SUPREME COURT OF QUEENSLAND

[1]  KEANE JA: On 22 October 2009 the Court concluded that the appeal in this matter should be allowed, the appellant’s convictions should be set aside and verdicts of acquittal entered.  I agree with the reasons for the making of those orders prepared by Fraser JA.

[2]  MUIR JA: I agree with the reasons of Fraser JA.

[3]  FRASER JA: After an eight day trial before a jury in the District Court, on 17 April 2009 the appellant was convicted of arson, fraud and attempted fraud.  The appellant was sentenced to four years imprisonment on the count of arson and to concurrent sentences of two years imprisonment on the remaining counts.

[4] The appellant has appealed against his convictions on the following grounds:

"1.That in all of the circumstances no reasonable jury would have found the appellant guilty of the offences;

2.That the improper submission by the Crown Prosecutor that it could be inferred that a Crown witness was not of bad character because of the absence of cross examination on that point was not properly cured by judicial direction in the circumstances of this trial;

3.That the jury [was] insufficiently directed that disproof of the appellant’s explanations should not act as a reversal of the onus of proof upon the Prosecution."

[5] After hearing argument on 22 October 2009 the Court allowed the appeal, set aside the convictions, and entered verdicts of acquittal.  The Court indicated that the reasons for those orders would be delivered in due course.  These are my reasons for agreeing that those orders should be made.

[6] It is necessary first to outline the Crown case. 

Outline of the Crown Case

[7] The Crown case was that the appellant had deliberately set fire to his house with the intention of defrauding the insurer who had issued a policy to his wife (count 1), he had defrauded the insurer of an emergency payment of $5,000 made in response to his wife’s claim on the policy (count 2), and he had attempted to defraud the insurer in relation to his wife’s claim on the insurance policy (count 3). 

[8] The evidence demonstrated that at about noon on Monday, 15 January 2007 the upstairs area of the appellant’s house was substantially damaged by fire.  The fire brigade was promptly called by a neighbour and fire crews quickly attended (there was a nearby fire station) and extinguished the fire. 

[9] The appellant did not call or give evidence at the trial, but many witnesses gave evidence of his version of events.  Very shortly after the fire the appellant gave an account of what had happened to police (Senior Constable Hallmark, Constable Michelle Baker and Detective Sheehan), two neighbours (Ms Woods and Ms Hole), a friend (Ms Fahey), and an officer of the Queensland Fire and Rescue Service who attended at the fire (Mr Heppell).  The appellant's account was materially consistent throughout.  The appellant and his wife had been renovating their house for many months, with a view to selling the renovated house.  On the day of the fire the appellant was painting part of the main bedroom, which was adjacent to the lounge room.  Drop sheets were in place in the upstairs section of his house.  The appellant told Detective Sheehan for example, that he used "old towels and sheets" as drop sheets.  The appellant was smoking a cigarette. (On his account he seems to have been a chain smoker.)  When the telephone rang he took some time trying to turn off his stereo, which was playing very loudly, before going downstairs to answer the phone.  After that delay, and in his haste to get to the telephone so as not to miss the call, he kicked over a tray of paint thinners.  It is important to note here that whilst most of the witnesses attributed to the appellant the use of the word “thinners” in that context, Mr Heppell was more specific: he said that the appellant told him that he "kicked the tray of white spirits – he called it – kicked the tray over as he sort of ran out of the room."  In later passages in Mr Heppell’s evidence he used the terms “white spirits” and “thinners” interchangeably.

[10]  The appellant told Mr Heppell that although he didn't recall dropping his cigarette before he went to answer the phone, he “could have knocked it or brushed it or dropped."  The other witnesses reported that the appellant communicated to them that he’d kicked over a tin or container of thinners and dropped his cigarette.  The appellant told Detective Sheehan, for example, that he "tripped on a drop sheet in the lounge room … knocked a container of thinners over … and, dropped my smoke … I didn't even give it a thought …"

[11]  The appellant then answered the telephone.  The call was from his wife.  She asked him about the dimensions of some downstairs bay windows for curtains which Mrs Dyke was to buy.  The appellant measured the area and gave the measurements to his wife.  After the conversation finished the appellant started to walk back up the stairs.  He was then confronted by smoke and flames which blocked his approach.

[12]  Mr Heppell gave evidence that he had completed a fire scene examination course over a nine month period.  He had about 38 years experience with the Queensland Fire and Rescue Service and its predecessor organisations.  He examined the scene after the fire had been extinguished and he made some enquiries of other experts about the properties of “white spirits”.  Mr Heppell could not exclude deliberate human action as the cause of the fire, but he expressed the opinions that what he observed was consistent with the account given by the appellant and that "a cigarette igniting paint thinners, for example" was a plausible explanation of the fire.  Mr Heppell’s reference to paint thinners was presumably intended to mean white spirits.  In re-examination he said that he had not himself come across a fire he thought had been started by a cigarette “in thinners or white spirits”, but he did not resile from the opinions favourable to the appellant he had expressed in his evidence-in-chief.

[13]  A few days after the fire an insurance investigator, Mr Gannon, conducted and recorded an interview with the appellant for the expressed purpose of investigating whether the insurer should meet Mrs Dyke's claim on the insurance policy.  The transcript shows that in response to Mr Gannon’s questions the appellant first said that he didn't really know how the fire started and he then substantially repeated the account I have summarised.  The appellant mentioned some additional details: he had moved the furniture in the lounge room into the middle of the floor; he had a drop sheet in between the coffee table and the TV stand, with the TV sitting on the floor; he had a plastic container cut in half with about an inch of “thinner” in it; there were rags and "all that sort of shit" in the room (apparently used in connection with the painting); the windows in his lounge were open; and he had a fan ventilating the rooms.  (A burnt out pedestal fan was discovered in the lounge room after the fire was extinguished).  When Mr Gannon pressed the appellant to say what caused the fire he said:  "…evidentially … the thinners, the paint, the cigarette, the fan.  But how it actually ignited I've got no idea … it never registered that I dropped the smoke, cause I've always got the smoke." 

[14]  The appellant told Mr Gannon that after he was confronted by the fire he went outside with his phone but a neighbour (apparently Ms Woods) told him that she had seen the smoke and rung the fire brigade.  The appellant told Mr Gannon that the flames were probably extending 10 to 15 feet out of the windows within a few minutes of the fire being discovered, so that there was nothing he could do.  Ms Woods and her sister, Ms Fitch, gave similar evidence: Ms Woods became aware of smoke pouring out of windows in the Dykes’ house and asked her sister to dial 000.  Ms Fitch described the fire she saw as "black smoke everywhere, flames everywhere". 

[15]  The appellant told Mr Gannon that he had tried to phone his wife but couldn't get through.  He said that Ms Hole's boyfriend, Mr Grant, tried to bring a hose from her house next door to put the flames out but the hose didn't even reach the fence. Ms Fitch and Ms Woods could not recall seeing the appellant with Mr Grant trying to get the hose to put the fire out, but neither denied that this had occurred.

[16]  The evidence I have so far summarised, particularly Mr Heppell's evidence, strongly suggested that the appellant was innocent of the offences charged against him.  The Crown case that the appellant had deliberately lit the fire with the intention of defrauding the insurer was based upon other evidence which, so the respondent’s counsel submitted in this appeal, was capable of being accepted by the jury as establishing the three planks of the Crown case:

(a) The appellant's account of how the fire started was improbable. Counsel for the appellant referred to evidence which was submitted to show that the appellant had no emotional reaction to the fire, and to expert evidence by Mr Smith.

(b) The appellant had a motive to burn his house down because he and his wife were in a desperate financial position, exacerbated by their excessive gambling. This proposition was based mainly upon expert evidence from Ms McKinnon, a chartered accountant employed by the Queensland Police Service. There was also the evidence of statements by the appellant to Mr Gannon that the appellant was in difficult financial circumstances, and documentary evidence that the local council had issued a notice to the appellant which asserted that he had not obtained the necessary approval for some of his renovations.

(c) Shortly before the fire the appellant said that he intended to set fire to his house. This plank of the case was based upon the evidence of Ms Hole's friend, Ms Alexander. 

[17]  I will refer to the evidence which is said to support those propositions in the course of discussing the appellant’s first ground of appeal.

Ground 1:  No reasonable jury would have found the appellant guilty of the offences

[18]  This ground of appeal raises the issue under s 668E(1) of the Criminal Code 1899 (Qld) whether the verdict of the jury should be set aside on the ground that it is unreasonable or cannot be supported having regard to the evidence.  In MFA v The Queen[1]  the High Court endorsed the statement in the joint judgment in M v The Queen[2] that under this ground of appeal the question is whether, notwithstanding as a matter of law there is evidence to sustain a verdict, upon the whole of that evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty.  This ground of appeal is "designed to afford a mechanism against the prospect that our community and its courts continue to regard as intolerable, namely that an innocent person has been wrongly convicted upon unreasonable and unsupported evidence and has thereby suffered a miscarriage of justice …".[3]

[19]  I will refer to the evidence under headings which correspond with the three planks of the Crown case.

Was the appellant's account of how the fire started improbable?

The appellant's calm demeanour

[20]  In support of the proposition that the appellant did not have the emotional reaction to the fire that one would expect of an innocent man, counsel for the appellant referred to some evidence given by Ms Woods.  She said that after she had seen smoke coming out of a window of the appellant's house and asked her sister to ring 000, she ran next door.  She saw the appellant, who was holding a cordless phone walking around the side of his house.  She thought he looked “just like a normal person walking around the side of the house.”  She said: "Do you realise your house is on fire?"  The appellant said "yeah".  Ms Woods asked him if he had called the "firies", and he said he hadn't.  Ms Woods then said, "Well, don't worry because we have done it."  Ms Woods then questioned the appellant about what had happened and he gave a brief account along the lines I have detailed.

[21]  This evidence was apparently meant to convey, so it was opened, that the appellant maintained a calm demeanour.  Whether or not that was so, the evidence was consistent with the appellant having been in a state of confusion or shock which would be unsurprising in a man confronted by an unexpected conflagration in his home.  Ms Hole said in her statement that the appellant “looked like he was in some kind of shock” and Ms Fahey gave evidence that the appellant was “devastated” by the fire.  Accepting that it was for the jury to assess Ms Woods’ evidence, it provided no worthwhile support for a contrary view.

Fire reconstruction evidence

[22]  Of much more significance was the opinion evidence given by Mr Smith.  The prosecutor opened Mr Smith’s evidence as establishing that the fire “in effect could not have started in the way described by Mr Dyke” and that “the fire could not have started by dropping a lit cigarette into thinners or into a drop sheet that might have been soaked with thinners that had been knocked over.”  Mr Smith's evidence did not live up to that opening but he did express opinions that tests he described (video recordings of which were played to the jury) and literature that he had read indicated that, though a fire might possibly have started accidentally in the scenario described by the appellant, that was unlikely.  The video recording of the tests showed men putting lit cigarettes and knocking ash off lit cigarettes onto material, some of which was wet by thinners and some of which was dry, and onto flooring, in a mocked-up room.  In no case did a fire result in that way, but a lit match applied to the material resulted in instant ignition.

[23]  Mr Smith was qualified to give expert evidence about the cause of the fire.  He gave evidence that he was a crime scene examiner, which included the examination of the cause and origin of fires; that he had a Bachelors Degree in Chemistry and a Masters Degree in Forensic Science; and that he held a Diploma of Fire Investigation, which he obtained after completing a University course involving the cause and origin of fires and assessment by written and practical tests.  He said that he was a member of the Queensland Association of Fire Investigators and the International Association of Arson Investigations, the latter being a "world-wide organisation".  The jury might well have been impressed by those qualifications of Mr Smith as an expert in the relevant field and accordingly placed particular weight upon his evidence.  That is particularly so because the trial judge did not give the conventional directions to the jury about expert evidence.[4]  In particular, the jury was not directed that an expert's opinion based on what the expert witness has learned of the facts may be of little value if those facts have not been established to the jury's satisfaction. 

[24]  Mr Smith gave evidence that he went to the appellant's house on 16 May 2007, some five months after the fire.  He and others "excavated" or "overhauled" the lounge room and found charring to the "real flooring" underneath the "floating flooring" in the area of the lounge room towards the main bedroom.  Mr Smith expressed the opinion that the fire probably originated in that "general area".  This is consistent with the appellant's account and with Mr Heppell's similar evidence.  MrSmith expressed the opinions that the fire probably progressed from the lounge room to the adjacent rooms in the upper story of the house; there was only one seat or point of origin of the fire.  He found furniture which had been effectively destroyed, including a television, entertainment units, and other items.  He found a tin that appeared to be consistent with a tin used for flammable fluids or ignitable liquids, of which paint thinners were an example.  He also found other tins of paint thinners in the lower story of the house.  He could not judge what had been in the tin he found upstairs.  No ignitable liquid residues were found in any of the samples he took.

[25]  Mr Smith could not express a view as to whether electricity potentially contributed to the fire, but when asked whether he formed an opinion about the probable cause of the fire, Mr Smith responded:

"Given that there was some time between my examination and the fire occurring, all the indicators I saw, the lack of electrical items within what I considered to be the area of origin, or – and the lack of evidence of electrical activity contributing to the fire, I formed the opinion that this was most likely – most likely to be a fire resulting from human involvement."

[26]  That answer was expressed in qualified and equivocal terms.  So too was the evidence of an electrician, Mr Adler, that “…it’s very hard to pick…but from what I saw…there was nothing there [in the area of origin of the fire] that would indicate to me that it had been caused by electrical [sic].” 

[27]  Video recordings of the “fire tests” were then tendered in evidence and played to the jury during Mr Smith’s evidence.

[28]  So much of the videos as purported to demonstrate that the fire might have been started by using a lit match were arguably inadmissible on the ground that they amounted to an out of court demonstration of an alleged crime which could not conceivably replicate what had in fact occurred,[5] but there was no objection to that evidence and nor was its admissibility the subject of any ground of appeal.  Rather, the appellant’s submissions focussed upon the experiments with burning cigarettes and Mr Smith’s opinion evidence that the fire was probably not started by a burning cigarette.

[29]  The respondent contended that the evidence of these experiments was admissible because, in the words of the Court of Appeal of New Zealand (Richardson, Casey, Hardie Boys, McKay and Tompkins JJ) in R v Harbour:[6]

“This was not an attempt to reconstruct events but a scientifically conducted experiment designed to establish not a fact but a probability”.

[30]  That court went on to cite authority[7] which emphasised the importance of ensuring that such an experiment was not likely to mislead the members of the jury who might attach exaggerated significance to the test, that there must be valid points of similarity between the experimental conditions and the actual conditions so as to render the evidence admissible, and that if the evidence was admissible any points of difference went to weight. 

[31]  In a recent decision which neither party cited, Birks v Western Australia,[8] the Western Australian Supreme Court of Appeal concluded that a video recording of an experiment designed to show that a burning cigarette was unlikely to start a fire was admissible in the appellant’s trial for arson.  Buss JA referred to authorities which supported that conclusion in the following passage:[9]

“[48] In R v Ireland (No 2) [1971] SASR 6, upon the trial of the appellant for murder, the prosecution tendered evidence of tests conducted by a police officer which involved the rolling of a body over the floor of a room, and the walking of a specified distance to ascertain the time taken in the walk. The Full Court of the Supreme Court of South Australia held that evidence of the results of the tests was admissible. Bray CJ, Hogarth and Wells JJ said, at 14 — 15:

"In practice, evidence of experiments, more often than not, is offered by experts and its purpose is frequently to confirm the opinions of those experts arrived at by an examination of real evidence that has been found and collected at the scene of the trial. But, of course, the use of experiments for the purpose of enlightening a jury is not limited to those so conducted. It seems to us that, given conditions for the experiment sufficiently similar to the conditions in which the act or event under consideration must have been done or occurred, an experiment carefully performed and conscientiously recorded and reported may frequently be of great assistance to the jury in its deliberations. Some experiments can be carried out only by experts, occasionally by experts with high qualifications and advanced skills, but others can be carried out by ordinary laymen by the application of common sense and the employment of such tools, materials and devices that are readily at hand. As with opinion evidence, much will depend on the degree of knowledge and skill required, and the degree of precision claimed for the results of the experiment. If a layman witness is seen by a trial judge to have blundered into arcane fields of science or learning with inadequate training, or to be claiming that his experiment produced results whose accuracy is plainly exaggerated, he will, no doubt, either recommend to the jury that they disregard the evidence or, in his discretion, exclude it. But there is a wide range of experiments whose purposes and execution are so easy to follow and to assess, and whose results are so plainly useful, that a jury can legitimately make reference to them to assist in ascertaining the truth, even though they are conscious that their results are only approximate."

Their Honours approved the following statement in the judgment of Edwards J in Shepherd v State (1931) 51 Okl Cr 209; 300 Pac 421:

"The general rule as to the admissibility of the result of experiments is, if the evidence would tend to enlighten the jury and to enable them to more intelligently consider the issues presented and arrive at the truth, it is admissible. The experiment should be under circumstances similar to those prevailing at the time of the occurrence involved in the controversy. They need not be identical, but a reasonable or substantial similarity is sufficient. Several Courts have held that the lack of identity of circumstances affects only the weight and not the competency of the evidence provided there is a degree of similarity which will assist the jury."

[49] Ireland (No 2) was followed in Thompson v R (1986) 13 FCR 165. In that case, the appellant had been convicted after a trial by jury in the Supreme Court of the Australian Capital Territory of murdering two sisters whose remains had been found in a burnt-out motor vehicle that had crashed against a tree and caught fire near the Monaro Highway in the Australian Capital Territory. The prosecution's case was that the appellant had murdered the sisters and feigned an accident resulting in the fire. The appellant gave an explanation to the police to the effect that he was driving on the Monaro Highway when he was dazzled by the lights of an oncoming vehicle. He said he drove his vehicle from the highway, without braking, and collided with the tree at a speed of 40 — 45 miles per hour. He also said his vehicle then burst into flames, the fire beginning in the engine compartment and spreading to the inside of the vehicle. According to the appellant, the flames engulfed the vehicle so rapidly that although he was able to extricate himself, it was impossible for him to assist the two sisters, who were his passengers. The police conducted experiments, which were referred to as the crash experiment and the fire experiments, to test the appellant's account of the events in question. The Full Court of the Federal Court held that the trial Judge had been correct to admit evidence of the experiments, and that the divergences between the circumstances of the experiments, on the one hand, and the circumstances of the appellant's version of the facts, on the other, went to weight rather than to admissibility. Forster, Everett and Miles JJ said, relevantly, at 173, in relation to the crash experiment:

"It seems to us that the difference in the loads carried by the two vehicles must have worked in the appellant's favour so that we may safely ignore that difference whatever it may have been.

In the event, it seems to us that the circumstances of the two collisions were sufficiently similar to make the evidence of the test results relevant. The weight to be put on that evidence was, of course, a matter for the jury."

Also see R v Baker [1989] 3 NZLR 635, especially in relation to the distinction between an experiment and a reconstruction.

[50] In R v Neilan [1992] 1 VR 57, the applicant was charged with the murder of his wife. His defence was that unknown intruders had committed the murder. The intruders had broken into the family home, assaulted him, placed him in the boot of a motor vehicle and shot his wife. During the trial, evidence was received of tests conducted by the police to determine whether the applicant could have seen the intruders approaching his home, as he claimed. The applicant was convicted. The Court of Criminal Appeal of Victoria considered the principles governing the admissibility of evidence of experiments and tests. Young CJ, Brooking and Marks JJ approved, at 74 — 75, the following observations in 29 Am. Jur. 2d, § 824:

"One desiring to make an experiment or test in court or to introduce evidence of an experiment or test made out of court should first show that the experiment or test is to be made or was made, as the case may be, under conditions and circumstances similar to those prevailing at the time of the occurrence involved in the controversy; otherwise, the courts will not, as a general rule, permit the making of the experiments or tests or the introduction of evidence thereof. It is clear, however, that the conditions need not be identical with those existing at the time of the occurrence, but it is sufficient if there is a substantial similarity of conditions. Minor variations in the essential conditions go to the weight, rather than to the admissibility, of the evidence.

There is no precise test or gauge to determine when the requirement of substantial similarity has been satisfied. This depends largely upon the purpose for which such evidence is to be introduced. Speaking generally, however, the measure of permissible variation of the conditions of the experiment or test from those of the occurrence is measured by whether such variation is liable to confuse or mislead the jury. When the conditions are so dissimilar from those of the occurrence in question as to tend to confuse or mislead the jury, the evidence of an experiment or test should be rejected. The question of similarity is one that lies within the sound discretion of the trial court, to be decided in the light of all the surrounding facts and circumstances."

Their Honours held, at 75, that in the case before them it had not been shown that there was such a similarity of relevant conditions on the night in question and at the time of the tests to render evidence of those tests admissible. They then said:

"The evidence of the tests was wrongly admitted. But after careful consideration we are satisfied that no substantial miscarriage of justice has occurred in consequence of the reception of this inadmissible evidence. It should at the outset be noted that the case is not one of the erroneous reception in evidence of material that is intrinsically highly prejudicial, such as evidence of bad character or evidence of the kind considered in Maric v R (1978) 52 ALJR 631. The evidence in question was no more than evidence of tests conducted with a view to ascertaining what could be seen in and from the house having regard to the state of the lighting, although it is true, as Mr Winneke submitted, that the evidence was led for the purpose of helping to persuade the jury that the accounts which the accused had given about the killing were not to be believed. The evidence was not given by experts of experiments carried out by them in a field within their expertise: it was evidence of lay persons about what they could see in or from a dark room and looking from the doorway of a dark house. The evidence had about it none of the possibly deceptive persuasiveness which might be thought to attach to the use of scientific instruments under conditions which did not reproduce those of the night in question. Moreover, because the admissibility of evidence of this kind is governed by a principle based on common sense as opposed to depending upon some rule of law that a jury might have difficulty in understanding, and because, similarity being a matter of degree, there is no hard and fast line between dissimilarity which merely affects (and may affect very seriously) weight, and on the other hand dissimilarity which leads to irrelevancy, the jury were in a position themselves to perceive the weakness of the tests. This weakness was well and truly exposed in cross-examination of the Crown witnesses concerned and was made much of by counsel for the defence in his final address. The Crown, without formally abandoning reliance on the tests, appeared in its final address to set no store by them. The unsatisfactory nature of the tests being an obvious matter which had been thoroughly exposed the jury, we do not think that the jury would have placed any weight on them. Moreover, the Crown case against the applicant was a very strong one."

The Court therefore held that in the circumstances there was no miscarriage of justice. The application for leave to appeal was dismissed.

[51] In the present case, there were differences between the "demonstration burn" and the fire which occurred in the early morning of 7 November 2003, relevantly, as follows:

(a) the materials from which the demonstration room was constructed were different from donga C25 in that the demonstration room comprised metal sheets for walls, whereas the donga had timber panelling;

(b) the demonstration room was a three-walled room with no front wall, whereas donga C25 was an enclosed room with its door partly open at the relevant time;

(c) the bed in the demonstration room had a steel frame with a timber headboard, whereas the bed in donga C25 had a steel frame and no wooden headboard;

(d) although the bedding used in the "demonstration burn" was obtained from Ms Tempra and was the kind of bedding which had been used in donga C25, the arrangement of the bedding in donga C25 at the relevant time was unknown, and the arrangement of the bedding therefore could not be replicated; and

(e) the weather conditions were different.

[52] The evidence of the "demonstration burn" was led by the respondent to establish that the appellant had lied in the interviews in relation to his having thrown a lit cigarette into donga C25. The respondent's case was that although the appellant's admissions in the interviews of his involvement with the fire were true, his admissions as to the means by which he started or thought he started the fire were false. According to the respondent, the appellant had entered donga C25, had deliberately set fire to the bedding (most likely with a cigarette lighter) and had then returned to his own donga and smoked a cigarette until the fire was discovered by Mr Higgins and the alarm raised.

[53] In my opinion, the circumstances of the "demonstration burn" were sufficiently similar to those of the fire in the early morning of 7 November 2003 to make the "demonstration burn" relevant to and probative of the facts which the respondent sought to prove by its tender.

[54] The differences between the circumstances of the "demonstration burn" and those of the fire in the early morning of 7 November 2003 did not have a tendency to confuse or mislead the jury. The dissimilarities in question affected the weight to be given to the evidence of the "demonstration burn", but did not render that evidence irrelevant or without any probative force, and therefore did not make it inadmissible. The differences between the fire in the early morning of 7 November 2003 and the subsequent experiment were explored in detail at the trial. Some of those differences made the creation of a fire in the course of the "demonstration burn" slightly more likely, and others made it slightly less likely. Significantly, however, the kind of bedding used in the experiment was identical to the kind of bedding in donga C25 in the early morning of 7 November 2003. The jury were in a position adequately to assess the differences and to determine the weight to be given to the evidence of the "demonstration burn", in the context of the purpose for which it was led.”

[32]  I accept that evidence of scientific tests designed to establish the likelihood of a particular event may be admissible, but the evidence in Birks v Western Australia did not include contrary opinion evidence (such as that given by Mr Heppell) and nor was it affected by deficiencies of the significance which detracted from the force of Mr Smith’s evidence.

[33]  Mr Smith explained his testing in these words:

"The purpose of this is to determine the - try to recreate the scenario to the best of our ability. Obviously, it's virtually impossible to absolutely recreate a situation. One, I wasn't there on the time it happened. We're limited by the resources we could obtain. Obviously, we couldn't obtain the exact furniture that was there. We couldn't recreate the room as it was. So we did the best we could with a shipping container and some material we purchased. We also couldn't - you know, we did this in the middle of the year, when the fire occurred in the hottest part of the year. So we're limited to - just by the fact this is the time we had available and the resources we could obtain. But as best as we could we tried to recreate the scenario as it was posited to me to see if we could achieve an ignition."

[34]  The prosecutor referred to Mr Smith's acknowledgement that there were differences between what he was able to do and the testing and what would actually have been happening when the fire started and asked Mr Smith whether that would result in any "significant difference in the results that you achieved".  Mr Smith responded: "I don't believe so, no." 

[35]  Examples of the difficulty thrown up by this answer emerged in cross-examination. 

(a) Mr Smith had used a cotton sheet as a drop sheet in his experiment.  When it was put to him that the likelihood of ignition would be greater if the drop sheet was hessian, plastic material, or a highly flammable polyester material, he responded, "possibly".

(b) When asked what sort of paint thinners were in the room on 15 January 2007, he responded that he had no way of knowing that.  When asked why he chose the two paint thinners used in his experiment, he said that he chose them because they were in the house.  That was confirmed by the investigating police officer, Mr Peake, who gave evidence that the tins he saw and "catalogued" at the house when he first went there in May 2007 were not necessarily in the house on the day of the fire.  Mr Smith said that he did not appreciate that there was evidence that a "white spirit" might have been the thinner.  He agreed he did not use white spirits in his experiment.  It is evident that he did not know that this was what the appellant told Mr Heppell was the paint thinner which had spilled.  Mr Smith and defence counsel used the terms "white spirit" and "mineral turpentine" as though they were interchangeable, and Mr Smith appeared to accept (he said "ok") defence counsel's suggestion that mineral turpentine has a lower ignition temperature than paint thinner, 250 degrees celsius as compared to 400 degrees Celsius. 

(c) Mr Smith said that he had used 500ml of thinner in his experiment only because, "we had to choose a volume to put on".  He agreed that he could not replicate the area covered by the thinner.

(d) Mr Smith agreed that the ambient temperature had a role to play because it would affect the amount of evaporation of the thinner.

(e) Mr Smith's experiment did not have a fan blowing on the lit cigarettes or over the top of the thinner.  He agreed that blowing on the cigarettes might create more combustion and a localised area of high temperature and possibly cause a fire.

(f) Mr Smith could not say whether or not the cigarettes smoked by the appellant provided the same resistance to heat transfer possessed by the cigarettes he used in the experiment.

[36]  These differences between the experiment and the scenario at the time were at least of sufficient importance to require analysis, but Mr Smith did not explain the basis of his expressed belief that the differences would not significantly bear upon the results.  Importantly, Mr Smith expressed that belief in evidence-in-chief, before he appreciated that there was evidence that the thinner involved in the fire was white spirit which, he seemed to concede, was much more volatile than either of the thinners used in the tests.  He did not give evidence that substitution of each of white spirit for the other thinners, mid-summer ambient temperature for mid-winter ambient temperature, an operating fan for a relatively still environment, and the cigarette smoked by the appellant for that used in the tests, might not make a significant difference in the experiment.  Furthermore, Mr Smith agreed in cross-examination that his report of his opinions was qualified by the statement that his opinions "are indicative only due to the extended period of time between the fire and the examination of the scene".

[37]  It may be arguable that there was no evidence capable of supporting a finding by the jury that the assumptions upon which Mr Smith’s opinions were predicated were sufficiently like the true facts to make his opinions useful, with the result that the evidence was inadmissible,[10]  but defence counsel did not object to Mr Smith’s evidence and its admissibility is not in issue in the appeal.  Nevertheless, the differences between the test scenarios and the conditions of the fire, and the absence of a cogent explanation about the effect of those differences (particularly concerning the nature and quantity of the thinners and the effect of an operating fan), make good the appellant's point that there were significant weaknesses in Mr Smith's evidence of his “indicative” opinion. 

[38]  The respondent's counsel acknowledged that there were such weaknesses in the evidence of the experiments but he contended that Mr Smith's opinion was supported by well regarded research.  The only evidence of that research was given by Mr Smith:

"Are you aware of any academic writing about cigarettes as a source of ignition?-- There are a number of textbooks and academic papers about cigarettes, because they are the cause of a lot of - they are - a lot of fires are attributed to them, and they are in fact the cause of a lot of fires.

In the circumstance that we're considering here where there is a flammable substance like thinners probably involved, thinners would evaporate and produce vapour?-- They do, yes.

Is there any academic writing on the likelihood of cigarettes igniting vapour?-- In general it's a very unlikely way to ignite a vapour mixture - a flammable vapour mixture. Having said that, in the right circumstances, yes, they could. But as I said before, it's a very unlikely way of doing it. There are a number of tests been conducted besides me over a number of years with almost uniform results, indicating that they do not result in an ignition, but there are reported cases where that has in fact been the source of ignition. So the problem with doing negative testing, as I've done, is that of course you need to do an infinite number of tests to cover an infinite number of possibilities or possible causes, so I can only - I was limited by the number of tests I could do. But the tests I performed and all of the literature I've read indicate that cigarettes are a very unlikely source of ignition in this circumstance."

[39]  The generality of that statement, including the absence of any description of the reported tests in which no fire resulted and of the reported cases in which a fire did result, rendered this evidence of limited probative value in the particular circumstances of this case.

[40]  The recorded tests graphically illustrated the failure of lit cigarettes to set fire to “thinners” and sheets soaked in thinners.  Those tests and Mr Smith's evidence of the literature might have weighed heavily with the jury, partly because of the impressive statement of Mr Smith’s qualifications as an expert and partly because the trial judge did not give the conventional directions about the limitations upon the value of expert evidence.  But for the reasons I have given Mr Smith’s evidence was incapable of excluding the hypothesis consistent with the appellant’s innocence which was supported by Mr Heppell’s evidence.

Did the appellant have a motive to burn his house down? 

[41]  In opening the case the Crown Prosecutor told the jury that the evidence of Mr and Mrs Dykes' financial difficulties, exacerbated by their excessive gambling,  provided a "clear and compelling motive" for the appellant to burn down his own house.  This second plank of the Crown case was based partly upon statements by the appellant to Mr Gannon to the effect that the appellant and his wife were in a poor financial position but more substantially upon expert evidence from Ms McKinnon, a chartered accountant employed by the Queensland Police Service.  She analysed bank and credit card statements and other accounting records which had been proved in the Crown case.  Her opinion evidence was said to demonstrate that the cash resources of the appellant and his wife had significantly worsened from July 2006 until the day of the fire.

[42]  In summing up to the jury the trial judge referred to the prosecution’s reliance on evidence said to show that the appellant and his wife were under financial stress contributed to by a fall-off in their cleaning business and excessive gambling and told the jury that this was relied upon to prove that the appellant had a motive to deliberately burn down the house with intent to defraud the insurer.  The trial judge directed the jury that the existence of motive without more would not be sufficient to found a finding of guilt, but that the existence of motive can be an important factual issue particularly in a circumstantial case like the present, where the prosecution asked the jury to infer that the appellant had deliberately burnt the house down to defraud the insurer. 

The financial evidence

[43]  Ms McKinnon expressed opinions that Mr and Mrs Dyke’s debts increased during the period 1 July 2006 to 30 June 2007; that business sales from the appellant's cleaning business, which were the major source of income, declined to a point where there was minimal income received after December 2006; and that their available cash resources declined leading up to January 2007, with Mr and Mrs Dykes' accounts on a consolidated basis being overdrawn, principally from January 2007.

[44]  The appellant’s counsel identified weaknesses in this evidence.  Particular criticism was made of the Crown’s reliance upon her evidence about the deterioration of the financial position in the financial year ended 30 June 2007, rather than for some years up to and ending on the date of the fire, and about weaknesses in her evidence to the effect that very large sums of money were withdrawn from ATMs in licensed premises. 

[45]  Counsel for the respondent properly acknowledged the validity of the appellant’s criticism that Ms McKinnon's analysis covered the whole of the financial year when the last five months of that period post-dated the fire.  The point may be illustrated by reference to a documentary exhibit tendered through Ms McKinnon.  It illustrated consolidated monthly closing balances derived by combining cheque and savings accounts which Ms McKinnon thought were used in the cleaning business.  The mid-point of a line drawn between the points marked for December 2006 and January 2007 indicated that the combined balance was about zero, but on closer analysis the only relevant entries were the points marked for the end of the months of July through to December of 2006.  That is so because Ms McKinnon acknowledged that daily balances fluctuated greatly and because all of the closing balances after December 2006 post-dated the fire (when, as Mr Gannon acknowledged, the appellant was ill).  The closing balance for July 2006 was about $8,000.  The closing balances thereafter fluctuated markedly, but at the end of December the closing balance was about $8,000 or a little more.  In short, whilst the document was superficially impressive it did not advance the Crown case that the appellant had a motive to destroy his house.

[46]  Another documentary exhibit indicated that the income of the business for January 2007 was of the order of only $5,000, as against the income of the order of $30,000 in August.  This is consistent with statements made by the appellant to Mr Gannon that a superior contractor, Mr Cullen, had refused to pay what the appellant claimed, but Mr Cullen denied, was a debt of about $40,000, and that the appellant had no other cash resources.  However, various reasons for the apparent decline in the business not being inconsistent with the overall financial health of the appellant and his wife were brought out in the cross-examination of Ms McKinnon.  She had no knowledge of the cleaning industry and therefore did not take into account the fact (of which the cleaning contractor Mr Cullen gave evidence) that December/January was generally a quiet month in the building industry in which the appellant provided much of the cleaning services; her analysis based upon accounting records was also unpersuasive because there may have been unbanked cash receipts and cash payments, including to employees, were common.  The latter point was directly supported by evidence adduced in Ms Fahey’s cross-examination and in Ms Hole's statement tendered in the Crown case.

Gambling

[47]  There was also an explanation for the very large amounts withdrawn from an ATM at one of the licensed premises referred to in Ms McKinnon's evidence.  The Crown relied upon that evidence to show excessive gambling, but this ATM was very close to the appellant’s house, from which the cleaning business was conducted (using cash in the way I have mentioned).  Counsel for the appellant referred to the evidence of employees at the licensed premises at which the Dykes gambled to the effect that they were not "problem gamblers".  The evidence suggested that they often spent some hours gambling, but this evidence was of limited significance.  Ms Lunt, who worked at a hotel, gave evidence that Mr and Mrs Dyke came there just about every day when she was working and they gambled for "a couple of hours" or "for a while".  She could not say how much money they gambled.  Ms Martin gave similar evidence.  Whilst Ms Lunt agreed in cross-examination that she would not classify Mr and Mrs Dyke as "problem gamblers", in re-examination she was permitted to say that they were "excessive gamblers for the fact that … they were doing high bets".  That opinion evidence was potentially prejudicial but it had no substantial probative value.  There was no reliable evidence that Mr and Mrs Dyke gambled beyond their means.

The Council notice

[48]  The prosecution also relied upon documentary evidence that the local council had given Mrs Dyke a notice to show cause why it should not issue an enforcement notice concerning some of the renovation work, which was alleged to have been done without the necessary development permit.  Mrs Dyke's response of 11 January 2007 to the notice read:

When we bought the premises the underneath of the house was already enclosed with tin (colour bond) this was done when the house was constructed approx 15 years ago. Over time the tin had rusted and needed to be replaced so we removed the sheets of tin and replaced them with blue board, and painted them.

The south east side of the .house was external stairs with a partial wall and an awning we were having problems with the neighbours at number 29 they were throwing rocks and making derogatory comments,  the police were called, several times so we finished the wall off and painted it and replaced the roof sheets from the awning this gave us privacy.

As we were not a new development just replacing an original one I was unaware that a development, permit was required. I was under the impression that I only needed a permit for new work and not for the replacement of the old tin and stair enclosure."

[49]  It seems to have been assumed that the appellant was aware of the notice to Mrs Dyke and her response, but the Crown did not attempt to prove that the reinstatement work sought by the notice would be difficult, expensive or disadvantageous in a re-sale.

The loss upon destruction of the house

[50]  Of even more significance than those points I have so far discussed is the fact, as was submitted for the appellant, that Ms McKinnon's accounting exercise did not bring to account the loss that the appellant would probably suffer by burning down his house. 

[51]  A document tendered for the Crown established that the property was purchased in February 2005 for $170,000.  (According to the documentary evidence, the property was owned by Mrs Dyke.) An employee of a finance company, Mr Osborne, gave evidence that in November 2006 the finance company leant Mrs Dyke $216,000 for the purpose of repaying existing debts, with the new loan secured by a mortgage over the property.  Mr Osborne said that this transaction required, as part of the finance company guidelines, that "we must insure that property (ie, the house) to the replacement value."  He agreed that the replacement value was $200,000.  He said that this was "one of our requirements and specific conditions".  In Mr Gannon's evidence in chief he agreed that on 30 November 2006 an earlier contract for the insurance of the house and its contents was replaced by a new policy in accordance with the requirements of this new mortgagee.  Mr Gannon gave evidence that as at 30 November 2006 the house was insured for $200,000 and its contents were insured for $40,000, and he agreed that in the immediately preceding policy it sounded "about right" that the house had been insured for $140,000 and the contents for $47,000.  He accepted that the increase in the replacement value of the house was not out of the ordinary and that the new policy did not contain an overstatement of the value of the replacement of the house or of the contents inside it.  There was evidence that the renovations were conducted by the appellant and his wife to improve the house for re-sale and were nearly finished when the fire burned the house down.  After the house was effectively destroyed in the fire the property sold for $190,000.

[52]  On the basis of that evidence it was contended for the appellant that upon the intended sale of the completely renovated house the appellant and his wife stood to make a tax free profit of some $100,000 to $200,000.  That proposition was not positively established because we were not referred to evidence that the assumed replacement value of the house of $200,000 would have been reflected in that amount upon a resale of the house and land; but having regard to the value which a house costing some $200,000 to replace might have contributed in any re-sale the Crown did not establish that the appellant’s overall financial position was likely to be improved by the destruction of the house by the fire and a successful claim for indemnity on the insurance policy.

Reinstatement or cash indemnity?

[53]  The Crown case also seems to have been premised upon an assumption that the appellant believed that the result of substantial damage to or the destruction of the house by fire would be that the insurance company would pay $200,000 or some other substantial cash amount to the appellant's wife, who held the policy.  That assumption was not verified by any evidence.  The policy and Mrs Dyke’s claim on the policy were not put in evidence.  There was no evidence of the terms of either document, save that Mr Gannon gave evidence that Mrs Dyke made a claim on an insurance policy effected in November 2006 which nominated a replacement value for the house of $200,000.

[54]  In circumstances in which the evidence of Mr Gannon and Mr Osborne was that the mortgagee had insisted upon that insurance as a condition of the mortgagee lending $216,000, the inference is available that the mortgagee or the insurance company would have insisted upon reinstatement or payment to the mortgagee, rather than any substantial cash payment, in the event of the house being damaged or destroyed by fire.  The Property Law Act 1974 (Qld) provides, in s 58:

"58 Insurance money from burnt building

Where a building is destroyed or damaged by fire a person who has granted a policy of insurance for insuring it against fire may, and shall, on the request of a person interested[11] in or entitled to the building, cause the money for which the building is insured to be laid out and expended, so far as it will go, towards rebuilding, reinstating, or repairing the building, unless—

(a) the person claiming the insurance money within 30 days next after the person’s claim is adjusted, gives sufficient security to the person who has granted that policy that the insurance money will be so laid out and expended; or

(b)the insurance money is in that time settled and disposed of to and amongst the contending parties to the satisfaction and approbation of the person who has granted the policy of insurance."

[55]  That section is a modern re-enactment of an eighteenth century English enactment which was designed to deter persons from deliberately setting fire to their houses to collect on insurance.[12]  In one form or another such a provision has applied in Queensland continuously since the beginning of the colony.[13]  Reference should also be made to ss91(3) and (4) of the Property Law Act, which empower a mortgagee to insist that all money received on an insurance of mortgage property against loss or damage by fire be applied by the mortgagor in making good the loss or damage or towards the discharge of the mortgage money. 

[56]  The Crown adduced no evidence that it was likely that the insurer or the mortgagee would not insist upon reinstatement rather than payment, or of the appellant’s belief in that respect.  On the evidence adduced in the Crown case the effect of arson by the appellant would likely have been the destruction of personal property and payment of its insurable value which had recently been reduced, (and which included items shown by the evidence to have real sentimental value for the appellant’s wife), the destruction of a great deal of work the appellant had organised and done over many months in renovating the house, the loss of the roof over his and his wife's head, the destruction of the place from which their cleaning business was conducted and their business records, the incurring of inevitable expense in finding alternative accommodation, and nothing from the insurance company other than the replacement of that which was destroyed or a payment to the mortgage of its replacement value. 

Conclusion as to motive

[57]  In these circumstances, evidence that the appellant and his wife were in a difficult financial position could not itself convincingly establish that he had a motive to destroy the house.

Did the appellant express an intention to set fire to his house?

[58]  In light of the weaknesses in the evidence relied upon in support of the first two planks of the Crown case, the third plank of the Crown case was quite critical.  This was apparent at the trial.  The trial judge concluded that the evidence of Ms Alexander was the only evidence of an intention to defraud.  The trial judge described the Crown case as being weak and as a case which, without the evidence of Ms Alexander, could not sustain a safe verdict of guilty. 

[59]  In summing up, the trial judge read the relevant passages from the evidence of Ms Alexander and commented that if the jury accepted that evidence it might well decide that it could be treated as an admission by the appellant that he intended to deliberately burn the house down with the fraudulent intent of claiming on the insurer.  The trial judge then directed the jury that before using the evidence in that way the jury must be satisfied beyond reasonable doubt both that Ms Alexander was truthful and accurate in recounting statements by the appellant after such a lapse of time between when the statements were said to have been made in November 2006 and when she reported them to the authorities in April 2008, and that the defendant meant what he said. 

[60]  Ms Alexander gave the following evidence.  She regularly visited her friends Ms Justine Hole and Mr Ron Grant, who she knew to be neighbours of the appellant and his wife.  She was aware that renovation work was being done to the Dykes' house during 2006 and she knew that Mrs Dyke worked for her husband cleaning, as did Ms Hole.  Ms Alexander remembered an occasion in November of 2006 when she was staying with Ms Hole when Mr and Mrs Dyke came to the house.  There followed these critical passages in the evidence in chief:

"Can you tell us about that? What happened and what was said?-- What really sticks in my mind was the attitude, because I - you know, I don't spend much time with her, and they just sort of floated in and acted like they owned the place. And that was the conversation with what they were saying about, you know, money's getting tight. I'm going to, you know - place might have to go up, sort of thing, and then talking about percentages of, you know, insurance on how much you would get and all that sort of thing, and then talking about actually how to put a place on fire. It just - and then saying, 'Oh, make sure we get the kids out,' you know. I was

like-----

…

And as best as you can recollect, what did he actually say?

What were the words that he used?-- I - on the first occasion it was - it was I'm – 'Money's getting tight. I can't find any cleaning contracts. They are not coming through. If this goes on any longer, this place will have to go up.' I'm, like, that's a pretty strange thing to say. But that's the first thing he said to me when I met him, so okay. Then on the second occasion he brought up paint tins and, like, setting the place up with paint tins, putting a dropsheet down with paint tins, and using, like, ethanol fluids or, you know, flammable liquids and leaving them alight over a dodgy light. And he also brought up, you know, electrical wiring. Because he knew that his dryer was already dodgy, so - yeah. And then on the second occasion he started discussing units of cash and - with the insurance money, and to make sure the kids weren't there. So, you know, I was, like, okay. I'm obviously witnessing someone planning something - in my opinion, but I can't be the judge of that. Was Justine Dyke present when these things were said?-- Yes. Yes, she was. Did she say anything?-- Did she say anything? Yes, she did. But she - those discussions weren't brought up to - with them, because that's quite threatening. In my opinion anyway.

Perhaps we'll move on?-- No, I mean - I mean, it's – when someone's planning something to you in your face, you're not really going to have an open discussion with them about it, sort of thing. You sort of let them leave, because they were only over for a cup of coffee. They didn't have any intentions of employing her at the time because, you know, they had no money. That was their explanation - given in their explanation for why she wasn't employed by them.

I think I might have confused them. When I said was Justine there, I meant to say was Justine Dyke present when Russell Dyke was saying things? Did she say anything, Justine Dyke?-- Oh, no. Not to my recollection, no.

Now, when was it that that conversation occurred?—Which conversation? Which one?

Well, that's what I wanted to clarify. You've told us of things that were said by Russell Dyke. Was that all said on one occasion, or was this more than one occasion?-- Just on - from my recollection, about two occasions, yeah. Two or three, because there was more-----

When were those occasions that things - this topic was discussed by Russell Dyke?-- When he had the - those big discussions, it was on a Friday afternoon and he had a couple of drinks. It was sort of after work and - yeah.

Could I approach it this way. We know there was a fire at the house on 15 January 2007?-- That's right.

The conversations you've mentioned, were they before or after that fire?-- Well before.

And how long before?-- Only by a couple of months, yeah.

And how many times do you think the topic was mentioned?-- Every time I saw them.

Well, how many times was that?-- Be five times all up, I think. About five times all up that I would have ever seen them. Never spoke to them otherwise.

What was said by Russell Dyke on those occasions?—Everything I just mentioned. Everything that's in my statement."

[61]  It will be noticed that in Ms Alexander's evidence in chief she first referred to two such conversations; she then said it occurred on two or three occasions; and she concluded by saying that it was said on each of the five occasions on which she was present when the appellant was at Ms Hole's house.  When Ms Alexander was asked in cross-examination to explain a similar discrepancy between her evidence and an earlier statement she responded with this unconvincing answer:

". . . he [Mr Peak] felt it was more important if we discussed the one conversation, the big one because I wasn't actually - well, as I saw it, I wasn't actually involved in those conversations, I can't say what was going on in the conversation because I was talking to my friend."

[62]  Furthermore in Ms Alexander’s evidence in chief she first agreed that Mrs Dyke was present and said something in the relevant conversation but Ms Alexander very quickly contradicted that by saying that Mrs Dyke did not say anything.  Ms Alexander's evidence in chief was also substantially inconsistent, both as to the time of any relevant conversation and as to its content, with a statement by Ms Hole.  Ms Hole said in a statement dated 3 November 2007, which was tendered by consent at the trial, that Ms Alexander stayed at her house over the weekend of 6 and 7 January 2007.  Mr and Mrs Dyke came over on Sunday, 7 January and Ms Hole remembered a conversation to the following effect:

"5)During our conversations I remember Russell and Justine were very concerned about their financial position. Justine said words to the effect 'I am worried that we are not going to get our money from Flexi-clean. He owes us about $40,000.00'. This was from the cleaning contract Justine and Russell had organised at the Swell Apartments at Burleigh Heads throughout 2006. They both looked very worried about what was going to happen as they had also been trying to renovate their house at the time. Some of his renovations were also causing Russell concerns because he said words to the effect 'Guy has complained to the Gold Coast City Council. They have told me I have to get approval for the stair case'. Russell had enclosed the outside stairs leading to lounge room of his house during 2006. This made the house look terrible but I did not say anything to him at the time because I didn’t want to hurt his feelings.

6)I remember that we continued to discuss what might happen when Russell said words to the effect 'If you go and set fire to our house it would be good because we need the money'. At the time he said this I took it as a joke because I remember we all laughed. We then went on to discuss other things and Russell and Justine went home about an hour after they had arrived."

[63]  On Ms Hole's evidence there was only one such conversation, the reference was to someone other than the appellant ("you") setting fire to the house, and all present thought that this was said in jest. (The respondent did not rely upon Ms Hole’s evidence as sufficient proof of the appellant’s alleged intention to burn down the house.)

[64]  It emerged in cross-examination that Ms Alexander was a "service provider in the adult entertainment industry" who used illicit substances.  In response to suggestions that she was taking illicit substances at the time of the conversation which she asserted Ms Alexander said, “at the time I was a little bit but not very much at all”, "I was actually sober during that conversation because it was in the afternoon", and "I was pretty sober and I – yeah."  It may be noted that Ms Hole’s evidence was that the relevant conversation occurred in the morning. 

[65]  Furthermore, cross-examination of the investigating police officer had extracted the information that Ms Alexander first gave a statement in about March 2008, which was some 14 months after the fire and some 16 months after the conversation she purported to recount.  Another very unsatisfactory feature of Ms Alexander's evidence was her explanation for that very lengthy delay in her giving a statement to police when, as she admitted in cross-examination, she had learned from Ms Hole about the fire (and, it may be inferred, the appellant’s account) on the very day it occurred:

"I want to ask you some questions, alright? Do you understand me? The question is this: why did you wait from November 2006, or at least from 15 January 2007 until April 2008 to tell anybody in authority about this conversation that - sorry, the statements you attribute to my client?-- Who would have been the proper person to discuss that with?

Well, I don't know. Did you ring anybody?-- Who am I supposed to discuss this with? Actually, I did and it was a police officer, and it was involved in another case and I can't discuss that with you.

Alright. Give me the name of the police officer, please?-- Mr Fisher.

Mr Fisher. Who is Mr Fisher? Does he have a rank?-- Victorian police that lost his job.

Victorian police that lost his job?-- Yeah.

Why did he lose his job?-- He was involved with a couple of cases and there was a drug bust that was going on in The Valley and-----

The Valley in Queensland or The Valley-----?-- In The Valley in Queensland but he was chasing some people around trying to get - have them locked up.

Yes?-- It was - it wasn't in his time to have that discussion with me regarding this case and, as he said-----

So you raised it with him?-- I did discuss it because I was very concerned. I didn't want to be put in this position.

When was this?-- It was just after the fire because I was just concerned that-----

Just after the fire?-- Because I was concerned.

Victorian policeman?-- It was either just before or just during.

So Victorian policeman is chasing criminals around the Fortitude Valley and you had a conversation with him, because he was busy with other things he decided not to take it any further?-- He told me that I should bring this up if I was approached upon it, yes.

But-----?-- What he did tell me is that if - "if you do hear that, you know, that it is possible, then the police will come and find you because you're the next door neighbour and Justine will inform the correct officer-in-charge to have this conversation with you."”

[66]  Ms Alexander’s evidence on this topic wears the appearance of fantasy.  Ms Alexander gave equally unimpressive evidence when she set out to explain what other topic (regarding an unidentified developer or builder she thought lived in Sydney) she was discussing with police in about April 2008 when she gave her statement in this matter:

“Just that I’ve – someone was saying things that are very strange to me and I didn’t quite understand what was actually going on at the time and he – he knew that, and on another turn he was sort of  talking to me in my mind’s eye as a police officer, sort of how ridiculous and silly that was, yeah.”

[67] Even attributing the fullest significance to the advantage which the jury possessed in seeing and hearing Ms Alexander give evidence, it is apparent that a conviction based upon her evidence must be regarded as unsafe. 

Conclusion

[68]  Setting aside a jury's verdict is a serious step, particularly having regard to the role of the jury as the constitutional tribunal for deciding contested facts.[14]  But bearing that in mind, this is in my opinion a case which falls squarely within the category of cases identified in M v The Queen:[15]

"If the evidence, upon the record itself, contains discrepancies, displays inadequacies, is tainted or otherwise lacks probative force in such a way as to lead the Court of Criminal Appeal to conclude that, even making full allowance for the advantages enjoyed by the jury, there is a significant possibility that an innocent person has been convicted, then the Court is bound to act and to set aside a verdict based upon that evidence."

[69]  The evidence, particularly that of Mr Smith, Ms McKinnon, and Ms Alexander, provided a basis for the Crown case, but there were such serious weaknesses and deficiencies in the evidence adduced to support each of the three main planks of the Crown case as to compel the conclusion that it was not open to a properly directed jury to be satisfied beyond reasonable doubt that the appellant was guilty of the offences of which he was convicted.  For the reasons I have given this is not a case in which there was no substantial miscarriage of justice such that the appeal might be dismissed by applying the proviso in s 668E(1A) of the Criminal Code.  Furthermore, because the evidence at the trial did not justify a conviction, the discretion to order a new trial should not be exercised.[16]  It is therefore unnecessary to give separate consideration to appeal grounds 2 and 3. 

[70]  For these reasons I concluded that the appeal should be allowed, the verdicts set aside and verdicts of acquittal entered.