Synopsis of the Kevin Sweeney case

Richard Gill, Dept. of Mathematics, Leiden University, 20 October 2007

Kevin Sweeney left Steensel (near Eindhoven) soon after 02:00 a.m., in the very early hours of Monday 17 July 1995, arriving at his old house at Bousval (near Brussels) around 04:00 a.m., where his and his wife Suzanne Davies’ three young children (two of his by a former marriage, one by a former relationship of hers) were being cared for by ‘weekend nanny’ McDaid. One of his children was sick and indeed the next day a mild pneumonia was diagnosed at the local hospital. The couple had been decorating their new house at Steensel and Suzanne stayed behind there after Kevin’s departure to Bousval.

At 02:38 an electrical fault in the burglar alarm system was registered at the security service centre. In response to this, between 02:47 and 03:00, two policemen and the housekeeper Smets and her husband Mooren walked all around the house, right below the balcony windows of Suzanne Davies’ bedroom, not noticing anything. Smets tried various doors with her keys but could not get in — they were all bolted from the inside. Everything seemed to be in order and they left. Seven months after the event, Mooren testified that he thought he had smelt smoke at the time. He had not said anything about it, nor had anyone else, neither at the time itself nor in the subsequent hours and days when a fire and a murder investigation were under way. However it was vital for the prosecution case that the fire was already started at this time, since they needed other allegations made by the witness (McDaid) who saw Sweeney arrive in Brussels, and she had already said when this was. By the way, the burglar alarm does not seem to have been triggered by the fire—it was in fact not even damaged by the (later) fire, despite claims of the police that it had melted. The house had recently been rewired and the alarm recently installed, and there were a number of problems with the electrics.

At about 03:45 a fire was reported: flickering light from flames in the bedroom was visible from the cafe across the road. Firemen arrived at 03:55. Note that no fire was reported an hour earlier, when the prosecution claims that it was raging. The firemen had to force entry. Suzanne Davies was unconscious and very near death. The ambulancemen determined a cardiac trace and attempted resuscitation. She was pronounced dead at 04:37. Cause of death was later determined to be carbon monoxide poisoning. She had second degree burns only.  She had used alcohol that night (in modest quantity). The bedroom was full of smoke and very hot; heat had broken the balcony windows, and there were flames at the foot of the bed. The fire had only consumed a small part of the bed and furnishings. Suzanne had apparently fled to the adjoining dressing-room where she was found crouched on the floor. Small house fires start rapidly and can lead to badly impaired reactions because of oxygen shortage within a short time; death can follow from asphyxiation by fumes and carbon monoxide poisoning in less than five minutes.  But Suzanne had been awoken enough by the heat to move to the dressing-room, though the fact she went there rather than leaving the room altogether suggest that her reactions were seriously impaired by oxygen depletion.  Since she appeared to have been still just alive when found, it is reasonable to suppose that the fire started roughly between 3:15 and 3:30. This matches the evidence of the police and the housekeeper and her husband who noticed nothing unusual, walking around the house in response to a burglar alarm, at 03:00. 

Suzanne had bought cigarettes and a lighter with her credit card in Steensel the Friday before, during a trip (alone) to the new house to start unpacking stuff, which the couple had brought together there the day before. Remains of a lighter and cigarettes confirm that she might well have been smoking in bed. The remains of cigarette buts in the bathroom washbasin were photographed by the insurance investigator though subsequently destroyed by the police, who claimed (on the basis of a statement of her mother) that Suzanne never smoked. Her brother first stated that she was a smoker, later he withdrew this statement. The main proof that she did not smoke, according to the police, was that there was no ash-tray in her bedroom.

A large proportion of house fires are caused by smoking in bed, and a large proportion of these fires result in fatalities [see CBS statistics].

No traces of fuel accelerants were found in Suzanne’s body. The first police on the scene of the fire later reported what they called “a strong smell of turpentine”. Such a smell, and similar GC/MS (gas chromatography/mass spectroscopy) traces, can be caused by burning of pine-wood (in furniture) and many synthetic materials (carpeting, bedding, matress, adhesives, plastic containers, nail varnish, lighter fuel).  Suzanne had been decorating and cleaning after painting work, and cleaning the bedroom carpet, which had been affixed to the concrete floor with adhesives. According to the police forensic laboratory, traces of “crude oil distillates” were “possibly” present in samples allegedly taken from the carpet. At the request of the prosecution, the word “possibly” was later replaced by “probably”. 

There is a lot of confusion around the word terpentine which means something different in English from what it means in Dutch. Terpentine (English-American usage) is not an oil distillate at all, but is manufactured from wood resin. What you buy in Holland under the name “terpentine” however, is an oil-distilate. Later, the prosecution switched their story from terpentine to petrol, alleging that a jerry-can of petroleum was used to start the fire, while at the same time the terpentine-like smell also remained as evidence that the fire was caused by arson.

Certainly no samples were taken from the carpet while it was still in situ, and if at all, samples were only taken weeks later. In the mean time, many petrol derivates would evaporate; contamination could have occured through firemen and police and insurance investigators walking over the carpet and moving objects about, during the fire and in the subsequent days. In fact it is known that the carpet was first cleaned in situ, then later taken to the police station, and samples were only cut from for lab analysis later still. Chemical cleaning of a carpet adds new petrol derivatives to it. It cannot be ruled out that the samples were in fact deliberately contaminated by the police investigators; it is known for sure that they falsified other evidence. Laboratory tests did not reveal what the substances found in the carpet were, which suggests that the traces were infinitesimal and the laboratory results inconclusive. Nail-varnish and lighter-fuel certainly got on to the carpet, since the remains of their containers were found on it. Large parts of the carpet, and a newspaper (the weekend edition of the Telegraph) lying on the carpet, were left unburnt, which could hardly have been the case if 8 litres of petrol had been poured over it and ignited.

Experiments done by TNO to reproduce the effects of the fire while initiating it by applying a naked flame to 8 litres of fuel, were a series of failures, though spectacular ones. In order to obtain anything like the actually observed fire damage, ventilation and total air space and ambient temperature needed to be drastistically reduced; humidity increased; less flamable fabrics and bedding materials, and less of them, were used. Windows needed to be strengthened since they otherwise shattered within seconds from intense heat. The length, technical content, and lack of clarity of the TNO report allowed its compiler to cover-up the actual message of the experiments and let the conclusion appear to support the prosecution. According to the prosecution, the fire was supposed to have been burning from soon after 02:00, when Sweeney was supposed to have left the house, apparently having bolted all the doors on the inside, behind him. Suzanne was supposed to have died soon after this time, in contradiction to the evidence of ambulance personel and doctors who attempted to reanimate her at 04:30.

Experiments done by the police, and supposed to prove that the fire could not have been caused by a smouldering cigarette on bed linen or clothing, were a farce. Applying a cigarette 8 times to bed linen without it catching fire, does not prove that fires are not caused by smoking in bed. Actual statistics of actual fires and fire-deaths were never consulted, and the court refused to admit them as evidence.

Despite the utter failure of the prosecution to support their case it was still claimed by the prosecution, and accepted by the court of appeal, that the fire was deliberately started by applying a naked flame, and that the accused had the day before rigged the bedroom with up to 8 liters of fuel, apparently without the victim being aware or its presence (the smell?) when she went to bed. Presumably the prosecution has in mind that the victim was drugged before the fire was started, but this is totally incompatible with all the evidence, which shows that Kevin Sweeney left the house long before the fire started, and that Suzanne Davies had only used a moderate amount of alcohol that evening, and had much later been sufficiently awakened by heat and smoke that she fled her bed, but unable to take a sensible route out of the room. Her behaviour and the foetus-like stance she was found in, crouched on the floor in the adjoining dressing-room, are very common in these circumstances.

The prosecution also dismisses without serious argument the possibility that the fire was caused by an electrical fault (a lot of rewiring had recently been done in the house), or that it was a parasuicidal gesture (Suzanne Davies certainly had a lot of serious problems in her life, especially at that time, and a long history of mental imbalance). However, in view of the natural explanation that the fire was caused by smoking in bed, we do not pursue these possibilities either.

The prosecution alleges that the fire was a “perfect arson” carried out by a manipulative, evil and highly intelligent killer (as evidence the court adduces his IQ of 134 and his consistent denial of guilt), who could simulate a charming personality while completely hiding his true, cold, self. Further evidence against the character of the accused has been adduced from witnesses who were led by the police to perjure themselves and who had financial and/or emotional motives to lie (Suzanne’s brother, and mother, respectively). Evidence for his motive, consisting of financial details of his and his wife’s companies, has evaporated along with the terpentine, and he was indeed not convicted of fraud. In fact Suzanne’s unexpected and accidental death caused him financial loss, not gain, since he stood guarantee for her various business ventures, which were not doing very well. He was well aware beforehand of her financial situation.