//
© 2026 Zoe. All rights reserved.
Zoe is a legal information platform. Always consult the Federal Register of Legislation for authoritative text.
[2019] HCA 37
appellant. Appeal allowed. Order of the Court of Appeal set aside and in its place the appeal allowed, conviction quashed and verdict of acquittal entered.
No linked legislation citations have been extracted yet.
Steven Fennell helped an elderly neighbour on a small Queensland island with her shopping and banking. After she was found murdered in her bedroom the police built a purely circumstantial case against him based on the idea he had stolen from her to fund gambling and killed her to cover it up. The only evidence really tying him to the crime was two witnesses saying a rusty hammer found in the mangroves was one they had lent him years earlier. The High Court decided this was not enough. The times when Fennell could have done it were either impossible or no different from what many other locals could have done, the supposed motive did not stack up because the withdrawals were authorised and his finances were basically in order, and the hammer story was too unreliable given how ordinary the tool was, how long since they had seen it, the quick flash on television, and the contradictory details the couple later gave. Because no reasonable jury could have been sure of guilt the conviction was overturned and Fennell was acquitted immediately.
AI-generated legal information, not legal advice. Zoe can make mistakes. Check the cited source, and for advice about your situation consult a qualified Australian lawyer.
2,464 words · generated 24/04/2026
Mrs Liselotte Watson, aged 85, lived alone in a neat two-storey house on Macleay Island in Moreton Bay. She had limited mobility and relied on local assistance, including from the Lions Club. Steven Fennell, a founding Lions Club member who delivered pamphlets for a living, helped her for one to two years with shopping, banking and yard work, visiting once or twice daily for 15–20 minutes. On 12 November 2012 Mrs Watson was seen alive around 9.30 am. She did not answer several afternoon phone calls, including one from Fennell at 4.22 pm. Newspapers delivered early on 13 November were not collected. That afternoon Fennell went to the local police officer’s house expressing concern that Mrs Watson had not answered her door; a welfare check revealed her body on the bedroom floor. She had been struck at least four times to the back of the head with a blunt instrument, probably while in bed in her nightdress, and had defensive injuries consistent with a hammer. The house had been ransacked—drawers opened, cupboards disturbed, a suitcase left open—yet only $290 was found despite widespread local knowledge that she kept large sums of cash.
A shaving bag containing Mrs Watson’s Westpac bank documents had been seen in the mangroves at Thompson Point days before the murder. After the murder police recovered her wallet, purse and a claw hammer about 15 metres from the bag. Two hammers were found inside her house but neither was alleged to be the murder weapon. Fennell’s house, vehicles, clothing and computer were forensically examined; no blood, DNA, fingerprints or excess cash linked him to the scene. His DNA was excluded from the shaving bag. He gave a detailed timeline corroborated in large part by phone records, computer usage, independent witnesses and his wife. He omitted only that he had gambled at the TAB on the Monday afternoon, an omission explained by his long-standing desire to hide gambling from his wife.
The Crown case at trial rested on three strands: a narrow window of opportunity on the Monday afternoon, motive to conceal thefts from Mrs Watson to fund gambling losses, and a miscellany of other matters, the most important being evidence from Mr and Mrs Matheson that the mangrove hammer was one Mr Matheson had lent Fennell one or two years earlier and which Fennell had never returned. The jury convicted Fennell of murder. The Queensland Court of Appeal dismissed his appeal. The High Court unanimously allowed the further appeal, quashed the conviction and directed an acquittal, holding that the verdict was unreasonable.
The joint judgment systematically dismantled the Crown case as a whole. On opportunity the Court found the evidence “extremely weak”. The only potentially viable sighting placing Fennell at the house for any length of time came from Ms McKie at around 2.00 pm on the Monday. Even accepting her evidence, the 20-minute window was insufficient for a premeditated hammer attack, ransacking, cleaning or disposing of bloody clothes and the weapon, all while Mrs Fennell was at home and without leaving any forensic trace. Evidence from Mr Robinson and Mrs Doolan was rejected outright: Mr Robinson’s fleeting sighting was a reconstruction after more than a year, contradicted by Mrs Fennell’s evidence that she had the utility that morning, and abandoned by senior counsel for the Crown on appeal; Mrs Doolan’s 6.00 pm sighting was irreconcilable with computer records showing Fennell (or someone using his account) logged into his LinkedIn and searched burns-recovery topics at the exact time she said he was at Mrs Watson’s house.
Motive fared no better. The Crown alleged Fennell stole to cover gambling debts. Yet forensic accounting showed his overall position was sustainable; large TAB wins were paid electronically and smaller ones in cash, and the pattern of betting had not changed. The $24,000 withdrawn from Mrs Watson’s account over three months, while larger than average, followed a five-year pattern of similar withdrawals. Every withdrawal slip was signed by Mrs Watson; most bore her explicit authority for Fennell to act as her agent. Bank staff gave evidence of verification phone calls. The supposed alteration of the final $8,000 slip could not overcome the inference that the bank had telephoned her. The disposal of documents at Thompson Point proved nothing distinctive about Fennell because he had never been shown to possess the shaving bag and his DNA was excluded from it. The $290 left behind actually supported a genuine burglary theory given the volume of cash Mrs Watson was known to keep.
With opportunity and motive reduced to “barely” distinguishing Fennell from any other island resident who knew about the cash, the Matheson hammer evidence became “essential”. The Court held it should have been given almost no weight. Mr Matheson claimed to recognise the hammer from a photo “flashed up” on the 6.00 pm news without context. Neither witness had seen it for one to two years (or four to five on Mrs Matheson’s version). The item was a mass-produced generic claw hammer now heavily rusted from salt water. Their subsequent statements contained minute details of “dags”, chips and drop marks yet they could not identify Mr Matheson’s own tool belt or other tools in a police line-up. Their accounts were inconsistent on who first saw the news item, the year the hammer was lent, and the precise cause of each defect. The identification occurred 12 days after the broadcast, in the context of a notorious local murder, after police showed them only the single hammer. The Court expressly applied the “realities of human experience” concerning memory plasticity, contamination and bias, concluding the evidence was “glaringly improbable”.
Because the only strand capable of converting weak surrounding circumstances into proof beyond reasonable doubt had negligible probative value, the verdict was unreasonable. The Court of Appeal’s contrary conclusion was infected by four errors: misstating Fennell’s record of interview about the $8,000, drawing unavailable conclusions from the accounting evidence, accepting the hammer identification as “detailed and consistent”, and relying on Mr Robinson’s discredited sighting for timing. These errors required the High Court to reassess the whole case and enter an acquittal.
Prior to this decision the statutory test under s 668E(1) of the Criminal Code (Qld) required an appellate court to allow an appeal if the verdict was unreasonable or could not be supported having regard to the evidence. The Court reiterated that in circumstantial cases the question is whether the jury must have had a reasonable doubt, assessed on the whole of the evidence rather than piecemeal. It drew on long-standing recognition of the fallibility of identification evidence, especially where memory is involved, and the risk that honest witnesses can be mistaken after time passes or after suggestive procedures.
After the decision the law is clearer on two points. First, when opportunity and motive evidence is neutral as between the accused and a pool of other persons with equal access to the same knowledge, any “essential” identification evidence must itself be capable of sustaining conviction; if it is glaringly improbable it cannot fill the gap. Second, appellate courts are expressly directed to bring “realities of human experience” and scientific understanding of memory contamination and bias to bear when weighing identification of ordinary objects seen fleetingly or after long delay. The judgment does not change the jury’s primary role but emphasises that an appellate court is not bound to defer where the evidence, objectively weighed, cannot support the necessary degree of satisfaction.
At the heart of the judgment is the statement that once opportunity and motive were seen to be extremely weak, “the evidence from Mr and Mrs Matheson linking Mr Fennell to the hammer became essential evidence that a reasonable jury would have been required to accept before convicting Mr Fennell. To use Wigmore’s metaphor, the evidence from Mr and Mrs Matheson became by far the most significant of the ‘strands in a cable’ supporting the conviction. But the evidence of Mr and Mrs Matheson should have had so little weight that, at best, it was barely admissible.” In plain English the Court is saying the prosecution case was like a rope made of many thin threads; when most threads were almost worthless the one thick thread had to be strong, but in fact it was so frayed it barely qualified as evidence at all.
Later the Court explains why the Matheson evidence was unreliable: “Mr Matheson’s evidence of his identification was from a photo that was ‘flashed’ briefly on a television screen without any context. It was a photo of a hammer of a type that was mass produced and generic. Also, Mr Matheson had not seen the hammer that he claimed to identify for one to two years … The Thompson Point hammer had been submerged in salt water … An accurate identification in those circumstances would have required an astonishing visual memory.” Translation: recognising a rusty everyday hammer you last saw years ago from a split-second news clip, then recalling microscopic defects while forgetting the brand and failing to recognise your own other tools, is simply not believable.
The dispositive conclusion is expressed without equivocation: “It was not open to the jury to be satisfied of Mr Fennell’s guilt beyond reasonable doubt. The Court of Appeal should have allowed the appeal, quashed the conviction, and entered a verdict of acquittal.” In ordinary language the evidence was so thin that no properly instructed reasonable jury could have convicted; the only proper outcome was immediate acquittal.
The Court’s treatment of memory science is captured in the passage: “the court may take into account the realities of human experience, including the fallibility and plasticity of memory especially as time passes, the possibility of contamination of recollection, and the influence of internal biases on memory. The court can also take into account the well-known scientific research that has revealed the difficulties and inaccuracies involved in assessing credibility and reliability.” Plain-English version: judges are not required to pretend that human memory works like a video recorder; everyone knows memory changes, can be altered by later information, and is affected by what we expect to see, especially in high-stakes cases.
This precedent is triggered when a conviction for a serious offence rests wholly or substantially on circumstantial evidence and the prosecution case contains one or more of the following features. First, evidence of opportunity is limited to a narrow temporal window that is contradicted by independent records (phone, computer, CCTV, alibi witnesses) or is equally available to numerous other persons sharing the same local knowledge. Second, motive is said to arise from financial pressure but the financial analysis, once all accounts and cash flows are considered, shows no unaccounted deficit or a deficit so small as to be consistent with normal living. Third, the decisive link is identification of a generic object (here a hammer) by a witness who last saw it years earlier, recognises it from a fleeting decontextualised image, provides minute detail after a delay, is not given a line-up, and whose account contains internal or inter-witness inconsistencies. Fourth, the identification occurs in an emotionally charged environment (local murder) after the witness has been shown only the single suspect item.
The precedent applies with greatest force where the Crown concedes, or the evidence compels the conclusion, that without the impugned identification the remaining strands cannot sustain a conviction. It is not limited to hammers or murder; any ordinary, mass-produced item said to be uniquely remembered after long absence in suggestive circumstances will engage the same scrutiny. Conversely, the case does not disturb convictions where forensic links, direct evidence, or compelling motive evidence exist independently of contested identification.
The judgment itself demonstrates how an intermediate appellate court’s errors in fact-finding and in weighing individual strands can require the High Court to conduct a fresh assessment of the entire circumstantial case. The Court of Appeal had accepted the hammer identification as “detailed and consistent” and had drawn accounting conclusions that were not open once all bank accounts and cash business receipts were considered. The High Court treated those errors as infecting the global reasonableness assessment, leading it to reverse the decision and substitute an acquittal rather than remitting the matter. Subsequent application therefore requires intermediate courts to avoid compartmentalising evidence and to test any “essential” identification strand against the known weaknesses of memory outlined in the reasons.
The decision reinforces that an appellate court is entitled, indeed obliged, to bring “realities of human experience” to bear even where a jury has seen the witnesses. Later courts are directed to treat evidence as “glaringly improbable” where it displays the combination of long delay, generic item, single-item presentation, inconsistent detail and absence of verification procedures that characterised the Matheson evidence. The judgment has been cited for the proposition that once surrounding circumstances are neutral, the decisive identification must itself carry sufficient probative force; if it is “barely admissible” the verdict cannot stand. Courts have also taken from it the need to examine whether an omission (such as Fennell’s failure to mention gambling) is truly inculpatory or is equally consistent with an innocent explanation (concealing gambling from a spouse).
The judgment leaves open precisely how much weight an appellate court may place on “well-known scientific research” concerning memory when the research itself is not led in evidence at trial. The Court refers to such research but does not annex any specific studies or require expert evidence in every case; the boundary between judicial notice of “realities of human experience” and impermissible resort to extra-curial material remains to be worked through.
Another open question is the minimum probative threshold for object-identification evidence in a circumstantial case that is otherwise evenly balanced. The Court held the Matheson evidence was “at best, barely admissible” but did not decide whether such evidence would ever be sufficient if the surrounding strands were marginally stronger. Future cases will need to explore where on the spectrum between “glaringly improbable” and “convincing” an identification must sit once motive and opportunity are conceded to be weak.
Finally, the decision assumes the honesty of the identifying witnesses. It does not address the situation in which contamination or suggestibility might cross into subconscious reconstruction that an honest witness cannot self-correct. How trial judges should direct juries on that distinction, and whether intermediate courts may substitute their own view on credibility where the trial judge has given orthodox warnings, are matters left for later elaboration. The judgment’s emphasis on the whole-case assessment suggests that even honest but unreliable evidence cannot be salvaged by careful jury directions if, objectively, the accumulation of improbabilities leaves a reasonable doubt.
(b) the appellant's conviction for murder be quashed; and
S C Holt QC with K B W Gover for the appellant (instructed by Anderson Fredericks Turner)
M R Byrne QC with C W Wallis for the respondent (instructed by Director of Public Prosecutions (Qld))
Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.
Criminal law - Murder - Appeal - Appeal against conviction - Where appellant convicted by jury - Where Crown case based entirely on circumstantial evidence - Where circumstantial evidence related to opportunity and motive and miscellany of other inculpatory matters - Where evidence of opportunity and motive extremely weak - Where evidence connecting accused to alleged murder weapon based on glaringly improbable identification evidence - Whether verdict unreasonable or cannot be supported having regard to evidence.
Words and phrases - "basis for an inference", "circumstantial case", "contamination of recollection", "credibility and reliability", "glaringly improbable", "identification evidence", "identification of object", "motive", "murder weapon", "opportunity", "unreasonable verdict".
KIEFEL CJ, KEANE, NETTLE, GORDON AND EDELMAN JJ. Introduction On either Monday 12 November 2012 or Tuesday 13 November 2012, Mrs Liselotte Watson was murdered in her home on Macleay Island, a small community of around 2,500 people in Moreton Bay, Queensland. At the scene of the murder, and in Mrs Watson's house generally, most of the drawers had been opened, the contents of cupboards were dishevelled, bags had been pulled out, and a large suitcase was left open. Numerous witnesses gave evidence of the common knowledge that Mrs Watson kept large sums of money in her house but only $290 was located. Mr Fennell was tried and convicted for her murder. An appeal to the Court of Appeal of the Supreme Court of Queensland was dismissed. In this Court, Mr Fennell's only ground of appeal was that the Court of Appeal erred by failing to conclude that the verdict was unreasonable or cannot be supported having regard to the evidence, in part because it made significant errors of fact. Mr Fennell's house was searched and his bank accounts were examined. No excess cash was found nor were any excess deposits identified. There was no evidence directly linking Mr Fennell to the crime scene. Neither his DNA nor his fingerprints were found at the bloody crime scene. His glasses, clothes, shoes, "postie" motorbike and utility vehicle were all examined, with nothing incriminating found. He was excluded from being a possible contributor of the DNA profiles found on a shaving bag containing Mrs Watson's bank documents which was found near other belongings of hers and a claw hammer in a mangrove area at Thompson Point, on the island. The Crown case against Mr Fennell relied upon opportunity, motive, and a miscellany of other matters said to be inculpatory. The most significant of the miscellany of other matters was evidence from two witnesses, Mr and Mrs Matheson, who purported to identify the hammer that was discovered in the mangrove area near Mrs Watson's belongings. Mr and Mrs Matheson said that the hammer had been lent by Mr Matheson to Mr Fennell. That evidence was very significant because a hammer was the likely murder weapon. The Crown case concerning opportunity and motive was extremely weak. Mr Fennell's opportunity was, at best, a very small window of time which required an assumption about the time of the murder that was contradicted by other evidence. Even with that opportunity, the evidence of opportunity and motive did not put Mr Fennell in a relevantly different position from any of the numerous other people who shared the common knowledge that Mrs Watson kept large sums of money in her house. As the Crown properly accepted in this appeal, once that conclusion is reached the evidence from Mr and Mrs Matheson linking Mr Fennell to the hammer became essential evidence that a reasonable jury would have been required to accept before convicting Mr Fennell. To use Wigmore's metaphor, the evidence from Mr and Mrs Matheson became by far the most significant of the "strands in a cable" supporting the conviction. But the evidence of Mr and Mrs Matheson should have had so little weight that, at best, it was barely admissible. At the conclusion of the hearing of this appeal, this Court unanimously ordered that the appeal be allowed, the order of the Court of Appeal be set aside and in its place order that: (i) the appeal be allowed; (ii) Mr Fennell's conviction for murder be quashed; and (iii) a verdict of acquittal be entered. The Court said that reasons would be published at a later date. These are our reasons for making those orders. Uncontroversial facts and unchallenged evidence Mrs Watson was an 85-year-old lady who lived in a two-storey house on Alistair Court which she kept very neat, and which one friend described as "immaculate". In the years before her death, Mrs Watson's mobility had become limited. She could walk, but not as far as the shops. She was assisted by the local Lions Club on Macleay Island, whose members do shopping and chores for elderly people. One member who had helped Mrs Watson, including delivering money for her to her daughter, Helen, on Lamb Island, was the then President of the local Lions Club, Mr Crook. Mr Fennell was a founding member of the local Lions Club. He was involved with the charitable activities of that club. For one to two years before her death, he also assisted Mrs Watson. Numerous witnesses gave evidence that Mr Fennell visited Mrs Watson, usually for 15 to 20 minutes, with accounts of his visits being almost every day or even up to twice a day. He would have conversations with her and assist her with her shopping and banking and, until another person took over, Mr Fennell also maintained her yard. Mr Fennell was contracted by various businesses in Moreton Bay to deliver pamphlets, which he did in his utility vehicle or on his "postie bike", which was a Honda CT110 motorbike. On Sunday 11 November 2012, a neighbour saw Mr Fennell delivering a pamphlet to Mrs Watson as part of his deliveries. On Mr Fennell's account, this was the last time that he saw Mrs Watson. Mrs Watson was seen outside her house at around 9.30 am on Monday 12 November 2012 by two witnesses, Ms Uzzell and Mr Barker. They respectively described her as wearing a grey house dress with shoes or a blue nylon dressing gown. That was the last time that Mrs Watson was seen alive. In the afternoon of Monday 12 November 2012, Mrs Watson did not answer telephone calls at 2.53 pm (from a caller with an international call prefix), 4.14 pm (from a caller with an international call prefix) and 4.22 pm (from Mr Fennell). On Tuesday morning, 13 November 2012, a friend, Ms Bowen, delivered newspapers to Mrs Watson's front verandah at 7.00 am or a little earlier. Those newspapers were never collected by Mrs Watson. At around 3.30 pm to 3.45 pm on Tuesday 13 November 2012, Mr Fennell drove to the residence of the Macleay Island police officer, Sergeant McDougall. He told Sergeant McDougall's wife that he had gone to see Mrs Watson that afternoon as he had missed an appointment to catch up earlier that day. Mr Fennell said that she had not answered her door. Mr Fennell said that he was concerned that she may have had a fall and might be injured in the house. Sergeant McDougall's wife told Sergeant McDougall that he may need to do a "welfare check". Sergeant McDougall accompanied Mr Fennell to the house. The aluminium windows on the front verandah were open about 5 cm but the front door was closed. Sergeant McDougall entered the kitchen of the house from a side door that was open. Sergeant McDougall said to Mr Fennell "It looks like a break. Don't touch anything ... Keep your hands in your pockets". Most of the drawers in her bedroom had been opened. Her television was lying on the floor. Mrs Watson's granddaughter gave evidence that there were a few cupboards where things were "dishevelled" and bags that were "pulled out that shouldn't have been or would not normally have been", and that a large suitcase, which Mrs Watson had described as her "storm bag", was open on a dining table at the far end of the lounge room. Sergeant McDougall discovered Mrs Watson on the floor in the main bedroom at the end of a hallway. The sheets were missing from the bed and an AM radio was playing with a lot of static. A forensic examination of her house with the assistance of an Australian Federal Police cash dog handler and cash dog found only $290 in the house although many witnesses gave evidence that Mrs Watson kept large sums of money in her house. Mrs Watson was found by Sergeant McDougall. She had been killed violently with a blunt murder weapon while in bed wearing her nightdress. She had "defensive" injuries to her wrist, forearm and fingers, including some with a crescent shape suggestive of the head of a hammer. She had been struck repeatedly on the back of the head, at least four and possibly six times. There was a lot of blood on her pillow. There were bloodstains on the bed, the floor, the wall behind the bed, and the curtain to the left of the bed. She was found lying on a towel on the floor, next to her bed, with a quilt partially wrapped around her. Both the towel and the quilt were saturated with blood. There were also blood splatters on the wall, on a table, on a clock radio, and on a tissue box on the table. The police later found two hammers in the house. Both of them were found in a box on a table in the entry to the kitchen. One was a standard claw hammer with a rubber handle and the other was a "mallet" type of hammer. The claw hammer had a transferred bloodstain on the handle with a DNA profile that matched Mrs Watson. At trial neither the Crown nor the defence suggested that this hammer was the murder weapon. The murder weapon was alleged by the Crown to be the claw hammer found at Thompson Point. Prior to the murder, on Saturday 10 November 2012, Ms Jensen was walking her dogs and stopped near the mangroves at Thompson Point. She saw a shaving bag sitting in the mud at low tide. It contained a rock and a folder of Westpac bank documents with Mrs Watson's name on them. The Westpac folder in the shaving bag contained a book of deposit slips and a book of withdrawal slips, each for Mrs Watson's accounts. There were printed customer receipts for withdrawals of $3,000 on 22 August 2012, $7,000 on 17 September 2012 and $3,000 on 28 September 2012. At midday on Thursday 15 November 2012, a police diver searching the water in the mangroves discovered a green Translink wallet, a black purse, and, 15 metres away, the claw hammer. The green Translink wallet contained a green transit card, two Medicare cards and two pension cards. The Medicare and pension cards were in Mrs Watson's name. The black purse was identified by Mrs Watson's granddaughter as belonging to Mrs Watson. An analysis of the DNA profiles found on the fabric adjacent to the zip of the shaving bag excluded Mr Fennell as a possible contributor of the DNA but found slight support for DNA profiles that matched three other people: Ms Jensen (who found the shaving bag), Ms Uzzell and Mr Cornell. Ms Jensen and Ms Uzzell were both called as witnesses at the trial. It was not suggested that either of them was responsible in any way for Mrs Watson's death. Mr Cornell, a postman who rode a red postie bike similar to Mr Fennell's, had died by the time of trial and it was also not suggested that he was responsible for Mrs Watson's death. It was not disputed at trial that, as Mr Fennell told the police, Mr Fennell had a brain injury from 1980 or 1981 from a house explosion which had left him badly burned and caused short-term memory loss. His wife also gave evidence of his memory problems. One witness, Mr Barker, gave evidence that Mr Fennell would often write things down and that Mr Fennell had explained that he did this because he had a bad memory. Mr Fennell provided the police with a typed timeline of his movements that he compiled with the help of his wife and by reference to a daily diary that he kept at home. Mr Fennell's timeline was as follows. He said that on Monday 12 November 2012, he telephoned Mrs Watson. After telephoning Mrs Watson and not receiving an answer, he went to her house and knocked on the door. She still did not answer. He said that he left a biscuit tin at the front door that he had brought to return to her. He then returned home, hung out washing, cleaned the kitchen, and made some telephone calls, including a long call to Redland Bay Hospital, before his wife returned from her craft class at 12.30 pm. He said that he then went out in the afternoon at around 3.00 pm to return a DVD and saw a friend, Mr Crook, in the late afternoon before returning home for dinner at 6.00 pm. The next day, Tuesday 13 November 2012, from 6.10 am he engaged in a number of tasks off Macleay Island including delivering pamphlets on Lamb Island and attending a motorcycle shop in Redland Bay. His timeline recorded that before leaving Macleay Island, he left a note in Mrs Watson's letterbox at 6.00 am saying that he would be away for the day and that he would see her that afternoon upon his return. His case was that this was informing her that he would be unable to meet her for tea at 10.30 am. Mr Fennell's account of his movements on Monday 12 November 2012 was corroborated by independent evidence. A summary of phone records showed a phone call from Mr Fennell's mobile to Mrs Watson's land line at 9.29 am which was unanswered. In his opening and closing addresses, counsel for the Crown suggested that her presence outside, as seen by Ms Uzzell and Mr Barker, might have been the reason that this phone call was unanswered. A neighbour, Mrs Doolan, gave evidence that she saw Mr Fennell come to Mrs Watson's house on his postie bike on Monday 12 November 2012 but that Mr Fennell "only stayed for a little while". Mrs Doolan said that he did not seem to stop in at Mrs Watson's house. A friend, Ms Vaney, gave evidence that, when taking a walk that Monday with her husband, which he confirmed, she saw Mr and Mrs Fennell and spoke with Mr Fennell at some time between 2.30 pm and 3.30 pm. Mr and Mrs Crook gave evidence, accepted by the Crown as "[n]ot controversial", that Mr Fennell visited them in the afternoon for a cup of tea and a chat, staying for half an hour to an hour, during a television show and leaving as the news came on at 6.00 pm. Finally, evidence of the use of the computer at Mr Fennell's home strongly supports the presence of Mr Fennell at his home from 6.20 pm. That evidence is discussed in detail later in these reasons. Mr Fennell's movements in his timeline on Tuesday morning were also corroborated by independent evidence. Mr Doolan's evidence was that Mr Fennell arrived at around 5.45 am on Tuesday 13 November 2012, and then "scooted off" within two to five minutes. The note that Mr Fennell described leaving in Mrs Watson's letterbox on Tuesday 13 November 2012 was recovered from the letterbox by Detective Senior Constable Strang. It read: "Mrs Watson, I telephoned yesterday to tell you that I have to go to the mainland today. I will call in on my return to get you any shopping that you need". At around 7.30 am to 7.45 am, after he had completed his deliveries, Mr Fennell was seen on Lamb Island by the owner of a convenience store on that island, for whom Mr Fennell delivered pamphlets. Around 11.00 am he was seen by the owner of a motorcycle repair business wearing a high-vis vest when Mr Fennell brought his postie bike in for repair. Mrs Fennell's evidence concerning Monday 12 November 2012 was also consistent with Mr Fennell's timeline. Her evidence was that she returned home from her craft class and Mr Fennell was at home with her until around 2.30 pm when he left, saying that he was going to return a DVD and to see Mr and Mrs Crook. Mrs Fennell also gave evidence that she had remained at home from the time Mr Fennell left until he returned at around 5.45 pm in time for their usual dinner at 6.00 pm and then Mr Fennell went through his usual routine before going to sleep. However, there was a clear omission from the timeline that Mr Fennell gave to the police. For years, Mr Fennell had gambled money and concealed his gambling from his wife. On Monday 12 November 2012, he was captured on closed-circuit television stills at Pub Paradise at times between 3.49 pm and 4.43 pm. He was wearing a high-vis vest and he placed bets in the TAB area and was embraced by others. After Mr Fennell was arrested on Friday 16 November 2012, the police conducted a search of his house. Nothing linking him to the murder was identified. The police also examined his bank accounts for the period from July 2010 to 19 November 2012. No excess cash deposits were identified. Police installed a listening device in his house and in the motel in which he stayed while his house was being searched. Nothing incriminating was heard. Police conducted forensic examinations of Mr Fennell's glasses, his clothes, his shoes, his postie bike and his utility vehicle. They had Mr Fennell's fingerprints on file and took DNA samples from him. They examined Mr Fennell's head, face, arms, upper body and legs, and took a series of photographs of him. Again, nothing linking Mr Fennell to the murder was located from any of these searches or examinations. A fingerprint of Mr Fennell's was located on the outside of Mrs Watson's kitchen window but, with the frequency of Mr Fennell's visits to Mrs Watson's house and the innocuous location of the fingerprint, the Crown did not suggest that the fingerprint was incriminating. The Crown case and its weaknesses The Crown case at trial was that Mr Fennell had stolen money from Mrs Watson and that he murdered her to avoid detection. The Crown case relied on three strands of circumstantial evidence: opportunity, motive, and a miscellany of other matters. Opportunity The Crown case was that Mrs Watson was murdered sometime after she had been seen at 9.30 am on Monday 12 November 2012 but sometime before Mr Fennell was captured on CCTV at Pub Paradise at 3.49 pm. A pathologist gave evidence that in estimating when, in the period between 9.30 am on 12 November 2012 and 4.00 pm on 13 November 2012, Mrs Watson was killed, he "preferred" a time of death at some time on Monday 12 November 2012 but he was unable to be precise about the time of death on that day. The Crown told the jury that they might think that Mr Fennell's movements on Tuesday 13 November 2012 were "reasonably well accounted for". Mr Fennell's movements were also well accounted for from 3.49 pm on Monday 12 November 2012. As explained above, the CCTV evidence placed Mr Fennell at Pub Paradise between 3.49 pm and 4.43 pm. The evidence of Mr and Mrs Crook placed Mr Fennell at their house between around 5.00 pm and shortly before 6.00 pm. The evidence of Mrs Fennell was that Mr Fennell returned home for dinner on Monday 12 November 2012 at the usual time of 6.00 pm and stayed at home that night. The evidence of computer usage associated with Mr Fennell provided a further strong basis for an inference that he was at home from 6.20 pm and left home at around 5.45 am the next day. There were therefore substantial obstacles to any conclusion that Mr Fennell had murdered Mrs Watson at any time after 3.49 pm. The Crown case was that Mrs Watson was murdered during the day on Monday 12 November 2012 in a window of time when Mr Fennell had the opportunity of committing the murder. Although Mrs Watson was found on the floor next to her bed in her nightdress, the Crown case was that this did not necessarily mean that the murder occurred at night. She was located with her dentures, or false teeth, intact in her mouth and her treating general practitioner gave evidence that there were occasions in 2012, including one in October 2012, when Mrs Watson was in her nightdress when a house call was made. Counsel for the Crown also pointed to the lack of any forced entry into the house and to three unanswered phone calls, at 2.53 pm and 4.14 pm (both with an international call prefix and suggested to be from "overseas telemarketers") and at 4.22 pm (from Mr Fennell). However, on this appeal, senior counsel for Mr Fennell pointed to Mrs Watson's limited mobility in a two-storey house as another possible explanation for the unanswered calls. There was also evidence of Mrs Watson's frustration with phone calls. In the days prior to her murder she told her neighbour, Ms McKie, that she had been receiving threatening calls and she had previously told her general practitioner that she had been receiving nuisance phone calls. Apart from the discovery of Mrs Watson on the floor next to her bed and wearing her nightdress, a further difficulty for the Crown theory that Mrs Watson was murdered during the daytime, and before 3.49 pm, was the evidence of Mr Matheson on this subject. Mr Matheson gave evidence that at 1.30 am or 2.00 am on Tuesday 13 November 2012 his dog woke him and his wife up by barking and standing underneath the window that faces Alistair Court. Mr Matheson said that was unusual because his dog had been trained not to bark and that it took a couple of hours to quieten her down. He said that while in bed, at around 2.30 am or 3.00 am, he heard the sound of a car travelling at great speed down Alistair Court and straight to the end of his street, which is opposite Alistair Court. He said that the car was a six-cylinder Holden or Ford, not a Japanese car with a "tinny exhaust note". This would have excluded Mr Fennell's Mitsubishi utility vehicle. In closing, counsel for the Crown said to the jury that Mrs Watson was "certainly not killed at 1 am or 2 am on the Tuesday" because her dentures were still in her mouth. The Crown theory that Mrs Watson was killed in the daytime also required the jury to conclude that evidence from Ms Dobson was mistaken. Ms Dobson lived three doors down from Mrs Watson and they spoke from time to time. In between Ms Dobson's house and Mrs Watson's house were Mr Dallas' house and Ms McKie's house. Ms Dobson said that at about the time in the afternoon of Monday 12 November 2012 that Mrs Watson would usually feed the birds, which was about 4.00 pm, she heard a car arrive and a dialogue between a man and a woman. She recognised the man's voice as Mr Dallas' and thought that the woman's voice sounded like Mrs Watson's (whose distinctive accent Ms Dobson knew). Ms Dobson glanced through her window and saw Mr Dallas at the corner of his property with a lady with grey hair. She saw the lady from behind and thought that she was Mrs Watson. Ms Dobson said that she identified the day as Monday because it was a nice sunny day, unlike the previous day, which was very windy. In his closing address to the jury, counsel for the Crown suggested that it was likely that Ms Dobson had seen and heard Mr Dallas on Sunday 11 November 2012, not Monday 12 November 2012, and that Mr Dallas had been speaking to his mother, not Mrs Watson. Mr Dallas' evidence was that his mother had visited on the weekend before Mrs Watson died. His mother left at midday on Sunday and he had a conversation at the front of his house that Sunday afternoon with Mrs Watson. The Crown case relied upon three occasions on Monday 12 November 2012 when witnesses said that they saw Mr Fennell, or his bike or his utility vehicle, at Mrs Watson's house, including two occasions when it was said that he could have been there for more than an hour. The evidence of those two occasions came from Mr Robinson and Mrs Doolan. The evidence of the other occasion came from Ms McKie, who suggested that Mr Fennell had been at Mrs Watson's house for 20 minutes. Mr Robinson's evidence was extremely problematic. By closing submissions the Crown case concerning opportunity placed little reliance upon him. The focus of the Crown case was that (i) Mr Fennell had killed Mrs Watson in her bed at around 2.00 pm in the 20-minute interval when he was seen by Ms McKie, and (ii) Mr Fennell had returned to the scene of the crime at 6.00 pm, when he was seen by Mrs Doolan, to "search or clean or tidy or do whatever else was necessary in order to put the crime scene into the shape in which it was found". (1) Mr Robinson Mr Robinson's evidence was that he saw a utility vehicle parked outside Mrs Watson's house at about 11.00 am on Monday 12 November 2012. Mr Robinson's alleged sighting of Mr Fennell's utility vehicle occurred, on his account, in the brief moment that he was driving past the end of Alistair Court, the street on which Mrs Watson lived. Any sighting would have been fleeting, of a vehicle 80 to 100 metres away, in circumstances in which there was no particular importance to it and in which Mr Robinson acknowledged that he "didn't pay any attention to it". Mr Robinson recognised the vehicle more than a year later in March 2014. On appeal to this Court, senior counsel for the Crown abandoned any reliance upon the evidence of Mr Robinson. He was correct to do so for four reasons. First, apart from the fleeting nature of the 11.00 am observation by Mr Robinson that was said to be of Mr Fennell's utility vehicle, there were significant discrepancies between Mr Robinson's description of the vehicle and Mr Fennell's utility vehicle. In particular, Mr Robinson described the utility vehicle that he saw as having Mr Fennell's name on the back of it. Mr Fennell's vehicle did not have his name on it. Secondly, it was more than a year after Mr Robinson saw the vehicle that he claimed to have subsequently recognised and identified it as Mr Fennell's vehicle. Thirdly, Mr Robinson's evidence was contrary to the unchallenged evidence of Mrs Fennell that she drove the utility vehicle that morning to the craft group that she attended every Monday from 8.45 am until 12.30 pm. Fourthly, and perhaps most significantly, in cross-examination Mr Robinson accepted that his purported recognition of the vehicle was "simply a reconstruction after the event". (2) Ms McKie The second witness that the Crown relied upon for opportunity was Ms McKie. Ms McKie had known Mrs Watson for 15 years. Ms McKie's house was separated from Mrs Watson's by one other house. Ms McKie's evidence was that at about 2.00 pm on Monday 12 November 2012 she looked out her bathroom window and saw a man whom she did not know, but whom she had seen delivering pamphlets and attending Mrs Watson's house regularly, arrive at Mrs Watson's front gate, park his "little red bike", and go into Mrs Watson's house. She did not see him leave but she said that she heard his bike start about 20 minutes later. The evidence of Ms McKie which suggested that Mr Fennell was at Mrs Watson's house at around 2.00 pm on Monday 12 November 2012 was contrary to Mr Fennell's timeline, which did not record him visiting Mrs Watson's house on Monday afternoon. It was also contrary to the evidence of Mrs Fennell and Ms Vaney, who were with Mr Fennell until at least 2.30 pm. One reason Ms McKie placed the 2.00 pm visit on Monday was that she said that she thought that she had heard Mr Fennell say to Mrs Watson on Sunday that he would be there on the Monday afternoon. Although Mr Fennell's counsel informed the trial judge that he was going to be running the argument that Ms McKie "clearly" was mistaken, it was not put to Ms McKie in cross-examination that the arrangement that she overheard was for Mr Fennell to meet Mrs Watson on Tuesday, about which Mr Fennell left a note of his apologies in Mrs Watson's letterbox at 6.00 am on Tuesday. Indeed, on Mr Fennell's timeline he said that he had gone to Mrs Watson's house on Tuesday 13 November 2012 shortly after he arrived back on Macleay Island from the mainland at 2.00 pm, knocked on the front door and walked around the back to see if she was hanging out washing. (3) Mrs Doolan The third witness who gave evidence relied upon by the Crown for opportunity was Mrs Doolan. Mrs Doolan also lived on Alistair Court and she could see Mrs Watson's house from her front patio. As discussed above, Mrs Doolan gave evidence, which supported Mr Fennell's timeline, that she saw him at Mrs Watson's house where he stayed very briefly in the morning of Monday 12 November 2012. However, she said that she also saw Mr Fennell again at 6.00 pm while she was watching the news and cooking tea. She said that she recognised his utility vehicle and saw him walk inside the gate. She said that she saw the utility vehicle leave at 7.30 pm when the X Factor television program started. One difficulty with this evidence from Mrs Doolan was that Mrs Fennell said that Mr Fennell had returned home for dinner at 5.45 pm and that they stayed at home. A further, and seemingly insurmountable, difficulty was the evidence of a police computer analyst. That evidence provided an extremely strong basis for an inference that Mr Fennell was at home, as Mrs Fennell had said, at the times described by Mrs Doolan. At 6.11 pm a user of the computer at Mr Fennell's home, who the Crown suggested was probably Mr Fennell's son, had logged out of the website Moshimonster.com. Then, between 6.20 pm and 7.25 pm, the Fennells' home computer was used in a manner from which a powerful inference could be drawn that the user was Mr Fennell: (i) at 6.21 pm a user logged in to the home computer with the username Steven and the ID islandlife59; (ii) the user logged in to a linkedin account with the name of Steven Fennell, and viewed and replied to messages in the inbox associated with that linkedin user; and (iii) the user visited various websites, conducted searches for burns recovery news and visited a website associated with fire burns. Mr Fennell had told the police in his record of interview that he had been burned in the explosion in 1980 or 1981 and had third degree burns to both legs. Summary of the weakness of the Crown case concerning opportunity No reasonable jury could have accepted the evidence from Mr Robinson or Mrs Doolan concerning opportunity. That left only the evidence of Ms McKie, who may have been mistaken about the day. But even if Ms McKie's evidence were accepted, the state of the crime scene casts serious doubt upon whether Mr Fennell could have had the opportunity to commit the crime in the short period of time that she described. Ms McKie also said that the man she saw did not take anything in with him and that she heard no noises or anything unusual during the 20-minute period when she thought he was in the house. Yet, on the Crown case, Mr Fennell must have gone to Mrs Watson's house at 2.00 pm with a hammer to commit a premeditated murder of Mrs Watson in her bed wearing her nightclothes. Within that 20-minute period, he must have done the following: committed the murder, caused the disturbance to the drawers and the house, and either driven off in bloody clothes on his postie bike or disposed of his clothes in a place where they were not found. On this appeal, senior counsel for the Crown also properly accepted that Mr Fennell would necessarily have had to shower and clean himself up despite Mrs Fennell being present in his house and not seeing him. He would also need to have concealed or disposed of the hammer before arriving at Pub Paradise, where he was captured on CCTV at 3.49 pm wearing a high-vis vest that he would sometimes wear to the hotel. He was again wearing a high-vis vest when he delivered pamphlets the next day, Tuesday 13 November 2012. The short period of time identified by Ms McKie, combined with the CCTV footage at 3.49 pm, made the evidence of Mrs Doolan particularly important to the Crown case concerning opportunity. The Crown relied upon the evidence of Mrs Doolan to show that Mr Fennell had returned to the crime scene at 6.00 pm for more than an hour. But that evidence was contradicted by the evidence of computer usage at Mr Fennell's house, Mrs Fennell's evidence, and Mr Fennell's timeline. At best, the evidence of opportunity showed only that Mr Fennell had briefly visited Mrs Watson's house on Monday 12 November 2012. For two reasons, that evidence of opportunity was, at best, a very weak strand in the Crown's circumstantial case against Mr Fennell. First, Mr Fennell's presence in Mrs Watson's house was entirely unremarkable since he visited her up to twice daily. Secondly, since the Crown case was that Mrs Watson was killed during the daytime on 12 November 2012, the same opportunity existed for any person who could have attended Mrs Watson's house on that day. Any opportunity that Mr Fennell had was a factor that barely set him apart from other members of the population of Macleay Island. Motive The Crown alleged that Mr Fennell's motive to murder Mrs Watson was to conceal thefts from Mrs Watson that she was soon to discover. First, in closing submissions to the jury counsel for the Crown observed that Mr Fennell was a gambler and that he had suffered gambling losses, such as $1,207 on Monday 12 November 2012. Secondly, the Crown pointed to withdrawals from Mrs Watson's bank account by Mr Fennell totalling $24,000 over three months. Thirdly, the Crown relied upon a withdrawal of $8,000 from Mrs Watson's account on 2 November 2012 where the withdrawal slip appeared to have been altered from $3,000 to $8,000. Fourthly, the Crown relied upon the disposal at Thompson Point of banking documents belonging to Mrs Watson. (1) Mr Fennell's gambling In his closing address to the jury, counsel for the Crown relied upon expert accounting evidence that had shown that Mr and Mrs Fennell's expenditures were almost equal to their income. He invited the jury to consider where the additional money could have come from to fund Mr Fennell's gambling. Witnesses gave evidence of regular gambling for at least five years by Mr Fennell at the TAB in Pub Paradise, usually on Monday or Tuesday, and anywhere from $50 to $500 although usually $50 to $100; he rarely drank alcohol and would sometimes win a couple of thousand dollars. The difficulty with this aspect of the Crown case was that there was no evidence from which conclusions could be drawn about Mr Fennell's net position from gambling prior to the time when he commenced Mrs Watson's banking or during that period. Large wins of $500 to $1,000 or more would be paid by the TAB into Mr Fennell's EFTPOS card but smaller wins would be paid in cash. As to the larger amounts paid electronically, the forensic accounting evidence from Ms McKinnon was that between July 2010 and November 2012 there was $25,880 transferred electronically from the TAB into Mr Fennell's account based upon staked amounts of $12,608. While some evidence was led as to cash bets placed at the TAB at Pub Paradise on Friday 9 November 2012, the Crown only relied upon evidence of amounts of cash that were gambled by Mr Fennell between 3.51 pm and 4.35 pm on Monday 12 November 2012. On that occasion there were 12 transactions at the TAB in Pub Paradise for a total amount of $1,357. Only one of those bets, for $80, was successful, returning a dividend of $149. But the results of one afternoon can hardly be taken as representative of a net position for months, still less over a period of years. The only broad conclusions that can be drawn from the evidence are that Mr Fennell's general gambling habits had not apparently changed and that any overall losses that Mr Fennell suffered over the period of his gambling appeared to be sustainable. By 12 November 2012, Mr and Mrs Fennell had paid more than $4,500 in additional voluntary payments of their mortgage and they had around $2,700 available to access across their accounts. (2) The withdrawal of $24,000 from Mrs Watson's account As to the withdrawal of $24,000 over three months between August and early November 2012, it occurred in amounts of $3,000, $7,000, $3,000, $3,000, and $8,000. Although the total of $24,000 withdrawn in those months was larger than usual, it was not extravagantly so. An exhibit summarising Mrs Watson's bank statements from November 2007 until November 2012 showed that over those five years she withdrew $148,297, usually in amounts between $2,000 and $5,000. Ms McKie also gave evidence that shortly before Mrs Watson was murdered Ms McKie had overheard Mrs Watson in her backyard speaking to tradespeople about "doing a pergola". The withdrawals amounting to $24,000 in the three months before her death in November 2012 had, after taking into account some deposits, reduced Mrs Watson's net balance from around $25,000 to around $8,000. But this was not novel. Between March and May 2010, she withdrew $18,000, which, after taking some deposits into account, reduced her net balance from around $22,500 to around $10,900. One of the withdrawals in the period from March to May 2010 was of $11,000. Mrs Watson told various of her friends that she had lent $11,000 to a Mr Holden and he had not repaid the money, although he eventually did. Most fundamentally, however, the evidence pointed to a significant likelihood that each of the five withdrawals that amounted to $24,000 was authorised by Mrs Watson. Although Mr Fennell asked the bank for authority to sign for withdrawals from Mrs Watson's account, he was never given that authority. The parties to this appeal helpfully presented an agreed position covering all of the evidence, including that of the handwriting expert called by the Crown. That agreed position can be summarised as follows: i. 22 August 2012: Mr Fennell wrote on the front "3,000.00", "22 / 08 / 12", "THREE THOUSAND DOLLARS ONLY". Mrs Watson signed the front and back of the slip. On the reverse side, she wrote "Permission given to Steven Fennell to cash this withdrawal". ii. 17 September 2012: Mr Fennell wrote on the front "7000.00", "17 / 9 / 12", "SEVEN THOUSAND DOLLARS ONLY". Mrs Watson signed the front and back of the slip. On the reverse side, she wrote "Please pay my agent". iii. 28 September 2012: Mr Fennell wrote on the reverse side "PLEASE PAY MY AGENT STEVEN FENNELL". Mrs Watson signed the front and back of the slip. On the front, Mrs Watson wrote "3000", "28 / 9 / 12", "three thousand dollars". iv. 5 October 2012: Mrs Watson signed the front and back of the slip. On the front she wrote "3000", "5 / 10 / 12", "three thousand dollars". On the reverse side, she wrote "Please pay my Agent". v. 2 November 2012: On the front, Mr Fennell wrote "02 / 11 / 12" and "EIGHT THOUSAND DOLLARS ONLY". The author of the figure "8000.00" on the front of the slip was indeterminable. Mrs Watson signed the front and back of the slip. On the reverse side, Mrs Watson wrote "PLEASE PAY MY AGENT STEVEN FENNELL". Further evidence of Mrs Watson's authority for the $24,000 of withdrawals between late August and early November was given by various employees of the Westpac branch where her account was held. One bank teller said that she telephoned Mrs Watson to confirm the first ($3,000) transaction and identified a note that she made to that effect on the back of the cheque. That teller also permitted the second ($7,000) withdrawal but she was not asked whether she had telephoned Mrs Watson on that occasion. Another teller gave evidence in relation to the fourth withdrawal ($3,000) that she waited with Mr Fennell while her supervisor telephoned Mrs Watson, although she could not recall whether she had telephoned Mrs Watson in relation to the third withdrawal ($3,000). A further teller, Ms Kershaw, was responsible for the final ($8,000) withdrawal. Ms Kershaw approached her supervisor, Ms Kickert, to discuss the transaction. Ms Kickert authorised the transaction but said that she believed that she had told Ms Kershaw to telephone Mrs Watson to verify the amount. Ms Kershaw confirmed from the words "sol auth" on the cheque that she looked up the computer to check Mrs Watson's signature. She recalled speaking with Ms Kickert about the withdrawal but could not recall whether she had telephoned Mrs Watson as instructed. The reasonable inference is that she would have made the telephone call as instructed. There was also evidence from a friend of Mr Fennell's, Mr Leahy, who went with him to the mainland on several of the occasions when one stop was for Mr Fennell to do Mrs Watson's banking. It was never suggested that Mr Leahy was part of a conspiracy to steal from Mrs Watson. Although Mr Leahy did not go into the bank, he described how Mr Fennell would bring with him a black pencil case. On one occasion, after the banking, Mr Leahy saw Mr Fennell go into Mrs Watson's house with the black pencil case and return without it. (3) The alteration of the $8,000 withdrawal slip Evidence was given by the handwriting expert that an amount of $3,000 on the 2 November 2012 withdrawal slip had been altered to $8,000 by changing the number three to an eight. The expert evidence was the subject of a strong attack but the likelihood of alteration seems apparent from ordinary examination of the withdrawal slip. The expert could not say who had changed the number or in what circumstances. However, before a conclusion could be reached that Mr Fennell had made that alteration and withdrawal without authority it would be necessary to reject the reasonable inference that Mrs Watson was telephoned by Ms Kershaw. Other matters also militate against a conclusion that Mr Fennell made the alteration without authority. One matter is the likelihood of discovery. Mr Fennell knew that Mrs Watson had been telephoned to confirm her authority on most, if not all, previous occasions, including the occasion immediately prior to him presenting the $8,000 withdrawal slip. He was also aware that, although very generous with her money, Mrs Watson paid close attention to it. Before she involved the police, Mrs Watson had also enlisted Mr Fennell's assistance to recover the $11,000 that she lent to Mr Holden. In Mr Fennell's record of interview with the police he said that Mrs Watson was "both extraordinarily generous and extraordinarily cheap ... to the tune of thriftiness of things". Counsel for the Crown also observed in his closing address that the $8,000 withdrawal occurred on the same day, 2 November 2012, that the new manager of the Russell Island IGA store, Mr Jones, cancelled Mr Fennell's pamphlet delivery contract, which deprived Mr Fennell of about $347 a week in income for his business. That observation was incorrect. The evidence from Mr Jones was that he spoke with Mr Fennell on Monday 5 November 2012 and paid him later that week. However, Mr Fennell had been on notice from Mr Jones since 22 October 2012 that the IGA contract could be lost. But the loss of the income from that contract, while significant, would not have been devastating and there was no evidence concerning whether Mr Fennell could have replaced that contract with other work. Conclusions about motive Ultimately, the evidence led by the Crown in relation to motive placed Mr Fennell in a position that was little different from any of the others on Macleay Island who had the common knowledge that Mrs Watson kept significant cash in her house and who might have had the opportunity to steal from her, but about whom there was never any suggestion of suspicion. The miscellany of other matters The third strand to the Crown case was a miscellany of allegedly inculpatory matters: (1) Mr Fennell's failure to mention to the police that he was gambling at the TAB in Pub Paradise on the afternoon of Monday 12 November 2012; (2) Mr Fennell's statements in his timeline and police interview about the biscuit tin that he said that he returned to Mrs Watson; (3) evidence that Mr Fennell clicked on a link on the Yahoo homepage; (4) the $290 in cash left in an envelope in an opened drawer at Mrs Watson's house; (5) the items belonging to Mrs Watson that were discovered at Thompson Point; and (6) the evidence of Mr and Mrs Matheson about the hammer that was found at Thompson Point. The sixth point was of particular importance. As the Crown properly accepted in this appeal, once the conclusion is reached that the Crown case on opportunity and motive was weak, Mr and Mrs Matheson's evidence about the hammer became essential evidence that a reasonable jury would have been required to accept before convicting Mr Fennell. Before dealing with that evidence, it is necessary to deal with the other allegedly inculpatory matters. (1) Mr Fennell's failure to mention his gambling at the TAB One matter, upon which counsel for the Crown relied as placing "serious doubt" upon Mr Fennell's accounts to the police of his movements, and reiterated by the trial judge in his directions to the jury as an allegation of a "telling omission", was the failure of Mr Fennell to mention in his police interview his visit on the Monday afternoon to the TAB. There is an obvious explanation for this omission that was not inculpatory. That explanation is the uncontradicted evidence of Mr Fennell's desire to conceal his gambling from his wife. Mr Papps was a good friend of Mr and Mrs Fennell for more than ten years. At about the time of Mrs Watson's murder, Mr Papps had a discussion with Mr Fennell in which Mr Fennell expressed concern that his wife would find out about his gambling. The obvious reason for the omission is that Mr Fennell's wife had assisted him to prepare the timeline for police and he was anxious to conceal his gambling from her. In any event, in light of the weakness of the Crown case concerning motive, the omission tended to undermine the Crown case rather than to support it. The CCTV footage provided Mr Fennell with an alibi from 3.49 pm for what ultimately became the Crown case that he murdered Mrs Watson on Monday 12 November 2012 in the period after he left home at 2.30 pm and before he arrived at the home of Mr and Mrs Crook between 5.00 pm and 5.30 pm. Mr Fennell's choice not to mention information that could have provided him with this partial alibi is not a circumstantial fact that could support any inference of guilt. (2) The biscuit tin The Crown also relied on what was described as "key" evidence from Mr Crook and Mrs Watson's granddaughter concerning the biscuit tin returned by Mr Fennell. In Mr Fennell's timeline and recorded interview, he said that on Sunday 11 November 2012, Mrs Watson had given him biscuits in a tin and that he washed the tin and put the biscuits in a container. He said that he visited Mrs Watson to return the tin at his wife's suggestion. His evidence in this respect was corroborated by Mrs Fennell, although she described the tin as containing cake rather than biscuits. There was a photograph referred to in evidence at trial, but not reproduced on this appeal, showing a receipt on the kitchen bench in Mrs Watson's house which referred to a purchase of two packets of chocolate biscuits on Sunday 11 November 2012. Ms McKie also described how Mrs Watson would serve biscuits in a deep container. However, counsel for the Crown, in closing, queried why Mrs Watson would use an antique tin to put biscuits in for Mr Fennell. In Mr Fennell's timeline he said that he had left the tin at Mrs Watson's house on the morning of Monday 12 November 2012 after his wife left for her craft class at 8.50 am. In his record of interview he was unsure whether he returned the tin in the morning of Monday 12 November 2012 or on Tuesday 13 November 2012. He said in his record of interview that he left the tin on top of a newspaper, which he assumed was Monday's newspaper. Mrs Watson's newspapers were delivered by a friend, Ms Bowen, before 7.00 am on both Monday and Tuesday. On Tuesday, Ms Bowen delivered Mrs Watson's newspapers, a local paper and the Courier Mail, to Mrs Watson's front verandah. She did not notice the presence of a tin or the Courier Mail newspaper she delivered on Monday but when shown a photograph of the front verandah taken after the murder she said that there were more papers on the verandah than she had left on Tuesday. The biscuit tin was seen by Sergeant Airlie when she conducted a forensic examination of the front entrance to the house. She described a copy of the Courier Mail newspaper, dated Tuesday 13 November 2012, next to the tin with the local paper, also dated Tuesday 13 November 2012, on top of the tin "with just a little bit [of the tin] poking out", showing as a "tiny corner" in a photo which was not tendered at trial. A photo that was tendered clearly showed the biscuit tin but that photo was described by Sergeant Airlie as having been taken after someone had moved the papers. Mrs Watson's granddaughter identified the tin as one that had been used "at one point" to keep receipts and other things in and said that she had not seen it with biscuits or food in it. Mr Crook also said that a tin that he saw on one occasion containing money and papers was one that "looked the same" as the tin that Mr Fennell had returned although he could not "verify that it was exactly the same tin". One difficulty with the Crown evidence in relation to the tin was the evidence from Mrs Watson's granddaughter that there were probably more than ten tins that Mrs Watson kept around her house of different shapes and sizes. Further, Mr Crook described the occasion that he saw the tin, when there was money and banking related documents in it, as being concerned with Mrs Watson's granddaughter's 18th birthday. That birthday was in August 2010, more than two years before Mrs Watson was killed. Even more importantly for the defence, the Crown submissions concerning the tin tended to undermine, rather than to reinforce, the Crown theory that Mr Fennell had stolen from Mrs Watson and disposed of her belongings in the mangroves at Thompson Point. As senior counsel for Mr Fennell expressed the point on this appeal: "Why steal the tin and then return it?" (3) The click on a link on the Yahoo homepage The Crown also relied upon evidence from the police computer analyst that a person logged in to the Fennells' home computer at 7.03 am on Monday 12 November 2012 with the username Steven and the ID islandlife59. At 7.47 am the user clicked on a link on the Yahoo homepage for an article entitled "Weird Places People Hide Money Around the Home". The Crown alleged, and there seems little doubt, that the user was Mr Fennell. The evidence that Mr Fennell clicked briefly on an article about money being hidden in the home is of almost no weight. He had not searched for that article. The article was dated 10 November 2012. Its gist was that hiding money around the home, including in false walls, could lead people to miss out on "great capital gains from investment accounts". The user navigated back to the Yahoo homepage within a second, although it is possible that the article could have remained open in a different tab. (4) The $290 in cash left in an envelope in an opened drawer A forensic examination of Mrs Watson's house conducted by Detective Senior Constable Strang with the help of an Australian Federal Police cash dog handler and cash dog found only $290 in an envelope in the top drawer of a chest of drawers in the master bedroom of Mrs Watson's house. In closing submissions, the Crown relied upon the discovery of $290 to suggest to the jury that the scene had been staged by Mr Fennell as a burglary to disguise his true motive of killing Mrs Watson so that she would not discover that he had stolen from her bank account. Counsel for the Crown said: "[w]hat kind of incompetent killer/burglar misses something like that, unless the motivation of the person doing this is something other than searching for jewellery or cash or things that might be able to quickly be pawned at the local hock shop in order to make some money?" The discovery of only $290 tended to support the likelihood of a burglary rather than undermine it. This is because of the substantial evidence that Mrs Watson kept large sums of money in the house. Properly characterised, the evidence suggested that the killer was likely to have taken large sums of cash and to have left behind only $290. The considerable evidence that Mrs Watson kept large sums of cash in her house was as follows. Mrs Watson's granddaughter, Emma, said that she knew there was a large sum of money in Mrs Watson's house. Emma gave evidence that four or five times a year, at Mrs Watson's house, Mrs Watson would give Emma's mother (Mrs Watson's daughter) amounts of money of between $1,000 and $3,000. Mr Robinson, who had come to know Mrs Watson well over the years, said that he knew that she had a lot of money in her house. He said that he had told her several times that she should put her money in the bank, although he knew that she did not trust banks. Mr Uzzell, who delivered groceries to Mrs Watson, said that she paid him in cash, sometimes from an envelope kept in a drawer and sometimes from cash kept in the drawer. Mrs Doolan remembered an occasion when she saw ten or more bundles of $50 notes on Mrs Watson's kitchen bench. Ms McKie said that it was common knowledge that Mrs Watson kept lots of money in her house. Mr Doolan said that Mrs Watson had told him and his wife that she had lent $11,000 to a person (presumably Mr Holden) who had done work for her at her house and that the money was repaid after she spoke to the police. (5) The items discovered at Thompson Point Counsel for the Crown relied upon the discovery by Ms Jensen at Thompson Point, prior to the murder, of the shaving bag containing Mrs Watson's bank deposit and withdrawal slips (including the withdrawal slips dated 22 August 2012, 17 September 2012 and 28 September 2012 but not the withdrawal slips dated 5 October 2012 and 2 November 2012) and the discovery after the murder by police divers of the green Translink wallet and the black purse belonging to Mrs Watson. The significance of the discovery of Mrs Watson's belongings at Thompson Point, near a hammer which might have been the murder weapon, went no higher than to give rise to the possibility that the murder was associated with theft from Mrs Watson. This might have increased the strength of a circumstantial case against Mr Fennell if there had been any force in the Crown case that he had stolen from her. But, as explained above, the allegation that Mr Fennell stole from Mrs Watson was no stronger than one that could have been made against any of the other persons who shared the common knowledge that she had substantial sums of money in her house and who could have accessed her home. Further, although Mr Fennell mentioned in his record of interview that Mrs Watson "was the sort of person" to pay a tradesperson from money that she kept in the shaving bag, there was no evidence that he had ever been in possession of the shaving bag, and his DNA was excluded as a possible contributor of the DNA profiles found on that bag. (6) The Thompson Point hammer There was evidence given at trial concerning a number of different hammers. Two hammers were found at Mrs Watson's house: a standard claw hammer with a rubber handle (which had a transferred bloodstain of Mrs Watson's blood on the handle) and a "mallet" type of hammer. A further five hammers were found at Mr Fennell's house; four of them were in the shed with lots of other tools and were sent for analysis. Another "quite rusty" timber-handled hammer was found under the slats of his house, photographed but not sent for any analysis. There was no suggestion at trial that any of these hammers was the murder weapon. The alleged murder weapon was the claw hammer found at Thompson Point, which was also an ordinary or generic claw hammer. The Crown relied upon the location of the hammer, about 15 metres from Mrs Watson's belongings, to associate it with the murder. Neither DNA nor fingerprints were found on the Thompson Point hammer. A photograph of the claw of the hammer shows extreme rust. Counsel for the Crown submitted that the salt water "would account for some of that". But no tests were done on the hammer to determine whether to the extent that the rust came from salt water it was possible to have occurred over the less than three days between the time that the Crown alleged that the hammer had been thrown into the water (at the earliest in the afternoon of Monday 12 November 2012) and the time when the hammer was discovered (around 12.00 noon on Thursday 15 November 2012). The location of the hammer at Thompson Point provided only an extremely weak basis to associate Mr Fennell with the murder. It showed only that a hammer consistent with the murder weapon was discovered within 15 metres of banking documents belonging to Mrs Watson to which Mr Fennell once had access - although there was no evidence that he had ever possessed the shaving bag in which they were found, and he was excluded as a possible contributor of the DNA profiles located on that bag. The evidence which was relied upon to establish a real connection between Mr Fennell and the hammer came from Mr and Mrs Matheson. In his closing address, counsel for the Crown described the evidence from Mr and Mrs Matheson as "important" and of "particular significance". He said: "Once you conclude that that's [Mr Matheson's] hammer as indicated to you by [Mr] Matheson, ladies and gentlemen, then you might think that it's very comfortable from that point to draw the conclusion that Mr Fennell was involved and, indeed, was the killer of Mrs Watson." The trial judge then directed the jury that "[t]he case against the defendant depends to a significant degree on the correctness of the identification of this hammer by each of Mr and Mrs Matheson". Mr Matheson gave evidence that he was watching the 6.00 pm news on 21 January 2013 when a photo of a hammer "flashed up" on the screen. He said that there was no context in the news report such as an investigation into a death on Macleay Island. Nevertheless, he said that he immediately recognised the hammer in the photograph as a hammer that his wife and children had bought for him in the mid-1990s. Mrs Matheson described the hammer that she had purchased as part of a second-hand purchase of a whole toolbox for $5 or $10 which contained an old hammer in reasonably good condition. Mr Matheson said that he had lent the hammer to Mr Fennell a year or two before Mrs Watson was murdered and that he asked Mr Fennell to return the hammer the week after he had lent it to Mr Fennell but that Mr Fennell had replied "I don't know where it is, I've lost it, couldn't find it". Twelve days after seeing the hammer on the news, when Mr and Mrs Matheson gave their statements to the police, they were shown the hammer. Both gave very precise accounts of the cause of various imperfections in the second-hand hammer that Mrs Matheson had acquired. Mr Matheson described slight damage to the head of the hammer as caused by knocking "dags" off the welding on some farm machinery. He described a mark on the shaft of the hammer, imperceptible in the photographs of the hammer, as having been caused by him dropping the hammer approximately 40 feet. He described chips in the claw of the hammer as having arisen from pulling out nails. The evidence from Mr and Mrs Matheson identifying the hammer was glaringly improbable for numerous reasons. Mr Matheson's evidence of his identification was from a photo that was "flashed" briefly on a television screen without any context. It was a photo of a hammer of a type that was mass produced and generic. Also, Mr Matheson had not seen the hammer that he claimed to identify for one to two years (on his evidence) or four to five years (on Mrs Matheson's evidence). The Thompson Point hammer had been submerged in salt water for a period of time and it was common ground that this had contributed to rust that was discovered on it. An accurate identification in those circumstances would have required an astonishing visual memory. This is particularly so given the precision of Mr Matheson's recollections of the most minuscule defects in the second-hand hammer. Yet in contrast with his recollections in relation to the appearance of the hammer, Mr Matheson was unable to recall the brand of the hammer and was unable to identify his own tool belt, screwdrivers and chisels from a line-up conducted by police. Secondly, the considerable detail in the evidence of both Mr and Mrs Matheson was in fact inconsistent. Their evidence differed in that Mrs Matheson said: (i) it was her who first identified the hammer from the photograph on the news, with Mrs Matheson's evidence being that it was her who "looked up at the TV" and said "that's your hammer"; (ii) the news report had provided the context of a murder on Macleay Island; (iii) the hammer had been lent to Mr Fennell in 2008 or 2009 after they had moved to Macleay Island in 2007; (iv) the damage to the head of the hammer had arisen when she caught her foot on a screw and "belted the nail", and damage to the claw of the hammer was caused by a boyfriend of her daughter, who "was trying to pull out a screw with the hammer". Thirdly, the identification of the hammer and its particular defects by Mr and Mrs Matheson occurred in a context that was prone to cause errors in memory. They did not identify the hammer for 12 days after the photo flashed up in the news report and potentially years after they had last seen it. They were aware of the context in which the identification was taking place, namely the murder of Mrs Watson. They identified the hammer by only being shown the single hammer by the police. In particular, neither Mr Matheson nor Mrs Matheson was shown any other hammers, including any of the five hammers that were located at Mr Fennell's house. Where a court of criminal appeal is called upon to decide whether it considers that, upon the whole of the evidence, it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty of the offence charged, the court must not disregard or discount either that the jury is the body entrusted with primary responsibility of determining whether the prosecution has established the accused's guilt or that the jury has had the benefit of having seen and heard the witnesses. At the same time, however, the court may take into account the realities of human experience, including the fallibility and plasticity of memory especially as time passes, the possibility of contamination of recollection, and the influence of internal biases on memory. The court can also take into account the well-known scientific research that has revealed the difficulties and inaccuracies involved in assessing credibility and reliability. And especially is that so in a case like this where the jury has been subjected to the seductive effects of a species of identification evidence that has in the past led to miscarriages of justice. For the reasons we have given, and without impugning the honesty of Mr and Mrs Matheson in any way, their evidence was glaringly improbable. The decision of the Court of Appeal In an appeal where a ground of appeal is that the verdict was unreasonable or cannot be supported having regard to the evidence, and particularly where the Crown case is based upon a number of matters of circumstantial evidence, it is necessary for the appellate court to assess the whole of the case and to weigh that case as a whole. A circumstantial case cannot be considered in a piecemeal fashion. A corollary of this principle, particularly in cases in which the conclusion is not clear-cut, is that a substantial error, or errors, by an intermediate court of appeal in the process of assessing the case as a whole can infect the entire process. The Court of Appeal (Gotterson JA, with whom Philippides JA and Byrne SJA agreed) properly engaged in a consideration of the case as a whole when assessing Mr Fennell's submission that the jury verdict was unreasonable or cannot be supported having regard to the evidence. However, and with genuine respect for an experienced judge, the reasoning of Gotterson JA involved errors in four areas which infected his conclusion. It is likely that a different process of reasoning in relation to any one of these areas would have led to a different conclusion. Mr Fennell did not say that he gave $8,000 to Mrs Watson's daughter His Honour said that Mr Fennell claimed to have taken some of the $8,000 withdrawn on 2 November 2012 to Mrs Watson's daughter on Lamb Island and that Mrs Watson's granddaughter had denied this. It appears that this was part of his Honour's reasoning towards the conclusion that there was an "evidential basis for concluding that he had stolen at least $5,000 from the deceased and was at risk that his theft would soon be discovered". Mr Doolan gave evidence that Mr Fennell had said that Mrs Watson had given Mr Fennell a "couple of thousand dollars here and there to take over to Lamb Island to her granddaughter or her daughter". But when Mr Fennell was asked in his record of interview whether he had given any of the $8,000 withdrawal to Mrs Watson's daughter, Mr Fennell was emphatic that he had not. Mrs Fennell also said that Mr Fennell had told her that he had withdrawn $8,000 and given it to Mrs Watson. The conclusions from the accounting evidence were not open The conclusions drawn by Gotterson JA from the forensic accounting evidence given by Ms McKinnon were not open. His Honour said that the evidence revealed a lack of commensurate withdrawals from Mr Fennell's bank account in order to fund his gambling habit. His Honour also had regard to evidence from Ms McKinnon that there was a likelihood that Mr Fennell had relied upon an external source of money in order to fund an alleged deficiency of $10,833.50 for the period between 1 July 2010 and 19 November 2012. This reasoning was important to his Honour's conclusions about motive. If Mr Fennell had had sufficient financial resources, his Honour reasoned, this would have "negatived any need to steal from the deceased to make ends meet". There are several reasons why this analysis was in error. First, there is no lack of commensurate withdrawals when all of Mr Fennell's accounts during the relevant period are considered. Between August 2012 and November 2012 he withdrew $15,280 from his own personal and business accounts. Further, the alleged deficiency of $10,833.50 arose in a context where Mr Fennell's business received cash payments for which accounting may not have been entirely accurate. Over the period of almost two and a half years, $4,500 of approximately $102,000 business income banked in Mr Fennell's business account was recorded as having been received in cash. Ms McKinnon acknowledged that her analysis could not take into account all of the cash the business received. Secondly, this financial analysis could only have relevance to motive if it were confined to the period over which Mr Fennell was doing Mrs Watson's banking. During that period the alleged deficiency, according to Ms McKinnon, was only $859, which was consistent with the pattern over the remainder of the period. Thirdly, the "deficiency" which was suggested to exist amounted to only $80 per week in the context of earnings of more than $100,000 a year. Since Mr and Mrs Fennell had been living within their means during the period of the "deficiency" and getting ahead on their mortgage, any such relatively small deficiency could not be a strong basis for an inference of a motive to steal. The reasoning concerning the Thompson Point hammer After his Honour's discussion of the significance of the Thompson Point hammer, Gotterson JA said that Mr Matheson's account that he lent a hammer to Mr Fennell, who did not return it, was "quite credible". That reasoning can be accepted. Indeed, that account was never challenged by counsel for Mr Fennell. However, Gotterson JA also reasoned that Mr Matheson's evidence of identification of the hammer as the Thompson Point hammer was "detailed and consistent" and "had an appealing practicality to it", and "[t]he jury could well have regarded it as convincing proof linking [Mr Fennell] to that hammer". For the reasons explained above, that conclusion cannot be accepted. The opportunity reasoning Finally, the reasoning of Gotterson JA concerning opportunity involved errors. His Honour said, in apparent reliance upon the evidence of Mr Robinson, that it was open to the jury to accept that Mrs Watson was killed at around the time when Mr Fennell's vehicle "was sighted at about 11.00 am". His Honour correctly observed that this evidence "had its imperfections". However, those imperfections were so substantial that the evidence should have been rejected by any reasonable jury. It does not appear that the Crown, which was represented by different counsel in the Court of Appeal, made the same concession about Mr Robinson's evidence as was made in this Court or that the Court of Appeal was directed to the fundamental problems with that evidence. Conclusion It was not open to the jury to be satisfied of Mr Fennell's guilt beyond reasonable doubt. The Court of Appeal should have allowed the appeal, quashed the conviction, and entered a verdict of acquittal. For the reasons above, we made orders to that effect.
Fennell
The Queen
Appeal allowed. Order of the Court of Appeal set aside and in its place the appeal allowed, conviction quashed and verdict of acquittal entered.