England and Wales High Court (Queen's Bench Division) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION

		Royal Courts of Justice The Strand London WC2A 2LL
		17 September 2012

This is the official judgment of the Court, and I direct that this written judgment may be used for all purposes as the text of the judgment, and that no further transcript of the judgment need be made.

John Randall QC, Deputy Judge

THE DEPUTY JUDGE:

1. This is a damages claim by a home owner against his insurers for (in effect) indemnity in respect of fire damage. The defendant insurers deny liability on the grounds that the fire, they allege, was deliberately started by a member of the home owner's family (his partner), and that the resultant damage therefore falls within an exception to the cover under the policy. The Claimant disputes this on the facts, and suggests, subject always to the burden of proof resting on the insurers, that the more likely cause of this fire was accident. The matter has come on for trial, pursuant to a direction of Master Leslie, as to liability only.

The property and its occupants

2. 23 Margaret Street, Brighton is a four storey, three-bedroomed terraced house, located in a residential part of central Brighton described by one of the witnesses as 'the gay village'. The Claimant, Mr Jonathan Smyth, bought it in March 2007, and took out a home insurance policy with the Defendant, through Halifax General Insurance Services Ltd.
3. The ground floor, a few steps above street level, comprises a front sitting room, with an informally created hallway to the side leading to a staircase to the upper floors, and past that to a room (known by the occupants as 'no man's land') which was unused save for the presence of a computer and it leading to the open style staircase to the basement.
4. The basement comprises a bedroom, accessed through a concertina-style 4 panel folding door, with a window onto the street, a bathroom, and a kitchen area, with a door from the kitchen area affording access to a small back courtyard. The seat of the fire was in the basement bedroom, more specifically at or very close to the foot of the bed.
5. The first floor comprises a bedroom, together with a toilet adjacent to the staircase, and the second floor another bedroom, with an en-suite shower room.
6. From purchase until after the fire, which occurred on Tuesday 22 January 2008, the Claimant occupied the second floor bedroom, together with his partner, a Mr Tony Howe. Within a few months of purchasing the property, the Claimant had informally rented out the other bedrooms: that on the first floor to a Mr Grant Eydmann, and that in the basement to a Ms Rebecca Alexander.

The terms of the policy and the issues for decision

7. The cover afforded by the policy was subject to exceptions in the usual way. So far as both buildings and contents (as defined) were concerned, the only material exception to simple fire cover was the amount of the excess [C/29/225 & 228], whereas cover in respect of vandalism was subject (so far as is material) to an exception in respect of loss or damage "caused by any of your family" [C/29/226 & 228]. The policy definition of "your family" included the policy holder and "… your spouse, your civil partner … or the person (whether or not of the same sex) with whom you are permanently cohabiting in a marriage-like relationship …" [C/29/220].
8. The Claimant realistically accepts that Mr Tony Howe falls within the policy definition of his family, and further that "if it is proved by the Insurer that Mr Howe deliberately started the fire, the 'vandalism' exclusion would apply" (opening skeleton paragraph 44; see also closing skeleton paragraph 1).
9. Having heard the evidence, I am quite satisfied, and I find:
10. 1. that there is no evidential support for any cause of the fire which does not involve human agency (in particular, the experts are agreed that there is nothing to suggest that some electrical fault might have been the cause);
11. 2. that there was no sign of forced entry to the house, and no realistic way in which, in the minutes after Ms Alexander left, an intruder entering the house through the front door can have gained access to the basement bedroom without being seen by Mr Howe, who throughout this period was either in the kitchen area or in the sitting room, through which anyone so entering the house would have to pass to get to the basement;
12. 3. that Ms Alexander and Mr Howe were the only 2 people awake in the house in the time leading up to when the fire started (Mr Eydmann having fallen asleep in his bedroom);
13. 4. that there is no realistic evidential basis for suggesting either that Ms Alexander started the fire deliberately (ultimately neither counsel has so contended; as the Claimant's case was put in closing, whilst she "could and might" have done so, "there is no cogent evidence to support [it]") or that Mr Howe started it accidentally;
14. 5. that nor is there any realistic evidential basis for suggesting that Mr Howe started the fire recklessly (as opposed to deliberately). Although Miss Nolten at one stage touched on the legal implications of this theoretical possibility, Mr Howe had no proper reason to be in Ms Alexander's room at all after she had left (prior to smelling smoke and detecting a fire which was already ablaze), let alone with a naked flame, and denied being so.
15. In these circumstances, the Defendant in turn has realistically accepted that nothing is added to its case by the other, more recherché lines of defence pleaded, such as fraudulent device and public policy.
16. Accordingly, the minimum question left for me to decide is whether Mr Howe started the fire. I now turn to the questions of whether that is all I should decide, and what is the correct approach to the evidence in such a case.

The burden and standard of proof

17. It is common ground that the burden of proof lies on the defendant insurers. This is because, absent proof of an excepted cause, the Claimant is entitled to recover: his policy was in force, his premium paid, and fire was an insured peril. Hence to avoid liability, the insurer has to prove that the damage resulted from an excepted cause (here, vandalism by Mr Howe).
18. As in any civil case, the standard of proof is on the balance of probabilities. The apparent simplicity of that statement, however, does not tell the whole story, for one of the two possible causes of the fire which has to be considered would involve the commission of a serious criminal offence by Mr Howe. Fortunately, the conundrum posed by the combination of on the one hand the familiar civil standard of proof, but on the other the principle that the more serious the allegation, the clearer the evidence required to support such a finding (see e.g. Hornal v Neuberger Products [1957] 1 QB 247 (CA)) has been addressed comparatively recently by the House of Lords in Re B (Children) (Care Proceedings: Standard of Proof) [2008] UKHL 35, [2009] AC 11. Lord Hoffmann dealt with the point in his speech at [14]-[15]:

14 Finally, I should say something about the notion of inherent probabilities. Lord Nicholls said, in the passage I have already quoted [from In re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, 586] that—

"the court will have in mind as a factor, to whatever extent is appropriate in the particular case , that the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability."

15 I wish to lay some stress upon the words [shown in bold]. Lord Nicholls was not laying down any rule of law. There is only one rule of law, namely that the occurrence of the fact in issue must be proved to have been more probable than not. Common sense, not law, requires that in deciding this question, regard should be had, to whatever extent appropriate, to inherent probabilities. If a child alleges sexual abuse by a parent, it is common sense to start with the assumption that most parents do not abuse their children. But this assumption may be swiftly dispelled by other compelling evidence of the relationship between parent and child or parent and other children. It would be absurd to suggest that the tribunal must in all cases assume that serious conduct is unlikely to have occurred. In many cases, the other evidence will show that it was all too likely. If, for example, it is clear that a child was assaulted by one or other of two people, it would make no sense to start one's reasoning by saying that assaulting children is a serious matter and therefore neither of them is likely to have done so. The fact is that one of them did and the question for the tribunal is simply whether it is more probable that one rather than the other was the perpetrator.

19. Mr Timothy Evans of counsel, appearing for the Claimant, emphasised that the burden of proof is on the defendant insurers, and that it is not about guessing. Accordingly, he submitted, if a number of theoretical possibilities are suggested, but there is no convincing support for any of them, the court should hold that the party carrying the burden of proving one of them has not discharged it: Rhesa Shipping Co. SA v Edmunds (The Popi M) [1985] 1 WLR 948 (HL), in which the decision of the trial judge (Bingham J, [1983] 2 Lloyds Rep 235) was overturned. Another, perhaps more surprising, example is afforded by Ashraf v Akram 22 January 1999 (CA), unreported (cited by Miss Nolten, and considered in Stephens v Cannon infra). In the present case too, Mr Evans submitted "it is not necessary to decide on any cause as the probable cause".
20. Miss Sonia Nolten of counsel, appearing for the Defendant, accepted the principle, and acknowledged that the Court should not commit what she called the " 'Sherlock Holmes' error" of "conclud[ing] in the case of a series of improbable causes that the least improbable or least unlikely is nonetheless the cause of the event... arriving at the least improbable cause and not the probable cause" (per Thomas LJ in Ide v ATB Sales Ltd [2008] EWCA Civ 424, [2009] RTR 8 at [4]). That is clearly right, though Miss Nolten's attribution of such error does not do full justice to Sir Arthur Conan Doyle, whose famous detective actually opined that "when you have eliminated the impossible, whatever remains, however improbable, must be the truth" (The Sign of the Four (1890), ch 6 – my underlining).
21. However Miss Nolten went on to stress that this problem only arises in "cases … where there may be very real uncertainty about the relevant factual background (as where a vessel was at the bottom of the sea) or the evidence might be highly unsatisfactory" (per Thomas LJ loc cit), and submitted that this is far from such a case.
22. She submitted that it is only in an exceptional situation that a civil court can properly resolve a disputed issue of fact by falling back on the burden of proof. Seven relevant authorities, including Morris v London Iron & Steel Co Ltd [1988] QB 493 (CA), were considered by the Court of Appeal in Stephens v Cannon [2005] EWCA Civ 222, [2006] RVR 126. Wilson J (as he then was), with whom Arden and Auld LJJ agreed, derived the following propositions from them (at [46]):

(a) The situation in which the court finds itself before it can despatch a disputed issue by resort to the burden of proof has to be exceptional.

(b) Nevertheless the issue does not have to be of any particular type. A legitimate state of agnosticism can logically arise following enquiry into any type of disputed issue. It may be more likely to arise following an enquiry into, for example, the identity of the aggressor in an unwitnessed fight; but it can arise even after an enquiry, aided by good experts, into, for example, the cause of the sinking of a ship.

(c) The exceptional situation which entitles the court to resort to the burden of proof is that, notwithstanding that it has striven to do so, it cannot reasonably make a finding in relation to a disputed issue.

(d) A court which resorts to the burden of proof must ensure that others can discern that it has striven to make a finding in relation to a disputed issue and can understand the reasons why it has concluded that it cannot do so. The parties must be able to discern the court's endeavour and to understand its reasons in order to be able to perceive why they have won and lost. An appellate court must also be able to do so because otherwise it will not be able to accept that the court below was in the exceptional situation of being entitled to resort to the burden of proof.

(e) In a few cases the fact of the endeavour and the reasons for the conclusion will readily be inferred from the circumstances and so there will be no need for the court to demonstrate the endeavour and to explain the reasons in any detail in its judgment. In most cases, however, a more detailed demonstration and explanation in judgment will be necessary.

23. Miss Nolten submitted that the present case does not give rise to such an exceptional situation, but is one where "[t]here is sufficient evidence … to enable the court to engage in an informed analysis of the possible causes of the loss and to reach a reasoned conclusion as to the probable cause. The fact that the experts were unable to reach a conclusion of their own does not preclude the court from reaching such a conclusion on the totality of the evidence and in the light of the findings of fact" (per Richards LJ at [67] in Datec Electronic Holdings Ltd v United Parcels Service Ltd [2005] EWCA Civ 1418, [2006] 1 Lloyd's Rep 279; upheld at [2007] UKHL 23, [2007] 1 WLR 1325, Lord Mance citing this passage at [43]).
24. I substantially accept Miss Nolten's submissions on this question. Whilst there are occasional, exceptional cases where the court can properly fall back on the burden of proof, without making any positive finding on the balance of probabilities, I do not consider the present case to be one.
25. As to the principle in Re B, Miss Nolten acknowledged that a common sense view of inherent probabilities is the proper starting point, and that one feature of this is that criminal acts are less likely to take place than non-criminal ones. However she submitted that this does not mean that the bare raising of a "possible" non-criminal cause will always suffice for a suggested criminal cause to be rejected, and cited National Justice Compania Naviera SA v Prudential Assurance Co Ltd (The Ikarian Reefer) (No 1) [1995] 1 Lloyd's Rep 455 (CA) as an example on its facts. She further submitted that to approach a civil case on that basis would effectively raise the standard of proof to the criminal one. As Stuart-Smith LJ, giving the judgment of the Court in The Ikarian Reefer (No 1), said (at 484):

The burden [on insurers seeking to prove that a fire was deliberately started (there on a ship)] is a heavy one; but the mere existence of the possibility that the fire was accidental does not mean that it has not been discharged. There must be a real or plausible explanation which is supported by the evidence, or at the least is not inconsistent with it.

26. While Miss Nolten's submissions are correct so far as they go, the point remains that on a common sense view of inherent probabilities it is less likely that someone (and particularly a resident and partner of the owner) would deliberately start a fire in the basement of a house, than that a smoker would carelessly discard or accidentally dislodge a lit cigarette in her bedroom (particularly if agitated and/or anxious to leave in a hurry at the time).
27. I approach the question of fact which I have to decide in accordance with the words of Lord Hoffmann in Re B, and Stuart-Smith LJ in The Ikarian Reefer (No 1) (a further citation from which appears below).

The principal witnesses of fact

28. The Claimant gave evidence himself, and called Mr Howe and a Mr Smart. The Claimant is a man in his forties, who works as an information security manager. At the time of the fire he was commuting from Brighton to London for work. Mr Howe had been his partner for 2¼ years when the fire occurred, and for over 6½ years by the time of the trial. Mr Eydmann described the Claimant as an entertaining person, whom everybody liked. The Claimant acknowledged that in 2008 he and Mr Howe had been quite heavy drinkers, principally at the weekends, which for drinking purposes they treated as starting on a Thursday evening. He answered questions in cross-examination fairly and realistically, and subject to points of detail as to which he was not sure (something wholly unsurprising for any witness being asked about events over 4 years ago, and which he was quite ready to acknowledge), I found him to be a reliable witness.
29. Mr Howe, who turned forty last year, was not in employment at the time of the fire. However he undertook the housework at the property, and did so conscientiously, being somewhat fastidious and house-proud in nature. A feature of his character which attracted some attention at trial is that from time to time he had tantrums or, as some referred to them, hissy-fits, despite being almost 37 at the time of the fire. His partner, the Claimant, affectionately referred to them as "Tony Bear Tantrums". These tantrums were generally in the nature of attention seeking behaviour rather than being dangerous or destructive; they were more likely to occur after drinking alcohol. Most were in the nature of having a moan, ending with a "Huh!" or some such, and walking or stomping off. More occasionally, they featured stamping feet and shouting. Once he caused property damage, by kicking and breaking some (already wobbly) spindles supporting the first floor bannister. This incident came after the Claimant's mother had (in the Claimant's words) been "winding him [Mr Howe] up" all day, something which was then aggravated by his arguing with the Claimant, and the Claimant hanging up on him while they were speaking by telephone. Mr Howe openly acknowledged what he had done (albeit that Mr Eydmann had been present at the time, making it difficult to do otherwise), and, the Claimant said, got it repaired. The possibility of his having broken a small, 6" square window pane in the kitchen on another occasion was raised, but I have not been satisfied on the evidence that he was responsible for this. Mr Howe volunteered evidence of an incident in a supermarket 3 years after the fire, when he could be said to have behaved like a 5 year old, but I do not find that of any significance for the question I have to try.
30. Mr Howe was warned of his privilege against self-incrimination with regard to cross-examination suggesting that he deliberately started the fire, but in the event he did not decline to answer any questions about it. When interviewed by the police the day after the fire (the delay in interviewing seems to have been due to the amount of alcohol found in his blood when samples were taken following his arrest), he was asked directly whether he had set fire to Ms Alexander's bed or anything else in her bedroom, and firmly denied it. Mr Howe's essential account of matters has been consistent throughout, although his recollection on matters of detail was not always correct, particularly after the passage of even a little time. It should of course be borne in mind that the evening of the fire was a traumatic one for all those directly involved. He gave evidence in a calm manner, even where the questioning was potentially difficult or embarrassing, for example in relation to his tantrums, or suggesting that he had started the fire. He did not take obvious opportunities to 'gild the lily' – e.g. as to the number of occasions on which he had seen Ms Alexander smoking in her room, whether she was smoking in the lead-up to the fire, or whether he had seen candles or similar items in her room. At the end of his evidence, when shown some of the photographs and asked to comment on them, he recalled (correctly, as it transpired) that there had been something at the foot of Ms Alexander's bed, which he believed to have been a box.
31. Miss Nolten argued that I should be cautious about accepting Mr Howe's denials, asserting that few people who have committed arson and not been prosecuted will admit it. That may well be so, and the same would presumably be true of other activities leading to substantial insurance claims which involve the commission of one or more serious criminal offences (arson, major criminal damage, fraud, perjury etc) at some stage in the process, such as scuttling ships. Miss Nolten cited a further passage from The Ikarian Reefer (No 1) (above) which is in point here, for Stuart Smith LJ (later on 484) went on to state that:

Most experienced Judges recognize that it is not easy to tell whether a witness is telling the truth … Where, as in this case, there is clearly evidence upon which a charge of arson can properly be advanced by the underwriters, the Judge should be careful to keep an open mind as to the honesty of those impugned. It imposes too high a burden on the underwriters to say that such witnesses must be telling the truth unless the underwriters prove that their accounts are impossible. Their evidence has to be tested in the light of the probabilities and the evidence as a whole.

32. I have borne Miss Nolten's argument, and that further passage from Stuart-Smith LJ, firmly in mind. Having done so, and having heard Mr Howe's evidence tested by cross-examination and by argument founded on other evidence both factual and expert, I nevertheless find Mr Howe's evidence to be convincing, given its consistency, and that it ties in with much of the other evidence, including as to the sequence of events, as well as the calm and realistic manner in which it was delivered, and it according with a common sense view of inherent probabilities. I have of course reached that view of Mr Howe's evidence independently, but having done so I note that the police officers who first investigated this fire arrived at a similar evaluation of Mr Howe's account [C/19/162-3].
33. Mr Derek Smart first met Mr Howe in 2005, through a social networking site called 'gaydar'. Through him Mr Smart in due course met the Claimant, Mr Eydmann and Ms Alexander, and they became friends. In due course he and Mr Eydmann entered into a relationship, which lasted from about April until September or October 2007; for most of this time Mr Eydmann was living in the property. While in that relationship, Mr Smart visited the property frequently, sometimes staying over. He often went to the basement of the house to make coffee for himself and others in the kitchen, becoming known in the house as 'the coffee bitch'. While doing so he would visit Ms Alexander, offering to make her a drink, and from time to time he smoked with her in her bedroom. On a majority of such occasions when he saw her in her bedroom, she was either smoking or there were signs of her having smoked. After his relationship with Mr Eydmann ended, Mr Smart still visited the property from time to time, though less frequently, and thenceforth he only saw Ms Alexander there every 2 or 3 weeks. Ms Alexander became one of his Facebook friends.
34. On Tuesday 3 July 2012, 6 days before the commencement of this trial, he received 'out of the blue' (not having heard from her for something like 4 years) a Facebook message from Ms Alexander. The resultant exchange, over a total period of about 15 minutes, ran as follows (after it had finished, Mr Smart e-mailed a copy of it to the Claimant's solicitors, and a hard copy print was inserted into the hearing bundle at B/3/17A-B):

RA: Why are you lying about me in court? Are they paying you to do it?

DS: what?

RA: Ive been told you going to court for jono and the fire

DS: yes I have been asked to be a material witness

RA: Ive read your statement and I never smoked 20 fags in my room and I hardly ever smoked in my room only when a party was happening and you didnt come in my room all the time. So what your saying is lies. And ive been told you could get in serious trouble. So jono much be payin you!

DS: Hopefully the truth will come out in court !

RA: Yes it will you lying sack of shit, I hope they lock you up, or laugh in your face, your just as bad as that murdering asshole tony

[Ms Alexander at least sometime refers to the death of the 3 household cats in the fire as murder].

35. Although Miss Nolten invited me to approach Mr Smart's evidence with "circumspection", I found him to be a straight-forward, balanced and reliable witness, whose measured, 'straight bat' responses to Ms Alexander's aggressive and abusive Facebook messages about his forthcoming appearance as a witness in this case reflect well on him, and are consistent with the calm manner in which he gave evidence in court. His evidence as to Ms Alexander's smoking habits ties in with that of other witnesses, as well as with the photographic evidence of the presence of numerous smoking materials in amongst the debris which at least largely came from Ms Alexander's room.
36. The Defendant called Ms Alexander and Mr Eydmann. Ms Alexander was 35 in July this year. She left her native city of Birmingham in about 1996, and moved to Brighton. At the time of the fire she was working in 2 jobs, one as a residential care worker in a children's home, and the other behind a bar. She was a very untidy person, as she herself at one point acknowledged 'out of sheer laziness', and her bedroom was very messy – a point on which I shall elaborate later.
37. Despite repeated denials by Ms Alexander, from the evening of the fire onwards, that she smoked in her bedroom other than very occasionally or during parties, there is an abundance of evidence that she did so quite frequently, and I so find. Realistically, and particularly since she had 2 ashtrays of her own (one red, one blue), she would have had no other reason to move the various "mobile ashtrays" from around the house to her room (from which they regularly had to be retrieved, full), and they did not fill themselves. Quite apart from the Claimant's witnesses, Mr Eydmann gave evidence that he, together with Mr Howe, had removed such ashtrays from her room and emptied them. In any event, I accept the evidence of the Claimant's witnesses as to Ms Alexander smoking in her room. The presence of numerous smoking materials amongst the debris - which I am quite satisfied mostly if not entirely came from her room – which ended up after the fire in the basement level courtyard, and was there observed by first Mr Cox and then Dr Forryan, supports this. Although Miss Nolten attempted to explain this away by seeking to establish that the brand names which were visible were not Ms Alexander's preferred brand, Ms Alexander herself accepted during her cross-examination that she probably did commonly discard cigarette packets in her bedroom. Few people take already empty cigarette packets into their bedroom, particularly when they have no waste paper basket there (when Ms Alexander used one at all, she used a bin in the kitchen). I also accept from evidence of the Claimant, Mr Howe and Mr Smart that smoking in the kitchen area was frowned upon by both the Claimant and Mr Eydmann. I do not accept the evidence of Ms Alexander that she regularly smoked at the foot of the stairs leading to 'no man's land', which appeared to me to be a response to the pressure generated by questions pointing out the evidential difficulties with her denials of smoking in her bedroom, nor her statement to the police that she rarely smoked in her room, and only smoked in the kitchen [C/6/30]; I also prefer that of the Claimant indicating to the contrary. The basic point is in my judgment quite clear on the evidence – Ms Alexander quite frequently smoked in her bedroom. That of course does not in itself mean that she did so that evening, nor a fortiori that one of her cigarettes started this fire, but it does undermine one of the bases on which early investigations of the fire appear to have proceeded.
38. Ms Alexander was warned of her privilege against self-incrimination when cross-examined about the messages she sent Mr Smart via Facebook, but in the event, and having at one point sought clarification about that privilege, she did not decline to answer any questions about it.
39. Ms Alexander was in my judgment, to put it at its kindest, an overly defensive witness (a view which I have of course formed independently, but which I note again ties in with the impression she created on the police officers investigating the fire). She was the one who made the accusations against Mr Howe which led to him being arrested. In her dealings with the investigating officers she was noted as having appeared jumpy and abrupt, been nervous when asked to make a further statement, and - when asked about the possibility of the fire having been accidentally started in her room - become "very agitated and loud" (despite being accompanied by her father) [C/19/163]. She was "adamant" that she did not smoke in her room (see e.g. the note of the police officer who interviewed her on 25 January 2008 at C/19/156), although this was clearly incorrect, and goes well beyond many smokers' tendency to under-estimate their own level of consumption. Although, as Miss Nolten submitted, Ms Alexander's conduct in the Facebook exchange with Mr Smart which she initiated (quoted above; see B/3/17A-B) may not have gone far enough to constitute the offence of witness intimidation under s.51(1), Criminal Justice & Public Order Act 1994 (citing R v Patrascu [2004] EWCA Crim 2417, [2005] 1 WLR 3344 – see in particular at [6] and [18] re the effect of ss.51(1)(c) and 51(7)), and it is unnecessary and undesirable that I make any finding about that, Ms Alexander was at the very least seeking to discourage Mr Smart from giving evidence in accordance with his witness statement, and doing so in terms which show her to have been overly defensive.
40. Overall, I regard Ms Alexander as a significantly less reliable witness than Mr Howe, or the Claimant's other 2 witnesses, and someone whose evidence on controversial matters ought only to be accepted if corroborated in some sufficient way.
41. Mr Grant Eydmann, a physiotherapist, had known the Claimant for quite a number of years prior to renting a room in the property from him in or about May 2007. He was referred to as having been the 'mother' or 'mother hen' of the house. His manner in the witness box tended towards the flamboyant, and at times he appeared to be 'playing to the gallery' by the manner in which his answers were given, if not by their content. In the event, however, little of his evidence has proved to be significantly controversial. Although his private nickname for Mr Howe (not used in front of him or the Claimant) was apparently "psycho Tony", Mr Eydmann made clear in his evidence that this was simply a colourful reference to the fact of Mr Howe's tantrums; he specifically said that he had not intended it to mean that Mr Howe was psychotic.
42. For completeness I should record that:

37.1. the contents of the witness summary [B/4/18-19] served on behalf of the Claimant in respect of Detective Sergeant Sian White, who was based at Brighton Police Station, and had charge of the police investigation into the fire (when still a Detective Constable), and the contents of the various (redacted) police documents [at C/19/146-169a] incorporated therein, were ultimately agreed as facts, subject to the exclusion of the documents at C/19/152-4 from the terms of paragraphs 3 and 4 of the summary, and subject to Miss Nolten emphasising that the subjective conclusions of police officers recorded therein were not thereby admitted to be correct conclusions; and

37.2. the evidence of Acquin Magee, a secretary at Crawfords who transcribed, as best she could, the digital recordings of interviews conducted by Mr Wintrip of that firm, and then submitted them to him to check and amend if necessary [B/7/31-32], was by agreement taken as read without her being required to attend for cross-examination.

The sequence of events

43. On the day of the fire, the Claimant had stayed in London after work to attend an early evening talk, and only caught a train back to Brighton at or shortly before 8.30pm. Mr Eydmann had got home from work at about 6.15-6.20pm. After getting something to eat, he went to relax and watch television in his room. Some time later Mr Howe rang him up and asked if he wished to join him at the nearby White Horse pub, but Mr Eydmann declined. Shortly afterwards he fell asleep.
44. Mr Howe had been at home during the day, and had cleaned and tidied it, including in particular the kitchen, which he had left 'spotless'. He had also cleaned the inside of the windows throughout the house, save for that in Ms Alexander's room.
45. After having worked a 24 hour shift at the children's home, Ms Alexander returned to the house at about 10.00am that morning, in her own words "pretty tired", not having had a full night's sleep. She had 60 to at most 90 minutes sleep, before leaving at about 11.30am to start a shift at the Stadium bar at 12 noon. She finished that shift at 7.00pm, and as her journey home, which included a bus ride, takes a minimum of 30 minutes, the earliest she can have got home was 7.30pm. There is nothing to suggest she was in any particular rush. It is likely that she got home a little after 7.30pm, but not later than about 7.50pm. She would by then have been very tired.
46. When Ms Alexander got home, she went on to cook herself a meal of fried steak and roasted vegetables. With or without smoking a cigarette first, it would have taken a few minutes to prepare the vegetables for roasting, and also a few minutes (quite likely concurrently) for the oven to warm up ready. She told me that she generally roasts vegetables for only 15 minutes, as she likes them crunchy. The steak will have been fried during the latter part of those 15 minutes. The dishes used were left in the kitchen when she took the meal to her bedroom, closing the door behind her. She did so by about 8.15pm, as her door was closed when Mr Howe got in and went to the kitchen. The overall time taken to prepare and then cook her meal would have been about 20-25 minutes (a little longer than Ms Alexander estimated).
47. She denied smoking a cigarette before eating, but given that she smoked regularly, including in her room, it would be unsurprising if she had simply forgotten doing something which she will have done many times. There was time for her to have done so before starting to cook. She also denied lighting a cigarette after eating, but again it would be unsurprising if she had simply forgotten doing so. She claims to have recalled that she had no cigarettes with her at home that evening, but I am sceptical about the reliability of that claimed recollection, and – particularly given that there is nothing to suggest that she had been planning to go out again that evening – I find it improbable that Ms Alexander, a regular smoker, would have walked past the Co-op store at the end of her street (where she often bought them), on her way home from the bus stop, with literally no cigarettes on her person or at home, without stopping in and buying some. It is also quite possible that when the horror of the fire ensued she simply became convinced that she cannot have been the cause of it. It is all too easy for a normally rational person who has been through a traumatic experience, even one as simple as, say, a road accident, quickly to come to view it and its causes with self-exculpatory hindsight.
48. Mr Howe went out drinking from about 4.30pm that afternoon, in anticipation of his birthday on Friday of that week (when he and the Claimant were planning to go away for the weekend). He visited 3 different pubs or bars. At the Marine Tavern he had 3 pints of lager and 2 or 3 Sambuca shots. At the Bulldog he had 3 vodka and cokes; whether singles or doubles he did not know. At the last, the White Horse, which is very close to the property, he had a further alcoholic drink (possibly another vodka). While there, he rang Mr Eydmann (as I have mentioned), and received a short phone call from the Claimant at 7.39pm (that time is taken from the Claimant's itemised mobile phone bill), who said something to the effect that he was heading home. Mr Howe finished his drink before walking home. Whilst for many what he had consumed would be ample to produce a state of drunkenness, I accept Mr Howe's description of himself at the end of it as (merely) "merry", given the Claimant's frank acknowledgment in evidence (backed by specific examples of how a typical weekend would unfold) that both of them were heavy drinkers at this time, his evidence that on his arrival back at Margaret Street Mr Howe appeared sober to him, and Mr Howe's own evidence that this was an average evening's consumption for him at this time; it also appears that Ms Alexander did not realise that he had been drinking.
49. Mr Howe returned home at about 8.15pm, and went more or less straight downstairs. Ms Alexander's door was closed, but he could tell that she was in because the light and the television in her room were on. He called out some sort of greeting to her, but she did not answer because her mouth was full. He got a pizza from the freezer, and put it in the microwave for something of the order of 4 to 6 minutes. He noticed that the kitchen, which he had earlier cleaned and tidied, was a mess, and mentally blamed that on Ms Alexander; though he may not have realised it, she had in fact only just cooked her own dinner. When he took the pizza out of the microwave the plate broke, and this caused Mr Howe, already cross with Ms Alexander because of the mess in the kitchen, and now cross with himself too, to shout out at her in annoyance. She had finished eating by then, and opened her door to challenge him. A short shouting match ensued, whereupon she returned briefly to her room, pulling her door shut behind her. The shouting continued, through the door. Mr Howe had no positive recollection of her having a cigarette in her hand at this stage, though he was clearly cross at the time, and as a fellow smoker of menthol cigarettes could not tell from sense of smell whether she had been smoking in her room. Piecing together the sequence of events up to 8.46pm, and bearing in mind the most significant timings of which there is reliable evidence (7.39pm as above; 8.36pm, when Ms Alexander sent her first text message to the Claimant (to which I shall come); 8.41-8.42pm, when the Claimant and Ms Alexander exchanged text messages (to which again I shall come); and 8.46-8.47pm, when Mr Eydmann's 999 call was made), this initial shouting match would have taken place at roughly 8.25pm.
50. Ms Alexander only remained in her room relatively briefly. She was very tired, upset because of the shouting match/argument, and just wanted to get out of the house quickly. She grabbed her coat, put it and shoes on, took her phone and 3 pounds, leaving a bag (containing tissues, make-up, papers and a lighter) on her bed, and went upstairs. This will have taken little time, and during that time she was agitated, and anxious to leave in a hurry. I do not consider that there would have been sufficient time for the fastidious and house-proud Mr Howe to have cleaned up after the broken plate, put the pizza on a fresh plate, cut it in half, carried his half upstairs, sat down and commenced eating it before she emerged, and I prefer his recollection that he was still in the kitchen when Ms Alexander left her room to her recollection that he was already upstairs in the sitting room eating his pizza; he chased her up the stairs, and (as is common ground) shouted abuse behind her as she left the house and set off up the street.
51. Piecing together the sequence of events as before, Ms Alexander is likely to have left the house by about 8.30pm. Whilst it was not until 8.36pm (applying the requisite 4 minute correction to the time shown on her then mobile phone handset, which she had with her when interviewed by the police the same evening) that she sent her first text to the Claimant, the evidence is that she had first tried to phone him at least once. In contrast to Mr Eydmann's 999 call, when he left a house which he knew to be ablaze, there was no especial urgency requiring that she make the call or send the text message to the Claimant immediately, and I am not satisfied that she did so. Her 8.36pm text message read "Big fat Tony tantrum. Sort it out, I'm on the receiving end this time". She was recorded as entering the nearby Co-op store in St James' Street, at the top of Margaret Street, on its CCTV at 8.41pm (to buy cigarettes, which she did not have with her; she may or may not by then have run out at home). As the experts note, the timing on that CCTV system appears not to have been verified by the police or anyone else. Although Ms Alexander exchanged text messages with the Claimant at about 8.41-8.42pm, given the overall sequence of events it seems that the CCTV timing will not have been many minutes out. She then sat smoking one of the cigarettes she had just bought at a bus stop near to the Co-op store.
52. In the meantime Mr Howe had closed the front door (ultimately both parties agree that some rather confusing evidence as to whether the door was then locked, and if so how, takes matters no further), and returned to the kitchen. Ms Alexander's bedroom door was closed, but I am satisfied that Mr Howe would have noticed if the fire had been ablaze, or if copious quantities of smoke were being produced, at this time. He re-heated the pizza, cut it in half, leaving half on a plate in the kitchen for the Claimant (as a fire brigade photograph [E/105] reminded him that he had done), and then went upstairs to eat in the sitting room. From shutting the front door after Ms Alexander had left until sitting down to eat, roughly 5 minutes will have elapsed. He may have made a short mobile telephone call to the Claimant at about this time (the Claimant's rather tentative evidence was that such a call was made) – if he did, it would have been at some time between Ms Alexander leaving the house and 8.41pm (when the Claimant texted her). On the Defendant's case, it must have been during this 5 minutes that Mr Howe also opened the door to Ms Alexander's bedroom, and set fire to something on top of, or immediately adjacent to the top of, her bed using a naked flame, such as that generated by a match or cigarette lighter, before closing the door again (because, as I shall come to, Mr Eydmann subsequently saw him open it).
53. Mr Howe had nearly finished his half of the pizza when he first became aware of the smell of smoke coming from downstairs, and I accept Mr Howe's estimate that he had been sitting down for 4 to 5 minutes at this stage, because it allows sufficient time for eating the relatively modest quantity of pizza he had, and possibly making a short mobile telephone call to the Claimant (if he did, but not while still in the kitchen), and fits in with the overall sequence of events. The structure of the house, with gaps between the floorboards, is such that the emission of copious quantities of smoke in the basement bedroom would have produced a detectable smell of smoke in the sitting room within quite a short time. Mr Howe did not immediately become concerned by the smell of smoke, but after finishing his food, which will only have taken a further minute or two, Mr Howe went downstairs to investigate.
54. From here on, events moved quickly. When Mr Howe got downstairs to the basement, he could tell from the combination of the smell of smoke and (by then) a red glow visible around the edges of the concertina door that there was a fire in Ms Alexander's bedroom.
55. I accept Mr Howe's evidence that he only opened the concertina door once after he was aware that the fire was ablaze. Although he could not recall whether he did so before or after he roused Mr Eydmann, it must have been afterwards, because Mr Eydmann saw him do it. Mr Howe ran back up to the ground floor, and shouted to Mr Eydmann from near the bottom of the stairs up to his first floor bedroom to awaken him and alert him to the danger. Mr Eydmann awoke (this must have been some minutes later than the 8.35pm estimate which Mr Eydmann gave in his police interview later that evening), and headed downstairs wearing only his shorts. He was (as he made clear in his oral evidence) part-way down the open stairs from 'no man's land' to the basement when he saw Mr Howe open the concertina door, and (when he did so) the fire in the bedroom, with flames rising up to the ceiling, and smoke billowing out of the room.
56. Assuming, understandably, that Mr Howe would follow him out of the house, Mr Eydmann turned and ran back up to his room, quickly put on a few outer clothes and grabbed his mobile phone, and then ran down to the front door to leave. He was in such a hurry that he did not even pause to take his wallet and keys. In his no doubt agitated state he struggled briefly with the lock before the door opened, emerged into the street, and immediately dialled 999. The house is not large, and I am satisfied that it will only have taken 2 to at most 3 minutes from Mr Eydmann observing the fire as the concertina door opened to his leaving the house. His 999 call commenced at 8.46pm.
57. Mr Howe's description of what he saw when he opened the door, given to the police when interviewed the day after the fire, was that the flames were at the bottom of the bed, but that all of it was alight. When interviewed by investigators for the Defendant about 6 weeks later, and asked whether he could see what was on fire, he answered "It was, well, it was the top of the bed", but when then asked how high the flames were, or how big the fire was, he responded "I dunno, it was just like a mass of red" [C/16/114]. In his oral evidence he said that it was still his recollection that the fire was on the top (meaning the surface) of the bed, near its foot; he did not notice things on the floor on fire, but the flames filled the room, up to the ceiling. I find that at the time Mr Howe entered the room, observed by Mr Eydmann from part way down the open stairs, the fire was well alight, the bed in particular was ablaze, and there were flames either coming or appearing to come from the surface of the bed towards its foot, but that it cannot safely be concluded that the first flame or flames which marked the onset of combustion came from the surface of the bed itself, as opposed to other materials either immediately adjacent to the foot of the bed (e.g. the cardboard box) or on the bed (e.g. the tissues or papers in Ms Alexander's bag).
58. Mr Howe made attempts to extinguish the fire, by a few times throwing water (which he had collected in a salad bowl or similar) onto the fire, before leaving the house. He was still inside when Mr Eydmann started his 999 call, but emerged within a minute or so, before that call had ended. He sat down in the doorway of one of the houses opposite, in tears. In the meantime, following Ms Alexander's request by text message sent at 8.42pm, the Claimant had rung her twice, both being brief calls, and timed at 8.43 and 8.46pm.
59. Shortly after that, and attracted by what appeared to be mist or smoke, Ms Alexander came back down Margaret Street. There she spoke to Mr Eydmann, before going across to Mr Howe. In effect, she accused him of starting the fire. She shouted at him "What have you done?" to which he replied "What have I done? What have you done?" She also struck him on the head with her hand, at least twice. This exchange, with its implicit denial in Mr Howe's response, is rather different from the impression Ms Alexander gave to the police officers who initially attended the scene (for which see C/19/151), although (even assuming they accurately recorded what they were told in a doubtless fast moving situation) that may well be attributable to the traumatic nature of the evening for Ms Alexander and others. Both she and Mr Howe realised that the household's 3 cats were still inside the house, and this exacerbated their emotional reactions to what was in any event the most disturbing and distressing sight of their home burning in front of them.
60. The first of four Fire Brigade crews attended by about 8.52pm. An experienced fire investigator employed by the fire service as a Group Manager, a Mr Peter Cox, attended that evening, and again the following morning. The house was treated as a crime scene, and secured accordingly, with a police presence for the next 2 days. At Mr Cox's instigation an accelerant detector dog was used extensively at the scene in the morning, but no evidence of the use of any accelerants (e.g. petrol) was found. In due course at least 36 photographs of the scene were taken by the fire service (E/71-106). In part on the basis of information relayed from Ms Alexander, including that she had not smoked in the room that evening, Mr Cox's initial conclusions were that the material which first ignited was "furniture – upholstered" (I infer the bed); that the maximum time period for development of the fire was 15-20 minutes (which would be about right, unless the smoulder had started while Ms Alexander was still eating, but at that stage produced insufficient smoke to be noticed by her, which is a possible scenario); that this was too short a time for a carelessly discarded cigarette end to be the source of the ignition (an understandable, if not necessarily certain, conclusion on the premise that it was upholstery fabric or the like (rather than a paper-based material) which first ignited – see my summary of the expert evidence below); and that the most likely cause of the fire was (therefore) deliberate [see C/4/11, 13b & 13c].
61. Largely on the basis of accusations made, or at least slanted information apparently given, by Ms Alexander, Mr Howe was arrested by the police in the house of a neighbour who had taken him in. The police investigation was led by DC White. Mr Smyth, Mr Eydmann and Ms Alexander (twice) were interviewed as witnesses. Mr Howe was interviewed under caution, but having made a favourable impression with his account of matters, which DC White assessed as "very credible" (with a colleague agreeing), was bailed without charge, and in due course told that he was not to be prosecuted.
62. Within a fortnight Halifax, for the defendant insurers, instructed the firm of Hawkins to investigate the cause of the fire. Dr Claire Forryan of that firm undertook an initial investigation on the insurers' behalf, although later she took up the role of an independent expert witness reporting to the court. On 8 February 2008 Dr Forryan visited the scene, and took at least 70 photos (E/1-70), including a number of the large pile of debris which she observed in the courtyard off the kitchen, comprising mainly burnt women's clothes which had obviously come from Ms Alexander's room. In amongst this she observed numerous cigarette packs and lighters (as the Fire Brigade's investigator Mr Cox had done on the morning after the fire); numbers of the photographs of this debris confirm this, whether or not any or all of the items which the Fire Brigade picked out and put into the box lid depicted in photograph E/106 had been amongst them. Although I accept that fire officers would probably have taken note of anything which they recognised as the remains of an ashtray (and it appears that identifiable molten remains of glass ashtrays can survive fires) in amongst this debris, the generally chaotic state and appearance of the debris afterwards is such that I do not consider that one can safely conclude from the absence of any positive report of such remains in the Fire Service documents that there had been no glass ashtrays in the area of the bed at the time the fire started. Dr Forryan's initial report to Halifax (dated 20 February 2008) included some limited accounts of the background facts which she herself had obtained from Mr Howe, Mr Eydmann and Ms Alexander, in the latter 2 cases by telephone (the police appear not to have released their witness statements and record of interview to her at that stage), and a reasonably detailed account of her own observations on her site visit. She eliminated an electrical fault as the explanation. Her understanding from her conversations with Ms Alexander and Mr Howe was that the fire had been discovered within 5 minutes of Ms Alexander leaving the house (I have found it to have been significantly longer than that), and on that basis (understandably) opined that a discarded cigarette is unlikely to have smouldered and developed into a flaming fire within such a short timescale. She also noted that the fire damage was most severe in the centre of the bedroom, around the foot of the bed, that Ms Alexander denied that she had smoked in her bedroom that evening, and that if such denial was correct the fire could not have started due to accidentally discarded smokers' materials. On this basis she considered it most likely that the fire was started deliberately, and at that stage left open whether that was by the actions of Ms Alexander before she left or of Mr Howe once she had done so.
63. The firm of Crawford, also acting for the defendant insurers, interviewed Messrs Smyth and Howe, and Ms Alexander, at their offices, roughly 6 weeks after the fire. Mr Eydmann was spoken to a couple of weeks later, by telephone. The 3 office-based interviews were conducted by Mr Robin Wintrip of Crawford, assisted by Mr Raj Soni of Halifax. The typed transcripts of them, and some manuscript notes of the telephone interview of Mr Eydmann, are in the court bundles, and extracts from the typed transcripts were frequently referred to in the course of the questioning of witnesses. Ms Magee and Mr Wintrip no doubt did their best with the process of transcription and review/approval of the draft transcripts, but the resultant transcripts are not always clear, and unfortunately Crawford's digital recordings underlying them have been lost from its IT system.

The expert evidence

64. Mr Chris Kenyon-Edwards and Dr Forryan were instructed and called by the Claimant and Defendant respectively. I have found sections 3 and 4 of their first joint statement, their second joint statement, and their oral evidence supplementing the same the most helpful. The first joint statement preceded the trial in the usual way, and I should immediately acknowledge that section 2 (as to the factual circumstances of the fire) very properly noted that the interpretation of witness evidence was a matter for the court. In the event, the oral evidence of the witnesses has significantly affected the relevant available material, and my findings of fact do not wholly follow what is postulated in section 2. The second joint statement was prepared (in manuscript) with the agreement of all parties during a break in Mr Kenyon-Edwards' re-examination, so as to enable the experts to supplement their initial joint statement with regard to the implications of certain evidence of fact which had emerged during the trial, and in particular the evidence (principally from Ms Alexander herself, but confirming the recollection of Mr Howe) of the presence of a cardboard box at the foot of her bed, and Ms Alexander's admissions as to the casual discarding of tissues which had been used for make-up removal at or in the area of the foot of her bed.
65. The experts agree that the patterns of fire damage indicate that the fire started at, or close to, the foot end of Ms Alexander's bed. That is consistent with the evidence of primary fact, and I accept it and so find. The experts also agree, in part from information from Mr Cox, that an electrical fault can be ruled out as the cause of this fire; again I accept this and so find.
66. Hence they conclude that "the only remaining plausible causes are associated with human agency, either deliberate ignition or accidental causes, for example carelessly discarded smokers' materials or candles". That is also my view, and for the reasons I have already given it effectively poses the key question which I have to address, namely whether the fire was deliberately started by Mr Howe (there being no other plausible candidate for having done so), or accidentally started by Ms Alexander (again, there being no other plausible candidate for having done so). As to a possible accidental cause, although there is some evidence that there were candle holders and (unlit) 'night-light' type candles in Ms Alexander's room (supported by photographs E/24 & 38), I consider this to be a considerably less likely accidental cause than one related to smoking materials, and accordingly will focus my evaluation of the 'accidental cause' limb of the key question on smokers' materials (and, more specifically, a putative lit cigarette).
67. In accordance with what is accepted in modern practice, Dr Forryan in her report dated 19 April 2012 expressed her own opinion on at least one aspect of the key question of fact (albeit one on which the scientific evidence is part of the available, relevant evidence), namely whether this fire was started deliberately or accidentally (her opinion was the former). This is of course ultimately a matter for the court itself, involving evaluation of the evidence of fact, as well as expert scientific opinion. Whilst there was nothing wrong with her doing so, (a) it inevitably runs the risk that the evidence of primary fact which informs that question will turn out at trial to be rather different from how it appeared 'on paper' beforehand, and (b) I do not accept Miss Nolten's submission that the evidence of Mr Kenyon-Edwards is to be treated as weaker because he did not do likewise. In what might be regarded as a more traditional approach to the legitimate scope of expert opinion, he limited his statements of opinion to what can and cannot be excluded scientifically, leaving the evaluation of factual evidence and relative probabilities to the court.
68. It is common ground that there would be little difficulty in starting a fire in combustible material such as bedding or clothes (or, no doubt, some paper-based material in close proximity to the bedding) by the application of a naked flame, for example one generated by a match or a cigarette lighter, without the need for the use of an accelerant, and that such a cause was unlikely to leave any residual physical evidence of itself.
69. As to an accidental fire caused by 'smouldering ignition' (starting with a smoulder, but culminating in flaming ignition), the experts agree that it is very unlikely that such ignition would follow from a lit cigarette left lying on a flat surface (whether fabric or cardboard). In addition to fuel and oxygen, such ignition requires a degree of insulation, and can (though is by no means sure to) follow where a lit cigarette becomes lodged between layers of clothes, or within crumpled bedding or paper. In their second joint statement (written, as I have mentioned, after the evidence of fact was complete) they identify 3 potentially relevant locations for a lit cigarette having the necessary oxygen, fuel and insulation to enable an initial smoulder to progress to flaming ignition, namely:

64.1. Into or onto a pile or piles of clothing (there being evidence that such piles of clothing were often on the upper surface of the box);

64.2. In one of the corners of the cardboard box, particularly if any combustible detritus (e.g. discarded tissues) had accumulated there, but not if the box was full of, or tightly packed with, CDs in the conventional plastic cases;

64.3. In between the side of the box and other material capable of smouldering combustion, such as clothing or the mattress/bedding (e.g. the 'furry, fuzzy' blanket of which Ms Alexander was particularly fond). I would add that this would be the more so if there were also discarded tissues there.

In my judgment there may well have been a fourth, namely:

64.4. Into the bag which Ms Alexander left on her bed, whose contents included tissues and papers (if it was left open – which, given that she told the police that before leaving she had got some money from her bag, and that her only other bag of which there was any evidence was described as a 'rucksack', seems quite possible).

Further, we simply do not know whether the blanket, duvets etc were sufficiently rucked up, or uneven, towards the foot of the bed for there to have been folds into which a cigarette could have rolled, and obtained the necessary degree of insulation. All that can be said is that, given the thoroughly messy state of the room, and Ms Alexander's untidy nature, this is a real, and not merely speculative or fanciful, possibility.

70. The published evidence of how long is taken before smouldering ignition occurs in such cases (if at all), and before substantial quantities of smoke are first generated, in various types of material seems to be somewhat sparser than might have been expected, and where it exists the range of variance is strikingly high. To summarise that referred to in the material placed before me which appeared to be of the most potential assistance:

65.1. Tests in respect of bedding material reported a transition time from smoulder to flame (which did not occur on every test) of between 51 minutes and 4 hours 10 minutes. The evidence before me contains nothing as to how long prior to flaming ignition copious smoke started to be emitted;

65.2. Smouldering ignition has proved difficult if not impossible to produce in many clothing fabrics. However an (otherwise unpublished) test conducted by Holleyhead (and reported by him in his article 'Ignition of solid materials and furniture by lighted cigarettes. A review.' (1999) 39 Science & Justice 75) involved the insertion of lit king size, filter tipped cigarettes into a fold in clothing material comprising denim jeans and a heavy gauge cotton shirt. In one such test out of (only) two, a flaming fire developed. This occurred after 28 minutes, with large volumes of smoke generated after 10 minutes;

65.3. In reported test results published in 1985 and cited by Holleyhead, a cigarette placed in amongst sheets of vertically arranged corrugated cardboard ignited the same after 15 minutes (Holleyhead does not record how much earlier, if at all, copious smoke started to be emitted). In 'The Ignition Handbook', Babrauskas, times ranging from 12 to 50 minutes for the transition from smoulder to flame in corrugated cardboard are reported. From recollection, Dr Forryan spoke of having, some years ago, assisted a colleague in conducting 3 tests of the potential ignition by one cigarette of cardboard sheets, vertically and tightly packed, in a plastic box. She had to speak from recollection, as the results were not published or otherwise formally recorded. In all 3 cases large quantities of smoke were produced "within minutes". However this evidence is of limited value because she and her colleague were not directing themselves to timings, and because in no case did flaming ignition result (the first two tests ended with self-extinguishment; she and her colleague extinguished the third, because it was producing so much smoke that they became concerned);

65.4. In a series of 300 tests with lit cigarettes in waste paper baskets containing a variety of rubbish reported by Winter (published in 1998), it appears that flaming ignition resulted in 5 cases; in those cases it took between 14 and 18 minutes to occur (Holleyhead does not record how much earlier, if at all, copious smoke started to be emitted). In 'The Ignition Handbook' supra, times ranging from 8 to 40 minutes for the transition from smoulder to flame in paper are reported. In another series of unpublished tests, Holleyhead, having produced charring but not flaming ignition in his laboratory in a number of tests where lit cigarettes were placed in various configurations of large, substantial paper tissues, records a separate independent test carried out by introducing a lit cigarette in amongst crumpled tissues placed in a metal waste paper basket in a draught-free basement. Here flaming ignition occurred in a little under 12 minutes; smoke was "visible" after 30 seconds, remained constant until 7 minutes had elapsed, but then increased over the following 3 minutes to "copious" after 10 minutes (with flames following some 90 seconds later).

71. The bottom end of the brackets quoted in the Ignition Handbook for the transition from smoulder to flame in corrugated cardboard (12 minutes upwards – see paragraph 65.3 above) and in paper (8 minutes upwards - see paragraph 65.4 above), and the c.12 minutes taken in Holleyhead's unpublished test conducted in a draught-free basement (see again paragraph 65.4 above) all contribute to the plausibility of an accidental cause of the fire in the present case, given (a) the time constraints which apply on the facts, but also (b) the emergence at trial of significantly greater evidence of the presence of such paper-based materials in the immediate area of the foot of Ms Alexander's bed than had thitherto been available or fully appreciated.
72. Although the test results as to when copious smoke is produced prior to flaming ignition vary, if reported at all, the experts' first joint statement fairly summarises the position in respect of the literature reported tests involving clothing and paper as being that the same was produced "in the few minutes leading up to the transition from smouldering to flaming ignition".

The cause of the fire

73. It is clear, not least from Ms Alexander's own evidence, that the state of her room on the evening in question made it a serious fire hazard, should anything occur to start one. She was on her own account a messy person, who was lazy about how she discarded things, and not tidying up. The room was a mess. Clothing was strewn over the floor. There was a cardboard box against the foot of her bed, which for at least part of the time had piles of folded clothes on it. The top was formed by 4 flaps folded together; there is no specific evidence as to whether that left any gap in the middle, or whether the resultant surface was entirely flat or tended (at least if and when it had items on top of it) to slope down somewhat towards the middle.
74. Ms Alexander simply discarded tissues which had been used for make-up removal, without using any form of waste paper basket (indeed, there was not one in her room). On the bed was a blanket she variously described as "furry, fuzzy" (to the insurers' investigators) and "soft, stretchy, multicoloured" (in her witness statement in these proceedings), of which she was particularly fond, but which she believed was probably very flammable.
75. The presence of paper based materials on or immediately adjacent to the foot of her bed -? including in particular:

(a) the cardboard box which, although it had been used to carry her CDs in when she moved in, I find did not have any significant number of CDs in it at the time of the fire. They had been moved onto her 2 CD racks, where their remains were found; the photographs show one fire damaged CD case (in particular, E/97), but only one, in the area at the foot of the bed where the box had been; there would have been no reason to sweep up many others but leave one; and there is no evidence of a significant number of fire damaged CD cases in amongst the debris (mainly clothing and smoking related material) which appears to have been swept into the basement level courtyard, and is the subject of a number of photographs – where they would be bulky enough to be readily visible. So far as relevant, I consider it unlikely that there were anything like as many as 300 CDs in Ms Alexander's room when the fire occurred, and that the probability is that most of her CDs were comprised in the 50-75 which were in her 2 CD racks;

(b) discarded tissues which had been used for make-up removal, some of which could have ended up inside the otherwise quite empty box, or in between the end of the bed and the cardboard box, as well as on the floor (although Mr Evans at one stage floated the possibility, there was no evidence that their flammability would have been heightened by the presence of make-up on them); and

(c) the tissues and papers within the bag which she left on her bed ?

was a significant matter for the investigation of this fire, of which the Fire Brigade and (on her initial investigation) Dr Forryan either were not fully aware or did not properly appreciate the significance.

76. Ms Alexander was a naturally careless person in other ways. Mr Smyth's evidence [B/1/4 at paragraph 15] was that she would often go out to the shops, having left the hob on in the kitchen. On the evening of the fire there were additional factors with the potential for contributing to carelessness, namely that she must have been very tired for the reasons I have described, and that by reason of the shouting match with Mr Howe (which he had started, not her) she was agitated and anxious to get out of the house quickly. It would not be surprising if a naturally careless person such as Ms Alexander, in the stressful circumstances in which she found herself that evening, had either carelessly discarded, or accidentally dislodged without noticing, a lit cigarette either when rapidly preparing to leave the house, or perhaps when getting up to challenge Mr Howe when he first shouted at her through the door.
77. By the time Mr Howe opened the basement bedroom door (observed by Mr Eydmann), the fire, and in particular the bed (which was covered by the blanket of which Ms Alexander was particularly fond, and believed probably to be highly flammable) was well alight. As I have said, (a) the seat of the fire was at or very close to the foot of the bed, and (b) despite the appearance, which Mr Howe described to Messrs Wintrip and Soni, of the "top of the bed" being on fire when he first opened the bedroom door, it cannot safely be concluded that the first flame or flames which marked the onset of combustion came from the surface of the bed itself (as opposed to paper-based materials in any of the other positions close to the foot of the bed mentioned in paragraph 64 above).
78. As follows from the findings which I have made as to the sequence of events, the available period of time between when Ms Alexander left her room on her way out of the house (shortly before 8.30pm) and when Mr Howe (having smelt smoke on the ground floor a couple of minutes earlier) observed the glow of the fire around the edges of the concertina door (just after 8.40pm) was sufficient (consistently with the shorter periods cited in the test results summarised in paragraph 66 above) to allow for flaming ignition of cardboard, paper or tissues following a lit cigarette being carelessly discarded or accidentally dislodged as Ms Alexander hurriedly prepared to leave the house. The available time would be slightly longer had this occurred as she first went to challenge Mr Howe's shouting, which is another real possibility. The state of Ms Alexander's room, including in particular the presence of strewn clothing, discarded tissues and what was believed to be a highly flammable blanket covering the bed, was such that, as a matter of common sense, once flaming ignition had occurred, a fire could well have spread very rapidly, and no evidence suggested otherwise.
79. An alternative hypothesis consistent with accident, on which I have not focussed, is combustion of some fabric(s). I find that to be significantly less likely than combustion of some paper-based material(s), because on the basis of the published test results described above there does not appear to have been enough time for the former to have occurred, nor for (in the minutes immediately prior to flaming combustion) copious quantities of smoke to have been emitted in the bedroom while there was nobody inside it (even an anxious and distracted Ms Alexander, hurrying to leave the property, would have noticed copious quantities of smoke). Given those time constraints, this alternative hypothesis appears unlikely to be correct (even though (a) as the experts rightly pointed out in their first joint statement (at paragraph 4.19), the types of clothing materials present are not known, and (b) the test results do not conform to a sufficiently clear pattern that one can entirely exclude it), unless the smoulder had started while Ms Alexander was still eating, but at that stage produced insufficient smoke to be noticed by her; this, as I have indicated, is another possible scenario.
80. The short period between about 8.25 and 8.30pm when Ms Alexander briefly returned to her room after the shouting match with Mr Howe had started, and went to grab a couple of items (which would have been on or very close to her bed) before getting out of the house quickly, is in my judgment the most obvious time at which a smoker who usually used her ashtrays safely enough (even though she often allowed them to become full, and did not empty them regularly) could easily have forgotten that she had been smoking, and either accidentally dislodged a lit cigarette from where she had left it (e.g. part in, part out of an ashtray in the usual way), or else have discarded one more carelessly than usual. Nor would it be surprising if either of those things had occurred a couple of minutes earlier, when Mr Howe first started shouting, and Ms Alexander got up to respond.
81. I have already discussed (in paragraph 64 above) sites sufficiently close to the foot of Ms Alexander's bed to be plausible seats of the fire, in which a lit cigarette could well have lodged and come into contact with cardboard, or paper, and/or tissues, as a result of being carelessly discarded or accidentally dislodged.
82. Simply by way of illustration, in a case where there are a myriad of possible permutations of the facts, and (as I would emphasise) neither the factual nor the expert evidence enables findings of fact as precise as this illustration to be made, one hypothetical but plausible timeline would be:

8.27pm – Ms Alexander dislodges a lit cigarette thitherto resting on the edge of a full or nearly full ashtray, as she grabs her things and prepares to leave the room. It comes to rest in one of the above-mentioned locations, in a confined space and in contact with one or more of cardboard, tissues or papers;

8.34pm – the level of smoke generated, thitherto low, starts to increase;

8.37pm – copious quantities of smoke start to be emitted;

8.39pm – the smell of smoke reaches the sitting room. Flaming ignition occurs, and the fire spreads, quickly igniting the 'furry, fuzzy' blanket of which Ms Alexander was particularly fond;

8.41pm – Mr Howe arrives in the basement, where the smell of smoke is increasing and a red glow from the flames is visible around the edges of the concertina door.

Having run upstairs and shouted to Mr Eydmann until he awoke, Mr Howe returns to the basement, and at 8.43pm opens the concertina door (observed by Mr Eydmann from part way down the open stairs), exposing a fire which is well alight, with flames from the bed reaching the ceiling.

Mr Eydmann runs back upstairs, quickly puts on some outer clothing, and leaves the house with his mobile phone at 8.46pm, dialling 999 as soon as he is outside.

Mr Howe spends a couple of minutes trying to fight the fire by throwing water from a salad bowl or the like (which he fills/refills with water in the kitchen) onto the foot of the bed, before giving up and leaving the property himself. He exits within a minute or so of Mr Eydmann, while Mr Eydmann is still on the phone to the emergency services.

83. Contrary to the basis on which witnesses were interviewed in particular by the Defendant's investigators in March 2008, the true circumstances as I have found them to be are that this is not a case where the fire must have been started deliberately, and the only question is by whom (cf Re B per Lord Hoffmann at [15] and Baroness Hale at [72]-[73]).
84. I remind myself that one aspect of the results of the scientific tests mentioned above is that they show that leaving a lit cigarette in contact with paper-based material or fabric with at least some degree of insulation will by no means always result in combustion. In one of the tests it appears to have occurred only 5 times in 300, although in another it occurred 1 time out of 2. On the other hand, deliberate lighting with a naked flame, such as that produced by a match or a cigarette lighter, would easily bring about an almost immediate fire. As Mr Evans neatly summarised what is a point for the Defendant, it is easier to start a fire by applying a flame than by discarding a lit cigarette.
85. Subject to various points of more tangential relevance with which I have already dealt, I find there to be no cogent reasons beyond the bare facts of the shouting match or argument which had occurred that evening, that Mr Howe was "merry", and that he had once kicked out the (wobbly) spindles supporting a bannister while in a tantrum, for concluding that he deliberately started the fire, and that the preponderance of the available evidence, as well as a common sense view of inherent probabilities, is against such a conclusion. As Mr Evans put what really comes to the same point, for Mr Howe to have done so would have been an act of extraordinary malice and recklessness. Furthermore, it would have taken a remarkably 'cool customer' to set fire to Ms Alexander's bed while reheating a pizza, to leave the half intended for the Claimant on a plate in the kitchen (where the fire brigade subsequently photographed it), and then to sit and eat his half of the pizza one floor above the fire which he had started, and I do not believe that Mr Howe would have acted as coolly as that, had he just started a fire in the room directly below.
86. Looking at the matter in the round, I find no "other compelling evidence" (cp per Lord Hoffmann) in this case to displace the starting point of a common sense view of inherent probabilities. Accident, and, more specifically, the careless discarding or accidental dislodging of a lit cigarette by Ms Alexander in her bedroom, when very tired, agitated and anxious to leave the house in a hurry following the shouting match, or simply very tired and agitated by Mr Howe's initial shouting, is a real or plausible explanation for the start of this fire. Not only is this "not inconsistent" with the evidence (cp per Stuart-Smith LJ), but it is supported by what I judge to be a preponderance of it.

Conclusion

87. For the above reasons, my conclusions are that:

82.1. The defendant insurers have not satisfied me that the fire was started by Mr Howe; and

82.2. It is more likely than not, and I therefore find on the balance of probabilities, that the fire was started accidentally, by a lit cigarette carelessly discarded or accidentally dislodged by Ms Alexander.

88. I therefore find for the Claimant on liability.

[END]