’s judgment, and the only passage from authority cited to us, that I would have cited myself. As he says, there are at least three compelling reasons for the conclusion that the learned judge failed to take advantage, as he could and should, of having seen not only Mr Yaqoob, but also all the other witnesses, and having seen the uncontested forensic evidence, in performing his task of deciding whether Mr Yaqoob had discharged the burden of proof upon him.

The first is the question of considering the implications and inferences to be drawn from the unchallenged forensic evidence. As my Lord says, the judge simply did not, on the face of his judgment, consider what characteristics that evidence compels one to assume as to the perpetrator of the fire. I would only add one point to those made by my Lord in support of the forensic evidence on the question of timing, namely that the extensive and clearly, if not well thought out, at least elaborate damage to the fire exit door must also have been done before the alarm was activated, because there simply would not have been time to do that after the alarm was activated and before the fire took hold. The conclusion from the forensic evidence is that Mr Yaqoob is a person who has the necessary characteristics and it is impossible to, nor indeed did Mr Yaqoob seriously attempt to, identify anyone, either individually or in terms of a class of persons, who would have shared those characteristics and who was not himself or someone who would have been operating in connivance with him.

The second feature is the evidence as to the stock of the restaurant at the time. My Lord has referred to the conflict of evidence between Mr Bott and Mr Rollison on the one hand and Mr Yaqoob on the other, which the learned judge stated but did not even purport to resolve.

The third is the conflict that arose between Mr Yaqoob’s evidence and that of Mr Chamberlain as to the time at which the works were done which resulted in the restaurant being un-useable as such. Mr Chamberlain’s report stated clearly that he had been told by Mr Yaqoob, unprompted and under no particular pressure at the time, that it was done in the months immediately after the fire. The effect of that work was to make it impossible to use the upper floor as a restaurant, unless of course all of that work had been reversed; and that is a conflict to which the learned judge simply did not refer in his judgment. It may be that if that had been the only respect in which the learned judge’s judgment was deficient, it might not have been possible to say that the only possible conclusion that the judge should have drawn was that Mr Yaqoob had failed to discharge the burden of proof upon him. But that is not so in respect of either the inference from the forensic evidence or the conflict in respect of stock, in respect of which, as my Lord has said, there was really no rational basis on which he could have accepted Mr Yaqoob’s evidence that there was a week’s worth of stock on the premises at the time and rejected that of Mr Bott and Mr Rollison, supported by contemporary evidence, that there was almost nothing on the premises.

That evidence, if accepted as, agreeing with my Lord, it seems to me that it must have been, shows, whatever view one might theoretically take as to Mr Yaqoob’s truthfulness in giving evidence at trial, that his story as to the conduct of the business was inaccurate and that the overwhelming likelihood was that he had prepared for the fire by running down or even possibly removing stock.

The learned judge came to his conclusion on the question of balance of proof at paragraph 34 of his judgment, which my Lord has read. Mr Taylor for Mr Yaqoob submitted to us that, although the judge had not made express findings of fact, particularly on the stock issue, he was in effect deciding that and all other conflicts of evidence compendiously in his acceptance of Mr Yaqoob as an honest and truthful witness. I would accept that, in a case in which there are very large numbers of conflicts of fact, great and less great, it is not necessary for a trial judge to go through every single conflict before explaining why he accepts the evidence of one side rather than that of the other. But it seems to me that, on the three matters to which I have referred, their importance was such that the learned judge must have addressed them expressly in explaining why he accepted the evidence of Mr Yaqoob as against that of three professional men with no incentive to say other than that which they honestly believed, and in the case of Mr Bott supported by notes made at the time which were entirely consistent with his evidence. It seems to me that on that basis, the expression of the learned judge’s conclusion at paragraph 34 cannot be taken as a (however abbreviated or succinct) finding in favour of Mr Yaqoob and a rejection of the evidence of Mr Bott, Mr Rollison and Mr Chamberlain, let alone a resolution of the conflict with which he did not seek to grapple arising from the inferences to be drawn from the forensic evidence.

For those reasons I agree with my Lord that the appeal ought to be allowed, and that it ought to be allowed by dismissing the claimant’s claim rather than by remitting the matter for a re-trial.

LORD JUSTICE WILSON: I agree with both judgments. In the course of his oral submissions to us, delivered as attractively as any advocate in difficult circumstances could have delivered them, Mr Taylor was constrained to make two concessions. The first was that the judge had failed to address the profile of the perpetrator inevitably revealed by the agreed evidence of the forensic engineer. The effect of his evidence was to go much further than the conclusion that the perpetrator had used keys in entering the property; for such a conclusion attributed no significance to his evidence that the perpetrator had devoted time, when inside the property, to damaging the external side door in order to make it appear that he had thereby broken into the property. The judge recited that evidence but he did not, at any rate overtly, reflect on its inevitable and considerable significance, namely that the perpetrator not only possessed keys to the property but considered that he was someone who was identifiable as having been in possession of such keys. In this way the forensic evidence dramatically reduced the pool of possible perpetrators. Insofar as, for example, there was evidence that Mr Clarkson had retained a set of keys, unfortunately labelled “Kashmir”, and that they had been stolen from him by burglars in 1995, the forensic evidence in effect excluded from suspicion the burglars, or all those to whom those keys might subsequently have been passed: for the burglars and their handlers would have no reason to consider that they were identifiable as having been in possession of the keys to the restaurant. In this way the forensic evidence drew the spotlight much more directly upon Mr Yaqoob than, if I may say so with respect, the judge appears to have realised.

The second concession was that the judge had addressed, yet failed to determine, the important, dramatic and arguably pivotal issue as to the amount of food and drink left on the premises on the night of 25 September 1996. In this respect the evidence of the constable and the fire officer raised against Mr Yaqoob’s case a very high index of scepticism. The evidence of the constable was to the effect, as my Lord, Lord Justice Chadwick, has explained, that there was a minimal amount of food at the premises and that, although they were licensed premises, there were no drinks, alcoholic or otherwise, there. The evidence of the fire officer was to almost precisely the same effect and, in his case, was reflective of his contemporaneous notes. When the gist of this evidence was put to him, Mr Yaqoob firmly denied it, saying that there was a lot of food on the premises that night; indeed, a week’s supply. In his judgment the judge noted this substantial issue, yet never purported to resolve it, nor even – which might itself have impacted upon the result – declared that he felt unable to resolve it in either way.

I therefore agree that in the above two major respects the central conclusion, namely that Mr Yaqoob had discharged the burden of proof, was flawed; for it was uninformed either by consideration of the first point or resolution of the second.

It seems to me therefore, as to my Lords, that the judgment in favour of Mr Yaqoob cannot stand; that the appeal has to be allowed; and that this court is required to determine whether the correct disposal is to remit the action for re-trial or to dismiss the claim. In this regard Mr Taylor has been particular eloquent. He has argued that the judge’s conclusion that Mr Yaqoob did not cause the fire nor connive at it was a conclusion of primary fact; that the issue, relating as it does to the perpetration of a serious criminal act, is one which this court should be particularly reluctant to resolve for itself; and that, on the contrary, the issue is pre-eminently one for determination by a court after it has seen and heard all the live evidence and, particularly, so it is said, as a matter of elementary fairness, seen and heard the evidence of Mr Yaqoob himself. These submissions have given me considerable pause for thought. But here it is important to remember where, in accordance with the conclusion of the judge not challenged by Mr Yaqoob in this court, the burden of proof lies. It lies upon him to prove that the fire was not caused by him nor with his connivance. This prompts two reflections. First, that in all the circumstances the burden makes very considerable forensic demands upon Mr Yaqoob; and second that a conclusion that he has failed to discharge the burden is not to be equated with a conclusion that he has, even on the balance of probabilities, caused the fire or connived at it. If there was any significant doubt in my mind as to whether, at a new trial, a judge properly directing himself could reach a conclusion other than that Mr Yaqoob had failed to discharge the burden upon him, I would necessarily have favoured a remission. But into the matrix of all the other significant circumstances referred to by my Lords, any judge hypothetically conducting a fresh trial would have to give such proper value to the forensic evidence as my Lords, and now also I myself, have sought to indicate. Furthermore the judge, in my view as in that of my Lords, could resolve only in one way the issue between Mr Yaqoob on the one hand, and the police and fire officers on the other, as to the presence or otherwise of food and drink on the premises; namely that in this working, licensed restaurant virtually no food or drink had been left that night. When the judge came to add those two features into the equation there would be only one reasonable conclusion open to him, namely that Mr Yaqoob had failed to establish the burden upon him. Remission would in my view be an idle exercise and I agree that, albeit unusually, we should now proceed to dismiss the claim.

LORD JUSTICE CHADWICK: The appeal is allowed, and the claim is dismissed.