Court of Criminal Appeal Supreme Court New South Wales

• Medium Neutral Citation: Gilham v R [2012] NSWCCA 131
• Hearing dates: 28, 30 November 2011, 1 December 2011
• Decision date: 25 June 2012
• Before: The Court at [1]; McClellan CJ at CL at [647]; Fullerton J at [662]; Garling J at [677]

Decision

1. Appeal upheld.

2. Convictions quashed.

3. Verdicts of acquittal entered on both counts

• Catchwords: CRIMINAL LAW - appeal - conviction - double jeopardy - incontrovertibility of verdict - whether conviction of applicant for murder of his parents controverted applicant's acquittal for murder of his brother - whether manner in which Crown Prosecutor conducted trial controverted earlier acquittal - whether trial judge failed to instruct jury to give applicant "full benefit" of earlier acquittal - whether trial judge erred by not staying proceedings - rule against double jeopardy not infringed - trial judge's directions denied applicant full benefit of earlier acquittal.; CRIMINAL LAW - appeal - conviction - expert evidence - admissibility - relevance - prejudicial effect - whether evidence of fire demonstrations and likely behaviour of fire elicited from Crown expert was relevant - whether probative value of evidence outweighed by prejudicial effect - probative value outweighed by prejudicial effect - evidence ought not to have been admitted - whether expert opinion evidence on similarity of pattern of stab wounds admissible - evidence of similarity admissible - opinion that similarity constituted an underlying pattern inadmissible - Crown Prosecutor's reliance on pattern of similarity amounted to coincidence reasoning - evidence of pattern of similarity not admitted as coincidence evidence under s 98 of Evidence Act 1995 - whether failure to call additional expert on issue of similarity on grounds of unreliability occasioned a miscarriage of justice - failure to call witness caused trial to miscarry - whether expert evidence on expected amount of blood on applicant and murder weapon relevant and admissible as expert opinion evidence - whether probative value of evidence outweighed by prejudicial effect - evidence admissible as expert opinion evidence - probative value outweighed prejudicial effect.; CRIMINAL LAW - appeal - conviction - whether Crown Prosecutor cross-examined applicant in improper manner - whether Crown Prosecutor addressed jury in an improper manner - whether Crown Prosecutor undermined directions of trial judge - aspects of Crown Prosecutor's conduct and address improper -no miscarriage of justice occassioned.; CRIMINAL LAW - appeal - conviction - whether verdict unreasonable or unsupported by the evidence - circumstantial evidence - circumstantial evidence to be considered as a whole - doubt capable of being resolved by jury's advantage in seeing and hearing evidence of applicant - no reasonable doubt on independent assessment of evidence.; CRIMINAL LAW - new and fresh evidence - whether evidence available at time of trial - whether evidence credible, plausible or capable of belief - whether evidence likely to have caused jury to have entertained a reasonable doubt about guilt of applicant - new evidence concerning carbon monoxide likely to have caused jury to entertain a reasonable doubt.; CRIMINAL LAW - appeal - conviction - whether applicant should be acquitted or retried - discretionary considerations.
• Legislation Cited: Criminal Appeal Act 1912; Criminal Appeal Rules; Director of Public Prosecutions Act 1986; Evidence Act 1995; Legal Profession Act 2004
• Cases Cited: Barca v The Queen [1975] HCA 42; (1975) 133 CLR 82; Burrell v R [2009] NSWCCA 193; Causevic v R [2008] NSWCCA 238; 190 A Crim R 416; Dasreef Pty Ltd v Hawchar [2011] HCA 21; (2011) 243 CLR 588; Director of Public Prosecutions v Shannon [1975] AC 717; Direction of Public Prosecutions (Nauru) v Fowler (1984) HCA 48; (1984) 153 CLR 317; Dyers v The Queen [2002] HCA 45; (2002) 210 CLR 285; Evans v The Queen [2007] HCA 59; (2007) 235 CLR 521; Everett v The Queen [1994] HCA 49; (1994) 181 CLR 295; Fleming v The Queen [1998] HCA 68; (1998) 197 CLR 250; Garrett v The Queen [1977] HCA 67; (1977) 139 CLR 437; Gerakiteys v The Queen [1984] HCA 8; (1984) 153 CLR 317; Gilham v R [2007] NSWCCA 323; (2007) 73 NSWLR 308; Gill v Walton (1991) 25 NSWLR 190; Haoui v R [2008] NSWCCA 209; (2008) 188 A Crim R 331; HG v The Queen [1999] HCA 2; (1999) 197 CLR 414; Jiminez v The Queen [1992] HCA 14; (1992) 173 CLR 572; Johnson v Miller [1937] HCA 77; (1937) 59 CLR 467; Justins v R [2010] NSWCCA 242; (2010) 204 A Crim R 315; Kanaan v R [2006] NSWCCA 109; King v The Queen [1986] HCA 59; (1986) 161 CLR 423; Livermore v R [2006] NSWCCA 334; (2006) 67 NSWLR 659; Longman v The Queen [1989] HCA 60; (1989) 168 CLR 79; M v The Queen [1994] HCA 63; (1994) 181 CLR 487; Makita (Australia) Pty Ltd v Sprowles [2001] NSWCA 305; (2001) 52 NSWLR 705; McCullough v R [1992] Tas R 43; MFA v The Queen [2002] HCA 53; (2002) 213 CLR 606; Paric v John Holland (Constructions) Pty Ltd [1985] HCA 58; (1985) 59 ALJR 844; Parker v The Queen [1997] HCA 15; (1997) 186 CLR 494; Peacock v The King [1911] HCA 66; (1911) 13 CLR 619; Plomp v The Queen [1963] HCA 44; (1963) 110 CLR 234; R v Abou-Chabake [2004] NSWCCA 356; (2004) 149 A Crim R 417; R v Anderson (1991) 53 A Crim R 421; R v Apostilides [1984] HCA 38; (1984) 154 CLR 563; R v Arp [1998] 3 SCR 339; R v Bikic [2002] NSWCCA 227; R v Carroll [2002] HCA 55; (2002) 213 CLR 635; R v Darby [1982] HCA 32; (1982) 148 CLR 668; R v Degnan [2001] 1 NZLR 280; R v Gibson [2002] NSWCCA 401; R v Gilham [2007] NSWSC 231; (2007) 190 A Crim R 303; R v Gilham [2009] NSWSC 138; R v Glennon [1992] HCA 16; (1992) 173 CLR 592; R v Grdic [1985] 1 SCR 810; R v Hilder (1997) 97 A Crim R 70; R v Hillier [2007] HCA 13; (2007) 228 CLR 618; R v Honeysett (1987) 10 NSWLR 638; (1987) 34 A Crim R 277; R v Kneebone [1999] NSWCCA 279; 47 NSWLR 450; R v Micallef [2002] NSWCCA 480; (2002) 136 A Crim R 127; R v Murre [2001] NSWCCA 286; R v O'Donohue [2001] NSWCCA 458; R v Parkes [2003] NSWCCA 12; 147 A Crim R 450; R v Pernich (1991) 55 A Crim R 464; R v Perry (1990) 49 A Crim R 243; R v Ronen [2004] NSWSC 1298; R v Storey [1978] HCA 39; (1978) 140 CLR 364; R v Taufahema [2007] HCA 11; (2007) 228 CLR 232; R v Thomas (No 3) [2006] VSCA 300; (2006) 14 VR 512; R v Wilkes [1948] HCA 22; (1948) 77 CLR 511; R v Wilton (1981) 28 SASR 362; (1981) 4 A Crim R 5; R v Z [2001] 2 AC 483; Ramsay v Watson [1961] HCA 65; 108 CLR 642; Randall v The Queen [2002] 1 WLR 2237; Ratten v The Queen [1974] HCA 35; (1974) 131 CLR 510; Reid v The Queen [1980] 1 AC 343; SKA v The Queen [2011] HCA 13; (2011) 243 CLR 400; Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650; Spies v The Queen [2000] HCA 43; (2001) 201 CLR 603; Velevski v R [2002] HCA 4; 76 ALJR 402; Vella v R (1990) 2 WAR 537; Ward v R (1981) 3 A Crim R 171; Washer v Western Australia [2007] HCA 48; (2007) 234 CLR 492; Whitehorn v The Queen [1983] HCA 42; (1983) 152 CLR 657; Wood v R [2012] NSWCCA 21
• Category: Principal judgment
• Parties: Jeffrey Gilham (Applicant); Crown
• Representation: Counsel: C Steirn SC/P Barham (Applicant); L Babb SC/N Noman (Crown); Solicitors: Paladin Law Pty Ltd (Applicant); Solicitor for Public Prosecutions (Crown)
• File Number(s): 2006/2766

Decision under appeal

• Jurisdiction: 9111
• Citation: R v Gilham [2009] NSWSC 138
• Date of Decision: 2009-03-11 00:00:00
• Before: Howie J
• File Number(s): 2006/2766

JUDGMENT

1 THE COURT: The applicant was convicted after trial of the murder of his parents. He appeals against his convictions and seeks leave to appeal against his sentence.

2 The Court is unanimous in its conclusions with respect to each of the grounds of appeal against conviction. However, we have reached different conclusions with respect to the appropriate orders. This judgment sets out our views in relation to the grounds of appeal. We have prepared and publish separate judgments in relation to the appropriate orders.

3 At or around 4 am on 28 August 1993, Mr Stephen Gilham and Mrs Helen Gilham were murdered in the home they occupied at Woronora with their two sons, Christopher, aged 25, and the applicant, Jeffrey, aged 23.

4 The family was highly respected in the community. Both Christopher and the applicant were well regarded by both friends and neighbours for their scholastic achievements, their generally exemplary behaviour and character, and their loving relationship with their parents. The applicant was gregarious and popular, with marked success as a yachtsman. He was apparently his father's favoured son. Christopher was more quiet and sensitive although not without a supportive circle of friends. He was a keen and successful fencer and a talented pianist.

5 The family home was built on a split level with the upper and lower floors connected by an internal spiral staircase. External stairs provided access through a garden to the Georges River.

6 A double carport and single garage were located at street level, which housed a yacht as well as the motor vehicle the applicant customarily drove and his father's motor vehicle. The upper level of the house was accessed via stairs from the street, which connected with a side deck (or wooden walkway) that extended the length of the house along the upper level to glass sliding doors through to a living area. A dining table was immediately to the left on entry, and in relatively close proximity, but further into the room, there was a piano. Mr and Mrs Gilham's bedroom was also on the upper level. A short hallway connected their bedroom to the living space. The upper level also housed the kitchen together with a second bedroom, formerly occupied by the applicant, and a computer room. Christopher's bedroom was on the lower level. The lower level also housed a billiard table and small kitchenette. A rear door opened from the laundry giving access to the garden.

7 The upper level was connected by external wooden stairs and a set of garden stairs to a boatshed, which the applicant used as his bedroom. An intercom system operated between the boatshed and the house.

8 Mr and Mrs Gilham both died from multiple stab wounds to the chest and back, after which a fire was lit on or near their bodies. It was accepted that the knife used to stab Mr and Mrs Gilham came from the kitchen in the main house. Mr Gilham sustained a total of 28 knife wounds of which 16 were to the front of the chest. Mrs Gilham sustained a total of 17 knife wounds of which 13 were to the front of the chest. Their badly burnt bodies were located on the upper floor of the house at approximately 4.54am, by fire officers who attended at 4.43am in response to a 000 call placed by a next door a neighbour at 4.34am, after the applicant alerted him both to the fire and the fact that his parents and his brother were dead. He said his brother killed his parents and set them alight and that he killed his brother.

9 Mr Gilham was found face down in the bedroom. Mrs Gilham was found lying on her back in the hallway outside the bedroom, close to a bookcase on the top shelf of which was located the intercom and a telephone and answering machine. Evidence at post-mortem confirmed that all wounds had the appearance of being haemorrhagic, which was suggestive of their infliction while the deceased were alive. The order of the infliction of the wounds could not be determined. Although the time of death could not be fixed with any precision, the evidence was that they both died within minutes of being attacked.

10 A short time after locating the bodies of Mr and Mrs Gilham, fire officers found Christopher's body on the lower level. He had sustained 17 knife wounds to the upper body, of which 14 were to the front of the chest. His wounds were also haemorrhagic. He was naked save for a towelling shave coat. His body was unaffected by the fire which had been set on the upper level. A knife was located beside his body. It was common ground that the knife had been used to inflict the wounds on each of the deceased.

11 On 29 August 1993 the applicant was charged with the murder of his brother. On 5 April 1995 he pleaded guilty to his manslaughter. The Crown accepted that plea upon the basis that it could not, as it was required to do, disprove that the applicant had been provoked into killing his brother upon discovering that Christopher had killed his parents. He was sentenced to 3 years imprisonment, which was deferred upon him entering into a bond to be of good behaviour for 5 years. That sentence was premised on the basis that Christopher had murdered the applicant's parents and that this conduct had provoked him to kill his brother.

12 In June 1995 the Coroner found that the deaths of Mr and Mrs Gilham were caused by stab wounds inflicted by Christopher.

13 Following a renewed police investigation in 1999 a second inquest was convened. That inquest was terminated in April 2000, the Deputy State Coroner having found that there was evidence capable of satisfying a jury that a known person (the applicant) had committed an indictable offence (the murder of his parents). In July 2000 the Director of Public Prosecutions notified the applicant of his decision not to prosecute, having concluded, on the available evidence, that there was no reasonable prospect of a conviction.

14 In May 2001 a private prosecution for murder was launched by the applicant's uncle. The following month those proceedings were taken over by the Director of Public Prosecutions pursuant to section 9 of the Director of Public Prosecutions Act 1986. The Director, as prosecutor, offered no evidence upon the information for murder and the applicant was discharged.

15 In October 2004 a third police investigation into the deaths of Mr and Mrs Gilham was commenced. In November 2005 a further brief of evidence was sent to the Director of Public Prosecutions seeking his advice as to whether there was sufficient evidence to charge the applicant with their murder.

16 On 21 February 2006 an ex officio indictment was filed in the Supreme Court charging the applicant with the murder of his parents. On 28 November 2008 a jury found the applicant guilty of murder and on 11 March 2009 he was sentenced to life imprisonment on each count.

17 The applicant appeals to this Court against his convictions and seeks leave to appeal against sentence.

18 The appeal against conviction relies upon 18 grounds, some of which allege error in the conduct of the trial and others which relate to evidence not led at trial and which the applicant submits has been productive of a miscarriage of justice. A further ground contends that the verdict is unreasonable and cannot be supported by the evidence. That ground requires an examination of the entirety of the evidence led at trial.

The Crown case at trial

19 Central to the Crown case were a number of related facts which the Crown Prosecutor submitted were proved by the evidence and which fixed a timeline within which the deceased were attacked and killed and the fire lit:

(a) Mr and Mrs Gilham, and Christopher, were attacked and died within five to ten minutes of 3.57am, that is, between 4.02 - 4.07am;

(b) the applicant knocked on his neighbour's door at approximately 4.30am to report the fire and the killings;

(c) the 000 call was placed at 4.34am;

(d) fire officers arrived at the scene at 4.43am;

(e) the upstairs bedroom window blew under heat and pressure at about 4.45am;

(f) the fire was under control between two to four minutes after water was applied through the blown window, that is, between 4.47 and 4.49am; and

(g) the fire had been ignited 20-30 minutes before being brought under control. Accordingly, the fire had been lit at the earliest at 4.17am and at the latest at 4.29am, the most likely time being 4.22 - 4.23am.

20 The prosecutor submitted that the timing of this sequence of events proved beyond reasonable doubt that the applicant murdered his parents and that his account of seeing them dead and alight within a minute or so of his mother's call for help is a lie.

The evidence of the neighbours

21 The Crown alleged that within a period of five to ten minutes of 3.57am the applicant murdered his parents and then killed his brother. This time frame was fixed by the evidence of Ms Christine Shaw hearing what she described as a strange combination of two voices, in a "screaming-type conversation shouting at one another". Ms Shaw's house was built on a gradient on the opposite side of the road overlooking the Gilham house. She had woken at 3.40am and was in bed reading by 3.45am when she was alerted to the sound of the voices, one which she described as a man's voice, that reminded her of someone being struck by lightning, and one which she described as a very stressed high-pitched woman's voice speaking in a restraining tone, as if desperately trying to reason with the man. She was unable to make out any of the words. After a while she went to the window but was unable to see anything. The voices stopped while she was at the window. She estimated that the shouting lasted for about five minutes, possibly longer, before it stopped. She returned to bed and continued reading. She said a dog started barking when the conversation started and that it kept barking after the voices stopped.

22 Mr and Mrs Baker's house faced the Gilham house. They were stirred from sleep sometime between 4am and 4.15am by their dog barking, which continued on and off for about half an hour despite efforts to quieten him. They did not hear any raised voices. At about 4.30 Mr Baker got out of bed and went outside where he saw smoke rising from under the roof tiles of the Gilham house and a flickering in the side window. He told his wife to call the fire brigade.

23 Ms Shaw only became aware of the fire after the arrival of the emergency services. The fire brigade arrived at 4.43am. She was unable to fix with any certainty the interval between when the voices stopped and when she was alerted to the arrival of the fire brigade by lights and men talking. She variously described it as being between 10 minutes and 20 to 25 minutes.

24 It was the Crown case that the female voice Ms Shaw heard was Mrs Gilham defending herself from the applicant's violence after he had attacked his father in the bedroom with such brutality that he had no time to defend himself. It was the Crown case that the male voice was Mr Gilham in the throes of death. The Crown submitted that the applicant was a calculated killer who said nothing at all as he stabbed first his father and then his mother. It was also the Crown case that there was an interval of at least 20 minutes after the applicant killed his parents (and then his brother) before he raised the alarm, during which time he took steps to destroy any evidence that would or might point to him as the murderer. These steps included sourcing accelerant and setting fire to the house, washing himself to remove any evidence of his parents' blood transferred to him during the knife attack, and washing the knife to remove his fingerprints. The Crown case was that he also washed off the blood which would have transferred to him when he stabbed Christopher to death.

25 What the Crown relied upon as the 20-minute window of opportunity was fixed by reference to Ms Shaw's evidence and the evidence of Mr and Mrs Warner, who lived next door to the Gilham house.

26 Mr Warner gave evidence that he was woken up by two people arguing, which to his best recollection were two male voices. He conceded in cross-examination it may have been a man's voice and a woman's voice. He described them as being very angry, very heated and very loud. He said the conversation lasted 20, 30 or 40 seconds and that he went back to sleep. Although he looked at his alarm clock, he was not able to recall the time with any precision. He said that it was "three something". He agreed in cross-examination that he told investigating police that it was about 3.40-3.50am but accepted it could have been 3.57am.

27 He went back to sleep but was then woken by the applicant knocking on his front door. He gave evidence that when he opened the front door the applicant, who was dressed only in boxer shorts, seemed very agitated and mumbled something about ringing 000 and the house being on fire.

28 Access to the Warner property from the Gilham property was either from the street via a side gate or via the back lawn. There was no fence separating the rear of the properties and although there were pool gates they were always kept open. Mr Warner gave evidence that there was a hose at the rear of his house and he believed there was a tap at the boathouse on the Gilham property. He said that the side gate had a noisy latch which had woken him in the past and that he was confident that he would have woken if the applicant had entered via the gate from the street. It was the applicant's case that he left the burning house and went to the street and then immediately next door to the Warners'. It was the Crown case that the applicant had washed himself before entering the Warners' property from the rear.

29 Mr and Mrs Warner described the applicant as being very cold to the touch but neither wet nor sweaty. They said it had been raining and there was mist in the air. Neither of them gave evidence of observing any blood on him at that time although, while awaiting the arrival of emergency services and in the process comforting him after giving him both a sweater and a blanket to keep warm, they both gave evidence of seeing blood (or what looked like blood) on his foot and on the lower part of his left leg, which gave the appearance of being either washed or wiped or both. Mr Warner recalled seeing blood in the quick of the applicant's fingernail on either one or both hands. Mrs Warner did not remember observing anything on his hands or fingernails. Neither of them saw any residue of blood on their light-coloured leather settee or on the kitchen floor where the applicant had been sitting from time to time, or any smears of blood on the creamy white-coloured telephone which the applicant handled when speaking to the 000 operator.

30 Neither Mr nor Mrs Warner described the applicant as out of breath when he first arrived. Mr Warner described him as speaking in a disjointed fashion, saying things such as "They're all dead. He's killed them, set up them on fire" and "Mum and Dad are both dead. He's burnt them. I've killed him for what he did to them". Mrs Warner gave similar evidence. She said that the applicant was agitated but not crying.

31 Although the time that the applicant arrived at the front door of the Warners' home could not be fixed with absolute precision, Mr Warner gave evidence that no more than two to three minutes (five minutes at the most) passed between the applicant knocking on the door and the call being placed to 000. That call was logged at 4.34am. In cross-examination Mr Warner said that the applicant attempted to dial the number but was shaking and distressed and pacing about and could not manage to put his finger in the dial, so he placed the call for him. Mr Warner gave evidence that the 000 operator asked him if the applicant had been drinking, so he lent over and smelled the applicant's breath. He said that he smelled what he thought was the smell of kerosene and a smoky smell which could also have been on the applicant's hair. He said this was consistent with what he had smelled when he opened the front door.

32 Mr Warner said the applicant came to the phone when the operator asked to speak to him. The applicant was asked by the operator to explain what happened. He said his brother killed his parents. He said he walked through the door; that his mother had called him; that he saw his brother with a knife; that his parents were on fire; and that he killed his brother after chasing him down the stairs. He said he did not know why he chased his brother. The Crown case was that the applicant was feigning breathlessness when speaking with the operator to give the appearance of truth to the lie that after stabbing Christopher he had immediately run from the burning house to raise the alarm.

33 It was also the Crown case that other aspects of the applicant's conduct that morning were a deliberate attempt by him to deflect attention from himself or the fire, and to feign concern for his parents. In the interval between the end of the 000 call and the arrival of emergency services Mrs Warner gave the applicant a burgundy sweater which he cuddled into with his hands without putting on. She said he got onto the floor in the kitchen in a foetal position. Mrs Warner also gave him a blanket to put around his shoulders at some stage while he was on the kitchen floor. It was the applicant's case that any residue of Christopher's blood would likely have been wiped off onto the sweater or the blanket. Neither was seized by police. Although according to the Warners the applicant did not say very much from his foetal position, at one point he grabbed a fire extinguisher and said he was going to fight the fire. Mrs Warner took the fire extinguisher from him. It was the Crown case that this was a feigned act of concern.

The arrival of the fire officers

34 Fire officers from Sutherland fire station arrived at 4.43am. They were the first of the emergency services to arrive at the scene. They observed smoke issuing out of the eaves of the Gilham house. At 4.45am police and backup services were summoned.

35 Fire Officers Langdon and Polson, under the command of Fire Officer Lowder, entered the property from street level. The applicant was at the base of the stairs leading from the street. He was described as extremely agitated, emotionally distraught and largely incoherent. He was saying that his parents were inside and that they were dead. Polson's evidence was that the applicant appeared to be delirious and disoriented and that he was crying. From his momentary dealings with the applicant Polson said his hair looked slightly wet and although there had been rain that morning it was not raining when they arrived at the house. It was an agreed fact that between 9pm on 27 August and 9am on 28 August 1993 2mm of rain was recorded at 3am at Sydney airport (15 km from Woronora).

36 Langdon was instructed to take the applicant to the street. Lowder and Polson travelled along the side balcony to the rear of the house. Langdon wrapped a blanket around the applicant's shoulders and left him in the care of Mr Warner. He made no observation of smelling accelerant on the applicant.

37 After the arrival of the fire services Mr Baker saw the applicant with Mr Warner. He described him waving his arms around and mumbling in gibberish. Mrs Baker said he was jumping from side to side and running from side to side along the house. She said he had a blanket wrapped around him and was shaking uncontrollably and that he was being held by her husband and then by Mr Warner.

38 Officer Wright was the designated pump operator who was at the rear of the fire truck when Langdon brought the applicant from the house. He said that when the applicant was about a metre and a half away from him he smelled petrol on him. His evidence was that he was familiar with petrol and was able to distinguish it from kerosene and methylated spirits and to distinguish petrol from a smoky smell infused with another accelerant. After the fire was knocked down and he entered the house to extinguish any spot fires, he did not smell any accelerant. He described the applicant as highly agitated, unable to sit still, and standing up and sitting down, but not crying. He was in the applicant's company for two to four minutes.

39 Fire Officer Pelham, who arrived from the Menai fire station, gave evidence that on his arrival smoke and fire were coming from the roof of the house. He said he saw the applicant standing near the garage alone, dressed only in shorts, and that he was highly agitated. The applicant said to Pelham: "Don't go in there, somebody's got a knife". It was the Crown case that the applicant was attempting to deflect the Officer's attempts to assist with fighting the fire in the hope that the fire would obliterate all evidence of his having murdered his parents.

40 Senior Constable Sands was the first police officer to arrive at the scene at 4.50am. He saw the applicant sitting on the footpath with a blanket wrapped around his shoulders, apparently in Mr Warner's company. He did not speak with him.

41 As Lowder and Polson reached the sliding glass doors that opened from the upper floor to the external balcony, the rear bedroom window blew. This was about a minute and a half after they first arrived, that is, at or about 4.45am. Lowder opened the closed glass sliding doors but was unable to enter due to the excessive heat. The only light in the premises came from the fire. It was the Crown case that the applicant had closed the sliding doors as a method of accelerating or hiding the fire, which was inconsistent with his account to police that he left the house in a hurry to seek the assistance of his neighbours, leaving the door open.

42 The fire was attacked through the main bedroom window from which flames were shooting. The bedroom was blanketed with water. The fire was reduced to a moderate level within minutes, permitting fire officers to then crawl into the house with another hose to extinguish the fire seated in the vicinity of Mrs Gilham's body on the floor on the right-hand side or southern corner of the room. Officer Pelham relieved Lowder and Polson, who had been forced to retreat from the excessive heat. They were both singed from the flashover effect of cold water contacting with open flame.

43 Pelham gave evidence that when he entered the house it was very smoky and hot. He located Mrs Gilham's body first and then as he continued to search for and extinguish the fire he found Mr Gilham's body in the bedroom. Fire officers with breathing apparatus assisted. Polson found an axe in the bedroom parallel to the bed. By this time the smoke had cleared. He also found Christopher's body downstairs. It was the Crown case that the applicant had taken the axe with him to his parents' bedroom in case his plan to kill his parents by stabbing them was frustrated for some reason, or in the event that Christopher came up the stairs roused by their struggle.

44 Under instruction from Lowder, Langdon went downstairs with Polson and Senior Constable Sands to check the lower level of the house, which they accessed by kicking in the rear door at the lower level. Langdon gave evidence that the area inside was dark save for the reflected light of a spot fire which was still burning upstairs.

45 Christopher's body was lying adjacent to a pool table with a knife next to one of his arms and near his chest. The pool table was in a living area with an open kitchenette. A bathroom and laundry were also located there together with Christopher's bedroom. The bed showed evidence of having been slept in. Lowder directed that the area be set up as a crime scene in accordance with the established protocol that nothing was to be moved by fire fighters.

46 Langdon also gave evidence that there was smoke at the ceiling of the downstairs living area. However, because of the largely uncontroversial evidence at trial that the levels of carbon monoxide in the bodies of the deceased indicated that they were all dead when the fire was lit, his evidence concerning the extent to which smoke from the fire upstairs penetrated the downstairs area, either before the fire was effectively extinguished or in the course of fire being fought, was not the subject of any testing. The crime scene officer Detective Sergeant Horn gave evidence that when he arrived at the house at about 7.12am there was no sign of fire or heat/smoke damage in the downstairs area, although there was fire debris on the staircase and on the tiled floor.

47 The evidence at trial was that the concentration of carbon monoxide in the bodies of each of the deceased revealed on autopsy was within normal limits. This evidence was then relied upon by the Crown to establish that each of the three deceased were dead when the applicant lit the fire on the upper level on or near the bodies of his parents, with the intention of engulfing in flames the house along with the bodies of the deceased, thereby destroying the crime scene and any evidence that would, or might, have implicated him as the murderer.

48 Although the concentration of carbon monoxide in Christopher's blood at six per cent was significantly greater than that of his mother and father at four and three per cent respectively, it formed no part of the Crown case at trial that the greater concentration in Christopher's blood evidenced his proximity to the fire before he died. It was the Crown case that after stabbing his parents the applicant stabbed Christopher as he was coming up the stairs (probably in response to the same voices Ms Shaw heard or the Barkers' dog or both) and that the applicant then stabbed him repeatedly in the chest on the lower level as Christopher retreated back down the stairs, trying to escape from the attack, such that Christopher was at no time on the upper level. This was in direct conflict with the applicant's account to police, first given within an hour of the events, and his evidence at trial, that Christopher had killed his parents on the upper level and was in the process of setting fire to his mother (on one view of his evidence with his father already alight) as the applicant entered the house from the boatshed in response to his mother's cry for help over the intercom. On the applicant's account, Christopher was in direct proximity to the fire (and it must be assumed inhaling smoke) before the applicant chased him downstairs where he was killed.

The crime scene and forensics testing

49 A number of police officers, including Detective Sergeant Horn, gave evidence of their observations of Christopher's body in situ and the location of the knife. There was some controversy at the trial about the location of the knife, whether it had been moved, and the extent to which the water which had been used to extinguish the fire upstairs had contaminated the downstairs area, including the body of the deceased and his shaving coat, both before and after a tarpaulin was used to cover his body. This evidence will be referred to when considering whether the applicant has made good his challenge to the admissibility of expert evidence as to the likelihood of there being blood on the knife and whether water may have removed evidence of blood from the knife and of his parents' blood from Christopher's body.

50 There were no swabs taken from Christopher's body or from the knife. Of the 17 areas on Christopher's shave coat that were subjected to testing in November 1993 (each of which was chosen because it was an area not obviously related to the knife injuries Christopher sustained), only one had blood which could only have originated from Christopher. In the remaining 16 areas there was a HP type common to both Christopher and Mrs Gilham but which could not be further confirmed by blood type or further testing. There were no properties in the blood consistent with it being the blood of Mr Gilham. At the time of trial neither the shave coat nor the knife were available for DNA testing.

51 Swabs were taken from the tiled lower level near the steps that led to the bathroom/laundry; from near the spiral staircase; from the railing of the spiral staircase; from the cupboard door below the sink to the right of where Christopher's body was found; and from an upright ironing board above Christopher's body. Human blood, which forensic testing confirmed could only have originated from Christopher, was detected on each of these swabs.

Blood stain pattern analysis of the blood on the lower level around Christopher and on Christopher's body

52 A plan marked with generalised areas of bloodstaining, with particular focus on eight areas around Christopher's body, was prepared for the purposes of trial, using crime scene photos and videos and photogrammetry plans. This material was used for the purposes of eliciting evidence from three expert witnesses in bloodstain pattern analysis: Sergeant Reynolds, Mr Raymond and Dr Culliford.

53 The blood droplets on the railing of the staircase included transferred blood and projected blood, including castoff blood that occurs when an object with blood on it is in motion. Sgt Reynolds said the most likely cause of the droplets on the railing was the placing of a bloodstained hand onto the balustrade, causing the blood droplets to be flung off, which would also transfer blood in the same process. He gave evidence that the blood spots on the front of the cupboards and the top of the ironing board were most likely castoff stains caused by a bloodied hand moving through the air, or from the weapon which was being used to stab someone on the ground causing blood to move through the air.

54 The Crown experts also gave evidence concerning the likely source of blood spatters on Christopher's chin and face, and more controversially, blood staining on his right hand, inner wrist and forearm. These areas of staining were revealed in crime scene photographs, which it was agreed were generally of poor quality. The reliability of any analysis was further undermined by the contaminating effects of water. The low spatter numbers and shapes of the stains operated as other impediments to any reliable analysis.

55 In relation to the spatter patterns on Christopher's neck and chin, it was generally agreed that the blood could have come from his wounds, especially those high on his body, or from Christopher hitting a bloodied surface. It was agreed that it was less likely to be the blood of his parents (or either of them) transferred in the process of Christopher attacking them, although that could not be discounted entirely.

56 In the view of two of the Crown experts, the directional spatter stains on Christopher's inner wrist and forearm gave rise to a reasonable possibility that this staining occurred in the process of Christopher stabbing his parents, although it was impossible to be certain of the origin of any of the blood without it being typed. The other possible source of this blood pattern was the defence injuries to the tips of Christopher's fingers, one of which went through the webbing between his thumb and forefinger, effectively running up his arm. Blood from under Christopher's fingernails on his right hand was collected and taken to the division of analytical laboratories, but was never analysed.

Blood stain pattern analysis of the blood on the applicant

57 Swabs taken from the applicant's left shin and right hand of blood or what appeared to be blood were not suitable for identification purposes when tested in 1993. A swab taken from the applicant's right foot was submitted for DNA testing in 1998; however, even at that time, DNA testing was still in its infancy. With the proviso that the results were below the laboratory reportable level, the Crown's expert witness gave evidence that there were strong indications that the DNA could be Mrs Gilham's. He gave evidence that it was definitely not the blood of Christopher or Mr Gilham.

58 Sgt Reynolds analysed the photographs of blood on the applicant's right foot which, on the Crown case, originated from Mrs Gilham. He said that the bloodstain would not have been the result of walking or stepping in blood. He was of the view that there appeared to be four individual areas of bloodstaining within the composite bloodstain, with voided areas between them. He said this pattern was either the result of contact between the applicant's foot and an irregular surface with blood on it, or from a blood drop of sufficient volume to travel under the force of gravity, landing above the flow area and running down the side of the foot. Mr Raymond gave evidence that the bloodstaining was not blood spatter but that it was consistent with contact between the applicant's foot and a bloody surface at the level of his foot, be it bloodstained clothing or some part of a body with wet blood. He said the more likely scenario was that blood had dropped from above and then looped onto the applicant's foot.

Other items seized and/or submitted for forensic testing

59 Spectacles were found in the downstairs bathroom. It was accepted at trial that they belonged to Christopher to correct his myopia and mild astigmatism. The Crown relied upon this evidence to support the submission that it was unlikely that Christopher went upstairs in the dark without his spectacles, and dressed only in his shave coat, to kill his parents, as distinct from him being roused from sleep and emerging from his bedroom in a panic upon hearing his parents being attacked upstairs. Although a former flatmate and a former girlfriend attested to Christopher having his spectacles at his bedside because he needed them to move from the bedroom to the bathroom, his ophthalmic surgeon gave evidence that in 1992 Christopher would have been able to negotiate a familiar environment quite well and perform everyday tasks without his spectacles (including going upstairs, pulling out a cutlery drawer, identifying one knife from another and identifying his mother as opposed to his father) but that he would have benefited from wearing his spectacles to view anything farther away than an arm's length. He gave evidence that Christopher's corrected and uncorrected visual acuity was below average but not far below average.

60 An empty packet of 24 Panadol tablets, and two empty 12-tablet packets of Panadol, were found in a clothes basket inside the laundry on the lower level, together with an empty 20ml syringe packet, a 25ml syringe with a white substance in it, a needle packet, a smoked glass cup with a white substance in it, and a silver table knife with a white substance on it. Subsequent forensic testing revealed the presence of paracetamol in the syringe and in the cup. Crime scene officers also photographed the contents of a plastic bag on the bottom shelf of a bookshelf in Christopher's bedroom, which contained two syringes and other items.

61 The items from the clothes basket were tested for fingerprints in September 1993. At trial, the fingerprint officer who performed the testing gave evidence by reference to a fingerprint examination record (only some of which was written by him). He said the items were in a clean state, without smudges, partial fingerprints or portions of fingerprint ridges. He gave evidence that if these items had been handled in a normal way he would have expected to see latent indication of that handling and that the absence of that evidence suggested to him that either the prints (or partial prints) had been removed or the person handling the item was at pains to ensure that no latent prints were deposited. He agreed that there were no glove marks or wipe marks. He also agreed that where he recorded "nil identifiable ridge pattern located", this did not mean an item was absolutely clean. It was also consistent with someone having handled an item without leaving sufficient fingerprint marks in residue to be later identified. He also agreed that in a statement made months after the examination, in which he wrote "I examined the exhibits and found nothing of evidentiary value in respect of fingerprints", he was intending to report that there was nothing to enable him to positively identify or eliminate a person who might have handled the items.

62 It was the Crown case that the applicant had planted the Panadol and syringe and other items in the laundry basket near his brother's bedroom to suggest that they belonged to him, and to reinforce what the applicant had predetermined would be the case that he would mount against his brother. This case was that his brother had murdered their parents because he was in a disturbed mental state. Dr Lawrence gave evidence that an injection of Panadol paste would result in illness, and ultimately liver failure if left unchecked for a couple of days.

63 Police located a 25-litre jerry can containing a quantity of leaded/super petrol and a length of garden hose with freshly cut ends in the small brick alcove area at the front of the house. The car used by the applicant operated on leaded petrol whereas the car used by Mr Gilham operated on unleaded petrol. A longer piece of garden hose was found in the garage. That hose was connected at one end to a tap. The other end also appeared to be freshly cut. Another section of garden hose was partly concealed beneath the small yacht and trailer. The exposed end of that piece of hose also appeared to be freshly cut. The freshly cut ends physically matched and under microscopic examination were confirmed to have been made with a knife.

64 It was the Crown case that the applicant had attempted, without success, to siphon petrol from his car by sucking on a section of garden hose he cut with a knife in order to set fire to the house, but that he was forced to use mineral turpentine as a substitute. The failed siphoning exercise was said by the Crown to be the source of the petrol that Officer Wright smelled on the applicant and what Mr Warner smelled on his breath, thinking it was kerosene. Three samples of carpet from the upper level of the house were removed for testing. Each was confirmed to contain mineral turpentine or a flammable hydrocarbon liquid similar to mineral turpentine. No container of mineral turpentine was found. Evidence was led in the Crown case that light-weight plastic containers would be readily consumed in a fire without leaving any discernible trace.

The applicant's account to ambulance officers and arresting police at the scene - in particular his claim that Christopher had possession of the knife and that the applicant took it from him

65 The ambulance arrived 4.57am and police at 5am, by which time the applicant had been directed by fire fighters to stay clear of the house.

66 Ambulance Officer Maudsley spoke to the applicant and asked him what had happened. He said: "My brother killed my parents; he had a knife and set fire to them". When he was asked by Maudsley why his brother would kill his parents, the applicant said: "There had been a lot of arguments".

67 Maudsley was also present when the applicant was spoken to in the rear of the ambulance by Senior Constable Parsons. When Parsons asked the applicant what happened, he said:

"Was asleep. Mum called me, she was yelling... on the intercom. I came up, he was sitting there, said he'd killed them, he'd lit them and I think I might have killed Chris."

When she asked where Chris was, he said:

"In the lounge room."

When she asked where his parents were, he said:

"On fire near the front door."

When he was asked what happened next, he said:

"I think I killed him. He said he killed them and I think I stabbed him. He was holding the knife."

When the applicant was asked where he was, he said:

"[I] stabbed him in the lounge room and then he [Christopher] ran."

68 Maudsley recalled the applicant telling the police officer that he struggled with his brother and that after he stabbed him (Christopher) he went down the stairs. When the police officer left Maudsley asked the applicant how his brother had set fire to his parents (Maudsley having not detected the smell of petrol). The applicant said: "He just set fire to them". The officer prepared a contemporaneous note which included the following:

"He had initially heard his mother calling for help over the intercom, and when patient arrived parents were alight. Patient states brother had knife. A struggle occurred and he stabbed his brother and his brother went down the stairs."

The applicant's account to police in the ERISP and the walkthrough interview and, in particular, his claim that he picked up the knife

69 The applicant was taken into custody at the house. The following morning he was spoken to by the officer in charge, and he repeated that he had responded to a call from his mother via the intercom and that when he got to the house he was told by his brother that he had killed his parents and set them alight. He was asked what he did then, and said: "I picked up a knife off the floor and chased him and stabbed him and this was [in] the pool room". He told the police officer that he did not know why his brother had killed his parents. He said there had been some "pushing and shoving" between his father and his brother.

70 At 8.22 am on 28 August 1993 the applicant voluntarily participated in a lengthy electronically recorded interview (ERISP) and, later that afternoon, voluntarily participated in a walkthrough interview at his parents' home, which was recorded on a video camera. The walkthrough commenced in the boatshed where he reconstructed his movements from that point through to leaving the upper level of the house en route to the Warners' home via the street to raise the alarm. The police also invited him to identify the knife and his keys (which he did) and invited his response to their finding of the cut hose.

71 In the course of the ERISP the applicant said that he was unsure of what his mother was saying over the intercom, but that she said nothing to indicate to him that she was being assaulted. He said she was just yelling or screaming. He said he did not hear any voices in the background, or any commotion in the house as he rushed up the garden stairs to the upper level. He said that as he came into the house through the glass doors his brother set his mother alight by throwing a match at her or by putting the match to her. He said that when he was near his mother he saw his father was already alight on the bedroom floor, although he did not see his brother go into the bedroom or see him set light to his father. He said the fire spread quickly from his mother to his father. He said he could not smell any chemical or accelerant and could not explain why Mr Warner had smelled kerosene on his breath.

72 He said he made no attempt to put out the fire because he did not think of it. He said he did not know whether his parents were alive or dead, and did nothing to check whether they were alive, but that neither of them were moving. (The Crown submitted that there was ample opportunity for the applicant to put out the fire and that his failure to check on the condition of his parents is inexplicable.) The applicant also said that while there was nothing to prevent him from running from the house to raise the alarm or get help, or using the telephone which was located near the intercom, instead he picked up a knife which was near the piano and chased Christopher down the spiral staircase and stabbed him in the chest. He said that nothing was said by either of them at this time and that he had said nothing in response to his brother declaring that he had killed their parents when he first entered the house.

73 He said his brother was standing right next to the piano, about a metre from his mother's body, when he (the applicant) grabbed the knife. He said he could not remember holding the knife and that he did not know what he did with the knife after stabbing his brother with it. He said after stabbing him he went back upstairs, which was by that time full of smoke to the height of the top of the piano. He said he then left through the glass doors and went up to the street to his neighbours' house to raise the alarm. He said he did not know the time that had elapsed between stabbing Christopher and leaving the house.

74 He told police that after his girlfriend left he watched football on television with his parents. He said he was not sure what time he went to bed but it was before midnight. When questioned about the syringe and the Panadol paste found by police in the downstairs laundry, he said he had some syringes that he was going to use to build a model as part of his civil engineering course. He was unable to account for the paste. Professor Geoffery O'Loughlin, who supervised the applicant as part of his university degree in 1994, gave evidence for the Crown that the syringes and tubing had nothing to do with the particular project in respect of which the Professor supervised the applicant. However, the Professor conceded that the objects may have had something to do with an earlier project or a project with which the Professor was not involved.

75 During the walkthrough the applicant identified the intercom. He said that when he heard his mother's scream he got out of bed, pulled on his shorts, and left the boatshed. He said this took less than a minute. He said that he habitually locked the door of the boatshed on leaving and that he would have locked the door before he went up to the main house in response to his mother's cry, taking the keys with him in the pocket of his shorts. The applicant's keys were found by police on the table immediately inside the glass doors of the main house. In the walkthrough the applicant said he must have removed them from his shorts and placed them on the table as he entered the room. Police also located a pile of clothes belonging to the applicant near a lounge chair in the lounge room. The clothes and upper soles of the shoes were partially burned. The applicant's then girlfriend gave evidence that the applicant was wearing these clothes and shoes on the night of 27 August 1993.

76 It was the Crown case that the applicant's claim that he locked the boatshed and took the keys with him immediately after responding to his mother's cry, and that he left his clothes and shoes upstairs after showering before going down to the boatshed, was a lie. It was the Crown case that the applicant did not go to the boatshed on the evening of 26 August but remained in the house in preparation for the planned murder of his parents. The Crown submitted that he had expected that his clothes would be incinerated in the fire and the fact that they were later found exposed his lie.

77 In the process of the walkthrough he demonstrated the route he took from the boatshed through the garden stairs to the upper balcony. He said that as he ran up the stairs he did not hear any fighting or arguing. (Evidence was led in the Crown case that it would take 22 seconds to run from the boatshed to the glass sliding doors of the main house and, at a steady jog, 30 seconds.)

78 He said that when he walked into the house his brother, who was standing a few metres into the room near the piano, looked up at him and that, as he approached, his brother said that he had killed their parents. He said that by the time he crossed the room to where his brother was standing beside his mother she was "alight". He said his brother had lit a match and put it to his mother as he had walked to where his brother was standing. He said when he reached his mother her feet were alight. He said the fire then spread through the main bedroom to where his father was lying so that both his parents were alight very quickly. He did not enter the bedroom. He said the knife was on the floor below the seat of the piano and that he picked it up by which time his brother was heading to the stairs. He chased his brother downstairs. He said he did not know whether he stabbed his brother on the way down the stairs or when he got downstairs. He said he turned the light on, cornered his brother and stabbed him, and that his brother fell to the ground. He said he thought he kept stabbing him after that but could not recall how many times. He repeated that he did not know what he did with the knife. He then went back up the stairs to street level and went to his neighbours'.

79 The Crown relied upon the inconsistencies in the applicant's various accounts of how and where he obtained the knife in further support of the Crown case.

80 The applicant told police that the previous night he and his father had tried to siphon some petrol out of the car to use for the boat (which was apparently kept near the river) but it was the wrong kind of petrol. He said that was why there was very little petrol in the jerry can that was found by police. During the walkthrough the applicant told police where the jerry can had been left.

81 It was the Crown case that this was a lie the applicant told to conceal the fact that he had unsuccessfully attempted to siphon petrol from his car to set fire to the house, forcing him to use turpentine as an alternative. The Crown submitted that turpentine, by comparison with petrol, was relatively non-volatile. The Crown led evidence from a number of people who knew Mr Gilham to the effect that although it was likely that Mr Gilham would mix his own two-stroke fuel for his boat (perhaps using super unleaded petrol, although standard fuel was more often used), being a meticulous person he would be unlikely to cut up a garden hose to siphon petrol from a car to prepare the fuel, particularly when, according to one witness, Mr Gilham had a siphon pump readily available.

Evidence of activities in the Gilham household and family interaction on 27 August 1993 and the preceding weeks

82 Evidence bearing upon this issue came from witnesses called in the Crown case as well as witnesses called in the applicant's case and the applicant himself.

83 The applicant's high school friend, Wayne Nolan, gave evidence that he went with the applicant to Cronulla after 5pm on 27 August to check surfing conditions. He gave evidence that the applicant told him that Christopher had been acting strangely in that he "was going off at the drop of the hat", especially towards his father, and that "it's come to the push and shove stage between them". When Mr Nolan asked what the applicant thought the conflict was about, he said that it could have been over money because his brother wanted to buy a car.

84 Mr Nolan said that the applicant had mentioned the problem in his family more than once and that he was surprised to hear that Christopher was pushing and shoving Mr Gilham, as it seemed to him to be out of character. He said that the applicant seemed to him to be concerned by his brother's behaviour. He did not get the sense that the applicant was overly dramatising the problem. He said that the applicant refused an invitation to join him to watch a video that evening because he was expecting his father to return from golf and, given the way his brother had been behaving, he wanted to get home. Mr Nolan said he dropped the applicant off at his home at about 8pm with plans to go surfing the next afternoon.

85 The applicant's then girlfriend, Ms Moskos, gave evidence in the Crown case that on 27 August she arrived at her boyfriend's home unannounced. Since the applicant was not at home she sat with his parents in the lounge room. She said Christopher came out from the computer room and said hello while she was talking with Mr and Mrs Gilham. He then took his coffee back with him into the computer room. She said the applicant's mother telephoned Mr Nolan to tell the applicant that his girlfriend was waiting for him. He arrived about 30 minutes later wearing the clothes that were discovered by police on the floor on the upper level the day after the fire.

86 She said they shared a meal and then played pool downstairs before the applicant walked her to her car. She said that she was at the house for about two hours. As the applicant walked to the car he said to her:

"What am I to do about my brother? It's pretty serious. I've never seen him so psycho. He is pushing my father around. I don't know what to do."

87 She said that the applicant seemed genuinely concerned. She said the applicant had told her about Christopher pushing and shoving his father and his general unpleasantness on at least three other occasions.

88 She said that on one occasion about three weeks earlier the applicant told her that Christopher had been stomping around the house throwing things and was "going off" at his parents. About two weeks earlier, during a telephone conversation, the applicant told her that he was going to use the other phone (which she understood to be downstairs) because Christopher wanted to use the computer room upstairs and was "going off again". During the course of that conversation the applicant said:

"Wait, Dad and Chris are having an argument. Chris is being really weird lately. Him and Dad have been pushing and shoving each other."

89 She asked the applicant if he knew what the argument was about and he said no. She suggested that he listen to the conversation but he said, "I don't know what's going on but I think it's to do with money". She asked him what money and he said "I don't know ... if there is a fight I'll have to break it up". She said he kept stressing that his brother had "lost it". He asked her advice about what he should do. She suggested that he tell his brother to grow up and that they (he and his father) should talk like adults.

90 She gave evidence that there were no signs of aggression from any of the family members while she was at their home that evening. She said the applicant's mood was normal. He was casual, relaxed and friendly and she did not sense any friction or tension between him and his parents or his brother. She gave evidence that Christopher seemed a little edgy and, while he said hello when he came out of the computer room, he did not interact with her (although this was not unusual). She discerned some underlying tension between Christopher and his parents. She said that to her observation Christopher spent a lot of time on the computer and normally kept to himself. She also gave evidence that she did not hear the applicant talk with his father about petrol for a boat, nor did she hear any conversation about a planned boating trip with his father the next day, although it was not unusual for the applicant to be on a boat on weekends. She agreed in cross-examination that any talk about boats or plans involving boats would not have particularly interested her and she would be unlikely to have paid it any attention.

The applicant's evidence

91 The applicant gave evidence largely consistent with the account he gave to police at the scene, as elaborated upon in the ERISP and the walkthrough he participated in the following day.

92 In his evidence, however, he said that as he entered the house through the glass sliding doors and walked over to his mother, his brother was standing near the bottom half of her body, which was then ignited by Christopher, and that, on the approach, he (the applicant) could see the fire spreading. He also said he could at that time see his father's body alight on the floor of the bedroom and that the fire was spreading fast. He agreed that when he was interviewed he told police that when he looked from where his mother was lying and saw his father on the floor in the bedroom, there was no fire in that room. He said he was dazed when he was interviewed by police but that he was now sure that his father was alight, although he had only seen his brother ignite his mother. He denied that he was trying to convey to police that the fire had travelled so fast that there was nothing he could do to stop it, although he did agree that his impression was that the fire gathered speed and moved quickly. He accepted in cross-examination that when his mother called him from the boatshed, only minutes before he entered the house, he knew she needed help. He could not, however, account for why he said nothing to the effect that he was there to help her, or to check to see whether she was injured but still alive. He said he did the wrong thing by not trying to put the fire out. He agreed that when Christopher said he killed their parents, even though it was out of character for him to have acted with such violence, he did not question him but simply assumed his parents were dead.

93 He said that when he chased Christopher he was not thinking that he had to kill him but rather that he had to "get him". He said he remembered chasing his brother and hitting him but did not remember stabbing him going down the stairs, or being in the pool room downstairs, although he accepted the evidence suggested that he stabbed his brother in the back on the way down the stairs. He agreed that he had stabbed him more times than was necessary to kill him. Although the red marks on his forearms observed by Dr Jennings at the police station after his arrest were consistent with him having struggled with his brother, and although the fact that the balls from the pool room table were found on the floor downstairs was also suggestive of a struggle, he said that he was unable to recall the details of any struggle.

94 He denied that the reason for the differing accounts he gave to emergency services officers, ambulance officers and police as to where the knife was when he came into the room was that he realised by the time he came to be formally interviewed by police that his brother could not have been holding a knife and lighting a match at the same time, and that he would need to change that aspect of his story to accommodate that fact. His evidence was that he did not remember seeing his brother with the knife but he did remember picking up the knife from the floor. He was unable to explain how he would have been able to see the knife on the floor in the dark. He was also unable to offer any explanation for why his brother would have left the knife on the floor knowing that their mother had called for his help on the intercom, or why his brother would run down the spiral staircase and not out of the house, making his escape by that route.

95 He said that as he was leaving the house after stabbing Christopher he could not remember any heat, but he did observe smoke. He said he had no difficulty breathing. He could not remember whether he looked towards his mother as he left to go to the Warners' house. He agreed that there were four phones in the house and that he had not used any of them to summon help. He said he went to the Warners' house via the front gate. He denied that he had washed himself or the knife before leaving or that he deliberately placed the knife close to his brother in order to implicate his brother in the murders of his parents. He was unable to account for why he was cold after running from a burning house. He denied that he was waiting outside the backyard area calmly and calculatedly in the night air for to the fire to take hold and to avoid being seen. He said he had no memory of anything that happened in the Warners' house.

96 As regards the clothes found by police in the upstairs lounge room, the applicant gave evidence that he would normally have a shower in the evening in the main bathroom of the main house, leaving his clothes behind because his mother did his laundry. He said he followed that routine on the night of 27 August and then went to the boatshed to sleep. He denied that he remained in the house to carry out a premeditated plan to kill his parents. He denied any knowledge of the syringes and the Panadol paste in the laundry basket downstairs. He denied carrying the axe to his father's bedroom (which on his account was usually kept on the lower level of the house, underneath his brother's bedroom) as a backup if the plan to stab his parents did not work.

The grounds of appeal

Ground 1: The trial judge erred in failing to stay proceedings.

Ground 2: The trial judge erred in failing to direct the jury so as to give full effect to the previous acquittal of the applicant for the murder of his brother.

Ground 3: The trial judge erred in allowing the Crown Prosecutor to controvert the previous acquittal of the accused for murdering his brother.

Ground 4: His Honour erred in permitting into evidence a fire demonstration conducted by the Crown's fire expert, Mr James Munday.

Ground 5: His Honour erred in allowing into evidence opinions drawn from fire demonstrations conducted by the Crown's fire expert, Mr James Munday.

Ground 6: His Honour erred by allowing evidence from Dr Culliford and Dr Cala regarding their expectations of how much blood should have been on the knife.

Ground 7: His Honour erred by allowing evidence from Dr Culliford regarding her expectation of much blood the applicant was likely to have on him.

Ground 8: His Honour erred by allowing evidence from Dr Culliford, Dr Cala and Dr Lawrence regarding the similarity of the pattern of stab wounds.

Ground 9: That his Honour erred in permitting the Crown Prosecutor to hold a knife before the jury and have the applicant demonstrate his dexterity with the knife.

Ground 10: The trial miscarried because the Crown Prosecutor engaged in cross-examination of the accused in an improper manner.

Ground 11: The trial miscarried because the Crown Prosecutor addressed the jury in a manner that undermined the effect of the directions his Honour gave concerning the need to scrutinise the fire demonstrations with great care.

Ground 12: The trial miscarried because the Crown Prosecutor addressed the jury in an improper manner.

Ground 13: That the verdict was unreasonable or cannot be supported by the evidence.

Ground 14: The applicant seeks to rely upon fresh evidence that the applicant's university timetable indicates that he would have been attending university on the afternoon that his mother left her place of employment two hours earlier than usual, indicating at that time that she was worried about one of her sons. The applicant seeks to rely upon fresh evidence from Steven Audet that the applicant attended a university lecture with him on the same Thursday afternoon. The corollary is that the son to whom the applicant's mother was referring was Christopher Gilham, not the applicant.

Ground 15: The applicant seeks to rely upon fresh evidence that there was a bloodied fingerprint on the intercom unit, thereby indicating that someone - who could only have been the now deceased mother of the applicant - had touched the intercom with her finger, which was consistent with the applicant's case at trial, and given contemporaneously to investigators, that she had contacted him by the intercom.

Ground 16: The applicant seeks to rely upon new evidence that the level of carbon monoxide in the lungs of the deceased Christopher Gilham indicates that he was alive and probably upstairs at the time of the fire.

Ground 17: The applicant seeks to rely upon new evidence that the fire is likely to have occurred in the manner described by the applicant and that the flames, far from being of a low height in the manner described in the evidence for the Crown by Mr Munday, were such as to prevent the applicant from any attempt to assist his mother.

Ground 18: The failure of the prosecutor to call Dr Cordner after the admission of the evidence of Dr Culliford, Dr Cala and Dr Lawrence regarding the similarity of the pattern of stab wounds, when viewed against the conduct of the trial taken as a whole, gave rise to a miscarriage of justice.

97 Under Ground 16 the applicant seeks to rely upon new evidence dealing with the level of carbon monoxide in Christopher's lungs at the time of his death as contradicting the Crown case and supporting his case. In light of evidence from various witnesses received on the appeal, this ground of appeal also deals with the levels of carbon monoxide in the bodies of Mr and Mrs Gilham. To the extent necessary, we would grant the applicant leave to amend this ground of appeal to accommodate the fact that the new evidence also concerns the levels of carbon monoxide in the bodies of his parents.

98 There are three grounds of appeal against sentence. However, having regard to our conclusion with respect to the appeal against conviction it is unnecessary for us to consider these grounds.

Ground 1: The trial judge erred in failing to stay proceedings.

Ground 2: The trial judge erred in failing to direct the jury so as to give full effect to the previous acquittal of the applicant for the murder of his brother.

Ground 3: The trial judge erred in allowing the Crown prosecutor to controvert the previous acquittal of the accused for murdering his brother.

99 The first three grounds of appeal deal with the question of whether the trial ought to have been stayed, or an acquittal ought to have been directed, because the conduct of the trial challenged, or else controverted, the conviction of the applicant for the manslaughter of his brother.

100 It is appropriate, as the parties did, to consider these grounds together. The first step is to understand the history of the prosecution of the applicant.

101 On 21 February 2006, the Director filed an ex officio indictment in the Supreme Court charging the applicant with the murder of his parents.

102 On 29 September 2006, the applicant sought, by notice of motion in the Supreme Court, a permanent stay of the proceedings against him upon the basis that it would be an abuse of process to permit the Crown to proceed against him on the charges of murder.

103 There were three grounds for that application. The first was that because of the delay in commencing proceedings, with accompanying prejudice to the applicant, it would be oppressive to subject him to a trial. The second was that it was not open to the Crown to seek from a jury verdicts that would be inconsistent with the applicant's conviction of manslaughter following his plea of guilty to that offence in 1995. The third ground was that it would be vexatious and oppressive to bring a further prosecution arising from the events of August 1993.

104 The notice of motion was heard by Howie J on 9-11 October 2006. On 21 March 2007, he delivered judgment, refusing to stay the proceedings: R v Gilham [2007] NSWSC 231; (2007) 190 A Crim R 303.

105 On 21 August 2007, an appeal to the Court of Criminal Appeal was heard against the refusal by Howie J to grant a stay. Judgment was delivered by this Court on 26 November 2007, dismissing the appeal: Gilham v R [2007] NSWCCA 323; (2007) 73 NSWLR 308.

106 On 8 February 2008, the High Court of Australia dismissed the applicant's application for special leave to appeal against the judgment of this Court of November 2007: Gilham v Regina [2008] HCA Trans 85. The Chief Justice expressly indicated that the applicant would not lose his right to raise the same issues in the event of his conviction after trial.

107 Pre-trial argument commenced on 4 February 2008 and the trial judge, Howie J, gave various rulings on evidentiary matters which he was asked to address in advance of the jury being empanelled. On 11 February 2008, a jury was empanelled to hear the trial. They were discharged on 18 April 2008 when they were unable to agree upon a verdict.

108 The applicant's second trial commenced on 13 October 2008. It was agreed in the course of this trial that rulings which had been given upon evidence in the first trial would be adopted without the need to re-argue the same issues. This, given the fact that the same trial judge was presiding, was a sensible and efficient way to manage the second trial.

109 The second trial commenced on 13 October 2008 and concluded on 28 November 2008, when the jury returned a verdict of guilty of both counts of murder. The jury had deliberated for eight days.

110 The applicant was remanded in custody. Submissions on sentence were made on 13 February 2009. On 11 March 2009, Howie J sentenced the applicant to life imprisonment on each of the charges of murder: R v Gilham [2009] NSWSC 138.

111 The foundation for the earlier appeal to this Court (Gilham v R [2007] NSWCCA 323; (2007) 73 NSWLR 308) was a submission that the trial judge erred in finding that, in instituting the proceedings, the Crown had not controverted the earlier verdict and had not, therefore, acted contrary to the rule against double jeopardy. Spigelman CJ resolved the appeal on a different basis to the other judges. His Honour held that where "the Crown accepts a plea to manslaughter in full discharge of an indictment for murder the appellant cannot be said to have been in jeopardy on the murder charge" (at [87]). The Chief Justice concluded that the evidence available to the prosecution could not support a conviction on the charge of murder (at [95]). It followed that the acquittal of the charge of murder "was not an acquittal which could attract the double jeopardy principle" (at [96]) and the breach of the incontrovertibility principle did not justify the grant of a permanent stay.

112 The other judges reached the same ultimate conclusion as the Chief Justice but their reasoning process differed. McClellan CJ at CL concluded that the applicant was "relevantly in jeopardy when acquitted of the murder of his brother" (at [197]). However, his Honour determined that although "the factual basis for both the applicant's acquittal for the murder of his brother and conviction for his manslaughter are in conflict with his prosecution for the murder of his parents" (at [209]) the prosecution for the murder of his parents would not be "an abuse of process" requiring the Court's intervention (at [222]).

113 Hulme J carried out his own analysis of the authorities and concluded that in the circumstances of the case and, in particular, having regard to the fact that the indictment alleged the applicant had murdered two persons, there was no principle of law which required that the trial be permanently stayed. Hidden and Latham JJ agreed with McClellan CJ at CL on these issues.

114 The applicant formally submitted that the earlier decision of this Court was wrong and renewed his right to pursue a further application for special leave.

115 At his trial, counsel for the applicant submitted that the trial judge should exclude all evidence which called into question the applicant's conviction for manslaughter and acquittal for the murder of Christopher. It was further submitted that the reasoning of the majority in this Court in the first appeal meant that the trial judge was required to direct the jury to acquit the applicant. The trial judge rejected these submissions, ruling that the appropriate warnings or directions, rulings on admissibility of evidence, and other orders would cure any possible prejudice and ensure a fair trial.

116 It is unnecessary to repeat the arguments put before this Court by which it was contended that the Court's previous decision was wrong. The novel issue in the present proceedings is the consequence for the applicant's trial of the previous decision with respect to the killing of Christopher.

117 The applicant submitted that he had been denied the full benefit of the acquittal for the murder of his brother: Garrett v The Queen [1977] HCA 67; (1977) 139 CLR 437. Because the issues in the trial involved a traversing of issues settled by the earlier verdict, it was submitted that the principle of res judicata would operate to prevent the trial proceeding: R v Storey [1978] HCA 39; (1978) 140 CLR 364 at 396.

118 There was debate at the trial as to the directions which the trial judge should give with respect to the prosecution of the applicant for the death of his brother.

119 The directions given by his Honour included the following:

"There have been a number of proceedings dealing with this matter over a long period of time. You will be aware by now that the killings occurred in 1993,15 years ago. There is a history as to why that is so, and you will learn about it. The history of it is relevant only as it relates to the evidence and the reliability of evidence. It has no other relevance in this matter. It may, however, explain why some evidence has been lost and is no longer available. It may explain why some people have given different evidence at different occasions. But that is the only relevance that the lengthy delay in this matter has. You should not be otherwise concerned at the delay or the history of the matter.

What happened is this: That the accused was charged initially with the murder of his brother. The three of them were killed: The mother, the father and brother. Three of them were stabbed to death. There was a fire set to the house which incinerated the mother and the father. The brother's body was downstairs. I think all of this was uncontroversial. Normally I know as much as the jury does about the trial and about the issues. But I was the trial judge in the earlier trial and therefore I have some knowledge of the matter and of the evidence that I expect to be called.

Anyway, the accused pleaded guilty to the manslaughter of his brother on the basis of provocation and was dealt with by this court. Over a long period of time there were questions as to whether or not he should be charged and dealt with, or charged with the murder of his mother and father, and ultimately the Crown decided that he ought to be.

Again, none of that has any relevance except as to the history of the matter. It explains to you why we are here in 2008 dealing with the matter in 1993 and, as I have said to you, it will explain to you why some evidence is no longer available. What you make of that we will talk about later. It may also explain why some witnesses have no memory any longer of the events and why they have to rely upon the statements they made back in 1993. Again, we will talk about that at a later stage when you are aware of what the evidence is.

So there will be during the course of this trial some reference, I would think, to other occasions when witnesses have given evidence, either before the Coroner or at some other Inquiry or in the last trial in this matter. You should not be concerned at all that a witness has given evidence on other occasions. The only relevance of that is that it may reflect upon the evidence they give on this occasion. The evidence that you are concerned with is the evidence that you hear in the witness box of this court or evidence that is in some other way placed before you. But the fact that a witness has given evidence on other occasions may be significant when you come to look at the reliability, that is, whether or not you can accept, over the long delay, the evidence that's given before you."

120 His Honour gave the jury the following further directions in his summing up:

"Now, you will be aware by now that the accused was not charged with the murder of his mother or father in 1993. In fact, he was not charged with those offences until 2006. You know that he appeared in the Supreme Court in 1995 on an indictment charging him with the murder of his brother. He pleaded guilty to manslaughter, and that plea was accepted by the Crown; the result being that the Crown no longer prosecuted or continued the prosecution for murder. As a result, he was sentenced in the Supreme Court for the charge of manslaughter.

You have heard the statement of facts that were tendered, and they are in evidence before you. You cannot put too much weight on those really before you finish the picture, as it were, of this history of the accused's involvement with the criminal law up to 1995.

I pointed out to you that one obvious statement in the statement of facts is clearly wrong, and that is that the statement suggests the accused never went back up to the first storey of the house. Well we know he did. We do not know how it was that that statement of facts got fashioned in that way. I do not think it mattered the slightest bit to the sentencing judge, that particular fact.

One use that has been made of that statement of facts is that Mr Boulten indicated to you, during the course of his address, that there was a concession by the writer of those facts that the accused's account was feasible. That is irrelevant. Whether or not a police officer who wrote those facts thought that the accused's account was feasible back in 1995 has absolutely nothing to do with this issue before you.

Similarly, the suggestion in the statement of facts that there was a complete and thorough investigation, you might think, is rather putting the hyperbole on what was in fact a pretty ordinary investigation, but again that is not a matter with which you are concerned.

I have already told you that the fact that the police decided not to charge him with murder of his parents in 1993 is neither here, nor there. That is a decision made by particular police officers with a particular set of circumstances. Whether or not they saw any significance in the stabbing of the three deceased and what is accepted now to be the similarity of the wounds, again is neither here, nor there. Whether the Crown was correct to concede that the accused should be sentenced for manslaughter of his brother is neither here, nor there. They are decisions made by individual persons for reasons best known to that particular person in the circumstance of the facts that that particular person had before him or her at that particular time.

However, you might conclude that it would have been absolutely absurd to have this trial conducted without any reference at all to the history of these matters. You can hardly ignore that there was another body downstairs, the body of his brother. You could hardly ignore in the circumstances of this matter the accused's explanation as to the killing of his brother which is so far a part, so much a part of his explanation and account of the events. It would have led to the position that nobody would have been able to come along here and tell you things that related to the fact that the accused was charged in 1995 and pleaded guilty to manslaughter. I have told you that it may explain why the exhibits were lost. It may explain why the accused now lacks a detailed memory of some of the events.

It is important, when it comes to the consideration of the question of his character, that he pleaded guilty. It has been put to you by Mr Boulten that it is a consistent approach that he has always taken. It was put to you that it was relevant that he has always, right from the start, said, in effect, that he killed his brother because of provocation. True it is that he was not able to tell the police in the record of interview why it was he killed his brother, but you might think that it may have been difficult at that time for him to enunciate in any degree of specificity what it was and what the emotions were that caused him to kill his brother, and it may, it is a matter for you, not be unreasonable for him to say in that situation, "I can't tell you why."

That is a matter for you, but it is all part of the circumstances in the history of this matter. As I have made it perfectly clear, this is not an inquiry and your verdict will not decide the correctness or otherwise of the police conduct in 1993 or the Crown's conduct in 1995.

There are two other reasons why it it is important that you have this history. One is that it allows people to come to this court and tell you facts and circumstances particular about the accused's character without any limitation based upon the fact that the accused was charged and dealt with for the manslaughter of his brother in 1995.

The other is that it shows a consistency in the approach of the accused and that it shows that he has always maintained the same story, the same account generally in relation to these events."

121 Counsel for the applicant sought stronger directions. His request was refused although his Honour recorded that objection had been taken and the applicant's position was accordingly protected.

122 At his first trial defence counsel submitted that evidence should be introduced of the fact that in 1995 the applicant was acquitted, by operation of law, of murdering his brother. His Honour refused that application. The submission was repeated in the second trial.

123 It was submitted to this Court that his Honour had erred by refusing to direct the jury that the applicant had been acquitted of murder. It was submitted that his Honour was erroneously concerned to avoid the need to direct the jury that they should give full effect to the applicant's previous acquittal for murdering his brother. It was further submitted that the proper application of the incontrovertibility principle would have the effect that his Honour should have directed the jury to acquit the applicant.

124 After failing to convince the trial judge either to direct an acquittal or to give directions that the applicant should be given the full benefit of the acquittal for the murder of his brother, the applicant argued, in the alternative, that his Honour should direct the jury that the Crown's acceptance of the plea of not guilty for murdering Christopher Gilham, and the Court's judgment and orders convicting the applicant of manslaughter, were relevant to the jury's determination of whether or not the applicant was guilty of murdering his mother and father.

125 It was submitted to this Court that his Honour's directions about the relevance of the 1995 proceedings were inadequate. The applicant submitted that the prosecution was contending that the applicant's explanation for killing his brother, namely that Christopher had killed their parents, was untenable, even though this account had been consistent from the applicant first reporting the killing to his neighbours and had been accepted as a reasonable possibility when the Crown accepted his plea of guilty to the manslaughter of his brother in full discharge of the murder indictment. The applicant referred to the fact sheet in the sentencing proceedings before Abadee J, which described the applicant's account as "feasible".

126 It was submitted that the jury were entitled to gauge the validity of the Crown's assertion that the applicant's version of events could not reasonably be true by having regard to the facts that demonstrated that the police regarded his version as feasible. Relevantly, the Office of the Director of Public Prosecutions accepted in 1995 that they could not negative it and the Court entered a judgment of "not guilty" of murder and "guilty" of manslaughter of Christopher Gilham on the basis that the prosecution could not negative the possibility that Christopher Gilham killed his parents. It was submitted that these were all facts that were relevant to an assessment of the applicant's guilt in the instant case. The applicant argued that fairness required that the jury be directed to weigh these facts in the balance.

127 In the course of the trial, the applicant pressed his objections, which included objection to the Crown Prosecutor advancing propositions in cross-examination of the accused to suggest that he had murdered his brother and that his brother had not murdered his parents. The applicant submitted that this amounted to running the case in a manner which was inconsistent with the sentencing proceedings and the applicant's acquittal of the murder of his brother.

128 The applicant submitted that problems emerged early in the trial. He identified that the Crown Prosecutor opened to the jury in the following terms:

"It's common ground, ladies and gentlemen, that means we all agree, that each of the killings was done by a member of the family. The Crown case is that the accused, Jeffrey Gilham, killed his father, his mother and his brother that morning. The defence case, I expect you will hear, is that Jeffrey Gilham killed his brother Christopher after Christopher had killed their parents."

129 Accordingly, it was submitted that from the outset of the proceedings, the prosecution was contending that the applicant not only murdered his parents but that he killed his brother in circumstances where there were no circumstances of provocation justifying a verdict of manslaughter. It was submitted that the Crown Prosecutor, in conducting the case in this manner, had accentuated the controvertibility issues. It was further submitted that in her closing address to the jury, the Crown Prosecutor highlighted the difficulties of incontrovertibility when she made the following submissions:

• that "it is inconceivable that two of them suddenly became killers in the space of 5 minutes";
• that "[the murderer] was Jeffrey Gilham calmly, calculatedly, over the period of three to four weeks, planning to kill his family and his plan was always to blame his quiet, older brother for the deaths of his parents";
• that "his plan included an outcome in which he could almost paint himself as something of a hero, something of a hero, that he couldn't help killing his brother because of the horror and grief that his brother had killed his parents";
• that "Chris Gilham needed his mum and dad" ;
• that "the accused didn't know, didn't care about his brother";
• that there "really is an undercurrent of disdain and a complete lack of respect for that older brother"; and
• that the jury should speculate as to the applicant's reason for killing his brother ("Did we see any respect for his brother? Far from it" and "why did he stab him, not once, and think "Oh my God, what have I done?" I've got him, he's falling." No....There was a lot of hate in that stabbing".

130 It was submitted that the Crown Prosecutor then explicitly controverted the applicant's acquittal for murdering Christopher by making the following submissions:

"The accused has admitted that when he stabbed his brother he, too, clearly intended to kill his brother". [T2.1405]

...

"The real situation, of course, is that it wasn't his brother with a knife, it wasn't his brother lighting the fire, it was this accused setting his dead brother up, and it doesn't stand together and it changes at that point". [T2.1430]

131 In summary, the applicant's submission has two limbs. It was first submitted to this Court that the trial judge was obliged to give full effect to the previous acquittal (for the murder of his brother), which would result in an acquittal by direction (for the murder of his parents). Secondly, it was submitted that the prior acquittal of murder and conviction of manslaughter was a relevant fact when determining whether the applicant murdered his parents.

132 The applicant relied upon what was described as the principle in Garrett v The Queen, where a complainant had, during a rape trial, been permitted to give evidence of a previous alleged rape for which the accused had been acquitted. The trial judge gave the following direction (at 442-3):

"So it is also with the evidence that you have heard about the past prosecution for rape and the accused's acquittal on that charge. Again it would be an entirely wrong use of the evidence you have heard about that, meagre as it is, to say on the one hand, well, she couldn't pin it on him last time but it is a different story this time, or on the other hand, to say - she cannot make it stick this time any more than she could make it stick last time. Such an approach would be plainly wrong. The fact is we do not know and cannot know the basis of the jury's verdict, and we are not to speculate about it. We are not trying that case; we are trying this one. We do not know whether the verdict means that the jury thought that the accused was innocent in the true sense, or whether they thought the charge was not proven. We do not know whether they believed [the complainant] or not. We do not know whether there was any corroboration-I shall talk to you about this later - and if there was not, whether the jury simply heeded the judge's warning that it might be unsafe to convict without it. So you are to draw no inference either for or against the accused or for or against Miss Golding from the fact of that prosecution and its outcome."

133 The High Court held that the trial judge had erred by admitting evidence in relation to the previous allegation of rape. The Court also held that the summing up was defective. Barwick CJ said (at 444-5):

"It is apparent, in my opinion, that the trial judge, in an understandable endeavour to dissuade the jury from embarking on a consideration of the facts relating to the former charge of rape, did in substance tell the jury that the acquittal was a neutral fact. But it was not neutral and, if the prosecutrix were rightly permitted to give the evidence she gave, the acquittal was a dominant fact of which the applicant was entitled to full credit. To have said that the acquittal was neutral was to deny the applicant the full benefit of the acquittal and not to lay that emphasis upon it for which the circumstances called. Further, having regard to what I have earlier said as to the admissibility of the prosecutrix's evidence as to the events of November 1975, his Honour's explanation of the presence of the evidence before the jury was, in my opinion, unacceptable. Consequently, had that evidence been admissible, I would have thought the portions of the summing up which I have quoted would have themselves been wrong to a significant degree." (Emphasis added)

134 The applicant submitted that the prosecutor, by building the Crown's case on the premise that the applicant's account of what happened on the evening was specious and inconsistent with human understanding, had effectively called into question the earlier acquittal of the applicant which the Crown consented to - "for if it defied logic to so accept the [applicant's] account, why did the Crown and the Court act as they did on the earlier occasion"?

135 These grounds of appeal raise issues of some difficulty. There is no case directly in point and plainly the facts of this case are unusual if not extraordinary. There is no manifest inconsistency in the verdicts of guilty of the manslaughter of Christopher and the murder of the parents. The fact that the applicant admitted that he was guilty of the manslaughter of Christopher is not inconsistent with his being accused of killing his parents. The issue in the applicant's trial was whether he had killed his parents. It was not in dispute that he had killed Christopher. It is of course true that the Crown accepted the plea to manslaughter in circumstances where the applicant asserted that it was his brother who killed their parents. However, there was no determination of that issue, the conviction for manslaughter evidencing only that the Crown accepted that it could not discharge the onus it carried to "disprove" provocation. Furthermore, there was no finding that the applicant had not killed his parents, although it must be assumed that the Crown concluded that the evidence which was available could not prove that Christopher did not kill his parents. As McClellan CJ at CL said in Gilham at [204]: "there was no verdict acquitting the applicant of the killing of his parents. Furthermore, he has never been in any sense 'in jeopardy' for the alleged killing by him of his parents."

136 In R v Carroll [2002] HCA 55; (2002) 213 CLR 635 the High Court considered a controversy concerning perjury proceedings which involved the same "ultimate and live issue" (at [101] (Gaudron and Gummow JJ)) that had been the subject of the acquittal of the respondent for the killing of a child. The Crown brought proceedings in which it alleged the respondent had lied when he denied that he was the killer (at [103]).

137 This Court discussed Carroll and the previous High Court decisions on this subject in its earlier decision. The inconsistency identified in Carroll, which had the consequence that the trial on the charge of perjury was permanently stayed, "arose because the prosecution based the perjury charge solely on the respondent's sworn denial of guilty" (at [649]). The verdict of acquittal necessarily meant that the jury had not rejected the respondent's denial that he killed the child. That inconsistency being manifest, Gleeson CJ and Hayne J said that "the case for a stay of proceedings was irresistible" (at [42]). A finding that the respondent had perjured himself would have conflicted with the previous verdict of the jury.

138 Gleeson CJ and Hayne J noted that the principle that an acquittal is incontrovertible rests upon an acceptance that what is decided in litigation is final. However, their Honours provided this important qualification at [50]:

"Finality of a verdict of acquittal does not necessarily prevent the institution of proceedings, or the tender of evidence, which might have the incidental effect of casting doubt upon, or even demonstrating the error of, an earlier decision. There may be cases where, at a later trial of other allegedly similar conduct of an accused, evidence of conduct may be adduced even though the accused had earlier been charged with, tried for, and acquitted of an offence said to be constituted by that conduct. R v Z [[2000] 2 AC 483], R v Arp [[1998] 3 SCR 339] and R v Degnan [[2001] 1 NZLR 280] are cases of that kind. In such cases, the earlier acquittal would not be controverted by a guilty verdict at the second trial."

139 In the present case the applicant was charged with the murder of his parents, an issue which has not previously been litigated and in respect of which there was no judicial determination. Of course, the earlier proceedings resulting in his conviction for manslaughter necessarily involved an acceptance by the Crown that it could not prove that the applicant had murdered his brother. However, there was no question raised, and no determination in the earlier proceedings, as to whether the applicant had killed his parents, which was the issue at trial.

140 Although rejecting the applicability of the civil law doctrine of issue estoppel to Australian criminal law, Gaudron and Gummow JJ recognised in Carroll that the common law protected defendants acquitted of an offence from retrial for a subsequent offence where (i) the elements of the two offences are identical or (ii) the elements of one offence are wholly included in the other (at [91]). Their Honours said:

"Notwithstanding the inapplicability of the civil doctrine of issue estoppel in Australian criminal law, the common law of this country has sought to protect defendants acquitted of an offence from retrial for a subsequent offence where (i) the elements of the two offences are identical or (ii) the elements of one offence are wholly included in the other. Pearce v The Queen [[1998] HCA 57; (1998) 194 CLR 610] determined that that degree of coincidence between the elements of the two offences will ground a plea of autrefois acquit at common law. Again, this Court held in Rogers [v The Queen [1994] HCA 42; (1994) 181 CLR 251] that the tender of evidence of admissions which had been found in a prior trial not to have been made voluntarily would challenge directly that prior judicial determination, which had become final once verdicts were returned, and, on that basis, would constitute an abuse of process. Similarly, in Garrett v The Queen, evidence of a previous charge of rape of the prosecutrix, for which the accused had been acquitted, was held by this Court to have been inadmissible in a later trial of the accused in which he was convicted of rape and abduction in respect of different allegations involving the same complainant. The evidence was said by this Court to be inadmissible because it "inevitably challenged the verdict of acquittal" [at 445]. The Court allowed an appeal against the later convictions and ordered a new trial.

However, in R v Storey, Stephen, Mason, Jacobs and Aickin JJ held that relevant evidence tending to show that the accused was guilty of an offence of which the accused had been acquitted may be admitted, but only if the jury can be and is directed not to interpret it in such a way as to deny the acquittal. In Storey, the two accused had been acquitted on a charge of forcible abduction; at their subsequent trial for rape, evidence was admitted which tended to show the forcible abduction of the victim, but without it being made clear to the jury that the evidence must not be taken as proving guilt on the previous charge. This Court dismissed the Crown appeal against the quashing of the conviction."

141 Analogously to Storey, evidence of the circumstances surrounding the killing of Christopher was relevant evidence which, on one view, tended to show that the applicant was guilty of an offence of which he had been acquitted. It was admissible if the jury could be, and was, directed not to interpret the evidence in such a way as to deny the applicant's acquittal of Christopher's murder.

142 It is instructive to consider the cases cited by Gleeson CJ and Hayne J in Carroll in support of the proposition that "Finality of a verdict of acquittal does not necessarily prevent the institution of proceedings, or the tender of evidence, which might have the incidental effect of casting doubt upon, or even demonstrating the error of, an earlier decision". One of those cases is the English decision of R v Z [2001] 2 AC 483. The defendant in that case was charged with rape. He had been prosecuted for that offence on four previous occasions, being convicted once. At his trial, the Crown proposed to call "similar fact" evidence from the four previous complainants in an attempt to negative the defendant's defence of consent or belief as to consent. The trial judge ruled, on the basis of the rule against double jeopardy, that the evidence from the three complainants of whose rape the defendant had been acquitted was inadmissible. The House of Lords reversed that decision. Lord Hutton delivered the leading speech. His Lordship reasoned that the evidence was not inadmissible merely because it showed or tended to show that the defendant was in fact guilty of previous offences of which he had been acquitted (at 504). Influential in Lord Hutton's reasoning was the fact that the evidence from the three complainants was not being used to show that the defendant was guilty of the offences for which he had been acquitted, but rather to prove, by similar facts, his guilt of the offence for which he was being tried (at 499). Lord Hutton held that in those circumstances, the principle of double jeopardy would not be infringed by the admission of the similar fact evidence.

143 The second case cited by Gleeson CJ and Hayne J in Carroll is the Canadian Supreme Court decision of R v Arp [1998] 3 SCR 339, which also involved similar fact evidence. The appellant in that case had been convicted of two murders three years apart, charged on a single indictment. The killings bore a striking similarity to each other. The trial judge instructed the jury that they could use the similar fact evidence to reason that whoever killed one of the deceased must have killed the other. On the appeal, the appellant argued that it was wrong for the trial judge to instruct the jury in those terms. He drew an analogy with cases where the Canadian Supreme Court had excluded similar fact evidence that underlay a prior acquittal. On the strength of those precedents, the appellant submitted that because it was possible for the jury to convict the appellant on one murder count, but not the other, the jury's verdict was "tainted" to the extent that it relied on the similar fact evidence. Cory J, writing on behalf of the Court, rejected that argument. The Court pointed out at [77] that the Canadian decisions which prohibited the use of evidence underlying an acquittal in a later trial of the same accused did so on the basis of the Court's decision in R v Grdic [1985] 1 SCR 810, where Lamer J said at 825:

"There are not different kinds of acquittals and, on that point, I share the view that 'as a matter of fundamental policy in the administration of the criminal law it must be accepted by the Crown in a subsequent criminal proceeding that an acquittal is the equivalent to a finding of innocence' ... To reach behind the acquittal, to qualify it, is in effect to introduce the verdict of "not proven", which ... has never been ... part of our law.

However, this does not mean that, for the purpose of the application of the doctrine of res judicata, the Crown is estopped from re-litigating all or any of the issues raised in the first trial. But it does mean that any issue, the resolution of which had to be in favour of the accused as a prerequisite to the acquittal, is irrevocably deemed to have been found conclusively in favour of the accused."

144 It will later be necessary to address the point made in Grdic that a verdict of not guilty is to be treated in subsequent criminal proceedings as a finding of innocence. That proposition, if accepted, would have obvious implications for the applicant's case. For present purposes, however, it is relevant to note that the Arp Court held that the principle in Grdic did not apply to verdicts rendered by the same trier of fact where the charges are tried together in the same proceeding (at [79]).

145 The third case to which Gleeson CJ and Hayne J referred in Carroll is the New Zealand decision of R v Degnan [2001] 1 NZLR 280, again a case involving similar fact evidence. The applicant in that case was charged with indecent assault on a male. The Crown sought to lead evidence from two other men that the applicant had subjected them to similar assaults. The applicant had been charged in relation to those alleged assaults. He was acquitted in one case and the other proceedings were stayed after two juries were unable to reach verdicts. The issue for the New Zealand Court of Appeal was whether similar fact evidence underlying a previous acquittal was inadmissible as a matter of law. The Court followed R v Z in holding that it was not (at [31]-[35]). The Court held that the proposed use of the similar fact evidence did not raise any question of double jeopardy. The applicant was entitled to the full benefit of the earlier acquittals, in the sense that he could never be tried again for the offences to which they related. But he did not have the added benefit of protection in future proceedings from relevant evidence that arose from events that were the subject of already concluded proceedings (at [33]).

146 R v Z and Degnan suggest that the rule against double jeopardy does not operate where the way the prosecution seeks to prove its case incidentally has the effect of calling into question or proving the error of a previous acquittal. It would otherwise be a breach of the rule where that was the purpose of the prosecution. The applicant's case does not fall foul of the rule against double jeopardy, so understood. The Crown seeks to prove that the applicant killed his parents. It is correct that if proved the consequence must be that the Crown was mistaken in accepting the applicant's plea to the manslaughter of Christopher. However, fundamental to the acceptance of that plea was not a finding that Christopher had killed his parents, but rather that the Crown could not prove that he did not. Proof that the Crown accepted the applicant's plea on a false premise would not relevantly contradict the verdict of manslaughter, although of course it would suggest that the applicant was guilty of Christopher's murder in so far as it indicates that there was evidence to negative provocation.

147 The Canadian decisions of Arp and Grdic suggest that an acquittal is an unequivocal finding of innocence and cannot be treated as a mere concession that the Crown could not make its case. However, this does not appear to be the position in Australia. In R v Darby [1982] HCA 32; (1982) 148 CLR 668 at 677, a majority of the High Court (Gibbs CJ, Aickin, Wilson and Brennan JJ) approved the following observations made by Lord Salmon in Director of Public Prosecutions v Shannon [1975] AC 717 at 764:

"An accused is entitled to be acquitted unless the evidence satisfies the jury beyond reasonable doubt that he is guilty. A verdict of not guilty may mean that the jury is certain that the accused is innocent, or it may mean that, although the evidence arouses considerable suspicion, it is insufficient to convince the jury of the accused's guilt beyond reasonable doubt. The verdict of not guilty is consistent with the jury having taken either view. The only effect of an acquittal, in law, is that the accused can never again be brought before a criminal court and tried for the same offence."

148 We note also the view of the New Zealand Court of Appeal in Degnan, which said the following in relation to the Canadian decisions of Arp and Grdic (at [34]):

"We find it difficult to endorse the approach taken in some of the Canadian cases that a verdict of not guilty is the equivalent of a declaration of innocence. That approach risks elevating perceived theory over the realities of criminal practice. In the vast majority of cases a jury, when returning a verdict of not guilty, cannot be taken as saying affirmatively they are satisfied the accused is innocent; what they are really saying is that they are not satisfied beyond reasonable doubt the accused is guilty."

149 The Court of Appeal's reasoning in Degnan has even greater force where, as here, an acquittal on a charge is the result of a sentencing court's acceptance of a plea on a lesser charge rather than a jury verdict. For in that scenario, no trier of fact has made a determination as to the merits of the Crown case. Neither the sentencing court by acquitting the applicant of murder, nor the Crown by accepting a plea to manslaughter in full discharge of the indictment for murder, can be taken to have affirmed the applicant's innocence of Christopher's murder. It follows that the trial judge did not err by failing to stay the proceedings or direct a verdict of acquittal. Accordingly, the first ground of appeal fails.

150 There remains the question, central to the second ground of appeal, of what affording the applicant the "full benefit" of his acquittal entailed in the context of the trial. As McClellan CJ at CL recognised in the previous appeal, and as was inherent in the reasoning of Spigelman CJ (at [31]), the trial of the applicant for the murder of his parents could not take place without reference to the death of Christopher and the fact that the applicant admitted killing him. That required the trial judge to direct the jury in a manner which made plain that the applicant had admitted killing Christopher but had always denied killing his parents. Having been provided with the statement of agreed facts which was tendered at the trial, the jury were made aware of the history of the matter and the applicant's consistent response to the Crown's allegations.

151 In our opinion the trial judge's directions on these issues were appropriate save in one crucial respect. The trial judge correctly identified that the applicant's plea to manslaughter in 1995 was relevant to establishing the context of the alleged offences, the consistency of the applicant's account of the incident, and the reasons for delay in the laying of charges against the applicant for the murder of his parents. But if the plea to manslaughter was relevant to an assessment of the facts in issue, so too were the consequences that flowed from it, one of which was the fact of the applicant's acquittal of Christopher's murder.

152 Once it became apparent that the acquittal was relevant, the applicant became entitled to the "full benefit" of that acquittal: Garrett at 445 (Barwick CJ). Where evidence of an acquittal is admitted, it is not enough for a trial judge to instruct a jury, without further explanation, that a defendant is to be given the full benefit of his or her acquittal. The jury must be told what it means in the circumstances of the instant case to give the defendant the full benefit of the earlier acquittal: Washer v Western Australia [2007] HCA 48; (2007) 234 CLR 492 at [31] (Gleeson CJ, Heydon and Crennan JJ). In the context of the trial, that required the trial judge to inform the jury that the applicant's acquittal of murder constituted a formal acknowledgement by the sentencing court that the Crown could not, as at April 1995, negative the reasonable possibility that Christopher had killed Mr and Mrs Gilham, and that in doing so he provoked the applicant to kill him. That this was an assumption which underlay the acquittal is apparent from the agreed facts before the sentencing court, which read in part:

"The accused has given an account of events to Police which in their simplest form are a feasible account of what may have happened. The investigation to date has been unable to refute that scenario, despite a thorough scientific analysis and evaluation of the scene and available evidence."

153 The "feasible account" to which the statement of agreed facts refers appears in the previous paragraph and reads as follows:

"The accused informed Police that he responded to a frantic intercom call from his mother after he was wakened from his dwelling, being a converted boatshed some 50-70 metres from the main house on the river line at the rear of the property. Upon entering the house the accused allegedly saw his brother Christopher standing beside their mother, who was lying on the floor, whom the accused presumed was deceased. As the accused approached his brother he (the accused) alleged his brother stated, words to the effect, 'I've killed Mum and dad', and with that saw a lit match in the brother's hand. As the accused drew nearer, his brother placed that match onto their mother's body which ignited immediately. The fire then spread rapidly to the main bedroom where the accused saw the presumed deceased of his father [sic] lying on the floor. With that, the accused stated that he responded and grabbed a knife, which was allegedly on the floor near to where his brother was standing, picked it up and chased his brother down a spiral staircase to a downstairs room where he fatally stabbed his brother a number of times ..."

154 Of course, the weight to be given to the assumptions that underlay the acquittal was entirely a matter for the jury. However, the jury were unable to consider whether to attach any weight to the fact of the acquittal because the trial judge did not inform them of it. More than that, the following direction given by the trial judge denied the relevance of one of the assumptions embodied in that acquittal, namely that the Crown could not disprove the circumstance amounting to provocation:

"Whether the Crown was correct to concede that the accused should be sentenced for manslaughter of his brother is neither here, nor there. They are decisions made by individual persons for reasons best known to that particular person in the circumstance of the facts that that particular person had before him or her at that particular time."

155 The applicant was denied the full benefit of his acquittal of murder. The second ground of appeal must therefore succeed.

156 However, the applicant was not denied the full benefit of his acquittal by reason of the manner in which the Crown Prosecutor conducted the trial. As the findings on Ground 1 have shown, the Crown was entitled to undermine the earlier acquittal solely as an incident of, or step along the way to, proving that the applicant murdered his parents. The Crown could not, and did not, mount a standalone argument that the applicant was criminally responsible for murdering Christopher. The Crown made that insinuation only to the extent necessary to prove that the applicant murdered his parents.

157 We reject grounds one and three of the appeal but uphold ground two.

Ground 4: His Honour erred in permitting into evidence a fire demonstration conducted by the Crown's fire expert, Mr James Munday.

Ground 5: His Honour erred in allowing into evidence opinions drawn from fire demonstrations conducted by the Crown's fire expert, Mr James Munday.

Ground 17: The applicant seeks to rely upon new evidence that the fire is likely to have occurred in the manner described by the applicant and that the flames, far from being of low height in the manner described by Mr Munday, were such as to prevent the applicant from any attempt to assist his mother.

158 The Crown called Mr Munday, who was the principal of a company which specialised in investigating fires.

159 He gave evidence of a series of experiments which he carried out, and filmed, the purpose of which was to assess the potential rate of progress of the fire. The Crown also relied upon the DVD of the fire experiments in an attempt to contradict the applicant's version that he was unable to assist his parents when he first came upon them.

160 The Crown Prosecutor argued in her final address to the jury, in rather forceful terms, that given the low height of the fire at the time it commenced, and its slow rate of spread, at least initially, the jury would not accept the applicant's evidence that he did not, or could not, go to his parents' assistance, and instead chased his brother down the stairs.

161 When dealing with the evidence of the fire, the Crown Prosecutor said to the jury:

"There were many things which he could have done, and ... I suggest to you if it was your parents in danger of being burnt, you'd put it out with your bare feet, you'd do something. He expects you to accept ... that an innocent man would do nothing to put out that fire. It doesn't matter if it's 30 centimetres high. Well that just doesn't make sense, it doesn't add up ... that's what I have to say really about the fire."

162 There was no challenge to the expertise of Mr Munday to undertake the series of experiments, or to express a variety of opinions about the behaviour of fire. However, defence counsel objected to the DVD of the experiments being put before the jury. A voir dire examination took place.

163 In his evidence, and before the voir dire took place, Mr Munday was asked, without objection, to describe in terms of seconds and minutes the progress of any fire which would have developed upon application of an accelerant. He said:

"If there were mineral turpentine present, then the fire would develop relatively slowly at first and would gradually accelerate as the fire became larger. But certainly for the first 15, 20 seconds, maybe up to 30 seconds, the fire would remain relatively small and would gradually increase in size and then would increase more rapidly in size as the fire size grew."

164 Further, in the presence of the jury, when asked to give a further description of the fire where the accelerant being used was mineral turpentine, Mr Munday said:

"In the initial stages of igniting mineral turpentine on an absorbent material, the flames are likely to remain under 30cm high for the first 10 seconds.

165 From the voir dire examination, it became apparent that there were a number of variables, any one of which could affect the size and rate of spread of a fire, particularly in the initial interval of between 10 and 30 seconds. They include:

(a) the volume or quantity of accelerant used. There was no evidence at all which established the volume, or quantity of accelerant. The Crown's submissions to the jury do not suggest any particular volume or quantity;

(b) the floor area across which the accelerant had been spread, and the pattern of the spread. Mr Munday said that from the evidence he had seen, there was some accelerant in at least three discrete locations, but he was unable to say which was the ignition point of the fire. In particular, it is of importance to know whether there was a trail of accelerant between Mr and Mrs Gilham. There was no evidence that such a trail existed;

(c) whether the accelerant had been spread onto the bodies of either Mr or Mrs Gilham, and if so, whether it had soaked into any of the clothing they were wearing;

(d) the nature, extent and material which made up the clothing which Mr and Mrs Gilham were wearing at the time of the fire, and hence the extent of its flammability, and the locations where the fire would have burnt the bodies; and

(e) whether the fire was ignited on a horizontal or vertical surface, and the flammability of the particular material in which the fire was started.

Experiments

166 The experiments were attempts by Mr Munday to simulate the behaviour of a fire in a variety of circumstances. As Mr Munday put it on the voir dire:

"What I tried to do was ... to formulate three hypotheses which would potentially explain the locations in which flammable liquid was identified, and also reflect the evidence given by other investigators."

167 The purpose of the experiments, he said, was to enable him to express an opinion about the speed at which, and the form in which, the fire spread.

168 Mr Munday accepted that the experiments did not:

a) replicate the carpet or underlay on the floor of the Gilham house because the nature and type of the carpet and underlay was unknown. In one experiment he used a nylon carpet and a polypropolene carpet in the others;

b) allow for the passage of oxygen under the carpet through the floorboards as in the Gilham house, because he conducted the experiments on a concrete floor;

c) accurately, or substantially, replicate the features of a body. In one experiment, PVC pipes were used to represent the legs of Mrs Gilham. In another experiment, pieces of timber were used. Nor did the experiments accurately replicate the clothing on the bodies. Mr Munday agreed that if there was accelerant-soaked clothing on the body, then the fire progression around the body would be far more rapid than if there were not;

d) replicate the temperature in the room at the time of the fire, which was unknown, whereas the ambient temperature in the cell where the experiments were conducted was 11.1 degrees. Mr Munday agreed that if the temperature was higher in the room when the fire occurred than in the experimental cell, then there would be a difference in the fire behaviour, although he thought that this was not significant;

e) establish where the ignition point of the fire was, that is, he could not say where the fire started. Yet in the experiments, the fire is started in the immediate vicinity of the simulated body position; and

f) establish whether the fire was initiated on a vertical or horizontal surface which would impact on the rate of growth of the fire. He said that the flames would be bigger and the rate of spread of the fire greater if it was on a vertical surface initially. He did not attempt to replicate any vertical surface upon which a fire was initiated.

169 In dealing with the objection to the DVD being admitted, the trial judge ultimately held:

"What I am minded to do is to allow this jury to have some idea of how the lighting of the substance: it's effectively up to the 10 to 15 [seconds] buildup."

170 Ultimately, having allowed the DVDs to be shown to the jury after an agreed editing process, the trial judge gave the following direction:

"But you still have to be very careful of experiments as they are called, and the reason is because it has been made apparent to you already that it was impossible to replicate the circumstances of this particular night, and you have heard that one of the problems is that nobody knows what the carpet was made of. As I understand it, nobody knows what the underlay of the carpet was made of. Nobody knows other matters that would possibly influence the way that the fire would progress. So what's been done here is an experiment, but a limited one, and it can only have limited assistance to you.

It seems to me that it is important to see how the fire progresses at least in the initial stage of the fire, and that is what we are talking about here, the initial growth of the fire until it starts on this exponential growth. But you should understand that it is not necessarily what would have happened on this particular night and certainly not necessarily what would have occurred to the accused or what the accused would have seen on this particular night and that is because we can't replicate all of the circumstances in which the fire took place.

Further, there is another slight problem with the experiments ... it is simply the basis of the material upon which [Mr Munday] was acting - and that is, there is no attempt to replicate in any way the female body, and particularly the clothing on the female body, and it may be a very important factor as to whether this fire was commenced on the carpet or on the body ...

So just understand this is a very limited experiment, it gives you limited assistance. Don't put too much weight upon it as being an exact replica or even a very close replica of what had occurred on this particular night, it [is] simply to give you some understanding of how a fire progresses in these circumstances, and as I understand, the fire, as seen, doesn't change very much from what you've already been told about the spread of the fire, but rather than being told about it from various witnesses, it's perhaps better that you can see an example of it, as long as you understand the limited basis upon which you can use that evidence at the end of the day."

171 The applicant submitted that the content of the DVDs was irrelevant or alternatively, by reason of the application of s 137 of the Evidence Act 1995, that they ought not to have been admitted.

172 Section 137 requires a court to weigh the probative value of the evidence against the danger of unfair prejudice to an accused.

173 The judge's directions based upon the evidence given by Mr Munday suggest that the experiments had very little, if any, probative value, in the absence of a sufficient correlation between what the evidence proved was likely to have occurred, and the experiments shown in the videos. However, as we have earlier noted, there was either insufficient evidence to prove a number of the variables upon which the experiments were based, or alternatively, the variables which were taken into account produced results which may or may not have replicated what actually occurred.

174 In those circumstances, the jury were left to do the best they could with a range of experiments which may or may not have coincided with the events that occurred.

175 For these reasons, we conclude that these experiments had very little, if any, probative value.

176 But the prejudicial effect of these experiments was, in our assessment, very high. The Crown sought to demonstrate that the jury should reject the applicant's account that when he arrived at the scene, it was, in effect, not open for him to have done anything about the fire. After all, the prosecutor submitted to the jury, anyone would have taken steps, when confronted by a fire of low height and slow spread, to have put out the fire before doing anything else.

177 In support of that submission, the prosecutor relied on the visual impact of the videos. But they did not paint an accurate picture, and probably painted a most inaccurate picture supporting the Crown's submission.

178 In those circumstances, where there was a real risk that the jury were being presented with, and were being invited to act upon, an inaccurate picture of what occurred, there was a strong prejudicial effect to the accused.

179 The balance of these matters weighed heavily on the side of the prejudicial effect being significant and the probative value being little. The videos ought to have been excluded by the application of the provisions of s 137 of the Evidence Act.

180 The next issue is whether the admission of this evidence caused a miscarriage of justice.

181 In his summing up, the trial judge gave directions to the jury in these terms:

"In this case as the matter has ultimately been left to you, much of the problems or the disputes of the issues in relation to the fire evidence has disappeared. I will talk to you later about the relevance of that evidence, but what might have been complex, what might have been difficult evidence where it might have been necessary for you to review in detail the evidence of the witnesses in relation to the fire and the spread of fire and perhaps even try to determine which witness you particularly thought was persuasive when it came to the fire evidence, all of that seems to me, with great respect, to have disappeared. At the end of the day, so far as it seems to me, the fire evidence can be reduced to a very, very confined area. So much of the evidence on the fire can be disregarded."

182 His Honour did not return, in summing up, to discussing any of the evidence about the fire, its size or rate of spread. He did mention the evidence of Mr Munday, but only in a general way and in connection with assisting the jury with the usual instructions with respect to expert evidence. Nothing which he there said dealt with the detail of the fire evidence, the variables which may have contributed to the size of the fire, the reliability of the evidence, or its probative value.

183 In all of the circumstances, although the evidence ought not to have been admitted, we cannot be satisfied that by reason of its admission there was a miscarriage of justice. This is particularly so having regard to the fact that the evidence seems to have ultimately assumed a minor role in the addresses of counsel and the summing up of the judge.

Ground 5: Admissibility of Mr Munday's opinion evidence

184 Ground 5 deals with whether the opinions of Mr Munday on the fire, which he drew in part from the experiments he conducted and in part from his broader expertise, were admissible.

185 In dealing with Ground 4, we have found that the evidence of the experiments which was tendered by the Crown ought not to have been admitted. But it does not necessarily follow that the whole of the expert opinion of Mr Munday is inadmissible.

186 An expert in formulating an opinion is entitled to draw upon a variety of sources, which may vary in their importance and the weight the expert gives them. For example, an expert was entitled at common law to give evidence of hearsay matters which went to demonstrate their expertise: Dasreef Pty Ltd v Hawchar [2011] HCA 21; (2011) 243 CLR 588 at [69] (Heydon J). However, there is no doubt that where, in order to give an opinion in a particular case, an expert relies upon assumptions of fact as the basis of the opinion, it is necessary to prove, in admissible form, those facts or facts which are sufficiently similar, in order to render the opinion of value: Hawchar at [66] (Heydon J); Ramsay v Watson [1961] HCA 65; (1961)108 CLR 642 at 648-649; Paric v John Holland (Constructions) Pty Ltd [1985] HCA 58; (1985) 59 ALJR 844 at 846; R v Perry (1990) 49 A Crim R 243 at 249; R v Hilder (1997) 97 A Crim R 70 at 79.

187 Mr Munday's expertise to give the opinion evidence which he gave was not challenged. He was adequately qualified. The general opinions which he expressed were within his area of expertise. They were relevant to the issues in the Crown case.

188 In so far as he based his opinion on facts about fire behaviour which he observed in the experiments and which were shown on the videos, our ruling that the videos were inadmissible would have the effect that any opinion based on that inadmissible material would itself be inadmissible. However, in so far as his opinion was based upon other assumed facts which were sufficiently proved in evidence, then it was admissible.

189 No attempt was made at trial when objection was taken to Mr Munday's evidence to separate out that part of his opinion based solely on the videos of the experiments and to seek to have the balance admitted. Nor was a similar exercise undertaken on the hearing of the appeal. It is not a matter for this Court to undertake the exercise unaided by the parties.

190 In those circumstances, we will proceed to consider this ground, as the parties' submissions did, upon the basis that Mr Munday's opinion was based upon the experiments which he conducted, the primary evidence of which was captured by the videos, and was to be proved by the tender of those videos.

191 On this basis, the opinion evidence of Mr Munday depended upon the admissibility of the videos of the experiments. In light of our ruling that the videos were inadmissible, we find that Mr Munday's opinion was similarly inadmissible.

192 However, for the same reasons expressed under the preceding ground, in light of the very limited use ultimately made of the opinion evidence, we are not satisfied that any miscarriage of justice has occurred.

193 This ground does not succeed.

Ground 17: New evidence about the behaviour of the fire

194 The applicant sought to rely upon new evidence on the appeal from Dr Anthony Green on the topic of fire behaviour.

195 We have already considered whether, at the trial, the DVDs of the experiments conducted by Mr Munday ought to have been admitted, and we concluded that they should not have been.

196 The evidence sought to be adduced on the appeal consisted of experiments undertaken to "more accurately" replicate the experiments that Mr Munday undertook.

197 In light of the earlier discussion, and applying the principles set out at [545], it is only necessary to say that in respect of this ground of appeal:

(a) the material advanced by Dr Green was available at the time of the trial, and if it was to be relied upon, ought to have been called at that time;

(b) the material does not amount to fresh or new evidence of a kind that permits its admission on appeal; and

(c) the material principally deals with matters which ought not to have been admitted at the trial in any event.

198 In those circumstances, we do not uphold the appeal on this ground.

Ground 6: The trial judge erred by allowing evidence from Dr Culliford and Dr Cala regarding their expectations of how much blood should have been on the knife [given its use in the three stabbings]

Ground 7: The trial judge erred by allowing evidence from Dr Culliford regarding her expectation of how much blood the applicant was likely to have on him [from killing Christopher]

The significance of the evidence concerning both issues in the Crown case

199 The Crown relied upon evidence that there was no blood (or no visible blood) on the knife, in combination with the fact there was too little blood on the applicant, as two of the fourteen circumstances proving the applicant's guilt. The Crown submitted that the absence of blood could only be explained by the applicant having washed himself to remove the blood of his parents (and to remove evidence of his fingerprints) before he placed the knife by the body of his brother to implicate him in their murder.

The evidence of blood on the knife

200 The evidence of what was visible (or not visible) on the blade, or at the hilt or on the handle of the knife, came from two sources. The first, from those officers who attended the crime scene and saw the knife, and secondly, from a number of expert witnesses who were invited to consider that evidence, together with a photograph of the knife in situ lying next to the body of Christopher and a photocopy of the knife made during the post-mortem, and to offer an opinion as to whether the appearance of the knife corresponded with their expectations of the amount of blood that would likely be in residue on the knife after the infliction of more than 60 stab wounds.

201 Senior Constable Sands, the first police officer who attended the scene at about 4.50am, did not recall whether there was any blood on the knife when he forced entry to the downstairs area where Christopher was found. He said Christopher was wearing a shave coat which was wet or moist. Another police officer who attended at about that time said she saw spots of blood in the vicinity of Christopher's body, but there did not appear to be any blood on the knife. She did not remember any water being in the vicinity of the body. Her evidence was in conflict with the evidence of Sgt Ahern, the police officer originally in charge of the investigation.

202 His evidence was that the amount of water that was used to extinguish the fire would be likely to have affected the integrity of the crime scene, including the potential for any visible blood on the knife to have been removed given the physical proximity of Christopher's body to the knife and the water on the floor around his body. He said that by the time he arrived water was dripping onto Christopher's body and onto his shave coat. He said that he was not sure whether he saw water falling onto the knife although he said it fell close to where the knife was lying. He said he did not look carefully at the knife although he believed there would have been some water on the knife at that time.

203 Detective Rowan, a crime scene officer, gave evidence at the inquest. He was not called at trial. His evidence was read. He made notes at the scene including a description of the knife and a rough sketch of where it was located relative to Christopher's body. He made no note of visible blood on the knife. He said the knife appeared to be clean, which could have been due to the amount of water that had been cascading down onto the scene. He also gave evidence that once the tarpaulin that had been placed over Christopher's body was removed a blood stain on the shave coat had apparently run because of water dripping onto the body before it was covered. He agreed that if there had been sufficient water to wash blood off the knife then it should also have washed blood off Christopher's body. His evidence was that there was still a good deal of blood on the body.

204 Detective Senior Sergeant Horn, another crime scene officer, had a cursory look at the knife and did not see any blood on it. When Christopher's body was moved to allow Dr Lawrence to take his rectal temperature he was unable to say what happened to the knife. He gave evidence that he looked at the knife in situ but did not touch it. He did not notice a great deal of blood on it. He also gave evidence that he picked up the knife to examine the body and put it somewhere on the floor.

205 The knife was eventually collected by police and wrapped in a rag before being submitted for fingerprint analysis. It had not at that time been analysed to determine whether there was blood on it. The following day the knife was produced by police during a walkthrough with the applicant, with a cloth around the handle. It was identified by the applicant. It was then returned to Sutherland Police Station and kept in Sgt Ahern's office. It was neither sealed as an exhibit nor entered into the exhibits register.

206 A crime scene officer conveyed the knife in a paper bag for fingerprinting and then took it to Dr Lawrence. The knife was used by Dr Lawrence during the post-mortem, where a black and white photocopy of one side of the knife was made.

207 The knife was not subjected to a formal examination to detect the presence of blood. The fingerprint examiner treated the knife with superglue in a sealed container and stained it under a poly light. No print or partial print was recovered either from the blade or from the handle. He agreed it was very common to find no evidence of a fingerprint or palm print on a knife and that the chances of recovering fingerprints from a wet knife were much lower than from a dry knife.

208 He also gave evidence that there were no signs of blood on the knife. He said that any foreign stain on the blade or the handle would have shown up under the poly light in which case he would have referred the knife back to the physical evidence officer before proceeding to fume it with superglue. He said that while the superglue process would have destroyed any potential for grouping if blood were present, it would have shown up any blood as a slight stain, which would not have been destroyed in the examination, although its concentration would have been diminished. In cross-examination he agreed that if there had been blood visible on the knife he would have written "a bloody blade" in his notes. He made no note of that kind. He also agreed that he expected that any blood stain (or possible blood stain) would have been tested before he received the knife and that the particular task that he was asked to perform was not concerned with whether there was any blood on the knife.

The evidence of blood on the applicant

209 Evidence regarding where blood was seen on the body of the applicant came from those who saw him at the scene, including neighbours and ambulance officers, and Dr Jennings, who examined him after he was taken into custody.

210 When the applicant arrived at the Warners' home he was wearing only boxer shorts. Mr and Mrs Warner agreed that it had been raining, that there was mist in the air, and that the ground was very damp. When he was sitting on the lounge Mr Warner said that he touched the applicant's hair, which was shiny but not wet. Mr and Mrs Warner both said he was cold to the touch but not wet. He was given a piece of clothing to wear but the applicant did not put it on; he just wrapped it around his hands. A blanket was put around him.

211 Mr Warner saw blood on the applicant's feet. He could not say whether it was on one foot or both. He also saw blood on the little toe of his right foot and on the shin of his left leg, and blood in the quick of his fingernail. Mrs Warner gave evidence that it appeared to her as if there had been an attempt by the applicant to wash the blood off his feet and hands, as the staining on his lower leg was a pale pink, not a bright red. He was in the Warner's house for about ten minutes. He left the blanket behind. It was not seized by police.

212 The applicant was outside the house when he was spoken to by fire brigade officers before the arrival of police and before he was taken into the ambulance. By this stage he had been given another blanket, which was wrapped around his shoulders.

213 An ambulance officer noticed a "spot of blood" on the top of one of the applicant's feet. Another officer observed a small dab of blood on his right foot.

214 A crime scene officer at Sutherland Police Station noted:

"I saw and photographed what appeared [sic] blood matter on his (1) inner right foot, swab A then (2) left shin swab B

(3) right hand swab C (fingernail)"

215 The same crime scene officer gave evidence at the inquest, which was read to the jury. He said that there appeared to be blood on the inside of the applicant's right foot, in one of his fingernails, and on the inside of his shin. This was the only blood he noticed after examining the applicant's entire body down to his boxer shorts.

216 Dr Jennings examined the applicant at 12.05pm after he had participated in an ERISP. She noted a small amount of dried blood on the nail of the applicant's left index finger and a small amount of dried blood on the palm side of his right middle and index fingers overlying the middle bone. She also detected a small amount of dried blood on the nail bed around the edge of his right index finger and a small amount of dried blood on the arch of his right foot, towards the front of the foot.

Evidence from experts concerning blood on the knife not the subject of challenge

217 Sgt Ahern gave evidence that he had attended several multiple stabbings over the course of his career and had never seen a knife used in a stabbing without visible blood. There was no objection to this evidence.

218 There was no objection to Sgt Reynolds, a crime scene officer, giving evidence of his experience as to the presence or absence of blood on a knife used in an assault or giving evidence that he had seen knives used in stabbing incidents, including of a person stabbed multiple times. Although he had no exposure to stab wounds inflicted on multiple victims in the one incident, he recalled an occasion where a knife had no visible blood on it after the infliction of a single stab wound. He described what crime scene officers call a "fat effect", which might explain why a knife used in a stabbing might be free of blood, in that the skin, being quite elastic, adheres to the knife blade and essentially cleans the blade as the knife is withdrawn from the body. He said clothing can produce the same effect.

219 He gave evidence that he was surprised that there was no visible blood on the knife considering that it had been used to inflict over 60 stab wounds, irrespective of whether the deceased were clothed or not. He concluded that something had happened to the knife blade after the stabbings to remove the blood. When asked about the likelihood of water contaminating the scene he did not think that was an adequate explanation. He said:

"... the rate that blood dries on a surface is a function of many variables, ambient temperature, wind speed, but it's always a function of the nature of the surface, porous, non-porous, rough, smooth, for example, and blood is very adhesive and persistent. If the blade has a reasonable amount of blood on it and 16 minutes later it was necessary to remove the blade (sic), my thoughts would be that it would have been washed in the sink, not just have water dribble on it."

220 There was no objection to Mr Raymond, a forensic expert, giving evidence of his expectation that there would be some blood residue on the knife used to inflict multiple stab wounds and the likely mechanism by which the blood staining on the applicant's right foot was transferred. He gave evidence that such photographs as were available from the crime scene were not very helpful in determining whether, and if so how much, blood was on the blade, the hilt or the handle. As to the likelihood of there being blood residue on a knife, he said that blood would run off a very smooth surface relatively easily unless the knife was very rusty, in which case residual blood would be a very thin smear. He said that in his experience the amount of blood typically found on a smooth knife was very little "but it is there". The knife in question did not appear to him to look particularly smooth and he was surprised not to see the appearance of blood. He accepted that may have been due to poor photographing of the knife. He said that oblique lighting will enable blood smeared on a blade to be viewed which would not be evident to the naked eye. He noted that it was very common to see blood on the handle of a knife because of the transfer of blood from the hand of the assailant. In this case the knife had a black handle so that blood would be difficult to see. When he was asked to offer an opinion as to whether a darkened colour at the serrated edge of the blade of the knife in the photograph was blood, he could not say one way or the other. He agreed that the mark was in a position where you might expect to see blood. In commenting upon the effect of water dribbling onto the knife, he said that blood not only clots but dries. He said a relatively thin smear would dry very quickly so that water that fell onto the smear would change the stain, probably making it difficult to see, but it would not eradicate it entirely unless the water had been running onto the blade for a period of time.

The applicant's argument concerning Grounds 6 and 7 in summary

221 Dr Culliford's evidence at the first trial was objected to on the basis that neither her training, study nor experience as a forensic physician qualified her to give evidence of her expectation that blood would be visible on a knife used to inflict multiple penetrating stab wounds, whether the blood consisted of a smearing of blood on the handle or on the blade, or was deposited at or near the hilt of the blade. It was also submitted that she was not qualified to give evidence that she would have expected more of Christopher's blood to be transferred to the applicant in the course of the stabbing for the same reason. Even if her evidence were admissible as opinion evidence, the applicant submitted, it ought to have been rejected in the exercise of discretion under s 137 of the Evidence Act.

222 Objection was also taken to Dr Cala giving evidence as to his expectations concerning blood on the knife and blood on the applicant on the same basis. However, since he was not asked to offer an opinion about either of these matters in the second trial, it is unnecessary to determine whether or not the trial judge's ruling on the voir dire in the first trial permitting Dr Cala to give that evidence was in error. Dr Cala was asked by the Crown whether he expected that the wounds inflicted on each of the deceased would have bled profusely, differentiating between those wounds inflicted through clothing and those not covered by clothing. This evidence was not objected to.

An additional basis for challenging the admissibility of Dr Culliford's evidence concerning the blood on the knife

223 The trial judge was satisfied that Dr Culliford had specialised knowledge, based upon her training and experience as a forensic physician, to offer an opinion as to the likelihood of blood being in residue on the knife, and that firm directions would address any possibility of prejudice to the applicant by reason of the jury giving undue weight to the evidence. This was a ruling delivered in the course of the voir dire in the first trial which also inquired into the question of the admissibility of her evidence concerning similarity in the stab wounds. At that time the ruling was based upon the assumption that there would be evidence capable of establishing that there was no blood on the knife so as to satisfy the requirement that her opinion was relevant (see Dasreef at [30]-[32]).

224 Before Dr Culliford was called at the second trial, objection was taken to her giving opinion evidence on an additional ground. Defence counsel submitted that it had been shown by the evidence led at the first trial that there was no evidence, or no sufficient evidence, of the foundational factual premise upon which it was agreed that Dr Culliford's opinion concerning blood on the knife was based, namely that there was no blood (or visible blood) on it, and for that reason her opinion was irrelevant. It was not clear whether the objection was because her evidence failed to satisfy the preconditions to admissibility in Makita (Australia) Pty Ltd v Sprowles [2001] NSWCA 305; (2001) 52 NSWLR 705 at [85], because of a failure to prove by admissible evidence the fact that there was no blood on the knife, or because her opinion failed to satisfy the test of relevance in s 55 of the Evidence Act, because the evidence about blood on the knife diminished the relevance of her opinion to the extent that it should not be admitted. (No further objection was taken to the admissibility of her evidence concerning the amount of blood she expected would be on the applicant. The applicant accepted that the ruling at the first trial applied to the second trial on that issue.)

225 Counsel reminded the trial judge of the warnings he had given to the jury in the summing up at the first trial concerning this aspect of the Crown case, which were given because of the state of the evidence as it then stood, in particular: the fact that the knife had been handled by police at the crime scene without ensuring that any blood that might have been on the knife was preserved; the fact that the knife was not tested for blood thereafter; the potential for water at the crime scene to have removed residual blood on the knife; and the fact that the knife was no longer available for forensic testing.

226 His Honour was satisfied that there was a sufficient factual foundation for the reception of Dr Culliford's evidence despite there being room for a legitimate attack on the reliability of the evidence upon which her opinion was based. He was satisfied that there was positive evidence, in particular from the fingerprint expert, that there was nothing to indicate the presence of any blood on the knife which, taken with the other evidence, was sufficient for the jury to consider Dr Culliford's evidence as part of the Crown's circumstantial case. He made it clear he would direct the jury accordingly, and even withdraw the issue from the jury's consideration if the evidence was further undermined.

227 Before Dr Culliford was called by the Crown, the trial judge warned the jury of the potential for some aspects of her evidence concerning blood on the knife to be of little weight. He said:

"The issue this witness is going to address, as I understand it, are two issues: Firstly, the absence or otherwise of blood on the knife. You know, of course, that there is an issue as to inspection and observation of the knife and whether anybody does know whether it had observable blood on it or not, and the possibility of the effect of water.

The second issue that this witness is going to give evidence about as an expert is about the blood that would be likely to be on a person who committed an act of violence with stabbing of another person. Again, it will depend upon the particular facts of this matter, and it is for you to decide what you make of this expert's opinion.

Now, it may well be the case that, under further consideration, I will give you directions at the end of the day as to what weight you can give the expert opinion evidence in light of the facts before you. That remains to be seen as the evidence pans out in this matter and on hearing counsel's addresses to you. But just at the moment, don't put too much weight on expert evidence even though it can sound very dramatic from the witness box. Just accept at the end of the day you are going to have to assess the expert's opinion based upon the facts that you will find or are unable to find, and there are two major issues here, as I understand it: One, the presence or otherwise of blood on the knife, and two, the question of the amount of blood on the accused. They are two matters which are very much open to debate and open to a conflict of evidence.

Just bear all of that in mind. We will hear expert evidence about blood splatter or blood resulting from various incidences such as stabbing. Keep in mind at the end of the day the expert is giving opinions and those opinions are based upon facts that you have to find and not the expert."

228 It was not suggested that the evidence at the end of the Crown case was in any different state from when counsel took the objection prior to Dr Culliford being called. The trial judge did not withdraw Dr Culliford's evidence concerning the absence of blood on the knife from the jury's consideration as he suggested he might. There was, however, no formal application that he do so. To the extent that the failure to do so is relied upon in support of Ground 6, which is not clear from the applicant's submissions, leave under rule 4 would be required.

229 It is not necessary to consider the question of leave as we are not persuaded that error has been demonstrated in admitting evidence of Dr Culliford's expectation that there would be blood in residue on the knife, whether on grounds of relevance or because her evidence failed to meet the criteria for admissibility under s 79 of the Evidence Act.

230 The trial judge did not distinguish between the sufficiency of her claim to specialised knowledge as a forensic physician to ground her opinion about blood on the knife and her opinion about blood on the applicant. He was satisfied that she was qualified to express an opinion about both issues. The applicant does not submit that she may have had specialised knowledge in one area and not the other.

Dr Culliford's evidence on the voir dire in the first trial

231 Dr Culliford based her opinion concerning the transfer of blood from Christopher to the applicant (and the blood from one or more of the three deceased to the knife) upon the knowledge of fundamental principles she acquired in forensic training undertaken as part of her Masters Degree and her training with police, as well as the application of that training in the course of her work over 20 years. (Dr Culliford's qualifications are set out in full at [278].)

232 She said one fundamental or guiding principle was "Locard's principle", which assumes that if the perpetrator of a crime has entered a crime scene and has come into contact with a victim, there will inevitably be an exchange of material of some kind from person to person. She also relied upon her examination of three perpetrators of multiple stabbings and her observation that in two of the three cases there was blood on the hands of the perpetrator, one with blood on his clothing and another with a lot of blood on his body and on his hands.

233 She also gave evidence of a number of factors she had encountered in the course of her experience as a forensic physician affecting the transfer of blood to an assailant in the course of a stabbing. These were:

(a) the nature of the surface in contact with the blood. Skin is a rough surface such that blood (which is a sticky material) is likely to remain on the hand or be embedded in the nails of the assailant unless it is wiped or washed off;

(b) the amount of blood available to be transferred - the greater the number and type of wounds, the greater the amount of available blood;

(c) the area of contact between the victim and perpetrator - in particular, clothing worn by the victim may obstruct the transfer of blood;

(d) the length of time the contact occurs;

(e) the number of contacts - a single stab wound is less likely to transfer blood than multiple contacts; and

(f) force of contact - the greater the force of contact (and depth of the wound) the more likely it is that blood will emerge from the wound.

234 She accepted that there is no necessary correlation between the amount of blood dispersed during a violent event involving a knife and how the blood might be distributed, including the amount of blood likely to be found on the perpetrator. She maintained her view, however, that in the circumstances of this case, where she understood the applicant's hand had come into repeated contact with Christopher's naked chest (or near his naked chest) and/or his shave coat, through which the knife passed to inflict some of the wounds, she would have expected a transfer of Christopher's blood onto the applicant's hand.

235 The trial judge was satisfied that Dr Culliford's expectation that blood would be transferred from Christopher to the applicant as a result of the injuries he inflicted on his brother was a matter within her expertise, but that she should not be permitted to give evidence about blood spatter after her concession that she did not have specialised knowledge in that field. His Honour was satisfied that she had applied scientific principle, acquired in the course of her training and experience, and in the context of the various factors that might affect the application of that principle in this case, and in that way she was qualified to offer her opinion about the blood that she expected would be visible on the knife, and in greater quantities on the applicant as a result. His Honour went on to say:

"I have no doubt that Locard's principle is a matter of scientific principle. It is a matter, according to the witness, that she has studied and used in the course of her training and her experience. It seems to me that she is entitled to seek to apply that principle and affect the factors of that principle in a particular case and form a view that she might have expected as a result of applying the factors and the particular circumstances in relation to this particular case.

She explained in the voir dire how she came to the view she has. In my view, she could say no more than she expected more blood upon the accused by way of transfer to him by the stabbing of his brother. Again, if I am required to do so, I am quite happy to give such warnings or directions to the jury after this evidence is given in order that the jury do not misuse it in an inappropriate way. Again, it seems to me it is merely one of the circumstances which does not prove that the accused did either wash his hands or in some way seek to eradicate blood from him, it is merely one of the factors that goes into the general circumstances to be considered by the accused [sic], that so far as this witness is concerned, that she would have expected in the circumstances of this matter for there to be more blood upon the accused than was actually on his person.

...

Other than that, it seems to me that this evidence is relevant in that the witness is capable of giving evidence as a result of her expertise, that it has some weight and that its prejudice is not such that I should exclude it in light of the directions that I believe I can give in order to ensure that the jury do not give it more weight than what it deserves."

236 We can see no error in his Honour's findings that Dr Culliford was qualified to give her opinion concerning this question, or that the evidence was of sufficient probative value (as one of a number of pieces of circumstantial evidence) to guard against the possibility that the jury might give undue weight to the evidence.

237 Her evidence was thoroughly tested in cross-examination. In the summing up the trial judge identified the points of departure between Dr Culliford's evidence and that of Mr Raymond and Sgt Reynolds, the two experts qualified to give blood spatter evidence. His Honour also made specific and repeated reference to the multitude of variables that can impact on the amount of blood that might be spilt or exuded from the body of a victim of a stabbing incident and transferred to a perpetrator. We are satisfied that by identifying these issues, and the conflict in the evidence between the expert evidence called by the Crown, there was adequate insurance against the risk that the jury might misuse the evidence in the sense contemplated by s 137 of the Evidence Act.

238 The trial judge also cautioned the jury against drawing any inference or conclusion that the applicant washed the knife based solely upon the fact that there was no visible blood on it, leaving it to the jury by other evidence in the Crown case to consider whether they were satisfied that the knife was washed. We note that in so doing he did not refer to Dr Culliford's evidence but to the evidence of the other experts to which no objection was taken:

"Let me say this... in relation to the visible blood on the knife. There are two problems in relation to this evidence that you would understand. The first is the lack of any testing or any scientific observation or examination of the knife. There is some evidence from which you may have been able to infer that there was no visible blood on that knife, and that simply comes from the evidence that people saw the knife and say they did not see any blood on it.

How reliable that is, is a matter for you, but you may think it is not very reliable in the circumstances of this particular case. Although I think there is some evidence, particularly from Sergeant Reynolds, that, all other things being equal, that he would have expected there to have been visible blood seen on the knife. He thought that some post-stabbing incident had occurred that removed any blood from that knife, if in fact there was no visible blood on that knife.

That brings me to the second point, and that is there may be a reason for there being no blood on that knife other than that the accused washed it, and that of course is the evidence of the water coming into the scene and falling into the vicinity of Chris and the knife at the time the fire was being extinguished. And, as I understand the scientific evidence, particularly from Sergeant Reynolds and I think Professor Raymond, it is not impossible that the falling of water during the course of putting out the fire did wipe any visible blood off the knife.

Now it may mean that there are still traces of blood on the knife that could have been shown up by the investigation, by proper examination and by proper testing, but as I understand, the Crown will correct me if I am wrong, Sergeant Reynolds accepted that one of the post-stabbing incidents that may have resulted in the lack of visible blood on that knife, contrary to his expectation, was the water coming down from the ceiling and impacting upon the knife and destroying any remnants of blood upon it.

In that regard then you should be cautious about drawing any inference or conclusion based upon the fact that there was no visible blood on that knife or that there was no visible blood on that knife because the accused washed it. You may be able to make that inference from other factors, for example, if you were able to infer as the Crown said you would, that the accused washed himself, that may support a suggestion that he also washed the knife. You might think that that would be a matter of commonsense, but you cannot reason alone safely that simply because of the evidence about the lack of visible blood on that knife, that the accused must have washed it. You could not, in my view, come to a view even on a possibility that that was so. Again, that is an expression of opinion. It seems to me, however, to be one to accord with the evidence and it seems to me, with respect, that it would be irrational and illogical to conclude otherwise that the knife was probably washed by the accused simply because it had no visible blood on it in the circumstances of this particular evidence."

239 We are not persuaded that Grounds 6 and 7 are made out.

Ground 8: The trial judge erred by allowing evidence from Dr Culliford, Dr Cala and Dr Lawrence regarding the similarity of the pattern of stab wounds.

Ground 18: The failure of the prosecutor to call Professor Cordner after the admission of the evidence of Dr Culliford, Dr Cala and Dr Lawrence regarding the similarity of the pattern of stab wounds, when viewed against the conduct of the trial taken as a whole, gave rise to a miscarriage of justice.

240 Ground 8 challenges the admissibility of the evidence of Dr Culliford, Dr Cala and Dr Lawrence as to what they each claim was a discernible similarity in the grouping or pattern of wounds in all three deceased. This evidence was relied upon by the Crown to prove that the applicant had killed all three deceased and was therefore guilty of the murders of his parents. The applicant submitted that their evidence failed to meet the prerequisites for admission as opinion evidence in s 79 of the Evidence Act. The applicant also submitted that the evidence of all three experts ought to have been rejected in the exercise of the discretion under s 137 of the Evidence Act.

241 Ground 18 of the appeal concerns whether the Crown should have called Professor Cordner to give evidence on the question of similarity, and whether the failure to do so was productive of a miscarriage of justice. This ground of appeal became inextricably linked to the question of admissibility raised in Ground 8.

242 Over the Crown's objection we received Professor Cordner's report of December 2010, prepared at the request of the applicant's solicitors for the purposes of the appeal. In the report he is not only critical of opinions expressed by the Crown's experts, but identifies aspects of their evidence which he argues are both factually wrong and scientifically unsound. He also expressed the view that the Crown Prosecutor's arguments based upon that evidence were unfounded and misleading.

243 Professor Cordner was called by the defence on the voir dire convened in the course of the first trial to consider the admissibility of the evidence of Drs Cala and Culliford. The applicant submitted that the views Professor Cordner expressed at that time were the same views he expressed on the appeal, but the subject of a more detailed analysis in his recent report.

244 The applicant invited the Court to accept that were Professor Cordner called by the Crown at the trial his evidence would have been consistent with his evidence on the appeal and, if his evidence were accepted by the jury, it would have very significantly undermined the Crown case. Although the Crown objected to the tender of the report, and to the applicant calling Professor Cordner to give evidence because his evidence was neither fresh nor new, the report was not tendered as new or fresh evidence. Grounds 14 to 17, which are concerned with the reception of fresh or new evidence, do not include any claim that Professor Cordner's report is in that category. The only basis for receiving Professor Cordner's evidence was to determine whether his current views concerning the question whether there was a pattern or relevant similarity in the knife wounds sustained by all three deceased are consistent with the views he held at or prior to trial. If we are satisfied that they are, and we are satisfied that the Crown was aware he held those views, the question arises whether the Crown was obliged to call him at trial, or whether the assertion in correspondence with the applicant's solicitors prior to trial that he was "plainly unreliable" relieved the Crown of that obligation.

245 Given the extent of Professor Cordner's criticism of the accuracy of the opinions expressed by each of the Crown witnesses, the Court also received additional evidence from Dr Cala and Dr Lawrence, both in response to Professor Cordner's criticisms and by way of further comment from them on the question of similarity of the wounds. They each gave evidence and were cross-examined. The Crown did not tender a further report from Dr Culliford on the appeal and did not call her to give evidence.

The applicant's argument concerning the admissibility of the evidence of similarity in summary

246 The applicant submitted that the evidence of Dr Cala and Dr Culliford concerning the similarity of wounds was not admissible as their evidence failed to meet the test of admissibility for opinion evidence under s 79 of the Evidence Act. The applicant submitted that this was so even if Dr Lawrence, the forensic pathologist who performed the autopsies on each of the deceased in 1993, was entitled to give evidence that he observed a grouping of similar stab wounds on each of the deceased in 2000.

247 The applicant further submitted that, even if admissible, their evidence ought to have been rejected in the exercise of discretion under s 137 of the Evidence Act, given the potential of the evidence from three Crown experts to overwhelm the jury's consideration of whether the Crown had proved the applicant killed his parents, where the probative value of the evidence used to assist in proof of that fact was slight. It was also submitted that there was a risk that the jury would adopt an inappropriate and unsafe process of reasoning to the conclusion that one perpetrator inflicted all the injuries from the mere appearance of similarity in the wounds that were inflicted.

The significance of the evidence of similarity in the Crown case

248 In opening the Crown case to the jury the Crown Prosecutor relied upon what she described as a concentration of multiple stab wounds on the centre of the chest of each of the deceased, from which she suggested that it would be open to the jury to find that each of the deceased were stabbed in a "very similar way", thereby supporting the Crown case that the applicant had murdered his parents in relatively quick succession and then stabbed his brother.

249 Her closing address focused on what she described as the "extraordinary similarity" in the pattern of stab wounds. She submitted that the pattern was such that the wounds could not have been inflicted coincidentally, further supporting the Crown case that it was the applicant who inflicted the wounds on each of the deceased. In a document which the jury were given for the purposes of their deliberations, the similarity of the stab wounds, referable to the grouping and number of wounds on each of the deceased, was the first of fourteen separate circumstances relied upon by the Crown in proof of guilt.

250 When the trial judge directed the jury as to what was meant by circumstantial evidence, and the way that they might approach an assessment of the various circumstances relied upon by the Crown in proof of guilt, he suggested that the amount of evidence called by the Crown bearing upon the similarity of the wounds, and the amount of time the Crown spent on the issue, might signify that it was a very significant part of the Crown case. The trial judge also added: "... and you might think that it is [significant] ...". He also went on to say that if the jury came to the view that they could not draw any conclusion from the evidence of similarity in the number and grouping of the wounds in light of the other circumstances in the case, or if they decided that the evidence did not assist them in determining whether the Crown had proved guilt and they disregarded that evidence altogether, that might make the Crown case a lot weaker than if another of the 14 listed circumstances were discarded by them.

The evidence the Crown relied upon to prove relevant similarity

251 The evidence from Dr Lawrence, Dr Culliford and Dr Cala concerning the issue of similarity was referable to the number of stab wounds on the body of each of the deceased, the grouping of the wounds on the chest and back areas, the configuration of the wounds, and the alignment of the wound tracks relative to the bodies of each of the deceased.

252 In different ways and with different emphases the experts each commented upon the similarity of the wounds referable to these features. In cross-examination, dissimilarities in the alignment of some of the wounds suffered by Mr Gilham and Mrs Gilham on the one hand and Christopher on the other were emphasised. Defence counsel also sought to neutralise the significance of the evidence of similarity by inviting each of the expert witnesses to concede that since it was common ground that the same knife had been used to inflict the wounds on each of the deceased, the shape of each wound would necessarily take the shape and dimensions of the blade and that the wound depths were also likely to be comparable for that reason.

253 Defence counsel also sought and obtained the concession, at least from Dr Culliford and Dr Lawrence, that the grouping of the wounds to the chest of each of the deceased was unremarkable given that the chest would be expected to be targeted by a person wielding a knife with an intention to kill.

254 The experts were not permitted to offer an opinion as to whether any of the features of similarity identified by them indicated or suggested that one or more than one person was involved in inflicting the injuries on each of the deceased. This was in accordance with the ruling given in the first trial that there was no field of specialised knowledge concerning the characteristics of stab wounds to render such an opinion admissible under s 79 of the Evidence Act. The trial judge also ruled at that time that none of the witnesses would be permitted to offer an opinion as to the degree of similarity between the grouped wounds, that being a matter exclusively for the jury to determine. There was no consideration given at that time, or during the subject trial, to whether the features of similarity relied upon by the Crown might provide a basis for the admission of the evidence as coincidence evidence under s 98 of the Evidence Act.

255 One of the questions raised by Ground 8, although somewhat tangentially, is whether the trial miscarried because of the use to which the prosecutor put the evidence of similarity when she urged the jury to find that the degree of similarity in the pattern and grouping of the stab wounds, which she at various points described as "extraordinary", compelled the conclusion that there was only one perpetrator. (Ground 12 is concerned with other aspects of the Crown Prosecutor's address without specific reference to her repeated use of adjectives of this kind.) Those parts of her address are referred to at [319]-[324]. Although the trial judge did not permit Dr Culliford, or any other Crown witness, to give that evidence, and later directed the jury that there was no "scientific basis" to support an inference about the number of perpetrators from the pattern of the stab wounds, he did permit the Crown to lead evidence from both Dr Culliford and Dr Cala that they shared Dr Lawrence's views that the groups of wounds on each deceased were similar.

How the issue of similarity emerged

256 It is important to trace how the evidence of similarity in the grouping or pattern of the stab wounds came to the attention of the prosecuting authorities. While that exercise will not be determinative of the question of admissibility, it does inform the question whether the Crown was obliged to call Professor Cordner as a witness.

Dr Lawrence's findings on autopsy and the preparation of diagrams showing the location of the stab wounds

257 Dr Lawrence performed the autopsies on each of the deceased in August 1993. The cause of death was reported as multiple stab wounds to the chest of each of the deceased. Extrapolating from the autopsy reports, Dr Lawrence gave evidence of the knife wounds on each deceased as follows:

Mr Gilham

Mr Gilham sustained a total of 28 knife wounds - 27 stab wounds and one incised wound on the left jaw. Of the stab wounds 16 were grouped to the front of the chest, 5 to the back of the left shoulder and two to the back of the right underarm near the armpit. The balance were to the face, neck and left upper arm. One or more than one of five wounds to the back would have proved fatal penetrating the heart that way.

The front of the body of the deceased was badly burnt and not all of the wounds to the chest were able to be tracked. Those that were distinguishable revealed an alignment from front to back, from left to right and downwards towards the feet.

The track of the stab wounds to the back of the right underarm was from left to right and slightly upwards penetrating skeletal muscle.

Mrs Gilham

Mrs Gilham sustained a total of 17 knife wounds - 15 stab wounds and two incised wounds on the left forearm and the left hand, probably defensive type injuries. Of the 15 stab wounds, 13 were grouped on the back of the chest with the wound track from back to front and slightly downwards and with most wounds in a vertical alignment. A further single stab wound to the front of the chest passed through the left lung, the aorta and the right lung and there was a further single stab wound to the front of the neck. The wound to the chest, together with one or more wounds to the back, would have been fatal.

Christopher Gilham

Christopher sustained a total of 17 stab wounds. There were 14 grouped wounds to the front of the chest and three to the back of the chest with a defensive type injury on the left hand and a superficial cut on the side of the face. Of the wounds to the back, two penetrated the skeletal muscle in the scapula (but not very deeply) while the third was close to the spine, penetrating the chest cavity and severing the lung and the blood vessels that run from the lung to the heart. The wounds to the chest were scattered across the central chest with the wound track from front to back and horizontally or slightly downwards in alignment.

258 During the autopsy the location of each of the stab wounds on each of the deceased, together with the length and depth of each of the wounds on the front and back of each of the deceased, was noted by Dr Lawrence on a pro forma anatomical drawing of a generic human figure. He did not make any observation of any similarity between the wounds on any of the deceased in the autopsy reports prepared in August 1993.

259 For the purposes of the second inquest seven years later, Dr Lawrence prepared a side view of a different generic anatomical drawing of the upper body on which he marked arrows indicating the general direction from which the stab wounds to the chest and back were delivered to each of the deceased. It was at that time that he also prepared a transparent overlay of the same pro forma anatomical drawing he used in the autopsy, on which he traced the wounds from the post-mortem drawings for each deceased. This was to enable a comparison to be made between the relative location of the grouped stab wounds on the front of the chests of Mr Gilham and Christopher. To enable a comparison between those wounds and the wounds on Mrs Gilham's back, Dr Lawrence overlaid her transparency over the other two. It was this exercise which at the time excited Dr Lawrence's observation that the grouped stab wounds on each of the deceased appeared to him to be "similar", an observation which he volunteered to investigators.

260 The diagrams prepared by Dr Lawrence at the autopsy and for the purposes of the inquest were tendered in both trials without objection. At the point of the tender of the transparencies in the second trial the jury were invited by the trial judge to place the transparency showing the location of Mr Gilham's and Christopher's wounds over the top of each other and, after turning over the transparency relating to Mrs Gilham (so that the wounds entering the chest cavity from her back were shown as wounds to the front), to overlay that transparency over the other two.

261 The Crown's intention to rely on what was claimed to be a discernible similarity in the pattern of grouped stab wounds, in proof of its case that the applicant was the person who inflicted all the wounds, was flagged well in advance of the first trial.

262 The views of Dr Cala, Dr Oettle and Professor Cordner concerning that question were first solicited by investigating police in 1999 for the purposes of the second inquest in 2000. They each gave evidence at the inquest, as did Dr Lawrence.

263 In the preparation of his report dated July 1999, Professor Cordner was invited to comment upon a number of areas reported upon by Dr Oettle, a forensic pathologist, who had by that time already provided his views to police about a number of matters considered by him to be of forensic significance. Commenting upon Dr Oettle's expressed view that "the pattern of stab wounds in all these three people was very similar being concentrated in the area involving the heart and lungs", Professor Cordner said:

"This is a true statement about stabbings in general. The vast bulk of homicidal stabbings involve the chest. The only similarity here is the number and the fact that they could all have been inflicted with the same knife. There is no specific peculiarity to tie the three together in my mind. Indeed, there are some differences. The mother had two incised wounds to the front of the neck. The brother had a superficial incised wound on the face. The father had an incised wound on the face and a stab wound on the neck. I do not think it is safe to conclude that the distribution and number of stabbed incised wounds in these three deaths are such as to indicate that the three deaths were by the one hand. I do not even think that it is a probable factor to be weighed in the balance - it is neutral in trying to conclude whether there was one or two murderers in this case. Conversely, one could say that as the wounds are consistent with being caused by the one knife, in this respect, the findings are consistent with Jeffrey's story. [This view became one of the bases of "unreliability" the Crown relied upon as justifying the decision not to call Professor Cordner at trial. It will be discussed later in this judgment]." (Emphasis added)

264 In a report also dated July 1999, Dr Cala took a different view to Professor Cordner. He said:

"The pattern of injury and indeed the causes of death in all three cases are extremely similar. The number of stabbings in two of the three victims are identical (17 times)..."

265 He went on to say:

"...That these people should die in extraordinarily similar ways suggests to me the possibility that only one person was involved in the three deaths."

266 When Dr Cala gave evidence at the inquest he confirmed that this remained his view. When he was invited at the inquest to comment on Dr Lawrence's transparencies, he said:

"Q. Having seen that, prepared by Dr Lawrence, in relation to the similarity of the wounds to each of those, are you able to make any comment?

A. None really other than what I have said both in my report and what I have just stated, that they are remarkably similar and for two people to have been stabbed in such a violent way an identical number of times is remarkable, in my opinion.

Q. Now the two people relate to the mother and the son?

A. Yes.

Q. So that's remarkable, you say, in one respect?

A. Yes.

Q. The number of wounds. Also between the father and the son the positioning of stab wounds, do you also, having compared the transparency between each of them, do you make any comment about that?

A. There are many stab wounds on the front of the chest of both Stephen and Christopher Gilham, and fewer, although quite a number on the back of both Stephen and Christopher Gilham and in my opinion they are as I said, strikingly similar.

(Dr Cala's evidence at the inquest was tendered on the voir dire. Neither Dr Lawrence's nor Professor Cordner's evidence was tendered.)

267 Dr Culliford's views were sought for the first time in June 2005 when the investigation into the deaths of the three deceased was under review. Her report of that date was doubtless included in the brief of evidence that was sent to the Director of Public Prosecutions in November 2005 in support of the ex officio indictment which was filed in February 2006. While it is reasonable to assume that Dr Cala's report and his evidence at the inquest, together with Dr Lawrence's post-mortem report and the materials prepared for the inquest, were also included in the brief of evidence, there is no evidence one way or the other as to whether Professor Cordner's report or his evidence at the inquest were brought to the Director's attention.

268 Dr Culliford was asked in June 2005 to review the forensic evidence and, in particular, to report upon whether the injuries sustained by the three deceased were likely to have been caused by one perpetrator rather than two. (The very issue that the trial judge ultimately determined was not a matter for expert opinion.) She was provided with copies of the post-mortem reports, photographs of the crime scene, and photographs of the post-mortem examination. The relevant extracts from Dr Culliford's statement dated June 2005 are as follows:

"iii. Number of wounds

...

The fact that multiple stab wounds to the chest was chosen to murder all three victims would tend to favour one perpetrator being involved in all three assaults rather than two. However, stabbing to the chest is a common site for unlawful wounding and homicidal stabbing, but multiple stabbing is much less common.

...

iv. Position and grouping of wounds on the body

The positions and grouping of the wounds showed significant similarities to each other. All three victims had both anterior and posterior wounds. It is possible that initial single stab wounds were made from a different position to disable the victim before the grouped stab wounds causing potentially more serious damage occurred in a different position. For instance, Christopher may have been stabbed in the back while being chased down the stairs to the games room but the main group of stab wounds occurred later when he was collapsed on the floor.

...

4. Possible scenario in each assault

...

Doctors CALA, LAWRENCE and OETTLE (Coronial report, item 3, pp 3 and 4) all acknowledged the similarity of the wound patterns. [ It would appear she was not provided with Professor Cordner's report.] Dr CALA [sic Dr LAWRENCE] prepared an overlay for the coroner's court and illustrated the similar patterns. I agree with these opinions. The similarity of the grouped wounds was such that I cannot believe that the violent acts required to stab three people virtually the same number of times in such a similar manner could be perpetrated by two different people. The similarities must strongly suggest that there was a single perpetrator who inflicted those wounds." (Emphasis added)

269 Dr Culliford's report was served on the defence in advance of the trial. It would seem that Dr Cala's report dated July 1999 was also served and that Professor Cordner's report, together with his evidence at the inquest, was either served or disclosed.

270 The applicant's solicitors were notified by letter of the Crown's decision not to call Professor Cordner in January 2008 because he was, in the Crown's assessment, a witness who was "plainly unreliable". Extracts of the letter are set out at [395]. Although the letter was signed by the Solicitor for Public Prosecutions, we proceed on the assumption that the decision not to call Professor Cordner was made by the then Crown Prosecutor with or without the concurrence of the Director of Public Prosecutions.

Dr Cala's evidence on the voir dire at the first trial

271 Dr Cala's evidence was taken on the voir dire prior to Dr Oettle, Dr Culliford and Professor Cordner giving evidence on the voir dire concurrently later in the trial.

272 At the commencement of the voir dire into the admissibility of Dr Cala's evidence, the prosecutor disclaimed any reliance on his view, reported in his July 1999 statement (extracted at [264]-[265] above) and repeated in his evidence before the Coroner, as to the likelihood of there being only one perpetrator because of similarity in the way the wounds were inflicted or their number. Despite this disclaimer, when the admissibility of Dr Culliford's evidence concerning the similarity in the pattern of wounds was raised, and the further voir dire was convened, which attracted a final ruling on the admissibility of the similarity evidence, the prosecutor claimed an entitlement to rely upon the expert opinion of Dr Culliford as going to this very issue. The reason why the Crown abandoned reliance on Dr Cala's conclusions but later pressed for admission of the same conclusion from Dr Culliford is not clear.

273 Dr Cala's qualifications include a Bachelor of Medicine/Bachelor of Surgery from the University of New South Wales (1982). After working in various public hospitals he commenced training in pathology in 1998 and took up the position of Forensic Pathology Registrar at the New South Wales Institute of Forensic Medicine at Glebe in 1994. In 1995 he worked at the Universities of Edinburgh and Glasgow as a forensic pathology fellow undertaking postgraduate studies. He returned to Australia in 1996 and was appointed a staff specialist at Glebe, where he remained until 2002.

274 Dr Cala's evidence can be summarised as follows:

(a) After accessing the autopsy reports prepared by Dr Lawrence, including his descriptions of the type of wounds, the location of the wounds, the depths of the wounds and the angle of the wounds, and photographs taken during the post-mortem examinations, it was his view that the pattern of injuries on each of the three victims was "very similar". It is of some significance that this is the first reference in the evidence to there being a "pattern of injuries" - a concept Dr Cala used in his report dated July 1999.

(b) He identified the similarities in the pattern as including:

i. the fact that the majority of the injuries (and those that were relevant for the purposes of comparison) were to the upper chest region on the front of the body in the case of Christopher and Mr Gilham, and to the back of the body in the case of Mrs Gilham;

ii. the fact that the injuries were to vital organs including the heart and lungs and the vessels that radiate from them; and

iii. the depths of the wounds, in that they all penetrated into the body to a depth of at least 150mm.

(NB: For reasons which are not apparent, on the voir dire Dr Cala did not refer to the number of wounds inflicted on each of the deceased as a feature of similarity. Although he made specific reference to the number of wounds to each of the deceased exceeding ten in his evidence at the inquest, and although the Crown tendered his evidence of the inquest on the voir dire, neither counsel made any reference to it in submissions. By the time of the second trial the number of stab wounds featured prominently in Dr Cala's evidence, which was given considerable emphasis by the Crown Prosecutor in her closing address. In the context of considering Ground 18, and the evidence led on the appeal from Professor Cordner, the relevance (or irrelevance) of the number of stab wounds is discussed. Professor Cordner's evidence at the inquest was not tendered by the Crown or the defence on the voir dire and was not made available for the purposes of the appeal.)

(c) In cross-examination Dr Cala accepted that in concluding that the pattern of wounds is "very similar in each of the deceased" he disregarded features of the wounds which were dissimilar. He also conceded that he was only referring to similarities in the pattern of wounds to the upper torso, as there were significant differences in the wounds to other parts of the bodies of the deceased.

(d) He accepted there were dissimilarities in the angle of entry of the wounds. The wounds to the back of Mrs Gilham's body had been inflicted in a downward direction and the wounds to the front of the chests of Mr Gilham and Christopher were predominantly horizontal. The length or depth of the wound tracks were also variable, with only some of the wounds being of similar depth in the sense that they could be described as "alike".

(e) Dr Cala conceded that, irrespective of the strength of the person wielding the knife or the person's state of mind, if two people use the same sharp knife to fatally stab a person, he would expect the wounds to be similar.

Dr Lawrence's evidence at the first trial concerning the similarity in the grouping of stab wounds

275 After leading evidence before the jury from Dr Lawrence concerning the location, grouping and alignment of the stab wounds he had marked on the transparencies, the Crown Prosecutor asked the following question without objection:

"Q. Doctor, what do you say about [the] comparison of the grouped stab wounds to these three victims.

A. I think they're similar."

276 Dr Lawrence was then asked whether he had been made aware of any other cases of multiple stabbings. Counsel for the applicant objected to that question, which was not answered. The further voir dire was then convened to resolve whether the Crown would be permitted to call evidence from Dr Culliford (and at that time Dr Oettle) confirming Dr Lawrence's opinion that the grouped stab wounds were similar. Dr Culliford gave evidence on the voir dire concurrently with Professor Cordner. Professor Cordner was called by the defence, no doubt as a consequence of the Crown's advice that they did not intend to call him as a witness because of his so-called "unreliability". (Dr Oettle also gave evidence for the Crown on the voir dire. However, because the Crown later abandoned any reliance upon his evidence, and because he was not called in either trial, his evidence will not be further referred to.)

277 The voir dire was also used to resolve the challenge to the admissibility of evidence from both witnesses in connection with blood on the knife and blood on the applicant, which are the subject of Grounds 6 and 7 of the appeal.

Dr Culliford's evidence on the voir dire at the first trial

278 Dr Culliford holds a Bachelor of Medicine/Bachelor of Surgery with Honours and was made a member of the Royal College of Physicians of the United Kingdom in 1974 after postgraduate medical training. She holds a Masters Degree in Forensic Medicine from Monash University and has undertaken a Master of Science (Crime Scene) course at Griffith University (2002). She said she had been involved in forensic medicine as a forensic medical officer since 1987. She described her work as a forensic medical officer in a number of ways. It involved attendance at crime scenes and working with police and police scientists. She made it clear that she did not perform autopsies although, with her expertise in toxicology and injury interpretation, she worked in cooperation with the pathologist at post-mortems. She also attended to live victims of crime and persons accused of committing offences to obtain forensic samples for the purposes of a prosecution. As at the date of her statement in 2005 she was a senior forensic medical officer in the Clinical Forensic Medical Unit in Brisbane. At the time of giving her evidence on the voir dire in February 2008 she had retired due to ill health.

279 She said her only previous experience of multiple stab wounds inflicted on a number of victims in the one incident was where there was no issue that all wounds had been inflicted by the same perpetrator. In one case a father had stabbed his children and his wife and in the other case a sibling had stabbed a brother and a parent. When asked about the degree of similarity in the stab wounds inflicted in those two cases as compared to the similarity in stab wounds sustained by the deceased in this case, she said that there was considerable variation in the appearance and spread of the stab wounds in both previous cases. In the first case this was due to the relative difference in the size of the father and the children, and in the second because the brother was not disabled very quickly. In effect, her exposure to these stabbing events did not provide any basis for comparison with the deaths of Mr and Mrs Gilham and Christopher. She was not cross-examined on this aspect of her evidence.

280 Dr Culliford said that the close grouping of between 15 and 17 stab wounds centralised in one area on the upper body of each of the deceased showed "an extraordinary degree of similarity", which may lead to a conclusion as to the number of perpetrators. This was a departure from her statement (reproduced at [268]) where she expressed her view as to the likelihood of there being one perpetrator in categorical terms. Ultimately, however, she deferred to what she recognised as the superior academic and professional credentials of Professor Cordner and accepted that despite what she maintained was a "marked degree of similarity in the grouping of the wounds", in the absence of any body of research into stab wound patterns, care should be taken before drawing a conclusion as to the number of perpetrators involved in inflicting the injuries. She expressed reluctance in drawing that conclusion. As to the significance of a pattern in the appearance of the wounds, she said:

"Certainly I took this to a grand round up in Queensland of experts in the John Tom Centre. I took it to other experts to say that there was a lack of body of evidence to determine whether you could link patterns of stab wounds with one perpetrator.

Professor Cordner is absolutely correct that there is no body of evidence stating that you can necessarily link it, but we throughout, in forensic medicine, when you're talking about injury interpretation, talk persistently about patterns of injuries, and from determining patterns of injuries you draw conclusions. I had this rammed into me when I did my masters degree, and, certainly since, that patterns of injuries lead you to make conclusions.

There are patterns of injury here that have great similarity. There are differences, but there are some great similarities. I think we have to think why there are those similarities.

Although I agree with some of what my learned colleagues have been saying, I have to, even though Professor Cordner is a very experienced and very long-standing revered pathology professor, he would have a lot more experience in hands on, when it comes down to interpreting patterns, I have read extensively about research that has occurred in the States. The States seemed to have managed to collect an awful lot of data that we don't have here. They have started to look at patterns of injuries. There are no set databases, but there are more conclusions saying that there are certain patterns that can be attributed to the nature of what happened. Whether it can be drawn to conclusions we are trying to reach here, I do not know, and I believe that Professor Cordner is correct in having doubts, but I still believe that the similarities still have to be explained. I think there are more similarities than dissimilarities, and I think that needs to be explained."

281 She adhered to the view that the grouped wounds on the chests of Mr Gilham and Christopher, and the grouped wounds on Mrs Gilham's back, were "remarkably similar" or "extraordinarily similar" - similarities which to her mind outweighed the dissimilarities. When she was asked against what data or index she assessed the similarities as "remarkable" or "extraordinary", she said her conclusion derived from a comparison of each group of wounds to the other two:

"... What I have tried to do in my statement is try and nut out the injuries that occurred around the disabling process and the injuries that occurred while the victim was still alert and awake and able to defend themselves or try and defend themselves and then the grouped injuries that appeared to occur during a process when almost certainly the victim was disabled and perhaps unconscious or even dead, and it's those groupings at that time which are similar. The numbers are similar and the grouping over a small area is similar. But whether or not I would be prepared to draw a conclusion from that, as I have said, because there is no body of evidence, I think we have to take care with drawing a conclusion even though they are extremely similar."

282 When she was asked by defence counsel what methodology she used in concluding that the pattern or grouping of injuries was "extraordinarily similar" as between each deceased (other than by reading Dr Lawrence's autopsy reports and reviewing his diagrams), she said:

"DR CULLIFORD. Yes, and I drew up my own diagrams too, but yes, I looked at the information that I was given.

Q. You were in the same position as the jury are in now, isn't that right? Have you used any particular technique or enhancement procedures to help you reach the same conclusion?

A. I looked at the dimensions. The directions, the positions of the injuries and then logged it - I drew it up on paper myself to look at how these things were similar. For instance, in some of the groupings, they seemed to be in a similar direction, which implicates (sic) that they occurred at a similar time, that is, one after the other. As his Honour says, I am not drawing any specific conclusion as to the significance of that similarity. What I am saying is I have looked at all the descriptions in the pathology very carefully. I spent weeks over my initial deliberation over this to look at the similarity or otherwise of these injuries, and I felt that there were groupings that were similar. However, drawing a conclusion myself as to, especially since it is lacking a specific body of evidence to confirm this, I am not so sure that I, as a forensic clinician, can in fact draw a conclusion as to whether it was one or two perpetrators."

Professor Cordner's evidence on the voir dire at the first trial

283 At the time of giving evidence Professor Cordner was Director of the Victorian Institute of Forensic Medicine and a Professor of Forensic Medicine at Monash University. He gave evidence by reference to statistical data that over the previous 20 years homicide by stabbing was the most common form of homicide in Australia and that the infliction of large numbers of stab wounds (14, 16, 19 stab wounds or more) is not uncommon. He confirmed that there was no literature or science comparing patterns of wounds other than references in the literature to signature injuries (the so-called "Mark of Zorro" injuries) or distinctive patterns of mutilation inflicted by mentally disturbed or psychotic offenders. He said that there was no database kept in Australia or overseas relating to patterns of wounds caused by knives. He went on to say:

"I am not aware of any work in the literature that has correlated the similarity of the number and distribution of stab wounds or injuries of any kind between multiple victims in one event and the number of possible assailants.

...I do not believe there is an area of expertise that enables a person to say that injuries are similar as between numbers of victims to allow a conclusion that there may be one or more assailants."

284 The trial judge then said:

"...I don't care what this witness' opinion is on that matter [the conclusion as to the number of assailants] at the present time. I am simply asking whether [Dr Culliford] has sufficient expertise to say whether these injuries are similar full stop, nothing more. All I seem to be getting is something about the information or the learning or the expertise to draw some conclusion about that. That's a separate question. Are you suggesting that Dr Culliford cannot, as a matter of expertise, tell me that these injuries or these wounds are similar? Is that what you're saying, professor?

Professor Cordner. ...Leave aside what is done with the conclusion, then I think you can draw - you can say that particular injuries on one deceased are similar to particular injuries on another deceased."

285 Professor Cordner gave evidence that he did not regard it as significant that the heart and lungs of each of the deceased were penetrated by stab wounds. He preferred to focus on the fact that the wounds were delivered to the back of one of the deceased (Mrs Gilham) and to the front of the other two deceased as important in determining the position of the deceased relative to the assailant when they were killed. As it was common ground that the same knife was used to inflict all wounds, he was of the view that this would inevitably result in a predictable level of similarity in the wounds that were inflicted and that he would give "no weight" to their number or distribution as constituting a pattern. He also considered that the dissimilarities in the number and distribution of the wounds were significant.

The submissions on the voir dire

286 Defence counsel submitted that whilst it was Dr Lawrence's evidence that he saw what he described as "a similarity" in the grouped wounds sustained by each of the deceased when using the transparencies as overlays (a submission from which it must be assumed that there was no objection to Dr Lawrence giving evidence of what he observed), it did not follow that the Crown should be permitted to call evidence from Dr Culliford and Dr Cala that they agreed with Dr Lawrence's observation and to give their opinion as to the extent of the similarity.

287 In the course of submissions defence counsel also made it clear that it was not the applicant's intention to call Professor Cordner to give evidence before the jury, as it was Professor Cordner's considered view that aside from so-called "Mark of Zorro wounds" there was no scientifically sound basis to engage in a process of analysing the degree of similarity between wounds, or groups of wounds, on the bodies of a number of deceased, or to attempt to identify or interpret a pattern of wounds and then to embark on a process of reasoning as to the number of perpetrators that might be involved by a comparison of the pattern of wounds inflicted.

288 The Crown Prosecutor then made the Crown's position clear. He said he intended to rely upon Dr Culliford's opinion that from a scientific perspective the similarities were "extraordinary", thereby supporting what he intended to submit to the jury, namely that they would reject as irrational that:

"... two people...by coincidence, happen to inflict almost identical wounds to different victims on that night in the same household using the same knife."

Again, it is noteworthy that it was not submitted that the evidence would or might be admitted under s 98 of the Evidence Act.

289 In the course of submissions his Honour made patent his view (later incorporated into his ruling on the voir dire) that since there was no reliable data or research on the prevalence of stab wounds or their patterning in cases of multiple or single homicides, he was not going to permit the Crown to lead evidence from any of the expert witnesses that because the wounds were similar there was a probability, or even a possibility, that the same person inflicted the wounds.

290 Nevertheless, defence counsel maintained his objection to Dr Culliford's evidence. He submitted that to the extent that her opinion was based upon Dr Lawrence's post-mortem findings in 1993, the materials he prepared at that time and the transparencies he prepared for the purposes of the inquest in 2000, she was simply restating his observation of "similarity" and her evidence did not satisfy the test of relevance under s 55 of the Evidence Act as the first portal to admission of her opinion. To the extent that Dr Culliford acted on her own diagrammatic analysis, counsel submitted that it was not shown to be based upon any demonstrated specialised training of any relevant kind.

291 He also submitted that because the degree of similarity in the pattern or grouping of stab wounds is not a recognised area of scientific expertise, and because Dr Culliford claimed no specialised knowledge in interpreting stab wounds, whether derived from her training, study or experience as required by s 79 of the Evidence Act, she was not qualified to offer an opinion about what Dr Lawrence observed.

292 In the alternative, he submitted that even if Dr Culliford were sufficiently qualified in forensic medicine to offer the opinion that the wounds were similar in the way that she explained, and her evidence was relevant on the same basis that Dr Lawrence was permitted to give evidence of his observations that the wounds were similar, then her evidence (and Dr Cala's evidence) should be rejected in the exercise of discretion under s 137 of the Evidence Act. Counsel submitted that because of material dissimilarities in the wounds, and because evidence of similarity could not be tethered to any conclusion about the number of perpetrators as the underlying fact in issue, the probative value of the evidence was low. This carried with it the risk of unfair prejudice because of its potential to provide scientific legitimacy to the Crown's case that the applicant inflicted all of the injuries. Defence counsel further submitted that the prejudice would not be cured by the trial judge's direction that the jury were not to approach an assessment of the expert evidence in that way.

The ruling on admissibility

293 The trial judge made express reference only to Dr Culliford's evidence when ruling on the issue of similarity. Objection was also taken to Dr Cala's evidence. It is common ground that the ruling applied equally to his evidence.

294 His Honour was satisfied that because there was a reasonable expectation that the accused would attempt to lead evidence of the dissimilarities between the wounds (despite defence counsel making it clear that the only reason that he would find it necessary to deal with dissimilarities would be if the Crown contended there were relevant similarities), the Crown was entitled to call further evidence "as to that similarity", and for this reason the evidence of Dr Cala and Dr Culliford satisfied the test of relevance. The following exchange between defence counsel and the trial judge reflects his Honour's view, which was incorporated in brief reasons published the following day:

"BOULTEN: Then, the only reason why that evidence is being adduced is to tell the jury one way or another that scientifically this man must have been the perpetrator of all three killings.

HIS HONOUR: Not at all.

BOULTEN: There is no other relevance to the similarity. The fact of the similarity cannot be relevant for any other reason.

HIS HONOUR: Well, of course it's relevant. Of course it is relevant to a determination if the one person killed the three people. Of course it is. It would be a nonsense to suggest it isn't. All that being led here is these witnesses' opinion that the relevant wounds are similar in their expertise and their experience and that those similarities, if you like, outweigh, very greatly, the dissimilarities and I can't see why they couldn't say that and that is without saying "remarkably" or using any other adjective.

BOULTEN: Then it really then is analogous to somebody coming along, say a fingerprint expert, to say that this mark here is similar to that mark there, but not then leading to a conclusion or offering an opinion about the relevance.

HIS HONOUR: That's right. There is nothing wrong with that. That is very often done. A person can come along and it is the perfect example from identification evidence, can come along and say these people are similar. Now nobody is going to say that that person identified the person; the Crown can't say that, but the Crown can say, as a matter of circumstance, a relevant factor is that the people were similar.

Now the matter here is a relevant circumstance, the opinion here by the experts that the wounding here is more similar to the three of them then dissimilar.

BOULTEN: Then the only logical way that that evidence about similarity can add to the Crown's circumstantial case is if they are prepared to reach a conclusion that the evidence points towards a single assailant.

HIS HONOUR: No, not at all. That is to take a piece of circumstance and to say that that circumstance must inevitably lead to the conclusion and we know that that is not how it's done. It is simply that one of the circumstances here may be the lack of blood and that may be explained; it may not be explained. Another circumstance is the similarity of the wounds. Another circumstance, I don't know - there could be all sorts of things the Crown is relying upon. If the Crown wants to put his egg[s] in one basket and say you can be satisfied beyond reasonable doubt that, because of the similarity of the wounds, the same person must have done it I will make a direction to the jury about that."

295 As to the question of Dr Culliford's expertise, his Honour held:

"It seems to me that the witness has the necessary expertise to give her opinion as to whether the wounds inflicted upon the three deceased persons are more similar than dissimilar and the reasons why she comes to that view. I have no doubt it is within her expertise, and so much was accepted by Professor Cordner, an expert who is relied upon by the accused even though Professor Cordner would come to a different view.

In light of the fact that there is a reasonable expectation that the accused will attempt to lead evidence or make submissions to the jury as to the similarity or, in his case, the dissimilarity of the wounds inflicted upon the three deceased, the Crown is entitled to call further evidence as to that similarity.

I do not believe that the witness ought go further and indicate the degree of similarity. It is a matter, in my view, for the jury once the similarities have been pointed out to the jury by the witness relying upon her expertise."

296 The fact that Dr Culliford may not have put her expertise into practice because of her exposure to a limited number of multiple stabbings did not, in his Honour's view, diminish her expertise where it was otherwise sufficiently based upon her training. In his view her limited exposure to multiple stabbings was a matter that went to the weight to be given to her opinion and was a matter for the jury.

297 The trial judge made no express finding as to whether Dr Cala had the necessary expertise to give evidence of his opinions regarding the issue of similarity. There was evidence on the voir dire of Dr Cala's experience as a forensic pathologist and the fact that he had conducted somewhere between 40 and 50 post-mortems involving stabbings over a 14-year period. There was no evidence as to whether any involved the infliction of multiple stab wounds or multiple victims of stabbing by one perpetrator.

298 In the course of argument on the voir dire, although not in the interlocutory judgment, his Honour also made clear his view that while the experts were not permitted to offer an opinion about the degree or extent of the similarity of the wounds, this would not prevent the Crown from advancing a submission to the jury that if, after considering the expert evidence, they were satisfied that there was relevant similarity, it was open to them to consider the degree of similarity in the pattern of wounds as part of the case against the applicant. On the appeal, the Crown submitted that this entitled the Crown Prosecutor to submit to the jury that the degree of similarity was "extraordinary" or "remarkable" and that for that reason they would be satisfied they were all inflicted by the applicant. For reasons which we will discuss later in this judgment, we reject that submission.

299 Finally, the trial judge was not persuaded that the discretion under s 137 of the Evidence Act should be exercised in favour of excluding the evidence. He said:

"I do not believe that the probative value of the evidence is outweighed by the likelihood of prejudice to the accused. The prejudice in that case has to be a likely prejudice. It seems to me, that any prejudice that might arise from this evidence would be likely to be the jury using the evidence [to form the] view that the expert had some opinion either expressed or not expressed or impliedly received by the jury as to the consequence of the finding that the wounds were similar. In my view, that prejudice, if it exists, is appropriately dispelled by directions to the jury."

The evidence of similarity in the second trial

300 The admissibility of the evidence of similarity was not revisited in the second trial. It was accepted that the parties would be bound by the ruling given at the first trial. Dr Lawrence and Dr Cala were the last witnesses called in the Crown case. Dr Culliford gave evidence earlier in the trial. A different Crown Prosecutor appeared for the Crown. There is no evidence as to whether she gave fresh consideration to whether the Crown should call Professor Cordner at the second trial, as we consider she was obliged to do having assumed the responsibilities and obligations of a Crown Prosecutor, as provided for under the Director's Guidelines and in the New South Wales Barristers' Rules.

301 In some respects, although perhaps only as a matter of emphasis, the evidence bearing on the similarity of the wounds called at the second trial was different from the evidence led on the voir dire. (The Court was not provided with the evidence led at the first trial because it had no bearing upon the challenge to admissibility). Although detailed analysis of the evidence led at trial cannot inform whether or not there was error in its admission, it is necessary to review the evidence that was led at trial in order to make an assessment of whether there was a miscarriage of justice by reason of the Crown's decision not to call Professor Cordner.

302 To a more limited extent, the way in which some of the evidence relating to similarity emerged at the second trial, some of it in response to questions asked by the trial judge, is also important. The trial judge's directions to the jury concerning the expert evidence also need to be considered as part of the inquiry into whether the applicant has made good the challenge to admissibility, given the trial judge's stated reasons for his refusal to exclude the evidence in the exercise of discretion.

Dr Culliford's evidence at the second trial

303 Dr Culliford was invited by the prosecutor to detail the injuries sustained by each of the deceased by reference to Dr Lawrence's findings on autopsy. She was first asked to detail the stab wound injuries suffered by Christopher. She referred to 14 stab wounds grouped on the front of the chest of which 9 were grouped very close together on the left side of the chest, following the alignment of the knife. In her view these wounds were inflicted in quick succession at a time when Christopher was disabled or unconscious.

304 She next referred to Mrs Gilham. Of her 15 stab wounds, 13 were in a group on the back of her chest. Because of their similar position and similar alignment, Dr Culliford said the wounds were likely to have been inflicted sequentially.

305 Of Mr Gilham's 28 stab wounds, she regarded 15 as being in a group on the front of the chest. She also considered that these wounds were inflicted sequentially because of what she described as their similar position, alignment and orientation. She was then asked the following questions:

"CROWN PROSECUTOR: But in relation to the grouped wounds that you have described, 14 for Christopher, 13 for Helen and 15 for Stephen, were you able to notice any similarity between that pattern of wounding in each case?

DR CULLIFORD: You have to be careful not to put too much importance on the fact that the chest was chosen. Chest injuries, when you stab someone and wish to kill them, you will tend to, I believe, go for an area where it is likely to kill the person; that is what they are trying to do. And you know that you have got the heart and lungs in the chest and that if you stab somebody in the heart ad lungs they are going to die. So in homicides, in Australia and all around the world, it is a very common place to kill somebody. To try and kill somebody you will go for a part of the body which is likely to disable and kill the person, if that is what you are intending to do. Certainly the multiple stab wounds that I have seen both at the scene and the incidents that I have been involved with, have involved multiple chest wounds, that is, I don't think that can be particularly correlated as being anything particularly special.

Q. What about the sheer number; is it common that there is a large number of stab wounds in homicides involving a knife?

A. No. I think that is - I don't think any particular significance can be correlated from that either. I think that if you are going to stab somebody a few times, you are making sure that that person is going to die and they are going to be grouped. I don't think that that necessarily can be correlated as anything special either. However, the numbers here are actually quite close together. Each one has, in the low teens, the number of stab wounds low. And they are grouped in the opposite side of the chest. They may have - that may have actually stopped the victim from moving. I just find that, the grouping on the opposite side of the chest, as a similarity between the three, but I don't wish to go any further on that.

HIS HONOUR

Q. What do you mean by "the opposite side of the chest"; opposite to what?

A. Opposite to the initial injuries that may have immobilised the victim in the first place.

CROWN PROSECUTOR

Q. Was there any similarity in relation to the depths of the grouped wounds?

A. There were a number of full thickness wounds in all the victims, although the detail given in Helen and Stephen wasn't as great in the autopsy report, and again Dr Lawrence will be able to explain them more fully. They all went to the whole length, the 150 millimetres in all three, which meant that there was a reasonable amount of force used in all three in the multiple stabbing.

Q. What about the directionality of the wounds in each case?

A. I think the direction of the groups I have described were in the same direction, which just confirms that they were likely to be sequential for those particular smaller groups, having taken out the other injuries in that description.

Q. Are there any other similarities upon which you wish to comment, doctor?

A. Only in summary that they had grouped - they each had grouped stab wounds. Christopher had 14 in the front chest and Helen had 13 in the back chest and Stephen had 15 in the front chest, a single possibly fatal wound on the side of the chest opposite to the grouped injuries; Christopher in the back, Helen in the front and Stephen in the back. There were other random injuries and defence injuries, as I have already described, that may be explained by the randomness of the initial assault. And there were defence injuries in two victims, Christopher and Helen."

306 Dr Culliford also identified what she described as a wound that hit "a very big blood vessel" in the heart of each of the deceased, which could have caused extensive bleeding. She went on to say that even though there were grouped wounds on "the other side of the chest" on each of the victims, which she relied upon as one of the similarities, the grouped wounds were not necessarily the wounds that caused death. She considered that that was another feature of similarity.

307 In summary, by the time of the second trial, Dr Culliford had discounted the concentration of stab wounds on the chest or trunk of each deceased and the number of stab wounds inflicted as individual features of similarity, or as constituting a pattern of wounding. Rather, it would appear that it was the grouping of the multiple stab wounds on the chest (or back of the chest) of each of the deceased, and some suggestion of a common alignment in those wounds, which she regarded as similar, together with the grouped wounds being as she described it "on the opposite side of the chest" of each of the deceased. The question asked by the Crown Prosecutor set out above, which elicited that evidence, assumes that the wounds inflicted on the deceased were in a pattern. The concept of a pattern in the wounds was not comprehended by the ruling on admissibility. The trial judge's short reasons for admitting the evidence dealt with similarity per se as distinct from whether the similarity suggested an underlying pattern or was a manifestation of it. There was, however, no objection to the question as framed in that way. Its impact on the jury was not lost, given the question asked of Dr Lawrence as the second of the three experts to give evidence (see [313]-[314]).

308 In cross-examination Dr Culliford confirmed that it was her understanding of Dr Lawrence's findings on post-mortem that the alignment of the wounds was similar as between Mr Gilham and Mrs Gilham, but that there was a difference in the alignment of Christopher's wounds. While the grouped wounds on Mr Gilham's chest and on Mrs Gilham's back were inflicted with the sharp side of the blade of the knife pointing downwards, in the grouped wounds to Christopher's chest (consisting of three wounds on the right of his chest and nine on the left), the blade was pointed to the left or the right. She then gave the following evidence:

"Q. The difference in the angle of those wounds on Christopher's body is explicable by a number of possible explanations?

A. Yes.

Q. You have mentioned some of them in your evidence so far. One is that the person who stabbed him might have moved in relation to the individual wounds?

A. Yes.

Q. So that when the left side wounds were inflicted, the person might have been standing in a different position than when the right side wounds were inflicted, is that right?

A. Yes, yes.

Q. Another alternative is one that you've suggested in your evidence already, namely that the wounds on the right side of the body might have been inflicted whilst the stabber had the knife in different hands?

A. Yes.

Q. Another explanation, I would suggest, is that the person stabbing Christopher might have changed his position simply by twisting his arm?

A. Yes.

Q. That's another explanation, isn't it?

A. There are all sorts of explanations, and that's what I explained as one of the differences between the parents and the son."

309 Dr Culliford also acknowledged that in discriminating between grouped wounds and ungrouped wounds, and in remarking upon the similarities between the grouped wounds, she assumed that there was no relative movement between the particular deceased and the perpetrator when the grouped wounds were inflicted. She also conceded that it was impossible to nominate the order in which any of the wounds were inflicted and conceded the possibility that one or more than one stab wound in an apparent group of stab wounds might have been inflicted at one time, with the majority inflicted more or less sequentially thereafter.

Dr Lawrence's evidence at the second trial

310 At the second trial Dr Lawrence's evidence on the question of similarity was more extensive. While he was asked precisely the same question as was asked at the first trial which elicited his observations of similarity in the grouped stab wounds (set out at [275]), he gave the following answer at the second trial:

"A: Well, when I first prepared it [the transparencies], I was surprised how similar they looked." (Emphasis added)

311 He was then asked by the prosecutor to expand upon his answer:

"Q: Do you say that the similarity is in terms of the location?

A: The location, the clustering on the chest.

Q: The angles?

A: Sorry, what do you mean by "the angles"?

Q: The general angles of those-

A: The alignment of the tracks relative to the body, yes. All of these things surprised me how similar they were.

Q: Lack of rotation or movement within the allowances which you have been given?

A: Well, we have somewhere there is some movement but, yes, predominantly they were.

Q: The depth, allowing that some were -

A: Well, I don't think there's ever been any issue that it wasn't the same knife that inflicted all three sets of wounds, so I don't think the depths are at issue.

Q: In each case all of the wounds are to the chest rather than any other part of the abdomen?

A: Yes, there's a very marked concentration to the centre of the chest. Many of the wounds are aligned to - directed towards the heart and the centre of the lungs.

Q: No wounds to the abdomen in any case?

A: There were no wounds to the abdomen at all, no."

312 There was no objection to this line of questioning, or to Dr Lawrence's expression of surprise at the degree of similarity, despite the ruling given at the first trial that the expert witnesses were not permitted to comment upon the fact or degree of similarity.

313 Before Dr Lawrence was cross-examined, the jury asked a question in the following terms:

"Is the pattern of chest stabbing in the three victims a common pattern for multiple stabbings?"

314 In the absence of the jury, defence counsel raised the potential for Dr Lawrence's use of the term "surprising" to transgress the ruling given at the first trial. He was also concerned to know in advance what the witness would say in answer to the jury's question, no doubt because Dr Lawrence had not given any evidence at any time about there being "a pattern of chest stabbing" and neither Dr Cala nor Dr Culliford had given evidence that such a pattern is common or uncommon, there being no scientific basis for an opinion of that kind. After taking some evidence from Dr Lawrence on the voir dire, in the presence of the jury the trial judge asked the following questions:

"HIS HONOUR:

Q. Is the pattern of chest stabbing in the three victims a common pattern for multiple stabbings generally?

A. Yes. Stabbings very commonly involve both chest and abdomen. In this case there are no stabbings to the abdomen but, yes, it is common to see the chest as a target for the stabbings. In reality, multiple stabbings of - by one person are not common and there is not a lot of data in regard to multiple stabbings, but if you - any lecture on stabbings will usually include the chest and abdomen as being the major targets.

Q. Well, did you say it was surprisingly similar. Why did you say that?

A. Because there is very much a focus on the chest as the target, not to the abdomen, not to the shoulder, to the arms or legs. In this case there is, by comparison with the other triple homicide that I looked at, a great concentration on the central chest.

...

CROWN PROSECUTOR

Q. Does the addition of the wounds on the opposite side to the same place in each case contribute to a similarity?

...

Q. Are you able to answer that?

A. As I said when I put them together, they looked very similar. Now, I don't have a lot of other cases where there are multiple stabbings. The one case I did put together, the pattern was much more widely scattered. That's all I can say in regard to that. There is not a large collection of data regarding multiple stabbings, as in one person stabbing multiple victims.

CROSS-EXAMINATION BY MR BOULTEN

Q. Just in relation to that area of your evidence first, Dr Lawrence, the situation is, and you can talk about this from your own experience as well as from your learning, both book learning and from what other people in your field have spoken about over years, stabbing is a very common way of killing someone in this country?

A. Yes

Q. If it's not the most common, it is one of the most common ways that people die violently?

A. Yes.

Q. Usually when someone kills somebody by stabbing them, they stab them in the chest or the abdomen?

A. Yes.

Q. More often more than once?

A. Yes, that's probably true.

HIS HONOUR

Q. As many times as over 10?

A. I'd have to look at my figures. I don't know the answer to that question."

315 In answer to further questions in cross-examination, Dr Lawrence confirmed that there were differences in the alignment of the wounds as between Mrs Gilham and Mr Gilham on the one hand (at least to the extent that he was able to distinguish the sharp end of the blade on their chest injuries, given that their bodies were extensively burnt) and Christopher on the other hand. This was a matter defence counsel had explored with Dr Culliford. Dr Lawrence also confirmed that given that the knife used to inflict all of the injuries was the same, the shape of each wound would necessarily follow the shape and dimensions of the blade and that for the same reason the depth of the wounds would be likely to be similar.

Dr Cala's evidence at the second trial

316 Dr Cala was the last of the three expert witnesses called by the Crown. He gave evidence that he had performed 4,100 autopsies since 1994, of which between a third and half involved stab wounds. He was then asked the following question (a question not asked of Dr Culliford and a question Dr Lawrence was asked by the trial judge but which was not answered):

"Q: Is it common to have, say, greater than 10 stab wounds as it is to have a fewer number, one two, or three, four?

A: It's uncommon to have more than 10. Usually, in cases of multiple stab wounds, my cases, and those others that I have seen, might be up to about six or eight, but, of course, every case is different from every other, but overall up to about six or eight would be usual or standard. More than 10 would be uncommon."

317 When he was invited by the Crown Prosecutor to comment on the comparison of the wounds sustained by each of the deceased, as described by Dr Lawrence in his evidence, Dr Cala said that the pattern of grouped stab wounds on each of the deceased on the upper chest (whether delivered from the front or the back) and covering a relatively small part of the chest was similar. He also took into account that at least one fatal wound was delivered to the opposite side of the body of each of the deceased, also into the chest, as contributing to the similarity. Although he noted similarity in the size and depth of each wound, he said that did not surprise him given that it was common ground that one knife had been used to inflict all the injuries. In cross-examination, he agreed that a clustering of wounds on the chest in a homicide by the use of a sharp knife of a particular length was likely to lead to fairly precise wounds in terms of shape and depth.

318 In re-examination, he was asked whether, in the context of the chest being a preferred target where multiple stab wounds are inflicted, it was common to see the number of stab wounds that each of the three deceased sustained in this case. He repeated:

"A. ...as I've said, my experience is that stabbings are usually up to about six or eight and rarely over about ten per victim.

HIS HONOUR

Q. Do I understand that the only considerations which lead you to believe that these stabbings are similar is the number of stab wounds, that is, being over ten and the grouping of them?

A. Yes, and their anatomical location in that they were largely grouped in the upper chest, that there were none in the abdomen.

Q. ...

A. It's the anatomical location and the grouping in the upper chest and the absence of stab wounds elsewhere, particularly in the abdomen, that makes me think the pattern was similar.

Q. Is the number an irrelevant consideration?

A. No, the number is relevant. The number of stab wounds for each is at least 15, which also again makes the pattern similar.

CROWN PROSECUTOR

Q. Are the number of fatal stab wounds in each case --

HIS HONOUR: That's irrelevant [sic] as I understand it.

Q. Is that right? The number of fatal wounds is not an irrelevant consideration?

A. No, I'm largely confining myself to the number of stab wounds on the skin's surface about the chest rather than confining myself to the fatal stab wounds for each of them. I'm more referring to the vast number of stab wounds that each sustained to the front or the back of the chest."

The use to which the Crown Prosecutor put the evidence of similarity in closing address

319 In her closing address the Crown Prosecutor identified the similarity of stab wounds as one of the "major areas" of the evidence that proved the applicant's guilt. In developing that submission she variously described the number of stab wounds as being "in a tight little group" to the front and back of the chest. She said that this circumstance, together with the fact that on each of the deceased there was a fatal wound on the opposite side of the body, was "quite amazing". She submitted that the presence of 16 grouped stab wounds to the front of Mr Gilham, 14 to Christopher and 14 to the back of Mrs Gilham's chest was an "extraordinary coincidence". She then submitted:

"But one of the things that also makes these stabbings extraordinarily coincidental and extraordinarily similar is the way they must have been inflicted, ladies and gentlemen. You have heard from Dr Culliford, her opinion that to get that sort of precision, you'd have to be - because your legs - because a human being's legs, along with his or her arms, you'd have to be down in a position kneeling or squatting to get the precision to make perfectly, or almost perfectly parallel close wounds into the middle of the body of those victims.

The Crown submits to you that the person who killed all three would have to have been kneeling because, if one is on one's haunches, if one is squatting, there's just not enough balance to get that sort of precision, you'd have to be kneeling at least on one leg, to get those wounds in those precise positions on the ground in each case. Because undoubtedly, the Crown submits to you, undoubtedly, each of those three people was lying on the floor when those grouped stab wounds were inflicted upon them and you just can't get those angled if you lean down and do it like that, it would be quite an oblique angle, you would be all over the place if you tried to do it on your haunches, well I would be, and I suggest anybody would be because there's no balance in that position, you would have to be kneeling on one knee at least to the side of the body while those wounds were inflicted."

320 She went on to submit that the pattern of wounding was "extraordinarily similar" to the extent that it would be "an extraordinary coincidence" that the stabbings happened as a coincidence as distinct from "the same person doing the same three stabbings".

321 After referring to Dr Lawrence's evidence, and his expression of surprise at the similarity in the location and grouping of the wounds and the alignment of the wound tracks, she said:

"...it is inconceivable that a coincidence like that could happen, this overkill having been applied to the parents. This man [the accused] doesn't know it but, by coincidence, does exactly as his brother does, a brother with whom, as he says, he was not very close."

322 She also submitted that because it is "really very, very unusual for one person to be stabbed more than ten times", the multiple stab wounds delivered into the chest cavity of each of the deceased were further evidence of a pattern of wounds inflicted by the applicant in a murderous rage. When referring to Dr Cala's evidence she said:

"Dr Cala, while I'm talking about the doctors, he's done 4,000 or something autopsies in his long experience as a forensic pathologist. He said it's uncommon to see someone with more than 10 stab wounds. Multiple stabbings might be up to 6 or 8, would be the usual standard number in his experience. So he said the number of stab wounds, being at least 15 in each case, makes the pattern similar between the three. And the pattern of injuries, Dr Cala said, was similar in all three cases. No abdomen, no skull, no leg wounds, a relatively small part of the chest, and all had wounds to other parts which would be expected to bleed both freely and profusely."

323 She also said it was "an extraordinary coincidence" that when one stab wound to the centre of the chest would be sufficient to kill, two perpetrators would inflict multiple wounds to the chest of their victims in the same way. She submitted that the multiple stab wounds inflicted on each of the deceased was "overkill" and another "extraordinary similarity".

324 Before leaving the issue of similarity and addressing other evidence in the Crown's circumstantial case, she said:

"So the Crown says, ladies and gentlemen, that the original attacks were to incapacitate each of them and, once incapacitated, the killer of each, the Crown says to you squarely it was the accused in each and every case got down on knees or knee and stabbed each member of his family in this tight little group of stab wounds. His father got 16 in the front of the chest, his mother got 13 in the middle of her chest through her back, his brother got 14 in the middle of his chest to his front. It's an extraordinary coincidence, ladies and gentlemen, it couldn't possibly - you couldn't possibly accept that that happened just as a coincidence; it was the same person doing the same three stabbings and it was a person who thought that that would never be obvious because he expected that the bodies of his parents would be so burnt that nothing would be able to be told about the stab wounds that they had inflicted."

325 Defence counsel did not complain that the Crown Prosecutor had exceeded the legitimate bounds of argument in seeking to persuade the jury of her expressed view that there was an "extraordinary similarity" in the pattern of wounds, or that it was an "extraordinary coincidence" that each of the deceased suffered more than ten stab wounds, coincidentally amounting to what she described as "overkill". He did not complain that she was inviting the jury to adopt coincidence reasoning when the evidence had not been admitted on that basis. Neither did he complain that the prosecutor's use of these expressions when she was dealing with the so-called pattern of injuries was calculated to undermine the ruling on admissibility, which prevented any of the experts from using descriptors of that kind. Were the evidence to have been admitted as coincidence evidence, a submission might legitimately have been mounted to the effect that the degree of similarity in the number of wounds (and perhaps other features of similarity) were such as to render it improbable that Christopher killed his parents and that the applicant killed Christopher in the one contextually-related killing.

326 There remains the question whether it was permissible for the Crown Prosecutor to submit to the jury that the wounds were strikingly similar in pattern and number, independently of any reliance upon the expert opinion evidence. On one view, the Crown Prosecutor was entitled to submit that the jury could apply their commonsense and ordinary experience to find that it was improbable that more than one stabber inflicted the wounds to each of the deceased. The grouping of the wounds and the number of wounds found on each body were facts that the jury were entitled to weigh in the balance and which did not need to be adduced through the opinion of an expert. Indeed, absent a field of specialised knowledge in relation to the similarity of stab wounds in multiple homicides, it is difficult to see how, if at all, the opinion of an expert on the question whether there was more than one stabber could be any more valid than that of the jury.

327 However, the Crown submission that a single stabber inflicted the wounds was an invitation to the jury to engage in coincidence reasoning. This was so even if the submission was based simply on the evidence of the number and grouping of wounds and not on opinion evidence. The Crown effectively asked the jury to find that it was improbable that stabbings which exhibited such similarities were coincidentally, or "just so happened", to be inflicted by more than one person. The coincidence reasoning involved was manifest in the Crown Prosecutor's closing address. The issue of coincidence reasoning was not raised at the trial, nor was it argued on the appeal. Nevertheless, it seems to us that the Crown Prosecutor's use of the evidence of similarity ought to have complied with the coincidence rule set out in s 98 of the Evidence Act, which provides:

"(1) Evidence that 2 or more events occurred is not admissible to prove that a person did a particular act or had a particular state of mind on the basis that, having regard to any similarities in the events or the circumstances in which they occurred, or any similarities in both the events and the circumstances in which they occurred, it is improbable that the events occurred coincidentally unless:

(a) the party seeking to adduce the evidence gave reasonable notice in writing to each other party of the party's intention to adduce the evidence, and

(b) the court thinks that the evidence will, either by itself or having regard to other evidence adduced or to be adduced by the party seeking to adduce the evidence, have significant probative value.

Note: One of the events referred to in subsection (1) may be an event the occurrence of which is a fact in issue in the proceeding.

(2) Subsection (1) (a) does not apply if:

(a) the evidence is adduced in accordance with any directions made by the court under section 100, or

(b) the evidence is adduced to explain or contradict coincidence evidence adduced by another party.

Note: Other provisions of this Act, or of other laws, may operate as exceptions to the coincidence rule."

328 The identification of the proposed use of the similarity evidence as falling within the coincidence rule is not purely academic. The coincidence rule requires that the court admitting the coincidence evidence be satisfied that the evidence has, either by itself or in conjunction with other evidence, "significant probative value". The coincidence evidence had probative value in the sense that it could have affected the jury's assessment of whether one person stabbed each of the deceased. Proof of that fact would have strongly supported the Crown case. But the question whether the evidence had significant probative value was never explored at the trial. Relevant to that question is the varying number of stab wounds; the fact that most stabbings are to the chest area; and the consideration that some of the similarities in the pattern of wounding may be accounted for by the use of the same knife, as distinct from their infliction by the same attacker. For the purposes of this appeal, it is not necessary to decide whether the evidence had significant probative value. We note, however, that that question may fall to be determined at a subsequent trial.

329 We do not consider that the failure of defence counsel to seek the trial judge's intervention after the Crown Prosecutor's address is fatal to the applicant raising these concerns on the appeal. Although Ground 12 complains about impropriety in the Crown Prosecutor's address without reliance on these aspects, the address must be considered as a whole when it is said to have resulted in unfairness to an accused productive of a miscarriage of justice (see Causevic v R [2008] NSWCCA 238; 190 A Crim R 416 at [6]). We will return to consider whether the Crown Prosecutor's address on the similarity of wounds was in breach of established standards of fairness and propriety when considering Ground 12.

Was the evidence from Dr Culliford, Dr Cala and Dr Lawrence regarding the similarity or the pattern of stab wounds wrongly admitted as opinion evidence?

330 In Dasreef Pty Ltd v Hawchar [2011] HCA 21; (2011) 243 CLR 588 the High Court reviewed the operation of the opinion evidence rule in s 76 of the Evidence Act and the exceptions to it, most notably s 79. Those sections provide:

"76 The opinion rule

(1) Evidence of an opinion is not admissible to prove the existence of a fact about the existence of which the opinion was expressed.

(2) ...

79 Exception: opinions based on specialised knowledge

(1) If a person has specialised knowledge based on the person's training, study or experience, the opinion rule does not apply to evidence of an opinion of that person that is wholly or substantially based on that knowledge.

(2) ..."

331 For the purposes of applying s 76 there is a clear distinction between evidence of "fact" and evidence of "opinion". As Kirby J recognised in Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650 at [52] the distinction is one of degree rather than of kind with difficulty of classification arising from the fact that, in one sense, all evidence is of the opinion of the deponent.

332 On the appeal it was not suggested that the evidence under challenge was evidence of a fact. Even if Dr Lawrence's observation when he first prepared the transparencies that the grouping of the wounds to the chest cavity appeared to be similar is capable of being interpreted as an opinion, it might equally be a conclusion or inference of fact that he drew after preparation of the transparencies. At trial it also seemed to be accepted by defence counsel that the evidence from each of the Crown experts that there was a similarity in the wounds (which by the time of the trial became a similarity in the pattern of the wounds) was opinion evidence which, by operation of s 76, was not admissible to prove the existence of the fact of similarity or a wound pattern, thereby invoking the operation of s 79.

333 In Dasreef the Court emphasised the importance of the party tendering the evidence making clear the basis upon which evidence of opinion is relevant, and in particular what finding the tribunal of fact will be asked to make based upon that evidence. Their Honours held at [31]:

"In considering the operation of s 79(1) it is thus necessary to identify why the evidence is relevant: why it is "evidence that, if it were accepted, could rationally affect (directly or indirectly) the assessment of the probability of the existence of a fact in issue in the proceeding". That requires identification of the fact in issue that the party tendering the evidence asserts the opinion proves or assists in proving."

Was the evidence of similarity relevant?

334 There can be no doubt that the Crown intended to rely, and to rely heavily, upon expert opinion from three medically trained forensic practitioners that there was a discernible similarity in the pattern of wounds from which the jury could conclude that the applicant inflicted all the wounds, and that the test of relevance in s 55 of the Evidence Act was met in that way. The Crown Prosecutor's closing address is the clearest articulation of the way the Crown ultimately used the evidence for that purpose.

335 Although it was the view of both Dr Cala and Dr Culliford in their reports to investigating police in 1999 and 2005 that the similar pattern of wounding supported the probability of there being only one killer, and despite the fact that the Crown had intended to call both doctors to give evidence of that opinion, the trial judge upheld defence counsel's objection to that aspect of their evidence. The question that remains is whether that limitation deprived their evidence of relevance for the purposes of admissibility. The trial judge was satisfied that it did not. He held that their opinion about the similarity in the pattern of wounds was relevant and sourced from their expertise, it being a question for the jury whether that evidence, taken together with other evidence in the Crown case, supported the inference that the applicant killed all three deceased and was therefore guilty of the murder of his parents. The trial judge must be taken to have been satisfied that if the jury accepted that there were similar wound patterns, that evidence had the capacity to affect their assessment of whether the applicant inflicted all of the injuries on each of the deceased and was therefore guilty of the murder of his parents. This is reflected in the trial judge's directions to the jury summarised at [250].

336 The trial judge's finding that the evidence of similarity between the wounds was relevant in this way obscured what we regard as a critical question, essential to resolve if the evidence was to meet the dual criteria for admission as opinion evidence under s 79. Neither counsel nor the trial judge grappled with this question on the voir dire. As we see it, that question is whether what Dr Culliford and Dr Cala identified as features of similarity in some of the wounds inflicted on the deceased were capable of supporting their further opinion (implicit at the time of the voir dire and explicit at the trial) that the injuries constituted a pattern of some discernible kind. The related question is whether that opinion was based wholly or substantially upon their specialised knowledge such as might rationally affect proof of the assessment by the jury of whether there was one stabber or two.

337 The question whether the trial judge was in error in admitting the evidence of the three Crown experts is complicated by the fact that Dr Lawrence was not called to give evidence on the voir dire as to the significance, from his perspective as the forensic pathologist who performed the autopsies, of the appearance of similarity in the grouped wounds when he prepared the transparencies in 2000. If the Crown Prosecutor made the deliberate decision to rely upon the other experts for that purpose, aware that it was Dr Lawrence's evidence at the inquest that he was unsure of the significance of the appearance of similarity, and that such similarity as he observed was of no probative value in any event in proving whether there was one assailant or two, he would have failed in his duty of fairness and candour. That question does not fall to be resolved on the appeal, there being no argument directed to it.

338 By the time of the second trial Dr Lawrence's evidence concerning what he observed on preparation of the transparencies was much more expansive and was given without any reservation or qualification as to its significance (as to which see [310]-[311]). This evidence was also given without objection. This has given rise to some problems in dealing with Ground 8, which has been framed to include a challenge to the admissibility of Dr Lawrence's evidence. For this reason a grant of leave under rule 4 is required in order to consider the admissibility of Dr Lawrence's evidence.

Was the evidence of similarity admissible as opinion evidence?

339 In Dasreef the High Court articulated the test in s 79 as follows (at [32]):

"To be admissible under s 79(1) the evidence that is tendered must satisfy two criteria. The first is that the witness who gives the evidence 'has specialised knowledge based on the person's training, study or experience'; the second is that the opinion expressed in evidence by the witness 'is wholly or substantially based on that knowledge'..."

340 In the course of the voir dire the trial judge accepted that there was no organised body of knowledge deriving from recognised principles of medical science which would enable either Dr Cala or Dr Culliford to express an opinion as to whether the wounds sustained by the deceased in this case were inflicted by one killer or more. He apparently accepted Professor Cordner's evidence on that issue. That being the case, the criteria for admission of what remained of the evidence translated into a requirement that Dr Cala or Dr Culliford be shown to have specialised knowledge in the interpretation of stab wounds based upon their training, study or experience as forensic practitioners, and that their opinion that there were discernible similarities in the wounds, manifesting as a discernible and distinct pattern in the wounds sustained by the deceased, was based wholly or substantially upon that knowledge.

341 It was not suggested that either doctor had undertaken any study of, or had utilised any body of research into, the characteristics or patterns of stab wounds in multiple homicides (or single homicides for that matter) from which, by comparison with recognised standards, they might have formed the opinion that the wounds inflicted in this case were in a pattern (or were sufficiently similar to create a pattern). Neither was there any training to which either Dr Cala or Dr Culliford referred that would have supported their opinion that there was a discernible and distinct pattern in the wounds. Accordingly, the only source of knowledge upon which they could have reliably based their opinion was their experience - in Dr Culliford's case, her experience as a forensic physician, and in Dr Cala's case, his experience as a forensic pathologist.

342 Dr Cala gave evidence on the voir dire of having attended and conducted post-mortems involving stab wounds. He did not, however, claim to have had any exposure to a multiple homicide where multiple stab wounds were inflicted to provide a basis for the opinion that the features of similarity in the wounds inflicted on each of the deceased constituted a pattern, even accounting for the dissimilarities. Dr Culliford's experience was limited to that of a forensic physician either attending at the scenes of crime or assisting forensic pathologists during a post-mortem. Such exposure as she had to the deaths of more than one person in a homicide by stabbing did not support her claim that there was, by comparison, a discernible pattern in this case. Indeed, she disclaimed any basis for undertaking that comparison.

343 The trial judge did not consider the question of Dr Cala's experience at all and regarded Dr Culliford's experience as adequate to entitle her to express an opinion about the appearance of similarity.

344 It is well-recognised that there is a wide variety of circumstances in which stab wounds might be inflicted, even where death is intended. In addition, the fluid dynamic involved in a fatal confrontation between a victim (or victims) and an assailant in a stabbing incident (even where the question of intent is at large) is frequently a relevant matter to be taken into account when a court is asked to determine whether the opinion of a forensic practitioner about how the wounds were inflicted and the relative positioning of the assailant and victim is admissible. That did not appear to carry weight in the trial judge's consideration of the question of admissibility in this case although, to the extent that it was relied upon by counsel in submissions, it cannot be said that the trial judge ignored it altogether. That said, the fact that the expertise of the forensic pathologist/forensic physician acquired experientially might be imprecise, requiring that great caution be exercised in admitting the evidence, is no barrier to admissibility under the Evidence Act (see Velevski v R [2002] HCA 4; (2002) 76 ALJR 402 at [153]-[158]). However, as Gleeson CJ observed in HG v The Queen [1999] HCA 2; (1999) 197 CLR 414 at [44]:

"...Experts who venture "opinions", (sometimes merely their own inference of fact), outside their field of specialised knowledge may invest those opinions with a spurious appearance of authority, and legitimate processes of fact-finding may be subverted..."

345 In the absence of any evidence of any relevant experience of fatal stab wounds inflicted by one killer on multiple victims, we are satisfied that Dr Cala's evidence failed to meet the second limb of the test for admissibility under s 79 of the Evidence Act. We are also of the view that in so far as Dr Culliford's evidence is concerned, her experience was insufficient to support her opinion that the wounds on the deceased showed sufficient similarity to constitute a pattern and, for this reason, the trial judge's decision to admit the evidence as opinion evidence under s 79 of the Evidence Act was in error.

346 Even were the evidence admissible under s 79, we are nevertheless satisfied that it ought to have been rejected in the exercise of discretion under s 137 of the Evidence Act. Properly analysed, the evidence of the Crown experts that the wounds "appeared similar" was of little probative value, while the risk that the jury would impermissibly use the collective force of the evidence from the three Crown witnesses to infer that the similarity created a pattern, which was explicable only if the applicant was the perpetrator, was overwhelming. This was a risk that the trial judge's directions could not protect against.

347 The trial judge directed the jury in conventional terms to be cautious before departing from expert scientific opinion in relation to any aspects of the evidence where the Crown and the defence were in disagreement, and which might be contrary to what the jury thought might be the case by the application of commonsense and experience. As regards the evidence concerning the similarity of wounds, he gave these further directions:

"...The scientific evidence is that there is no scientific basis for assuming that people who have the same type of wounds, similar wounds, had the same perpetrator. But you put that in context. It may be the case that, if there were three people killed, one each week, over different parts of the city, that had the same wounds to them generally as these three persons had, that you could not on the scientific basis conclude that it was the same person who inflicted those wounds. But you have a look at this question in this particular context, that is, of three people in the one family being killed roughly at the same point in time, in the same circumstances, and you ask yourself: Is there anything about these particular facts and circumstances that might indicate that the general scientific view about whether you can reason from the number of and type of wounds that they were the same perpetrator, applies in this particular case. I am not saying whether it does or whether it does not, but you are not bound by scientific learning, scientific experience and scientific evaluation. You take it into account.

In some cases, as I say, such as psychiatric illness, you are bound by it because there is nothing upon which you can depart from it. But in this particular matter, particularly when we come to the question of the wounds, their similarity or otherwise and what that means, you are not bound by the scientific opinion that you could not conclude from that, and other facts, seeing it in the circumstances surrounding it, that it was the same perpetrator. And this is where we come to deal with what are circumstances and what onus of proof there is upon the Crown in relation to circumstances, but we will talk about that shortly."

348 In directing the jury to use their commonsense in this way the trial judge gave uncritical endorsement to another aspect of the Crown Prosecutor's closing address, which we do not consider was properly grounded in the evidence or open as a matter of argument given the way the trial was conducted:

".....In effect, the Crown says to you, "As a matter of common sense and of your experience in life, the accused's account can't be true, it just doesn't happen that way". That goes to the fact that the three victims had what the Crown says are remarkably similar wounds, so similar it can't be coincidental."

349 We are also satisfied that the Crown Prosecutor's repeated use of various adjectives to describe the similarity in the wounds, which was a submission unsupported by an application for admission of the evidence under s 98 of the Evidence Act, exceeded the legitimate bounds of a closing argument by a Crown Prosecutor. Her approach is the more egregious where the trial judge had not admitted the evidence under s 98 of the Evidence Act and had expressly prevented the expert witnesses from using these very words, or words like them, when describing the extent of similarity in the pattern of injury about which they were permitted to give evidence.

350 Were the trial judge invited to rule upon whether the Crown would be permitted to address the jury in terms that there was an "extraordinary similarity" or "remarkable similarity" in the pattern of stab wounds; that it was an "extraordinary coincidence" that two killers would stab in the same way; or that there was coincidence in the "overkill" by the infliction of multiple stab wounds to vital organs in each of the deceased, this being evidence of a pattern of killing, he should have ruled that these submissions were not permissible unless the evidence of similarity was admitted as coincidence evidence under s 98.

Was the Crown's failure to call Professor Cordner as a witness concerning the similarity of the pattern of stab wounds productive of a miscarriage of justice?

351 The applicant's solicitors obtained a report from Professor Cordner for the purposes of the appeal.

352 In addition to various materials he had already in his possession (presumably as a result of the report he provided to police in 1999), Professor Cordner was provided with the three autopsy reports prepared by Dr Lawrence and related diagrams; the evidence of Dr Culliford, Dr Cala and Dr Lawrence at the second trial; and the complete transcript of the Crown's closing address. His stated objective was to evaluate the proposition contended for by the Crown at trial that there was a similar pattern in the wounds sustained by each of the deceased of such significance that it supported the conclusion that the applicant killed all three and was therefore guilty of the murders of his parents.

353 Given what Professor Cordner identified as the paucity of literature to inform the matters under inquiry (and none to support the assumption inherent in the Crown case theory that a person will stab multiple victims in the one incident in the same way such as to produce a pattern of injury), he undertook his own research.

354 In summary, he reported that after reviewing the materials provided to him, and after considering the results of his independent research, there was no distinctive quality to the pattern of stab wounds revealed on autopsy to support a description of them as similar, still less to support a conclusion that one person was responsible for inflicting them, and that the prosecutor's submission to the jury to the contrary was wholly mistaken.

355 He disagreed with the views of Dr Culliford, Dr Cala and Dr Lawrence that there were sufficient features of similarity such that the dissimilarities were rendered insignificant (in particular between Mr Gilham and Mrs Gilham, who were on any view killed by the same person whether it be Christopher or the applicant), thereby justifying the identification of the wounds as constituting a "pattern of injury". In our view, he correctly identified the issue as one relating to the pattern of injury and said:

"The issue in this case relates to 'the pattern of injury'. The Shorter OED defines pattern, in the sense being used here, as "a typical instance, a significant example". The use of the word begs the question: the pattern of injuries (said to be similar between the three cases) is a typical instance of what? In my view, it is a typical instance of multiple stab wounds. What elements of the stab wounds are being relied upon to establish the pattern which is so similar between all three deceased that the inescapable conclusion is that there is one killer?"

356 He then identified the elements relied upon by each of the three experts to establish the pattern, which was relied upon to support the Crown case that there was only one killer, as follows:

(a) the number or multiplicity of injuries;

(b) the location of the injuries;

(c) the locations free of injury; and

(d) the grouping or clustering of the injuries (that is, the high number of injuries in one location).

357 Professor Cordner made a comparison of the injuries to Helen, Stephen and Christopher Gilham, represented in the following table:

	Total number of stab wounds	Stab wounds to front of chest (% total)	Stab wounds to back of chest (% total)	Dominant clustering	Number in cluster	% of total wounds in cluster	Wounds not to chest
Helen	15	1 (7%)	13 (87%)	Upper left and upper right back	9	60%	Neck, L forearm
Stephen	28	15 (54%)	7 (25%)	Mid upper chest and central chest	13	46%	Face, neck, L upper arm
Christopher	17	14 (82%)	3 (18%)	Mid upper chest and central chest	14	82%	Hand

358 He then undertook a comparative analysis of the injuries to each of the deceased against these data and concluded that the only similarities (and even then only between Mr Gilham and Mrs Gilham and at a level of generality) were the multiple stab wounds to the front and back of the chest, with some of the wounds being grouped and parallel to each other, and that the only conclusion capable of flowing from that observation was the likelihood that the deceased were not moving when the grouped stab wounds were inflicted. (The same conclusion follows from Christopher's grouped but horizontally aligned wounds to the chest.)

359 Data relating to the stab and incised wounds in 9 incidents of multiple homicides in New South Wales and Victoria between 2002 and 2009 (involving 20 deceased in total, with 9 incidents involving two deceased and one incident involving four deceased) was extracted from the National Coroners Information System and analysed by Professor Cordner.

360 He considered that some of the conclusions able to be drawn from that data are as follows:

"5.2 The data derived from incidents in NSW and Victoria of multiple homicides where death is due to stabbing does not support the assertion that there will be similarity in the pattern of injury (ie in the number and distribution of the stab wounds) in the victims of one incident of multiple homicide.

...

5.4 The data supports the contention that in multiple homicides where death is due to stabbing, it is often the case that the abdomen will not contain stab wounds, and that there is a variety of distributions of wounds to other parts of the body."

361 Professor Cordner also found that in a sequential series of 50 Victorian homicide cases displayed when the National Coroners Information System was searched for "multiple stab wounds" (a search which would exclude single stab wounds and probably some cases with small numbers of wounds), 31 cases (62 percent) involved more than 10 stab wounds excluding injuries to the hands and forearms. In a sequential series of 165 homicidal stabbings from the Victorian Institute of Forensic Medicine database (which included cases of single stab wounds), 48 (29 percent) involved ten or more stab wounds and incised injuries, again excluding incised injuries to the hands and the forearms. Obviously, if injuries to the hands and forearms were included, the number of stab wounds would be higher.

362 Professor Cordner also pointed out that this data directly contradicts what he described as Dr Cala's unsubstantiated, and almost certainly wrong, evidence at the trial that stab wounds rarely exceed ten for any one victim, evidence which was then used by the prosecutor to advance the following submission:

"You might think as a matter of logic ... the more stab wounds there are the fewer cases that have many. You heard Dr Cala say that it's very very unusual for someone to be stabbed more than 10 times because if you're in the centre of the chest, you only need - you certainly only need one stab wound to kill someone, and the greater the number, the fewer cases there are... There was overkill [in this case], if I can suggest that to you, ladies and gentlemen, so many more wounds needed than was necessary to kill any one of them, and that is an extraordinary similarity." (Emphasis added)

363 By reference to the same data, Professor Cordner says that the prosecutor's "overkill" submission is equally misconceived and certainly not an "extraordinary similarity" or an "inconceivable coincidence" that links the three deaths to the applicant.

364 Professor Cordner also suggested that the research data shows that the prosecutor's invitation to the jury to use their commonsense was misplaced. She said in her closing address:

"We have heard of many stab wounds in our daily lives... we just don't hear of this additional infliction of all these extra wounds after the fatal wounds, the Crown would submit to you. It's just so extraordinarily different from our usual experience, as a matter of common sense, from what you would ever see.

...

Ladies and gentlemen, it is inconceivable that a coincidence like that could happen, this overkill being applied to the parents. This man does not know it but by coincidence does exactly as his brother does, a brother with whom, as he says, he was not very close."

365 Professor Cordner reviewed 40 cases of homicide in Victoria in which death was due to stabbing, of which there were 10 or more stab and incised wounds. 35 of the cases occurred between 2000 and 2010 and were listed on the National Coroners Information System, and five were collected manually before that time. From that data, several observations were made.

366 In a sample of 40 homicides in Victoria where the deceased had more than 10 stab wounds:

• 8/40 had more than 28 stab wounds in total.
• 16/40 had between 16 and 28 wounds in total.
• 16/40 had between 10 and 15 stab wounds in total.
• Stab wounds to the front (36/40) and back (27/40) of the chest are common: 25/40 had stab wounds to both the front and back of the chest.
• The 36 victims with stab wounds to the front of the chest had a total of 863 stab wounds, averaging 24 each. 285 (or 33 per cent) of these stab wounds were to the front and side of the chest, averaging 7.9 each.
• The 27 victims with stab wounds to the back had a total of 763 stab wounds averaging 28 each. 214 (or 28 per cent) of these stab wounds were to the back of the chest, averaging 7.9 (sic) each.
• The 25 victims with stab wounds to both the front and the back had a total of 695 stab wounds, averaging 28 each. The victims had 218 wounds to the front and 162 to the back; that is, 380 wounds to the chest, representing 55 per cent of the wounds in total.
• 16/40 cases had no stab wounds to the abdomen; 14/40 had no stab wounds to the head (eg the face); 16/40 had no stab wounds to the arms; 24/40 had no stab wounds to the legs; and 11/40 had no stab wounds to the neck.
• 40/40 had some injuries elsewhere.

367 Using the same data, but including the autopsy records of the three deceased for comparative purposes, Professor Cordner tabulated the location of the dominant clustering of injury (defined as the area where at least 25 percent of stab injuries were sustained), the number of wounds in each cluster as a percentage of the total injuries sustained, and wounds within the group or cluster that were parallel to each other, assessed as a percentage of the total number of wounds in each cluster. Fourteen of the 40 cases were not supported by photographs such as to enable any detailed analysis. With that limitation in mind, he concluded that the grouping of the injuries sustained by each of the deceased in the chest cavity, and their relative location to one another, with two other wounds in close proximity, placed them within the range of injuries one would expect of homicidal deaths from multiple stab wounds generally, such that similarity of the kind the Crown contended for (that is, a pattern of wounds of notable similarity) could not be accurately claimed.

368 In Professor Cordner's view, this must also be set against the prosecutor's submission to the jury:

"So in the case of Stephen Gilham, ladies and gentlemen, 16 grouped stab wounds to the front; in the case of Helen Gilham, 14 groups stab wounds to the back of the chest and in the case of Christopher Gilham, 14 grouped stab wounds to the front. The Crown says that is an extraordinary coincidence. It is also extraordinarily coincidental the extraordinarily similar way they must have been inflicted. To get that sort of precision you'd have to be down in a position kneeling or squatting to get the precision to make perfectly or almost perfectly parallel close wounds into the middle of the body of those victims."

369 Professor Cordner regarded Dr Culliford's opinion (extracted at [305]-[306]) that there was similarity evident from the grouped wounds on the opposite side of the chests of the deceased, this being reflective of a pattern, as being of no weight at all. Professor Cordner said that view should have been disregarded for two reasons. First, Dr Culliford's assertion that there was "a" single wound to the heart was contradicted by the autopsy reports. Christopher and Mr Gilham had, respectively, 14 and 15 and 7 and 3 wounds to the back and front extending to both the heart and lungs. Secondly, it was pure speculation for Dr Culliford to elevate the mere potential for injuries to have been inflicted in a particular order to the level of an underlying pattern. In Professor Cordner's view, the way the prosecutor used this evidence in further support of the Crown case that there was only one killer was also misleading.

370 Professor Cordner was not cross-examined by the Crown on the appeal to suggest that his criticisms of the evidence of Drs Cala and Culliford was unfounded, or that the Crown Prosecutor's reliance upon their evidence was in fact justified from a scientific perspective. This may be explained, in part, by Dr Lawrence's evidence on the appeal, which in material respects did not challenge Professor Cordner's findings or his conclusions. Professor Cordner was cross-examined about his approach to the preparation of his report to police in 1999, and his evidence at the inquest, with a view to defending the Crown's decision not to call him as a witness at trial. This evidence will be discussed when we are considering whether there has been a resulting miscarriage of justice.

Dr Lawrence's evidence on the appeal

371 Dr Lawrence was invited by the Crown to review the report prepared by Professor Cordner for the purposes of the appeal. He did so in an affidavit dated 4 February 2011 and a further affidavit dated 1 September 2011. He referred repeatedly to "the pattern of stab wounds" in both affidavits.

372 He emphasised that the comparison of the stab wounds by the use of transparencies was a visual comparison only, as it was not possible to compare photographs of the bodies. This was because of the extensive superficial burns to the bodies of Mr Gilham and Mrs Gilham, which obscured the stab wounds visible in the photographs. He said for this reason that the knife wound tracks on the two deceased were grouped, but he could not describe them individually. He agreed with Professor Cordner that this circumstance could impose an apparent similarity on the wound tracks when the wounds were reproduced on the transparencies.

373 He said that he prepared the diagrams in this way in 2000, having recently returned from working for the United Nations International Criminal Tribunal for the former Yugoslavia, where, as part of his report to the Tribunal about the victims of a massacre, he prepared diagrams showing the location of all gunshot wounds sustained by the deceased on one single body diagram. This involved approximately 60 to 300 bodies across a number of mass graves. He said this process revealed some apparent "patterns of [gunshot] injury", which we take to mean similar parts of the body at which shots were fired. He said because of the questions raised during the second inquest he used the same method with the stab wounds to the deceased. He said he was surprised at what was revealed but he did not at that time venture a view about its significance.

374 He went on to say he no longer knows "what the diagrams mean" and that his interpretation of the pattern was not based on science. When cross-examined on the appeal he said that he was really reporting an observation that there was a "similarity simpliciter" between the wounds and nothing else. He said that he did not carry out any further research at the time of preparing the transparencies or, at the time that he reported his observation to the prosecuting authorities or before giving his evidence.

375 He set out in his affidavit his evidence at the inquest on the issue:

"I have indicated in the past that while I thought the patterns were similar, I didn't know what the significance of that observation was. At the inquest...BREWER: "Well what do you say about the proposition that the pattern of wounds is not a sufficient basis for concluding that one person killed them and it's neutral"? LAWRENCE: I think it's the crux of the matter isn't it? I don't completely agree with that. I have never seen a group of wounds where I've had the - you look at them and they are very similar. I agree it is not - it's of difficult weight to evaluate that similarity. But I honestly cannot recall a group of wounds that - a group of three sets of wounds that look this much the same. And I have dealt with other triple homicides before. But is it of probative value? I don't think so. It certainly impressed me but does it prove that there was one assailant or two? I don't think so." This is still my position."

376 This evidence was not before the trial judge on the voir dire.

377 He was then asked to agree or disagree with the passage in Professor Cordner's report dated July 1999 (set out in full at paragraph [263]) to the effect that it was unsafe to conclude that the distribution and number of stab wounds indicated that there was one assailant, or that it was even a probable factor to be weighed in the balance, as distinct from the number and distribution of the wounds being entirely neutral on the question. He agreed with Professor Cordner's views.

378 He also agreed with Professor Cordner that there were a number of features of the comparative analysis of the stab wounds of the deceased which might increase the likelihood of the injuries appearing similar, including that the same knife caused all of the wounds; that the chest is a common site for the infliction of stab wounds, and therefore a clustering of wounds on the chest may have less significance than those inflicted on other locations on the body; and that Mr and Mrs Gilham's wounds only appear similar if the posterior body diagram of Mrs Gilham is compared with the anterior body diagram of Mr Gilham (the same, so it seems to us, must be said of the anterior body diagram of Christopher).

379 In his affidavit dated September 2011, he agreed with Professor Cordner that three cases of homicide with multiple stab wounds in the chest "could arise by chance".

380 In his report dated February 2011, Dr Lawrence said he did not regard the number of stab wounds as "highly significant". In his later affidavit he said his opinion regarding the similarity of the pattern of stab wounds was not based on the number of wounds. This is in direct conflict with his evidence at trial.

Dr Cala's evidence on the appeal

381 Unlike Dr Lawrence, after reading Professor Cordner's report, Dr Cala was not moved to moderate his views. Instead he cautioned against using statistics. It would seem that despite Dr Lawrence's revised opinion as to the significance of the comparative exercise using the transparencies, Dr Cala remained of the view that there was a "remarkable similarity to the pattern of injury that exists in all three cases with respect to the number of stab wounds". He did not consider that the differing alignment in stab wounds between Mr Gilham and Mrs Gilham on the one hand and Christopher on the other counted against there being an underlying pattern, despite the dynamic situation at play in a stabbing incident where the assailant and victim are changing positions relative to each other.

382 While Dr Cala conceded that the statistics produced by Professor Cordner were at odds with Dr Lawrence's observation of similarity (which we take to mean his opinion also), he claims that this does not prove that his (Dr Cala's) opinion is wrong.

Was the Crown obliged to call Professor Cordner in the Crown case?

383 The primary obligation imposed on a Crown Prosecutor in discharge of the function which it is the prosecutor's to perform in a criminal trial finds expression in the Director's Guidelines and the rules of professional practice enshrined in the New South Wales Barristers' Rules, as well as the principles relating to the calling or non-calling of witnesses in the collected authorities. That obligation is to act fairly by ensuring that the Crown case is presented with fairness to the accused.

384 While it is the Crown Prosecutor alone who bears the responsibility for deciding whether a person will be called as a witness for the Crown, that decision must be made consistently with the overriding obligation to act fairly and in strict accordance with what was then Rule 62 of the Barristers' Rules. This rule was binding on Crown Prosecutors as legal practitioners (as the identical Rule 82 is currently) by virtue of s 711 of the Legal Profession Act 2004. That rule provided:

"A prosecutor must fairly assist the court to arrive at the truth, must seek impartially to have the whole of the relevant evidence placed intelligibly before the court, and must seek to assist the court with adequate submissions of law to enable the law properly to be applied to the facts." (Emphasis added)

385 Section 13 of the Director of Public Prosecutions Act 1986 empowers the Director to furnish guidelines to Crown Prosecutors, which guidelines bind them in the prosecution of offences. The Prosecution Guidelines of the Office of the Director of Public Prosecutions for New South Wales, issued in June 2007 and in force at the time of both trials, provide as follows:

"The prosecution should generally call all apparently credible witnesses whose evidence is admissible and essential to the complete unfolding of the prosecution case or is otherwise material to the proceedings. Unchallenged evidence that is merely repetitious should not be called unless that witness is requested by the accused.

If a decision is made not to call evidence from a material witness where there are identifiable circumstances clearly establishing that his or her evidence is unreliable, the prosecution, where the accused requests that the witness be called and where appropriate, should assist the accused to call such a witness by making him or her available or, in some cases, call the witness for the purpose of making him or her available for cross-examination without adducing relevant evidence in chief (see Rule A.66B of the [Barristers'] Rules - Appendix B)."

386 Rule 66B of the Barristers' Rules then in force provided:

"A prosecutor must call as part of the prosecution's case all witnesses:

(a) whose testimony is admissible and necessary for the presentation of the whole picture;

(b) whose testimony provides reasonable grounds for the prosecutor to believe that it could provide admissible evidence relevant to any matter in issue;

(c) whose testimony or statements were used in the course of any committal proceedings; and

(d) from whom statements have been obtained in the preparation or conduct of the prosecution's case;

...

provided that:

(h) the prosecutor is not obliged to call evidence from a particular witness, who would otherwise fall within (a)-(d), if the prosecutor believes on reasonable grounds that the testimony of that witness is plainly unreliable by reason of the witness being in the camp of the accused; and

(i) the prosecutor must inform the opponent as soon as practicable of the identity of any witness whom the prosecutor intends not to call on any ground within (f), (g) and (h), together with the grounds on which the prosecutor has reached that decision." (Emphasis added)

387 The Director's Guidelines relevantly provide:

"Mere inconsistency of the testimony of a witness with the prosecution case is not, of itself, grounds for refusing to call the witness. A decision not to call a witness otherwise reasonably to be expected to be called should be notified to the accused a reasonable time before the commencement of the trial, together with a general indication of the reason for the decision (eg The witness is not available or not accepted as a witness of truth). In some circumstances, the public interest may require that no reasons be given. Where practicable the prosecution should confer with the witness before making a decision not to call the witness."

388 A decision by a prosecutor not to call a particular witness will constitute a ground for setting aside a conviction if, when viewed against the conduct of the trial taken as a whole, it could be seen to give rise to a miscarriage of justice (see R v Apostilides [1984] HCA 38; (1984) 154 CLR 563 at 575). As the High Court made clear in Apostilides, reference to misconduct was omitted in this connection to broaden the approach to the question of miscarriage so as to focus on the consequences, objectively perceived, that the failure to call a witness has had on the course of the trial and its outcome. The Court went on to say at [577]:

"...It is not necessary to postulate misconduct of the prosecutor as an essential condition precedent to a miscarriage of justice. No doubt in the great majority of cases of this kind an appellate tribunal which finds a miscarriage of justice to have occurred will trace that miscarriage to a wrong exercise of judgment by the prosecutor which led to the witness not being called. In cases where there has been no error of judgment there will be less likelihood of a miscarriage resulting from the failure to call the witness..."

389 In Whitehorn v The Queen [1983] HCA 42; (1983) 152 CLR 657 Deane J said at 663:

"Under the adversary system which operates in a criminal trial in this country, it is for the Crown and not the judge to determine what witnesses are called by the Crown. That is not to say that the Crown is entitled to adopt the approach that it will call only those witnesses whose evidence will assist in obtaining a conviction. Prosecuting counsel in a criminal trial represents the State. The accused, the court and the community are entitled to expect that, in performing his function of presenting the case against an accused, he will act with fairness and detachment and always with the objectives of establishing the whole truth in accordance with the procedures and standards which the law requires to be observed and of helping to ensure that the accused's trial is a fair one...

The observance of traditional considerations of fairness requires that prosecuting counsel refrain from deciding whether to call a material witness by reference to tactical considerations. Whether or not their names appear on the back of the indictment or information, all witnesses whose testimony is necessary for the presentation of the whole picture, to the extent that it can be presented by admissible and available evidence, should be called by the Crown unless valid reason exists for refraining from calling a particular witness or witnesses, such as that the interests of justice would be prejudiced rather than served by the calling of an unduly large number of witnesses to establish a particular point..."

390 These principles were considered again by this Court in R v Kneebone [1999] NSWCCA 279; (1999) 47 NSWLR 450. In the judgments of Smart AJ and Greg James J (with whose judgments Spigelman CJ agreed) the circumstances in which a decision not to call a witness would or might be defensible were identified. Greg James J made some observations as to how a prosecutor might ascertain a witness' unreliability at [49]-[51]:

"Since both experience and logic confirm that merely because a witness' evidence is inconsistent with or contradicts other evidence, it need not be untrue, it is necessary that a prosecutor whose decision is under examination be able to point to identifiable factors which can justify a decision not to call a material witness on the ground of unreliability: see R v Apostilides (at 576); Director of Public Prosecutions Guidelines, at least if the suggestion of attempting to obtain an improper tactical advantage is to be avoided. It is therefore necessary for the prosecutor to take appropriate steps, including, where necessary interviewing witnesses to be able to form the opinion.

In reaching a view as to reliability, it is clear that it is not an adequate basis to conclude that the witness is unreliable, merely because the witness' account does not accord with some case theory which is attractive to the prosecutor. An approach, whereby the witness is not called at all or is left to the defence to call because the witness' evidence is seen as not fitting the prosecution's view of the case is likely to lead to a miscarriage of justice. R v Apostilides deals with the consequence of such an approach. A case theory should accord with the evidence. The prosecutor should not espouse a theory and tailor a case accordingly:R v Anderson (Hilton Bombing case) (1991) 53 A Crim R 421.

The advisability, if not necessity for a conference is reinforced by the decisions in Tran v Magistrates' Court of Victoria [1998] 4 VR 294; R v Armstrong [1998] 4 VR 533, where the prosecutor refused to interview the witness or to consider notes proffered by counsel for the defence indicating the substance of the evidence it was anticipated the witness would give; and R v O'Brien (1996) 66 SASR 396 particularly in the judgment of Doyle CJ (at 398-399). There the prosecutor had sensibly spoken to and proofed the witness before concluding that there was a clear allegiance on the part of the witness to the accused."

391 In Velevski v The Queen, Gleeson CJ and Hayne J observed that prosecutorial fairness does not require that there be a "head count" of experts holding differing opinions. That observation was made in the context of a complaint that the Crown had failed to obtain reports from qualified experts who, as the Crown was aware, favoured a murder/suicide hypothesis rather than the murder/murder hypothesis advanced by the Crown. The Crown also declined to call them as witnesses at trial. Gleeson CJ and Hayne J held at [47]:

"... In a case such as the present, where several experts were to be called to give evidence, the prosecution, not already being in possession of evidence from other expert witnesses, was not bound to seek it out by having the witness qualify himself or herself to form an opinion and then to call the witness to give evidence of the opinion that was formed." (Emphasis added)

392 Gummow and Callinan JJ were of the view that despite the fact that the applicant was fully apprised of the views of the experts who were not called, the Crown should have called them. In their Honours' view, however, there was no miscarriage of justice as the trial judge's directions operated to protect the accused against any disadvantage he might otherwise have suffered (at [176])

393 Where there is a question whether the prosecution has discharged its obligation of fairness in the context of calling expert evidence, particular considerations may intrude, not least that an expert witness can give evidence only by reference to facts that must be independently established (see Velevski at [47] (Gleeson CJ and Hayne J)). However, we do not consider (and the Crown does not submit) that has any bearing on this case, where each of the experts called by the Crown was asked to consider the materials produced at post-mortem and invited to offer an opinion about what Dr Lawrence reported as his findings, including, in particular, whether there was any similarity in the grouped injuries that might suggest that there was one perpetrator or two.

394 Unlike the situation in Velevski, in this case, Professor Cordner's report was solicited by the prosecuting authorities in July 1999. Also unlike Velevski, the decision not to call Professor Cordner was made because the Crown had assessed that he was "plainly unreliable." This assessment was made for the purposes of the first trial by one Crown Prosecutor and, it is open to infer, not reconsidered when another Crown Prosecutor appeared in the second trial. What is clear is that there was nothing to suggest that she gave her own consideration to whether to call Professor Cordner. It appears she uncritically adopted the approach of the first prosecutor - an approach which we consider was fundamentally flawed. The question for this Court is whether the decision that Professor Cordner not be called as a witness can be justified.

The sufficiency of the reasons Professor Cordner was not called to give evidence

395 When the applicant's solicitors requested that the Crown explain the basis for their view that Professor Cordner was a witness who was "plainly unreliable", they were informed by letter dated 30 January 2008 as follows:

"Professor Cordner has openly expressed that his opinions are based upon a complete acceptance of the accused's account of what happened (see for example paragraph A1 on page 5 of his report dated 28 July 1999).

'1. The account given to the Police by Jeffrey Gilham ... is in conflict with evidence of external witnesses (neighbours) as to the duration of his involvement in the episode.'

I do not regard it as my function to try and describe any defects in Jeffrey's story other than those which, if demonstrable, show that he killed his parents or that Christopher did not kill his parents. Other defects would simply be defects which could be mistakes or lies and which they are, and their significance, are generally outside the province of the forensic pathologist and are best evaluated by the decider of fact. I will try to steer clear of these and concentrate on matters closer to the core issues. In other words, I start from the premise that Jeffrey's account is given in good faith.

Professor Cordner's opinions are often argumentative (see for example paragraphs 4(ii), 5, 10). The Crown is of the view that Professor Cordner's views are at variance with the views of other experts it intends to call. The Crown is also of the view that the extent of Professor Cordner's limitations and the inadequacies of his report would only be disclosed during cross-examination by the Crown. The defence would have no interest in eliciting weaknesses in his methodologies and conclusions. Unlike civilian witnesses, the Crown does not believe that it is appropriate to call an expert whom it intends to discredit by obtaining leave to cross-examine under s 38 of the Evidence Act. Based upon the decision of the High Court in The Queen v Apostilides (1984) 154 CLR 563 the Crown will make Professor Cordner available for you to call."

It is necessary to consider each of the purported bases of unreliability in turn.

396 It is not suggested, nor could it be, that Professor Cordner has other than an esteemed academic and professional reputation nationally and internationally. The only source for what the Solicitor for Public Prosecutions considered was Professor Cordner's patent unreliability was his statement of July 1999. This was prepared at a time when there was a renewed police investigation into the deaths of Mr Gilham and Mrs Gilham and solicited by them. Because of the way the investigation had been handled at the time of the deaths of the deceased, there were many unanswered questions in respect of which the investigators considered that the expertise of a forensic pathologist might be of assistance.

397 The decision not to call Professor Cordner as a Crown witness in January 2008 was made without the Crown Prosecutor (or anyone else acting on behalf of the Director of Public Prosecutions) having conferred with Professor Cordner in order to determine the extent to which his views were actually at variance with those of the other pathologists. The extent to which the Crown experts might have been content to moderate their views or even defer to Professor Cordner, as Dr Lawrence ultimately did on the appeal, might have been explored in conference. There is no explanation proffered as to why Professor Cordner was not spoken to as the Crown prepared its case for trial in January 2008; before or after the voir dire in March 2008; in the course of the first trial; or by the new Crown Prosecutor before the commencement of the second trial in October 2008. For that reason alone the assertion in correspondence that his evidence was at variance with those of the other experts and that he was unreliable for that reason cannot be defended.

398 The assertion that Professor Cordner's views are based upon "a complete acceptance of the accused's account", either in the passage extracted or more generally in his report, simply does not withstand scrutiny. Even if by a skewed reading of his report that might have been thought to be the case, after Professor Cordner gave evidence on the voir dire in March 2008 concerning his approach to the preparation of his report, an objective and detached prosecutor would have rejected any suggestion of bias in his methodology or his conclusions.

399 The trial judge permitted questions to be asked by defence counsel on the voir dire (over the objection of the Crown Prosecutor) concerning Professor Cordner's approach and the structure of his report in the event that the applicant was convicted and an appellate court was invited to find that there had been an irregularity in the trial by reason of the Crown's failure to call him as a witness.

400 Professor Cordner said when he prepared his report in July 1999 that he used the applicant's version of events given to the police as a starting point, not uncritically accepting it as the truth but in the sense of it consisting of a series of propositions which forensic pathology may or may not be able to say something about. He said his preferred approach as a forensic pathologist is to apply the findings on autopsy, and such observations of the crime scene that are known, to some specific set of circumstances rather than coming to any conclusions (presumably about the cause or manner of death) in the abstract. He offered the view that as a general proposition:

"...forensic pathology...makes a stronger contribution when it is dealing with particular propositions than when it is asked to, without any proposition at all, recreate circumstances from say the autopsy findings alone..."

401 The Crown Prosecutor did not ask any questions of Professor Cordner on this issue on the voir dire.

402 On the appeal Professor Cordner agreed with the proposition that a working assumption he adopted in 1999 in the renewed investigation into the deaths of Mr and Mrs Gilham was that the applicant's version was given in good faith. He also said that he was invited to respond to comments from Dr Oettle about a number of puzzling aspects of the events surrounding the murders and the applicant's account of what occurred. He gave the following evidence:

"...I am really, as you can see, responding to comments that were made by another pathologist in relation to this case and I'm really saying that these are the sorts of imponderables and suppositions and speculations that forensic pathology has a lot of difficulty with and I conclude my comment in that particular section by saying Jeffrey's account is, you know, an assessment of it is not for the forensic pathologist to say..."

403 None of the other expert witnesses were asked to comment upon whether or not Professor Cordner's approach was contrary to or consistent with the accepted methodologies of the forensic practitioner.

404 There is a more fundamental defect in the Crown's decision not to call Professor Cordner because it was, in part, expressly based on the fact that he held a different opinion from that advanced by the witnesses the Crown intended to call. The Crown is simply not entitled to discriminate between experts, in particular between those whose views they have sought, calling only those that advance the Crown case, any more than it is entitled to call some lay witnesses and not others for some perceived tactical advantage. The fact that there is disagreement between the experts from whom the Crown have sought an opinion provides no basis for the Crown to elect to call only those experts who will give evidence that supports the Crown case, or whose views are consistent with the Crown case theory, and to refuse to call those witnesses whose views do not support the Crown case. This is particularly so where the subject matter about which the opinions are expressed is controversial. The Crown Prosecutor's obligation is to call all relevant evidence at his or her disposal. That obligation is a continuing obligation which persists until the Crown case is closed. The Apostilides principles are not the rules of a game. They are rules designed as a protection against unfairness or the abuse of prosecutorial power (see R v Gibson [2002] NSWCCA 401 (Sully J at [49], Wood CJ and Howie J agreeing)), quoting Randall v The Queen [2002] 1 WLR 2237 at 2243).

405 The expert witness has independent obligations to the Court under the Expert Witness Code of Conduct (see Schedule 7 to the Uniform Civil Procedure Rules 2005). The paramount duty of the expert witness is to the Court and not to any party to the proceedings. An expert is not an advocate for a party. It is in the discharge of the different but allied obligations of the expert and the Crown Prosecutor that the jury is educated and informed about matters in issue between the Crown and the accused which are beyond the jury's experience. Where the views of the experts differ, the extent and basis for disagreement can then be tested, if necessary, with the Crown seeking leave to cross-examine where the evidence might prove to be unfavourable under s 38 of the Evidence Act (as occurred in Velevski). It is in that process that the interests of justice are preserved and advanced. The assertion by the Solicitor for Public Prosecutions that it is not appropriate to utilise s 38 for the purpose of calling an expert witness who the Crown intends to discredit by obtaining leave to cross-examine is not soundly based, whether in the context of this case or generally. On the appeal it was only faintly suggested that this approach to the construction of s 38 provided justification for the Crown's decision not to call Professor Cordner.

406 As noted by Ipp JA in R v Parkes [2003] NSWCCA 12; (2003) 147 A Crim R 450, there is ample authority that the use of s 38 is not limited to circumstances where the unfavourable evidence is unexpected. And as noted by Hunt AJA, Buddin and Hoeben JJ in Kanaan v R [2006] NSWCCA 109 at [84], the greater availability of cross-examination of a Crown witness by the Crown Prosecutor has placed more emphasis on the Crown's obligation to call relevant witnesses. Their Honours noted the remarks of Whealy J in R v Ronen [2004] NSWSC 1298 at [32]:

"...the increasingly common use of s 38 has often demonstrated its value in getting to the truth of the matter, although great care must be taken to ensure that the trial does not become side-tracked by a collateral issue carrying with it the real possibility of prejudice to the accused..."

407 Ipp JA observed in Parkes at [71]:

"Special considerations, however, may arise in regard to s 38 in circumstances where the applicant well knows that the witness is likely to give unfavourable evidence on a particular issue (but wishes to elicit evidence from the witness on other issues), deliberately refrains from asking about the unfavourable evidence, and when it emerges in cross-examination, makes the application."

408 His Honour recognised the Court's disapproval of the use of s 38 (by the Crown) as a tactical or forensic device, that is, the deliberate use of s 38 to manipulate trial procedure so as to gain an unfair or forensic advantage that, but for s 38, would not arise.

409 There could have been no resistance to an application by the Crown for leave to cross-examine Professor Cordner at trial, to explore what the Crown thought was a persisting bias or imbalance in his approach. This is assuming the Crown Prosecutor maintained that view after leading evidence from him concerning the question of similarity or the pattern of injuries, and after giving his views the careful consideration they deserved.

410 Finally, the criticism that Professor Cordner's opinions are argumentative, with nominated paragraphs of his report cited to exemplify that criticism, is baseless. It is unnecessary to set out paragraphs 4 and 10 of his report in full to appreciate what we are inclined to think was a determination on the part of the Crown Prosecutor to look for a way of not calling Professor Cordner. Professor Cordner was expressly invited to offer his views about a wide range of issues that had been the subject of comment by Dr Oettle, including the relatively small amount of blood transferred to the applicant, which might have been explained by the applicant washing. To this Professor Cordner raised the question, which we note was then the subject of considerable focus at the trial, whether the applicant could have stabbed Christopher 17 times without being spattered with blood. He said:

"Assuming that Jeffrey was spattered with blood, why wash - and if he washed where and how did he do this? The inference is that he washed because he had some of his mother's and father's blood on him too. He had no need to wash (assuming he is acting rationally, which may be a significant assumption) if it was only his brother's blood because he is going to concede that he killed his brother. But why wash the knife too? Blood from all the victims would be on it however the murders happened. (Incidentally, I can see no report of a formal examination of the knife to confirm the absence of blood, although as far as one can see from the photographs, there is no obvious blood on it. In the absence of any blood, an additional question would need to be asked - was it indeed the weapon?)."

411 We regard these questions and those set out in paragraph 10 of his report as sensible questions posed interrogatively for the attention of the investigators. They were questions that Professor Cordner thought might have been overlooked. He recommended submitting Christopher's shave coat and the applicant's shorts to analysis for hydrocarbons and accelerants if they were still available (they were not) and reviewing the formal report of the examination of the knife (there was none). He also suggested that the applicant's clothes and shoes found upstairs be submitted for examination, particularly for hydrocarbons or blood (they were no longer available, having been disposed of by police in 1995).

412 We are satisfied that Professor Cordner's report fairly reflects the opinion he held at the time of the trial in October 2008 and that the Crown's failure to call him as a witness cannot be justified. The emphasis given to the issue of similarity as a piece of circumstantial evidence (a similarity which was then elevated in the evidence of the Crown's experts to the extent of an underlying pattern), and the Crown Prosecutor's ultimate articulation of the Crown case theory that the applicant inflicted all the wounds in a frenzied state of murderous rage in substantial reliance upon that evidence, cannot be doubted. The failure to call Professor Cordner to give evidence that in his opinion that analysis lacks a legitimate scientific foundation constitutes a miscarriage of justice.

Ground 9: That his Honour erred in permitting the Crown Prosecutor to hold a knife before the jury and have the applicant demonstrate his dexterity with the knife.

413 The knife used to kill Mr Gilham, Mrs Gilham and Christopher was lost prior to the trial. During cross-examination of the applicant the Crown Prosecutor presented the court with a knife similar to the knife used in the murders. The Crown Prosecutor then presented the applicant with the knife in front of the jury and asked him to comment on its similarity to the murder weapon:

"Q. (Approached and shown Exhibit AD.) Would you agree that that handle has roughly the same type of indentation - leave aside the blade of the knife, that's not important?

A. Yes, okay.

Q. But the handle?

A. Yes and no, I think it's a little bit different.

Q. Yes, you're quite right. It's a similar type of affair for the hand to use, isn't it?

A. I think it's smaller and it's got a small bit that's there, so it's different, yes."

414 The Crown Prosecutor then asked the applicant a series of questions about how the knife could be used and the significance of gripping the handle in different ways while using it in a "stabbing motion":

"Q. All right. What I wanted to ask you though in relation to it is if this knife is being used to carve or to cut, then that part of the handle is of use to assist the holder to keep some hold of the knife when it's being used?

A. Yes, that's what it's designed for, yes, I agree.

Q. Filleting, carving, slicing?

A. Yes.

Q. And your fingers run along that handle section?

A. That's right.

Q. If, however, it's being used in a stabbing motion like this (indicated), it doesn't really matter where your fingers are, does it, because at any point you've got the leverage of your thumb on top and your grip really around the main part of your hand to hold on to that knife?

WITNESS: I don't understand the question.

Q. The question is: You can move the knife in your hands while you're using it, meaning that the sharp edge will turn--

A. Oh, if you're saying could you spin it in your hand, I'd say yes, you could spin it in your hands, yes.

Q. Without necessarily having to move your position in relation to the thing or person being stabbed?

A. Yes, I think you could move your - spin the knife around in your hand, yes. Is that what you're trying to ask me?

Q. There's not a particular way to hold the knife with the handle like that when you are using it in a stabbing motion?

A. I don't think they're designed for that, no, if that's what you're asking me."

415 Counsel for the applicant raised concerns about the Crown Prosecutor's use of the knife in cross-examination. However, he did not in terms object and the issue was not directly addressed by the trial judge. Counsel said:

"I don't know what we're doing here. I don't think that it is very orthodox for a Crown Prosecutor to be handling a knife like this."

416 In-court demonstrations do not fall within the scope of Part 2.3 of the Evidence Act: Evans v The Queen [2007] HCA 59; (2007) 235 CLR 521 at [30] (Gummow and Hayne JJ), [63] (Kirby J), [195] (Heydon J), [225] (Crennan J). Accordingly, the admissibility of an in-court demonstration turns upon the application of the relevance provision found in s 55 of the Evidence Act, the discretionary and mandatory exclusionary rules contained in ss 135 and 137, and the common law rules of evidence.

417 The respondent submitted that the Crown Prosecutor's use of the knife in the courtroom was relevant to an understanding of the various directions of the stab wounds that were inflicted on each of the deceased. It was submitted that the manner in which the knife could be held and manipulated was relevant and explained the divergent stab wounds of Mr and Mrs Gilham and Christopher.

418 The applicant submitted that the applicant's dexterity in holding and using a knife was not a fact in issue. It was further submitted that even if the applicant's dexterity was a relevant issue, any object with an appropriate handle could have been used for the same purpose.

419 These submissions should be accepted. The applicant's "dexterity" with the knife was not a fact in issue at trial. At no time was it suggested that he was physically incapable of wielding the knife in a manner consistent with the wounds sustained by the victims. However, the wielding of a knife and its use in a stabbing motion before the jury was plainly likely to provoke a response adverse to the applicant. The demonstration with the knife had no relevance to the issues in the trial and, given the likely consequence of the demonstration, it was unfairly prejudicial. This ground of appeal is upheld, although it would not alone justify the conclusion that there was a miscarriage of justice requiring the conviction to be quashed.

Ground 10: The trial miscarried because the Crown Prosecutor engaged in cross-examination of the accused in an improper manner.

Ground 12: The trial miscarried because the Crown Prosecutor addressed the jury in an improper manner.

420 The applicant submitted that the Crown Prosecutor cross-examined on topics for which there was no evidence and on topics that were irrelevant and highly prejudicial. It was also submitted that the Crown Prosecutor caused a mistrial by addressing the jury in an improper manner.

421 At the outset of her cross-examination the Crown Prosecutor asked the applicant a number of questions about how well he knew his brother:

"Q. What was your brother's date of birth?

A. I think it was the - oh, I can't remember.

Q. Year of birth?

A. The year would have been a few years before me, 1968 I'd say.

Q. You didn't know much about him, did you?

A. I think we were - we moved in different circles, yes.

Q. What do you mean by that?

A. We had different interests."

422 She then put a number of questions to the applicant about the relationship between troubled young people and their parents:

"Q. You see, when there are problems in a young person's life outside, they need their mum and dad more than ever, don't they?

A. I don't know.

Q. You're a father now, aren't you?

A. Yes, I am.

Q. When a kid's got problems outside the house, they need their mum and dad more than ever?

A. That's right, but I think sometimes when children become adults they probably think they can resolve things in other ways.

Q. So you're saying, are you, that because your brother Christopher had some sort of problem with his teaching and didn't like it, that he took that out on the two people who had loved and supported him, and still were, in the way that you have told this court?

A. No, I'm not saying that.

Q. You are suggesting a reason--

A. No, I'm not."

423 The Crown Prosecutor then questioned the applicant about his failure to take certain actions upon entering the house and seeing his mother on the floor:

"Q. So you got up there within a minute or on (sic) a minute?

A. Yes.

Q. Walked in and saw what?

A. Walked into the house and saw my brother over near my mother.

Q. Did you fly over there to your mother and say, "Mum, I'm here."?

A. No, I didn't.

Q. Did you whisper it into her dying or dead ear, "Mum, I came to help you."?

A. No.

Q. Did you let her know you had come?

A. No, I think I - I didn't know what was happening when I got up there, but my brother told me what he had done.

Q. You didn't check, did you?

A. I didn't check, no.

Q. No? Your mother was face down, wasn't she?

A. Yes.

Q. Didn't you just rush over and pick her up and see whether she was living?

A. No, I didn't.

Q. You didn't check whether she was dead?

A. No. My brother told me that they were both dead, so, no, I didn't check.

...

Q. You didn't run over and say, "Mum, what's he going on about?"?

A. No, I didn't.

Q. "Dad, what's happening?"?

A. No, I didn't."

424 Aspects of these sections of cross-examination were then relied upon by the Crown Prosecutor during her closing address to the jury, in which she said:

"He [the applicant] didn't even, as you know, ladies and gentleman, fly to his mother who he says had called him, grab her and say to her "Mum, I came, I came to help you". He didn't do that."

425 And:

"[depression] does make you down on the world, but it doesn't make you down on the two people who have always supported you, always seen it your way, done everything for you all the time, Chris Gilham needed his Mum and Dad."

426 No objection was taken to the relevant sections of the cross-examination of the applicant or to the Crown Prosecutor's closing address. Rule 4 of the Criminal Appeal Rules applies.

427 The respondent submitted that the cross-examination of the applicant on each of these issues was relevant and did not result in unfair prejudice to the applicant.

428 The respondent submitted that the question whether the applicant had "set his brother up" by telling people in the weeks leading up to the murder that Christopher was acting aggressively towards his parents was a fact in issue. It was further submitted that the relationship between the applicant and Christopher was relevant to this issue and that "the absence of knowledge by the applicant of his brother's date of birth was directly relevant to that aspect".

429 This submission is rejected. The applicant's ability to recall his brother's date of birth some 15 years after his death cannot be relevant to, or a fair indication of, the nature of the relationship between the applicant and Christopher in 1993.

430 The respondent contended that questioning of the applicant about whether a "kid" with problems needs their mother and father was also relevant to a fact in issue (whether Christopher had "problems") because it went to undermine the version of Christopher's emotional state provided by the applicant.

431 This submission is also rejected. Issues of relevance require careful consideration: Evans at [23] (Gummow and Hayne JJ). While it is correct that Christopher's mental state was a fact in issue, it is plain that the applicant's view on whether a child with problems needs his or her parents is not rationally capable of informing that fact. The Crown Prosecutor's submission in closing that depression "doesn't make you down on the two people who have always supported you, always seen it your way, done everything for you all the time" was similarly inappropriate. There was nothing in the evidence to support it.

432 The respondent further submitted that the line of questioning about whether the applicant had "whispered" into his mother's "dead or dying ear" was relevant to an assessment of the applicant's version of the events that took place upon the applicant entering the house. The applicant's evidence was that he did not approach his mother to verify that she was dead even though she had been alive only a minute or so earlier. The respondent submitted that the Crown Prosecutor was entitled to challenge the applicant's account by questioning him about his failure to check whether his mother was dead and to challenge it in a way that raised with the applicant a particular suggested response.

433 It is plain that the Crown Prosecutor was entitled to challenge the evidence of the applicant and to raise particular expected responses. However, what was formerly Rule 64 of the NSW Barristers' Rules (now Rule 84) provides that "A prosecutor must not, by language or other conduct, seek to inflame or bias the court against the accused". The words used by the Crown Prosecutor and the manner in which her challenge was put to the applicant were inflammatory and inappropriate. The Crown Prosecutor's use of similar language in her closing address was similarly inappropriate.

The issue of "disdain"

434 The applicant also referred to aspects of the Crown Prosecutor's closing address to the jury that dealt with the applicant's alleged "disdain" for his family. These included statements that:

• "he [the applicant] didn't know his brother at all";
• the applicant was "the odd man out in the family";
• "During cross examination he [the applicant] was given ample opportunity to show you that he had a close and loving relationship, a regard, a respect, for his parents. Did we see it? Did we see any respect for his brother? Far from it"; and
• "he [the applicant] was prepared to talk about his [father's] work in such a dismissive, such a condescending fashion, you might think, and that's the context of this family, that's the context even after the passage of years, even after perhaps any unpleasantness would have dropped out of memory - leaving aside the dreadful events of that night - all the bad qualities would drop out and all the good qualities would be there and that's the best that we can have about those three people who all died that night, from this accused in this court."

435 The Crown Prosecutor's description of the applicant talking about Mr Gilham's work in a dismissive and condescending fashion related to answers given by the applicant to questions about his father's qualifications as a scientific instrument maker:

"Q. You see, your father was a scientific instrument maker, wasn't he?

A. Yes, he was.

Q. His whole life was about precision, isn't that right?

A. No, not really. I think he - scientific instrument makers, in his lifetime, weren't required.

Q. Weren't required?

A. No, the actual trade disappeared, so he wasn't - he wasn't - I can't remember him being a scientific instrument maker. He did the Tech course, of course, but as a teenager or a young man he was all working for the RTA in the traffic counter section installing the counters on the road. They're a box with a black tube coming out of them that counts the cars.

Q. They're a scientific instrument, are they?

A. No, they're actually pre-made and you install them on the road, you nail the black strip down, set them till--

HIS HONOUR

Q. Didn't we have evidence from your uncle that your father was a scientific instrument maker?

A. He did a - that's his trade qualification.

Q. What, did he never practice in that qualification?

A. He would have before I knew what he did for a living, but my memory of him as a young man was - his job was - that's what his job was. He used to work in the southern pylon of the Harbour Bridge and that's where they were stationed.

Q. What did he tell people he was when they said, "What do you do for a job Steve"?

A. I work at the traffic counter section.

Q. You see, Ted Warner knew his job as a scientific instrument maker, didn't he? You've heard his evidence?

A. I don't remember if Ted said that, but I know what he did for a living."

436 The applicant was further cross-examined on this point the following day:

"Q. You said that his trade had virtually ceased to exist as an instrument maker?

A. Yes, that's right.

Q. And you said all he did really was put the boxes out on the roads?

A. I went through his work on a number of occasions and I talked to him about what he did and I knew that his trade was no longer taught.

Q. Did you know that he actually constructed those boxes for the roads?

A. No, I didn't.

Q. Calibrated them?

A. No, I didn't know that, but my--

Q. And used his technical expertise to make sure that they were performing for the function that they were meant to perform?

A. No, I didn't, but when I did go to his work he told me what he did and that was my understanding of what he did.

Q. He told you, "My trade ceased to exist and all I do is put boxes on the roads," did he?

A. No, he was not unhappy with his job. I don't think he felt his job was a fulfilling job but he had lots of other things in his life that were fulfilling for him but I think he didn't not enjoy his work, but when I went to his work, that's what he explained he did. I went to his work on a number of occasions.

Q. He still used his technical expertise in the use and preparation and deployment of these scientific instruments that he was responsible for placing in strategic locations around the State?

A. I think the process is someone maps out where they are supposed to go.

Q. Do you disagree with that?

A. Disagree with--

Q. That he would have used his skills?

A. I think he would have used his skills, yes, I agree with that, yes. But I think his role was to - I agree, to calibrate them, yes, and to put them where they should go, in a suitable location. But when I talked to him about it, what he did, that's my understanding of his job, you know. If you tell me he built them, I'm not aware of that, but-"

437 In the absence of the jury counsel for the applicant objected to the Crown Prosecutor's closing remarks in the following terms:

"Your Honour, I complain about several aspects of the propositions that have been advanced about the accused having had disdain for his brother, mother and father. There was no basis in the evidence for it and it has never ever been put to him in anything like those terms. There's just no proper basis for those submissions."

438 The applicant submitted that the Crown's submissions to the jury about the applicant's "disdain" for his family were inflammatory, prejudicial and without a proper evidentiary foundation. The applicant also sought leave to tender to this Court the affidavit of Anthony Cariola, the Engineering Manager of the Traffic Signal Workshop of the Roads and Traffic Authority (RTA) at Rhodes in Sydney during the late 1980s and early 1990s. Mr Cariola worked with Mr Gilham during this time. In his affidavit Mr Cariola describes the work done by Mr Gilham in the RTA Traffic Counting section. His evidence is consistent with the evidence given by the applicant at trial.

439 The respondent submitted that it was appropriate for the Crown Prosecutor to comment on aspects of the applicant's evidence that related to his family members. The jury had the benefit of observing the applicant in interviews taken on the day of the killings and also in his evidence given at trial. The respondent submitted that the manner of the applicant's responses were just as significant as the content of his responses, and that it was permissible for the Crown Prosecutor to make submissions about the applicant's views on his family that were informed by the way in which he gave evidence about them.

440 The respondent also objected to the affidavit evidence of Mr Cariola on the ground that it constituted new evidence rather than fresh evidence. The respondent submitted that the evidence given in the affidavit could easily have been obtained from Mr Cariola or another source at trial. The applicant was cross-examined on 6 and 7 November 2008 about the nature of Stephen Gilham's employment. It is plain from the extract set out at [435]-[436] that he was being challenged about the "down-skilling" of his father. The defence continued to call evidence until 11 November 2008. The respondent submitted that the four days between 7 November and 11 November were sufficient to gather and call evidence responding to the issues raised in cross-examination.

441 The applicant submitted that the evidence of Mr Cariola, or similar evidence from another source, could not have reasonably been obtained during this time period. The applicant submitted that neither the applicant nor his representatives were in a position to search for, locate, and bring to court a witness such as Mr Cariola during the final four days of a long trial in circumstances where the issue of the applicant's supposed "disdain" for his family had not been foreshadowed in the Crown opening or relied upon in the first trial. The applicant further submitted that the issue of the applicant's "disdain" for his family is an issue that goes to motive and is of significant importance.

442 The test for the admission of fresh or new evidence is set out at [545]. Having regard to the manner in which, and the time when, this issue was introduced, we accept that it would not have been reasonably possible for the applicant to have obtained Mr Cariola's evidence for tender at the trial. Given the motive suggested in the Crown case the issue was significant. The fresh evidence strongly supports a conclusion that the Crown was not entitled to submit that the applicant had denigrated his father's abilities and achievements and had accordingly shown disdain for his parents. This submission lacked a reasonable foundation. However, we do not believe that the receipt of the evidence would have caused the jury to have a reasonable doubt about the applicant's guilt.

The items on the bookshelf

443 The applicant also referred to a section of the Crown Prosecutor's closing address in which she referred to the items on the bookshelf in the main house, including the intercom, as being "all in a neat" line. She said:

"The other thing about the scene, when you have a look at the photograph of the intercom, which the accused tells us his mother used about a minute before he got there, that is, about 4.25, you might think, if it's true that it takes 5 minutes to get up there and do what he did with his brother. Have a look at those pieces of equipment there on top of the encyclopedias. There's the phone. There's the intercom. And there's the answering machine. The phone is not even knocked off its cradle, they're all in a neat line.

Now, if that lady is stabbed to the neck and maybe through the heart, if she's trying to make that call, wouldn't you expect that those pieces of equipment to have been knocked off, the phone at least to have been off the hook or something, even the intercom would be a bit out of place if they're all in a neat line. Have a look at your photographs of that. It's just so unbelievable that that should have been the case."

444 The Crown Prosecutor's statements to the jury were made without reference to the fact that the crime scene video clearly shows that the crime scene investigator had removed the calendar off the top of the intercom unit and had lifted the answering machine, the intercom and the telephone and put them back without reference to their previous positions. The photographs referred to by the Crown Prosecutor were taken after the items on the shelf were moved.

445 The applicant submitted that, in light of the interference with the items on the shelf, the Crown's submission that the items being "all in a neat line" was inconsistent with the applicant's case was misleading and unfair. The respondent concedes that the crime scene video confirms that the items were moved. However, the respondent asserts that in spite of the movement shown in the video, the observation that all of the objects on the bookshelf were in a neat line holds true. This submission should be accepted. The crime scene video shows that prior to the movement the items did not appear to have been disturbed. In particular, the phone handset was in the cradle. It was open to the Crown Prosecutor to argue that this could not be reconciled with Mrs Gilham frantically handling an item in this area after being stabbed or while being stabbed.

Attempts to dissuade Fire Officer Pelham from entering the House

446 The Crown relied on evidence given by Officer Pelham, the first fire brigade officer on the scene with a breathing apparatus, that the applicant had tried to persuade him not to enter the burning house. Officer Pelham's evidence was that he saw the applicant standing alone near the carport and that the applicant had said: "Don't go down there. Someone's got a knife". The Crown used this evidence to argue that the applicant had falsely claimed that there was someone in the house with a knife to prevent Officer Pelham from entering for fear that the fire would be put out too rapidly to destroy any evidence that linked the applicant to the murders.

447 Officer Pelham's evidence was in conflict with, or contradicted by, the evidence of a number of other Crown witnesses, including Officers Lowder, Polson and Langdon, Mr Baker and Mr Warner. The applicant submitted that the Crown had failed to put the applicant on notice that Officer Pelham's evidence was to be used as a key plank in the Crown case theory. The nature of the Crown's reliance on Officer Pelham's evidence during the second trial was first made apparent when it was put to the applicant during cross-examination that he had "told one of the fire officers not to go into the house due to someone having a knife in order to dissuade him from entering the house so as to keep the fire blazing". Officer Pelham's evidence was not used in the same manner during the applicant's first trial. The applicant submitted that the late notice of the issue resulted in the applicant and his counsel being unable to properly meet the Crown's submission, as all the relevant Crown witnesses had been discharged by the time the applicant was aware of the significance of Officer Pelham's evidence in the Crown case.

448 This submission should be rejected. Defence counsel made no objection to the Crown Prosecutor's use of Officer Pelham's evidence in this way. Rule 4 applies. Furthermore, no application was made by defence counsel to recall any of the relevant witnesses or for the Crown to reopen its case to allow further questioning.

The use of the knife in the Crown's closing address

449 During her closing address the Crown Prosecutor again made reference to, and wielded, the knife that had been presented to the applicant during cross-examination:

"Now, something might be said about the fact that in Christopher Gilham's case, one can clearly see - and Dr Lawrence could clearly see - that three of the wounds to his chest had the sharp end of the knife facing the opposite way. I put this, as you might recall, to the accused using a real knife. The type of knife which was used, the Crown submits to you, was a knife which is made to hold in a forehand position, because that's why the moulded handle is shaped as it is. I have it here. With his Honour's leave may I demonstrate?

HIS HONOUR: Yes.

CROWN PROSECUTOR: I have that knife there with the same handle, forget about the blade, that's not the same at all, but the same type of handle with an indentation for each end of the hand really. That works for slicing, it works for cutting up one's food and it works for carving, but for stabbing, and the knife goes really straight in from front to back in those middle of the chest wounds, you've got to hold it in a backhand grip, you've got to hold the knife in a backhand grip, and when you're doing that, the handle of the knife can turn, it doesn't really matter how you're holding it because you're holding it with your fist tightly around it. It makes no difference, the grip doesn't work that way, the grip is to your fingers, but when it's in your fist, it doesn't work. You can turn it around.

It turns around itself. You can maintain purchase with the thumb on the end; any position equally easily.

So it's not suggested to you, ladies and gentlemen, that there was any change in position in any of these cases. It doesn't matter what ends - the sharp end was facing because of the way one holds that knife when one has it in the stabbing position, as opposed to using it in the way that we use knives every day cutting up our food. So that's a red herring the

Crown would submit to you. It means nothing. It doesn't mean the stab wounds in Christopher's case are any different from the other two. There could well have been a change in their stab wounds too. Dr Lawrence was hindered by the fact that the two bodies were so severely burnt he couldn't make it out, even in the majority of the stab wounds, certainly he was specific about that as far as Stephen Gilham went."

450 The applicant submitted that the demonstration and the speculation engaged in by the Crown Prosecutor were impermissible: Livermore v R [2006] NSWCCA 334; (2006) 67 NSWLR 659; Causevic v R [2008] NSWCCA 238; McCullough v R [1992] Tas R 43; Vella v R (1990) 2 WAR 537; R v Pernich (1991) 55 A Crim R 464.

451 The respondent argued that the Crown Prosecutor used the knife with the leave of the judge and that her handling of the knife was directed to addressing the various ways in which a knife with a handle similar to that used in the killings could be held. The respondent submitted that the demonstration was permissible and merely added a visual aid to what would otherwise have been permissible to describe to the jury.

452 The use of the knife by the prosecutor during her final address was in substance a repetition of the manner in which she used the knife when cross-examining the applicant. The trial judge having allowed that questioning, the subsequent submission cannot be criticised. However, since we are of the opinion that the questioning of the applicant with the aid of the knife should not have been allowed, there was no foundation for the prosecutor's submission in her closing address.

Conclusion on Grounds 10 and 12

453 Some of the complaints of the applicant with respect to Grounds 10 and 12 are made out. However, we are not satisfied that the issues in respect of which complaint is justified support the conclusion that the conduct of the Crown Prosecutor caused the trial to miscarry.

Ground 11: The trial miscarried because the Crown Prosecutor addressed the jury in a manner that undermined the effect of the directions his Honour gave concerning the need to scrutinise the fire demonstrations with great care.

454 During the course of Mr Munday's evidence the jury was expressly warned by the trial judge about how to approach the experiment evidence. The warning was given in the following terms:

"Well, ladies and gentlemen, this is a matter which was dealt with by me earlier on when you weren't here and it was my decision that the DVD should be edited because in my view it was not representative of what might have happened in the fire after the fire started to grow, and so it seemed to me it was unhelpful to you and therefore I made the decision that it shouldn't be shown to you. But you still have to be very careful of experiments, as they are called, and the reason is because it's been made apparent to you already that it was impossible to replicate the circumstances of this particular night and you've heard that one of the problems is that nobody knows what the carpet was made of. As I understand it, nobody knows what the underlay of the carpet was made of. Nobody knows other matters that would possibly influence the way that the fire would progress, so what's been done here is an experiment, but a limited one, and it can only have limited assistance to you.

It just seems to me that it's important to see how the fire progresses at least in the initial stage of the fire, and that's what we're talking about here, the initial growth of the fire until it starts on this exponential growth. But you should understand that it is not necessarily what would have happened on this particular night and certainly not necessarily what would have occurred to the accused or what the accused would have seen on this particular night and that's because we can't replicate all of the circumstances in which the fire took place.

Further, there is another slight problem with the experiments - and this is not the fault of Mr Munday, it's simply the basis of the material upon which he was acting - and that is there was no attempt to replicate in any way the female body, and particularly the clothing on the female body, and it may be a very important factor as to whether this fire was commenced on the carpet or on the body. And there'll be evidence about that no doubt in cross-examination, or particularly Mr Munday will give evidence about the difference that might have made to this fire had the female body, particularly if it had been clothed and where it had been clothed, had been lit rather than the carpet alongside the body.

So just understand this is a very limited experiment, it gives you limited assistance. Don't put too much weight upon it as being an exact replica or even a very close replica of what had occurred on this particular night, it's simply to give you some understanding of how a fire progresses in these circumstances and, as I understand it, the fire, as seen, doesn't change very much from what you've already been told about the spread of the fire, but rather than being told about it from various witnesses, it's perhaps better that you can see an example of it as long as you understand the limited basis upon which you can use that evidence at the end of the day."

455 During the cross-examination of the witness the trial judge re-stated part of the warning:

"That's what I have explained to the jury, and the jury should keep it in mind, that this is a very limited experiment designed merely to show a very limited amount of information."

456 The applicant submitted that the Crown Prosecutor directly undermined the force of the trial judge's directions during her closing address by asserting knowledge of the extent to which fires progress beyond the initial stage of ignition and by asserting that the first 10 to 15 seconds of the fire experiment shown to them was something approximating what the applicant may have seen.

457 In relation to the assertion about the extent to which fires progress, the applicant points to a passage of the Crown Prosecutor's closing address where she said:

"So you see, ladies and gentlemen, we do know that fires do take off pretty fast, once they reach that exponential thing. By the time he's killed Christopher downstairs, he'd have to crawl along the ground like Mr Lowder."

458 The applicant submitted that there was no evidentiary basis on which the Crown was entitled to make that submission to the jury. The applicant further submitted that contrary to the Crown Prosecutor's assertion, the evidence given by the applicant that the smoke was at the height of the piano as he left the house is consistent with the evidence given by Mr Munday.

459 The Crown Prosecutor then went on to address the jury about the applicant's reaction to seeing his brother set his parents' bodies alight. She said:

"But, ladies and gentlemen, it doesn't really matter, the Crown submits, about how long this fire took to take hold because the accused says he saw it when it was still on the end of a match, he saw it right from the start, and he expects you to believe, he expects you to accept, that an innocent man would not have done a single solitary thing to try to put that fire out. It really doesn't matter how big it was getting. He didn't do a thing to try to put those fires out and he hadn't, he says, checked where his parents are alive. He's not burnt at all, he didn't try to do a thing. There's a linen closet a little way away, there's his clothes he's taken off.

There were many things he could have done and, ladies and gentlemen, I suggest to you if it was your parents in danger of being burnt, you'd put it out with your bare feet, you'd do something. He expects you to accept, ladies and gentlemen, that an innocent man would do nothing to put out that fire. It doesn't matter if it's 30 centimetres high. Well that just doesn't make sense, it doesn't add up, ladies and gentlemen. That's what I have to say really about the fire.

Don't forget Professor Stern gave evidence that he didn't really see any evidence of a continuous application of accelerant between the body of the female and the male. He said that if it's turps it's a slow spread; whereas petrol would flash-over in seconds, but he said petrol, if there's vapour there, it will go over in seconds; mineral turps will not do that."

460 The applicant submitted that the Crown Prosecutor's statements that "it doesn't matter how big it was getting" and "It doesn't matter if it was 30 cm high" implied that the experiments conducted by Mr Munday should be relied upon as the outer limit of what the applicant experienced in seeing the fire progress from ignition, contrary to the directions of the trial judge.

461 The applicant also submitted that the Crown Prosecutor made submissions to the jury about human reactions to fire, about which there was no evidence. The Crown Prosecutor said:

"And we have, of course, the evidence from Mr Munday, that no matter what really, even with a lot more turps than the Crown suggests was available and a lot bigger spread than the Crown was suggesting was available, he still says there's 10 or 15 seconds, which, if you stop and time it, it's quite a long time. I'm not suggesting that's exactly it, but it's quite a long time, there is time to do something. Jeffrey Gilham wasn't drunk or on drugs or anything, there was nothing impairing his ability, he was a strong, fit young man. He didn't do a thing to try to stop that fire and that's not, the Crown submits to you, what an innocent man, a man who had not killed his parents, would do in those circumstances. It doesn't accord with any natural human instinct."

462 This statement is in conflict with the evidence given by Mr Munday about the ability of a person to put out a fire shortly after ignition:

"Q. Would it be possible for someone standing within a few metres of such a fire as it is set to put the fire out with any degree of ease in the first 10, 15, 20, 30 seconds?

A. Well, that would depend on what they had available to put the fire out and their state of mind. I'm not entirely comfortable surmising on that.

Q. Could someone simply walk out and, even with bare feet, stamp it out?

BOULTEN: I object to this."

463 No objection was made by defence counsel to the Crown Prosecutor's closing remarks in relation to the fire. Rule 4 applies.

464 It is correct, as the applicant asserts, that some of the Crown Prosecutor's statements were not founded in the evidence. However, to our minds, they had little force in any event. If it was the case that the applicant came upon his parents after they had been brutally stabbed many times by Christopher, who was in the act of lighting the fire, it is difficult to say how one would expect him to have reacted. His account is that he was so affected by what he saw that he took up the knife and, losing self-control, killed Christopher. The jury were in a position to evaluate how a person in the applicant's position may have reacted. The trial judge was careful to counsel the jury against setting too much store by the demonstration. We reject this ground of appeal.

Ground 13: The verdict was unreasonable and cannot be supported by the evidence

465 Section 6(1) of the Criminal Appeal Act 1912 provides that a verdict may be set aside if it is unreasonable or cannot be supported having regard to the evidence. As the High Court recently confirmed in SKA v The Queen [2011] HCA 13; (2011) 243 CLR 400 at [14], when considering whether or not the verdict of a jury was unreasonable the appellate court is to make its own independent assessment of the evidence both as to sufficiency and quality. The question that an appeal court is required to determine is whether, notwithstanding that there is evidence upon which a jury might convict, "nonetheless it would be dangerous in all the circumstances to allow the verdict of guilty to stand": M v The Queen [1994] HCA 63; (1994) 181 CLR 487 at 492-493. Only in circumstances where a doubt experienced by a court of criminal appeal can be resolved by the jury's advantage in seeing and hearing the evidence can the court conclude that no miscarriage of justice has occurred: M at 494 quoted with approval in SKA at [13].

466 The case against the applicant is a circumstantial one. As McClellan CJ at CL discussed in Wood v R [2012] NSWCCA 21 at [50]-[53], where a case is circumstantial "the jury cannot return a verdict of guilty unless the circumstances are such as to be inconsistent with any reasonable hypothesis other than the guilt of the accused": quoting Barca v The Queen [1975] HCA 42; (1975) 133 CLR 82 at 104 (Gibbs, Stephen and Mason JJ), citing Peacock v The King [1911] HCA 66; (1911) 13 CLR 619 at 634; see also Plomp v The Queen [1963] HCA 44; (1963) 110 CLR 234 at 252. However, a circumstantial case must be considered holistically. The court must weigh and consider the totality of the admissible evidence before reaching a conclusion, on the whole of the evidence, that it was open to the jury to be persuaded beyond reasonable doubt of guilt or, in the alternative, that a doubt persists such that it would be dangerous to allow the verdict of guilty to stand: R v Hillier [2007] HCA 13; (2007) 228 CLR 618 at [48]-[49] (Gummow, Hayne and Crennan JJ). Tortuous reasoning in order to explain away every individual circumstance as being consistent with innocence should not be engaged in: R v Micallef [2002] NSWCCA 480; (2002) 136 A Crim R 127 at [42] (Dunford J); Burrell v R [2009] NSWCCA 193 at [55] (Giles JA).

Issues raised by the applicant

467 The applicant submitted that the evidence led at trial left open a reasonable hypothesis consistent with his innocence. The applicant relied upon 19 factual issues in support of this submission. They may be summarised as follows:

(1) The pattern of the stab wounds: The similarity in the pattern of stab wounds of each of the deceased did not compel the conclusion that there was one perpetrator given that:

(a) death by stabbing is common;

(b) the chest or abdomen is where stab wounds are usually inflicted, as affirmed by the evidence of Dr Lawrence;

(c) the similarities in the wounds can be attributed to the same weapon being used by two young men of roughly the same age and level of physical fitness; and

(d) there were differences in the angle and direction of the wounds to Mr and Mrs Gilham on the one hand and Christopher on the other.

(2) The relative life circumstances of the applicant and Christopher at the time of the murder: The applicant's academic life and social life were stable while Christopher had career and relationship problems. These factors were recurrent themes in the evidence given by family members, friends and associates. Viewed as a whole, this evidence supported the applicant's case that he was not labouring under a psychological condition or a depressive illness and was not socially isolated. It did not support the Crown Prosecutor's assertion that the applicant was uncommunicative with his family or that he had a poor relationship with his father. The same body of evidence suggested that Christopher was experiencing some disappointment at his inability to find work as an engineer, was having difficulties in his work placement as a teacher, and had a relationship breakdown earlier in the year.

(3) Christopher's relationship with his parents: There was evidence that Christopher had a tense relationship with his parents. This evidence included the applicant's evidence that Christopher had been arguing with his father, with some pushing and shoving in the period leading up to the murders, and the evidence of friends of the applicant that he had told them about these incidents.

(4) Lack of motive: The Crown was unable to nominate any motive for the killings despite a police investigation that extended over 13 years from the date of the murders to the filing of the ex officio indictment in 2006.

(5) Timing: The timing of the sequence of events from 3.57 am culminating in the applicant raising the alarm at approximately 4.30 am, as contended for by the Crown, was weak. It relied heavily upon Ms Shaw's evidence that the voices she heard at 3.57 am were the only sounds uttered by the deceased and the killer. Even if her evidence were reliable, it did not negate the applicant's account, given for the first time in the early hours of the morning, that he was only alerted to the attack on his parents when his mother called him on the intercom and that they were dead or dying when he entered the house. Ms Shaw gave evidence that she heard a man and woman arguing at about 3.57 am. It was the applicant's case that the male voice she heard was not that of his parents fighting with him, as the Crown contended, but Christopher arguing with his mother upon her realising that he had murdered Mr Gilham and was intent upon killing her and that the encounter and the argument may have extended for much longer than the Crown suggested. Ms Shaw gave evidence that the arguing continued for about 5 minutes, maybe longer, and the Bakers' dog was unsettled for some time. If the Crown's contention was correct, where was Christopher at that time? Why did he not intervene or attempt to intervene? If the likely order of events was that Mr Gilham was attacked first and then Mrs Gilham, as the Crown submitted, the duration of the argument and its audibility raised unanswered questions as to Christopher's role in the events of that night.

(6) Evidence that the accused washed to remove the blood of the deceased: The evidence that the accused washed or showered before raising the alarm was weak. It consisted of Dr Culliford's evidence that after the applicant killing his brother she would have expected blood on the ulna region of his right hand or wrist, and evidence from Mr and Mrs Warner that the blood they noticed on the applicant's foot and shin appeared to be washed or smeared. The inference that the applicant should have had more blood on him after killing Christopher was contradicted by the expert evidence of Dr Raymond and Sgt Reynolds that the amount of blood shed in any violent assault varies widely, as does the manner in which it is dispersed. In any event, the fact that there was some visible blood on the applicant diminished the force of the Crown theory that he had washed himself prior to going to the Warners'.

(7) Evidence that the applicant washed the knife: The evidence that the applicant had washed the knife was also weak. The crime scene was contaminated with water and the knife was not handled with care by police. The fact that the knife was lost further undermined the Crown case.

(8) The lack of blood on the applicant: Neither Mr Gilham's nor Mrs Gilham's blood was definitively identified on the applicant. The blood on the applicant's foot, which may have been Mrs Gilham's, was capable of being explained in a manner consistent with his account.

(9) The blood on Christopher's shave coat: Christopher's shave coat was tested and found to have blood on it that may have belonged to Mrs Gilham. The fact that the shave coat has also been lost, thereby preventing any DNA testing that may have identified the blood of his parents, counts against the Crown case.

(10) The blood spatter on Christopher: The blood spatter on Christopher's right hand, wrist and forearm was consistent with him stabbing a bloodied body with that hand. Uncontradicted evidence to this effect was given by Dr Raymond.

(11) The fire demonstration had no scientific value: Even accepting the fire demonstrations were a reliable indication of how the fire behaved, there remained a real possibility that the fire may have started as the applicant described in his evidence.

(12) The fire evidence did not challenge the applicant's account: The fire experts gave no evidence which rendered the applicant's account implausible. This was acknowledged by the trial judge in his summing up where he said that "so much of the evidence on fire can be disregarded" that the relevance of the material "seems to have disappeared".

(13) Character: The applicant was a person of good character.

(14) The existence of a consistent and exculpatory account: The applicant provided an exculpatory and consistent account from when he was first spoken to by police on the morning of the killings. He remained unshaken in cross-examination.

(15) Carbon monoxide: Dr Lawrence's evidence that "Christopher did not inhale significant amounts of smoke" was at odds with the Crown's assertion in its closing address that the applicant had murdered his family and set fire to the house and that Christopher did not make it to the top of the stairs before he was attacked.

(16) Implausibility of premeditation: The Crown's assertion that the murders were premeditated is unlikely for the following reasons:

(a) the use of the filleting knife from the kitchen is antithetical to any degree of planning;

(b) the evidence concerning the use of accelerant points to an even greater lack of preplanning. Petrol has greater volatility than mineral turpentine. The Crown case was that petrol was siphoned from a car into a jerry can by the applicant. This is uncontentious as the applicant admitted to siphoning the fuel the previous evening for use in the preparation of 2-stroke fuel for a motor boat. There was 1 litre of petrol in the jerry can located by police. The Crown case does not address why a litre of mineral turpentine was used to set the fire instead of a litre of petrol. Had the applicant been the murderer he would have used the available petrol; and

(c) the supposed preplanning undertaken by the applicant involved him openly admitting to police that he murdered his brother. The applicant did not claim that Christopher threatened him. There was therefore no possibility of self-defence being raised. There was no evidence that he was versed in the law. There was no certainty that he would not serve a substantial term of imprisonment for killing his brother. He did not engage a lawyer or exercise his right to silence before speaking with police. He fully cooperated with police.

(17) The height and temperature of the fire: The Crown asserted that the applicant had used one litre of mineral turpentine to set the fire. However, the calculations of the height and temperature of the smoke relied upon the use of four litres of mineral turpentine.

(18) Prosecutor misconduct: During her address the Crown Prosecutor referred to a defence argument run during the first trial that Mr Gilham went to bed with an axe beside him because he was frightened of Christopher Gilham. This was not part of the defence case in the second trial.

(19) The Panadol Paste: The Panadol paste found in the bathroom, used almost exclusively by Christopher, was a strong piece of evidence pointing to the likelihood that he was contemplating suicide, which supported the theory that he was the murderer.

468 A number of these issues have been dealt with in relation to other grounds of appeal. In dealing with Grounds 4 and 5, we accepted the applicant's submission that little weight can be placed on fire demonstration evidence (issues 11,12 and 17). The evidence that there was no blood on the knife is also of little weight because of the way it had been handled and the likelihood of its contamination by water (issue 7).

469 As discussed in relation to Ground 8, the purported similarity of the pattern of stab wounds on the three victims is not available to support an inference that there was only one perpetrator. As we observed when dealing with Ground 8, the evidence was not admitted under s 98 of the Evidence Act such as to permit the jury to apply coincidence reasoning. Moreover, we are satisfied that the evidence of similarity of the stab wounds, or evidence suggestive of a pattern of wounding, was wrongly admitted as opinion evidence. The Crown placed significant reliance on this evidence. It must be disregarded when considering whether the verdict was unreasonable.

470 With the exception of issue 18, which raises a procedural rather than evidentiary issue and is also of no relevance to a consideration of whether the verdict was unreasonable, each of the remaining factors relied upon by the applicant address the circumstantial case mounted by the Crown at trial. We are mindful that each piece of evidence must ultimately be considered and weighed along with the whole of the evidence presented at trial. However, we will give separate consideration, where appropriate, to each of these issues in order to assess their cumulative contribution to the case against the applicant.

Evidence that Christopher was the more likely killer

471 Plainly, either Christopher or the applicant was responsible for the murder of their parents. Neither brother had an identified motive for the killings. Both were of good character. Numerous witnesses attested to their general sociability and their individual talents.

472 There was evidence that Christopher was experiencing some difficulties in his life before his death. He had been unable to get a job in his chosen field of engineering and had commenced studying for a Diploma of Education in the hope that he would be able to get work as a teacher. A month before the killings he had had an unpleasant teaching experience while undertaking a practical component of his course. There was also evidence that Christopher's former girlfriend had recently ended their relationship.

473 Evidence was led at the trial that Christopher was becoming increasingly aggressive towards his parents and had assaulted his father in the weeks leading up to the killings. The applicant gave evidence that he had seen his brother push his father more than once in the four-week period prior to the murders. The applicant gave evidence that he was uncertain about the cause of the arguments.

474 Mr Nolan, a friend of the applicant, gave evidence that the applicant had told him that Christopher was behaving aggressively in the weeks leading up to the murders. Mr Nolan said that on the afternoon before Mr and Mrs Gilham were killed the applicant said:

"Chris has still been acting really strange and going off at the drop of a hat, especially towards dad. It's come to the push and shove stage between them. I don't know what's wrong with him."

475 Ms Moskos, the applicant's former girlfriend, also gave evidence that the applicant had expressed concerns about Christopher's behaviour on multiple occasions in the weeks leading up to the deaths of his parents. She said:

"About three weeks [before the murders] Jeff told me in a telephone conversation, 'Chris has been stomping around the house. He has been throwing things down and going off at mum and dad.' Jeff had told me on previous occasions that when Chris gets angry he often stomps around and slams the door - slams doors. About two weeks [before the murders] when I was speaking to Jeff on the phone he was in the computer room. During the conversation Jeff said to me, 'Can you just hang on a minute, I will just go down to the other phone.' A few seconds later Jeff got back on the phone and I said to Jeff, 'What was all that about? Why did you have to move?' He said, 'Chris wanted the computer room and Chris is going off again.' Jeff and I then had a general conversation and after about fifteen minutes Jeff interrupted me and said, 'Wait, dad and Chris are having an argument. Chris is being really weird lately. Him and dad have been pushing and shoving each other'."

476 Ms Moskos, who had visited the applicant at his family's home the night before the murders, gave evidence that the applicant had again expressed concern about Christopher on that evening. Ms Moskos said that as she was leaving the Gilham residence the applicant had said:

"What am I going to do about my brother? It's pretty serious, Hayley. I've never seen him so psycho. He's pushing my father around. I don't know what to do.'"

477 Ms Moskos gave evidence that when she arrived at the Gilham house that evening she had a cup of coffee with Mr and Mrs Gilham while she waited for the applicant to arrive home. Ms Moskos said that she had also briefly seen Christopher, who had come out of the family computer room, "said hello" and returned to the computer room without engaging with her. Ms Moskos said that she had formed the impression that there may have been "some degree of tension" between Christopher and his parents, but she did not see any signs of pushing, shoving or aggression. She also gave evidence that she had not heard anything in the background of her earlier phone conversation with the applicant that supported his statement that Christopher was "going off again".

478 The evidence that Christopher was experiencing difficulties may suggest that he was more likely to have murdered his parents than was the applicant. However, relationship breakdowns and difficulties in identifying a chosen career path are problems that are commonly faced by young adults without operating as precursors to uncontrolled or premeditated violence. It must be remembered that all of the evidence about Christopher's aggressive behaviour towards his parents originated, either directly or indirectly, from the applicant. That evidence contrasts sharply with other evidence given by those who knew Christopher well. Ms Kubish, Christopher's former girlfriend, described him in particularly favourable terms. She said that Christopher was:

"just a calm, gentle, sincere person. He - I remember him as being very interested in everything. As I mentioned before, we just spoke about practically anything really and he spent so much time on public transport, and whenever he was interested in something he would just go and borrow a book from the library and read all about it. I think he would have made a wonderful teacher. He had a good sense of humour. As I mentioned before, he was very interested in jazz and we went out and saw many sort of jazz bands in that way. I always found him very mature. He was certainly older than me at the time, but he also seemed very, very astute, very perceptive, and someone who ... was able to say what he was thinking and say what he was feeling. He wasn't a, you know, he wasn't a repressed person or anything like that, just a calm, gentle person really."

479 The Crown asserted that what the applicant reported to his friends about Christopher's behaviour and his evidence at the trial to the effect that Christopher had been behaving aggressively were designed to frame his brother as his parents' killer. The Crown also relied upon this evidence to support the submission that the killing was premeditated. The lack of independent evidence that Christopher was behaving erratically lends some support to the Crown case.

480 Another factor bearing on the likelihood of Christopher being the murderer was his attire. He was wearing a shave coat, no underwear and was without his glasses. The crime scene video reveals that his glasses were on a windowsill in the downstairs bathroom.

481 There was evidence that Christopher was reliant upon his glasses. Dr Davies, an ophthalmic surgeon who had last treated Christopher in 1992, gave evidence that Christopher was moderately short-sighted and suffered from mild astigmatism (an irregularly shaped cornea) and bilateral congenital nystagmus (an involuntary, irregular jerky movement of the eyes). Dr Davies said that the nystagmus only had a "mild" impact on Christopher's vision. Dr Davies gave evidence that a person with Christopher's level of vision impairment would have been able to perform normal tasks around the house without his glasses and that it would have been "quite feasible for him to go about his daily life without glasses at that time". However, Dr Davies gave evidence that, in his opinion, someone with Christopher's vision would have benefited from having glasses to see anything further than an arm's length away. There was also evidence from Christopher's friends, neighbours and former girlfriend that he always wore his glasses and evidence that he experienced greater vision problems at night, and did not drive after dark for this reason. In light of this evidence, the fact that Christopher was not wearing his glasses lends significant support to a conclusion that he was not responsible for the murders of his parents.

482 The Crown submitted that Christopher's attire, which exposed his genitals and would likely have impeded his movement, was inconsistent with him being the killer and was more consistent with him being roused from bed by the noise upstairs and dressing with haste. While it might also be said that his unusual attire is capable of supporting the applicant's case that he was the killer, it being easy for him to dispose of a bloodied and possibly fire damaged shave coat after killing his parents, we consider that his attire lends significant support to a conclusion that he was not responsible for the murders of his parents.

Timing and the carbon monoxide evidence at trial

483 There are inconsistencies between the timing evidence given at the trial and the applicant's account of what occurred that morning. On the Crown case, Ms Christine Shaw, who was awake in bed reading at 3.45am, overheard voices at 3.57 am that continued for approximately five minutes, perhaps a bit longer. She described the first voice as male, "primal" and "guttural" and the second voice as female, "high pitched" and "very stressed". It was the Crown case that the voices were those of Mr and Mrs Gilham screaming while they were being killed.

484 Two other neighbours, the Bakers and the Warners, also gave evidence in support of the timing of the events for which the Crown contended. Mr and Mrs Baker were woken by their dog barking sometime between 4 am and 4.15 am. They did not hear any voices coming from the Gilham residence. Mr Warner gave evidence that he was awoken by the sound of two people arguing in the early hours of the morning. He was in doubt as to whether it was two men or a man and a woman. He described the voices as being "very angry", "very heated" and "very loud". He was uncertain as to the exact time that he was wakened but he believed that it was after 3.30 am, most probably about 3.40 am to 3.50 am. He agreed that it may have been 3.57 am. After he heard the voices he went back to sleep and was woken again by the applicant knocking loudly on his front door.

485 The combined effect of this evidence, and in particular Ms Shaw's evidence, fixed the most likely time of the raised voices as being between 3.57 am and 4.02 am. It is likely that both parents were killed at approximately this time or shortly after. Mr Gilham had no defensive wounds and his position in the bedroom and the location of his injuries on his upper torso supported the inference that he was stabbed while he was asleep in bed. It is probable that Mrs Gilham was killed a short time later, as she had moved only a short distance into the lounge room from the bedroom where she was later found by fire officers. It was the Crown case that the applicant then went downstairs and killed Christopher, probably as he was roused from sleep by the commotion upstairs, this explaining his attire. On the Crown case all three deceased were killed between 3.57 am and roughly 4.12 am.

486 Dr Lawrence gave evidence that, having regard to the wounds sustained by the three victims, it would have taken no more than a few minutes for each person to die. Relying upon the Crown's asserted timing of events, this placed the time of death of Christopher, being the last of the deceased to be killed, at about 4.15 am at the latest.

487 This analysis was supported by Dr Lawrence's evidence that each of the deceased had died before the fire was ignited. Dr Lawrence's evidence was as follows:

""Q. Now I want to ask you about carbon monoxide levels. You gave evidence before lunch in a question led in these terms by the learned Crown Prosecutor, accepting the proposition that Christopher died before the fire was lit?

A. Yes.

Q. There are other possibilities, are there not?

A. Possible explanations for--

Q. Yes, possible explanations?

A. As I have said, I don't think there's a significant difference between the carbon monoxide levels in any of them to be quite frank.

Q. They're all been (sic) normal levels?

A. Yes.

Q. Stephen's level, Helen's and Christopher's levels are all within reasonable limits?

A. Yes.

Q. So really, apart from concluding that none of them had inhaled a lot of carbon monoxide before they died, no other conclusion can be safely drawn; do you agree with that?

A. Yeah, that's a reasonable assessment, that none of them had absorbed significant amounts of carbon monoxide.

Q. And that's the only scientific conclusion one can draw?

A. In the normal course of things in looking at a person with stab wounds and fire damage, a level of less than 10 per cent would be assumed to be that the person had died before the fire started."

488 The applicant submitted that Dr Lawrence's evidence that Christopher had "not inhaled significant amounts of smoke" was inconsistent with his evidence that Christopher must have died prior to the fire being lit because, if that were correct, he would not have inhaled any smoke at all. However, to our minds the two propositions are not inevitably in conflict.

489 It was the applicant's case that his mother and Christopher were still alive when he was called on the intercom and that he immediately ran from the boatshed to the house. He said that upon entering the room, being informed by Christopher that his parents were dead and seeing the fire on his parents spreading, he chased his brother downstairs and killed him. He said that he then ran upstairs and out of the smoke-filled house directly to the Warners' house. The 000 call was logged at 4.34 am and was placed, on Mr Warner's evidence, within minutes of the applicant entering their house. Taking into consideration the time required to run from the boatshed to the house (timed at between 22 and 30 seconds), the relatively short time the applicant was in the house chasing and then stabbing Christopher, and the time required for him to run to the Warners', the Crown submitted that on the applicant's account he must have arrived at the main house from the boatshed at or shortly before 4.25 am. However, the evidence of the neighbours all but compelled the conclusion that by this time his parents were already dead.

490 More fundamentally, even putting the timing evidence aside, Dr Lawrence's evidence that none of the deceased (in particular Christopher) inhaled significant amounts of carbon monoxide effectively negated the applicant's account that Christopher set his parents alight, having just murdered them, and that the fire was alight by the time the applicant arrived in the house and was spreading quickly. Dr Lawrence's evidence that Christopher and his parents were already dead when the fire was lit effectively ruled out the possibility that anyone other than the applicant lit it. This evidence, which was not contradicted at the trial, lent very powerful support to the Crown case.

Inconsistent statements about the location of the knife

491 The applicant gave inconsistent versions about the location of the knife and whether Christopher was holding it when he arrived at the house from the boatshed. During the 000 call the applicant told the operator that Christopher had a knife when he entered the house:

"Operator: Easy. Calm down, calm down and just tell me mate.

J Gilham: I walked in...through the door...

Operator: You've walked in from where? Out..being outside somewhere?

J Gilham: Mum called me...called me...

Operator: Right.

J Gilham: And... I came up...and...I see him there...with a knife."

492 Detective Parsons gave evidence that when she interviewed the applicant in the back of an ambulance at the scene he said: "he [Christopher] killed them and I think I stabbed him, he was holding the knife." Det Parsons' evidence was supported by Ambulance Officer Maudsley, who gave evidence that "Gilham also told [Det Parsons] 'My brother had a knife'".

493 In the electronically recorded interview with police several hours later the applicant's account was different. He told police that when he entered the house he had seen the knife "near the piano" and had "picked it up and chased [Christopher] downstairs". During the walkthrough of the house he again told police that the knife was near the seat of the piano.

494 The applicant's inconsistent statements concerning the knife impact negatively upon his credibility.

Evidence that the applicant did not return to the boatshed

495 The applicant gave evidence that he was asleep in the boatshed before his mother called for him over the intercom. However, the clothing and shoes that he had been wearing the night before were found by police in the main house. The Crown relied upon this fact, together with other evidence, to submit to the jury that the applicant had not gone to the boatshed to sleep on the night of the murders but had remained in the main house, having planned to kill his parents and his brother. The Crown further submitted that the applicant took off his clothes and shoes before carrying out the killings, so as to ensure that they did not become bloodstained.

496 The Crown relied upon the presence of the applicant's keys in the main house to further support its case that the applicant remained in the house on the night of the murders and did not go to the boatshed. In the video walkthrough of the house conducted with the applicant the day after the murders, a set of keys can clearly be seen on a table on the upper floor of the house. The applicant was questioned about the keys by Detective Sergeant Ahern on the video:

"Q: Jeff, can you indicate as to who owns these keys?

A: They're, they're my keys.

...

Q: So it would be fair to say that these are the keys that you obviously have left here last night after leaving the boatshed?

A: Yeah, I would have."

497 According to the Crown, the applicant's response conflicted with a statement made earlier in the video that after he was called by his mother on the intercom he locked the boathouse door, took the keys, placed them in his pocket and ran upstairs. It was implausible, the Crown appears to have suggested, that the applicant removed the keys from his pocket and left them on the table either upon arriving at the house and being confronted with Christopher admitting to having murdered his parents, or after killing Christopher and before leaving to go to the Warners'.

498 The inference that the keys were left on the table is contradicted by other evidence. A crime scene video was taken of the house in the hours before the video walkthrough was conducted. Footage of the table in the crime scene video does not show any keys on it. During cross-examination, the discrepancy between the crime scene video and the walkthrough video was drawn to the attention of Det Sgt Ahern. He admitted that he did not know where the keys were first found or who had found them:

"Q. As it happens, do you know who found the keys?

A. No, I don't.

Q. Do you know where they were found?

A. No, I don't."

499 The applicant gave evidence that his agreement with Det Sgt Ahern's proposition that he had left the keys on the table the night before was based on an assumption that the police had found them on the table. The applicant said that he had no independent recollection of where the keys were found. In light of this evidence we do not believe that it can be said with any certainty where the keys were found or how they came to be upon the table.

500 The applicant explained that his clothing was found in the house because it was his practice to shower there before returning to the boatshed, leaving his clothes behind to be washed by his mother. This explanation is not improbable.

Siphoning of petrol

501 Mineral turpentine was used as an accelerant for the fire. A cut garden hose was found by the police along with about a litre of leaded petrol in a jerry can. The Crown relied upon this evidence to suggest that the applicant had attempted to siphon petrol from a motor vehicle to use as an accelerant and that it was only after the applicant had failed to successfully siphon enough petrol that he had decided to use mineral turpentine to light the fire.

502 The applicant told police that he siphoned petrol from a car with his father the night before the murders. On his account, the siphoning was for the purpose of acquiring petrol to be used to make two-stroke fuel for one of the Gilhams' motorboats. He explained the presence of the cut hose to police in the video walkthrough of the crime scene the day after the murders:

"Detective Sergeant Ahern

Q: All right. Jeff, it's been pointed out to us that there's been a - a siphon of some - a siphoning hose of some description...

A: Mm.

Q: Can you throw any light upon that at all?

A: That's from the top hose, the hose up near the boat in the carport.

Q: And did you use that hose last night?

A; Yeah, me and my father used that hose last night.

Q: Can you tell me for what purpose?

A: We were going to get some petrol for the boat which is down - down the bottom, but it's - it was the wrong kind of petrol so we didn't use it, so we didn't have to siphon anything out. We siphoned some out, but, not...

Q: So can you indicate to me as to where the petrol would have been placed here last night after it was siphoned out?

A: There wouldn't have been any, actually, probably only that much siphoned out from the jerry can out near the bin."

503 The applicant then indicated that there would have been approximately 3 to 4 cm of petrol in the jerry can.

504 Mr Elton, a close friend of Stephen Gilham, gave evidence that he did not believe that Mr Gilham would have siphoned petrol using a cut garden hose. He said it was his understanding that Mr Gilham had a pump or pumps fit for that purpose. Mr Elton said the following in his evidence in chief:

"Q. What do you say to the suggestion that Stephen would syphon (sic) petrol out of his car with a piece of hose just cut out of his garden hose?

A. I would think that Steve (sic) would have the proper equipment in his garage to syphon petrol.

...

Q. Have you been in Steve's garage?

A. In the upstairs - what do they call them - landing area where they used to park the boat and the car.

Q. Have you seen any syphoning equipment?

A. I haven't personally seen any, no.

Q. Do you know whether or not Stephen Gilham possessed a syphon?

BOULTEN: A syphon?

CROWN PROSECUTOR

Q. A pump?

A. A pump, yes.

Q. Do you know whether he had one or more than one?

A. I believe he had more than one.

Q. How do you come to that belief?

A. Just in talking.

Q. Was he the type of man who would have cut up a good garden hose?

A. I can't believe that.

Q. And why is that?

A. Well, why would you, if you have the facilities to use the proper equipment?

Q. Was he the sort of man who looked after his things?

A. Yes, yes.

Q. Was he a precise man?

A. Well, he was an instrument maker and, yes, I think he was precise."

505 As Mr Elton did not give evidence that he actually saw a pump in Mr Gilham's garage, little weight can be attached to his evidence that Mr Gilham had a pump. Moreover, to the extent that Mr Elton based his evidence on his opinion that Mr Gilham was "precise", "the sort of man who looked after his things", and not "the type of man who would have cut up a good garden hose", his evidence is not highly probative of whether Mr Gilham would have siphoned petrol from the vehicle in the manner suggested by the applicant. Mr Elton's evidence does not undermine the applicant's account to any significant degree.

506 The applicant gave evidence at the trial that he had no independent recollection of siphoning petrol from the car, but that he had told the police the truth in the video. This siphoning must have occurred late in the evening as the applicant's then girlfriend, Ms Moskos, did not leave the Gilham house until approximately 10 pm and the applicant was not yet home when she arrived at his house. In his electronically recorded interview with police, given before the walkthrough, the applicant made no mention of siphoning petrol from the car with his father after Ms Moskos left the house. When giving an account of the previous evening he told police only that he had watched television with his parents.

507 Significantly, there was no evidence that the applicant or his father had plans to go boating the following day. There was evidence that the applicant was scheduled to meet with Mr Nolan to go surfing and that Mr Gilham was to attend a golf day. The absence of evidence of plans to use the motorboat the following day raises questions about why the applicant and Mr Gilham would have been siphoning petrol from one of the cars at such a late hour. The implausibility of the applicant's account and the applicant's failure to disclose to police during the earlier interview that he had engaged in that activity raise doubts about the accuracy of his version of events. The applicant's account of the siphoning was properly available to the jury as a lie going to consciousness of guilt.

508 There was evidence from Mr Warner that he had smelled kerosene on the applicant's breath when he was asked by the 000 officer to smell his breath for alcohol. Fire Officer Wright gave evidence that he had smelled petrol on the applicant. The applicant's evidence that he had showered before bed made it unlikely that he would have smelled of petrol in the early hours of the morning if, as he claimed, he had been siphoning petrol with his father late the previous evening. However, we put little weight on this aspect of the evidence. Kerosene and petrol are different substances with different odours. In any event, on the applicant's account he had moved through the upstairs area of the house after Christopher had spread accelerant and had engaged in a struggle with him. It is quite possible that the smoke infused with mineral turpentine or the turpentine itself could have transferred to his hair and clothing at that time and that both Mr Warner and Fire Officer Wright had mistaken the smell of mineral turpentine for kerosene or petrol.

The plausibility of premeditation

509 There are aspects of the evidence which tend to undermine the Crown's submission that the crime was a highly premeditated one. The Crown was not required to prove that the crime was premeditated in order to establish the applicant's guilt. But given that premeditation was integral to the Crown's case theory, a failure to prove premeditation weakens the Crown case.

510 The applicant submitted that it is implausible that preplanning for the murder of his family would involve him openly admitting to killing his brother, the more so without claiming that it was done in self-defence. That defence is a well-known and relatively uncomplicated one. It was the more obvious argument for the applicant to have raised to seek to escape liability on a charge of murder or manslaughter. The defence of provocation is far less commonplace, and a person in the applicant's situation was unlikely to be aware of the way it operates at law.

511 In the electronically recorded interview conducted with police the following day, the applicant did not suggest that he acted in self-defence:

"Q: ...why would you kill your brother?

A: I don't know, I was just there and I just...

Q: Was he armed at the time?

A: I don't think so.

Q: Well, what do you mean, you don't think so? Was he armed or wasn't he armed?

A: I didn't see anything?

Q: So you're saying that your life was not in fear at the time that you stabbed your brother?

A: No, no it wasn't.

Q: So why did you stab your brother?

A: I just - just did. I saw everything there and I picked up the knife and chased him."

512 At the time of the murders it was most unlikely that the applicant could have expected, much less have planned for, the killing of his brother to be resolved by a plea of guilty to manslaughter by reason of provocation. Furthermore, even in the unlikely event that he knew that a verdict of manslaughter by reason of provocation was a possible legal outcome, he could not have expected, much less predicted, that he would not face a significant gaol term.

513 These matters, together with the use of a knife from the kitchen as the murder weapon, render the Crown submission that the murders were premeditated highly improbable.

514 The applicant's counsel also questioned the plausibility of the Crown's theory that the applicant attempted to siphon petrol from one of the vehicles after the killings took place. The theory that the applicant went to the carport (one of the few areas of the Gilham property that is visible from the street), cut a garden hose and attempted to siphon petrol from one of the cars with no guarantee of success and with a reasonable possibility that he may have been observed by neighbours sits uneasily with the Crown case theory that the applicant was engaged in the last phase of a premeditated crime. Having said that, as we noted above, the applicant's account of when and why the petrol was siphoned is also problematic.

515 Not having used the petrol that was siphoned from the car also creates a difficulty for the Crown case. In her closing address, the Crown Prosecutor put to the jury that it was unlikely that there was more than a litre bottle of mineral turpentine available in the household for use as an accelerant. If, as we are prepared to accept, 3 to 4 cm of petrol in the jerry can equates to approximately one litre of petrol, and even if the applicant had thought this insufficient to ensure that the bodies of the deceased were incinerated in the fire, it could be expected that he would have used the petrol he had siphoned in conjunction with the mineral turpentine. Furthermore, once he started siphoning the petrol there would be little reason for him to stop, unless of course the petrol tank was virtually empty (a matter about which there was no evidence). The inferences that are open to be drawn from this evidence are favourable to the applicant.

Blood-related evidence

516 A considerable amount of evidence was given at the trial about the blood (or lack thereof) found on the applicant and the knife, and the blood spatter on Christopher's body and in the downstairs area of the house.

517 Detective Rowan, a police officer who examined the applicant at Sutherland Police Station, gave evidence that there appeared to be blood on the inside of his right foot, on one of his fingernails, and on the inside of his shin. Det Rowan said that he had examined the whole of the applicant's body down to his boxer shorts and did not see any other markings that resembled blood. Mrs Warner gave evidence that she observed a small amount of blood on the applicant's lower leg that appeared to be pale and smeared. Mr Warner said that he noticed blood in the quicks of the applicant's fingernails.

518 The relative lack of blood on the applicant was relied upon by the Crown to support the submission that he had washed before raising the alarm. Dr Raymond and Dr Culliford were asked to offer an opinion as to whether the applicant should have had Christopher's blood on him, accepting that the applicant had stabbed his brother. Dr Raymond gave evidence that the amount of blood on the applicant was not inconsistent with him having stabbed Christopher, but during re-examination he expressed an opinion that he would have expected the applicant to have had blood on the ulna writer's palm aspect of his hand after stabbing a body 17 times. Dr Culliford gave evidence that she too would have expected more visible blood on the applicant, in particular on the ulna surface of the perpetrator's hand. In contrast to this evidence there was also evidence from Sgt Reynold and Dr Raymond that there was little projected blood where Christopher was found, which was consistent with there being minimal blood transferred to his killer. Sgt Reynolds also gave evidence that he had seen cases where considerable violence had been inflicted with very little blood on the offender, although in these cases there was a vast amount of blood at the scene and on the victim. To our minds, the inference that by reason of the lack of blood the applicant must have washed was not a compelling one.

519 The evidence suggested that there was little, if any, blood on the knife. This was relied upon by the Crown in support of the submission that the applicant had also washed the knife. We do not regard this as a submission of any weight. As we have already observed, by the time the knife was retrieved from beside Christopher's body, the area had been doused with water by the fire brigade and there was evidence, particularly from Sgt Reynolds and Dr Raymond, that it was not impossible that the water had resulted in any visible blood being washed off the knife.

520 Evidence from tests conducted shortly after the killings showed that blood on Christopher's shave coat was consistent with either his blood or the blood of his mother. The test results were not conclusive either way. The shave coat has since been lost. There was also evidence that Christopher had blood spatter on his right arm which was not tested. Dr Raymond gave evidence that the spatter could have come from either Christopher stabbing a bloodied body or his own bloodied hand.

521 For these reasons the evidence of the observed blood is equivocal. It does not strengthen the Crown case.

Dissuading Fire Officer Pelham from entering the house

522 Fire Officer Pelham gave evidence that the applicant had attempted to dissuade him from entering the house by telling him that someone was inside with a knife. At the time that this was said the applicant would have been aware that all occupants of the house were already dead. As discussed in relation to Ground 12, the Crown relied upon this evidence to argue that the applicant had attempted to prevent Mr Pelham from entering the house and putting the fire out too quickly, before it had destroyed any incriminating evidence.

523 Mr Pelham did not give a statement directly after the event and was first approached by the police in 2005.

524 None of the other fire officers present at the scene gave evidence that the applicant had made any attempt to prevent them from fighting the fire. As discussed in relation to Ground 12, the issue was not put to them directly as defence counsel was not made aware of the way in which the Crown intended to rely on the evidence until the applicant was cross-examined.

525 Mr Lowder, the most senior fire officer, gave evidence (consistent with a statement prepared at the time of the events) that the applicant was "extremely agitated" at the scene of the fire and that he had placed him in the care of Fire Officer Langdon:

"At the base of the stairs I observed a young male person approximately 6 foot tall with a solid build and he was only wearing a pair of shorts. I checked and saw that he was not armed and he was extremely [agitated] and screaming that his parents were inside the house and that they were dead.

Because of his close proximity to the house, which was now becoming totally involved in fire, I instructed Fire Fighter Langdon ...to remove him to the street level and to look after him."

526 Fire Officer Polson, who was in the company of Mr Lowder, referred to notes made after the fire in which he described the applicant as "delirious and disoriented" when he was encountered by fire fighters at the bottom of the stairs. Polson also gave evidence that the applicant was saying "something about people being stabbed". Fire Officer Langdon's evidence, also taken from a contemporaneous statement made to police, confirmed that he escorted the applicant to the street:

"As I was proceeding down the stairs I observed a young male person at the bottom of the stairs. The area was dimly lit, but I saw that he was young. He had mousey brown hair and would have been approximately 6 foot tall with a solid build. The person was of an Australian appearance. He was only wearing a pair of blue board shorts. He was screaming and appeared to be extremely agitated. He was screaming words to the effect of, 'Oh know' [sic]. He was very incoherent.

Station Officer Lowder then told me something. I then walked to the base of the stairs and grabbed him. I then walked him up to the external stairs and placed him against a vehicle at the front of the premises. I wrapped a blanket around him. He was continually mumbling but I cannot recall what he was saying. I then had a conversation with a neighbour and I left the male person in his care."

527 Fire Officer Wright gave evidence that he remained in the applicant's company at the top of the stairs for a few minutes after he was brought up to the front of the property by Langdon. Like Pelham, Wright was only approached to make a statement in 2005 and did not make any contemporaneous notes detailing the events at the fire. Wright gave evidence that the applicant was extremely agitated, but did not indicate that he had attempted to prevent him from fighting the fire:

"Q. While you were there did you see a young man who was clothed only in a pair of shorts?

A. Yes.

Q. Did you come to know that man's name?

A. Yes.

Q. Do you recall it now?

A. Jeff Gilham.

Q. Jeff Gilham?

A. Yes.

Q. What did you do with him that morning, if anything?

A. The other fire fighters went down to the fire and took a line of hose down to the fire to extinguish the house fire. One fire fighter brought Jeff Gilham back and put him with me behind the motor until other resources turned up.

Q. How far away were you from him during this time?

A. Fire fighter Langdon brought him up to me and told me what to do with him, and we put him on the edge of the kerb which is right near the back of the truck.

Q. And that's what you did?

A. Yes.

...

Q. Did you notice anything about the way that Jeffrey Gilham was behaving or did he just sit quietly and do nothing?

A. No, he was highly agitated.

Q. What was he doing that suggested to you that state?

A. Well, he couldn't sit still. He was standing up, sitting down, standing up.

Q. Did you notice he was upset or distressed?

A. Yes.

Q. Was he crying?

A. No.

Q. What distress could you see then if he was not crying?

A. He was just agitated; you could just see it. If your house was on fire you'd be agitated."

528 The evidence that the applicant had attempted to dissuade Pelham from entering the house in an attempt to ensure that the fire continued to burn may also be contrasted with other evidence that suggests he was anxious to ensure that the fire was put out. During the call to 000 made from the Warners' the applicant explicitly requested that the 000 officer send the fire brigade:

"Operator: Listen. Look, I'm going to send the police out and see you. You stay there with Ted.

J Gilham: Fire br - Fire brigade.

Operator: Yeah, I'll send the fire brigade too. Alright? Ok?

J Gilham: Yeah."

529 Mrs Warner gave evidence that after the 000 call was made the applicant grabbed a small fire extinguisher and attempted to take it to go and fight the fire. Mrs Warner said that she dissuaded the applicant from attempting to fight the fire by telling him that the fire brigade was on the way. The Crown portrayed the applicant's determination to fight the fire as a feigned act of concern.

530 Ultimately, even if it is accepted that the applicant told Pelham not to enter the house, we place little significance on this evidence. It is clear from the evidence of other fire officers present at the Gilham house that morning that the applicant was in an extremely distressed state and was speaking incoherently about his parents and the manner of their death. When viewed in this context, it is difficult to view his comments to Fire Officer Pelham as a calculated effort to prevent him from fighting the fire.

The applicant's appearance and demeanour at the Warners

531 The applicant was said to be noticeably cold to touch when he arrived next door. On the applicant's account he had left his burning house after struggling with and killing his brother and had then run directly to his neighbours. The Crown suggested that the applicant's body temperature was not consistent with his account and was more consistent with him having washed and spent time outside in the night air waiting for the fire to take hold before raising the alarm. We place little weight on this evidence. On both the Crown scenario and the defence scenario the applicant arrived at the Warners' house after a series of traumatic events. The evidence of those he had contact with at the scene plainly suggested that, whether or not he killed his parents, he was in shock. It was a cold winter's night and he was scantily clothed. It would be no surprise that in the circumstances the applicant would have seemed cold to touch and may have been damp when he arrived at the Warners'.

532 It was submitted by the Crown that it was telling that the applicant was not observed to be breathless upon his arrival at the Warners' house, but only commenced sounding breathless when speaking to the 000 operator on the phone. We do not regard this evidence as significant either. Whether he had killed three people, or only one, a range of physical responses to the situation, even extraordinary responses, could be expected.

Panadol and syringes

533 It was the Crown case that the applicant planted the Panadol paste and syringes in the downstairs laundry basket in order to implicate his brother in his parents' murder. In support of this theory, the Crown pointed to syringes and plastic tubing that were retrieved from the upstairs bedroom previously occupied by the applicant.

534 In his interview with police the applicant said that he intended to use the syringes found in the upstairs bedroom in a model he was required to create as part of his university studies in civil engineering. In cross-examination, the applicant gave evidence consistent with this account. He elaborated on that account by explaining that he had asked his mother to obtain the items for use in a university project that related to water flow in gutters. The following exchange took place in cross-examination:

"Q. All right. Well, then, let me ask you then about Panadol, Panadol paste, and the objects that were found by the police in the laundry basket on the vanity next to the basin in that bathroom downstairs?

A. Mm.

Q. Did you put any of those objects there?

A. No.

Q. Did you carry the laundry basket with those objects into that room?

A. No.

Q. Did you have any prior knowledge about how the Panadol was drawn into a syringe before the police told you about it when you were at the police station?

A. No.

Q. Did you do any of it?

A. No.

Q. There were syringes and needles and plastic tubing apparent in the upstairs bedroom, that is, the room that you used to occupy. Do you remember seeing those items on the crime scene video?

A. Yes.

Q. Did you know that those objects were there at this time?

A. Yes.

Q. How did they come to be there?

A. I asked my mother to get them for me.

Q. Why?

A. I was doing a project at Uni to do with water flow, and I wanted something that would drip in some sort of ink into the water.

Q. What was the project about?

A. It was about flow in gutters, but it was different to the one I ended up doing in the end after I went back to Uni.

...

Q. All right. Did you know that your brother had in his room a plastic bag that also contained similar sorts of objects?

A. No, I didn't.

Q. Do you know any reason why your brother would have such objects?

A. No."

535 The Crown sought to contradict the applicant's explanation of the use to which he had planned to put the syringes and plastic tubing. Professor O'Loughlin, who supervised the applicant in relation to a project he undertook while a student in 1994, was asked about whether the applicant's project required the use of syringes and plastic tubing. He gave the following evidence:

"Q. Have you been provided with a series of photographs by a detective depicting a syringe with a hypodermic needle and some white coloured paste inside it, an empty blister packet of Panadol with a glass cup?

A. Yes. As far as I can remember that was about three and a half years ago.

Q. Were you also shown some photographs of blue coloured tubing and syringes?

A. Yes.

Q. You were told, were you not, sir, that the accused had said, "I've got some syringes, but I was going to use them at Uni for my model"?

A. Yes.

Q. Did you supervise a project that the accused did in 1994?

A. Yes, I did.

Q. Did those items in the photograph have anything to do with that project?

A. No.

HIS HONOUR

Q. Did syringes generally have anything to do with that project?

A. Not that particular project; possibly an earlier project.

CROWN PROSECUTOR

Q. It's the case, isn't it, that things likes syringes and the blue tubing are commonly used in engineering projects at university to inject into models and simulate water movement and so on?

A. Yes, at that time a number of projects were done in laboratories and those types of things would have been used in various experiments with models of flowing water."

536 Professor O'Loughlin's evidence was put to the applicant in cross-examination:

"Q. But we have heard from Associate Professor O'Loughlin in this trial that in 1994 you did undertake a project, but that was about the flow of water in gutters, is that correct?

A. That's right.

Q. Was that the same project that you were thinking about doing in 1993?

A. I don't - it might have been similar, but it might have been - I think it's probably a different project, but similar - there was a lot of projects about gutter flows at that time to do with roofs, if that was at all what they were doing.

Q. Where did you get the idea to use a syringe and plastic tubing?

A. I don't know exactly. I may have got it from one of the lecturers. I think the lecturer - I can't remember his name at the moment. It escapes me."

537 In our view, the suggestion that the applicant lied about the use to which he proposed to put the syringes and plastic tubing is not persuasive. Professor O'Loughlin could not rule out that the applicant may have used the objects in earlier projects or projects in relation to which he did not supervise the applicant. This was consistent with the applicant's evidence that he had undertaken more than the one project during his time as a student and that many projects at the time related to gutter flow. There is also the consideration that the applicant undertook the 1994 project with Professor O'Loughlin well after the killings took place in 1993.

538 In the result, we are of the view that the presence of the Panadol-filled syringe in the downstairs bathroom does little to support either case. As the applicant suggested, the presence of the syringe may suggest that Christopher was intending to use it to commit suicide after killing his parents. On the other hand, the lack of fingerprints on the syringe may support the Crown's submission that the applicant was engaged in an elaborate plan to frame his brother as the perpetrator of his parents' murder.

The position of the sliding door

539 The applicant told police that when he ran up to the house from the boatshed the glass sliding door to the house was open. When the fire brigade arrived the door was closed. The Crown submitted that the applicant having closed the door was inconsistent with him leaving the house in a hurry and suggested that the applicant had closed the door in an attempt to accelerate the spread of the fire before waiting outside for the fire to take hold. The applicant was unable to explain why he closed the door behind him as he made his way out of the burning house. On the other hand we do not see that the door being closed adds materially to the weight of circumstances in the Crown case probative of guilt, or that the door being open when the applicant arrived the house adds very much to the applicant's exculpatory version of events (unless perhaps it gave Christopher a ready means of escaping the fire).

Conclusion on Ground 13

540 There are some features of the evidence that individually and collectively support a hypothesis consistent with the applicant's innocence. These include the lack of any credible motive and the evidence tending to suggest that the applicant did not premeditate the killing of his parents. We note also the neutrality or weakness of much of the evidence that was said to establish guilt, such as the blood (or lack thereof) on the applicant, Christopher and the knife; the applicant's being wet or cold to the touch on arriving at the Warners'; the applicant's closing the door as he left the house; and his supposed attempt to dissuade Fire Officer Pelham from fighting the fire. On the other hand, there are other features of the Crown case that are probative of guilt. The fact that Christopher was dressed only in a shave coat and was not wearing his glasses; the applicant's inconsistent statements about whether Christopher was armed; the location of the knife he claimed to have armed himself with to attack his brother; and his explanation for siphoning petrol the previous night are in this category.

541 The most significant evidence persuasive of guilt is the carbon monoxide evidence as it was presented at the trial. Dr Lawrence gave evidence that, given the concentration of carbon monoxide in his blood, Christopher was dead before the fire was started, which according to Mr Munday was sometime between 4.17 am and 4.29 am and most likely at about 4.22-4.23 am.

542 Dr Lawrence's account cannot be reconciled with the applicant's account that his brother ignited first his father and then his mother and that the fire was spreading when he arrived at the house. The cumulative force of the timing evidence and the carbon monoxide evidence, taken with the other evidence to which we have referred, excludes the applicant's account as a reasonable hypothesis consistent with his innocence.

543 Moreover, it must be borne in mind that this is a case where the jury did have a real advantage in being able to see and hear the applicant give evidence. They alone had the opportunity to evaluate the credibility of his denial of the Crown case and his explanation of the events which are prima facie inconsistent with his account. We have not had the opportunity to assess the applicant as a witness. In order to have been satisfied of the applicant's guilt, the jury must have rejected the applicant's evidence as untrue. Moreover, the case against the applicant was not so inadequate or lacking in probative force that the jury's advantage can be put to one side: M v The Queen at 494; MFA v The Queen [2002] HCA 53; (2002) 213 CLR 606 at [56] (McHugh, Gummow and Kirby JJ). We accordingly reject this ground of appeal.

544 The resolution of this ground of appeal does not by itself determine what orders the Court should make in disposing of the appeal. The other grounds of appeal, in particular that relating to the new carbon monoxide evidence, remain to be considered.

Ground 14: The applicant seeks to rely upon fresh evidence that the applicant's university timetable indicates that he would have been attending university on the afternoon that his mother left her place of employment two hours earlier than usual, indicating at that time that she was worried about one of her sons. The applicant seeks to rely upon fresh evidence from Steven Audet that the applicant attended a university lecture with him on the same Thursday afternoon. The corollary is that the son to whom the applicant's mother was referring was Christopher Gilham, not the applicant.

Ground 15: The applicant seeks to rely upon fresh evidence that there was a bloodied fingerprint on the intercom unit, thereby indicating that someone - who could only have been the now deceased mother of the applicant - had touched the intercom with her finger, which was consistent with the applicant's case at trial, and given contemporaneously to investigators, that she had contacted him by the intercom.

545 The question whether evidence not led at trial by an accused is fresh evidence or new evidence is decided according to settled principles. Those principles were set out by McClellan CJ at CL in Wood v R [2012] NSWCCA 21 at [706]-[714]:

"The general principle is that parties to litigation, including a criminal trial, are bound by the manner in which they presented their cases at first instance: Khoury v The Queen [2011] NSWCCA 118 at [104] (Simpson J, Davies J and Grove AJ agreeing). In a criminal trial there is an obligation on the prosecution to disclose all relevant evidence to the accused. There is no obligation on an accused person to seek out information which the prosecution is obliged to produce: Mallard v The Queen [2005] HCA 68; (2005) 224 CLR 125 at [16]-[17]; Grey v The Queen [2001] HCA 65; (2001) 75 ALJR 1708 at [23].

The law makes a distinction between "new evidence" and "fresh evidence." "New evidence" is evidence that was available and not adduced at the trial. "Fresh evidence" is evidence which either did not exist at the time of the trial or, if it did, could not then have been discovered by an accused exercising due diligence: R v Abou-Chabake [2004] NSWCCA 356; (2004) 149 A Crim R 417 at [63] (Kirby J, Mason P and Levine J agreeing).

In Ratten v The Queen [1974] HCA 35; (1974) 131 CLR 510 at 516, Barwick CJ (McTiernan, Menzies, Stephen and Jacobs JJ agreeing), in his analysis of what may constitute a miscarriage of justice, referred to a category of instances of miscarriage as including the "production of evidence not available to the appellant at his trial." His Honour said:

'The rule in relation to civil trials is that evidence, on the production of which a new trial may be ordered, must be fresh evidence; that is to say, evidence which was not actually available to the appellant at the time of the trial, or which could not then have been available to the appellant by the exercise on his part of reasonable diligence in the preparation of his case. However, the rules appropriate in this respect to civil trials cannot be transplanted without qualification into the area of the criminal law " (emphasis added).

His Honour went on to say at 517 that:

"It will not become an unfair trial because the accused of his own volition has not called evidence which was available to him at the time of his trial, or of which, bearing in mind his circumstances as an accused, he could reasonably have been expected to have become aware and which he could have been able to produce at the trial. Great latitude must of course be extended to an accused in determining what evidence by reasonable diligence in his own interest he could have had available at his trial, and it will probably be only in an exceptional case that evidence which was not actually available to him will be denied the quality of fresh evidence " (emphasis added).

And later at 520:

"To sum up, if the new material, whether or not it is fresh evidence , convinces the court upon its own view of that material that there has been a miscarriage in the sense that a verdict of guilty could not be allowed to stand, the verdict will be quashed without more. But if the new material does not so convince the court, and the only basis put forward for a new trial is the production of new material, no miscarriage will actually be found if that new material is not fresh evidence" (emphasis added).

In Gallagher v The Queen [1986] HCA 26; (1986) 160 CLR 392 at 395, Gibbs CJ said in relation to s 6(1) of the Criminal Appeal Act :

"The circumstances of cases may vary widely, and it is undesirable to fetter the power of Courts of Criminal Appeal to remedy a miscarriage of justice. I respectfully agree with the statement of King C.J. in Reg. v. Mclntee [(1985) 38 SASR 432 at 435], that 'appellate courts will always receive fresh evidence if it can be clearly shown that failure to receive such evidence might have the result that an unjust conviction or an unjust sentence is permitted to stand.'

The authorities disclose three main considerations which will guide a Court of Criminal Appeal in deciding whether a miscarriage of justice has occurred because evidence now available was not led at the trial. The first of these, that the conviction will not usually be set aside if the evidence relied on could with reasonable diligence have been produced by the accused at the trial, is satisfied in the present case, and need not be discussed, although it should be noted that this is not a universal and inflexible requirement: the strength of the fresh evidence may in some cases be such as to justify interference with the verdict, even though that evidence might have been discovered before the trial ..." (emphasis added).

The Chief Justice later said at 399:

"It seems to me, with all respect, that where the trial was by jury, the accused was entitled to have the question of his guilt determined by the verdict of the jury, and that the Court of Criminal Appeal, in considering the effect of the fresh evidence, should consider what effect it might have had upon a reasonable jury. It is not enough that there is a bare possibility that a jury might have been influenced by the evidence to return a verdict of not guilty. On the other hand, it is too severe, and indeed speculative, a test, to require that the court should grant a new trial only if it concludes that the fresh evidence was likely to have produced a different result, in the sense that it would probably have done so."

There is a difference in approach between fresh evidence and relevant evidence not disclosed by the prosecution to the defence in a criminal trial. Where evidence has not been disclosed, the discussions in Grey and Mallard are authoritative and apply. The prosecution must disclose all relevant evidence to an accused, and a failure to do so may require the quashing of a verdict of guilty: Mallard at 133 [17] (Gummow, Hayne, Callinan and Heydon JJ); Grey at 1713 [23] (Gleeson CJ, Gummow and Callinan JJ). In relation to evidence to which this common law obligation attaches, "there [is] no reason why the defence in a criminal trial should be obliged to fossick for information of this kind and to which it was entitled": Grey at [23] (Gleeson CJ, Gummow and Callinan JJ).

Where such evidence remains unpresented at trial, it is not the function of an appellate court to "seek out possibilities, obvious or otherwise, to explain away troublesome inconsistencies which an accused has been denied an opportunity to exploit forensically": Mallard at 135 [23]. Where evidence has a capacity to discredit the prosecution case, these matters are of significant forensic value: Mallard at 135 [23], 141 [42].

546 Even were we satisfied that the timetable evidence and the intercom evidence are fresh evidence as contended for by the applicant, the cogency of that evidence in the context of the evidence that was led at trial bears upon whether we are satisfied that there has been a miscarriage of justice justifying an order for a new trial. We were not invited to find that the timetable evidence or the intercom evidence is of such cogency that we should be satisfied that there is a reasonable doubt about the applicant's guilt to warrant the convictions being quashed and verdicts of acquittal being entered.

The relevance of the evidence concerning the applicant's university timetable

547 The defence called evidence from members of the applicant's extended family, and a number of his friends, to establish that Christopher was experiencing emotional difficulties on a number of fronts in the weeks preceding the killings. These included the breakdown of the relationship with his girlfriend and his dissatisfaction and frustration with a recent practical teaching course that his parents had encouraged him to undertake with a view to him securing employment as a teacher when his preferred career path in chemical engineering did not pan out. The applicant gave evidence that the conflict between Christopher and his parents had erupted on one occasion into a physical confrontation between Christopher and his father which involved pushing and shoving. The applicant's girlfriend gave evidence that on the night of the murders she sensed that Christopher was "edgy", that he was not making eye contact with her, and that there was some degree of tension between Christopher and his parents.

548 It was also the defence case that while Christopher was something of a loner, the applicant was socially and emotionally well-adjusted, making solid progress through his course in engineering at the University of Technology, Sydney (UTS). The applicant gave evidence that he had a good relationship with his parents and a particularly strong relationship with his father. This was supported by the observations of a number of witnesses.

549 It was the Crown case that the applicant had manufactured an undercurrent of conflict and dissent between his brother and his parents as part of a premeditated plan to kill his family and then blame his brother for the murders of Mr and Mrs Gilham.

550 In the first trial the Crown sought to lead evidence from Mrs Obsorne, one of Mrs Gilham's former work colleagues, that she (Mrs Gilham) had left work early in the afternoon of Thursday, 26 August 1993, two days before she was killed. Mrs Osborne said that during the course of the day Mrs Gilham had appeared agitated and had expressed concern about one of her sons. She was reported to have said that she was "very worried and didn't know what to do" and to have been uncharacteristically abrupt with a client before leaving work two hours before her usual departure time of 5 pm, saying she had to get home. Mrs Osborne did not claim that Mrs Gilham said she was going home to meet with the son she was concerned about, although that is an available inference.

551 The applicant successfully objected to Mrs Osborne's evidence on the basis that she was unclear as to whether it was the applicant or Christopher that Mrs Gilham had expressed concern about (although she thought it was the applicant). The same ruling applied in the second trial.

552 In conjunction with other evidence not led at trial, the applicant now seeks to rely upon Mrs Osborne's evidence to establish that his mother must have left work early to go and meet with Christopher. According to the applicant, his university timetable of the time shows that he was at a lecture at UTS in Ultimo on Thursday afternoon between 1.30 and 4.30 pm. He submitted that taking into account the traveling time from the city to Sutherland, he could not have returned home to meet his mother until the early evening at the earliest. This evidence is then relied upon in further support of the defence case that Christopher was in fact a troubled person, that he was in a disturbed state of mind on the Thursday before the murders and, in that state of mind, killed his parents in the early hours of Saturday morning.

The "fresh" evidence

553 Carolyn Hunt and Stephen Audet were fellow students of the applicant in 1993. Stephen Audet was a friend of the applicant at that time, up to and including the trial. There is no evidence that the applicant knew Ms Hunt. Professor Geoffrey O'Loughlin was a member of the academic staff at UTS in 1993. They each gave evidence on the appeal by affidavit. The Crown objected to the tender on the basis that their evidence was neither fresh nor new. The Court received the evidence provisionally. The witnesses were not required for cross-examination.

554 Although the applicant claimed the evidence was fresh evidence it was not submitted, nor was it suggested, that any of the witnesses were unavailable to be called to give evidence at his trial. Rather, it was submitted that because Mrs Osborne's evidence had been excluded, Mr Audet's evidence that he recalled the applicant being with him at a lecture at the time Mrs Gilham left work on the Thursday before she was killed was simply not relevant. For that reason no inquiries were made at that time with a view to supporting his evidence as to the applicant's whereabouts by independent evidence, including locating the relevant lecture timetable. Those inquiries were not made until July 2009 when the applicant's wife made email contact with Professor O'Loughlin for the purposes of the appeal. The timetable was not located by Ms Hunt until March 2010.

555 In the initial email to Professor O'Loughlin the applicant's wife said she was working from the applicant's academic transcripts for 1993 with a view to understanding his movements in the week before his parents were killed and, in particular, which days of that week he attended lectures. Professor O'Loughlin was able to locate a list of the subjects the applicant was enrolled to study in the Spring semester of 1993 but was unable to locate any reference to lecture times or the identity of the applicant's project supervisor. He said he would continue to make inquiries.

556 After the concerted efforts on the part of Professor O'Loughlin and a number of members of academic and administrative staff to locate this information, at the end of 2009 he informed the applicant's wife that there was no material available from the university or the university staff to confirm the precise time at which the lecture in transport engineering was held in the Spring semester of 1993.

557 In November 2009 the applicant's wife forwarded a group email to a large number of former students at UTS who may have been enrolled in transport engineering for the Spring semester, and who may have been able to give evidence that the lecture was scheduled on Thursday afternoons, whether by reference to diary entries or other documentation or from memory. No evidence was forthcoming.

558 Ms Hunt first searched for the timetable in November 2009 upon receipt of the group email, which was forwarded to her by a friend. She was unable to locate the timetable at that time. In March 2010, she located her UTS civil engineering course timetables for the period 1990 to 1995 (excluding 1994) by chance among her belongings, which she had stored before leaving Australia to work first overseas and then interstate. According to the 1993 Spring semester timetable the relevant lecture was held between 1.30 and 4.30 pm on Thursday afternoons.

559 The applicant submitted that the combined effect of Ms Hunt's and Mr Audet's evidence, in the context of Mrs Osborne's evidence, constitutes fresh evidence that supports the defence case and that its unavailability at the time of trial was productive of a miscarriage of justice.

The fresh evidence concerning the presence of a bloody fingerprint on the intercom

560 An edited version of the video of the crime scene was tendered at trial. It was of 50 minutes' duration and filmed in real time. At about the forty-first minute, and over a two-second interval, the gloved thumb of a crime scene investigator can be seen to pass over the front panel of the intercom, apparently making contact with a latent mark or impression of some kind and appearing to smear it. The intercom was located on a sideboard above where Mrs Gilham's body was found. The sideboard and the intercom were both damaged by the fire. The underside pages of a document which appear to have fallen from the wall above the sideboard protected part of the face of the intercom from the fire.

561 The applicant relies upon that two-second segment of the crime scene video, extracted and separately tendered on the appeal, and slowed down to one-tenth of real-time speed, together with a series of still photographs of successive frames from the video, as constituting fresh evidence. He submitted that the mark on the intercom is consistent with a fingerprint deposited by a finger wet with blood, which was smeared when it was touched by the investigator's gloved hand. He relies upon that evidence as supportive of the account he gave to investigating police at the time of the killings, and repeated in his evidence at trial, that he received a call for help from his mother over the intercom at a time when Christopher must have been in the process of attacking her or his father. He submitted that the bloodied fingerprint left by her as she raised the alarm was evidence of that fact.

562 The Crown case at trial was that the applicant manufactured his account of responding to his mother's call for help over the intercom in order to implicate his brother in the murders.

563 It is common ground that the segment of the crime scene video showing the gloved hand appearing to smear a latent mark on the intercom did not attract the attention of investigators at the inquest, or the Crown or the defence in the first or second trial. It was also common ground that it was identified for the first time by Mr Mackie, a friend of the applicant, in July 2009 when he was viewing the video to assist the applicant's wife to prepare for this appeal.

564 The appearance and composition of the mark has been the subject of extensive attention by forensic experts retained by the applicant and the Crown since that time. In the result, and in summary, there is no conclusive evidence that the mark on the intercom was a fingerprint before it was smeared, or apparently smeared, or that the mark was imprinted in blood or by blood, although that cannot be discounted as a possibility.

565 The applicant called Professor Spring to give evidence of the assumption inherent in the applicant's claim that the segment of the video is fresh evidence, namely that the mark is consistent with a bloody fingerprint. A number of questions were posed for his consideration, which he addressed in his report:

"Is the substance seen to smear on the intercom blood or a substance like blood?

Does the mark prior to it being smeared resemble or has (sic) the characteristics of a fingerprint or transferred finger mark in blood?

Is there any identifiable ridge details or other hallmarks consistent with the mark being a fingerprint?

Are any of the attributes of the mark consistent with this being a fingerprint or finger mark with reference to shape and size?

Could a mark be created by another mechanism, for example blood spatter?"

566 The Crown objected to Professor Spring giving evidence of his opinion concerning most of these matters because he was not shown to have the necessary expertise. The Crown also objected to his evidence on the basis that it was neither fresh nor new.

567 Professor Spring gave evidence that he held the position of Associate Professor of Scientific Photography at RMIT University, Melbourne, where he has been actively engaged in the analysis of photographic materials and images for forensic purposes for the last 22 years. Prior to coming to Australia, he worked at the University of Texas South Western Medical School for 11 years, where he documented autopsies photographically as well as photographing a variety of objects allegedly used in the commission of criminal offences. His duties included the documentation of blood drops and blood splatter. He said photographing blood specifically, or as an artefact or at a crime scene or on a weapon, was a routine part of his daily activities. He said he was required to produce accurate photographs of blood whether it was dry, wet, blotted, absorbed or separating as well as photographs designed to accurately reproduce the size, shape and thickness of blood on a variety of surfaces, including on body tissue and on hard and soft surfaces. He gave evidence that he worked under the direction of the medical examiner and forensic scientists and, through this exposure to their expertise, and in discussion with them, he had learnt to recognise a deposit of blood in order to reproduce it photographically, and to interpret photographs taken of items of forensic interest to determine whether or not they revealed deposits of blood.

568 He also gave evidence that he used various photographic processes to enhance the appearance of latent fingerprints. He said that he has photographed objects believed to contain latent fingerprints, but he has no expertise in fingerprint analysis or comparison.

569 After hearing argument we received Professor Spring's evidence, but limited that evidence to the following conclusions which we were satisfied he had the relevant expertise to make:

(a) traditional photography and videoography are unsuitable for identifying whether the mark or substance on the intercom is blood; and

(b) fingerprint ridge marks cannot be seen on the video as the resolution of the original video camera, and the scanning technology used to produce the image on the video, would most likely obscure the fine details in a fingerprint if it were present.

570 The Crown called evidence from Dr Raymond, the chief scientist in the Forensic Sciences Services Branch of the Forensics Services Group of the New South Wales Police Force. He was formerly Director of the National Institute of Forensic Science and, prior to that, Director of the Forensic Services Group. He holds a number of postgraduate degrees and a doctorate from La Trobe University, Melbourne, for a dissertation entitled "Trajectory reconstruction from blood stains at a crime scene". He has had significant involvement in developing blood pattern analysis curricula for institutional purposes as well as extensive practical involvement in blood pattern analysis and interpretation.

571 He examined the extracted portion of the crime scene video and the still colour prints of the suspected fingerprint at various magnifications. He said that while it appeared to have a general shape and dimension similar to a fingerprint, there was nothing to indicate any finger mark direction or ridge detail which would confirm the mark as a fingerprint. He described the unsmeared mark as elliptical in shape and of a consistent colour, best described as brown or maroon. He said that the absence of a scale or accurate reference point to assess the size of the mark relative to a finger, together with the curve on the front panel of the intercom having the capacity to distort the shape of the mark and limit its utility for comparing it against the size or shape of Mrs Gilham's finger, were additional limiting factors.

572 Dr Raymond agreed that the video showed the crime scene examiner's finger brushing through the mark five times in one direction and in the opposite direction at least once. He said it was impossible to indicate what pressure was being applied by the operator's finger but that the smearing only appeared after the fourth pass across the face of the intercom. He said this gave rise to two interrelated questions. The first is what, if anything, that exercise revealed about what the smear was comprised of and the second is whether the mark that was smeared was consistent with a bloodied fingerprint.

573 He said that the way a bloodied fingerprint is deposited, and the amount of discernible ridge detail, will depend on the amount of blood available to be deposited, its consistency and spread over the finger, and the degree of pressure that is applied. He said that he was unable to determine what material the mark or the smeared mark was comprised of by viewing the video or the stills, or whether the smearing was as a result of water on the intercom or adjacent to the mark, or the result of the residue of water or some other material on the operator's gloved finger passing through the mark. However, if the smearing was because of fluid picked up by the investigator's hand, he would have expected to see the concentration of blood in the mark lessen as it moved from the perimeter of the print to what he described as the "two horns" within the mark (which he marked on the still photographs).

574 He said that it was well established that latent marks, particularly blood marks, generally maintain their shape in a fire and that latent fingerprints, whether bloodied or not, can even be enhanced in a hot fire. To test the resilience of blood under heat he conducted a series of controlled tests on a number of devices similar to the intercom. He applied blood to the test devices for minutes, hours and days and then exposed them to radiant heat of up to about 80 degrees for approximately ten minutes, after which the unit started to soften, buckle and ultimately scorch.

575 With the qualification that the tests were not determinative, given the significant number of variables, he confirmed that the results demonstrated that blood is resistant to smearing, even to pressure by the application of a dry finger, once it has been baked. The blood stains also darkened after baking. He said that resistance is a function of temperature and the duration of exposure to heat. On this basis he concluded that unless the investigator's gloved finger was moist the smeared mark was unlikely to be blood, assuming the mark was exposed to sufficient ambient heat before the fire was extinguished for it to be baked hard. Under test conditions, when blood stains that were left for between ten minutes and one hour and then baked for ten minutes were subject to a wet glove, they were completely removed by the application of pressure. In addition, he said if the investigator's gloved hand were dry, and if the mark were dried blood applied within hours of being touched by the investigator and baked by the fire in the interim, he would have expected to see some abrading of the mark as the gloved finger passed over it.

576 Although Dr Raymond could not categorically eliminate the possibility that the mark was left by a bloodied finger, he said:

"The mark here has the appearance to me of being a pressure mark where, if it were a blood mark, it would be quite a bit of blood that has been pressed on by the finger lightly onto the intercom. That means that there's a reasonable amount of blood that has been heated, but because there's a reasonable amount there, it is far more likely when it is rubbed with a dry gloved hand for the particles of blood to actually break off or be removed or disturbed. I must hasten to add it depends very much on how long it has been there and how hot it has been baked because the blood increasingly fixes with temperature and length of time that it is at a reasonably hot temperature."

Is the timetable evidence fresh evidence or new evidence?

577 The applicant submitted that the timetable is fresh evidence because it could not have been located with reasonable diligence at the time of trial. This is said to follow logically from the combined effect of Professor O'Loughlin's evidence, the negative response to the group email sent in November 2009, and the fact that the timetable was only discovered by chance in March 2010 after Ms Hunt had returned to Australia.

578 Given the latitude which must be extended in determining what evidence was discoverable by reasonable diligence, we accept that the timetable is fresh evidence in the relevant sense. Although the respondent resisted that categorisation of the evidence, it did accept that the production of the timetable by Ms Hunt is new evidence, but submitted that it was of so little weight in the context of the evidence led at trial that its non-production was not productive of a miscarriage of justice.

579 However, we are unable to regard Mr Audet's evidence that he was with the applicant in a lecture on the Thursday before his parents were killed as fresh evidence, as it was referred to in a character reference tendered in the applicant's sentencing proceedings and, it must be assumed, was something which was always known to him. The only explanation for not calling Mr Audet to give that evidence at trial appears to be that the successful objection to Mrs Osborne's evidence in the first trial, a ruling which was not revisited at the second trial, made his evidence irrelevant. It is of course the applicant's intention to rely upon Mrs Osborne's evidence at any retrial, in which case Mr Audet's evidence would be relevant, and it is only in that sense that his evidence is new evidence.

Is the evidence "credible" or plausible or at least capable of belief?

580 On the assumption that Mrs Osborne's evidence is to be relied upon by the applicant at any retrial, we are satisfied that Mr Audet's evidence, supported by reference to the timetable produced by Ms Hunt, is of sufficient cogency to support the conclusion that Mrs Gilham did not leave work early to meet with the applicant and may have left early to meet with Christopher. Together with other evidence relied upon by the applicant at his trial suggestive of Christopher's unsettled state of mind, we are also satisfied that this evidence provides some support for the defence case that Christopher murdered his parents.

When examined in the context of the evidence that was led at trial is the fact that the timetable evidence was not called at trial productive of a miscarriage of justice such as to justify an order for a new trial?

581 Despite the fact that the timetable evidence is, in part at least, fresh evidence and that it is credible and capable of supporting the defence case, when weighed together with the other evidence called at trial, we are not satisfied that it can be confidently said that the evidence would have been likely to have caused the jury to have entertained a reasonable doubt about the guilt of the applicant such that a miscarriage of justice resulted and a new trial should be ordered.

582 Accordingly, we are not satisfied that Ground 14 has been made out.

Is the intercom evidence fresh evidence or new evidence?

583 The fact that the suspect mark on the intercom was not seen at trial or at any earlier time in the course of the investigation, and for that reason was not the subject of investigation for the purposes of leading expert evidence about it at trial, is not decisive of its categorisation as fresh evidence. Neither is the fact that it was first identified by a friend of the applicant in the process of assisting in the preparation of the appeal. What cannot be avoided is the conclusion that the mark on the intercom was always available to be relied upon by the applicant, assuming a reasonably diligent eye was looking for some evidence of Mrs Gilham having touched the intercom, and that someone, upon seeing it, formed the view that the elliptical-shaped mark in brown or maroon might be a bloodied fingerprint.

584 A review of evidence in a circumstantial case of the kind assembled in this case will inevitably uncover matters of fact overlooked. That is what has occurred here. New evidence has been identified. We are not satisfied that reasonable diligence on the part of the applicant would not have revealed the mark such as to satisfy the test for fresh evidence. It is not suggested that it was first noticed by Mr Mackie because the speed of the video was altered or because he used any special optical devices to enhance his viewing of it, or that he had any particular level of visual acuity.

Is the evidence of Dr Raymond credible, plausible or at least capable of belief?

585 Dr Raymond's evidence, taken at its highest, cannot discount the theoretical possibility that the mark on the intercom was a bloodied fingerprint before it was smeared and largely obliterated by the investigator's finger, although he thinks it improbable, for the reasons he articulated in his evidence. Professor Spring's evidence adds nothing to that assessment. We are not satisfied that the evidence is of a sufficient quality to meet the test of plausibility or cogency in the context of the evidence led at trial to qualify for reception as new evidence.

586 Accordingly, Ground 15 is not made out.

Ground 16: The applicant seeks to rely upon new evidence that the level of carbon monoxide in the lungs of the deceased Christopher Gilham indicates that he was alive and probably upstairs at the time of the fire

587 This ground of appeal seeks to rely upon new evidence with respect to carbon monoxide levels revealed in the deceased on post-mortem. The starting point for the consideration of this ground of appeal is to identify the evidence at the trial bearing upon the question and the Crown case with respect to it.

588 In her closing address, the Crown Prosecutor told the jury:

a) he [Stephen Gilham] was dead before any fire started; and

b) she [Mrs Gilham] was also negative for smoke inhalation, so "all her wounds had been inflicted while she was still living ...", but she had not inhaled any carbon smoke, so was dead by the time the fire started.

589 Although the Crown Prosecutor made no submission to the jury concerning whether Christopher was alive or dead at the time the fire was ignited, or whether he could have inhaled any smoke, it was the Crown case that Christopher was not upstairs during the attack on his parents or at the time the fire was lit or gathering momentum. The Crown submitted that Christopher was killed downstairs, after the applicant had killed his parents.

590 Dr Lawrence performed the post-mortem examinations of each of the deceased in 1993. The Crown did not establish the extent of his experience as at 1993; however, by the time of the trial Dr Lawrence estimated that he had performed somewhere between 3,000 and 4,000 post-mortem examinations.

591 Dr Lawrence was asked about a particular aspect of the post-mortem examination of Mr Gilham that touched upon the level of carbon monoxide in his blood. The evidence was as follows:

"Q. Now did you cause the usual screening to be done in relation to Stephen Gilham on the question of ingestion of alcohol and drugs?

A. Yes, toxicology was performed.

Q. Was it the case that there were no drugs or alcohol detected in his body?

A. There were no drugs or alcohol detected and there was a carbon monoxide level of four per cent.

Q. Four per cent carbon monoxide?

A. Yes.

Q. What does that mean in terms of smoke inhalation?

A. It means for all intents and purposes there was no smoke inhalation.

Q. Does that mean then that Mr Gilham was dead when the fire commenced?

A. Yes."

592 Dr Lawrence also gave evidence about the post-mortem examination of Mrs Gilham. On the issue of carbon monoxide, he gave this evidence:

"Q. In relation to Mrs Gilham, was it the case that she also had a nil result for the presence of alcohol and drugs?

A. Yes, that's correct.

Q. And what about her level of carbon monoxide ingestion?

A. Her carbon monoxide level was measured at three per cent.

Q. Is that a negative result for your purposes?

A. Yes, for my purposes, I would regard it as a negative result.

Q. In that case, does it appear that she was not alive when the fire was set?

A. Yes."

593 In dealing with the post-mortem examination of Christopher, Dr Lawrence gave this evidence:

"Q. What about his carbon monoxide level?

A. His carbon monoxide level is six per cent. Again, I would be inclined to view that as being a negative result.

Q. So he was dead when the first started?

A. Yes, I think so."

594 There was some brief cross-examination of Dr Lawrence about the carbon monoxide levels:

"Q. Now I want to ask you about carbon monoxide levels. You gave evidence before lunch in a question led in these terms by the learned Crown Prosecutor, accepting the proposition that Christopher died before the first was lit.

A. Yes.

Q. There are other possibilities, are there not?

A. Possible explanations for ...?

Q. Yes, possible explanations.

A. As I have said, I don't think there is a significant difference between the carbon monoxide levels in any of them to be quite frank.

Q: They're all been normal levels?

A. Yes.

Q. Stephen's level, Helen's and Christopher's levels are all within reasonable limits.

A. Yes.

Q. So, really, apart from concluding that none of them had inhaled a lot of carbon monoxide before they died, no other conclusion can be safely drawn, do you agree with that?

A. Yeah, that's a reasonable assessment, that none of them had absorbed significant amounts of carbon monoxide.

Q. And that's the only scientific conclusion one can draw?

A. In the normal course of things, in looking at a person with stab wounds and fire damage, a level of less than 10 per cent would be assumed to be that the person had died before the fire started.

Q. I know you've said that, but I want to remind you of evidence that you gave ... are you sure about this topic at the inquest?

A. Yes.

HIS HONOUR: Is it that he died before he inhaled smoke?

BOLTON: That's what I'm asking the witness.

A. Yes, he did not inhale significant amounts of smoke.

HIS HONOUR: That's all you can say isn't it? You can't say anything about when the fire started, but that he didn't, that is, Christopher downstairs, didn't inhale a significant amount of smoke before he died.

A. Yes, that's correct."

595 Dr Cala, a forensic pathologist, was also called to give evidence in the Crown case. He obtained his qualification as a pathologist in 1994 (that is, a year after the post-mortem examinations were undertaken). By the time of the trial he was an experienced forensic pathologist.

596 He was not asked any questions on the carbon monoxide levels in the deceased revealed upon post-mortem.

597 From the evidence led at trial it was generally accepted that:

(a) the levels of carbon monoxide in the bodies of Mr and Mrs Gilham and Christopher were four per cent, three per cent and six per cent respectively;

(b) these were within normal limits; and

(c) each of them was dead when the fire started.

598 The applicant adduced new evidence on the appeal concerning these questions.

Professor David Penney

599 Professor Penney is a specialist toxicologist. He holds a Doctorate in Philosophy from the University of California, Los Angeles (1969), having obtained earlier primary qualifications in biology and chemistry from Wayne State University, Detroit (1963). He has been studying the effects of carbon monoxide since at least April 1974, when he received a research grant from the United States Public Health Service to study the chronic effects of carbon monoxide on the heart.

600 He is the author of over 100 peer-reviewed articles published in respected scientific journals since 1963, most of which deal with issues related to carbon monoxide inhalation, toxicity and poisoning. He has authored chapters in a number of publications dealing with carbon monoxide toxicity and is the author of four books which deal with all aspects of carbon monoxide poisoning. He has given lectures and presentations on carbon monoxide toxicology for the World Health Organisation and various educational bodies in many countries around the world.

601 The Crown did not challenge his expertise or the relevance of the opinions he expressed in his report. He was not required for cross-examination.

602 In the course of submissions, the Crown accepted that Professor Penney's evidence was new evidence and of a quality that this Court would be entitled to consider in determining whether there has been a miscarriage of justice.

603 In Professor Penney's opinion:

"... none of these [carbon monoxide] levels [in the blood of the deceased] should be considered normal for normal, adult, non-smoking humans. ... [Dr Lawrence's] conclusion that 'none of them had absorbed significant amounts of carbon monoxide' is in error, because the values are above normal, meaning they had to have absorbed significant exogenous CO [carbon monoxide] to reach these COHb [concentration of CO in blood] levels. All three individuals appear to have absorbed some CO from the fire smoke/fumes, with Christopher Gilham absorbing the most. ... In my opinion, Christopher Gilham inhaled a significant amount of smoke before he died. This is in sharp contrast to the opinion of Dr Lawrence that he did not ... CO remains in the body for a very short period of time (that is, half life in air four to five hours) leaving quickly when the breathing of fresh air occurs."

604 Professor Penney considered the normal concentration of carbon monoxide in blood and noted:

"... the normal blood level of COHb in healthy non-smoking adults is 0.4-1.4 per cent. The handbook of the instrument used to measure the Gilhams' COHb, the Radiometer Hemoximeter gives a range of 0.0-0.9 per cent. Normal COHb cannot be said to extend as high as 10 per cent. That is the COHb level seen in humans exposed to 70-80 ppm CO for 10-12 hours. That is not normal, is not allowed by law, and we now know leads to serious health risk. That is the COHb level seen only in a few of very heaviest cigarette smokers. Again, 10 per cent COHb is not normal."

605 He went on to conclude:

"1. That the carboxyhemoglobin (COHb) values for blood of normal, non-smoking, adult humans is routinely observed in the range 0.4% to 1.4%, encompassing as it does some 99% of such individuals.

2. That the Radiometer Hemoximeter, Model OSM2, instruction manual (handbook) states the COHb 'fraction' (i.e. saturation) for adults (12 subjects) is 0.0-0.9%.

3. That a normal, non-smoking, adult human found to have 6% COHb, alive or dead, has an abnormally high (i.e. elevated) level.

4. That such an individual in 3 must of necessity have recently taken up the additional CO load from an exogenous source.

5. That Christopher Gilham inhaled a significant amount of smoke ... before he died, because CO remains in the body for a very short period of time, leaving quickly when the breathing of fresh air occurs.

6. That in my understanding, the only likely or probable source of respirable CO at the house ... on the night of August 28, 1993 was fire purposely started in that structure.

7. That based on the 'inaccuracy, repeatability, and uncertainty' parameters published for the Hemoximeter, Model OSM2, COHb saturations observed in the normal range (above) for Christopher Gilham, could not have been indicated (i.e. read) as 6% by this instrument, i.e. again, his COHb level was abnormally elevated.

8. That dead bodies do not take up additional CO after death, only before death, and that the COHb level measured after death is the COHb saturation that was present at the instant of death."

606 In a second report dated 12 November 2011, discussing the report of Dr Anthony Green dated 16 December 2010, Professor Penney said:

"...this consultant is in substantial agreement with Dr Green's modelling of CO uptake over short periods of time (2-4 minutes) and the concentrations of CO in the air required to produce six per cent COHb."

607 From this it is open to conclude that it was also Professor Penney's opinion that in order for Christopher to have the six per cent level of carbon monoxide in his blood after death, he needed to have inhaled smoke for a period of between two to four minutes prior to death.

Dr Anthony Green

608 Dr Green's report dated 16 December 2010 was tendered on the appeal. Dr Green did not give evidence at trial. In this report, he came to the following conclusions:

"(a) The conditions in the house prior to the explosion of the windows, did not allow smoke and carbon monoxide to move against the ventilation flow that was coming from below.

(b) It was not possible that carbon monoxide from the upper level was responsible for the raised carbon monoxide level in Christopher Gilham if he had been downstairs at all times.

(c) It is necessary to assume that Christopher Gilham was inhaling carbon monoxide from the fire over a two to four minute period, in order for the level of carbon monoxide in his blood to be six per cent.

(d) The raised levels of carbon monoxide in Christopher Gilham's blood was (sic) compatible with his being in the vicinity of the fire in the upstairs lounge room."

609 In this report, Dr Green supported his conclusions, in particular that relating to the two to four minute time period, with modelling of carbon monoxide uptake. It was this modelling to which Professor Penny referred, and with which he agreed.

610 In his report dated 15 November 2011, Dr Green reaffirmed his view that a level of six per cent carbon monoxide in the blood was not a normal level, indicating that it is 4.5 standard deviations above the level of a non-smoker.

611 He expressed the opinion that the flued wood-fired heater, which was located in the downstairs area of the Gilham house, was not responsible for the levels of carbon monoxide recorded in the blood of any of the deceased. He set out three reasons for this: first, that there was no evidence that the stove was lit; second, that because the heater was flued, any carbon monoxide would be vented to the outside of the house, above the rooftop level; and third, that the differential levels of carbon monoxide in the deceased indicated that they were not exposed to carbon monoxide from the heater, because one would have expected their exposure to result in similar levels.

612 He regarded this analysis as reinforcing his earlier opinion and the view that the only source which could have exposed Christopher to the elevated level of carbon monoxide was the fire on the upper level of the house.

613 Dr Green gave evidence and was cross-examined on the appeal. He did not moderate his view.

614 He was cross-examined also about the possibility of carbon monoxide leaking from the wood-fired heater, in the event that the flue was blocked. In re-examination, Dr Green expressed the view that although carbon monoxide was odourless, if it were to be produced from a fire lit in the wood-fired heater, and the flue was not working properly so as to prevent ventilation through to the atmosphere above the roofline, someone in the house would notice both the smell and the smoke associated with the wood which was being burned.

615 In re-examination, he said that if the flue were not working, because it was blocked, it would be an obvious malfunction because byproducts of a fire, such as fumes and smoke, would be readily detected visually and by the smell.

Dr Christopher Lawrence

616 As set out earlier, Dr Lawrence's evidence concerning the levels of carbon monoxide can be reduced to two related propositions:

(a) The carbon monoxide levels of all three members of the Gilham family were within normal levels because they were each less than 10 per cent.

(b) Each of the deceased was dead before the fire started.

617 Dr Lawrence was called by the Crown to give evidence on the appeal, by which time he had read Professor Penney's report. The Crown led the following evidence from him:

"Q. In conference on the telephone with me last week, did I ask you some questions in relation to carbon monoxide haemoglobin levels?

A. Yes you did.

Q. Did you indicate to me in conference that those levels may indicate that each of the three people may have been alive when the fire started?

A. Yes, if you look at the totality of the evidence here. There is no, there is no macroscopic evidence or no visible evidence of smoke in Stephen's airway, but there is some blood there which could obscure a little bit of smoke. Under the microscope there is a very small amount of smoke and --

Q. A small amount of --

A. A small amount of carbon --

Q. Whereabouts did you observe a small amount of carbon?

A. In the lungs.

...

Q. Then in relation to Stephen Gilham in the second report, what did you note in relation to observations in the second autopsy report?

A. An examination of the lung under the microscope - there is a small amount of carbon or black pigment in, around the bronchi.

Q. Where is that in your report?

A. That is in the microscopic examination on page 6 of the final report.

Q. There is some black pigment around the bronchi?

A. Yes.

Q. At page 7 of that report, at point 6 of the report, you noted that there was extensive post-mortem fire damage with extensive burning of the anterior chest wall, face, arms and legs and noted (a) no evidence of smoke inhalation.

A. Yes.

Q. This is under - what's that a reference to? No evidence explaining ...

A. Well, its', I'd modified the thing - I probably should have qualified that to being minimal evidence of smoke inhalation. I think the fact is that the naked eye examination revealed no soot. The histological examination revealed a little bit of soot, but I'm not 100% certain about the significance of histological examination.

Q. Why is that?

A. It is possible to contamination of it, but I think in light of the CO2, the CO level and so forth, I think that it's quite possible that he had taken some breaths during the time of the fire.

Q. That's a combination of the microscopic examination ....

A. Yes.

Q. And the carbon monoxide ...

A. There's also quite a lot of blood in the upper airways and it's possible that the blood could obscure the carbon on direct examination.

Q. In relation to the upper airways?

A. Yes.

Q. You made no observations of any ...

A. I didn't see any, I didn't see any soot and at the time that I did the case, it was my impression that he was dead at the time the fire started. In light of the 4% CO and the carbon in the lungs, it is possible that he did breath a small amount during the fire."

618 The Crown then dealt with the position so far as Mrs Gilham was concerned. The relevant evidence was as follows:

"Q. So going on to the final autopsy report for the Coroner, at page 2, did you confirm your observation that on an external examination, the nose contained a small amount of soot.

A. Yes.

...

Q. At page 5 of your final report, under the heading Respitory System, did you note there is no obvious soot in the upper or lower airways?

A. Yes.

Q. Can you tell me is this a macroscopic or microscopic ...

...

A. Sorry, macroscopic examination.

Q. Macroscopic examination, thank you. And at page 7, is that where you detailed your microscopic examination?

A. Yes. The microscopic examination, the lung shows haemorrhage. There is a small amount of soot and blood in the bronchus.

...

Q. At page 9 of that report, did you state under the heading "Opinion", "it appears that the deceased was dead when the fire started as there is no soot in the trachea and a very low level of carbon monoxide in the blood" ?

A. Yes, and I probably qualified that to say there could have been some terminal breaths that had occurred at the time the fire started, so it may be that there was a small amount of smoke inhalation.

....

Q. Could that have been the case if the person was dead before the fire started?

A. There was soot on the face and I have specifically commented about soot in the nostrils, so I think there was probably more than perhaps could be accounted for by the soot on the face. In relation to the interpretation of the microscopic examination, as I have said, it is quite sensitive but not very specific and it could be contaminants. I tend to work on my visual inspection rather than my microscopic, but in alight of a 3% carbon monoxide level, soot in the nostrils and soot in the lungs, I think it's possible that she could have inhaled some fire gasses before she died.

...

Q. And what you observed was microscopically small amounts?

A. Small amounts of it, but nonetheless, I think the most likely explanation for it is that there may have been a little bit of smoke inhalation."

619 Dr Lawrence confirmed that on both macroscopic and microscopic examination during Christopher's post-mortem he did not see any soot. He proffered a possible explanation for the absence of soot:

"Again, given the presence of a large amount of blood, it is possible that a small amount of soot may not be visible, but in this case when I looked under the microscope I didn't see any soot in the lungs under the microscope."

620 Dr Lawrence was cross-examined by counsel for the applicant on the appeal. He gave this evidence:

"Q. And without taking you to that evidence again, do you agree, generally speaking, that it was wrong to say that the CO level was not significant given what you know now?

A. I think it's probably better to say I don't know what the significance of the level is. In most of our cases we deal with either levels of zero or over 10%. I deal relatively infrequently with cases between zero and 10% because I don't mostly - mostly see live people and so my experience and knowledge of interpretation of that sort of level is small."

621 Under further questioning Dr Lawrence conceded that he was not adequately qualified to offer an expert opinion about the significance of a level of carbon monoxide between zero and 10 per cent. He also said that he did not inform the prosecutor of his limited level of expertise before giving evidence at the trial. He agreed that the view which he expressed in his evidence on appeal was different from his evidence at the trial.

Other possible sources of carbon monoxide

622 On the appeal, the Crown pointed to other possible sources of carbon monoxide which might explain the varying carbon monoxide levels in the blood of the deceased and which would leave intact the Crown submission that the deceased were dead when the fire was lit, or at least not lead inevitably to a finding that they were alive at that time and that Christopher was alive for between two and four minutes while the fire was burning.

623 The evidence establishes that smokers will have higher carbon monoxide levels than non-smokers and that smokers may have levels of up to ten per cent, depending upon the extent of their use of tobacco. On appeal, Mr Nolan (the applicant's friend) deposed to the fact that none of the family members were smokers. Although the Crown objected to this affidavit upon the basis that the contents of it constituted neither fresh evidence nor new evidence and therefore were not admissible, it did not otherwise challenge the facts Mr Nolan deposed to. The Crown did not adduce evidence to suggest that any of the deceased members of the Gilham family were smokers, nor did it cross-examine Mr Nolan to suggest that his evidence was wrong.

624 We are satisfied that cigarette smoke can be excluded entirely as a possible explanation for the higher than normal carbon monoxide levels in the blood of the deceased.

625 Dr Green confirmed that carbon monoxide was neither lighter nor heavier than air, but rather about the same weight, and accordingly, ambient room temperature controlled where carbon monoxide levels would be concentrated.

626 On the appeal, the Crown called Mr Warner, who gave some evidence about the general use of pot-bellied stoves in the suburb of Woronora and the surrounds. He was unable to say whether, on the evening of 27 August 1993, the pot-bellied stove in the Gilham house was or was not lit. There was no evidence from any witness at trial or on the appeal which suggested that the wood-fired heater had been lit on that evening.

627 There was evidence that within the Gilham house alternative sources of heating from a number of electric heaters, particularly downstairs, were available. There is no evidence as to whether the electric heaters were or were not switched on that evening, although given it was winter there is an available inference that some heating was used.

628 But even if the pot-bellied stove was operating on the evening of 27 August 1993 and into the early hours of 28 August, the Crown would have to demonstrate two further matters in order for the stove to be a source of exposure to carbon monoxide such as to explain the elevated levels in Christopher's blood. Those two matters are whether there was an undetected defect in the fluing system which prevented the ordinary convection processes from drawing the carbon monoxide up the flue and releasing it into the atmosphere, and that this defect affected the flue on the lower level. There was no evidence of either of these two matters.

629 On the evidence on appeal, this Court is unable to find that the wood-fired heater was a source of the carbon monoxide found in the blood of the deceased members of the Gilham family, there being no evidence that the stove was actually lit on the evening in question; no evidence that there was any defect in the fluing system, let alone one in the downstairs section of the house, which could have caused or permitted the escape of carbon monoxide into the atmosphere downstairs; and no evidence at trial or on the appeal which cast any light on either issue.

630 Accordingly, the only possible source of the carbon monoxide levels detected at post-mortem was the fire that was deliberately lit and which substantially destroyed the house. Although it was the Crown case at trial that each of the deceased were dead at the time the fire started, if, having regard to Professor Penney's evidence, Mr and Mrs Gilham were alive when the fire started (even if in the throes of death), as their bodies were located at or near the seat of the fire, that remains the obvious source of the carbon monoxide levels in their blood. No damage is done to the Crown's case by this analysis.

631 However, the same cannot be said of Christopher. On the Crown case he was never near the seat of the fire or where it was raging, the applicant having killed him downstairs. Accordingly, in light of the new evidence, and assuming no other theory is advanced, it was necessary for the Crown to satisfy the Court that there was sufficient smoke that travelled downstairs from the fire upstairs to enable him to passively inhale the smoke for between two and four minutes before his death. If he did not, a fundamental factual premise underpinning the Crown case falls away.

632 Dr Green and Mr Munday agreed that there were two possible mechanisms whereby smoke from a fire commencing upstairs in the house could travel to the lower level. These mechanisms can be described, in a shorthand way, as the pressure mechanism and the entrainment mechanism.

633 Dr Green noted that a considerable amount of water was sprayed onto the top level of the house in order to bring the fire under control, to the extent that water soaked through from the upper level of the house to the lower level, both via the stairs and the stairwell. He said that water moving by this mechanism actually entrains smoke, with the effect that smoke is carried to the lower level, but that he would not expect a person downstairs to breathe in the smoke that was entrained in this way.

634 He was cross-examined about whether there could have been smoke downstairs after the fire was lit upstairs as a consequence of a mechanism other than entrainment by water. He was asked whether pressure, which had built up in the upstairs area prior to the windows blowing out, could force the smoke downstairs. His answer was this:

"I don't think so. The first, prior to the blowing of those windows, was in an unventilated state in that it didn't have enough air supply. So one of the major sources of air supply is actually from the floor below and up the stairwell. I think the pressures involved in that movement would not have been overcome by pressure building up within the upper of the fire in the upper floor."

He said that in his opinion it was not possible for smoke to have been forced downstairs to the extent that someone on the floor downstairs could have been breathing in smoke.

635 The Crown accepted that this theory did not advance the Crown case in any event because any movement of smoke by this mechanism could only have occurred after 4.45 am, that being the earliest point in time that water was sprayed on the Gilham house, and it was not feasible that Christopher was still alive at that time having regard to the number and location of his chest wounds.

636 The second possible method for the transmission of the smoke generated from the fire on the upper level was by pressure. As we understand the mechanism, given that the fire commenced upstairs, it depended upon the build-up of pressure on the upper level to the extent that the following occurred:

(a) The upper level windows were still intact so that the upper level was effectively a sealed level.

(b) The upper level had filled with smoke at a pressure higher than the pressure of the smoke-free atmosphere on the lower level.

(c) The smoke held under higher pressure on the upper level was forced by the differential pressure down to the lower level.

(d) Having entered the lower level, and because it was still hotter than the air temperature at the lower level, sufficient time had elapsed for the smoke to be distributed, under continuing higher pressure, throughout the lower level so that it reached within a short distance of the floor where it would have been inhaled by a supine Christopher Gilham.

637 There are a number of difficulties with this mechanism as a means of explaining the levels of carbon monoxide in Christopher's blood. First, Dr Green did not express the view that this mechanism was likely. His attention was drawn to the observation of the first fire officer to enter the downstairs level (apparently at a time after the bedroom window on the upper level had exploded, thereby allowing the pressure upstairs to dissipate and equalise with atmospheric pressure) that the room was only partially smoke-filled. The smoke was observed to be visible at ceiling level.

638 Second, Mr Munday described the pressure theory in this way:

"Q. What do you consider as the most likely explanation for why there was smoke in the bottom area of the main house?

A. Well, the other potential explanation is that the upstairs loungeroom/bedroom area became pressurised as a result of hot combustion gasses being evolved during the fire and that that pressurisation, prior to the windows failing, was sufficient to push some smoke downstairs against the prevailing convection up the spiral staircase. That would be an alternative. I can't say which is more probable."

639 Mr Munday did not embrace the pressurisation theory as a practical explanation for the inhalation of carbon monoxide by the deceased. He was asked the following question by the Crown:

"Q. When smoke had been forced downstairs, if a person was downstairs and standing, is it possible that they could inhale carbon monoxide?

A. If a person was standing in the downstairs part of the house at the point where pressurised smoke was being forced down, yes it is possible, although I would expect that smoke to gather initially under the ceiling, so I would think it's not likely."

640 Third, Mr Munday in cross-examination agreed that he would expect that had smoke been present downstairs, there would be a defined stain on the walls caused by the condensation of the liquids within the gas layer combined with the carbon particulates. He gave the following evidence:

"Q. And if for any reason the fire had descended to the lower floor, would you expect to see a smoke mark on the walls of the lower floor?

A. Yes, I think I would expect to see a smoke horizon at the depth of the smoke of it had come down before the windows failed.

Q. Yes, so therefore if there was no smoke mark on the lower floor, you could assume that the smoke came down at a later stage?

A. Yes, that's true."

641 There was no evidence at trial of the existence of any mark on the walls indicating the lowest level which the smoke reached. The crime scene video does not demonstrate the existence of any smoke mark on the downstairs walls.

642 At its highest, the evidence at trial was that fire officers saw the downstairs room partially filled with smoke at or around ceiling level. There was no evidence that the smoke approached within a few inches of the floor, which the Crown accepted on the appeal was essential if the Crown case advanced at trial that Christopher had at no time been upstairs, or involved in the lighting of the fire, was to remain undisturbed.

Summary

643 We are satisfied that the evidence enables the following conclusions to be safely drawn:

1. Consistent with the proper concession by the Crown, the evidence of Professor Penney was admissible on the appeal as new evidence (as defined) and is of such quality as to be available for considering as to whether there has been a miscarriage, as was other evidence on the appeal dealing with the same subject matter.

2. Contrary to the way in which the case was put at trial, the evidence before this Court demonstrates that each of the deceased members of the Gilham family were alive when the fire which destroyed the house was lit.

3. Christopher was exposed to the byproducts of fire, including carbon monoxide, for between two and four minutes prior to his death.

4. There is no rationally available, or acceptable, alternative source of carbon monoxide to which Christopher was exposed, and which could account for the level of carbon monoxide in his blood, other than the fire which destroyed the house.

5. There is no evidence, or persuasive inference, which is available to suggest that Christopher could have been exposed for the necessary period (at least two minutes) while downstairs in the house, let alone while supine.

644 These conclusions contradict two central elements of the Crown case as presented at trial. First, that Christopher was never upstairs, although he may have been briefly on the lower part of the staircase, and second, that each of the members of the Gilham family were killed within five to ten minutes of 3.57 am, as the earliest that the fire could have been started was sometime shortly after 4.15 am, probably around 4.22 am.

645 By contrast, the applicant's evidence that Christopher was upstairs at the time that he entered the house from the boatshed, and that he was at that time setting his parents alight, provides a plausible explanation for the level of carbon monoxide in Christopher's blood. Whether the new evidence corroborates the applicant's account to the extent that we are satisfied of his innocence is unnecessary for us to determine. What can be said is that the new evidence is in closer conformity to the applicant's exculpatory account than the Crown case theory advanced at trial.

646 It is inevitable from this that we reach the conclusion that the applicant has lost a fair chance of acquittal, and that a miscarriage of justice has occurred.

647 McCLELLAN CJ at CL: Having regard to the findings the Court has made it is necessary to consider the orders I should make. Section 6(1) of the Criminal Appeal Act 1912 provides: "Subject to the special provisions of this Act, the court shall, if it allows an appeal under section 5(1) against conviction, quash the conviction and direct a judgment and verdict of acquittal to be entered." One of the "special provisions" to which s 6(1) refers is s 8(1), which provides:

"On an appeal against a conviction on indictment, the court may, either of its own motion, or on the application of the appellant, order a new trial in such manner as it thinks fit, if the court considers that a miscarriage of justice has occurred, and, that having regard to all the circumstances, such miscarriage of justice can be more adequately remedied by an order for a new trial than by any other order which the court is empowered to make."

648 Whether there should be an order for a new trial under s 8(1) is a discretionary consideration. However, the discretion is only to be exercised if the court determines that the evidence presented at trial was sufficiently cogent to justify a conviction, for if it was not, an acquittal must follow as a matter of course: Director of Public Prosecutions (Nauru) v Fowler [1984] HCA 48; (1984) 154 CLR 627 at 630; Gerakiteys v The Queen [1984] HCA 8; (1984) 153 CLR 317 at 322 (Gibbs CJ), 331 (Deane J). For the reasons discussed under the unreasonable verdict ground of appeal, the evidence before the jury was sufficiently cogent to justify the applicant's conviction. It is therefore necessary, in view of the other successful grounds of appeal, to consider whether the applicant ought to be acquitted or retried.

649 The relevant principles were discussed in Reid v The Queen [1980] 1 AC 343 and Fowler: R v Anderson (1991) 53 A Crim R 421 at 453 (Gleeson CJ). The overriding consideration is whether the interests of justice require a new trial: Fowler at 630. Unless the interests of justice require the entry of a verdict of acquittal, an appellate court should ordinarily order a new trial of a charge where a conviction in respect of that charge has been set aside but there is evidence to support the charge: Spies v The Queen [2000] HCA 43; (2001) 201 CLR 603 at [104]. The court determines where the interests of justice lie by considering various factors, including:

• the public interest in the due prosecution and conviction of offenders (R v Taufahema [2007] HCA 11; (2007) 228 CLR 232 at [49]; Anderson at 453; Reid at 349);
• the seriousness of the alleged crimes (Anderson at 453; Reid at 350; Haoui v R [2008] NSWCCA 209; (2008) 188 A Crim R 331 at [164] (Johnson J));
• the strength of the Crown case (Anderson at 453);
• the desirability, if possible, of having the guilt or innocence of the accused finally determined by a jury, which, according to the constitutional arrangements applicable in New South Wales, is the appropriate body to make such a decision (Taufahema at [51]; Anderson at 453; Reid at 350);
• the length of time between the alleged offence and the new trial, and in particular whether the delay will occasion prejudice to the accused (Taufahema at [55]; Parker v The Queen [1997] HCA 15; (1997) 186 CLR 494 at 520 (Dawson, Toohey and McHugh JJ); Anderson at 453);
• whether the grant of a new trial would impermissibly give the prosecution an opportunity to supplement or "patch up" a defective case or to present a case significantly different to that presented to the jury in the previous trial (R v Wilkes [1948] HCA 22; (1948) 77 CLR 511 at 518; King v The Queen [1986] HCA 59; (1986) 161 CLR 423 at 433 (Dawson J); Parker at 520 (Dawson, Toohey and McHugh JJ); Reid at 350; Fowler at 630; Anderson at 453; Taufahema at [59]);
• the interests of the individual accused, and in particular whether it would be unduly oppressive to put the accused to the expense and worry of a further trial (Spies at [103]; Reid at 350);
• whether a significant part of the sentence imposed upon conviction has already been served (Jiminez v The Queen [1992] HCA 14; (1992) 173 CLR 572 at 590 (McHugh J));
• the expense and length of a further trial (Reid at 350);
• whether a successful appellant to the Court of Criminal Appeal has been released from custody (Taufahema at [55]; Everett v The Queen [1994] HCA 49; (1994) 181 CLR 295 at 302 (Brennan, Deane, Dawson and Gaudron JJ); R v Wilton (1981) 28 SASR 362; (1981) A Crim R 5 at 367-68 (King CJ)); and
• whether an acquittal would usurp the functions of the properly constituted prosecutorial authorities, which are entrusted with responsibilities and discretions to act in the public interest in the initiation and conduct of criminal prosecutions (R v Thomas (No 3) [2006] VSCA 300; (2006) 14 VR 512 at [27]).

650 The applicant relied heavily on the authority of this Court in Anderson to argue that a third trial would give the Crown an opportunity to reconstruct its case, which it cannot do. It is necessary to set out the facts of that case. The Crown had relied on the testimony of a witness whose evidence was proved during the course of the trial to be unreliable. The Crown Prosecutor, in his closing address, modified the Crown's case theory in support of the guilt of the accused such that it did not depend on the witness' evidence. On appeal, the Court of Criminal Appeal held that the appellant's convictions for murder were unsafe. The Court declined to order a retrial, partly on the basis that the prosecution ought not to have been given the opportunity to "patch up" its case against the appellant: at 453 (Gleeson CJ, Finlay J and Slattery AJ agreeing).

651 The applicant submitted that for the same reason a new trial was not ordered in Anderson, a new trial should not be ordered in this case. It was said that the Crown, by advancing a case theory that was based on a wrong view of the significance of the levels of carbon monoxide in the blood of the deceased, had bound itself to that case theory. The applicant said that a new trial cannot be ordered as a matter of fairness to him, as it would require the Crown to argue for a different timeframe and sequence of events than that presented at the previous trials. The applicant submitted that the position with regard to a new trial, based on a revised case theory, is no different from that which obtains when the Crown impermissibly attempts to change its case theory during the course of a trial. In each case, the accused has been denied the right to know of the case that he or she must meet, including the factual transaction that underlies the charge: Johnson v Miller [1937] HCA 77; (1937) 59 CLR 467 at 489 (Dixon J).

652 Whether an acquittal should be ordered on the basis of the principle identified in Miller and applied in Anderson turns on what it means for the Crown to put on a "new case". In Taufahema, a majority comprising Gummow, Hayne, Heydon and Crennan JJ reviewed the authorities on this subject and concluded at [67] that "the difference between the case relied on in a first trial and the case to be relied on in a second trial must be substantial if the difference is to stand as a bar to an order for a second trial". The plurality held that an order for a second trial was appropriate because the Crown proposed to rely on the same evidence as was called at the first trial, the only difference being that the prosecution would "seek to characterise the facts which that evidence may establish in a different way, but not a radically different way" (at [68]). Their Honours distinguished the case before them from Jiminez, where the Crown proposed to retry the accused in a case that would necessarily rely on "events different in time, place and quality": Taufahema at [65].

653 The question is whether the case theory on which the Crown proposes to rely at a third trial is substantially different from that which it relied on at the previous trials. This is largely a question of degree, to be answered in the light of the way in which the prosecution wishes to conduct a further trial: Taufahema at [54].

654 The Crown indicated that if the case were to run again, and if the Crown accepted Professor Penney's evidence, it would be forced to make certain concessions. The Crown would be forced to concede that Christopher was exposed to carbon monoxide emanating from the fire for two to four minutes before his death. The Crown would therefore have to acknowledge that Christopher was upstairs when or after the fire was lit, which is consistent with the applicant's exculpatory version of events. The Crown would also have to acknowledge that Mr and Mrs Gilham inhaled some carbon monoxide emanating from the fire before they died. This latter adjustment to the Crown case is less problematic than the former, as the bodies of Mr and Mrs Gilham were located at or near the seat of the fire. It is arguable that they inhaled the carbon monoxide after they were stabbed but before they finally died. But the contention that Christopher was exposed to carbon monoxide before his death marks a more significant change to the Crown case. As originally framed, the Crown case assumed that the applicant killed Christopher downstairs, away from the seat of the fire and before it was lit.

655 The proposed adjustments to the Crown case, while not insubstantial, are not so substantial as to preclude an order for a new trial. As in Taufahema, the Crown proposes to rely on the same evidence that was led at the previous trials, albeit with the qualifications that flow from Professor Penney's interpretation of that evidence. The Crown proposes to present the sequence and timing of events that the evidence is said to establish in a different way, but the difference is not so radical as to give the Crown the opportunity to make a new case. The basic factual premises that underlie the Crown case - that the applicant killed his parents and his brother and lit a fire to cover his tracks, all within a short space of time - will remain the same at any new trial. All that would change in light of Professor Penney's evidence is the order in which the events allegedly took place, the location of Christopher, and the time that elapsed between the killing of Christopher and the applicant's going next door (which, if Professor Penney's evidence is accepted, must be significantly shorter than the Crown argued at the trial). These changes would certainly pose problems for the Crown case. The applicant would no doubt direct the jury's attention to those problems. But the decisive consideration is that the applicant would not be asked at any new trial "to meet a significantly different case to that the jury were asked to consider": Parker at 520 (Dawson, Toohey and McHugh JJ). This is not a true instance of the Crown attempting to supplement or reconstruct its case. Finally, it is relevant that Professor Penney's evidence was not practicably available to the Crown or to the applicant at the time of the trial.

656 There remain other considerations that are relevant to the question whether a new trial should be ordered. The charge of murder is the most serious crime known to the law. The Court must give appropriate weight to the public interest in the due prosecution and conviction of persons accused of murder. As Spigelman CJ said in Justins v R [2010] NSWCCA 242; (2010) 204 A Crim R 315 at [120], "the criminal involvement of one person in the death of another raises a serious issue of moral culpability that ought be determined". The involvement of one person in the death of multiple others raises an even more serious issue of moral culpability. A further trial would not be unduly lengthy or expensive, given the seriousness of the charges. Nor is the time already served by the applicant a bar to a further prosecution, for the applicant has served only a relatively small part of his sentences of life imprisonment on each of the counts of murder: Taufahema at [55]. The gravity of the charges and the relatively short period of time served are powerful reasons for the Court to leave the decision to retry the applicant to the prosecutorial authorities, which are better placed to determine whether a further trial would be in the public interest: Dyers v The Queen [2002] HCA 45; (2002) 210 CLR 285 at [85] (Kirby J); Fleming v The Queen [1998] HCA 68; (1998) 197 CLR 250 at [47]; Longman v The Queen [1989] HCA 60; (1989) 168 CLR 79 at 109 (McHugh J).

657 There are also considerations that tell against a new trial. The strength of the Crown case could be diminished by the version of events it may have to present at a new trial. If Professor Penney's evidence is accepted, that version of events must assume that Christopher was upstairs when or after the fire broke out. So framed, it would be consistent with the exculpatory version of events put forward by the applicant. Further considerations that militate against a new trial include the applicant's release from custody by order of this Court, as well as the prejudice that would be occasioned to the applicant by the lapse of time between the alleged offences and any new trial. It has been some 19 years since the offences allegedly took place. Some evidence that may assist in the applicant's defence is no longer available. These include Christopher's fingernail scrapings and shave coat, the murder weapon, samples from the bedding, the jerry can, the petrol sample, the garden hose, clothing recovered from Mr and Mrs Gilham, carpet samples, the intercom on which the bloody fingerprint is said to have appeared, Christopher's computer, and a set of keys. However, none of these items were available at either of the trials. The further lapse of time has not occasioned any additional prejudice by reason of the unavailability of these objects. I accept, however, that the recollections of the witnesses may have been further diminished by the passage of time, while those of others may have become entrenched.

658 It is also relevant to the exercise of the discretion that a new trial would be the third to which the applicant is subjected. In R v Honeysett (1987) 10 NSWLR 638; (1987) 34 A Crim R 277, Hunt J, with whom McInerney and Allen JJ agreed, said at 646 that the "decision whether a third trial should proceed even if ordered can often safely be left to the good sense of the Attorney-General and the Director of Public Prosecutions, but the fact that it would be a third trial will in some circumstances be sufficient to refuse an order for a third trial". The Court in Honeysett declined to order a new trial because it would be the appellant's third trial and because the appellant had already served the whole of the non-probation period that was imposed upon conviction. It was those two factors in combination that led the Court to order an acquittal: Honeysett at 647 (Hunt J). In circumstances more analogous to this case, this Court has declined to order a third trial where an appellant was twice tried for murder and had already been in custody for three years: Ward v R (1981) 3 A Crim R 171 at 179 (Street CJ), 195 (Lee J). Lee J said at 195: "A third trial in those circumstances should only be ordered if the court is convinced that no other course consistent with the interests of justice - meaning thereby the interests of the community and the interests of the accused - is open". The Court held that in view of the appellant's defence of diminished responsibility, the interests of justice were better served by substituting a verdict of manslaughter for the verdict of murder: Ward at 179 (Street CJ), 196 (Lee J). These cases, and others, suggest that the fact that a retrial would be a third trial is not of itself a bar to a further prosecution, but it may in conjunction with other considerations justify an acquittal: see also R v Murre [2001] NSWCCA 286 at [30]-[32] (Adams J, Giles JA and Hulme J agreeing); R v O'Donohue [2001] NSWCCA 458 at [36] (Bell J, Heydon JA and Dowd J agreeing).

659 There remains the issue of whether a third trial would be fair in the light of the considerable media attention and notoriety generated by the allegations against the applicant. This consideration, though not irrelevant, is not compelling enough on its own to warrant an acquittal. Brennan J noted in R v Glennon [1992] HCA 16; (1992) 173 CLR 592 at 613 that "some degree of risk, albeit not a substantial risk, to the integrity of the administration of criminal justice is accepted as the price which has to be paid to allow a degree of freedom of public expression when it is exercised in relation to a crime that is a topic of public interest". The law presumes that jurors will adjudicate fairly and impartially: Glennon at 614-615.

660 Weighing all of these considerations, I am of the view that the miscarriages of justice identified in the Court's judgment can more adequately be dealt with by an order for a new trial than by an order for an acquittal. The seriousness of the charges, the life sentences imposed, and the proportionally short period of time served by the applicant point to this being an appropriate case for the exercise of prosecutorial discretion. I am mindful that in considering whether or not to file a fresh indictment, the Director will give due weight to the various considerations discussed in the Court's judgment. No doubt he would carefully consider whether it is in the public interest to proceed with a third trial in circumstances where the applicant has already served time in custody and the Crown case may not be as strong as was previously thought.

Orders

661 The orders I would make are:

1. Appeal upheld;

2. Conviction for the murder of Helen Gilham quashed;

3. Conviction for the murder of Stephen Gilham quashed;

4. A new trial be ordered.

662 FULLERTON J: Whether there should be an order for a retrial or an acquittal in this case is to be decided by reference to the various factors to which McClellan CJ at CL has referred in his Honour's review of the authorities. As Dawson J said in King v The Queen (1986) 161 CLR 423 at 433, s 8(1) of the Criminal Appeal Act 1912 confers a broad discretion which is exercised in accordance with settled principles with the overriding consideration being whether the interests of justice require an order for a retrial.

663 In R v Taufahema [2007] HCA 11; (2007) 228 CLR 232 at [54] a majority of the Court comprising Gummow, Hayne, Heydon and Crennan JJ held that the question whether there should be an order for a retrial in that case was to be approached in light of the way the prosecution wished to conduct its case. At [60] their Honours recognised that while the Crown should not be given the opportunity to make a new case the authorities offered little explicit guidance as to what is meant conceptually by "a new case which was not made at the first trial". After a review of four authorities (none of which were analogous to the present case) they concluded at [67] that "the difference between the case relied on in a first trial and the case to be relied on in a second trial must be substantial if the difference is to stand as a bar to an order for a second trial".

664 It is not necessary to restate the competing arguments of counsel on the question whether a third trial will afford the Crown the opportunity of making a new case. I agree with McClellan CJ at CL that because the basic factual premises which underpin the Crown case remain the same, the applicant will not be asked at any retrial to meet a significantly or radically different case to that which the jury were invited to consider, despite the fact that the timing of critical events will need to be reformulated. The allegation that the applicant killed his parents in the family home within a relatively short time in the early morning of 28 August 1993 before attempting to destroy the evidence by setting fire to the house remains intact.

665 However because the new carbon monoxide evidence necessitates an adjustment to some of the factual allegations upon which the Crown case depended at trial, I am of the view that the strength of the Crown's circumstantial evidence case is weakened. In Anderson, the fact that the Crown case was said not to be strong weighed against the public interest considerations in having three counts of murder determined by a jury according to the applicable constitutional arrangement.

666 In this case the Crown will be forced to concede not only that Christopher was upstairs when or after the fire was lit and that he remained there for two to four minutes, but to abandon the argument that he was unlikely to be the killer because he was unable to negotiate his way in the dark and without his spectacles. While the carbon monoxide evidence does not have the same impact when considering the timing of the death of Mr Gilham and Mrs Gilham, it being possible that they inhaled the carbon monoxide after they had been stabbed by the applicant but before they finally died, given Dr Lawrence's evidence that they would have died very quickly, the Crown will be forced to abandon the case that the applicant killed his parents and his brother and then set about siphoning petrol from the car only settling on mineral turpentine because the siphoning was unsuccessful.

667 If the Crown were to continue to rely at any retrial upon the evidence of the cut hose and the smell of petrol on the applicant's breath and what were said to be his lies about the boat trip the following day, it seems to me that the case would need to be reformulated to contend that the attempt at siphoning the petrol was done before the killings and that the mineral turpentine was settled upon as an alternative when the siphoning was unsuccessful.

668 When invited on the appeal to identify the case it would advance were a retrial ordered, and to suggest how the carbon monoxide evidence might otherwise fit with the timing of other events upon which the Crown case was mounted at the trial, the Court was not provided with a detailed expose of the case that would be put beyond recognising the need for an adjustment to the Crown case theory. That is perhaps understandable given the range of issues that were dealt with on the appeal and, in particular, the Crown's decision during the hearing to make no challenge to Professor Penney's evidence. Even so, the Crown's failure to offer a detailed particularisation of the case it would open to the jury, in particular, a case accommodating Christopher being upstairs (unusually attired and without his spectacles) for two to four minutes while the fire was alight, and how it was that his parents also inhaled some smoke before they died, it is difficult to conclude other than that the new evidence weakens the Crown case. It must be accepted that Christopher's presence upstairs inhaling carbon monoxide is consistent with the exculpatory version of events the applicant gave to police at the time of the killings and to which he adhered in his evidence. It is also more closely allied to the objective evidence as to when the fire was likely to be lit relative to when he raised the alarm.

669 Not only has the carbon monoxide evidence undermined the Crown case theory in a material way but evidence of a pattern in the stab wounds suffered by each of the deceased was wrongly admitted as expert evidence and, for that reason, is not available to be used by the Crown at any retrial, further weakening the Crown case.

670 A third trial will unquestionably be an ordeal for the applicant and one that he would be subjected to through no fault of his own. Even were the new evidence ground as it concerns the issue of carbon monoxide not successful, the appeal would have been upheld on other grounds, including one at least that involved a finding of prosecutorial misconduct in the marshalling of the evidence to support the proposition that there was a pattern in the similarity in the knife wounds - a circumstance used by the Crown as the central motif in its case before the jury. In Anderson the most compelling factor which resulted in verdicts of acquittal being entered was that the trial miscarried due to the failure of the prosecuting authorities to satisfy itself of the integrity of the Crown case theory which was compounded by an unfairness in the way the Crown argued for a conviction.

671 We have been critical of the Crown Prosecutor in the present case in a number of respects in her conduct of the Crown case, and for her part in the decision not to call Professor Cordner which was productive of a miscarriage of justice. While not determinative of whether a retrial should be ordered, in my view her conduct is also a factor weighted against an order for a retrial. While I do not see that there is any unfairness in permitting the Crown to attempt to reformulate its case to account for the impact of Professor Penney's evidence, since both the applicant and the Crown proceeded at trial on the mistaken assumption that Dr Lawrence's evidence was correct, the same cannot be said of the Crown having a further opportunity to recast their case with a different argument, or potentially different argument, as to the admissibility of the evidence of similarity.

672 A further matter weighing against the order for a third trial is that before the ex officio indictment was presented in 2006 the applicant had been informed three times by the prosecuting authorities that it had determined not to prosecute him for the murder of his parents, first in September 1999 after the first inquest, then in July 2000 after the matter was referred to the Director of Public Prosecutions by the Coroner following the second inquest and then again the following year after the Director assumed conduct of the private prosecution. It appears likely that the evidence of similarity in the knife wounds played a very significant role in the ultimate decision to prosecute. I regard it as onerous to require the applicant to wait in anticipation of a further determination by the same prosecuting authority as to whether he will be prosecuted a third time.

673 The delay before any third trial could be convened in this Court, coupled with the fact that it has been almost 19 years since the offences allegedly took place and almost four years since the second trial, operates to the prejudice of the applicant. It is notorious that memories fade with the passage of time while the recollections of others may become entrenched. Other evidence has been lost or destroyed. While it is true that was the case at the earlier trials, it does not for that reason cease to be relevant as a discretionary consideration when considering whether an order for a retrial should be made in the interests of justice at this time.

674 Finally, the combined length of the first and second trials was 15 weeks with the second trial being shorter because the pre-trial rulings at the first trial were adopted for the purposes of the second trial. A third trial would not be expected to be of any greater length although with new evidence and the potential for different evidentiary rulings to be sought it is unlikely to be shorter. In addition, the expense of a third trial, with some expectation of the need to retain experts in an expanded range of disciplines in light of the new evidence led on the appeal, will not be inconsiderable.

675 I accept unreservedly that in the discharge of his statutory functions the Director of Public Prosecutions would carefully consider the available evidence, and the forensic consequences of the need to make appropriate adjustments to the Crown case before determining that the public interest dictated that the applicant be tried a third time. However, the question posed by the operation of s 8(1) the Criminal Appeal Act is different. After giving full weight to the fact the applicant is alleged to have murdered his parents and that after verdict he was sentenced to life imprisonment, and that the decision not to order a retrial will usurp the role and function of the Director of Public Prosecutions as the entity entrusted with the discretion to initiate and conduct of criminal prosecutions, for the reasons set out above, I am not persuaded that the interests of justice require that a new trial be ordered.

676 Accordingly, while I agree with Orders 1, 2 and 3 as proposed by McClellan CJ at CL I would direct that verdicts of acquittal be entered.

677 GARLING J: I agree with the first three orders proposed by McClellan CJ at CL. I respectfully differ from the terms of his proposed fourth order. Like Fullerton J, I would order that acquittals be entered with respect to both charges.

678 The final relief, upon which all of the members of the Court are not agreed, involves the exercise of a discretion provided by the terms of the sections of the Criminal Appeal Act 1912 which are set out in the judgment of Fullerton J. As the exercise of the discretion is one upon which the Court is not unanimous, it is appropriate for me to set out my reasons for favouring the result that the applicant should be acquitted of the murders of both of his parents.

679 The applicable authorities are adequately set out in the judgment of McClellan CJ at CL and so there is no need for me to repeat them.

680 In addition, I note that the discretion granted by the Criminal Appeal Act 1912 is not restricted by the terms of the section. The section does not impose any condition upon the exercise of the discretion. For example, it does not require that special or exceptional circumstances must be found to exist before the discretion can be exercised.

681 On the contrary, the width of the language of the statute will inevitably give rise to reasonable differences of opinion, which result from the call to an individual value judgment required by the unconfined nature of the discretion.

682 The authorities to which McClellan CJ at CL refers each record matters which may be relevant in the exercise of the discretion. It is appropriate to have regard to these authorities, in the knowledge that the matters to which they refer may each be relevant, but they do not constitute an exclusive list of the matters which may be relevant for consideration in any particular case. That is necessarily so because the ultimate question is whether the interests of justice in the particular case require the entry of a verdict of acquittal or else the ordering of a new trial.

683 My consideration commences with an acknowledgement that the murder charges which the applicant faces are of the most serious kind and involve two deaths. There is accordingly a real public interest in the due prosecution of the individual accused of those crimes. I also accept that it is desirable that a jury determines the guilt or innocence of the applicant, and that this Court does not usurp the ordinary function of the executive government. These are all matters which stand in favour of an exercise of the discretion to order a new trial.

684 But these matters including the weighty issue of the public interest are not matters which overrule all other considerations. Other considerations may decisively tell against ordering a new trial because if a new trial would be an instrument of injustice, then there is no public interest which mandates a trial in those circumstances.

685 I now turn to a consideration of these other matters.

686 An assessment must be made of the strength of the Crown case to be presented at any retrial. Such an assessment is made on the basis of all of the evidence which is now available, and not just the evidence which was before either of the two previous juries. In making that assessment, I take into account the way in which the Crown approached the evidence which was led before this Court on the appeal.

687 Of particular importance is the approach which the Crown took to the new evidence of Professor Penney about the significance of the carbon monoxide readings. Initially, it challenged that evidence and indicated that it proposed to adduce expert evidence to contradict it, but during the hearing of the appeal, it changed its approach and indicated, specifically, that it accepted Professor Penney's qualifications and did not wish to cross-examine him or lead any evidence in reply to the opinions which he expressed in his reports. In cross-examination of Dr Green, who expressed similar conclusions to Professor Penney, the Crown did not challenge his expert opinion on the carbon monoxide levels.

688 As well, the Crown led evidence from Dr Lawrence, the effect of which was that he accepted Professor Penney's expertise in the area of carbon monoxide readings in preference to his own limited learning on the subject. He then said that based on Professor Penney's opinion and report, he would need to change the evidence which he gave at trial. The details of this further evidence are to be found in the principal judgment.

689 In any new trial, having regard to this further evidence which, in my assessment, the Crown would be obliged to call, the essential elements of the Crown case as presented to the two previous juries could not be accepted by a jury. In simplified summary those features were: the deaths within a short time of the three members of the Gilham family at or about 4am, followed by a period of 15 minutes or more during which the applicant took elaborate steps to destroy the evidence of the crimes and to set up evidence implicating his brother, then to set alight the premises and proceed to wash himself and go next door seeking help.

690 As well, the additional evidence led on the appeal provides significant independent corroboration of the applicant's case. This evidence includes the opinion of Professor Penney and the further evidence of Dr Lawrence, which together provide objective material supporting the initial account given by the applicant to the investigating police, namely that his brother was upstairs in the house, and alive when the fire was being lit. There is other new evidence which suggests, unequivocally, that Mrs Gilham was sufficiently concerned about her relationship with Christopher to have told her workmates that it was necessary for her to leave work earlier than usual to return home and address the issue. This evidence provides corroboration of the applicant's expressed concerns about the behaviour of his brother to others in the period leading up to the murder of his parents. It was the Crown case that these concerns were fabricated by the applicant as part of his plan to divert attention from his involvement with the deaths. Such an allegation would now be difficult, if not impossible, to maintain.

691 As has been demonstrated in the principal judgment, the Crown case was, and remains, a circumstantial one. The Crown must exclude all reasonable hypotheses consistent with the applicant's innocence. The Crown case at the two earlier trials depended acutely on the time at which events were said to have occurred. As the Crown Prosecutor put to the jury in her closing submissions on a number of occasions: " ... time does not tell lies ...". As well, the submission on the timing of the events was relied upon to advance the contention that the account given by the applicant was implausible, and hence, unbelievable.

692 Any new Crown case cannot rely upon that timing construct. A new case theory will need to be advanced. The Crown, although invited so to do, did not really formulate such a theory in the course of the appeal. It is very difficult to see that a satisfactory theory can be now developed.

693 I have concluded that, having regard to the entirety of the new evidence, if the Crown was still able to mount a case against the applicant, it would be far less convincing than that mounted at the two earlier trials. Any new Crown case would be significantly weakened when compared with that presented at the earlier trials. I would not describe any such case as a strong one.

694 Any new trial would be held nearly 20 years after the murders occurred. This delay has not been caused by the applicant. It is a matter of common experience and commonsense that the lapse of such a long period will adversely affect the applicant's capacity to " ... wage a forensic contest in defence of [earlier] conduct ...": Gill v Walton (1991) 25 NSWLR 190 at 198 (Gleeson CJ).

695 It is inevitable that the passage of almost 20 years will affect the quality of the evidence at any new trial. Unless a witness has compiled contemporaneous notes of what they saw or heard, or else given a reasonably contemporaneous statement, then it is unlikely that any remaining recollection will be truly reliable. Adverse effects upon the accuracy of evidence and the consequent deterioration in the quality of justice are well recognised. Additionally, a number of the exhibits are no longer available. They have been disposed of. This includes the murder weapon, the items said to have been used by the applicant to siphon petrol which was said to be intended for use as an accelerant, clothing from all three of the deceased, carpet and bedding samples and other items. Importantly, having regard to the submissions made on appeal with regard to Ground 15, the intercom unit is no longer available for examination.

696 In my assessment, the quality of justice which will result from any new trial will be severely diminished. All of these matters tell against ordering a third trial.

697 There are issues which are personal to the applicant which must be weighed in the balance. He has already spent time in custody. However, having regard to the sentences which he received, I do not regard this as a factor of any real weight.

698 However, the conduct of the prosecuting authorities (which term I use to include the police, and the Director of Public Prosecutions) towards the applicant, with the inevitable personal anxieties which it has caused and the raising of false hopes which it has created, and which has continued over a lengthy period of time, is of real significance. It tells strongly against ordering any further trial. It is appropriate to enumerate these matters:

(20) First, in 1993, the prosecuting authorities charged him only with the murder of his brother and not with the murders of his parents. He was entitled to assume (and conduct his affairs accordingly), that, after the results of their investigations were concluded there had been a decision that he would not be charged with the murders of his parents, because the available evidence did not warrant such charges. There is no suggestion that by the standards of the time, the police investigation was anything other than thorough.

(21) Secondly, in 1995, in accepting the applicant's plea to manslaughter, the Crown did so because it could not disprove the applicant's account that his brother had murdered his parents. Put differently, the Crown did not have sufficient evidence to charge the applicant with the murders of his parents, and accordingly he was not charged. Any assumption which the applicant was entitled to make in 1993 was confirmed.

(22) Thirdly, in June 1995, a Coroner found on the available evidence, gathered by the police investigation, that Christopher Gilham had murdered his parents. The prosecution authorities did not seek to challenge this outcome, but accepted it.

(23) Fourthly, five years later in July 2000, after a second coronial inquest had been terminated because the Coroner formed the view that there was evidence capable of satisfying a jury that the applicant had committed the murders, the Director of Public Prosecutions informed the applicant that there would be no prosecution because there was no reasonable prospect of any conviction.

(24) Fifthly, about a year later in May 2001, the Director, having taken over a private prosecution against the applicant launched by his uncle, decided to terminate the prosecution, and offer no evidence on the matter when next before the Local Court. The applicant was discharged.

(25) Sixthly, it was over five years later that, in 2006, the Director served an ex-officio indictment upon the applicant charging him with the murders of his parents.

(26) Finally, now, a further six years has expired during which the applicant has faced two trials, one of which ended without a verdict because the jury could not agree, and the circumstances are that the result of the second trial must be set aside because a miscarriage of justice has occurred. In significant part, this miscarriage has been brought about by the manner in which the trial was conducted by the Crown. In addition, in the course of the judicial process, the applicant has brought two appeals in this Court, in one of which he has been successful, and one unsuccessful application in the High Court of Australia.

699 The trials which the applicant faced have not been short trials. Nor would they be expected to be having regard to the complex issues of fact and law involved in them. The first trial, including pre-trial argument, the trial itself and the period of jury deliberation occupied over ten weeks; the second trial and jury deliberation occupied over seven weeks. Any further trial is likely to be of a similar length, namely between seven and ten weeks. This raises afresh questions of significant additional public and private expense, the inconvenience to the public involved both as witnesses and jurors, and the impact which any further trial may have on the limited public resources which are available in the criminal justice system.

700 This Court is asked to exercise its discretion to order that the applicant be put on trial for a third time, on the basis of a Crown case theory which, although not yet even formulated, the Crown accepts must necessarily differ from the first two trials. Any new trial is on a Crown case which, in my assessment, is significantly weaker than the Crown case at the two previous trials has been. The fact that any new trial would be the third time to which the applicant has been subjected to the trial process is a matter which it is important to weigh in the balance.

701 I am satisfied that this is a wholly exceptional case. By reason of the combination of the factors to which I have referred above, but in particular, that in my opinion, having regard to the fact that any new Crown case would not be a strong one, and the history of the applicant's treatment at the hands of the prosecuting authorities throughout his involvement in the judicial system, intervention by this Court is justified because otherwise the continued operation of the criminal justice system upon the applicant would constitute a source of unacceptable oppression or unfairness.

702 Accordingly, I agree with the orders proposed by Fullerton J, including that verdicts of acquittal ought be entered in favour of the applicant on the two charges of murder.

**********

Decision last updated: 18 October 2013