England and Wales High Court (Technology and Construction Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
TECHNOLOGY AND CONSTRUCTION COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		20th August 2008

MR. JUSTICE AKENHEAD:

Introduction

1. On 16 December 2002, a devastating fire occurred in the early evening at a warehouse at Wayside Business Park, Wilson Lane, Coventry ("the warehouse"), owned by Fosse Motors Engineers Limited ("Fosse"), the First Claimant. By the time that the fire brigade arrived shortly after 7.15 p.m., there was little which it could do to save the warehouse although the attached offices were saved. Fortunately no-one was hurt. The four female "Agency" workers, who had been the only people working at that time in the warehouse, were able to vacate the premises. The Second to Twelfth Claimants rented space in the warehouse from Fosse and their goods were lost in the fire.
2. Those four workers were actually employed by the Second Defendant, Phoenix Industrial Recruitment Limited ("Phoenix") which was an employment agency which had seconded them to Conde Nast and National Magazine Distributors Limited ("Comag"). Comag also rented space in the warehouse from Fosse. Issues arise as to whether Comag is vicariously liable for the negligence if any on the part of these four workers. Judgment has been entered against Phoenix in default but it is anticipated by the Claimants that it will be unable to satisfy any award of damages against it.
3. The main issues in the case however surround how the fire started, in particular whether it was started by a cigarette carelessly discarded by one of the four Agency workers whilst no-one else was at the warehouse, that is, after about 6 p.m. or by someone else before 6 p.m. or by an intruder who gained entrance before or after 6 p.m. or by some electrical equipment. The relatively unusual feature of the case is that the expert forensic evidence was not able to identify a cause of fire from the physical wreckage left as a result of the fire. Thus, the case and the defence is dependent upon the reliability of the witnesses of fact and in particular upon the veracity (or otherwise) of the four Agency workers. The case has some features of a "whodunit" and certainly what "done it".

The Contract between Fosse and Comag

4. The contract between Fosse and Comag was contained in and/or evidenced by a quotation dated 9 April 2002 prepared by Melissa Undy (as she then was) and sent to Mr Byrne of Comag. She submitted various rates for his consideration. She identified that Comag's requirements were:

"2 … Provide Office, Canteen and floor area for a production line and storage within our site at Wayside Business Park."

Rates were quoted for the use of 350 square feet of offices at £49 per week, 800 square feet of canteen space at £96 per week and 14,500 square feet of floor area at £0.14p per square foot per week. There was reference to the need for customers to insure their goods against all risks. There then followed this under the heading "Conditions":

"To accept this quotation, please sign & return a copy of this Agreement. The contract between us will be subject to our conditions which are attached. We recommend that you read these carefully and you will see that we ask for written acceptance of these and of our rates before any goods are placed into store."

5. On 19 April 2002, Mr. Byrne on behalf of Comag signed the pro forma attachment in the following terms:

"Having been supplied with, and read the conditions of contract, I/We accept them together with the rates submitted by you. I am a responsible official and am authorised by my company to accept these terms and rates."

There is a reference to the quotation in Miss Undy's letter of 9 April 2002.

6. On ordinary principles of contract law, there can therefore be no doubt that, following the communication of the acceptance by Mr Byrne, there was a contract between Fosse and Comag. There is nothing in Fosse's Conditions of Contract which materially assist with regard to any of the issues in the case.
7. It is however pleaded that in effect there was a term of the original agreement or a collateral agreement whereby Fosse and Comag agreed that Comag, its servants or agents would not smoke in the warehouse.
8. Following the entering into of the basic contract, on 24 April 2002, Mrs Undy sent to Comag some further "information" in relation to "a few details that must be adhered to for the smooth running of the site". These "details" related to staff parking, unloading, skips, staff, food, cleaner and this, in relation to smoking:

"SMOKING The building has [a] NO SMOKING policy, this includes all tearooms and internal rooms. There is an extremely sensitive fire alarm that has already been activated by smoking. Please advise your staff that smoking must be done in your car park ONLY."

At the end of this letter she wrote as follows:

"I apologise if this letter seems YOU MUST – YOU WILL, but on a multi-user site like Wayside, we must make sure that all our tenants are looked after to the same standard. Your first point of reference will still be Melissa, who will be moving to the site within the next few weeks."

9. There is no pleaded suggestion that Comag positively or expressly agreed to this. It is however the case that Comag sought to comply with the no smoking policy.
10. I am satisfied that the letter of 24 April 2002 did not and was not intended to have any contractual effect. The wording of the letter suggests that the matters raised were to be applied for the smooth running of the site. Comag's Counsel used the appropriate expression "house rules" to describe what this letter was addressing. I do not consider that the content of that letter was intended to give rise to a legally binding contract. The parties had just entered into a signed binding contract with extensive conditions which did not include any requirement as such not to smoke at the site or to comply with house rules or similar requirements. The letter does not ask for Comag's agreement to the rules, albeit that there is an expectation. The requirements in the letter represented management requirements from Fosse which were not intended as such to have absolute or binding legal effect.
11. That this view is correct is supported by the evidence of Melissa Smith (née Undy) who confirmed in evidence that Fosse had a standard contract in place which Fosse adjusted for each and every customer's requirements sending out a quotation with the terms and conditions to be accepted. She later said that, if there had been an agreed change in the contractual terms and conditions, she would want to record this in writing as being such a change.
12. It cannot be said (and indeed it was not argued with any force by Fosse) that there was an implied term of the earlier signed agreement that Comag its servants and agents would not smoke in the warehouse. It is not necessary or necessarily reasonable to imply such a term. Whilst one would readily imply a term that Comag would exercise reasonable care in and about its occupation of the premises, such a term would be apt to provide some protection so far as smoking is concerned.
13. Accordingly, I am satisfied that there was no term of the Comag agreement or indeed any agreement to the effect that Comag, its servants or agents would not smoke in the warehouse.

Description of the Warehouse

14. In or shortly before April 2002, Fosse acquired the warehouse. The warehouse itself comprised some 81,000 square feet in area with an adjoining two storey L-shaped office block of approximately 23,000 square feet. The total site area was some 6 acres and it was situated close to Junction 3 of the M6 near Coventry, indeed adjacent to a roundabout off the M6. The warehouse and offices were built in 1976. The buildings were served by three car parks, one at the front and one at the rear of the offices and a third on the other side of the entrance road to the site.
15. The warehouse comprised an open area having dimensions of approximately 105 metres by 70 metres, with the sides of 105 metres being on the west and east sides. The office block was on the north side. The external walls were brickwork with brick buttresses up to approximately 3.125 metres in height with profiled steel sheet from there up to the eaves at approximately 6.7 metres height. Internally, block work walls were infilled between the brick buttresses. The warehouse was divided into three bays of equal width by two lines of steel columns supporting the roof structure, which comprised three shallow roof pitches orientated along the length of the building (that is north to south) formed by steel joists and purlins overlaid with corrugated asbestos cement sheets with intermittent translucent roof light panels.
16. What has been called the Comag Bay comprised approximately one-third of the floor area on the east side and ran the whole length of the building (105 metres) north to south. There were some fire escape doors (including Door 11) on the eastern wall. At the northern end, access was provided through a door (Door 1A) and a roller shutter door for bringing in and taking out goods. At the southern end of the Comag Bay was what has been called the "video refinishing unit" which was an area contained by block work walls in which the four Agency workers worked. At the northern end of the building was the canteen which Comag and its workers were allowed to use (and forming part of the area rented to them). That was a separate room which was accessed by a door at right angles to the Comag Bay. There was another working door at the other end (the western side) on to a corridor. Turning to the right would take one through a security door (Door 3A) and then an outer door (Door 3) to the outside area. Turning left out of that canteen door would take one down towards the body of the warehouse used by persons other than Comag. One would then need to turn right towards the ladies and gentlemen's lavatories, which were a few paces down on the right. Between the canteen and the outside wall were offices used by Comag and Fosse.
17. In the body of the warehouse not occupied by Comag was extensive racking (erected by Fosse) constructed of steel some two inches by four inches in dimensions. This was high bay storage racking on which customers' palletised goods were to be stored. There were aisles between the racks to allow for the passage of fork lift trucks. The bulk of the central and western bays of the warehouse were filled with racking and customers' goods on pallets by the time of the fire. In the area adjacent to the southern wall of the lavatory block and the canteen were what have been called Racks 1 and 2. Racks 1 and 2 were divided by an aisle, Aisle 1, to enable fork lift trucks to load pallets on to the racks. Each rack was some 5 foot to 5 foot 6 inches above the ground or lower rack as the case may be. On the south side of Rack 2 was another rack back to back with it and another aisle for the passage of fork lift trucks. In all probability it was in Rack 2 on Aisle 1 that the fire emanated. Towards the southern end of the warehouse were placed light beam smoke detector reflectors at eaves level. These were numbered 1 to 6 with the unit No 6 being placed close to the central line of the building and broadly opposite the lavatory block described above.

The Background History before the Fire

18. In or shortly after April 2002, Fosse set about providing the racking upon which goods were to be stored. From April 2002, it was always the case that there was a no smoking policy within the building. As the letter of 24 April 2002 from Mrs Undy stated, there was an extremely sensitive fire alarm which was installed by Fosse. It was not true that by the time that letter was written it had already been activated by smoking: the fire alarm had been activated by the fumes from a fork lift truck. Although a bus shelter was eventually provided some 70 metres from the north side of the warehouse across the car park for smokers to go to smoke, external ash trays were provided in two places within the courtyard between the north end of the warehouse and the L-shaped office block attached on the north side as well as another ash tray on the west side of the office block.
19. Comag's use of the warehouse was initially only involved with the taking in of numerous pallets of magazines and with some 50 to 60 staff (mostly women) putting insertions into or onto the magazines (by way of advertising materials or special offers). The pallets were brought in through the roller shutter doors at the north end of the Comag bay and, doubtless after the magazines had been dealt with by the staff, then removed on pallets through the same area.
20. Fosse retained a company known as Meridian Security Services ("Meridian") to provide security at night (6 p.m. to 7 a.m.) and also at weekends. Mr McNicol was the director in charge. By mid June 2002, Comag began to report that various types of stock had been removed from the warehouse: in effect Comag believed that the stock was being pilfered by people other than their workers. Fosse wrote to Mr McNicoll on 19 June 2002 in the following terms:

"We are writing to inform you that over the past few weeks Comag have been reporting that varying types of stock have been removed from the warehouse that they cannot account for.

As this is a very serious matter they have now informed the police who will be interviewing everybody that has any access to the warehouses.

Can you please give all the names and address[e]s of all the security guards that work on our site when the police contact you.

We have now shown you that the guards no longer have any need to enter the warehouse, they still have access to set the alarm but no entry further than that."

This last quoted entry reflected two matters. The first was that at least the security guards were under suspicion (justified or not) with regard to the pilfering. Certainly up to that point, the security guard at night had had access into the warehouse at night. Secondly, it reflected the fact that by this stage Fosse had introduced certain security doors including Door 3A which had a key-pad lock in respect of which the number was known only to a few people excluding the security guards and Comag's employees.

21. At a meeting between Comag and Fosse representatives on 23 July 2002, a number of matters were discussed. It was confirmed that a "sheltered area" was being built to give cover from the elements for smokers. That was to be erected by early October 2002. Secondly, it was recorded that security cameras were to be erected at the rear and side of the building within the next fortnight. This aspiration was not achieved prior to the fire. However, there was on the corner of the office block overlooking the main car park an old CCTV camera which was not in fact linked or used. However that fact does not seem to have been generally known by those who worked in the warehouse. The CCTV camera, albeit not operating, appeared to cover the route through one of the car parks between Door 3A and the smoking bus shelter arrangement some 70 metres away.
22. By the end of July 2002, Comag had expressed interest in using the self-contained unit at the south end of the Comag Bay for use as a more secure area. Comag wanted to use the area to "clean" videos, the idea being that the videos could then be used again. On 30 July 2002 Melissa Undy wrote to Comag offering the area of approximately 4,000 square feet at £0.14p per square foot. This was agreed to by Comag. Fosse cleared out the room, concreted in an existing vehicle inspection pit and put in an internal shutter door to connect the room with the remainder of the Comag Bay. The videos were delivered through Door No 14 on the south side of the building direct into the room.
23. Comag usually worked between about 8 a.m. in the morning finishing by or before 6 p.m. From time to time however as from August 2002, Comag had some of their staff working late, usually up to about 10 p.m. This gave rise to a potential and occasionally actual problem because the Fosse staff had left by 6 p.m. It was reported to Miss Undy in early August 2002 that doors were not being closed at the end of each working day by the Fosse workers. She wrote to Mr Holes of Comag on 2 August 2002 in the following terms:

"It has been brought to our attention that doors are not being closed at the end of each working day.

The door through the corridor next to your kitchen must be closed, as this is the main security door [Door 3A].

The fire exit at the side of the building is only supposed to be open for your mailing collections and as soon as they are collected…This is to be shut. For the past two weeks our foreman has had to shut it.

The lights have also been left on.

Our staff are on site until 6 p.m. and all doors are shut by them as they leave, if you are to have staff on site after this time you must accept responsibility for making the building secure. Can you please ensure that doors are shut and lights turned off."

I refused a late application by Fosse for permission to plead a contractual obligation on the part of Comag in effect to ensure that the building was secure at all times.

24. By 3 September 2002 another problem had emerged which was to do with the accumulation of rubbish, primarily paper and plastic at the northern entrance to the Comag Bay. At a meeting attended by Mr and Miss Undy and various Comag representatives, the following was noted:

"Housekeeping

Fosse raised the issue of general levels of tidiness in that part of the warehouse controlled by COMAG. It was agreed that COMAG management would issue guidelines to staff regarding this and that improvements would be made."

25. This issue was not immediately or fully addressed by Comag. On 6 September 2002, Fosse wrote to Comag in the following terms:

"In the meeting we had on Tuesday 3 September we mentioned housekeeping; this not only applies to the warehouse but also the outside loading bay.

This week the area around your bins has got increasingly worse, I mentioned to Adrian Slade on Wednesday about getting this area tidied up, which he did instruct your staff to do, but today once again it is even worse. There are bits of paper, plastic bandings, old shrink-wrap holders and broken pallets just thrown near to the skip.

As we explained to Comag before you moved on to this site, this is a multi-occupied site with various different tenants, and these bins and rubbish at the very front of the warehouses are making it an eyesore.

We will monitor this situation for the next week and if there is no improvement we will instruct our staff to tidy the area, neaten up all the stacks of useable pallets and sort through and get rid of all the broken pallets and rubbish. We will charge Comag for this.

I will speak to you in the week to let you know that the problem has been rectified."

Comag's skips were at the northern end of the warehouse just outside the roller shutter doors. Comag did take steps to clean up the warehouse on a reasonably regular basis together with the area outside the warehouse, but there is no doubt that from time to time thereafter there was, inevitably, a certain amount of paper and other debris left lying around both inside and outside the warehouse.

26. On several occasions in August and going into October, Comag had some of their staff working until 10 p.m. Comag told Miss Undy that Adrian Slade of Comag would be the supervisor. This information was transmitted to the security guard who, coming on duty at 6 p.m. would need to be aware of this fact.
27. By 9 October 2002, the video room was available to be used. As from 9 October 2002, Comag introduced a two-shift system with regard to the work in the video room, 6 a.m. to 2 p.m. and 2 p.m. to 10 p.m. There was no agreement that after 6 p.m. the Agency Workers who were to work in the video room would be supervised by Comag supervisors.
28. By late October 2002, Miss Undy was complaining of "lots of rubbish blowing around the site, pieces of scrap, cellophane wrapping, it also once again is an eyesore outside the building around your skips". It was clear that the earlier problem had re-emerged. Again, Comag took steps to clear up the rubbish, paper and other materials.
29. By late October or early November 2002, concern was being expressed by Comag about the temperature within the warehouse; there was some wholly ineffective or even non-operative heating. Comag had taken to using what was thought to be a dangerous blow heating arrangement. On 28 October 2002, Fosse obtained a quotation from Midland Combustion, a firm of industrial heating engineers to provide a number of heaters which were to be gas-fired, albeit electrically operated. Later in November 2002, Fosse retained Midland Combustion to carry out this heating work.
30. On the morning of 14 November 2002, fire alarms were activated. Miss Undy went into the ladies' lavatory and smelled air freshener. She saw ash on the tile space above the actual toilet. She reached the conclusion that a woman had smoked in the toilet and activated the smoking alarm. She e-mailed Mr Holes of Comag reciting the facts and saying:

"As we have advised you before there cannot be any smoking in the building.

1)It's a warehouse 2) the smoke/fire alarm system is linked to the fire station and the tiniest bit of smoke will set it off.

This is now the second time this has been activated by people smoking in the warehouse and as you may or may not be aware we are charged by the fire service for every false call out …

We really cannot accept this any more and if you cannot control what your staff are doing then we are going to be looking at other options. … We am aware that you are using agency staff at the moment and that this is where your problems are seeming to stem from, perhaps if you spoke to your agency contact as surely it is their duty to send you trustworthy staff."

31. Mr. Holes responded within an hour to that e-mail in the following terms:

"As you well know controlling staff is easier said than done. Steps have been taken in the past but obviously to no avail. So another approach has been taken, now only time will tell if that works. If you have any suggestions that you could recommend to me to eliminate this problem, I would be very grateful. This situation annoys me as much as it does you."

It seems clear (and Mr. Holes did not seriously dispute it) that one of the Comag staff or its agency workers had smoked in the lavatory that morning. Given the timing of this incident, none of the four Agency workers who worked the later shift was involved. Mr Holes gave all the staff a very serious "talking to" about how important it was not to smoke. He also emphasised to them that there were areas of the warehouse into which they could not go and he told them something along the lines that there was a £50,000 smoke alarm in the warehouse which was extremely sensitive and that they would not get away with smoking in the warehouse undetected so they should not do it.

32. Notices were put up in the toilets shortly thereafter in the following terms:

"To all personnel using these toilets

Can you please be aware it is strictly no smoking within the buildings.

There are smoke sensors in all toilets that are linked to the fire alarm system which is linked directly to the fire station.

The smallest amount of smoke sets this off.

Thank you.

Melissa Undy Warehouse Manager."

33. The four Agency workers who were in the building on the evening of the fire were Michelle D, Susan W, Charlene C and Laura B. Given that serious allegations are made against these ladies, I am withholding in this judgment their surnames. I will refer to them hereafter simply by their Christian names. All four worked on the second shift, that is 2 p.m. to 10 p.m. As to each of them:

(a) Michelle was 31 and had worked for Phoenix for two to three months working elsewhere. She started work at Comag's warehouse in October 2002 working in the section which dealt with magazines; she packed the magazines into boxes. Towards the end of October 2002 she was asked by Mr Slade of Comag to work in the video room on the 2 p.m. to 10 p.m. shift. She agreed to do so and worked there on that shift until the evening of the fire. She was the supervisor of the others.

(b) Susan was 35 years old and had worked as a hairdresser for some 13 years after school. She was employed in October 2002 by Phoenix at the warehouse, initially working in the magazine section. Again in late October or early November 2002 she moved to the video section on the later shift.

(c) Charlene was 17. She had clearly left school when she was about 16 and had worked for several companies before being employed by Phoenix in the late summer of 2002. She worked on the later shift throughout in the video room as from October 2002.

(d) Laura was the youngest of the four and also 17. She left school at 15 and also worked for several other companies before being signed up with Phoenix in November 2002. She and Charlene had known each other since birth as their families were close friends. She and Charlene sometimes referred to themselves as being cousins which was not strictly speaking correct. She began work in the magazine section in or around mid-November 2002 and several weeks later joined Charlene, Susan and Michelle in the video area on the second shift.

34. Only Laura and Susan were smokers. By the time of the fire, Susan was trying to give up. The other two were not smokers. All four however were aware not only of the no smoking policy but also that there was or they believed there to be a very sensitive smoke alarm at least in the lavatories. They were also aware that they were not permitted to be in any part of the warehouse other than the Comag Bay, the Comag canteen, the corridor outside the canteen leading to the exit doors (3 and 3A) and to the lavatories. They all believed that if they were caught smoking anywhere in the warehouse or they were found in areas of the warehouse where they were not entitled to be, that would be considered most seriously and would probably result in their dismissal.
35. By 10 December 2002 the heaters which were to be provided had been installed but the gas remained to be connected and finished off. It is clear that this heating system remained to be commissioned. The installers were working on Friday 13 December 2002 and left a tower scaffold in the Comag Bay near the top of Aisle 1 between Racks 1 and 2. The evidence is such that I cannot make any findings as to precisely where it was but I will return to this aspect of the matter later in this judgment. The installers did not return on 16 December 2002.
36. On 16 December 2002, the day of the fire, Miss Undy wrote to Mr Holes by e-mail asking that his staff pick up "all the rubbish that has blown around to where the Fosse cars had parked".
37. By 16 December 2002, Fosse had not installed either sprinklers or any compartmentation within the warehouse. However Fosse had spent some £250,000 to £300,000 at the warehouse. This included a new fire alarm system, linked to a monitoring agency which would normally and should have passed through to the requisite fire service any material alarm activation. The fire alarm system installed by Fosse relied upon a beam which would become interrupted by smoke.

The Day of the Fire

38. It seems to have been "just another day" at the warehouse. Nothing untoward particularly happened. The four agency workers came on duty at 2 p.m. Sue Wilcox of Mercia Print & Packaging Ltd (the third claimant) visited the premises on 16 December to inspect some of their stock for stock-checking purposes. She says in her witness statement (which I accept) that she visited the warehouse during the morning of 16 December 2002 to conduct a stock check. The stock check took several hours and when she left site it was beginning to get dark. From that I infer that she left about 3 to 4 p.m. At no time whenever she was in the warehouse did she see or smell any sign of fire or smoke.
39. By about 5.55 p.m. the Fosse employees had left. These included Melissa Smith (nee Undy), Mr Seymour the warehouse foreman and at least two other male employees of Fosse. Apart from Mrs Smith (who smoked about one cigarette a year), all the other Fosse male employees smoked.
40. At about 6 p.m. Mr Kariuki, the Meridian security guard, recorded himself as having arrived at the site and installed himself in the guard hut near the main entrance. The front gate which gave access was shut.
41. At about 6 p.m., the four agency workers proceeded from the video room along the eastern bay into the canteen where they had their "dinner break" which was to be for half an hour. Laura and Charlene told the court and the fire officers that at about twenty past six they left the canteen, turned right went through Doors 3 and 3A, across the car park to the bus shelter for Laura to smoke a cigarette. Charlene went with her, she said, to keep Laura company. It was, they said, as cold in the warehouse as outside (given the lack of heating); there certainly was not much difference between the two. They say that they returned at about 6.25 p.m. There is an issue as to whether Door 3A (the key-pad door) was propped open or whether they banged on the door to attract the attention of the others to let them in. At some stage towards the end of the dinner break, Susan told the court that she went to the ladies lavatory and then returned to the canteen. At about 6.30 p.m. all four then went back to the video room.
42. The four gave evidence that at about 6.45 to 6.50 p.m., Laura told the others that she needed to go to the lavatory which she did. She said that she followed the official route along the bay through the canteen out into the corridor and along to the ladies toilet. She says that before going into the toilet she thought she might have smelled smoke but thought nothing of it, went to the lavatory and came out. She said that, as she was leaving the ladies lavatory, she saw through Rack 1 an incipient fire in Rack 2 and ran shouting to tell the others who were in the video room.
43. I am satisfied that what happened then was that all four went along the bay and stood at the Comag Bay end of Racks 1 and 2 by Aisle 1 and looked down and saw that there was indeed a fire at the first level several bays down on Rack 2. Michelle told the others to leave the building but she had forgotten her mobile telephone so she went back to the video room to get it. She then followed the others out and contact was made with Mr Kariuki. The fire alarms began to go off in the order set out below. The timing thereafter is as follows:

19.04: smoke detection, zone 6 (beam detector) activated

19.06: smoke detection, zone 3 (beam detector) activated

19.08: smoke detection, zone 2 (beam detector) activated

19.08: fire alarm, first floor west (detector type unknown)

19.13: West Midlands Fire & Rescue Service receive an alert on the 999 system

1915: Warwickshire Fire & Rescue Service passed alert they had received from the auto dial assistant to central control

1917: the first West Midlands Fire & Rescue Service appliance arrived at the scene and found the fire through the roof.

1918: the second West Midlands Fire & Rescue Service appliance arrived and Station Officer Thomas, at that time the Incident Commander, immediately requested the attendance of an additional six pumping appliances and two hydraulic ladder platforms.

Beam 6 was centrally located and was in line with the area where the fire was noticed by the Agency workers.

44. Thereafter the Fire Brigade was unable to save much of the warehouse but it was successful in ensuring that the adjacent office building was not materially damaged by the fire. Essentially the warehouse and its contents were damaged beyond repair.
45. Some time after the Fire Brigade arrived, the four Agency workers were taken home but were brought back to the warehouse site several hours later to be interviewed by Alan Pellowe who was the Station Commander at the local fire station holding the rank of Assistant Divisional Officer. Thus these interviews were carried out at about 10 to 10.30 p.m. Each was interviewed sitting in the back of a car with Mr. Pellowe in the front seat. He noted, materially, the following in relation to each of the ladies:

"Laura B (17) … 3rd week of work in video section – break 18:00-18:30. Went to smoke outside in car park @ 18:20 with Charlene. Came back to canteen. Joined mates and returned to work – 19:00. went to toilet, smel[led] smoke, saw fire in pallets (1st shelf up) sparking like electrical – ran back to friends shouting Michelle (she made big thing about how she had got everybody out). Took friends back to fire – we got out – alarms went off as we left building.

Michelle D (38) – team leader … worked for Phoenix six months – after 18:30 break Sue went to toilet, we waited + all went back together – Laura went to toilet @ 19:00 (5 – 10 minutes) She came back saying there was fire (thought she was joking at first). We went with her. Saw fire in pallets by canteen – nothing unusual about flames – no sparks. But they were coming from each end of pallet (white cardboard boxes of paper) – Alarms went off. We left building.

Susan W (35) … Worked there since 22^nd October 2002 – returned to work area after break @ 18:30 with others – Laura went to toilet @ about 19:00 – raised everyone when she came back stating fire – went back with everyone to fire in pallet by canteen – boxes on pallet – fire coming from centre – looked as though it was sparking – fire on first shelf – outside when alarms went off – trying to give up smoking – had saved last two cigs for 20:00 break.

Charlene C (17) (very nervous) … – been working here for two months – nobody else in building Comag workers left @ 18:00 – Laura went to toilet a@ 19:00 – came back after 5 mins saying there was fire – we all went to see – fire in pallet by canteen not that big in centre of pallet. We left building alarms went off outside."

46. Mr. Pellowe also wrote in his notes:

"Smoking Materials Charlene and Laura smokers Susan W …( trying to give up), company has strict no smoking policy. …

Arson/Accelerants not ruled out yet

First Impressions – Careless disposal of cigarette – possible deliberate by Laura or unknown."

47. A number of other people had come down after the Fire Brigade was called. They included Mr and Mrs Undy, their daughter, Mr Seymour, Mr. McNicol, Miss Sweeney of Phoenix and Mr Stockley of the contractors who had been working there on the heaters the previous week. Mr Greenhough of Fosse was also on site.
48. On 18 December 2002, the respective insurers of Fosse and Comag had appointed respectively Mr Bourdillon and Dr Graham as investigative fire experts. They visited the site on that day but the fire was still burning. With other representatives of the loss adjusters, and Mr Lynch from the Fire Service, they visited again on 19 December 2002 where they continued their investigations together with a Mr Pugh another expert acting on behalf of the insurers of one of the firms which owned goods stored at the premises. On that day they interviewed the four Agency workers with Dr Graham taking the lead in the interview and also Mr Kariuki for whom Mr Bourdillon took the lead. Notes were taken by Messrs Bourdillon, Dr Graham and Mr Pugh. There are some differences between what the four ladies said on that day compared with what they are recorded as having told Mr Pellowe.
49. As it had not been practicable for the experts to investigate the debris which remained in particular in the area where the four ladies had indicated they saw the fire, it was not until 26 February 2003 when a colleague of Mr Bourdillon (Ms Jo Goodridge) and Dr Graham were able to attend and see the removal of the material in the area. I accept Dr Graham's evidence that as a matter of fact this clearance operation was not done in a way which was particularly helpful so that it was not possible to determine whether there was present any electrical equipment, in particular that which might have been left by the heating contractors, in the area of the fire. The contractors had used for the clearance operation a large tractor digger with a substantial RSJ attached to the front which was a wholly indelicate tool to enable effective sifting of the debris.

The Witnesses

50. As much of this case depends upon the credibility of the four Agency workers, it is right that I should give my impressions of those witnesses when they gave their evidence in court. Without intending any disrespect to any of the four ladies, I bear in mind that they were not sophisticated or well educated people. Certainly the two younger ones had left school at the first available opportunity. None of them, I understood, had given evidence in any court before.
51. I also bear in mind that they were subjected (within wholly proper bounds) to what was in effect a relatively hostile cross-examination. In effect, Counsel for Fosse was seeking to establish that the four ladies had in effect "put their heads together" in an exculpatory way to seek to come up with similar stories about what had happened. In effect he sought to argue that one or more of them was lying. In particular he sought through cross-examination to explore the possibility that it was either Laura or Susan who had carelessly discarded a cigarette in the area at which the fire was later observed by all four of them on their way out of the building. However, in their different ways, they all stood up very well to relatively intensive cross-examination.
52. Although I accept that there are some discrepancies between the ladies' accounts to Mr. Pellowe, their interview on 19 December 2002, their witness statements and their oral evidence, I was, broadly, favourably impressed by each one of the four witnesses in question. I believe and accept that they were seeking to tell the truth and that none of them was in substance lying to the court or being anything other than truthful in making clear that they were not involved in any way in smoking within the warehouse or carelessly discarding any cigarette into the area where the fire originated. My impressions, other than that, as to the individuals were as follows:

(a) Michelle: she had somewhat impaired hearing and on some occasions needed to have questions repeated, particularly when Counsel or I was not facing her when speaking. She gave me, very strongly, the impression that she was essentially an honest person. She gave her evidence generally in a firm and clear way and gave me no impression that she was lying or gilding the truth. She made concessions in effect against herself which was not the hallmark of someone who was obviously lying. For instance she accepted that the external door 3A was occasionally propped open and that it was wrong to do so (T3/202). Where there were discrepancies between her oral evidence in court and earlier statements or interviews, I formed the very strong impression that these were mostly attributable to the lapse in time (over five and a half years) since the night of the fire.

(b) Laura: she was clearly very nervous when she came into the witness box to give evidence; indeed, she could hardly speak. However she settled down after some questioning in chief and gave her evidence, particularly under cross-examination, with confidence. I was favourably impressed with how this unsophisticated young lady stood up with such confidence to the cross-examination which she underwent. Again she, like Michelle, made concessions when required. For instance,she had given evidence that there was a metal railing dividing the route to the toilets from the canteen corridor area from the warehouse area. She conceded (T4/82) in the light of cross-examination that she was obviously wrong in her recollection.

(c) Charlene: she was a relatively quiet witness who gave her evidence with some assurance and confidence. I formed the view that she was telling the truth. She did not always support Laura who was a good friend of hers in some of the evidence which Laura gave. For instance, Charlene's evidence about who heard some noises in the warehouse one night when Charlene was working did not support what Laura had said. Similarly, she made it clear that she could not remember an occasion when Laura had said she had found a security door open at night. That was not consistent with Laura and Charlene having "put their heads together" to concoct a story.

(d) Susan: she was the least unsophisticated of the four Agency workers. She gave her evidence in a clear and confident manner and, guilelessly, in my view. She conceded that she could not remember whether there was a steel railing dividing the area by the lavatories from the warehouse; she had effectively said in her witness statement that such a steel railing was there. It would have been easy for her to say, if she was lying, that she stood by her witness statement.

53. There does not appear to have been much time for the four Agency workers to have concocted a story (see below). There would be no particular motive for any of the four who was entirely innocent of what is now alleged, namely the careless discarding of a cigarette, to have stood by the one (if it was the case) who had discarded the cigarette. It was suggested by Fosse that they were trying to exculpate themselves but it would have been easier, safer and simpler for the innocent ones to tell the truth. For instance if Michelle and Susan knew that Laura and Charlene had gone not to the outside bus shelter but into the body of the warehouse to have a cigarette, there is no obvious reason why they would conspire to fabricate a story which exculpated those two. They were work colleagues; they were not friends and they were divided by about 17 years in age. They were in any event not going to be able to work at Comag any more given that the warehouse had burned down.
54. I would have been surprised in the circumstances if there were not some discrepancies in the four workers' recollection of events as relayed to Mr Pellowe, at the interview on 19 December 2002, in their witness statements and indeed in evidence. It must have been at least somewhat distressing and upsetting to have to vacate a building which is on fire, then see the fire brigade arrive and the building being destroyed in flames. I am not surprised as to discrepancies as against what they told Mr Pellowe: for them the circumstances of the interview at or after about 10 p.m. in the back of the car whilst the fire was still raging would seem odd and unsettling to most people. This applies particularly to the two 17 year old girls who must have found the experience extremely strange. Discrepancies between the interviews on 19 December 2002 and the remarks made to Mr Pellowe are explicable by the fact that the later interviews seem to have been more structured and logical; they went into more detail. The witness statements at least for two of the ladies apart from Michelle seem to have been prepared in 2005, over two years after the incident and, again, discrepancies would not be altogether surprising. Michelle's and possibly Laura's were prepared somewhat later and was being signed in January 2008; there is a greater lapse in time until then.
55. I do very much bear in mind that these statements of the four ladies were, clearly, drafted by Comag's solicitors, doubtless based on what they understood the witnesses would be able to say. However the language and grammar are clearly that of the solicitors as opposed to the witnesses. I do not imply any criticism of the solicitors who doubtless wanted the statements to be comprehensible to the court and the lawyers. Some discrepancies between the witnesses' oral evidence and their earlier statements or interviews again are not surprising given the lapse of time. Much was made in closing by the Claimants' Counsel as to similarities between the statements; that however is not surprising if they were broadly telling the truth and linguistic similarities are explicable by the fact that the statements were prepared by the solicitors. If the statements have been "tailored" as suggested by Fosse, that would implicate the solicitors in an attempt to pervert the course of justice; I simply reject that assertion as a serious charge for which there is no factual justification.
56. There seems to have been little opportunity in practice for the four to have colluded. On the night of the fire, they were taken to their homes for several hours before the interview. It is unlikely that during the period at the site with other people around they would have had much opportunity to concoct a story. Similarly, whilst they were at their separate homes, no real opportunity arose. It is always possible that in the van back from their homes to the site, or vice versa they could have discussed matters; indeed it would not be surprising if they did talk to some extent about the fire and possibly how lucky they were to escape safely. They were not together between the night of the fire and 19 December 2002 when the interviews took place. It is extremely unlikely that there was any contact between Michelle and Susan on the one hand and Laura and Charlene on the other during this period. It is distinctly possible that there was some contact between Charlene and Laura who were and whose families were close but it would not be at all surprising in those circumstances if they did not discuss the fire and what might have caused it. That said, I consider that it is most unlikely that the basic version of events which each of them gave on interview was the result of them "putting their heads together" to concoct what would be a fabricated and untruthful story.
57. As to the other witnesses of fact, I broadly found them to be honest and straight with the court. I was not wholly convinced however by some aspects of Mr McNicol's evidence albeit that that may have little impact on the outcome of this case.
58. As to the two main experts, Mr Bourdillon is a well-known and experienced fire expert of many years' standing whilst Dr Graham has been an expert for somewhat less time. The importance of their evidence is somewhat limited because the effect of the fire was so all pervading that no forensic evidence remains as to the cause and to some extent even the seat of the fire. Thus what, primarily, the two main experts have had to do is to consider the various witnesses' evidence (and particularly that relating to the four agency workers) in order to determine what the more or less likely causes of the fire were. Criticisms particularly of Dr Graham by Fosse were unjustified: I formed the view that he was impartial in the way he gave his evidence, as was Mr Bourdillon.
59. Of course, the experts have provided some valuable assistance to the Court. They have assisted in determining, from the information available from the fire alarms, the period of time before they were activated that is likely to have gone by before the fire started in earnest. They were able to provide evidence about the conditions which would or might have to exist for a discarded cigarette to have caused a fire and the period which would be required to cause material to break into flame such that smoke would be given off that led to the fire alarms being activated. In this context Mr Nimmo, a second expert called by Comag, and expert in fire testing and investigations, provided useful information about how easy or otherwise it would be for a randomly discarded cigarette to catch fire on the top of packaging.

Causation

The law

60. There was a very substantial debate as to what principles could or should be applied so far as causation in relation to the fire was concerned. There can be no doubt that in a civil case such as this, it is incumbent upon the Claimants, to prove on the balance of probabilities that the fire was caused by a cigarette carelessly discarded by one of the four agency workers. That is their pleaded case (paragraph 5 of the Re-Amended Particulars of Claim). There was an allegation that the Agency workers were negligent for failing to take reasonable precautions to prevent a fire starting and/or spreading; that has not been pursued. There is another allegation that if the fire was started by an intruder entering after 6 p.m. that was attributable to the negligence of the Agency workers leaving Door 3A open.
61. Causation is essentially a matter of fact. The courts over the years have not laid down any strict rule of causation as to how damage or loss or an event has been caused is to be proved.
62. In Rhesa Shipping Co v Edmunds [1985] 1 WLR 948, the House of Lords considered the case which involved the sinking of a ship, the Popi M which sank in calm weather in the Mediterranean in deep water when laden with a cargo of bagged sugar. The issue was in effect what had caused it to sink. Lord Brandon gave the lead judgment. He said at page 951A-G as follows:

"… the appeal does not raise any question of law, except possibly the question what is meant by proof of a case 'on a balance of probabilities'. Nor do underwriters challenge … any of the primary findings of fact made by Bingham J. The question, and the sole question, which your Lordships have to decide is whether on the basis of those primary findings of fact, Bingham J and the Court of Appeal were justified in drawing the inference that the ship was, on the balance of probabilities, lost by perils of the sea.

In approaching this question it is important that two matters should be borne constantly in mind. The first matter is that the burden of proving, on a balance of probabilities, that the ship was lost by perils of the sea, is and remains throughout on the shipowners. Although it is open to underwriters to suggest and seek to prove some other cause of loss, against which the ship was not insured, there is no obligation on them to do so. Moreover, if they choose to do so, there is no obligation on them to prove, even on a balance of probabilities, the truth of their alternative case.

The second matter is that it is always open to a court, even after the kind of prolonged enquiry with a mass of expert evidence which took place in this case, to conclude, at the end of the day, that the proximate cause of the ship's loss, even on a balance of probabilities, remains in doubt, with the consequence that the shipowners have failed to discharge the burden of proof which lay upon them.

This second matter appears clearly from certain observations of Scrutton L.J. in La Compania Martiartu v. Royal Exchange Assurance Corporation [1923] 1 K.B. 650. That was a case in which the Court of Appeal, reversing the trial judge, found that the ship in respect of which her owners had claimed for a total loss of perils by sea, had in fact been scuttled with the connivance of those owners. Having made that finding, Scrutton LJ went on to say, at p. 657:

'This view renders it unnecessary finally to discuss the burden of proof, but in my present view, if there are circumstances suggesting that another cause than a peril insured against was the dominant or effective cause of the entry of seawater into the ship … and an examination of all the evidence and probabilities leaves the court doubtful what is the real cause of the loss, the assured has failed to prove his case.'

While these observations of Scrutton L.J. were, having regard to his affirmative finding of scuttling, obiter dicta only, I am of opinion that they correctly state the principle of law applicable …"

63. Lord Brandon then went on to consider the approach to the evidence adopted by the first instance judge and referred to the well-known saying of Mr Sherlock Holmes:

"How often have I said to you that, when you have eliminated the impossible, whatever remains, however improbable, must be the truth?"

Lord Brandon considered that it was inappropriate to apply this dictum and indeed set aside the lower courts' finding.

64. At page 955H to 956F, he continued:

"The first reason [why it is inappropriate to apply Mr. Holmes' dictum] is one which I have already sought to emphasise as being of great importance, namely, that the judge is not bound always to make a finding one way or the other with regard to the facts averred by the parties. He has open to him the third alternative for saying that the party on whom the burden of proof lies in relation to any averment made by him must be able to discharge that burden. No judge likes to decide cases on burden of proof if he can legitimately avoid having to do so. There are cases, however, in which, owing to the unsatisfactory state of the evidence or otherwise, deciding on the proof is the only just course for him to take.

The second reason is that the dictum can only apply when all relevant facts are known, so that all possible explanations, except a single extremely improbable one, can properly be eliminated. That state of affairs does not exist in the present case: to take but one example, the ship sank in such deep water that a diver's examination of the nature of the aperture, which might well have thrown light on its cause, could not be carried out.

The third reason is that the legal concept of proof of a case on the balance of probabilities must be applied with common sense. It requires a judge of first instance, before he finds that a particular event occurred, to be satisfied on the evidence that it is more likely to have occurred than not. If such a judge concludes, on a whole series of cogent grounds, that the occurrence of an event is extremely improbable, a finding by him that it is nevertheless more likely to have occurred than not, does not accord with common sense. This is especially so when it is open to the judge to say simply that the evidence leaves him in doubt whether the event occurred or not, and that the party on whom the burden of proving that the event occurred lies has therefore failed to discharge such burden.

In my opinion Bingham J adopted an erroneous approach to this case by regarding himself as compelled to choose between two theories, both of which he regarded as extremely improbable, or one of which he regarded as extremely improbable and the other of which he regarded as virtually impossible. He should have borne in mind, and considered carefully in his judgment, the third alternative which was open to him, namely, that the evidence left him in doubt as to the cause of the aperture in the ship's hull, and that, in these circumstances, the shipowners had failed to discharge the burden of proof which was on them."

65. This case was picked up in the recent case of Ide v ATB Sales [2008] EWCA Civ 424 by the Court of Appeal. Thomas LJ giving the judgment of the court reviewed the Rhesa Shipping case and also a more recent House of Lords decision in Datec Electronic Holdings Ltd v United Parcels Service Ltd [2007] 1 WLR at 1325. He then said this:

"6. As a matter of common sense it will usually be safe for a judge to conclude, where there are two competing theories before him neither of which is improbable, that having rejected one it is logical to accept the other as being the cause on the balance of probabilities. It was accepted in the course of argument on behalf of the appellant that, as a matter of principle, if there were only three possible causes of an event, then it was permissible for a judge to approach the matter by analysing each of those causes. If he ranked those causes in terms of probability and concluded that one was more probable than the others, then, provided those were the only three possible causes, he was entitled to conclude that the one he considered most probable, was the probable cause of the event provided it was not improbable."

66. I do not consider that Thomas LJ was suggesting that there was anything wrong with Lord Brandon's conclusion; indeed the House of Lords decision was binding on the Court of Appeal. He was simply considering what might happen if there were three possible causes, by weighing up the inherent probability or improbability of each of those causes and come to a conclusion on the balance of probabilities as to what the cause was.
67. What is not acceptable, at the very least in a case like the current one, is to identify that there are, say, (as here) five possible causes, rank them each in percentage terms as possibilities and then select the possibility with the highest percentage as the probable cause. The only circumstances in which it would be legitimate would be if the highest ranked cause was the one which on all the evidence the judge was satisfied was the probable cause of the incident or loss in question. This proposition was, I believe, accepted ultimately by Counsel for both parties. I consider that it is dangerous and generally a fruitless occupation to seek to rank possibilities or probabilities in percentage terms in any event. If there are five possibilities of which four are remote or extremely improbable, that conclusion may go to support a judge's finding that the remaining "possibility" is in fact the probable cause or explanation for the event in question.
68. There are numerous cases at first instance and in the Court of Appeal which deal with causation largely as a matter of fact. One such case was Kiani v Land Rover Ltd and Others [2006] EWCA Civ 880. The unfortunate facts of that case were that Mr Kiani went to work at the Land Rover plant; his dead body was found in a tank in the area in which he worked. He had died of asphyxia. His personal representative sued on the basis that Mr. Kiani had accidentally fallen into the tank; Land Rover suggested that his death was suicide. There were thus two possible explanations. The first instance judge had found suicide to be a less than probable explanation, he found that it occurred as a result of accident because the tank had its hatch left open and that Mr. Kiani had probably gone over to have a look, overbalanced and fallen in. Waller LJ giving the lead judgment, dismissing the appeal, said this at paragraph 30:

"It seems to me that some of the criticisms made of the recorder are on any view not justified. First it does not seem to me legitimate to say that [certain] evidence established that an accidental fall was 'impossible'. … Second it is not in my view fair to criticise the recorder for not setting out precisely how any accident occurred anymore than it would be fair to say to the defendants that they should show precisely how a deliberate act of suicide would have occurred. As long as accident can be demonstrated to be possible, it is open to a court which has discounted any other possibility to be of the view that accident has been proved on the balance of probabilities. That must be particularly true where a breach of duty, a duty to guard against the very type of injury with which the case is concerned, has been established. Third, I do not myself think that it is false logic to reason that where only two possibilities are under consideration both of which seem unlikely, if one seems much less likely than the other, the less likely can be discounted thus making the first likely to have happened on the balance of probabilities. …"

I do not consider that this approach is different broadly from the views which I have expressed above.

Causation

The Facts

69. There are essentially five possible causes of the fire:

(a) A cigarette carelessly discarded by one of the agency workers;

(b) A cigarette carelessly discarded prior to 6 p.m. by a non-Comag employee;

(c) An intruder (bent on arson) securing entry to the warehouse before 6 p.m.;

(d) An intruder (bent on arson) entering the premises after 6 p.m.; and

(e) Electrical equipment left by the heating contractors.

70. The experts are agreed, properly and logically, in their First Memorandum of Agreement signed on 30 April 2008 as follows:

"We agree that the available evidence shows that the most probable explanation for the outbreak of fire involves human agency; either carelessly discarded smokers' materials or deliberate ignition. [Dr Graham] considers other alternative credible explanations cannot be eliminated on the basis of the physical evidence; other possibilities include battery chargers for portable equipment or mains powered equipment that was fed from an extension lead."

71. Based on the evidence of the four Agency workers the experts were agreed that "the fire originated within a stock of boxes containing documents and paper publications that were stored on a timber pallet. The pallet was positioned on steel racking at approximately eye level". The position of the seat of the fire was in Rack 2 in Aisle 1.
72. At Paragraph 9 of the Memorandum the experts agreed as follows:

"We agree that if the fire was initiated by a smoulder that subsequently underwent transition to flame, then it is most likely that the cause of fire was accidental. We further agree that there exists no physical evidence that demonstrates that the fire necessarily started as the smoulder. Therefore this proposition is reliant on the recollections of Laura … who reportedly smelled a burning odour as she made her way to the toilet and on the Court finding that that smoke emanated from the same fire, which underwent transition to flame by the time Laura … observed flames."

73. They went on at Paragraph 10:

"It follows we agree that notwithstanding the account of events given by Laura …, if the Court were to find that the fire was not preceded by a smouldering fire then we agree that the cause of fire could have involved a flaming ignition; a possibility that cannot be rejected on the basis of the physical evidence, in isolation. We further agree that in the event the fire was started deliberately, the only credible explanation for the outbreak of fire involves deliberate ignition at, or very shortly before, Laura made her observation of the flames. This is because the reported observations of the witnesses (Laura … initially, and thereafter, the other warehouse personnel) indicate that the fire had not reached significant proportions by the time they observed the flames that had evidently been burning for only a short period of time."

74. At Paragraph 11 of the Memorandum the experts set out their views in relation to timing. Whilst Mr Bourdillon believed that the period of 20 to 40 minutes explained and indicated a period of smouldering prior to transition to flame, Dr Graham and Mr Nimmo considered that the evidence was also entirely consistent with the careless disposal of smokers' materials up to about an hour before the fire was discovered.
75. I am satisfied that the expert evidence in this case does not definitively (on a balance of probabilities) point in itself to any one cause being probable (or more probable than not).
76. I will first address the question relating to the carelessly discarded cigarette. If the cause was a carelessly discarded cigarette, one needs to determine whether the timing of fire first being seen (initially by Laura and then no more than a few minutes later by the other three ladies) is consistent or inconsistent with a cigarette having been discarded in effect between about 6.20 and 6.30 p.m. (that is by one or other of the Agency workers, more specifically Laura or Susan) or a cigarette which might have been discarded before Fosse's employees left for the evening (shortly before 6 p.m.).
77. There was a substantial amount of evidence about this. For the fire to have started at the position agreed as the most likely seat of the fire by the experts (Aisle 1 Rack 2 eye level), there would have to exist conditions which enabled a fire to occur even if a cigarette was discarded. Mr Nimmo's work on this aspect of the matter was not seriously challenged and I broadly accept it. He used 100 part smoked Embassy No 1 cigarettes (the type of cigarette smoked by Laura) thrown on to various filled cardboard boxes to determine whether they would cause a fire. He concluded that the suggested smoulder time for a half smoked Embassy No 1 tipped cigarette would be about eight and a half minutes; it would be more or less depending on whether it was more or less than half smoked. He found, based on his experiment, that out of the 100 partially smoked lit cigarettes, not one led to the progressive smouldering ignition of the corrugated cardboard boxes, their contents or shrink wrap. It is therefore most unlikely in my view that if it was a carelessly discarded cigarette it was carelessly discarded on to cardboard or the shrink wrap which enveloped most of the pallets.
78. Thus, if the fire was caused by a carelessly discarded cigarette, the probability must be that it was discarded or placed into the gap in the pallet (the source of the fire) into which the prongs of the fork lift truck had gone to lift the pallet and load it into place at eye level on the rack.
79. It must also follow from the expert evidence that fire could only be generated if within that area of the offending pallet there was other material, such as paper, upon which the cigarette would lie leading to smouldering of the paper and of sufficient paper and other material to cause a flame to erupt. I will return later in this judgment to the probability of there being such conditions within the pallet within the gaps created by the prongs of fork lift trucks.
80. There were several publications referred to by the experts. In Babrauskas-Ignition Handbook, the author states as follows at page 718:

"The effectiveness of cigarettes as ignition sources for wastepaper baskets has been examined. For waste baskets filled with papers, snack wrappers, fast-food bags and polystyrene foam coffee cups, ignitions were not observed. Oily paper towels turned out to be ignitable, but out of a total of 300 tests of dropping cigarettes into waste baskets, flaming occurred in only 5 instances; the times to flaming raged from 14 to 18 minutes. A German Study provided more comprehensive results (Table 22), but the results are based on only 12-15 trials. In fact, much longer ignition times have been observed in real fires. Figure 20 shows documentation of a fire that occurred due to cigarette disposal in a rubbish container. The time between the last human activity at the place of origin and the eruption of flaming was 192 minutes …"

The experts explained that Figure 20 showed a real fire that happened to have been captured on CCTV. Table 22 (the German Study based on only 12-15 trials) indicated that the times for ignition for corrugated cardboard and paper were, by way of minimum, 12 and 8 minutes respectively and maximum 50 and 40 minutes respectively.

81. R Holleyhead (a former partner in Mr Bourdillon's firm) produced a well- respected article in Science and Justice Volume 39 No 2 (1999) April – June. The paper discussed the mechanism of ignition of solid materials by cigarettes, the smouldering process and the transition of the smoulder to the flaming combustion. He said this at page 84:

"Attempts to light fires with an assortment of popular brands of cigarettes have been made…Wastepaper baskets with whatever rubbish was in them such as office, kitchen and laboratory waste were collected and more than 300 tests were conducted. On the 132nd test ignition was achieved in a popcorn sack. In previous tests there had been smouldering but no transition to flaming was observed. The tests from 133 onwards were extended to rubbish from fast food restaurants. The perfect combination was found to be dried paper hand towels which had been used the previous day to wipe the hands impregnated with oils associated with gas piping [this was a reference to the tests referred to in the Babrauskas quotation above] …

Paper tissues are often items discarded in waste bins and tests have been carried out on these at the author's laboratory. Lighted cigarettes are placed in various configurations of large substantial paper tissues, which included placement on top of a pile of sheets or buried inside several crumpled sheets. The cigarettes charred the paper on every occasion but in still air no flaming ignition occurred. A separate independent experiment was carried out with tissues crumpled and placed in a metal wastepaper basket located in a draught ash free basement. A flaming fire developed in a period of just short of 12 minutes on the introduction of the lighted cigarette among the mass of paper. Smoke was visible after 30 seconds and the quantity remained constant until 7 minutes into the test. After 10 minutes copious amounts of smoke were observed, followed by flame just over 90 seconds later (unpublished observations).

Other paper products studied include cardboard and toilet tissue. The latter involved the placing of whole lighted cigarettes in a variety of orientations at different locations in and on toilet rolls. In some cases paper was unravelled from the roll and crumpled up adjacent to the roll.

In the 20 tests, all the cigarettes scorched the tissue and on 4 occasions flaming combustion occurred. The time from the start of a test to flaming ignition varied between 16 and 26 minutes. Needless to say the orientations where ignition occurred was where the heat was conserved with the cigarette having been placed within the bulk of the material. One of the tests involved the placing of the cigarette inside the inner cardboard tube of the toilet roll. Flaming combustion occurred after a total of 37.5 minutes. A cigarette in this position is almost certain to start a smoulder (a perfect substrate) and the transition to flaming some time later is also very likely to occur, owing to heat conservation. The second series of tests involved the placement of lighted cigarettes on the upper surface of packs of toilet rolls. 28 cigarettes were placed on the rolls and 3 flaming ignitions were observed after 92 minutes, 98 minutes and 102 minutes [unpublished observations].

No flaming ignition was observed when lighted cigarettes are placed on flat, horizontally orientated cardboard."

82. In effect Mr Bourdillon had to accept that, for his theory that there was a carelessly discarded cigarette and that this was the cause of the fire occurring within 20 to 40 minutes, there had to exist conditions similar to those found in wastepaper basket type experiments which would lead through smouldering to flame within the 20 to 40 minute period. The problem, as I see it, with that approach is that it is dependent upon the exact conditions existing within the holes created by the fork of the fork lift truck to enable that to take place. It is equally clear that, if other conditions exist, for instance something similar to the presence of toilet rolls, times can be much longer than 20 to 40 minutes. The disadvantage with Mr Bourdillon's theory and timing is that most experiments in wastepaper basket environments do not lead to a fire at all.
83. That leads me on to consider whether or not it is likely that the conditions existed within the pallet to enable a fire to break out within 20 to 40 minutes after the Agency workers' alleged discard of a lit cigarette. The evidence was that generally all pallets came to site in shrink-wrapped plastic which covered not only the goods on the pallet but the pallet itself. As Mrs Smith told me (and I accept), the pallets were unloaded at the outside near the northern end of the Comag Bay. The fork lift trucks would take the pallets direct from the lorries and put them on to the requisite rack. Thus forks of the fork lift truck would pierce the plastic of the shrink wrap on the lorry. There would thus be only two ways in which or occasions on which other material could find its way into the gap within the pallet created by the pierced shrink wrap. Either there could be paper or other material on the end of the fork as it entered the shrink wrap (thus pushing such material into the gap) or once it was in place on the rack materials of one sort or another were introduced into the gap. In that latter case, material could have been introduced either accidentally, as wind or drafts blew paper or other fragments into the gap or if people stuffed various pieces of paper such as tissues and the like into the gap.
84. I consider that it is most unlikely that there was a significant amount of material introduced into the gaps created by the fork lift trucks. Whilst it is always possible, I would suspect strongly that most fork lift truck drivers would want to ensure a quick and easy entry of the forks into the shrink wrap to the wooden part of the pallet and would be unlikely to allow paper and other material blowing around to get into the gap. It is of course possible but there would have to be a statistically low chance of that having happened for the particular pallet which first caught fire; it was doubtless one of hundreds of pallets in the warehouse at the time.
85. As to the other possibilities of the introduction of flammable materials into the gap in the offending pallet, there is no suggestion that Aisle 1 was kept anything other than clean. Indeed Ms Wilcox who gave evidence said that the site was tidy. That seems to have been her experience. Whilst there was some evidence (see earlier in this judgment) that Comag was not very good about the clearing up of rubbish and discarded paper and plastic, it seems inherently unlikely to me that material would in some way have blown down Aisle 1 and found its way in any appreciable quantity into the gaps (which would have been relatively small) created by the forks of the fork lift truck on the shrink wrap of the pallet. It is of course always possible that one or more people over the weeks if not months before had used this particular pallet to deposit paper products such as tissues or old newspapers or the like. However I cannot quantify this possibility as approaching anything like a probability.
86. On balance, I consider that it is more probable than not that the fire was not caused by a discarded cigarette at all. The scientific tests that have been carried out suggest that only in a relatively limited number of occasions will there be a fire created by a discarded cigarette even in wastepaper basket type conditions. I am not satisfied on the evidence that the conditions, which would produce even the statistically limited number of occasions for a conflagration to occur in those circumstances, existed at or around the particular pallet which formed the seat of the fire. A carelessly tossed cigarette would have a low chance of finding its way into the gaps. To establish a higher possibility of a cigarette causing the fire, Fosse suggested that the smoker could well have wrapped the lit cigarette in a tissue and stuffed it in one of the gaps in the pallet. That would have been a very stupid thing to do and I seriously doubt that even an illicit smoker would have done that; they might have stubbed it out and then put it in one of the gaps but then it would not have caused the fire. There has to be major speculation to achieve a scenario which could produce a fire caused by a discarded cigarette.
87. Given these findings, it is certainly possible that a discarded cigarette might have led to a longer smoulder period because the conditions within the gaps in the pallet were somewhat less than the wastepaper basket conditions considered in the publications referred to above. It then follows that it is impossible to say how long such a fire would take to break out from the time that a lit cigarette was discarded. It will be appreciated that, unless the lit cigarette causes some other flammable material to smoulder, the cigarette will simply go out as in Mr Nimmo's 100 experiments, and there will be no fire. As Dr Graham and Mr Nimmo said in evidence (which I accept), there may well be a series of flammable materials which ignite in a smouldering sense from another smouldering item. In theory, a fire may take a very considerable time. Indeed some of the research would suggest that in practice a period of two to three hours could arise before a fire broke out (if it broke out at all).
88. Thus, I am not satisfied on any balance of probability test that it is more probable than not that the fire started as a result of a discarded cigarette discarded after 6 p.m. as opposed to one discarded before 6 p.m. or at all.
89. It was accepted by Mrs Smith that all her male employees smoked, and as the facts have been found, two of the agency workers smoked. Thus it is, at best for the Claimants, equally possible that the fire if started at all by a discarded cigarette, stemmed from a cigarette discarded either by a Fosse employee or one of the Agency workers. However, based on the expert evidence, I find that it is more probable than not that wastepaper basket type conditions did not exist within the gap in the pallet at the seat of the fire. It follows from this that on a balance of probabilities I am satisfied that the fire was not caused by a discarded cigarette discarded after 6 p.m. or at all.

The Evidence

90. I have already indicated that I found that the four Agency workers gave credible evidence. Although there were some discrepancies in their evidence, about which I have commented above, their essential evidence was that Susan did not smoke at all during her break (between 6 and 6.30) and that Laura and Charlene left the building for Laura to smoke during that break. I accept that evidence. It therefore follows that Fosse's primary case fails.
91. I should however address in some detail the main points made by the Claimants and their Counsel as to the respects in which they argue that the four agency workers in effect lied. For their final written closing submissions, they produced an Appendix 1 which was entitled "Table of Palpable Falsehoods". The word "palpable" was accepted to mean "demonstrable".
92. One common complaint against all four workers was that they appeared to have changed their story so far as timings were concerned between their discussion with Mr Pellowe on the night of the fire and the interviews on 19 December 2002. They all told Mr Pellowe that Laura went "to the toilet" at 1900 hours, whereas on 19 December, they each gave times of about 18:45 or 18:45 to 18:50 as the time that she went. There are some discrepancies in the various different people's notes of the 19 December 2002 meeting. For instance, Mr Pugh records Charlene as saying that it was "about 6.45" that "Laura went to the toilet" whilst Dr Graham records her as saying this happened at 18:45. I did not conclude from these relatively minor differences in the evidence either that any one individual was lying or that two or more of them were acting in concert. People's recollection of times is often confused. As the interviews on 19 December were more structured, I am not surprised that each of the ladies recalled or concluded that Laura did go to the lavatory at somewhere between 6.45 and 6.50. That is not inconsistent with what happened thereafter. If Laura went to the lavatory at, say, 6.50, it would take a minute or slightly more to get there; she said she was in the lavatory for "a few minutes". When she came out she observed the fire and then went back to the others which again might have taken another minute. By this time the time would be 6.56 to 6.58. Initially, the others appear to have thought that Laura was joking when she said that there was a fire. It is not at all unlikely therefore that it was about 7 p.m. that they came down and saw the fire. Whilst the others left the building Michelle went back for her mobile phone. It is therefore likely that they finally vacated the building at about the time that the first alarm was activated at 19.03 approximately. That is then consistent with all the events which followed thereafter. Thus, I apprehend, on a balance of probabilities, that the second lot of times which they gave were the same (18:45 to 18:50) broadly because that is more likely to have been right than the 1900 hours time which they gave in circumstances which would have been strange to them on the night of the fire.
93. Again, the Agency workers were of relatively short stature, with Michelle, Laura and Charlene being about 5 foot or 5 foot 1 inches. It is suggested that they lied in court by suggesting that the fire was above them: Michelle said it was 7 to 8 feet above ground level whilst Laura said that she would not have been able to reach up. There was evidence which I accept that the height of the pallet above ground was 5 foot to 5 foot 6 inches. I bear in mind however that the fire was in all probability observed by Michelle from some distance away: she stood at the Comag Bay end of Aisle 1 and she observed fire and flames some way up the pallet. It would probably have seemed to her to be some 7 to 8 feet above her albeit that she was not beside the pallet whilst it was on fire. A similar comment would apply to Laura's evidence in this context.
94. I will now deal in relatively brief terms with all the other "palpable falsehoods" put forward by Fosse in this case. First, in relation to Michelle:

(a) Michelle said in evidence that Laura was on her mobile phone to the Fire Brigade in the warehouse. Michelle had said nothing about this in her witness statement but there are notes of her interview on 19 December that Laura called after they had all left the building. I do not consider that this was a lie or deliberate untruth by Michelle. It was well over five years since the fire. This is just the kind of detail which someone like Michelle might get out of order.

(b) In evidence, Michelle said that when she first saw the fire it looked "more of a sparky sort of flames, you know, little sparks coming out of the pallet". In her witness statement (Para 28) she said "the flames looked normal in appearance ... they were catching very quickly … there did not appear to be much smoke". She is recorded as having told Mr. Pellowe that there was "nothing unusual about flames – no sparks". During the interview on 19 December 2002 she told of there being small fires on each side of the pallet with 2 feet flames. I am not convinced that there is an enormous discrepancy. In evidence she confirmed that there were flames but she gave evidence of what sort of flames they were in her recollection. In any event the fire took place a long time before she gave evidence.

(c) It was Michelle's oral evidence that when all four went down to see the fire "we were like, kind of, like all panicking amongst us". This is said to be inconsistent with Paragraphs 28 and 29 of her witness statement which indicated that she went back to retrieve her mobile. That is not a real inconsistency. I am not at all surprised that the panicking started when they saw the fire. Michelle running back to get her mobile is not inconsistent with somewhat irrational behaviour when faced with a dangerous situation or perhaps not wholly appreciating how quickly a fire might accelerate. It is, possibly symptomatic of her not thinking straight at the time, which is not surprising in the circumstances.

(d) Michelle in evidence said once out of the building they all ran over to the security guard in his guard box but "he had earphones on and we were banging the door, telling him that there was a fire in the factory"; they had to bang on the door because he had earphones and was sitting in his box with his feet up. She said he eventually caught on. That does not contradict her witness statement as such because she said there at Paragraph 31 that "…we went for the security guard and told him that the warehouse was on fire. At first I thought the security guard did not believe us …" It is not contradicted by the interviews given on 19 December 2002. It is said to be contradicted by the statement of Mr Kariuki but he was not available to give oral evidence. Susan gave oral evidence that the ladies went over to see Mr Kariuki, that he came out of the door and that he had seen them. That suggested that in Susan's recollection there was no banging on the door. Laura gave evidence that they tapped on the glass on the Portakabin. This issue is a relatively minor detail. I certainly considered that what Michelle said had the ring of truth about it. It was corroborated by what she told Mr Lynch on the 19^th December 2002: he also recorded Michelle as saying that she "banged on security guard box". I will return later to consider the effect of the absence of Mr Kariuki overall.

(e) In evidence, Michelle accepted that she occasionally propped Door 3A open but on the night of the fire it was not propped open and that she let Charlene and Laura back in after they had been for their cigarette after "they just banged on the door". Dr Graham records her saying in the interview on 19 December 2002:

"Normal exit door is used to exit for smoking; propped open"

Fire Officer Lynch who was also present recorded as follows:

"When go out for smoke the door is propped open whilst outside".

It is said to be inconsistent because, although Laura and Charlene could not remember in evidence whether the door was or was not propped on the night in question, their witness statements indicated that they came back into the building and joined Michelle and Susan in the canteen; inferentially, that, it is said, suggested that Michelle or Susan did not need to let them in so the door must have been propped open. This is stretching matters somewhat. Michelle does not appear to have been asked in interview in December 2002 what happened that night in terms of whether the door was propped or not. What she was asked about was what happens normally. I did not get any impression that Michelle was lying when she gave her evidence about the banging on the door and the letting in of Charlene and Laura. It would be more logical on a cold night for the door to be shut as opposed to propped open and, as Door 3A was fairly close to the canteen, any banging on the door would have been heard by those in the canteen. It is equally possible that the door was left open. Some evidence later in the trial suggested that the door did not need to be propped open at all because it would stay open when opened. This is again a somewhat minor detail of the evidence when, given the lapse of time, it is hardly surprising that errors in recollection might occur.

(f) Michelle gave oral evidence that there were windows all round the canteen so that one could see into the "little corridor that led to the outside door". It seems clear that there were no windows on to the corridor albeit that there windows in the canteen out on to the warehouse. When photographs were shown to her that suggested that this could not be right she wholly accepted that she was mistaken. This is just the type of thing which one could get wrong five and a half years after the event. She said however that the door on to the corridor was open so that she could see in which direction Laura and Charlene turned to go to smoke.

(g) In evidence under cross-examination she said that she recalled that there were flames coming out of the pallet at either side. She had told those interviewing her on 19 December 2002 that there were flames either side of the pallet. Again there has been a substantial lapse of time since she gave evidence. She gave the answer under cross-examination when it was suggested to her that, if she had not gone down the aisle, she could not necessarily have seen the fire. Her answer that there was fire just on one side is consistent with someone being asked in effect to carry out an after the event reconstruction. She was, I felt, simply trying to rationalise what she could remember that she would have seen from the top of the aisle.

All in all, I found Michelle a convincing and honest witness.

95. As to the so-called "palpable falsehoods" told by Laura:

(a) Laura said in evidence that she "got told there were cameras, so I didn't want to get into trouble. I always lit my cigarette at the bus shelter". There were in fact cameras on the outside, albeit that, unbeknownst to the four ladies, the cameras were not working. At least one camera was very obvious and would have covered the car park area. It is suggested that her evidence about this conflicts with a reason given for not smoking in the car park which was that there was a chemical factory next door. In interview, and indeed Dr Graham recorded that Laura and Charlene went out for a cigarette:

"went to the bus stop in the main car park. + chemical factory next door – so can't just smoke generally".

The evidence does not conflict. I presume that she was simply telling the interviewers that one was not expected to smoke just anywhere in the car park but only in the bus shelter; one of the reasons was the presence of a chemical factory nearby.

(b) In oral evidence, Laura recalled what happened when going to and from the lavatory:

"On the way to the toilet, I'm not sure if it – I can't remember if it was in my statement that I walked to the toilet and it were like – I don't know how to describe, I just like sniffed and as I was like, you know, what's that and I just took no notice of it, and then I went back – I went to the toilet, erm, come out of the toilet and then it was like I can say – it weren't like a burning, like how a fire is. It was like a smouldering smell, and then I was looking for smoke and I think it was out of my glasses I just caught like a light and I looked and then I seen. It weren't like a flame …"

She told Officer Pellowe:

"1900 Went to toilet. Smelled smoke – saw fire in pallets (first shelf up)."

Laura is recorded variously by different people as saying this:

"6.45 Went to the toilet. Go through the canteen smelled burning as went to the toilet but took no notice" (Pugh)

"About 1845 went to toilet – Walked through the canteen – went to the toilet – As I came out of door into warehouse and could smell burning …" (Bourdillon)

At 18.45 went to toilet, through canteen to toilet. As came out of door from canteen into warehouse to go to the toilet smelled burning. Carried on to toilet, then came back out of toilet …" (Lynch)

"… walked to canteen, through, out and to toilet – as she came out, she smelled burning. Took no notice of the smell, went into Ladies …" (Dr Graham).

Assuming that the version of events she gave in interview on 19 December 2002 was broadly correct, I would have assumed that, if she had really thought there was a fire before she went into the lavatory, she would have raised the alarm immediately. It is consistent with her oral evidence that she thought possibly there was some smoke, thought better of it and went into the lavatory. Given that she saw the fire afterwards, some after the event recollection is not surprising of the "I thought I might have smelt something like smoke before going to the lavatory but thought nothing of it" variety. I did not find her oral evidence inconsistent or inconsistently given.

(c) Laura, like Michelle, indicated in oral evidence that the ladies had to attract the attention of the guard. She said "I think all of us were shouting. … I don't know if he could hear us or – and I'm sure we like tapped on the glass on the Portakabin. I don't know if he didn't understand us or he couldn't understand English …" As in the case of Michelle this is said to conflict with the evidence of Mr. Kariuki and Susan. For the same reasons as indicated above, I do not consider that there is a material conflict. I am not at all surprised that the ladies had to rouse Mr. Kariuki from his guard box.

(d) Much is made by the Claimants's Counsel as to Laura's evidence that she remembered working in the video refinishing unit during the evening and hearing noises in the warehouse which seemed to be coming from somewhere in the storage racking. She remembered Michelle using her mobile phone to call Ms Sweeney of Phoenix who attended at the premises with a male relative of hers; despite a search of the premises nothing was found. All four women gave similar evidence. However Laura also explained that she had only been working in the video finishing unit for about two weeks before the fire. Charlene gave evidence that another worker called Ann was present when the noises were heard but Laura did not ever work with Anne. Thus, the Claimants assert, Laura must be lying because she could not have been there when the noises were heard. I am satisfied that Laura was not lying. Notes of the interviews on 19 December 2002 record:

"Two weeks ago it was thought that there was someone in the building (no one found)." (p 807)

"About two weeks ago, I heard noises which I believed to be someone in the building – no one found inside". (Mr Lynch).

Thus those who attended the interviews mentioned noises some two weeks before. That is entirely consistent with the noises issue arising during Laura's time with the other three ladies in the video unit. Laura readily accepted that she may have got the date (November) wrong in her evidence. It is likely that Charlene's evidence about Ann was simply mistaken. The very fact that all four ladies broadly said that there was a problem experienced with noise in the warehouse leads me to the view that that recollection is true. That is corroborated by what one or more of them clearly told the interviewers in December 2002.

(e) It was said that Laura lied in giving evidence about the signing of her statement. Laura's witness statement was signed by her but dated by somebody else 5 April 2008. There appears to have been some delay in the service of her statement. Laura said in evidence that she did not receive it initially because she had had trouble with her post. Comag's solicitors wrote to Fosse's solicitors on 20 March 2008:

"Turning to the issue regarding the Witness Statement of Laura …, as you know Laura confirmed that she has signed dated and returned the same to us. We understand that there was recently a major bomb scare at the main sorting office in Coventry which may account for the delay in receipt. However, we have arranged for an additional copy of Laura's statement to be delivered to her for her signature and immediate return to us."

Laura indicated that she only signed the statement once. She said she never received the first copy but when she received the second copy she signed it. It seems to me that all that has happened here is that there has been some mix up. Comag's solicitors may be mistaken about what Laura may or may not have done with the first copy. Suffice it to say, it does appear that Laura signed a statement shortly after this letter and that was the one that was served. It is impossible to conclude that Laura was lying about this. There would be little or no point in lying about this.

(f) In evidence Laura explained that the door, 3A, had to be propped open. In interview on 19 December 2002, Dr Graham records:

"Security door left open – has a ratchet détente to hold fully open".

Dr. Graham accepted that she did not use the expression "ratchet détente". The discrepancy is said to be that if there was a ratchet détente the door would not have to be propped open because it would simply stay open. Photographs after the fire show the door simply open without any propping. I find no particular discrepancy in this. Even if the door would stay open, one might want to prop it to prevent it closing in any draught. There are different types of door mechanism such that one person might describe it as being propped open if there is some mechanical catch or ratchet which holds it open. The witnesses were not really asked as to what they meant by the expression "prop open" and how that would be achieved. I do not see that this was a material discrepancy.

(g) Much was made by the Claimants of Laura's description to Dr Graham in interview in December 2002 that "scaffolding was present in front of the racks". Dr Graham sketched something in his notes which indicates the scaffold structure a distance away but perpendicular to Aisle 1 and Racks 1 and 2. She could not remember in evidence whether scaffolding was "in the way or not". I see absolutely no discrepancy. The sketch which Dr Graham showed does not show it as such blocking the aisle. It shows a gap between the scaffolding and the end of the aisle: it would have been possible (bearing in mind that the sketch is not in any way supposed to be to scale but is simply a rough sketch) for people to come between the scaffolding and the head of the aisle to observe what was going on down the aisle. This is and was a bad point taken by the Claimants.

(h) Laura gave evidence that she believed that there was a yellow railing across the area outside the toilets, that is delineating the boundary of the way through to the toilet and the body of the warehouse in which are Racks 1 and 2. She said she was sure there was such a railing but she accepted that the photographs showed otherwise. She was "obviously wrong". That showed candour in my view and not deception. It is the case that at least one witness for Fosse gave oral evidence that he thought there was a barrier in that area. If Laura's recollection is wrong, she was not the only one.

(i) Laura said that she had been told that "there were cameras all throughout the warehouse". For that reason: "that's why I stuck to every rule, because I didn't want to lose my job". Whilst she had not said in her statement or before that there were cameras inside the warehouse, I am not surprised that she may have been confused about this. It is improbable that there were cameras within the warehouse but it is also possible that she was told that there were cameras within the warehouse. I am not satisfied that she was deliberately lying.

(j) Laura gave evidence that when she first saw the fire after coming out of the lavatory she saw it in effect through the racking in Rack 1. That is consistent with what she said in interview on 19 December 2002. In interview, Dr Graham produced a sketch which showed Rack No 1 extending to a point just in front of the middle of the ladies' lavatory. This was in fact wrong because the rack stopped some 3 to 5 yards short thereof, but that discrepancy may have been due to Dr Graham misunderstanding what he had been told. It is said by the Claimants that she must be lying because in all probability she could not have seen through Rack 1: there would have been so much on Rack 1 that the view would have been blocked. I do not accept that. It is inevitable that pallets such as were in place here would not have been packed up horizontally right up against each other, for no reason other than it would have been extremely difficult to get them in and out of the racks if they were packed that tightly. There must have been gaps and I consider that it is wholly feasible for Laura to have seen an incipient fire in a darkened warehouse breaking out in a gap within Rack 1.

(k) It is also suggested that Laura tried to exculpate herself by suggesting to Mr Pellowe that the fire was "sparking like electrical" and in evidence that the cause was electrical. This point has no weight: she was just trying to explain what she saw: the fire could have looked as if it was sparking to some extent.

I found Laura to be in general terms an honest and truthful witness.

96. In relation to Charlene apart from any criticism that there was some collusion in relation to the changes of time from 7 p.m. to 6.45 p.m. to 6.50 p.m. (see above), the only "palpable falsehood" is said to be that when she and Laura got back from the bus shelter where Laura had had her cigarette that when she went back in she went "back to the canteen to Michelle and Susan". It is said that if Susan was in the lavatory as she said she was only Michelle would have been in the canteen. This is a very minor detail which I am not surprised that a witness may or may not have got right. By all accounts Susan went to the lavatory at about 6.25. She may well have been in the canteen when Charlene and Laura came back, having or having not been to the toilet. It is equally possible that she was not in the canteen when Laura and Charlene came back but she must have been back in the canteen within a very short time after they get back. It is equally possible that Susan was wrong in saying that in effect she was in the lavatory when Laura and Charlene came back. I found Charlene generally to be a reasonably impressive witness.
97. As for Susan, the main "palpable falsehood" is that she told Mr Pellowe that she had two cigarettes left for a break at 8 p.m. By 19 December 2002 she was saying that she was saving "her last cigarette" in effect for later. It was that latter evidence which she repeated in evidence. There is an inconsistency between what she told Mr Pellowe and what she said thereafter and in evidence. This is the major inconsistency which was put to her and it could only be relevant if it was Susan who, instead of going to the lavatory, went into the warehouse had a cigarette and carelessly discarded it in Aisle 1. I found her to be a believable witness. It is equally possible that she simply got it wrong when she told Mr Pellowe that she had two cigarettes left for the 8 p.m. break or even that Mr Pellowe misunderstood what she said.
98. Most and probably all of the discrepancies asserted by the Claimants in the evidence of the Agency workers were relatively minor and, if discrepancies at all, are not inconsistent with the lapse of time and slightly different recollections by four disparate witnesses of what was an event which happened over a relatively short period of time. I would have expected there to have been greater substantial and irreconcilable differences which could only be explained by dishonesty if Fosse wanted to establish perjury and in effect a conspiracy between two or more of the four workers. Most of the factors relied upon would only give rise to suspicion if one started from a belief that one or more of these ladies were involved in collusive perjury.

The Cause of the Fire

99. I am not in a position to form a view as to what probably was the cause of the fire. However, I am able on the basis of what primarily was the evidence of Michelle, Laura, Charlene and Susan coupled with some of the other evidence to form the view that it was probably not caused by a carelessly discarded cigarette from one of them.
100. That leaves one of four other possible causes. I discount immediately the possibility that it was caused by an electrical fault or electrical equipment. The only ground for this is that the contractors installing the heating equipment had left some rechargeable piece of electrical equipment at or close to what turned out to be the seat of the fire. Both experts accepted in principle that, if such equipment connected to the mains had been left there since Friday, it could have overheated to such an extent that a fire could or would break out. However, it is inherently unlikely that those contractors would have left some electrically rechargeable tool at that location as most tradesmen take their tools away with them at the end of the working day. If, by accident or even deliberately, one or other of them left some piece of electrical equipment recharging I can see no possible reason why it should have been placed in the middle of the first aisle on Rack No 2 at the first level which would have been a point at some distance from any electrical point to which the cable which carried the charge for the recharging would stretch. There was little room on the racking to place any equipment to be charged: it would certainly have been precariously located on some part of the relatively narrow metal racking. Even Dr Graham accepted that this was at best the least likely of all the possible explanations. It is inherently improbable.
101. One then turns to the possibility that an intruder with some malicious intent (including an intent to commit arson) entered the building before or after the Fosse employees left just before 6 p.m. It is unlikely that any intruder secured entry after 6 p.m. At best, there was a very short period of time in which the door (Door 3A) would have been open whilst Laura and Charlene went to the bus shelter for Laura to smoke. The area of the car park was lit and there was a guard on duty after 6 p.m. who had been walking around before about 6.15 p.m. An intruder would have had to have been hiding on the off chance that the door would be opened and left unwatched for a period of time. It would be most unlikely that an intruder would have been able to get in, let alone been around on the off chance of getting in, whilst Laura and Charlene were out of the building. It remains unclear to me (at least on a balance of probabilities) whether Door 3A was in fact shut by Charlene and Laura or left propped open.
102. Thus, if it was an intruder which caused the fire, it is more likely (although not probable) that it was an intruder who secured access prior to 6 p.m. Given that before 6 p.m. there were only the four Agency workers working in the video unit and no more than a handful of Fosse employees in this large warehouse, an intruder could have secured entry through one of several doors and hidden in the warehouse until the Fosse employees left. It is not unlikely however that an arsonist, who thought that the Agency Workers had gone back to the video room, might have felt it necessary to start the fire when one of the workers returned to go to the lavatory; whilst she was in the lavatory, he or she might well have felt that he or she had better get on with it. The Fire Brigade found two doors open which would have included Door 3A. There is no reliable evidence as to which the second door was. If the other door was one of the warehouse doors, that could well have been the possible arsonist's escape route. With Mr Kariuki not available to give evidence and Michelle's evidence that Mr Kariuki was in effect not concentrating on looking out of his guard box, the fact that an intruder was not seen by Mr Kariuki does not work against the arsonist theory. Mr Seymour of Fosse accepted that an intruder could have come in and hidden before the Fosse people left.
103. Laura's evidence about possibly smelling smoke before going in to the ladies' lavatory only raises the possibility that there was smoke already in the air and renders it improbable (as the experts agreed) in those circumstances that an arsonist started the fire. I am not satisfied that Laura did actually smell smoke before going into the lavatory or that there necessarily was smoke the smell of which would have reached her at that stage and, consequently, it is still possible and not inherently improbable that an arsonist entering before 6 p.m. started the fire.
104. I consider that it is unlikely that, if the fire was started deliberately, it was started with a firework. Unless it was children, I can not see that a firework would be an obvious choice for an arsonist. Although several witnesses said that it could be a firework (with Laura referring to a possible Catherine Wheel and Susan saying that it looked somewhat like a sparkler), I consider that they were just trying to rationalise what they had seen.
105. I am not materially assisted by evidence that in the months leading up to the fire a number of possibly suspicious things had happened. There is no doubt that a "gypsy" or "gypsies" had been found in the warehouse some weeks before the fire; however there was some tangential evidence that the "gypsies" in a local encampment had moved on well before the fire. The incident of the noises in the warehouse some two weeks before the fire does not assist: notwithstanding a search of the warehouse on the evening in question no one was found in the warehouse. Finally, there was some evidence that a front door was found open one evening with no obvious explanation as to why it was left open. I accept Laura's evidence about this but it does not establish that there was a pattern of breaking in. There was some evidence that some months before Comag had experienced a problem of pilfering but that would not necessarily explain why a pilferer would want to burn the warehouse down in December. If someone with malicious intent wanted to burn the warehouse down, I have no doubt that he or she would have been able to secure access if so desired.
106. That leaves the possibility that the fire was caused by a cigarette carelessly discarded by a Fosse employee prior to 6 p.m. For reasons explained above, I consider that this is distinctly possible but I am not in a position to say that it was the probable cause of the fire. Still less am I able to "point the finger" at any individual.
107. Reviewing all these possible causes, whilst I can exclude on a balance of probabilities the fire being started by a cigarette discarded by any of the ladies, the electrical cause and the intruder after 6 p.m., I am unable to distinguish in terms of likelihood between an intruder who gained access before 6 p.m. causing the fire and a carelessly discarded cigarette from a non-Comag employee before 6 p.m. Unlike Sherlock Holmes, in those cases in which there is an "either or" pair of causes, I am drawn to the conclusion that, whilst the cause of the fire was not attributable to the Agency workers, it is not possible on the balance of probabilities to determine that the fire was caused by either of the two remaining feasible causes. The outcome or explanation is anything but "elementary". However, I am in no doubt that Fosse has failed to prove its case on a balance of probabilities.

Miscellaneous matters on causation and negligence

108. A number of matters were raised during the trial which I should address:

(a) I would accept that, if one of the four Agency workers had discarded a lit cigarette within the warehouse, that would have involved a breach of a tortious duty by that worker: there was a clear no smoking policy and it was obvious that there were flammable goods around.

(b) If Laura or Charlene left the security door (Door 3A) open whilst they went to the bus shelter, that was also a breach of tortious duty by them: it was clear that the door was to be kept shut except for going in and out and there was a method of getting back in, by attracting the attention of those inside to let them back in.

(c) There was no failure of supervision with regard to smoking, if there had been any on the premises. It had not been agreed that there would be Comag supervision of the Agency workers after 6 p.m. There was no need for there to be such supervision as Michelle was the supervisor of the other three and in total only four workers were then involved. In any event, Michelle or any other supervisor could not reasonably be expected to escort the likes of Laura off the premises to be assured that she was smoking outside or the likes of Susan to the lavatory to see that she did not go into the warehouse to smoke. There was no history of any of these four workers misbehaving before. I am not satisfied on a balance of probabilities that, even if Comag had provided a supervisor on the evening of the 16 December 2002, he or she or his or her presence would have prevented either a cigarette being smoked within the warehouse or Door 3A being left open whilst Laura went for a cigarette outside.

109. Various other matters and evidence were relied upon which did not take the case much further forward:

(a) It was argued that it was less cold in the warehouse so that a smoker would have been more likely to have wanted to smoke inside than out. It was cold in the warehouse however: there was no heating and the Agency workers worked with warm coats on and mittens. There was likely to be little difference between the temperature inside and outside. Smoking inside was as unappealing as smoking outside; of course smoking inside was more unappealing because it could trigger smoke alarms and lead to the dismissal of the errant worker.

(b) Mr Kariuki did not give evidence; he was proofed in late 2007 but left the country for Africa shortly thereafter supposedly to attend the African Nations Cup football competition. A draft statement was prepared based upon his proofing. I find it to be of little weight: it could not be tested by cross-examination. His departure gives rise to some suspicion albeit unquantifiable because he had worked for Meridian for over five years before and there is no obvious reason for him not to come back or stay in touch. There was at least one significant discrepancy between what he said in a written statement prepared for the 19 December 2002 interviews which he also attended and what he told the solicitors: in the former he said the Agency workers approached him shouting whilst in the later he said that they were not shouting and were relaxed. There was some (at best) confusing evidence that Mr Kariuki may well have had a master key which could have secured entry in to the warehouse. Whilst Mrs Smith gainsaid that, it remains a real possibility which could well have been cleared up if he had come to give evidence.

(c) A draft witness statement from Divisional Officer Smith from the Fire Brigade found its way into the court bundles to which I attach no importance at all. Some of it is hearsay on hearsay. There appears to have been no difficulty in calling him if his evidence was required and he could then have been cross-examined.

(d) A cigarette was found on 19 December 2002 in the video room. It was suggested that this was proof that unauthorised smoking had gone on in the warehouse. This establishes nothing. It could as was accepted by several witnesses have been brought in on the sole of someone's shoe after the fire; that was not at all unlikely because cigarette stubs were found outside the warehouse just outside the video room. The cigarette was not of the type smoked by Laura or Susan.

110. One aspect of the evidence to which I attach no importance is the suspicion which Mr Pellowe had on the night of the fire that Laura may have started the fire deliberately or at least knew more than she was saying. In an interview with Mr Bourdillon in 2004, he said that Charlene she was very nervous and very "clammed up". As for Laura, he said that she was:

"was almost cocky with her description of the incident, which, as I say, gut feeling again, it just rang a few warning bells. In that her story was very precise, about her actions and what she did, and she made a big play about having got everybody out of it. She'd sort of saved the situation, got everybody out…And from experience, sometimes that suggests that this person is covering up an action that they've undertaken, which is probably behind the cause of it. But that is supposition on my part. It's purely speculation".

He went on to accept that adrenaline may have been an explanation for her behaviour. I consider that it would be wholly wrong to attach any weight to what Mr Pellowe felt in this regard. Laura had actually raised the alarm and helped the others to evacuate the building and, if she was somewhat "cocky", it might well have been justified or at least explicable. More importantly, she had just been through what must have been an unusual and frightening experience and she was brought back to the warehouse some hours after the fire to be interviewed on her own as a seventeen year old in the back of a car by someone she had never met before which must have been very unsettling. It is most unsurprising that she may have behaved in a way which was not exactly calm. As Mr Pellowe said beforehand and effectively accepted in evidence his suspicion was pure speculation and I treat it as such.

Contributory Negligence

111. It is unnecessary for me to decide the issues relating to contributory negligence. However, I summarise the reasons for my conclusion that there was no contributory negligence, albeit that a suitable sprinkler system would have seriously limited the spread of fire:

(a) It was asserted that much of the fire damage was attributable to the absence both of a suitable system of sprinklers and of compartmentation at the warehouse.

(b) There was no statutory requirement (such as Building Regulations) for a building of this size built in 1976 to have sprinklers or compartmentation.

(c) None of the literature at the time of the fire makes it clear that for an existing building of this age these are required.

(d) There are numerous warehouses which do not have such systems in. Put another way, there seems to be a respectable body of opinion amongst warehouse owners that they are not necessary.

(e) Fosse was not required by the Fire Brigade who checked the building over some time before the fire or their insurers to install sprinklers or compartmentation. Fosse had taken some steps to provide an expensive new fire alarm system and other safety measures.

(f) The evidence on compartmentation was not satisfactory. There was no attempt to explain what compartmentation was needed as a reasonable minimum or the extent to which it would have limited fire damage. If it was extensively compartmentalised, the damage would have been much less; if it was limited to means of escape for people, little of the building would have been saved

Vicarious Liability

112. In view of my findings on causation, it is again unnecessary for me to address in any detail what my findings would have been on vicarious responsibility. I summarise what they are in the context that there are always two sub-issues:

(a) Whether the person who commits the tort in question was or is to be treated as if he or she was employed by the defendant;

(b) If so, whether he or she acted in committing the tort in the course of his or her employment.

113. As to the first sub-issue and having regard to the authorities relied upon by the parties, I formed the view that the four Agency workers are to be treated as being employees of Comag. They were employed by an employment agency which provided them to work for Comag. Although not paid by Comag, they were working under the direction of, to the rules laid down by and for the benefit of Comag. There was evidence from Fosse and indeed from Michelle which I accept that it was Comag supervisors like Mr Slade who told the four what to do and how to do it; it must have been Comag supervisors who told them what the rules were also and when and where breaks were to be taken. From 2 p.m. to about 6 p.m. the four were working under the supervision of Comag. The fact that they were trusted to work without Comag supervision after 6 p.m. does not detract from that. There is nothing inherently unusual about the possibility that there might be two entities that have dual vicarious responsibility for one employee.
114. As to the second sub-issue, the test in law is summarised in Gravil v Carroll and another [2008] EWCA Civ 689 at Paragraph 21 in the Master of the Rolls judgment:

"As we see it, the authorities show that the essential question is that posed in Lister and adopted in Mattis, namely whether the tort is so closely connected with the employment, that is with what was authorised or expected of the employee, that it would be fair and just to hold the employer vicariously responsible. In answering that question the court must take account of all the circumstances of the case, as Lord Steyn put it, looking at the matter in the round. The authorities show that it will ordinarily be fair and just to hold the employer liable where the wrongful conduct may fairly and properly be regarded as done while acting in the ordinary course of the employee's employment (per Lord Nicholls). This is because an employer ought to be liable for a tort which can fairly be regarded as a reasonably incidental risk to the type of business being carried on (per Lord Steyn [in Lister v Hesley Hall Ltd [2001] UKHL 22]).

115. In my judgment, if the fire had been caused by a carelessly discarded cigarette, such a tort was not so closely connected with the employment of the Agency worker in question as to make it fair and just to hold Comag liable. There was a clearly laid down policy well understood by the Agency workers whereby there was to be no smoking in the warehouse and no entry was permitted within the part of the warehouse where the fire started; they knew that if they did offend against these requirements they would probably be dismissed. They would be acting outside the course of their employment if they did so offend. There was no history of these workers smoking on the premises; it is not as if Comag had condoned or shown a "blind eye" to the smoking on the premises.
116. Conversely, if the fire had been caused by an intruder negligently being permitted to enter whilst Door 3A had been left open (if it was) when Laura went for her cigarette, that would have been sufficiently connected with the employment of the Agency worker in question as to make it fair and just to hold Comag liable. This is because the workers were entitled to a break and to smoke if they went outside. No procedure was laid down by Comag as to what should be done with Door 3A when this happened. It was wholly foreseeable that that door would be left or propped open.

Conclusion

117. It follows from the above that the Claimants' claims fail both on liability and causation. The Claim is dismissed and there will be judgment in favour of the First Defendant.