HKSAR v. Khaw Kim Sun

Read the full judgment text of CACC 309/2018 on BabelCite. This Court of Appeal judgment was delivered on 7 June 2022 before Poon CJHC, Macrae VP and Zervos JA.

Criminal law – murder – two counts – carbon monoxide poisoning – circumstantial evidence – burden and standard of proof – summing-up – expert evidence – directing jury – leave to appeal. The applicant, an Associate Professor in Anaesthesia and Intensive Care at the Chinese University of Hong Kong, was convicted of two counts of murder after his wife and 16-year-old daughter were found dead from carbon monoxide poisoning in a yellow Mini Cooper on 22 May 2015. The prosecution alleged the applicant filled a yoga ball with carbon monoxide under the guise of sham medical research, brought it home, and placed it in the car to kill his wife. The applicant denied placing the yoga ball in the car and relied on his out-of-court statements and expert evidence, contending that his research was genuine, that he intended to use the CO to kill rats at home, and that the deaths were possibly an unexplained accident or caused by someone else (possibly his daughter Lily). The court considered 20 grounds of appeal under four headings. Held, dismissing the application for leave to appeal: (1) the trial judge's directions on burden and standard of proof were correct and followed the Specimen Directions in Jury Trials; (2) the summing-up was comprehensive and balanced, fairly presenting the defence case including the timeline defence, the genuineness of the research, the missing stopper issue, and the possibility of an unexplained accident; (3) the trial judge did not improperly remove the issue of unexplained accident from the jury, but properly directed the jury on the main issue of who placed the yoga ball in the car and why; (4) the trial judge's directions on expert evidence were thorough and balanced, fairly summarising the conflicting evidence on both the research protocol and the diffusion rate of carbon monoxide. The court found the applicant had a thoroughly fair trial, that there was cogent and compelling evidence to justify his convictions, and that none of the 20 grounds disclosed any arguable error. Application for leave to appeal refused; appeal dismissed.

Legal issues: Application for leave to appeal against conviction for two counts of murder

Outcome: Application for leave to appeal against conviction refused; appeal dismissed

Cited by 5 cases · Cites 6 cases

Case No.CACC 309/2018[2022] HKCA 802
Court
Court of Appeal
Date07 Jun 2022
JudgePoon CJHC, Macrae VP and Zervos JA
Case Document
100%Judiciary

CACC 309 /2018

[2022] HKCA 802

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF APPEAL

CRIMINAL APPEAL NO 309 OF 2018

(ON APPEAL FROM HCCC NO 374 OF 2017)

________________________

BETWEEN    
  HKSAR Respondent
  and
  KHAW KIM SUN (許金山) Applicant

________________________

Before: Hon Poon CJHC, Macrae VP and Zervos JA in Court
Dates of Hearing: 16 and 17 December 2020
Date of Further Written Submissions: 10 February 2021
Date of Judgment: 7 June 2022

____________________

J U D G M E N T

____________________

The Court:

Contents

Introduction

The prosecution case

The defence case

The issues at trial

The relationship between the applicant and his wife

Lily

The character of the applicant

The expert evidence

The applicant’s research on rabbits

Events from 20 to 22 May 2015

Tests done on the Mini

The diffusion rate of carbon monoxide

An overview of the grounds of appeal

(i)    Proof, defence evidence and the right of silence

(ii)   Failure to summarise the defence case

(iii)  Removal of issues from jury

(iv)  Expert evidence

The form and content of the grounds of appeal

The “keys” argument

The reference to a “crime”

The financial arrangements

The burden and standard of proof

The applicant’s right of silence

The evidential directions

The directions on the expert evidence

The applicant’s research

The reinjection of the blood

The timeline defence

The missing stopper

Conclusion

Introduction

1.On 19 September 2018, the applicant was unanimously convicted after trial before Barnes J (“the Judge”) and a jury of nine of two counts of murder, contrary to common law.  He was sentenced to life imprisonment on each count and now seeks leave to appeal against his convictions. Having reserved our decision at the hearing of the appeal, this is the unanimous judgment of the Court.

2.On 22 May 2015, Ms Tong Yuk Ling (PW1) (“Ms Tong”), a nurse, was on her way home when she noticed that a yellow Mini Cooper (“the Mini”) had stopped at a bus stop outside Sai O Village in the New Territories. There appeared to be two people asleep inside the car.

3.On arriving home, Ms Tong decided to go out for a run.  Whilst on her run, she again saw the Mini and its occupants, but on this occasion she did not stop.  On her way home, however, she passed the vehicle again and this time she noticed that the windscreen wipers were operating, even though it was not raining.  Ms Tong knocked on the window of the vehicle. There was no response and, accordingly, she made a report to the police.

4.A police sergeant, who happened to be in the area, soon arrived at the scene and smashed one of the windows of the Mini; thereafter, an ambulance arrived.  The two occupants, each apparently unconscious, were removed from the vehicle and taken to hospital, where they were both certified dead.  The cause of death of both persons was later established to be carbon monoxide poisoning.

5.The two occupants of the car were the applicant’s wife, Madam Wong Siew Fing (“Madam Wong”), and her 16-year-old daughter, Khaw Li Ling (who was known, and referred to at trial, simply as “Lily”).  The police initially suspected that the carbon monoxide must have emanated from the Mini, yet after investigation they could find no mechanical defect with the vehicle to account for the tragedy.  In the boot of the Mini was a partially deflated yoga ball.  It was not apparent at that stage that the yoga ball had been the source of the carbon monoxide.

6.The applicant, who was an Associate Professor in the Department of Anaesthesia and Intensive Care at the Chinese University of Hong Kong, had filled the yoga ball with carbon monoxide and brought it to his family home on the night before the incident.

7.About a year later, on 12 May 2016, the applicant was arrested inside his office at Prince of Wales Hospital by Lam Kam Cheung, Detective Sergeant 16404 (PW15) (“DSgt 16404”) in respect of two offences of murder.  Under caution, the applicant replied in Cantonese:

“I did not know why this incident happened. I intended to use the yoga ball for killing mice. There were many mice at my residence. I took (it) home (and) placed (it) at where Lily did exercise. (After) she had just finished doing exercise, I told her that (it) contained poison (and) not to touch (it). (It) would be used for killing mice at the weekend. This matter was only known to Lily. I did not know why the ball ended up in the car on 22 May either. Perhaps Lily wanted to (commit) suicide.”

8.On the following day, 13 May 2016, the applicant gave the first of three video recorded interviews under caution, all of which were conducted by two officers, one of whom was DSgt 16404.  The applicant was throughout represented by both counsel and an instructing solicitor.  No issue as to the voluntariness or admissibility of his out-of-court statements was taken by the applicant at trial.

The prosecution case

9.The case for the prosecution was that the applicant had murdered both Madam Wong and Lily by deliberately placing a yoga ball filled with carbon monoxide gas inside the Mini and removing the stopper or plug, thereby causing the gas to escape and fatally poisoning them both.  The prosecution suggested the applicant had intended to kill his wife but, tragically, he had also killed his daughter in the process.

10.A subsequent examination of Madam Wong and Lily revealed that the carboxyhaemoglobin detected in their blood was at a level of 50% and 41% respectively with the time of their deaths estimated at between 12:45 and 2:45 pm on 22 May 2015.  Carboxyhaemoglobin is a consequence of the ingestion of carbon monoxide into the body and the levels of carboxyhaemoglobin suggested that both Madam Wong and Lily had ingested a substantial quantity.

11.The prosecution alleged that the applicant had ostensibly embarked on some purported medical “research”, which was in fact a sham, in order to obtain a supply of carbon monoxide for the purpose of killing his wife[1].  Their alternative position was that if the jury were to find that this research may initially have been genuinely undertaken and the carbon monoxide thereby obtained for a legitimate purpose, he had subsequently decided to use the gas to kill his wife.  It was argued by the prosecution that the applicant’s suggestion that carbon monoxide could be used as an effective means for killing mice or rats was a “silly idea”, which defied critical examination[2].

12.On 20 May 2015, the applicant was observed by his research fellow, Dr Dick Chow Ho Kiu (PW9) (“Dr Dick Chow”), filling two empty yoga balls, one blue and one grey, with carbon monoxide from the cylinder in the laboratory.  The applicant explained to Dr Dick Chow that he had done this so that a friend could check the purity of the carbon monoxide.  The applicant subsequently admitted in a video recorded interview that this explanation was untrue.  He said that he intended to use the carbon monoxide to kill mice or rats at his home.

13.The applicant accepted that he put the two yoga balls in his Toyota Alphard, which was parked at the Prince of Wales Hospital. He left the car there overnight and went to his girlfriend’s home nearby, where he stayed for the night.  He had also installed two carbon monoxide detectors in the vehicle.  He did not get back to his vehicle until after work on the following day, which was a Thursday.  He noted the detectors were emitting an alarm and found that the blue yoga ball was leaking carbon monoxide.  He released the carbon monoxide from the blue yoga ball and then went back to his family home.  Having arrived home, he left to play in a tennis match and took the Mini instead.  On subsequently returning home, and after the children had gone to bed, he retrieved the grey yoga ball from his Toyota Alphard.  He placed it in the exercise room, intending in due course to connect it to a drainpipe in order to kill rodents.  Lily came into the exercise room and saw the yoga ball.  He told her it was filled with carbon monoxide and she then asked him whether it was to kill rats “or something like that”.

14.It was usual for Madam Wong to pick up the two younger children from school in the afternoon, which was part of her normal routine.  On the Friday, Lily did not have to go to school and, accordingly, stayed at home.  The applicant accepted that he had told Lily to finish her outstanding homework and not to go out.  He said he also asked his wife to ensure that Lily stay at home to finish her homework.  However, Lily joined Madam Wong and they went out together in the Mini. 

15.The family’s domestic helper, Ms Siti Maesaroh (PW6) (“Siti”), testified that she saw both Madam Wong and Lily leaving the house and getting into the car at “around 2:00 pm”[3].  She had watched them from the upper floor of the house as they got into the car[4].  She said it was raining at the time[5], but not very hard[6]. She did not see either of them carry anything to the car.  The car also had two transponder keys: the last updated record for the car key of Madam Wong was at 2:18 pm on 22 May 2015[7]. These facts, when taken together, said the prosecution, indicated that it must have been very shortly after leaving home that Madam Wong stopped the Mini at the bus stop.  As will later become apparent, this time frame was variously addressed at trial and this appeal, and has been termed by the applicant in argument before us as the ‘timeline defence’.

16.Ms Tong, who first happened upon the Mini stopped at the bus stop, said that it was 3:35 pm when she first noticed the car on her way home, and that the windscreen wipers were on but that there was no rain.  Having arrived home, she changed and went for a run.  At about 3:55 pm, she saw that the Mini was still in the same position.  On her way home, she again passed the Mini at about 4:15 pm.  It was then that she had knocked on the window and, after getting no response, called for help[8].

17.It was an admitted fact that the location where Madam Wong and Lily were found was about 1.6 kilometres from the house[9] where they both lived with the applicant and two other children of the marriage.  The eldest child, Khaw May Ling (PW11) (referred to as “May May” at trial), was away in Malaysia, where she was studying medicine.  It was common ground at trial that it would have taken about 15 minutes to drive from the family home to the place where the Mini was found. 

18.The prosecution did not have any direct evidence that it was the applicant who had placed the yoga ball inside the boot of the Mini.  There was no eye-witness evidence of anyone placing the yoga ball inside the vehicle.  Instead, the prosecution case proceeded on the basis that, by a process of elimination, the applicant was the only one who could have placed the yoga ball inside the Mini and removed its stopper or plug.

19.The prosecution further commented that it was highly convenient that the only other person, whom the applicant claimed knew about the yoga ball being filled with carbon monoxide, was Lily, who could not, of course, confirm this assertion.

The defence case

20.The defence case was the applicant did not put the yoga ball into the Mini.  There was no motive for him to kill his wife; moreover, the research he was conducting was not a sham.  While the applicant did not give evidence himself, he relied mainly upon what he had said in his three video recorded interviews, and on the evidence of various expert witnesses.

21.The late Mr Gerard McCoy SC, with him Mr Kim McCoy, on behalf of the applicant at trial, argued that there was no clear motive for him to murder his wife or daughter.  Money did not provide a sufficient justification, since the bank accounts of his wife and himself were joint accounts, from which he could have removed any money he wanted at any time[10]. Nor was his relationship with another woman a sufficient motive, since his wife had come to terms with his affair and reconciled herself to his infidelity[11]. He added that the applicant would not have put in jeopardy his successful professional career, or the care of his children and home, which were attended to by his wife[12].

22.Mr McCoy listed a number of factors to show the applicant was not a person likely to have murdered his wife and daughter.  He submitted that the number of people (some twelve in all) who would have known about the applicant purchasing carbon monoxide, and the clear paper trail linking the applicant to the gas, both militated against him having planned the murder of his wife[13]. He argued that the applicant was a person of good character, with a high professional and academic standing as a doctor, whose work involved saving lives, not taking them.  It was also pointed out that if the applicant had really intended to kill his wife in this way, he ran the obvious risk that he might kill the two children she was due to collect from school as well[14].

23.Mr McCoy commented that it was some six months after the incident on 22 May 2015, that the police turned their attention to the yoga ball as the source of the carbon monoxide.  During that period, the Mini had been stored by the police and was variously examined in order to determine the source of the carbon monoxide that killed Madam Wong and Lily. 

24.When the police eventually turned their attention to the yoga ball, it was located in a collapsed state in the boot of the Mini.  The stopper or plug of the yoga ball was not apparently found by the exhibits officer Chan Ching Lun, DPC 7021, (PW17) (“the exhibits officer”).  Since he had mistakenly numbered some exhibits, Mr McCoy queried whether the exhibits officer had also made a mistake about not finding the stopper.  He invited the jury to consider how the items, particularly the yoga ball, were only found in the boot of the Mini after a lapse of some six months, suggesting that things may have been moved around during this period.  He asked them to consider whether the yoga ball had been found in a “neatly folded state”[15]

25.Leading counsel invited the jury to consider other possibilities for the deaths on the evidence[16]:

“What could have happened? Just think about this on the prosecution’s own case because they’ve only tried to really eliminate suicide which is not my positive case. We don’t have one. We just say look at possibilities. We don’t plead a positive case. It’s anything that can exist.”

Whilst accepting that Lily had not committed suicide, Mr McCoy did advance one possible explanation, namely, such was Lily’s phobia of insects that she might have brought the yoga ball out of the house, unaware of the extreme danger of its contents, in order to exterminate insects she may have seen in the Mini, before setting off with her mother to collect her siblings from school[17].

The issues at trial

26.There were two major areas of dispute at trial, both of which heavily depended on the evidence of experts.  The first area was the applicant’s research on the effectiveness of using oxygen to rescue live rabbits suffering from carbon monoxide poisoning.  The prosecution contended that the research was a sham, in order to provide an excuse for the applicant to obtain a supply of carbon monoxide.  The prosecution therefore sought to show that the experiments conducted by the applicant were of no scientific utility or benefit: the defence argued otherwise.

27.The second, which we have briefly alluded to, involved the timeline defence.  The argument advanced was that, in order for the carbon monoxide concentration to have been lethal enough to kill, the yoga ball would have had to have been placed in the Mini not long before they left home.  If that was the case, then it would rule out the applicant as the person who put the yoga ball inside the car, since he was teaching at the Chinese University at the relevant time.

The relationship between the applicant and his wife

28.Before turning to the issues raised by this appeal, it is helpful to say something of the dynamics within the family relationship, since this also was a significant aspect of the trial.  Madam Wong was a housewife and it was clear from the evidence that her relationship with her husband had become an unhappy one.  A friend of the family testified that the two did not talk to each other, but communicated by messaging.  May May said that sometimes she would have to relay messages back and forth between her parents[18].

29.That Madam Wong knew of her husband’s affair with another woman was confirmed in evidence by Madam Foong Wai Yee (PW7) (“Madam Foong”), a friend of Madam Wong.  In late 2013, Madam Wong had been diagnosed with moderate depression and anxiety[19]. However, Madam Foong said she had got over this and had been happier during the first few months of 2015 before her death[20]. There was evidence that Madam Wong had joined an organisation to improve herself (May May described it as a “cult”), and the defence further produced Madam Wong’s diary, in which she had listed all her shortcomings[21]

30.Dr Victor Cheong Mun Khan (DW2) (“Dr Cheong”) was a family friend and doctor from Malaysia.  He considered that Madam Wong had changed in recent years, and that she had begun to spend more money[22].  

31.The applicant accepted that his relationship with his wife was not good.  This was particularly so after May May had been diagnosed with a blood disorder.  He maintained that he and his wife had discussed divorce, but each felt they could not cope with the four children individually, so they had decided to stay together as parents in order to look after them.  He acknowledged that his wife knew about his affair but did not seem to want a divorce[23].

Lily

32.A number of witnesses[24] testified that Lily was clever, energetic and full of life.  May May said Lily had been diagnosed with Attention Deficit Hyperactive Disorder (“ADHD”) and dyslexia when she was about 6 or 7 seven years of age.  There was one apparent reference in the evidence to the subject of suicide.  In an e-mail exchange between May May and the applicant, May May had told him, “I don’t want to drag Lily into this but there are things that even she wouldn’t bother bringing up with you, which is relieving, because I don’t want the same thing to happen to her”[25]. Nevertheless, the Judge remarked that there was no actual evidence that Lily was suicidal[26].

33.There was, however, evidence that Lily was terrified of rats and insects.  Ms Eunice Li Lok Ting (DW3) (“Eunice”), Lily’s school friend, described how Lily would jump up on a chair, or run out of class, if she saw an insect.  She also testified that she had seen something of Lily during Easter 2015, and she appeared happy.  Dr Cheong said he had seen Lily only five days before her death, and he also described her as happy and full of life.  There were Facebook chats produced between Lily and Ms Sarah Niu (PW18) (“Sarah”), another of Lily’s school friends, on the Friday she died, in which Lily had suggested going out with Sarah on the following Monday[27].

34.The applicant had told the police that he thought Lily wanted to commit suicide, and he did not know whether this was because she felt that he had put too much pressure on her to study, or whether she had possibly had a disagreement with her mother[28].  Against this, the prosecution submitted that if, indeed, it were true that Lily intended to commit suicide, then she must have intended to kill her mother as well[29].

The character of the applicant

35.The applicant was a highly competent professional and the sole financial support for his family of six.  He was a professor at the Chinese University and a senior medical officer at the Prince of Wales Hospital.  According to May May, he had pushed his children to better themselves.  It was her view that he did not understand ADHD, something from which both she and Lily suffered.

36.Professor Leung Tak Yeung (DW5) (“Professor Leung”) explained that whenever high risk cases were referred to the Prince of Wales Hospital, he would ask the applicant to assist him; for example, if there was a particularly complicated pregnancy to deal with.  He said the applicant was somewhat avant-garde in his approach to new methods and was well ahead of others in this respect[30]. Dr Cheong said the applicant was a clever man and devoted to his family.  He said he was a man well able to control his temper[31].

The expert evidence

37.As we have already noted, the expert evidence in this case principally focused on two issues.  The first, in relation to which three expert witnesses were called, namely Professor Kwong Yok Lam (PW21) (“Professor Kwong”), a professor in haematology at the Department of Medicine, University of Hong Kong, Professor Gavin Matthew Joynt (PW22) (“Professor Joynt”), the Chairman of the Department of Anaesthesia and Intensive Care at the Chinese University, and Professor Michael Garnet Irwin (PW23) (“Professor Irwin”), a professor in Anaesthesiology at the Faculty of Medicine, University of Hong Kong, for the prosecution[32]; and Professor Nicholas Allan Buckley (DW1) (“Professor Buckley”), a professor in Clinical Pharmacology at the Medical School, University of Sydney, Australia, and Professor Wang Chi Chiu (DW6) (“Professor Wang”), the Divisional Head of the Department of Obstetrics and Gynaecology at the Chinese University, for the defence, concerned whether or not the experiments done by the applicant on rabbits amounted to genuine research.  The second issue, in respect of which the prosecution called Dr Wong Koon Hung (PW20) (“Dr Wong”), a Forensic Government Chemist, and the defence, Professor Li Chi Kwong (DW4) (“Professor Li”), an engineer at the Hong Kong Polytechnic University, concerned the diffusion rate of carbon monoxide inside the Mini; which issue was relevant to the timeline defence.

38.The prosecution experts essentially maintained that they did not see any clinical benefits from studying the elimination half-life[33] of carbon monoxide in a rabbit during oxygen therapy.  They were also critical of the methodology used by the applicant in his experimentation, pointing out that people who suffered from carbon monoxide poisoning would have inhaled the gas and not had it injected into their bloodstream.  Accordingly, they queried if there was any point in a study in which blood was extracted, infused with carbon monoxide and then directly reinjected back into the subject[34].

39.The prosecution experts stated that when treating a patient who had inhaled carbon monoxide, the standard procedure was to use 100% oxygen, with or without pressure, in order to save the patient’s life.  There would be no question of using 60% or 80% oxygen, as the applicant had apparently used in his experiments on rabbits[35]

40.On the other hand, the defence experts stated that there was some utility in supplying a lesser percentage of oxygen because of the harm caused in giving a patient 100% pure oxygen.  The defence case was that the carbon monoxide was being used to produce a state of hypoxia, in order to test the concentration of oxygen in the blood[36].  The applicant’s curriculum vitae indicated that he was doing research into ‘supplementary oxygen’: his experiments on rabbits could be considered consistent with such research.  Professor Buckley explained that there was some clinical relevance in drawing blood, infusing it with carbon monoxide and then reinjecting it into an animal; it might provide some interesting insight as to how carboxyhaemoglobin was dealt with by the body[37].  When asked, in cross-examination, whether the injection of carbon monoxide into the blood of an animal was of any utility, Professor Buckley responded that “the use is not immediate, but there might be an insight as a result”[38].  He believed that the study could be of some academic utility[39]

41.Professor Wang was a colleague of the applicant and co-author of various academic papers with him.  He stated he had no track record in carbon monoxide research, and that he was providing expert opinion specifically on the study of protocols and related documents[40].  He testified that the applicant’s study protocol was an animal study, not a clinical study.  He said it was relatively short and simple but was still a developing one, and from a clinical point of view the protocol was very good[41]. In cross-examination, Professor Wang maintained that the method used had direct clinical use.  He was of the view that the applicant was nonetheless trying to establish an “acute carbon monoxide poisoning model”[42].  However, his evidence appeared to be rather undermined when he suggested that blood infused with carbon monoxide had been reinjected into the rabbits.  He accepted that the method of reinjecting saturated carbon monoxide blood had no direct physical relevance.  He also accepted that the normal method of acute carbon monoxide poisoning was by breathing[43]. It is to be noted that Professor Wang’s expertise was not in carbon monoxide research.  He said he thought he could determine reinjection by the oximeter readings which demonstrated that the applicant had injected saturated carbon monoxide blood back into the rabbit.[44] He later stated in cross-examination that he believed reinjection took place because of two facts, namely he had repeatedly asked the applicant and the changes shown on the oximeter readings.  This was contrary to the testimony of those present at the experiments, as well as the documentary evidence. 

The applicant’s research on rabbits

42.The applicant had proposed a research project which involved the infusion of carbon monoxide into rabbits.  As noted earlier, it was part of the prosecution case that this research was a sham in order to prevent questions being asked about the applicant’s procurement of carbon monoxide.  Further, the prosecution submitted that this research was outside the speciality of the applicant and his known field of research.

43.In about December 2014, Professor Qin Ling (PW8) (“Professor Qin”), a professor at the Chinese University, said he remembered playing tennis with the applicant, during which the applicant mentioning his desire to use animals for research.  Therefore, Professor Qin asked his assistant, Dr Dick Chow, to liaise with the applicant about the experiments[45]. Thereafter, Dr Dick Chow was the one who forwarded the relevant forms to the applicant for him to submit to the Ethics Committee, and who helped the applicant order carbon monoxide.

44.Professor Joynt was the head of the applicant’s department. He was the applicant’s supervisor and it was he to whom the applicant, in the normal course, would have had to forward his research.  He said that the first time he had seen these documents was when the police had shown them to him.  This was not challenged by the defence.  The applicant admitted in his video recorded interview that the relevant proposal forms were never actually submitted to the department[46].

45.Professor Joynt explained how he had looked at the applicant’s curriculum vitae and saw no direct link between carbon monoxide poisoning and any of the applicant’s particular interests[47]. He said it was normally his job, if a proposal came to him, to vet it.  Had he seen this proposal, he would have sent it back and there was no way he would have passed this on to the Ethics Committee, since it was too crude and rudimentary[48]. He thought the applicant’s animal research protocol was not well thought through.

46.Professor Buckley was of a similar view, saying in cross-examination, that if he had been on the Ethics Committee, he would have sent it back too, as it was still in draft form and there were question marks in respect of it[49].

47.Professor Buckley, when asked about the protocol in examination-in-chief, testified[50]:

“… The research outlined … it was focused on using various concentrations of inhaled oxygen and a rather unusual model of carbon monoxide poisoning involving removing and reinjecting blood. There was no explanation as to why this reinjection model was planned and I do not believe it is widely used. It would mean the maximum carboxyhaemoglobin concentration achieved would be around 20 per cent with the model that was outlined in the protocol. I could see some advances to an infusion model that could reliably achieve a higher specific concentration of carboxyhaemoglobin sufficient to cause tissue hypoxia, but it is not clear how that could be achieved with that protocol. So it is reasonable to question whether this particular protocol would end up being a useful model and taken any further. Other research on oxygen therapies for carbon monoxide generally uses inhalation for exposure and typically aims to achieve much higher levels, around 70 per cent carboxyhaemoglobin.”

48.In cross-examination, however, Professor Buckley elaborated as to why he would have sent the protocol back.  He pointedly expressed his view in the following exchange[51]:

“A. Yes. If this came to me, I’d be saying, ‘Take it back, try again.’

Q. Yes. And being Professor Nick Buckley, you’d probably have a multitude of annotations on it as to what you were concerned about?

A. There would be a bunch of things that would be addressed, required to be addressed, yes.

Q. Yes. Now, I suggest to you that this research protocol is not of the standard you would expect of an experienced researcher and a clinician at a respected university?

A. If this was the final version submitted to an ethics committee, I would be very unimpressed, yes.

Q. Yes, and you’d be even less impressed if they started dealing with rabbits, perhaps not completely starting the experiment, but withdrawing blood, undertaking tests, is that right?

A. Yes, I think that would be the sort of issue that would be very seriously taken, referred to whatever disciplinary committee…”

49.Ms Tam Tsui Ping (PW23) (“Ms Tam”), the applicant’s laboratory assistant, described two experiments, one on 13 May and the other on 20 May 2015, where the applicant, it would appear rather unsuccessfully, sought to infuse rabbit blood with carbon monoxide.[52]  This was also the evidence of Dr Dick Chow.  In neither experiment was it apparent that carbon monoxide blood was in fact infused back into the rabbit[53].  The prosecution said, of the experiment on 13 May, it “wasn’t much of an experiment at all”[54].

50.Whilst there was evidence of the applicant using carbon monoxide for his research in these two experiments, there was no evidence of him studying any patients who had been poisoned by carbon monoxide; it was apparent he had done research on oxygen therapy, or what is called ‘supplemental oxygen therapy’, but this was all in relation to saving pregnant women who were giving birth[55].

51.After the second experiment, the applicant was observed filling up two apparently empty yoga balls, one blue, and one grey, with carbon monoxide from the cylinder in the laboratory.  The applicant explained to Dr Dick Chow that he was filling up the yoga balls with carbon monoxide in order to take them to a friend to analyse their “purity”, but the applicant subsequently admitted in his video recorded interview that he had “made up a story” so that he would not challenge him for removing laboratory materials[56].

Events from 20 to 22 May 2015

52.The applicant admitted that he was the one who filled the two yoga balls with carbon monoxide; and that this followed the failed experiment of 20 May 2015.

53.The applicant said he had tried getting rid of the rats at home using sticky pads, but this was unsuccessful.  He had also tried using cages but, again, to no avail.  He considered using poison but, since he had pets at home, he did not think it was a good idea.  However, after having done the experiment using carbon monoxide, the idea came to him that he could use gas to kill rats[57].

54.He maintained that it was safe to use a yoga ball to contain the carbon monoxide since the material was tough.  His plan was to fill the yoga ball with carbon monoxide and then connect a tube to the ball and pump carbon monoxide from the yoga ball into the drainpipe from which the rats were emerging.

55.He said that after he had filled the two yoga balls with carbon monoxide, he left them in his car, a Toyota Alphard, which was parked at the Prince of Wales Hospital, and then stayed the night at his girlfriend’s home, which was near the hospital.  He had also installed two carbon monoxide detectors in his car.

56.The applicant did not get back to his car until after work on the Thursday (19 May).  He then noted the carbon monoxide monitor was sending out an alarm and he discovered that one of the yoga balls was leaking.  Consequently, he emptied the carbon monoxide from this particular yoga ball, leaving the other one still with its carbon monoxide contents in the car.  The prosecution made the point that his ability to do this safely indicated the applicant knew what he was doing in handling a poisonous gas[58].

57.To put these assertions in their proper context, the prosecution described the carrying of yoga balls filled with carbon monoxide as “crazy”[59] and said that a yoga ball full of carbon monoxide let loose in a court room would likely have killed everyone inside in 5 to 10 minutes[60]; whereas the defence countered that “in some ways only a person that brilliant can be that stupid to bring it home, because none of us would have done that”.  Further, someone who was that clever would not have left so many “obvious signposts that it was him”[61].

58.Having picked up his Toyota Alphard, the applicant then drove home, where he left his car and went off to play a tennis tournament driving the Mini[62]. When he returned home from the tennis tournament, he said he joined his children in the exercise area.  Having sent the two younger children to bed, he saw that Lily was still up exercising on the treadmill.  When he thought the treadmill had stopped and that Lily had gone to bed, he went out to the Toyota Alphard and brought the yoga ball filled with carbon monoxide inside the house.

59.It so happened that Lily then returned and saw him place the yoga ball in the corner of the exercise area[63].  He told Lily that the yoga ball contained poisonous gas, with which he intended to kill rats over the weekend.  She had replied, “Oh, is it very bad, very poisonous?  Would it kill the rats or something like that?”.  His thinking was even if there was leakage it would only constitute a small volume of the entire house, so it would not be dangerous[64].

60.The applicant said that, on 22 May, he got up between 9:30 and 10:00 am.  His wife was still asleep.  He made himself some coffee and wandered around the garden, before doing some work in the study.  When he later came down, he saw his wife and Lily both eating.  He then left for work at about 12:30 to 1:00 pm[65]. He said he did not notice if the yoga ball was still inside the house on 22 May after he got up, and he had no idea how the yoga ball had found its way into the Mini.  Since Siti did not see the applicant leave the house, the prosecution submitted it was a fair inference that the yoga ball had been placed in the Mini at about 12:00 pm without its plug.

61.The applicant admitted he did not tell the police investigator, who came to the house as part of the investigation on 24 May, that his wife and daughter were killed by the carbon monoxide from the yoga ball, which he himself had filled with the gas; he admitted being a coward in letting people think the carbon monoxide came from the engine of the car[66]. He also said he did not tell the police because he did not want to get people such as Dr Dick Chow into trouble.  Nor did not want his children to be without a parent should he be taken away.

62.The prosecution submitted the applicant knew full well the danger posed by the carbon monoxide in the yoga ball, and yet he introduced it into his family home without telling anyone, as part of the plan to kill his wife.  Mr Andrew Bruce SC, with him Ms Christal Chan, for the prosecution, commented rather graphically, “he said that he knew that the kids sat on the yoga balls.  Well, that’d be good wouldn’t it?  Sitting on a yoga ball full of carbon monoxide. Boom”[67]. This conduct was said to be in contrast to the applicant’s caution in ensuring that there were two gas monitors in his own car to warn if carbon monoxide was leaking from the yoga ball[68].

63.The defence answer to this point was that the applicant was the sort of person who was “nerd-like” and viewed things with “tunnel vision”; or, as the Judge characterised the argument, “he does not think the way how we think.  He just did not even consider that it could be dangerous to do such a thing.  It is a matter for you”[69].

64.Siti testified that she had never seen any rats or evidence of rat droppings[70]. On the other hand, May May testified that many rats emerged at night, but they quickly disappeared.  However, as the Judge pointed out, May May had been pursuing her studies in Malaysia since the beginning of the year, so would not have been familiar with the position prior to the incident in May 2015[71]. Photographs of the drain[72] at the family home were produced and shown to the jury.  The applicant also produced a sketch of the house, with the drains shown, and used it to describe how he was planning to use the yoga ball, connected to a tube, to pump carbon monoxide into the drains.

65.It is also relevant to note here that, according to the evidence of the exhibits officer, the plug or stopper of the yoga ball was not found when he searched the car.  His reliability, however, was challenged on this issue and it became clear that he had evidently mixed up the exhibit numbers of the two car keys[73]. On this issue the Judge directed the jury, “… if you are sure that the exhibits officer was telling you the truth that there was no plug, then whoever had put the yoga ball in there had removed the plug totally”[74].

Tests done on the Mini

66.Dr Wong, for the prosecution, tested the Mini to see, amongst other things, whether there was any leakage of gas, what the dissipation rate of gas from the yoga ball was and how much concentration there was at different time intervals[75]. His methodology was relatively simple.  He filled up a yoga ball with carbon monoxide but, when he tried to place it in the boot of the Mini, he found he was unable to close it.  Accordingly, he let out some of the contents in order to fit it into the boot, removed the stopper and recorded the diffusion rate.

67.Dr Wong found there were variables which affected the amount of carbon monoxide inside the vehicle; for example, opening a door.  A later test was to reveal a huge leakage of gas once the doors were opened[76]. It should be remembered that the window of the Mini had been smashed when the police arrived and, at the time of the tests, it had simply been covered.  The Judge reminded the jury that the carbon monoxide concentration would have been higher on the day the deceased died before the window had been broken[77].

68.Professor Li, for the defence, used a different Mini Cooper of the same year and model for his experiments.  He also found that a yoga ball with a volume of more than 65 litres would not fit into the car[78]. Instead of carbon monoxide, Professor Li used nitrogen, which has the same density, for his tests.

69.Mr McCoy, in his closing address to the jury, made the point that if someone was aware of how dangerous carbon monoxide was, he could not have safely squeezed it into the boot; whereas, “someone who doesn’t understand how dangerous it is may well have done it.  It’s a ridiculous notion … somebody standing there with their hand, the lid halfway down, trying to squeeze it out”[79].

70.Professor Li was critical of Dr Wong’s methodology, describing it as misleading.  His main complaint was that the yoga ball would not have deflated at a constant rate.  He also observed that when one of the doors of the Mini was opened, there had been a huge drop in carbon monoxide concentration.

The diffusion rate of carbon monoxide: the timeline defence

71.There was no dispute that both Madam Wong and Lily had died from carbon monoxide poisoning.  As we have already mentioned, the carbon monoxide concentration in the blood of Madam Wong was 50%; and in Lily, 41%.  The way the car came to rest, with its rear protruding out into the road, suggested that the carbon monoxide concentration must have affected Madam Wong’s ability to drive properly[80].

72.The defence sought to show, from the evidence of the diffusion rate, that the yoga ball must have been put in the car shortly before they left the house, suggesting that it could not have been the applicant who placed it there, since he was teaching at the time.  Professor Li testified that there had been a 40% reduction of volume in the first half hour of his test[81].

73.Siti testified that Madam Wong had taken the children to school at 7:30 am and returned home at 8:30 am.  At about 10:00 am she had gone upstairs.  According to his video recorded interview, the applicant did not have to go into work too early, but he did have to attend a student’s presentation at the Chinese University at about 1:30 pm[82]. The admitted fact dealing with this matter did not specify the exact time but merely averred that the applicant was at a presentation on 22 May 2015[83].  Furthermore, he was on duty at about 4:00 pm later that day in an operating theatre[84]; and it was while he was working, of course, that he received news about his wife and daughter.

74.The Judge noted in her summing-up that the window of opportunity for the applicant to have put the yoga ball in the Mini was between the time when Madam Wong returned from dropping the children at school, and before he left to do the presentation at the Chinese University[85]. She pointed out that the jury might have thought it unlikely he could have put the yoga ball in the car before 10 am, since Madam Wong was in the garden admiring flowers or playing on her mobile telephone in the sitting room[86].

75.Siti stated that she saw Madam Wong and Lily walking to the Mini carrying nothing, certainly not a yoga ball.  Siti said it was raining at that time and the car was not parked under cover, indicating that one would get wet when getting into the car; although the Hong Kong Observatory records suggested it did not start raining until 3:00 pm.  This was all relevant in weighing up how long it would have taken someone to get into the car in such conditions and how long the yoga ball might have been there.  Madam Wong had only managed to drive the car for 1.6 kilometres. 

76.Accordingly, the jury were required to consider the expert evidence against the factual matrix, in deciding when the yoga ball might have been placed in the Mini, and by whom.

An overview of the grounds of appeal

77.In order to have an overall appreciation of how this appeal has been presented and argued, it is necessary briefly to examine the litany of complaints and points which have been made by and within the numerous grounds of appeal.  In the course of this overview, we will make some obvious responses and general remarks which will be followed, where appropriate, by a more detailed discussion in respect of the specific complaint or point that is being addressed.

78.The grounds of appeal have been grouped under four general headings[87]. The first general heading is “Burden and standard of proof/approach to defence evidence/right to silence”.  There are in total 11 grounds of complaint under this heading. 

(i) Proof, defence evidence and the right of silence

79.Ground 1 complained that the Judge misdirected the jury as to the burden and standard of proof in that: (a) her directions suggested there was a burden on the defence requiring, variously, “reliable evidence” or evidence “of sufficient reliability and strength” or “good evidence” in order to cast doubt or destroy the prosecution case; and (b) her directions on speculation erroneously suggested that identical burdens and standards of proof applied to both the prosecution and the defence. 

80.We are of the view that there was nothing improper in what the Judge said when they are read in context.  She gave the correct directions to the jury on the burden and standard of proof in accordance with the Specimen Directions in Jury Trials, promulgated by the Hong Kong Judicial Institute, and went into considerable detail in explaining their task and function as judges of fact[88].The references to “reliable evidence” or “good evidence” in the context in which they were given did not shift the burden of proof as alleged.

81.Ground 2 suggested that the Judge misdirected the jury as to the burden and standard of proof by: (a) effectively inviting the jury to choose between the evidence of the experts whose evidence was conflicting; (b) reducing the issue of whether there was a rodent problem at the applicant’s home to a choice between the evidence of Siti and the applicant; (c) reducing the question of whether blood had been reinjected into rabbits during the preliminary experiments conducted to a choice between the evidence of Dr Dick Chow and Professor Wang; and (d) directing the jury about the conflicting prosecution and defence evidence as to whether it was raining, without reminding them that the defence evidence which might be true would suffice to create a reasonable doubt. 

82.In our view, the Judge fairly presented these issues by setting out very clearly the factual and contextual bases in which they arose.  The jury were consistently reminded that it was a matter for them how they assessed and evaluated the evidence in resolving the issues they were to consider.  The differences in the expert evidence and the issues to which they related were delineated by the Judge, and this included whether reinjection took place in the two experiments.  The Judge also canvassed the issue in respect of the rodent problem at the applicant’s home and the method of exterminating them by releasing carbon monoxide from the yoga ball; and the question of whether it was raining when Madam Wong and Lily got into the car.

83.Ground 3 contended that the Judge failed to direct the jury to acquit the applicant if his exculpatory statements in his video recorded interviews might be true. 

84.However, in our view, the Judge gave a clear and precise direction that if the account put forward by the defence “is or may be true”, they should find the applicant not guilty[89].  The Judge also stressed the applicant’s good character both in terms of credibility and propensity in respect of his answers in the video recorded interviews[90].

85.Ground 4 averred that the Judge wrongly directed the jury in respect of the drawing of inferences to the effect that: (a) Madam Wong’s ignorance of the fact that the yoga ball contained carbon monoxide was a reason for excluding her as the person who put the yoga ball in the car; (b) if there was no stopper for the yoga ball in the car, Lily could in effect be ruled out as the person who placed the yoga ball there because there was no reason for her to throw away the stopper; and (c) the jury might infer that it was raining from the fact that the windscreen wipers were on when the car was found, when it was equally likely that the wipers had been switched on clumsily and accidentally.  

86.In our view, these were all valid propositions arising from the evidence for the jury to consider, particularly as to whether Madam Wong or Lily had put the yoga ball in the car.  It was the defence suggestion that Lily may have put the yoga ball there because of her phobia of insects.  In order to assist the jury, the Judge gave the appropriate directions on circumstantial evidence and inferential reasoning.  She clearly explained how inferential reasoning worked and how it could be employed in order to assist the jury to make findings of fact.  The Judge emphasised that it was for the jury to consider and ultimately resolve the issues mentioned.  She also told them that they were free to reject any view she might express on the evidence unless it accorded with their own view of the evidence[91].

87.Ground 5 complained that the Judge failed to direct the jury not to draw an adverse inference from the applicant’s failing to tell the police that the yoga ball contained carbon monoxide, and to remind them of the applicant’s right to silence. 

88.We are of the view that the issue of the applicant exercising his right of silence, did not arise, as he voluntarily answered initial inquiries by the police and he subsequently acknowledged that he failed to disclose relevant information to them.  Furthermore, no point was taken at trial that the applicant had a right of silence or was purporting to exercise it at the time he was answering initial inquiries from the police.  We shall deal with this ground in greater detail later.

(ii) Failure to summarise the defence case

89.The second general heading is “Failure to summarise the defence case/unbalanced summing-up/inaccurate summing-up”.  There are in total 9 grounds of complaint under this general heading. 

90.Ground 6 alleged that the Judge failed to remind the jury of the defence case that the timeline and possible levels of carbon monoxide concentration meant that if the applicant had placed the yoga ball in the car, the carbon monoxide level would have been such that Madam Wong could not have been overcome so quickly.  

91.In our view, this complaint cannot be justified. The Judge comprehensively set out the evidence and the respective cases of the prosecution and the defence, referring in detail to the criticisms of the prosecution case and the range of possibilities as to what might have happened as submitted by the defence.  More particularly, she referred to the factual matters and empirical data relating to the possible levels of carbon monoxide concentration as well as the times of relevant events and other factors for the jury to consider[92]. It was accepted by Mr McCoy in his closing address that there were a number of variables and unknowns in the case which the jury had to bear in mind when assessing the evidence and the propositions he was putting forward for their consideration.

92.Ground 7 contended that the Judge failed to remind the jury that the applicant’s research was genuine, and in particular failed: (a) to summarise the defence case as to the extra haemoglobin effect hypothesis, by directing the jury that Professor Wang had not addressed the matter and that it was not important; and (b) to remind the jury of the defence case that the focus of the research was not on carbon monoxide or carbon monoxide poisoning but on supplemental oxygen and tissue hypoxia. 

93.The Judge presented an extensive overview of the scientific evidence in respect of the applicant’s research, and hence the subject of this complaint only focuses on a small aspect of it.  The overview was in fact very detailed and appropriately addressed the evidence and the salient issues at trial[93]. This included the evidence as to the extra haemoglobin effect hypothesis and the nature and terms of the research protocol.  Relevantly, one issue before the jury was whether the applicant’s research was a sham, and even if it was not, whether it was a means by which he sought to obtain carbon monoxide.  The Judge, on several occasions, reminded the jury of the defence case that the applicant’s research was genuine and that he openly engaged in experimenting with carbon monoxide[94]

94.Ground 8 argued that the Judge incorrectly summarised the evidence as to whether blood was reinjected into the rabbits when choosing between the evidence of Dr Dick Chow and Professor Wang. 

95.It was highly relevant that Dr Dick Chow was involved in the two experiments and therefore able to give a first-hand account of what took place, whilst Professor Wang was not involved.  We cannot see how the Judge can be criticised for noting this fact for the jury to consider when addressing this issue.  In any event, the jury were well aware of the difference in the evidence between the prosecution witnesses (Dr Dick Chow and Ms Tam) and the defence expert, Professor Wang, as to whether reinjection took place when the two experiments were conducted.  Professor Wang’s evidence was an opinion based on his interpretation of the carbon monoxide oximeter readings. 

96.Ground 9 complained that the Judge’s summing-up was unbalanced in repeatedly stating or implying that “a crime” had been committed. 

97.The applicant stood accused before the jury of having committed the crime of murder.  There would have been no conceivable misapprehension or misunderstanding on the part of the jury that they were to determine whether the applicant was guilty or not guilty of the murder of Madam Wong and Lily.  The reference to “a crime” would not have impacted adversely on the applicant.  We would add that no complaint was made by leading counsel for the applicant at trial about this matter.  We shall deal further with this ground later.

98.Ground 10 suggested that the Judge misdirected the jury as to the financial benefit to the applicant in the event of Madam Wong’s death.  There is no substance in this complaint.  This issue was appropriately dealt with by the Judge in her summing-up.  The jury heard evidence as to the financial status of the applicant and Madam Wong and they were reminded of the defence case on this issue in the closing address of leading counsel.  The Judge, in any event, explained that the financial information was included in the admitted facts and that there was no evidence that the applicant wanted to get rid of his wife in order to get the money.  She told the jury: “So that is another aspect that you have to consider; no apparent motive for the defendant to kill his wife”[95].

99.Ground 11 concerned the Judge’s directions in respect of the stopper to the yoga ball. 

100.It is complained that the Judge misdirected the jury as to the yoga ball stopper and whether it was missing on 22 May 2015[96] by: (a) omitting to mention possible errors and the unreliability of the exhibits officer in not finding the stopper; (b) omitting to direct on the possibility that the stopper had been mishandled or misplaced in the six months which had elapsed before the exhibits officer had searched the Mini, and inviting the jury to speculate as to whether the stopper was missing when Madam Wong and Lily were found; and (c) directing the jury that if they were sure there was no stopper, there was no reason for Lily to throw it away but omitting to mention that if it had been the applicant, he too would have had no reason not to leave the stopper in the Mini. 

101.We will address this ground in greater detail later.  The points raised by the applicant about the exhibits officer’s handling of the exhibits and possibly misplacing the stopper were canvassed by Judge in her summing-up.  We are satisfied that the Judge properly summed up the evidence on this issue.

(iii) Removal of issues from jury

102.The third general heading is “Removal of issues from jury”.  There are in total 5 grounds of complaint under this general heading. 

103.Ground 12 alleged that the Judge erred by effectively removing the issue of unexplained accident from the jury and in doing so: (a) usurped the function of the jury; (b) placed a burden on the defence to prove the applicant’s innocence; and (c) failed to leave the defence case of unexplained accident to the jury.

104.There is simply no substance in this complaint. In short, the prosecution case was that the applicant had placed the yoga ball in the Mini and released the carbon monoxide from it, knowing his wife would be using the car.  As part of the defence case, alternative possibilities were presented to the jury, principally that Lily put the yoga ball in the car because of her phobia of insects.  The Judge made it abundantly clear to the jury, without demur from the defence, that the main issue for them to decide was who put the yoga ball filled with carbon monoxide inside the Mini and why[97]?  The Judge also reminded them that in addressing the two counts of murder against the applicant, they had to consider the ingredients of the offence, and whether they were sure that he put the yoga ball in the Mini and removed the stopper, so that the carbon monoxide leaked into the cabin of the car.  If they were not sure, then the prosecution case had failed[98]. There was a strong case against the applicant that it was he who placed the yoga ball in the Mini.  In his closing address to the jury, Mr McCoy stated: “Now, let me tell you straight away the defendant in this case is morally responsible for the deaths of his daughter and wife.  If that guy had not brought that home, it couldn’t have been there for it to end up in the car.”[99]  Although the case against the applicant was strong, leading counsel raised other possibilities as to who might have placed the yoga ball in the car and why[100].

105.The applicant relied on R v Henderson and Others [101]for the proposition that there was a realistic possibility of an unknown cause of death in the present case and the jury should have been reminded of that possibility.  Henderson and Others involved three appeals which concerned what is termed ‘shaken baby syndrome’, where the cases turned solely on the evidence of experts.  Whilst expert evidence featured predominantly in the present case, it was not the sole evidence on which the case fell to be determined.  The cause of death was known and the issues between the respective experts were clearly defined.  The focal point at trial was who put the yoga ball in the Mini and released its deadly contents. 

106.We are bound to say that there was little if any evidential basis to support a realistic possibility that the deaths of Madam Wong and Lily were an accident.  The applicant had filled two yoga balls with carbon monoxide and stored them overnight in the Toyota Alphard, placing carbon monoxide detectors inside the vehicle.  The next day he found one of the yoga balls had released its contents of carbon monoxide but took the other yoga ball to his home residence.  According to the applicant, he was intending to release the deadly gas to kill rodents at the home.  He said he left the yoga ball in his home for that purpose and went off to work.  He did not install any carbon monoxide detectors at home, nor did he warn anyone about the deadly contents of the yoga ball, except to tell Lily that it contained a poisonous gas, with which he intended to kill rats over the weekend[102]. The yoga ball ended up in the Mini and its poisonous gas was released, killing Madam Wong and Lily.  Clearly someone put the yoga ball in the Mini and removed its stopper.

107.Despite the applicant’s complaint in this ground, we note the Judge left the defence argument that the death of Madam Wong and Lily was an accident for the jury to consider.  When taking the jury through the evidence, she told them that they “may think” they could eliminate Madam Wong and Lily as being responsible for putting the yoga ball in the Mini.  However, she also reminded them of the defence suggestion that Lily was experiencing difficulties and may have accidentally killed herself and her mother because she wanted to kill insects inside the Mini[103]

108.Ground 13 complained that the Judge erred in directing the jury that the yoga ball was in the car before Madam Wong got into it, thereby usurping the function of the jury and removing from them the defence case that it might have been done by Madam Wong or Lily. 

109.The evidence of Siti was that she observed Madam Wong and Lily were not carrying anything with them when they got into the car. The Judge carefully guided the jury through the evidence as to whether Siti, Madam Wong or Lily may have placed the yoga ball in the car as well as addressing the case for and against the applicant[104].  As already noted, the Judge reminded the jury of the defence argument that the death of Madam Wong and Lily was an accident as well as reminding the jury of Lily’s character and fear of insects, which were relevant to the defence theory of Lily having attempted to kill insects inside the Mini[105]. There is simply no merit in this complaint.

110.Ground 14 contended that the matters set out in Grounds 6 to 13 resulted in the summing-up being unbalanced in that it so undermined the defence case as to be tantamount to a second prosecution closing address. 

111.Mr Grounds took issue with the Judge’s directions in respect of (i) Dr Wong’s evidence concerning the gas concentration and dissipation in the Mini and the criticisms of his tests by Professor Li; (ii) the level of carbon monoxide in the Mini under certain conditions, such as if it was raining and if the car window was broken; (iii) the exhibits officer’s evidence about not finding the stopper to the yoga ball and Lily having no reason to throw the stopper away if she put the yoga ball in the Mini; and (iv) the rat problem at the family home and May May’s evidence on this issue.

112.These matters were mentioned by the Judge in her summing-up.  She addressed the evidence of Dr Wong and Professor Li and the tests they had performed.  She discussed the levels of gas concentration and dissipation in the Mini as variously explained by the experts.  She referred to the challenge to the exhibits officer’s evidence and the issue as to his reliability of what he found in the Mini.  She summarised the evidence of May May and took the jury to the defence photographs of the home in relation to the rat problem[106].

113.In our view, this ground is unarguable: the summing-up was plainly not unbalanced, nor did it conceivably amount to a second prosecution closing address.

(iv) Expert evidence

114.The fourth general heading concerns “Expert evidence”.  There are no less than 19 grounds of complaint under this general heading. 

115.Ground 15 alleged that the Judge failed to summarise the expert evidence correctly, nor was her summary properly balanced.

116.This complaint is expressed in extreme language and is without substance or justification.  There is no basis to suggest that the Judge’s summary of the expert evidence was unbalanced.  In our judgment, the Judge gave a very comprehensive summary of the expert evidence, which properly reflected the salient matters and issues raised by or between the respective experts.

117.Ground 16 complained that the Judge in respect of the expert evidence of the applicant’s research: (a) wrongly suggested that only Professor Buckley disputed the validity of the hypothesis in relation to the extra haemoglobin effect when Professor Wang had also done so; (b) erroneously directed the jury that as the applicant had not given evidence there was no evidence that blood had been reinjected into the rabbits and did not remind the jury of the applicant’s statements on the matter in his first video recorded interview; (c) failed to direct the jury that Professor Kwong’s conclusions were vitiated to the extent they were based on hearsay as to whether reinjection had occurred; (d) failed to summarise Professor Wang’s evidence by omitting to mention the significance of the research; (e) failed to instruct the jury as to the clinical relevance of the research; and (f) stressed procedural and formal criticisms concerning the research by: (i) unfairly contrasting the bulk of exhibits P62 and P63 with the “slim protocols” for the research (exhibits P113A, B and C); (ii) directing the jury to consider whether the applicant and his research protocols should have justified the methodology of extracting blood from rabbits, rather than focusing on the stated purpose and the opinions as to the value of the research. 

118.This ground essentially focuses on the issue of the genuineness of the applicant’s research involving carbon monoxide.  The complaints under this ground are not justified and we have concluded that the Judge very fairly summarised the evidence and identified the relative arguments of the parties in respect of this issue.  The specifics of this ground will be addressed later in the judgment.  We shall also refer to the relevant parts of the Judge’s summing-up in respect of the applicant’s research, because it reveals that the matters now complained about by the applicant were fully and properly addressed by the Judge.

119.Ground 17 contended that the Judge failed to sum up the expert evidence as to the rate of deflation of the yoga ball, the levels of carbon monoxide in the Mini and the rate of diffusion or dissipation of the carbon monoxide, and consequently as to the relevant timeline, in that she: (a) erred in directing the jury as to the actual carbon monoxide concentration in the Mini on 22 May 2015, which would have been higher when allowance was made for leakage from an imperfectly sealed broken window of the car; (b) summarised the evidence of Professor Li in a confusing and inaccurate manner; (c) erred in directing the jury to consider photographs 4 and 5 of exhibit P121, which showed the yoga ball in different positions in the boot of the Mini, when deciding how the yoga ball had deflated, because: (i) the photographs had been taken several hours after the discovery of the deceased in the car, but there was no evidence as to which photograph had been taken first; (ii) there was no evidence as to the position or state of the yoga ball when the deceased were first discovered; and (iii) it was apparent that the yoga ball must have been handled at least between the taking of photographs 4 and 5; (d) erred in directing the jury that they could choose between the evidence of Dr Wong and Professor Li as to the test performed about deflation, leakage and dissipation; and (e) failed to provide proper guidance as to the tables relied on by Dr Foo Ka Chung (PW31) (“Dr Foo”) and Dr Wong as to the effects of exposure to carbon monoxide in different concentrations and for different periods, and did not properly remind the jury of the defence case as to the timeline. 

120.This ground is concerned with the timeline defence, which to the extent it was argued at trial, was properly addressed by the Judge in her summing-up to the jury.  Most of the points raised by this ground were part of the factual matrix set out by the Judge in her summing-up. This is another ground where we will have to recite in full the relevant passages of the Judge’s summing-up in order to dispel the criticism that the Judge did not fairly or properly present the defence case on this point.

121.Ground 18 complained that the Judge failed to direct the jury that since the evidence was circumstantial, if there was a reasonable explanation of the facts consistent with innocence, they should find the applicant not guilty. 

122.The Judge not only gave the standard directions to the jury on inferential reasoning, but also repeated, when it was contextually relevant, that an adverse inference could only be drawn against the applicant when it was the only reasonable inference from the facts proved[107].  She explained to the jury that in addressing a circumstantial case they should “consider whether the evidence upon which the prosecution relies in proof of its case is reliable and whether it does prove guilt, or whether on the other hand it reveals any other circumstances which are or may be of sufficient reliability and strength to cast doubt upon or destroy the prosecution case.”[108] 

123.Ground 19 averred that the Judge failed adequately to: (a) direct the jury as to how to approach the expert evidence and how to identify areas of dispute; (b) summarise clearly and in a balanced manner the evidence for and against the propositions advanced; and (c) provide guidance to the jury as to how to evaluate conflicting expert evidence, explaining the basis for distinguishing reliable from unreliable evidence. 

124.This is a general complaint without any substance. The Judge gave a comprehensive summing-up in respect of the expert evidence and the jury would have been well aware of the issues and matters they had to consider and resolve.

125.The last ground of appeal, Ground 20, was in general terms and asserted that in all the circumstances of the case the applicant’s convictions were unsafe and unsatisfactory.  It was contended that the applicant was deprived of a fair trial and there remained a lurking doubt as to the correctness of his convictions.

The form and content of the grounds of appeal

126.As we have already noted, no less than 20 numbered grounds of appeal have been advanced in this matter.  However, several of the grounds have sub-grounds and sub-sub-grounds, which are in many cases separate and discrete complaints, and which in reality take the overall number of grounds to more than 40 grounds of appeal spanning some 22 pages.  The grounds are contained in a document entitled Amended Perfected Grounds of Appeal Against Conviction, signed by counsel, Mr Christopher Grounds and Mr Andrew Bullett, who, together with Mr Franco Kuan, represented the applicant at the appeal.  All three counsel were new to the appeal, as were their instructing solicitors.  We wish to say something about the form and content of the grounds of appeal in this case, since they demonstrate a growing and unacceptable trend in appeals before this Court.  It is necessary to remind practitioners that an appeal should identify only those errors of ultimate fact or law which affected the result.  It should not condescend into complaints about minutiae or espouse new and unsupported theories about the case.

127.We must begin by reciting and emphasising section 7(1) – (8) of Practice Direction 4.2 concerning Criminal Appeals to the Court of Appeal (“PD 4.2”), which was signed by the Chief Justice on 19 February 2019, and came into effect on 11 March 2019.  Under the heading ‘Perfected Grounds’, section 7(1) – (8) of PD 4.2 stipulates:

(1)  Perfected grounds of appeal should not be settled by counsel or solicitors unless they consider that they are properly arguable.

(2)  Perfected grounds of appeal must be provided as a single document.

(3)  Perfected grounds of appeal must be prepared with sufficient precision to enable the Court to identify readily and clearly the issues and the matters relied upon.

(4)  Perfected grounds of appeal must commence with a statement of the conviction and/or sentence from which leave to appeal is sought; and a brief statement of the key issues at trial relevant to the application or appeal.

(5)  Perfected grounds of appeal must be numbered consecutively and must be specific rather than expressed generally.  Where particulars are required to identify the matters relied upon, the particulars must be concise.

(6)  Perfected grounds of appeal must state as concisely as possible the complaint which constitutes the ground of appeal.  It should not normally be necessary to state the point in more than one or two sentences; thus, for example, specifying the statement or finding of the Judge said to be in error and, in its barest essence, what the suggested error was; or what he should have said that he did not say.

(7)  Perfected grounds of appeal must not contain submissions in support of the grounds.

(8)  Where perfected grounds of appeal are filed which do not accord with these directions, they are liable to be returned for correction with a direction that the case be not listed until a compliant set of perfected grounds is filed.

128.With respect to counsel’s endeavours, and acknowledging that the case was lengthy, albeit, we think, not unduly complex, the Amended Perfected Grounds of Appeal infringe almost all of the above sections of the Practice Direction.  Firstly, what we assume was an attempt to provide “a brief statement of the key issues at trial relevant to the application or appeal” (in accordance with section 7(4) of PD 4.2) in fact runs to 8 pages and goes far beyond what was necessary to state the key issues at trial; one of which, for example, was that there was a conflict of expert evidence as to when the yoga ball may have been placed in the Mini so as to overcome the occupants by carbon monoxide poisoning.  Instead, there is an extensive analysis of the respective expert evidence to the following effect:

“IV. Expert evidence of a technical and medical nature was called as to the rate of deflation of a yoga ball containing CO, the rate of leakage of CO from the Mini Cooper or a similar model of car, the likely consequent CO concentration levels in the car at different times and their likely medical consequences, etc.

V. Prosecution evidence as to these issues was given by Dr Wong Koon Hung [PW27], a government chemist, over one and a half days. Tables as to the effects of various ambient CO concentrations were included in the autopsy report by the pathologist, Dr Foo Ka Chung [PW31], and the report by Professor Kwong Yok Lam [PW36] [exhibits P59, P55A at B162, 319], suggesting that for a person to be overcome in a few minutes required a CO concentration level in the region of 6,400 parts per million (“ppm”), a level which on Dr Wong’s evidence would not have been possible, given the sharp drop in gas concentration observed once the car doors had been opened to allow the Deceased to board, if the yoga ball had been placed in the car an hour or more before the Deceased set out. As the car was not airtight there would also be leakage of CO from the car when doors and windows were closed and this would be greater when the car was in motion. The yoga ball when fully inflated could not fit into the car boot, where it was found.

VI. Defence expert Dr Li Chi Kwong [DW4, exhibit D13], an engineer, pointed to various shortcomings in Dr Wong’s experiments, including poor methodology, inaccurate instruments and unexplained discrepancies in results. He considered that Dr Wong’s conclusions as to CO concentration levels were unreliable and over‑stated. Dr Li’s experiments showed that:

a. natural leakage from the same model of stationary closed car would be at the rate of 50 litres per minute (also derived from Dr Li’s reading of Dr Wong’s experiments) from a total internal car volume of 2,306 litres;

b. the yoga ball’s internal volume was 100 litres but a ball containing more than 65 litres of CO would not fit in the car boot, a figure with which Dr Wong was unable to disagree;

c. the yoga ball could not completely deflate by passive deflation and without external force.”

129.On another issue of whether the applicant’s research was a sham or genuine, an issue which could likewise have been expressed in a single sentence, the Amended Perfected Grounds of Appeal again contain a detailed analysis of the respective experts as follows:

“VII. In addition, medical experts were called by the prosecution in support of the prosecution argument that the Applicant’s proposed research project at CUHK involving oxygen therapy and CO poisoning (the “Research”) was not genuine but a sham in order to obtain CO; although it was not disputed that the Applicant had brought home CO in a yoga ball: the prosecution arguing that this was not in order to kill rodents but to murder Madam Wong. Professor Kwong Yok Lam [PW36], a haematologist, argued that the mechanism of CO poisoning was well understood and extensively researched; that the standard clinical treatment was normobaric (100% oxygen) or hyperbaric (200-300% oxygen) oxygen therapy in order to remove carboxyhaemoglobin (“COHb”), the product of CO binding with haemoglobin, which impairs the carriage of oxygen to the tissues; that the neurological problems often experienced by survivors were attributable to so-called extra-haemoglobin effects; and that the clinical outcomes did not correlate well with COHb levels, rendering the Research not useful; that the Research was poorly designed, lacked originality and clinical translatability and was not within the expertise and usual research area of the Applicant.

VIII. Defence experts as to the Research, Professor Nicholas Buckley [DW1, exhibit D11], a specialist in clinical pharmacology and toxicology, and Professor Wang Chi Chiu [DW6, exhibit D15], a medical researcher with an interest in oxidative stress and free radicals, took issue with various matters, arguing, in particular, that:

a. research protocols typically started as outlines and underwent change before finalisation and this was not a valid criticism of the Applicant’s draft research protocols. Any flaws in the Research did not imply it was sham;

b. the Applicant had both an interest in hypoxia, which requires oxygen therapy, and experience in animal experimentation;

c. the attribution of neurological problems of survivors of CO poisoning to extra-haemoglobin effects was not supported by evidence but was a mere hypothesis not widely accepted;

d. those who did accept the extra-haemoglobin effect hypothesis might be influenced by financial considerations as hyperbaric oxygen therapy was lucrative;

e. Professor Kwong was not a toxicologist and had no background in CO or supplemental oxygen treatment. Some of his criticisms made no sense, e.g. his argument that since there had been many animal studies no more were required, or his statement that the half-life of CO poisoning in humans was well established, supported by citation of a study indicating a wide variation for unexplained reasons;

f. Professor Irwin relied for his argument on a single, controversial, paper written by people with an apparent interest in denying any controversy about oxygen therapy. That paper’s practice recommendations were not endorsed by e.g. the American College of Emergency Medicine or by NICE (the UK public body which provides guidance and recommendations on medicines and treatment for the National Health Service);

g. an alternative theory which had respectable support and merited research such as that of the Applicant was that such neurological problems were caused or contributed to by oxidative stress from free radicals generated by the oxygen therapy itself, thus it was valid and worthwhile to research the effects of different levels of oxygen;

h. the use of animals allowed a standard degree of CO poisoning to be achieved and was justified by the ethical and practical difficulties of studying oxygen therapy using different levels of oxygen in humans;

i. the optimal treatment of CO poisoning, including the use of oxygen, remained the subject of much research. It was unknown whether treatment with a lower oxygen level could improve outcomes by reducing tissue damage;

j. the various iterations of the Research protocols were concerned with CO poisoning or haemorrhage, the latter often occurring in pregnancy or childbirth, an area of clinical interest of the Applicant. Both conditions led to tissue hypoxia or hypoxic cellular damage and the standard treatment for both included supplemental oxygen;

k. the Applicant is a clinical anaesthesiologist and medical researcher. He was active in research relating to oxygen therapy for haemodynamic disturbance in obstetrics cases, the relevance of CO poisoning to this being that both caused tissue hypoxia;

l. CO inhalation methods did not achieve consistent baseline COHb levels, whereas these could be achieved by re-injection, as proposed. Experiments with 60% oxygen, about half way between 100% oxygen in normobaric oxygen therapy and 21% atmospheric oxygen, to see if this could be as effective with fewer adverse effects from the oxygen therapy itself, were scientifically justified;

m. the proposed Research was a basic animal study not a clinical study and had clinical and scientific potential.

IX. The defence thus contended that the Research was genuine; and further contended that the yoga ball had indeed been brought home in order to address a rat and mouse problem at the Applicant’s home, other methods having failed or been unsuitable; that the yoga ball may have been placed in the car by one of the Deceased: yoga balls were used by the family and Madam Wong went to yoga classes; and that the circumstances made accident of some unexplained kind likely and murder by the Applicant impossible.  Those circumstances included: the copious paper trail clearly leading to the Applicant as the likely source of the CO; the risk of killing not only Madam Wong but also any of the Applicant’s younger children; the timeline in light of estimated gas dissipation and leakage rates; the possible CO concentration levels when the Deceased boarded the car; the unpredictable range of outcomes from exposure to such concentration of CO; and the unpredictable consequences of driving after such exposure.”

130.All, with respect, that was needed was a short, concise statement of the essential issue to which the grounds related. Such issue could have been stated, as we have done, in one sentence.

131.Secondly, some of the grounds themselves are peppered with comment and submission.  Plainly, they do not in any way comply with the requirement that the complaint constituting the ground of appeal should be stated “as concisely as possible” (as stipulated by section 7(6) of PD 4.2).  While section 7(6) states that “[i]t should not normally be necessary to state the point in more than one or two sentences”, we accept that occasionally, it may be necessary to go further.  This is not such an occasion.  Ground 16, for example, reads as follows:

“16. In relation to the expert evidence as to the Applicant’s Research, the Learned Trial Judge:

a. erroneously directed the jury that only Professor Buckley disputed the validity of the hypothesis attributing neurological effects to the extra haemoglobin effect when Professor Wang had also done so in his report [B729-731, exhibit D13, pp 3-5];

b. erroneously directed the jury that as the Applicant had not given evidence there was (apart from the oximeter printouts [B351-363, exhibit P130H,]) no evidence that blood had been re-injected into the rabbits [A173-174]; whereas the jury should have been reminded of the Applicant’s statements on the matter in his first video interview [C136-137. Exhibit P43 counters 2071-2098];

c. on the other hand, gave unduly favourable treatment to the prosecution evidence as to the Research by failing to direct the jury that Professor Kwong’s conclusions were vitiated to the extent to which they also were based upon hearsay as to whether re-injection had occurred [A97, exhibit P96, p5, B329];

d. failed adequately to summarise the evidence of Professor Wang Chi Chiu [DW6] [A169, 170-171, 172, B726 exh D15], in particular omitting mention of the significance of the Research and the relevance of the professional interests and experience of the Applicant to the Research;

e. failed to direct the jury in clear terms that whether the Research had clinical relevance was not the only mark of its genuineness but that academic or theoretical value could also justify the Research [A166-170];

f. unduly stressed procedural and formal criticisms concerning the Research, by:

i. contrasting the bulk of exhibits P62 and P63 [B364-484] unfavourably with that of exhibits P113A, B and C, the slim protocols for the Research [B338-344], but in doing so [at A175] failing to point out to the jury that the former were past research applications to the Clinical Research Ethics Committee (“CREC”) of CUHK, whose purview was human experiments the requirements for which were complex, whereas the latter were draft applications for the Research to the Animal Experimentation Ethics Committee (“AEEC”) of CUHK which were required to be no more than 500 words in length [exhibit P124E, B366]. The comparison was thus inapposite and highly prejudicial;

ii. directing the jury to consider whether the Applicant in his Research protocol should have justified the methodology of extracting blood from rabbits [A175], rather than directing the jury to focus on the stated purpose of the Research and the opinions as to the value of the Research of the various experts.  In doing so the Learned Trial Judge misdirected the jury by implying that a proper protocol should have acknowledged that the Research intended to use methodology already used on dogs, as stated in an article, not in evidence, which had only been published after the protocol was prepared [A170-171].”

While Ground 17 is pleaded as follows:

“17. The Learned Trial Judge fail to sum up the expert evidence as to the rate of deflation of the yoga ball, the levels of CO in the Mini Cooper car and the rate of diffusion or dissipation of the CO therefrom, and consequently as to the relevant timeline, in a correct, balanced, fair and comprehensible manner. In particular the Learned Trial Judge:

a. erred in directing the jury that the actual CO concentration in the Mini Cooper on 22.5.15 would have been higher than was shown by the data collected by Government chemist Dr Wong Koon Hung [PW27] when allowance was made for leakage from an imperfectly sealed broken window of the car when the CO dissipation tests were conducted, whereas Dr Wong had testified that by then there was no significant gas escape and the leak had been sealed or further sealed [A180-181, 194, 198, B10, 12 , 20, 33, 35, 42, 70];

b. summarised the expert evidence of Dr Li Chi Kwong [DW4] in a confusing and inaccurate manner by saying that the yoga ball would not totally deflate without ‘passive pressure (sic) from outside’ [A189, 204, B80; exh D13 pp 4, 7, 8, C730, 733, 734];

c. further erred in directing the jury to consider photographs 4 and 5 of exhibit P121 [C586, 587], which showed the yoga ball in different positions in the car boot, when deciding whether or not the yoga ball had deflated naturally, or with pressure, or by folding, although it was not safe to do so since:

i. the photographs had been taken several hours after the discovery of the Deceased in the car, there being no evidence as to which photograph had been taken first;

ii. there was no evidence as to the position or state of the yoga ball when the Deceased were first discovered; and

iii. it was apparent that the yoga ball must have been handled at least in between the taking of photographs 4 and 5 [A200-201, 204, 91, 92];

d. erred in law in directing the jury that they could choose between the evidence of Dr Wong and Dr Li as to the tests performed about deflation, leakage and dissipation, instead of directing the jury that if the evidence of Dr Li on a particular issue might be correct the jury should treat the prosecution case on that issue as to that extent unproven [A193];

e. failed to provide proper guidance as to the tables relied on by Dr Foo and Dr Wong as to the effects of exposure to CO in different concentrations and for different periods [A143, 144, 198, 200], so that the jury was not properly reminded of the defence case as to the timeline.”

132.The prolix form of these (and, indeed, other grounds) raise a third concern, which is that some of the grounds put forward are in fact a number of different grounds masquerading as a single ground of appeal.  Ground 16, for example, complains of the Judge’s misstatements of fact, her many failures to remind the jury of evidence, her favourable treatment of prosecution evidence, her prejudicial comparison of the parties’ expert evidence and her failure to exclude hearsay evidence.  Ground 17 makes the same complaints, but in addition avers that the summing-up was confusing as well as inaccurate, failed to give proper guidance to the jury and erred in law by allowing the jury to “choose between the evidence of” a prosecution and a defence expert witness, an issue which goes to the burden and standard of proof.

133.Fourthly, there is considerable repetition and obvious overlap between the various grounds of appeal.  For example, Ground 16(d), which complains of an incorrect direction as to the burden and standard of proof where expert evidence was in conflict, is already part of Ground 2(a).  Ground 4, which takes issue with the inferences in respect of who might have put the yoga ball in the Mini, overlaps with Grounds 12 and 13, which complains that the Judge suggested that Madam Wong and Lily could be eliminated as suspects. Ground 6, which takes extensive issue with how the jury were directed on the evidence about the diffusion rate of carbon monoxide, is much the same as Ground 17(a) – (e), in particular Ground 17(e), and Ground 19(a) – (c).

134.Ground 1, which complains about the directions on the burden of proof, inferences, and how the Judge wrongly placed a burden of proof on the applicant is virtually the same as Ground 18 and substantially overlaps with Ground 3.  Ground 7 focuses on the applicant’s experiments on rabbits and the Judge’s directions thereto, which in substance are the same allegations in Ground 16(a) – (f).  Ground 14, which is a complaint of an unbalanced summing-up, has significant overlap with Ground 6 and the complaints concerning the timeline defence; as well as with Ground 4 concerning inferences, one of which concerned whether or not it was raining.  And Ground 8 overlaps considerably with Ground 16(b) and (c), which concerns whether or not blood was reinjected into the rabbits.

135.We must make clear that if, in future, perfected grounds of appeal are filed which do not, as these grounds manifestly do not, comply with PD 4.2, they will be “returned for correction with a direction that the case be not listed until a compliant set of perfected grounds is filed” (as stipulated by section 7(8) of PD 4.2).

136.There are a number of important reasons why this Court must insist on perfected grounds of appeal being filed properly and in accordance with PD 4.2.  The first important reason is that the Court of Appeal is an appellate court, not a trial court.  Much of counsel’s arguments in this appeal were involved with solely factual matters, or arguments as to why the evidence of one witness undermined that of another witness, or why one piece of evidence should have been accepted and another rejected.  And, on several occasions during the appeal, defence counsel seemed to be making a jury speech to us.  The discipline of PD 4.2 requires appellate counsel to focus on “properly arguable” grounds of appeal which have a reasonable prospect of unsettling the conviction (or reducing the sentence).  Appellants are not on trial: they have already been convicted.  It therefore behoves appellate counsel appealing against conviction to identify, not why the appellant should not be convicted, but why the conviction should be set aside as unsafe and unsatisfactory, or because of a wrong decision on a question of law, or because of a material irregularity in the course of the trial.

137.The following grounds of appeal demonstrate complaints that were entirely without substance and simply not properly arguable.

The “keys” argument

138.It was argued before us, in a 19-page Reply signed by Mr Grounds and Mr Kuan, that it was “an admitted fact of great significance” that the applicant did not use his key to the Mini on the date of the offence[109]. Therefore, “it had to be either Madam Wong and/or Lily who had put the yoga ball in the Mini for a yoga related purpose, pesticide, suicide or any other reason.  It was an unfortunate and unexplained accident.  The applicant is innocent”[110]. In argument before us, it was submitted that since it was an admitted fact that “the last updated record for the car key of Deceased 1 (Madam Wong), Exhibit 2 was at 14:18 on 22 May 2015”, it could not have been the applicant who placed the yoga ball in the boot of the Mini[111].

139.This argument is not adverted to anywhere in the 20 Amended Perfected Grounds of Appeal: it came in by way of the applicant’s Reply and fell to Mr Kuan to present it.  Yet, the point made before us on appeal was never one engaged by either leading counsel at trial: for the defence, the one reference to the time on Madam Wong’s key in Mr McCoy’s speech to the jury was in the context of trying to establish a rough time at which Madam Wong and Lily must have got into the car and arrived at the bus stop[112]. Nor was the issue, which is now said to be so significant and demonstrative of the applicant’s innocence, regarded as such at trial, no doubt for the very simple reason that there was no evidence as to whether the Mini was locked or, if it was, when it was unlocked, prior to Madam Wong and Lily setting off to pick the two children up from school, or what, in technical terms, “the last updated record for the car key of Deceased 1” actually meant; for example, whether it recorded the time of the unlocking of the car, the turning on of the engine or, even, the turning off of the engine.  With respect to counsel, this is not a submission (let alone a ground, which it was not) that should have been put forward on appeal before this Court.

The reference to a “crime”

140.Mr Grounds then took specific issue with the Judge’s expressed or implied references to the fact that a “crime” had been committed (Ground 9), notwithstanding that some of the references are verbatim recitals by the Judge of standard Specimen Directions making reference to the commission of “the crime” (for example, in relation to circumstantial evidence).  The jury would have been well aware that it was the prosecution case that the applicant had murdered his wife and daughter; and that it was the applicant’s case that he had had nothing to do with their deaths, which may simply have been a tragic accident.  No one reading the summing-up as a whole could conceivably have thought that the Judge was putting the case against the applicant any higher than an allegation by the prosecution and it is frankly absurd to suggest otherwise. 

141.This ground should never have found its way into the Amended Perfected Grounds of Appeal to be argued before this Court.

The financial arrangements

142.It was further argued that the Judge’s reference to the applicant’s and Madam Wong’s financial arrangements misdirected the jury that there may have been a financial motive on the applicant’s part for killing his wife (Ground 10).  However, there were Admitted Facts as to the parties’ financial arrangements[113] and the Judge was making a point, in a section dealing with the absence of motive on the part of the applicant[114], that was in fact in favour of the applicant.  She said[115]:

“There is evidence, of course, about their financial arrangement. It is in the Admitted Facts. Of course, naturally, if anything happened to either one of them and if there is no divorce, the other party would get the share of the deceased one, but there is no evidence, direct evidence before you that the defendant wanted to get rid of the wife in order to get the money. So that is another aspect that you have to consider; no apparent motive for the defendant to kill his wife.”

No complaint was made at trial about the Judge’s direction by leading counsel for the applicant and one can readily understand why.  Read properly and in context, it forms part of a section in the summing-up that was actually favourable to the applicant, inasmuch as it instructed the jury that there was no evidence that either mistress or money were motives for the applicant to kill his wife.

143.Yet, Mr Grounds submitted that the mere mention of this matter, notwithstanding that the evidence was contained in the Admitted Facts, had the contrary effect on the jury, for it “excited” by way of “innuendo” the possibility that the applicant did have a financial motive for killing his wife.  This argument is wholly untenable and unrealistic, and flies in the face of what the Judge was at pains to make clear in the applicant’s favour.  It, also, is not a properly arguable ground of appeal.

The burden and standard of proof

144.Equally unarguable is the submission made under Ground 2(a) and (c) that the Judge misdirected the jury as to the burden and standard of proof, by inviting the jury to choose between the evidence of the experts, when it may not have been possible to reach a firm conclusion as to who was right and who was wrong.  The Judge in fact gave the standard direction on expert evidence, in accordance with Specimen Direction 32.  She said[116]:

“A witness called as an expert is entitled to express an opinion in respect of his findings or matters which are put to him and you are entitled and would no doubt wish to have regard to this evidence and to the opinion expressed by the expert when coming to your own conclusion about this aspect of the case. You should bear in mind that if, after having given the matter careful consideration you do not accept the evidence of the expert, you do not have to act upon it. Just because the evidence comes from the expert does not mean that you must except it, no. You consider it and you think that, ‘No, I am not going to accept any particular experts evidence’, you are perfectly entitled to do that.

Now, you are, of course, aware there are conflicting views expressed by some experts for the prosecution and for the defence. So where two or more experts have conflicting views, it is for you to decide are you accepting either one or neither, up to you. You decide who’s evidence you accept, whose evidence you do not.” (Emphasis added)

145.At the commencement of her summing-up, the Judge had given a general direction on how the jury were to approach the evidence of witnesses[117]:

“You and you alone must decide what evidence you accept, what evidence you do not accept and what evidence you are not sure about. And when you come to consider your verdict, you and you alone must do that. You do not have to decide every point which has been raised, only such matters as will enable you to say whether the charges laid against the defendant have been proved. You will do that by having regard to the whole of the evidence and forming your own judgment about the witnesses on which evidence is reliable and which is not. The evidence consists of Admitted Facts, testimony of various witnesses and various exhibits.” (Emphasis added)

146.The argument that the jury may somehow, despite these directions, have decided that they must accept one or other of the experts where their evidence on a particular issue was in conflict is an untenable proposition, as well as wholly unrealistic in the context of a trial where the jury would have well understood the areas of agreement and disagreement between experts.  Again, with respect, this argument should never have found its way into this set of grounds of appeal.

147.A second important reason why practitioners must comply with the discipline of PD 4.2 when filing grounds of appeal is that where, as here, appellate counsel are new to the case, their duty to settle grounds which are properly arguable may require them to consult previous trial counsel.  In HKSAR v Apelete (No 1)[118], this Court made it clear that:

“…the duties which we have set out in respect of complaints of flagrant incompetence, shall also apply to any ground of appeal which seeks to assert that something happened or did not happen at trial, which should or should not have occurred. In such a situation, fresh counsel on appeal will also be required to make inquiries of the previous legal representatives as to what was done or not done, and why. Having done so, they are likewise required to certify in the grounds of appeal themselves that they have complied with this duty.”

Fresh counsel at an appeal, particularly where the solicitors are also new to the case, will seldom have a full appreciation and proper understanding of the dynamics of the trial and the real issues between the parties, and the tactical and forensic decisions employed by trial counsel in conducting the defence, which are usually derived from trial counsel’s anticipation of what witnesses are likely to say in accordance with their depositions, from their assessment of the actual evidence as it emerges, from their perception of how the jury is receiving that evidence, from their instructions and, sometimes, from their discussions with the prosecution either inside or outside court, unless they ask trial counsel.

148.The difficulties which arise when fresh counsel on appeal advance new arguments impugning the conduct of the defence case at trial has recently been discussed in HKSAR v Abedin Minhajaul[119].  In that appeal, defence counsel was required to appear before the appellate court to explain his conduct of the defence case at trial, notwithstanding that no complaint of flagrant incompetence had been levelled against him by the appellant’s new legal team.  Macrae VP identified two particular problems that commonly arise in such circumstances:

“56. … The exercise has illustrated two particular problems that appeal courts face whenever fresh counsel and solicitors appear at an appeal in place of the trial lawyers. Firstly, fresh counsel at appeal will invariably have an imperfect and incomplete understanding of the real areas of tension and dispute at trial, the tactical measures employed by counsel to deal with them and the dynamics and considerations underlying such measures. Moreover, tactical decisions taken as the evidence is unfolding may well change or shift their emphasis as the case develops, just as cross-examination may alter course in the light of subsequent evidence, or evolving or changing instructions. The process of trial is not static, as it is on appeal. It is, therefore, very easy for fresh counsel at an appeal to question, or be critical of, a line of defence that seems not have been sufficiently explored or emphasised during a trial, when viewed with the benefit of hindsight and in the light of all the evidence, the speeches of counsel, the judge’s summing-up and any questions from the jury. However, such criticism of counsel’s conduct at a particular point in the trial, or on a particular issue, can be woefully unrealistic and highly artificial, especially when considered in isolation or out of context.

57.      Secondly, there is, regrettably, a continuing tendency for fresh counsel on appeal to impugn the tactical decisions of trial counsel as ones they would not have made themselves, whilst at the same time not being prepared to characterise them as flagrantly incompetent.  The familiar cry by appellate counsel of an ‘unfair trial’, as seen through the prism of the appeal papers, is believed to trump all other considerations of competence and tactic on the part of trial counsel.  Yet, how trial counsel sees the issues and the best way to present his case in dealing with them is a vital part of the function of professional counsel in an evolving trial and should not be lightly disregarded …” 

149.Consulting trial counsel would have been an important step before advancing Ground 5, in which Mr Grounds questioned before us Mr McCoy’s decision to leave the passages in the first video recorded interview of the questions and answers between DSgt 16404 and the applicant as to why he did not tell the officer about the missing yoga ball, or the fact that it was full of carbon monoxide, at their meeting at the applicant’s home a year earlier.  Mr Grounds has maintained that he would have asked for the passages of the interview in question to be redacted, although he initially accepted before us that they are not inadmissible.

150.Mr Grounds quite properly did speak to Mr McCoy about the matter.  Accordingly, he was able to inform us that Mr McCoy had made a specific decision that he did not want the impugned passages excised from the record of interview.  We will return to the significance of this matter when we deal with Ground 5 in due course.

151.The third reason that PD 4.2 must be properly complied with is that it will concentrate the minds of counsel for the applicant and the respondent, as well as the Court itself, on the real and significant issues which fall for consideration at the appeal.  The Court is not assisted by jury points masquerading as legal argument, or by a multitude of repetitious, unfocussed or plainly inconsequential grounds of appeal which fail to recognise how and why the trial was conducted in the way it was, or cannot sensibly affect the outcome.  We venture to suggest that many of the grounds of appeal in this case could not have been put forward by leading counsel at trial, who, despite his experience and stature as a criminal practitioner, made no complaint whatsoever about the summing‑up.  Moreover, it seems to us, with respect, that several of the grounds put forward by fresh counsel at the appeal do not reflect a proper appreciation and understanding of the real issues engaged at trial.

152.For these reasons, which this Court forecast in HKSAR v Chung Chun Fung and Another[120], we wish to state the position for all future appeals in murder cases. Hitherto, there has been a practice by which applications for leave to appeal against conviction for murder have been dealt with at the same time as the hearing of the appeal proper, rather than going through the Single Judge leave application process.  This routine practice will now cease and all appeals against conviction for murder, where issues of fact or mixed law and fact are concerned, must be dealt with in the same way as any other appeal requiring a separate leave application before a Single Judge, in accordance with the Practice Direction, unless the Single Judge orders otherwise.  Had that happened in the present case, most of the 20 grounds of appeal (in reality, some 40 grounds of appeal) would have been refused leave to appeal as not being reasonably or properly arguable, which ought to have resulted in the streamlining of the appeal as well as the focussing of counsel’s oral submissions.

153.We have been concerned for some time, and this case further illustrates our concern, with allegations made at appeals by fresh counsel on appeal, which too readily question the conduct of trial counsel: some go so far as to allege flagrant incompetence, others stop short and are merely critical of decisions taken by trial counsel, which appellate counsel maintain they would not have taken.  Again, these allegations are often made without a proper understanding of the real issues at trial and the tactics employed by trial counsel to deal with them.  We would respectfully endorse the recent comments of the Court of Appeal of New South Wales, presided over by Bathurst CJ, in Roach v The Queen[121]:

“It must be said that this application has the flavour of an ‘armchair appeal’, where counsel not involved in the trial have gone through the record of the trial in minute detail looking for error or possible arguments, without regard to the manner in which the trial was conducted: R v Fuge (2001) 123 A Crim R 310 at [40]-[45]; Darwiche v The Queen (2001) 209 A Crim R 424 at [170]. The additional step taken in the present case was the addition of a generalised claim of incompetent representation by trial counsel (ground 4) which was sought to be used, without success, as a type of broad antidote to the r 4 difficulty confronting the other grounds of appeal. Neither the use of ‘armchair appeals’ nor dubious claims of incompetent representation is to be encouraged by this Court.”

154.Although we are critical of the form and volume of the grounds of appeal filed in this case, many of which are plainly “armchair” grounds of appeal, we wish to make clear that such criticisms are by no means confined to this particular case; which is why we have considered it necessary to give clear guidance to the profession in conducting future appeal cases.  Moreover, such criticisms as we have made do not in any way affect the obligation of the Court to address the issues raised by this appeal, which were fully, comprehensively and, if we may say, articulately addressed by the applicant’s counsel in the course of a two-day appeal.  Although there were many grounds of appeal which were plainly not reasonably or properly arguable, there were some grounds which were.  The danger of fresh appellate counsel putting everything they can think of into a set of grounds of appeal, without the distillation which should come with appellate counsel’s expertise, is that the arguments sometimes “lose sight of the wood for the trees”, and that grounds which may have individual merit become lost in a morass of unmeritorious and inconsequential grounds of appeal and argument.  One of the more arguable grounds to emerge in this appeal was that concerning the alleged infringement on the applicant’s right of silence (Ground 5).  However, looking at the way the Amended Perfected Grounds of Appeal were drafted by counsel, and the short single paragraph it merited in the accompanying written submissions[122], one does not immediately appreciate its significance.  Indeed, at the end of argument concerning Ground 5, even Mr Grounds was moved to say of this particular ground, “it may be that the point is much, much bigger than even I had thought of … as one of the grounds”.  Given that it became such a prominent point of argument, we shall begin by addressing this matter in greater detail.

The applicant’s right of silence

155.The complaint in Ground 5 is that the Judge was wrong to remind the jury of the applicant’s failure to tell the police in his early dealings with them about the missing yoga ball containing carbon monoxide and his connection with it, without at the same time directing the jury not to draw adverse inferences from such failure, and without directing them that the applicant had a right of silence.  Thus put, it was not averred that the evidence was inadmissible per se: indeed, Mr Grounds, in answer to a specific question from the Court, originally accepted that the particular impugned answers in the first video recorded interview concerning the applicant’s failure to tell the police about the carbon monoxide inside the yoga ball were not inadmissible.  His complaint was that because the applicant had a right of silence, and had not therefore been obliged to tell DSgt 16404 anything about the missing yoga ball or its toxic contents when spoken to in May 2015, roughly one year before the interview, it was unfair and prejudicial to the applicant for the Judge to refer to this matter in her summing-up.

156.Before addressing this ground of appeal, we consider it necessary to set out the text of the questions and answers in the video recorded interview in question, which came towards the end of the interview.  In the transcript of interview, ‘D’ represents DSgt 16404, who was one of the two interviewers, while ‘B’ was the applicant.  DSgt 16404 asked first about the yoga ball[123]:

“D. On the 22 May 2015, that means, the day, er, when your wife and Lily met with mishap, I did go to your house. It was the first time we met. … Er, why didn’t you tell me at the time that a yoga ball had gone missing?

B. Because I was too scared to, uh, admit that, um, uh, I was responsible for putting the yoga ball in the fitness area.

……

D. In fact, when you – on the night of – 22 May or in the small hours of 23 May, at the time when you returned to 84 Tai Tung Village, that means, the – the place where your home (was situated), … in fact, is it the case that at the time you had already noticed the yoga ball had gone missing?

……

B. Yes.”

DSgt 16404 then moved to the question of why, at another meeting a few days later, the applicant had not mentioned that the yoga ball was filled with carbon monoxide[124]:

“D. When the incident happened, er, on 22 May 2015 (and) a few days or one week after the incident occurred – well, a few days or one week (after), er, the 22nd, … I did go to your home to meet you. … At that time it was raining heavily. I suppose it was a Sunday. … (I) told you clearly that your wife and daughter had passed away due to carbon monoxide poisoning. … Why did you not point out to me at the time, er, actually, it was you – the yoga ball you brought back had leaked the gas, causing the death of your wife and daughter?

B. Um, maybe because I was, first of all, too scared of facing the consequence of admitting that I was responsible for bringing the yoga ball back. … I knew that, if I had, uh, admitted it, I would have been in a lot of trouble … and I would, probably, also end up getting, uh, possibly, uh, Dick Chau (Chow), my colleague, or the laboratory technicians into some trouble. … I was also very, uh, worried that, if I was, uh, um, arrested, then I would not be able to look after my children. … Um, I, I feel very ashamed of my coward(sic) behaviour … but I was, uh, I was trying to hide when a convenient excuse that the carbon monoxide came out of the car was suggested by many people.”

157.In their closing speeches to the jury, neither Mr Bruce for the prosecution, nor Mr McCoy for the defence, said anything at all about these exchanges; although Mr McCoy did make the comment that “the police hadn’t worked it (the source of the carbon monoxide) out for six months until when they seized the yoga ball.  When did they seize the yoga ball?  22 May? No, November 2015, six months later.  There’s no one accounting for the ball in the police control for six months.  Who was looking in it during that six months?”[125]. The point he was making was that since the police did not realise that the yoga ball was the source of the carbon monoxide, its original position in the Mini had not been properly noted, nor had the ball been isolated or examined for some six months.  One obvious rejoinder to this point is that the reason the police took six months to work out where the carbon monoxide had come from was that the one person who knew the answer chose not to tell the police.

158.The context of the passage in the summing-up about which complaint is now made was a section of the Judge’s directions dealing with what the applicant had said to the police during his arrest and questioning.  Thus, the Judge began this part of her summing-up by explaining[126]:

“Now, I would like to remind you again what the defendant had said to the police after his arrest.”

She first dealt with the applicant’s response as recorded in DSgt 16404’s notebook, which we have recited above[127], before going through the salient features of what the applicant had said in the three video recorded interviews that followed.  In that process, she reminded the jury[128]:

“He was in the operating theatre when he learned from the wife’s good friend, that is Ms Foong, that something had happened to his wife and daughter, then he rushed to the Accident and Emergency Department of the Prince of Wales Hospital and he saw the medics trying to resuscitate them, the wife and the daughter, but in vain, not successful. He had no idea how the yoga ball ended up in the Mini Cooper.

He admitted he did not tell the police investigator that his wife and daughter, who died of carbon monoxide poisoning, were killed by the carbon monoxide from the yoga ball which he had filled with carbon monoxide.  He admitted being a coward and he let people to think that the carbon monoxide came from the engine of the car.  He did not tell the police because he did not want to get people like Dick Chow in trouble.  He also did not want his children to be without a parent should he be taken away. He said he did not notice if the yoga ball was still inside the house on 22 May 2015 after he got up; and he did not know how the yoga ball ended up in the Mini Cooper.”

159.Accordingly, leaving aside for the moment the legal issue of whether the applicant had in law a right of silence at his two meetings with DSgt 16404 at his home on the night of 22 May 2015 and a few days later, the Judge was in fact doing no more than summarising the applicant’s answers in his first video recorded interview, which commenced on 13 May 2016; answers which leading counsel for the defence at trial in fact wanted to be included, and must, therefore, have accepted were admissible.  Moreover, they formed part of an interview, during which the applicant not once declined to answer a single question.

160.Mr Grounds evidently disagrees with the course adopted by Mr McCoy.  However, with respect, he was not counsel at trial.  Mr McCoy had deftly painted the applicant as a brilliant[129], highly skilled[130] individual of positive good character[131], with an international reputation for medical excellence[132]; but also as someone who was weird in a strange way[133], nerd-like[134] and one‑dimensional[135], whose focus or concentration was abnormal[136] and tunnel-visioned[137], who was dogmatic[138] and only able to see one side of things[139], who took a narrow approach and saw things in his own way, and who lacked basic human empathy[140].  The failure to tell the police about the yoga ball or the carbon monoxide may well have corresponded with Mr McCoy’s depiction of the applicant, whose decision to bring the yoga ball home in the first place he characterised to the jury as “crazy”[141].  Moreover, we can well understand why Mr McCoy might have wanted the applicant’s answers before the jury.  The explanations for not telling DSgt 16404 about the yoga ball or the carbon monoxide were not necessarily implausible or incomprehensible on a human level, given the position the applicant found himself in and the exigencies of the moment.  Had the applicant felt guilty about introducing the yoga ball with its deadly contents into his home, or “morally responsible” as Mr McCoy candidly put it to the jury[142], he may well have realised the seriousness of his position and chosen to hide behind an investigation of the Mini as the source of the carbon monoxide, rather than imperilling himself and his colleagues and, more particularly, his young family, who would have been without a parent if he had been arrested.  His answers to DSgt 16404 did not mean that he was the one who placed the yoga ball in the Mini.

161.Ultimately, even assuming that a right of silence attached to the applicant in May 2015, we think it very unlikely that Mr McCoy, as an experienced tactician, would have preferred the jury to be told that they should not hold the applicant’s failure to assist the investigation into the deaths of his wife and daughter by withholding relevant information against him in any way, rather than be deprived of the applicant’s perhaps ill-judged but nevertheless understandable explanation as to why he did not assist the police in the circumstances; an explanation which may have tied in with the type of person he was.

162.However, as Mr Grounds developed his argument in reply, it appeared to metamorphose into one of intrinsic inadmissibility; either on the basis that the questions and answers were more prejudicial than probative, or because the applicant had a right of silence in May 2015 and, therefore, his lack of assistance to the police about the disappearance of the yoga ball and its toxic contents meant that the evidence should never have been in front of the jury in the first place.  While, for the reasons we have already adverted to, we do not think the evidence was more prejudicial than probative, and Mr McCoy made no such submission to that effect, nor did he seek to have the evidence redacted from the first video recorded interview, it is nonetheless necessary to address the argument as to whether the applicant did in fact have a right of silence when spoken to by DSgt 16404 in May 2015.

163.It is common ground that in May 2015, the applicant had not been arrested; he had not been cautioned; he was not even a suspect.  If anything, he was a witness who might have been able to assist the police in piecing together what had happened to his wife and daughter.  When DSgt 16404 visited the applicant’s home on the night of 22 May 2015 and a few days later, he would have believed he was dealing with a grieving husband and father in respect of a tragedy which, at least on the face of it, neither the applicant nor the police could explain.  The applicant’s status at the time he was asked questions in May 2015 is vital to understanding whether he was someone to whom a right of silence attached and, if so, what the right entailed.

164.Mr Grounds submitted that no citizen is obliged to cooperate with, or assist, the police by saying anything when questioned at any stage of an enquiry or investigation.  He has a right to remain silent, and that right obtains whether he is a suspect or not.  Leaving aside the limited inroads which the legislature has made concerning the obligation on persons in certain circumstances to provide answers to questions posed by an investigating authority, we think Mr Grounds is correct as a matter of law.  The point has long been established by the venerable decision of Rice v Connolly[143], where Lord Parker LCJ explained[144]:

“It seems to me quite clear that though every citizen has a moral duty or, if you like, a social duty to assist the police, there is no legal duty to that effect, and indeed the whole basis of the common law is the right of the individual to refuse to answer questions put to him by persons in authority, and to refuse to accompany those in authority to any particular place; short, of course, of arrest.”

165.However, as Lord Parker CJ went on to make clear, the right of silence is exactly that: it is not a right to lie or mislead[145]:

“In my judgment there is all the difference in the world between deliberately telling a false story – something which on no view a citizen has a right to do – and preserving silence or refusing to answer – something which he has every right to do. Accordingly, in my judgment, looked at in that perfectly general way, it was not shown that the refusal of the defendant to answer the questions or to accompany the police officer in the first instance to the police box was an obstruction without lawful excuse.”

166.In the Scottish case of Boyes v MacLeod[146], presided over by Lord Rodger, the Lord Justice General (later Lord Rodger of Earlsferry SCJ), two accused[147] were passengers in a motor car together with a woman when she was injured by a missile thrown into the car in which they were travelling.  Accordingly, the accused drove to a police station to make a complaint and were both interviewed about the incident, without being cautioned.  Subsequently, a complaint was made against the accused that before the missile was thrown into their car, they had thrown a missile (or missiles) from their car at a group of pedestrians in the street.  The accused objected to the proceedings arising out of that subsequent complaint on the ground that they were oppressive because the accused should have been cautioned in connection with the throwing of missiles from their car at the group of pedestrians.  However, they had not been cautioned and had thereby disclosed their defence to the police.  On appeal, the Court held[148]:

“We take from the justice’s report that she took the view that at the relevant time the appellants were being interviewed in the context of an investigation of the matter relating to Tina Mohammed and, more widely, in the context of an investigation of what had been going on that evening. But we also take from it that her conclusion is that the appellants were not suspects at that time when they were interviewed. In these circumstances, it cannot be said that there was any requirement on the part of the police to caution the appellants when they interviewed them. In his submission counsel homed in on a particular question which his client Jonathan Boyes had been asked in relation to an allegation that the occupants had thrown a bottle from the car. In reply to that question his client had said that he had dropped the bottle out of the car window but had not thrown it. But in our view, given that the police officers did not suspect the appellant at that stage, there was no need to caution the appellant before putting the matter to him.

In the circumstances, we consider that all that one has here is the situation where certain replies were given and certain things were said in the course of an investigation where these appellants were witnesses rather than suspects.  For that reason we take the view that there was nothing improper in what the police did.”

167.Boyes v MacLeod is an illustration of the point that there is nothing improper in a police officer asking questions of someone who is not a suspect, and who has decided of his own free will to speak. Accordingly, what an un-cautioned non-suspect says to the police in answer to an enquiry is not inadmissible per se.  Of course, where it is alleged that the police have tricked someone into believing he was not a suspect so as to obtain information or evidence or an admission from him, when he was in reality a suspect, the information or evidence thus obtained might be objectionable but would depend on the issue, properly engaged in evidence, as to whether that person was a suspect and should have been cautioned.  The key is whether or not the person was, at the time of the answer, a suspect, thus warranting his being cautioned.

168.The position of the appellant in the leading authority in this jurisdiction of Lee Fuk Hing v HKSAR [149]was very different.  There, the appellant was intercepted while driving a light goods vehicle along a road in the New Territories.  As soon as the police revealed who they were, the appellant leapt out of the vehicle and tried to escape.  However, he was caught and held by the police.  When his vehicle was searched, two bags of heroin were found.  He was duly arrested for trafficking in a dangerous drug.  At his trial, the appellant gave evidence that the heroin must have been left behind in his vehicle by one Tsang Kei, to whom he had lent his vehicle, and from which he claimed Tsang Kei had alighted shortly before the appellant was intercepted by police.  The Judge asked the appellant whether he had ever mentioned the existence of Tsang Kei to the police before and if not, why not.

169.Of these questions, the Court of Final Appeal held[150]:

“There is a right to silence. And in consequence of this right, suspected persons in Hong Kong should be – and routinely are – given a caution informing them in unqualified terms that they need not speak. In the whole of these circumstances, it is inappropriate in Hong Kong to use a person’s silence against him in any way.” (Emphasis added)

The Court went on to explain that[151]:

“A person’s right against self-incrimination (his right of silence) would otherwise become a possible source of entrapment. It is unfair for a person to have the right to remain silent, and usually to have been reminded of this right through the caution, and then for his silence to be put against him at trial.”

170.It is clear that the Court of Final Appeal was dealing with someone who had been arrested but failed to give an innocent explanation at the time, but which explanation was later relied upon at trial. Indeed, from the moment the appellant jumped out of his vehicle in order to escape, he would no doubt have been suspected of committing a criminal offence.  In the case before us, it was not until November 2015 (six months after the conversation between DSgt 16404 and the applicant) that the police made any connection at all between the yoga ball and the deaths of the deceased six months earlier.

171.In his final reply, Mr Grounds produced the further judgment of the Court of Final Appeal in HKSAR v Ata Asaf[152] in support of his contention that everyone, whether suspected of a criminal offence or not, has a right of silence and cannot be criticised for refusing to assist the police in their enquiries.  Thus, he contended that even a potential witness has the right to remain silent when asked a question by the police, which cannot later be the subject of any adverse comment.  Again, it is important to bear in mind the facts of the case in Ata Asaf and the stage at which comment was made about that appellant’s failure to give an explanation.

172.The appellant had been arrested on a street in Kowloon for the offence of theft and subsequently, upon a search of his person, for the offence of possession of a dangerous drug, after which he was cautioned.  Having been cautioned, the appellant declined to say anything. Given the quantity of narcotic concerned, he was later charged with trafficking in the drugs seized from him in the street.  The appellant offered to plead guilty to simple possession of the drugs in question, but this offer was refused by the prosecution.  At his trial, during cross-examination, prosecuting counsel asked the appellant why, at a subsequent search of his premises following his arrest and caution, he had not mentioned to the police that his alleged drug consuming apparatus for smoking “Ice” had not been pointed out to the officers conducting the search[153].  He responded that he had not been asked by the officer concerned.  The Court found that such question and answer were inadmissible.  It also held that a further point made in prosecuting counsel’s speech that the appellant had “for the first time” in evidence claimed he had gone to work that morning with $7,000 on his person was improper[154].

173.In its judgment, the Court held that there were two ways in which a person’s right of silence could be infringed[155]:

“The first is where his silence is used to infer guilt of the particular offence of which he is suspected or charged. This may occur, for example, where a suspect is caught red-handed and, when questioned as to what he is doing, declines to answer. His silence in these circumstances may be said to give rise to an inference of guilt.

The second way that a person’s right to silence may be infringed is where it is suggested that a line of defence or evidence which is helpful to the suspect or accused is a late invention because it was not mentioned earlier. This suggestion of recent invention may be said to undermine the credibility of the line of defence or evidence tendered.” (Emphasis added)

The Court went on to note that it was common ground that the offending evidence in the appellant’s premises about his drug-taking equipment was inadmissible[156]:

“because it clearly infringed the appellant’s right to silence in the second way referred to above. That right, being an aspect of the right against self-incrimination, is a fundamental right of a person suspected of or charged with a criminal offence as recognised in this court’s decision in Lee Fuk Hing v HKSAR.” (Emphasis added)

174.We pause to note the words we have italicised in the above two passages from Ata Asaf and Lee Fuk Hing.  They were adopted for good reason.  If the right of silence is a privilege against self‑incrimination and is not to become a source of entrapment, then it must attach to those suspected of or charged with a criminal offence in the first place.  This mirrors Specimen Direction 45 used in Hong Kong in respect of a defendant’s right of silence:

“Any person suspected of a criminal offence or charged with one is entitled to say nothing when asked questions about it. You must not hold his [silence] [refusal to answer questions] against him. The exercise of the right of silence cannot amount to an admission of any kind nor can it be taken to reflect a guilty conscience.” (Emphasis added)

175.If the right of silence were intended to be applied to anyone who may be connected with an investigation but who is not himself suspected of complicity in the offence, or indeed of any wrong-doing, then we wonder how the ‘caution’ should be framed.  It would involve the police officer saying to someone who is not a suspect but who is being asked to help the police with their enquiries as to what has happened, “I would like to ask you some questions.  You are not suspected of involvement in this or any other offence, but nevertheless you have the right not to help me with my enquiries”.  We suggest this would be an unwarranted, unnecessary and somewhat absurd, development of the common law.  Of course, under normal circumstances, a witness can still decline to assist police enquiries without penalty or compulsion.  However, if he volunteers information as a witness, before becoming a suspect, we do not see why comment cannot be made about what he has volunteered.  Prosecution witnesses are often cross-examined by the defence on the basis of what they said or did not say to the police; and so are defence witnesses, by the prosecution.

176.The fact is, however, that no comment was made by the Judge in this case.  All she did was to recite what the appellant had said in his video recorded interview, which leading counsel for the defence actually wanted to go before the jury.  Nor was it being suggested by counsel on either side, or by the Judge herself, that the applicant had failed to put forward an innocent explanation when given an opportunity to do so.  All he had refrained from doing was helping the police discover that the yoga ball had contained the carbon monoxide which killed his wife and his daughter, a yoga ball which he throughout denied ever placing in the Mini.

177.With respect, we do not see that the statements of principle in Lee Fuk Hing and Ata Asaf have any relevance to the circumstances of the present case.  The questions of DSgt 16404, and the answers which they elicited, were not inadmissible; the Judge merely referred to them in the context of her review of the video records of interview without comment; and, in any event, the defence wanted the impugned questions and answers before the jury.  As Mr Bruce has correctly pointed out, the applicant was represented by both counsel and solicitor during each of his video recorded interviews and “in the run-up to trial and during the trial by lawyers of the highest calibre”; moreover, “there was never any issue on the voluntariness of taking of the video recorded interviews or their admission as evidence at trial”[157].

178.In the course of our consideration of this ground of appeal, it seemed to us to be relevant whether or not the applicant had in fact ever declined to assist the police with their enquiries on the night of 22 May 2015 and, again, a few days later.  For, if he had chosen to say nothing, or signified his wish not to answer or assist with police enquiries, then at least the factual premise of the applicant’s contended-for right of silence might be established.  Without knowing the position, we were in effect being asked to assume an exercise by the applicant of a right of silence in May 2015 when questioned DSgt 16404, which had thus been infringed by the same officer’s questions a year later, when in reality the applicant had never purported to exercise any such right in the first place.  We had been informed that Mr McCoy had, for tactical reasons, wished the impugned questions and answers to be included in the video recorded interview which was placed before the jury.  It seemed to us relevant to know why counsel might have taken that view, which would depend upon many factors, including his instructions and also, perhaps, the statements or depositions of the officer making the enquiries, which would have shaped the way Mr McCoy dealt with this issue.  If, for example, DSgt 16404 had asked questions of the applicant and received answers, at a time when the applicant was not a suspect but was being treated as a witness, the applicant could hardly be said to be availing himself of a right of silence, assuming one even attached to him at that stage.  On the other hand, if the applicant said nothing or declined to answer any questions or assist the police, Mr Grounds’ argument that the applicant was entitled to say nothing at that stage, even as a witness, might at least have some foundation.  As it was, we were being asked to assume such a foundation, for which the Judge’s subsequent directions amounted to a material irregularity.

179.Accordingly, we wrote to the parties on 28 January 2021 to invite written submissions as to whether the Court could or should look at the statement(s)/deposition(s) of DSgt 16404, which were filed with the trial court at committal in order to better understand the factual and contextual basis for Mr McCoy’s tactical decision.  Both parties responded in writing.  The respondent had no objection to that course and repeated that there was never any issue as to voluntariness or admissibility of the video recorded interviews: the defence submitted it was not appropriate for the Court to examine the material now, and whether or not the impugned questions and answers were admitted, it was still incumbent on the Judge to inform the jury that the applicant had a right of silence in May 2015.  Having considered the respective submissions, we concluded that it was relevant to establish whether or not the applicant had declined to answer questions, from which later questions in interview may have been a derogation of any right of silence; or whether he had answered questions and appeared to cooperate with police enquiries, from which later questions would not be a derogation of any such right.  Such information would also have been very relevant to Mr McCoy’s decision as to how to deal with the impugned passages in the video record of interview.

180.DSgt 16404 made two witness statements, one on 1 June 2016, the other on 2 October 2017, both of which were served on the defence in the committal bundle and, therefore, also on the court of trial. There were three sections of the statement of 1 June 2016, which were relevant to the present consideration, namely:

“4. At 0020 hours on 2015-05-23, witnessed by the head of household Khaw Kim-sun, 11 items were seized for examination. No information was provided (by Khaw Kim-sun). The detective team left at 0025 hours the same day.

……

5. At 1700 hours on 2015-05-24, (I) paid a visit to Khaw Kim-sun at his home at G/F, No 84 Tai Tung Village and asked if he recalled anything else or had any clues but Khaw Kim-sun did not have any additional information or clues to provide.

6. At 1855 hours on 2015-07-13, (I) arrived at G/F, No 84 Tai Tung Village to get a copy of (the deceased’s) insurance policy and the other key to the private car GT 1162 from Khaw Kim-sun for the investigation of this case.  The car key was then sent to the manufacturer for examination of the chip record, which showed that the key provided by Khaw Kim-sun had been used before the incident.”

181.The excerpts from the above witness statement, in conjunction with DSgt 16404’s evidence and, in particular, his cross-examination, demonstrate that at no stage of the trial was there any issue whatsoever as to the voluntariness or admissibility of what the applicant told the police qua suspect in his video recorded interviews; nor was there any hint of the applicant purporting to exercise a right of silence qua suspect during the video recorded interviews or, indeed, qua witness during earlier police enquiries.  We do not agree with Mr Grounds’ proposition that in order “to ensure a fair trial for the applicant, on the evidence available to the jury, given the emphasis placed on the failure of the applicant to inform the police of his earlier actions, the jury needed to be directed that he was under no obligation to volunteer information to the police and that no adverse inferences should be drawn”[158].  As we have said, no emphasis was in fact placed on this matter by the Judge; but, in any event, there was no duty on her to direct the jury that the applicant had a right of silence qua witness in May 2015 when this was never an issue at trial, and when leading counsel actually wanted all of his answers in the video recorded interviews in May 2016 to go before the jury.  When fully analysed, there is no merit in Ground 5.

182.We wish to make clear that we would have arrived at this conclusion even without recourse to the deposition of DSgt 16404.  As we have said, we were not prepared to make an assumption, in the absence of any evidence or even suggestion, that the applicant had exercised a right of silence in May 2015 when he was a mere witness as to the unexplained deaths of his wife and daughter.  Our examination of DSgt 16404’s statement (which the Judge herself would have seen, since it was among the depositions sent to the High Court) merely reinforces the fact that there never was a right of silence exercised, in respect of a matter about which there was never any issue taken by the applicant’s leading counsel at trial.

The evidential directions

183.We now turn to Ground 1 of the Amended Perfected Grounds of Appeal, which concerns pure matters of law, although sub-paragraphs (a) and (b) in fact engage different points of law.  Ground 1(a) effectively takes issue with the correctness, as a matter of law, of Specimen Direction 21 in respect of circumstantial evidence, since it is said to impose a similar burden and standard of proof on the defence to provide “good evidence”[159], or evidence “of sufficient reliability and strength to cast doubt upon or destroy the prosecution case”[160]. A further complaint concerns the direction given at the outset of her summing-up, which is taken from Specimen Direction 1 dealing with the respective functions of the Judge and the jury, that the jury were to form their “own judgment about the witnesses on which evidence is reliable and which is not”[161]. It is said that this was also a misdirection; for it gave rise to the risk of the jury rejecting defence evidence which might be true.

184.In making this argument, Mr Grounds faced an obvious hurdle, inasmuch as the directions about which he complained are in fact standard Specimen Directions, which have been approved by the Hong Kong Judicial Institute and applied in countless cases where the directions on circumstantial evidence are required for very many years.  Nonetheless, Mr Grounds is entitled to take the point and we have examined it carefully to see whether the criticisms of the relevant Specimen Directions are valid.

185.We may deal first with the complaint about that part of Specimen Direction 1 which directs the jury to make their own judgment about the reliability of the evidence of witnesses.  With respect, the context of Specimen Direction 1 is the opening direction of a summing-up concerning the respective functions of the Judge and the jury.  It tells the jury that they must look at all the evidence in order to determine what evidence they are going to rely upon in deciding whether the charge laid against the defendant has been proved.  The argument that the jury might from this general direction reject unreliable evidence that may in fact be true is specious semantics, which ignores the other later directions that are routinely given to juries, and were given to the jury in this case: that the jury had to be sure of guilt beyond a reasonable doubt[162]; that if there was a reasonable doubt arising from the defence case put forward, which included his own account in interview and the evidence of witnesses, he must be found not guilty[163]; and that even if the defence case were to be rejected, the jury still had to be sure that the prosecution had established the guilt of the defendant[164].

186.The direction on circumstantial evidence, about which complaint is then made, has been approved by the Hong Kong Judicial Institute and is derived from same direction endorsed by the Judicial Studies Board used in England and Wales.  In R v P[165], Thomas LJ (later Lord Thomas CJ), giving the judgment of the Court of Appeal of England and Wales, held:

“Thirdly, it is clear, in our view, that the direction which the Judicial Studies Board has advised should be given in appropriate cases weaves together the various approaches in the authorities which are set out in Blackstones, but also in Cross on Evidence (11th ed (2007), pp 31 and following). The standard direction is this:

‘Circumstantial evidence can be powerful evidence, but it is important that you examine it with care and consider whether the evidence upon which the prosecution relies in proof of its case is reliable and whether it does prove guilt.  Furthermore, before convicting on circumstantial evidence you should consider whether it reveals any other circumstances which are or may be of sufficient reliability and strength to weaken or destroy the prosecution case.  Finally, you should be careful to distinguish between arriving at conclusions based on reliable circumstantial evidence and mere speculation.  Speculating in a case amounts to no more than guessing or making up theories without good evidence to support them; and neither the prosecution nor the defence nor you should do that.’”

187.The direction approved by Thomas LJ is virtually indistinguishable from the direction given by the Judge in this case, which is itself taken verbatim from Specimen Direction 21 approved for use by the Hong Kong Judicial Institute.  With respect to Mr Grounds’ argument, it does not make sense to talk of circumstantial evidence being “truthful” as distinct from “reliable”, since circumstances are, by their very nature, indirect or objective conditions or external events, which are not usually dependant on their internal veracity or the truthfulness of a witness’s testimony.  The question will normally be whether such circumstantial evidence can be relied upon in proof of guilt and whether it does in fact prove guilt; or whether, on the other hand, it undermines the prosecution case.  And when the Specimen Direction speaks of such evidence revealing any other circumstances, which are or may be of sufficient reliability and strength to cast doubt upon the prosecution case, it is applying the concept of reasonable doubt to circumstantial evidence.  We do not agree that the Judge’s directions wrongly assimilated the prosecution and defence standards of proof.  On the contrary, they emphasise the importance and necessity for the jury to consider whether such evidence does in fact prove guilt.

188.There is no merit in this ground of appeal.

The directions on the expert evidence

189.We will deal with the complaints about the Judge’s failure to summarise the defence case and the expert evidence together as they overlap considerably and seek mainly to reargue or argue afresh scientific and technical issues in answer to the allegations against the applicant.

190.The essence of the applicant’s submission is that the Judge failed to properly summarise the expert evidence and hence did not fully present the defence case to the jury[166].  It is generally complained by the applicant that the Judge failed adequately to summarise the expert evidence, “doing so variously in insufficient or excessive detail, without proper balance and with various errors.”  The specifics of this ground are contained in the grounds in respect of the applicant’s research and the timeline defence.  It is argued that the Judge wrongly directed the jury that they had to choose between the conflicting evidence of the experts[167] and did not properly address the defence expert evidence in respect of the timeline defence (the dissipation rate of carbon monoxide)[168] and the genuineness of the applicant’s research[169].

191.Of the failure to summarise the defence case, the applicant further complained about the Judge’s summary of evidence in relation to the reinjection of blood into the rabbits[170]; and the issue of when the stopper of the yoga ball went missing[171].

192.The Judge, having reminded the jury in some detail that it was their responsibility to Judge the evidence, went on to tell them to consider the closing speeches of counsel and whatever they thought was important in the evidence, even though she may not herself have mentioned it[172]:

“Now, you have all sworn either an oath or affirm that you will deliver a true verdict according to the evidence. And therefore, you must decide the case only on the evidence which they had been placed before you. There will be no more. You must not speculate about what evidence that might have been and allow yourself to be drawn into speculation. The facts of this case are your responsibility. You will wish to take account of the arguments or submissions in the speeches you heard on Thursday and Friday from the prosecution and defence, but you are not bound to accept them. You must remember that. That is just their submission, their argument. You are not bound to accept them.

Equally, if in the course of my review of the evidence I appear to express my views concerning the facts or emphasise a particular aspect of the evidence, do not accept those views unless you agree with them. In other words, you do not agree with me you can of course, by all means, because you are the ones who have to decide on the facts, not me. You must remember you and you alone will decide on the facts, all right?

So and if I, in my review of the evidence to you, did not mention something which you think is important, still consider those points because you must realise it is not possible for me to repeat everything that had happened in this court, right?  So even though if I do not mention it, you think it is important, you must also consider it and give the evidence such weight you think fit, how much weight you put on a piece of evidence whatever you think fit.  When it comes to the facts of this case it is your judgment alone that counts.”

193.It is complained that the Judge told the jury they had to choose between the conflicting evidence of the experts, and therefore excluded the possibility of them not reaching a firm conclusion on the relevant issue(s)[173]. In our view, the Judge did not give any such direction.  She very carefully, and correctly, explained how the jury should approach the expert evidence without limiting how they were to consider it but noting that there were conflicting views between the experts which they had to consider.  She said[174]:

“A witness called as an expert is entitled to express an opinion in respect of his findings or matters which are put to him and you are entitled and would no doubt wish to have regard to this evidence and to the opinion expressed by the expert when coming to your own conclusion about this aspect of the case. You should bear in mind that if, after having given the matter careful consideration you do not accept the evidence of the expert, you do not have to act upon it. Just because the evidence comes from the expert does not mean that you must accept it, no. You consider it and you think that, ‘No, I am not going to accept any particular expert’s evidence’, you are perfectly entitled to do that.

Now, you are, of course, aware there are conflicting views expressed by some experts for the prosecution and for the defence. So where two or more experts have conflicting views, it is for you to decide are you accepting either one or neither, up to you. You decide whose evidence you accept, whose evidence you do not. You must remember that the expert evidence only relates to part of the case not the whole case, so you must consider the expert evidence but also consider all the evidence and to see whether the expert evidence assists you in your final conclusion reaching your verdict after considering all the evidence.” (Emphasis added)

194.We observe that the Judge by her remarks above followed Specimen Direction 32 in respect of expert evidence.  It is important to bear in mind that in the context of a criminal trial, experts do not decide cases, juries do.  Expert evidence is admitted only in respect of matters that lie beyond the common experience and knowledge of the jury, and in order to assist them to understand scientific and complex facts at issue during the trial: R v Turner[175]. Often experts will differ in their opinions and it is for the jury to decide between the experts, by reference to all the available evidence and what is in issue: R v Kai-Whitewind[176]; R v Brennan[177].

195.The Judge presented to the jury a very comprehensive summary of the expert evidence, noting the issues which the experts sought to address as well as the issues that arose between them.  When appropriate and relevant, she dealt with the expert evidence in the context of the evidence of the witnesses and related exhibits.

196.It might be stating the obvious, but a summing-up is not meant to be a full recitation of all the evidence, nor a complete delineation of all the issues.  It is to provide a clear and precise overview of the salient facts and circumstances of the case, as well as addressing the main issues and matters that warrant consideration by the jury, with appropriate directions or guidance as the law requires in determining the count or counts on which they are to return a verdict.

197.The expert evidence at trial focused primarily on two issues.  The first was the applicant’s research in respect of carbon monoxide poisoning and whether it was a genuine research project.  The second was the leakage rate of carbon monoxide from the Mini and the deflation rate of carbon monoxide from the yoga ball, which was relevant to the timeline defence. These two issues were crystallised for the jury by the Judge’s early remarks to them[178]:

“So the main issue you have to decide is who had put the yoga ball filled with carbon monoxide inside the Mini Cooper and why, who and why? Was it the defendant or was it someone else other than the defendant who had put the yoga ball there and why? Now, in deciding this main issue of who and why, there are sub-issues - I will call them sub-issues - that you have also to consider and that is why did the defendant fill two yoga balls with carbon monoxide and took them home? Was his purpose of doing so to kill rats?

Was the research experiment involving the acquisition and the use of carbon monoxide a sham or a genuine research?  And who had the opportunity of putting the yoga ball inside the Mini Cooper between the time the defendant brought the yoga ball back home and the time mother and daughter were found unconscious inside the car?  These are all issues that you have to consider and decide.  So, at the end of the day, can the prosecution prove that it was the defendant who had murdered his wife and his daughter?  That is just a general introduction.”

198.The Judge then explained how she would address the expert evidence in respect of the two issues between the parties.  She said[179]:

“Now, I will now move on to remind you of the evidence about the proposed research, the actual experiments done, that is, about the protocols, the tests done with the yoga ball and the Mini Cooper and expert evidence that you have heard and so on.

Now, you have heard a lot of expert evidence on two areas. The first area is about carbon monoxide poisoning, about the effect of carbon monoxide on the human body and the treatment administered to people suffering from carbon monoxide poisoning. That is the first part. The second area is the tests done to determine leakage rate from the yellow Mini Cooper and another blue Mini Cooper, experiments conducted with the yoga ball in question, P17, and other yoga ball, which is not the yoga ball in question, and the dissipation rate of carbon monoxide from yoga ball and the concentration of carbon monoxide in the cabin of the Mini Cooper. So I call these item 1 and item 2, two areas.

So in item 1, that is basically about the effect of carbon monoxide, carbon monoxide poisoning and so on.  You have Professor Kwong, Kwong Yok-lam, Professor Gavin Joynt, Professor Michael Irwin, for the prosecution, three experts.  And you have Professor Nicholas Buckley and Professor Wang Chi-chiu for the defence.  For item 2, that is the test with the yoga ball and the Mini Cooper, you have Dr Wong Koon-hung, for the prosecution and Professor Li Chi-kwong for the defence.”

199.She went on to explain to the jury that the expert evidence was before them[180]:

“… as part of the evidence as a whole to assist you with regard to these particular aspects of the evidence; namely, for item 1, whether the proposed research and the experiments conducted by the defendant with the use of carbon monoxide on rabbits is a genuine research or just a cover up, a sham to obtain carbon monoxide; for Item 2, what was the dissipation rate of carbon monoxide leaked from the yoga ball and the concentration level reached inside the cabin of the car at various stages of time lapse.”

The applicant’s research

200.Mr Grounds complained that the Judge failed to remind the jury of the defence case that the applicant’s research was genuine for a myriad of reasons that mainly consist of comment or submission on the case for and against the genuineness of the applicant’s research[181].  Some of these points were made by leading counsel for the applicant in his closing speech to the jury[182], while others are new and selective.  The Judge in her summing-up emphasised the respective cases advanced by the prosecution and defence at trial[183].  The jury had the benefit of seeing and hearing the witnesses and receiving detailed closing speeches from the parties.  They would have been well aware of the arguments on the issue of the genuineness of the applicant’s research.  As a result of the numerous criticisms that have been made on this subject by the applicant, we need to address them by extensive reference to the evidence, the arguments at trial and the summing-up.

201.We will begin our consideration of the applicant’s research by quoting the Judge’s summary of the evidence of the applicant’s initial discussions with Professor Qin concerning his proposed study and the arrangements that followed[184]:

“… You heard from Professor Qin. He is a professor of the Chinese U and he worked at the Orthopaedics and Traumatology Department, you know, the bones. He plays tennis and it was on one occasion about December 2014, when he was playing tennis with the defendant, that the defendant mentioned about wanting to use animal for research. So Professor Qin asked his underling, Dr Dick Chow, to take over; so basically, Dick Chow was the first person who liaised with the defendant about the experiment.

You also heard that Dick Chow was the one who helped the defendant to order the carbon monoxide from this company called Hong Kong Oxygen, but the defendant had already obtained a quotation from Hong Kong Oxygen in October 2014 before he talked to Professor Qin. So when the defendant gave the quotation to Dr Chow, Dr Chow followed up and found that the price had changed. There had to be a new quotation. You remember all that evidence.

You also heard that there were procedures about application for licence because animals involved and so you need the licence from the government to allow you to do the experiment and also need the approval of the Ethics Committee of the Chinese U itself. So Dick Chow told you how he forwarded the forms, telling the defendant, ‘You have to fill this and fill that’ and so on.

Then the protocols, the draft, came to him, the first one of 113A and then he made some comments and sent it back. That is why in 113B there were all these suggestions or ways of improving, however you call it, on two. So basically that is his involvement. He did not know what else happened. He expected the defendant would have tidied it up and sent it onto the relevant department authority for approval.

The chain of how Professor Qin was approached and then told the defendant to contact Dick Chow and tell Dick Chow what happened, you can actually see the link in the emails. It is 127 and 124G. Just reading them you can see backwards and forwards and basically Professor Qin was not even interested in attending, I suppose one can call it. He was never there, although Dick Chow said, ‘The experiment is going to take place now, or this is going to take place now. Are you coming or not?’ Professor Qin, his only involvement was he was told about the animal research experiment plan and he just put it onto Dick Chow.

The defendant himself, in the video-recorded interview, I did not mention this part but I will now remind you, he admitted that those proposal forms were never submitted to the department.  When he was shown, remember during the interview the police officer showed him, right, ‘This is the protocol.  Is this the...’ so and so, so and so, he admitted that he did not submit them to his department.”

202.The Judge went on to deal with the study protocols for the research[185]:

“You also heard from Professor Gavin Joynt, the head of the defendant’s own department. His evidence was that he first saw these documents when the police showed it to him. He had not received them. He had not seen them. So his evidence again was not challenged, quite apart from the defendant’s own admission during the video-recorded interview that he had not. So, basically, those proposals had never gone any further to whoever is in authority to say, ‘Yes, I’m going to pass this’ or ‘I’m not going to pass it’, all right?

So he told you that in a normal procedure, if a proposal comes to him as the head of the department he will vet it and he will see whether it will pass him, right? Basically, if there was funds involved, personnel involved, so whether there is any sort of advantage by this proposal and if he vets it okay then he will pass it onto the Ethics Committee to see if they would agree that animals can be used, and so on and so forth. So I told you he had not received any of these, so these drafts, the application for the Ethics Committee, that is 130F, P130F, again, that form had not been passed onto any Ethics Committee.

Now, Professor Joynt also commented on the protocol but I will leave that later when I deal with all the experts, what they say about the protocol itself.  I just want to remind you at this particular stage that he, as the head of the department, had not seen these protocols.  Now, we come to the experiment.”

203.The Judge next addressed the evidence in respect of the two experiments on 13 and 20 May 2015[186]:

“First experiment was on 13 May. According to Dick Chow, apart from the defendant, this other person called Perpetua Tan who had not come to give evidence but who was described in the Admitted Facts, she was there. She was the person doing the oximeter to do the measurement of the blood. Shara Lee was not there at the beginning, only came quite late when the experiment is almost finished.

Now, Dick Chow told you that the carbon monoxide was inside. You can see from the photograph the big red tank was there and there was a regulator on top to, you know, regulate how much CO1, carbon monoxide was coming out. They were pure, 99.9 per cent carbon monoxide. The regulator was a safety requirement. You must get that, so remember when they purchased the carbon monoxide, they had to buy the regulator as well. There were also monitors all around for safety reason because the carbon monoxide itself is odourless and tasteless. You would not even know it is filling the room, so that is what he told you.

Also, the experiment had to be done in what is called a fume cupboard. Again, you can see from the photograph it is something like with a hood and because it had ventilation, so the idea is that if there is any leakage from the carbon monoxide, it will be sucked out of the room not to endanger those inside the room.

So he said that there were already two monitors, carbon monoxide monitors inside the lab but the defendant brought along eight, so altogether there were 10 for this particular experiment. Dr Chow told you that they wanted to do a trial run like a rehearsal, not the actual test itself and they want to see if they can draw blood from the animal, from the rabbit, without hurting the animal. So Dr Chow anaesthetised the rabbit, put the rabbit to sleep, and he used what he called a butterfly needle. He said the shape is like a butterfly, but there was some difficulty in drawing the blood out so in the end only 10 ml, millilitre, was extracted.

After the blood was extracted, the defendant was the one who tried to saturate the rabbit blood with carbon monoxide. At first, he said the carbon monoxide was blown into a tube of blood but it was not successful. There was a lot of wastage, according to Dr Chow, so then they decided to try with the blood still in the syringe. Again, the defendant did it directly into a syringe and he found that it worked better. Then the carbon monoxide concentration in the blood was tested by Perpetua Tan and another technician, of course you now know that is Miss Tam, with the oximeter. After some tests, then they worked out the best way to put carbon monoxide into the blood. That is all he told you on 13 May.

On 20 May, that is the next rehearsal, again there was a rabbit and the rabbit was anaesthetised by him. On this occasion a catheter was used instead of the butterfly needle. The butterfly needle was not very good so another gadget was used. Again, only 10 ml of blood was drawn. The carbon monoxide was blown directly onto the blood in the syringe to infuse the blood. The carbon monoxide infused blood was tested by the two technicians with the oximeter. Saline, that is salt water, was injected back into the rabbit to test the tube to make sure that it is clear. Carbon monoxide infused blood was not reinjected into that rabbit. So he was specifically asked about carbon monoxide infused blood, was it reinjected into the rabbit. He said no. The rehearsal finished around 4.30. Shara Lee came around 4. So that is the evidence of Dick Chow about these two experiments.

What did Miss Tam say, the lab assistant? That is the one that was read out to you under section 65B. She did not come personally. Miss Tam said she and her supervisor, that is, Perpetua Tan, that is Pat Tan, assisted the defendant and Dick Chow in the experiments on 13 May together, but on 20 May she alone assisted. Pat Tan was not there. According to her, on both occasions the procedure adopted was the same. The rabbit was given carbon monoxide to inhale before blood was drawn and then the blood sample was analysed by the use of the carbon monoxide oximeter. She handed the result to the defendant for his record and she kept the printout for reference and she had given a printout to the police upon request. Of course you now have the copies of the printout.

So you will notice that the description of what was done to the rabbit was quite different between Dr Chow and Miss Tam. Dr Chow said blood was taken out. The blood that was outside of the body of the rabbit was infused with carbon monoxide. No mention of the blood going back into the rabbit on the 13th. On the 20th positively not going back, the saline that had gone back into the rabbit, so what was being tested was blood with carbon monoxide but not from the rabbit on both occasions, if that is correct. But according to Miss Tam, there was not any situation of blood being taken out and then infused with carbon monoxide. The rabbit inhaled carbon monoxide so that the carbon monoxide got into the blood first and then the blood was tested. They are quite different, you would think.

Now, it is a matter for you, of course, but you may think that Dr Chow is the very person helping the defendant. He had looked at the protocol. What did the protocol say? The protocol says drawing blood out, infuse it with carbon monoxide, put it back in the rabbit to see the effect. So you may think that Dr Chow would know better that it was blood taken out rather than inhaling, whereas, Miss Tam, she was there to just use the oximeter to test the content of carbon monoxide in the blood. Maybe she did not pay as much attention. Of course, it is a matter for you but you do have these two very different descriptions.

Her evidence, of course, has not been challenged because she did not come to give evidence. So you have her saying, ‘Oh, the rabbit inhaled carbon monoxide on both occasions’, and you have Dr Chow saying blood was taken out, infused with carbon monoxide but not put back into the rabbit. That is what happened during the two experiments.” (Emphasis added)

204.The Judge addressed at length the separate issue of whether the applicant’s research was a sham.  This was the main focus of the applicant’s complaint about the Judge’s treatment of the expert evidence.  It is therefore necessary that we set it out in full.  One of the criticisms of the Judge’s summing-up was that she did not address the defence case in respect of the “extra haemoglobin effect”.  As revealed from the following passages, she did in fact deal with this issue and appropriately reminded the jury about it[187]:

“Now, the next part I want to deal with you is was the research a sham, because that is what the prosecution says. It is a sham, just a way of getting hold of carbon monoxide. So let us examine what is the evidence on that. You have reports by various experts. I have already mentioned who they are; three from the prosecution, two from the defence, about you know the protocol, about carbon monoxide poisoning and so on and so forth, so please look at all these reports carefully. They speak for themselves anyway and the experts have come into court to further explain to you. I will just try to give you a summary.

Now, all the experts, that is, both prosecution experts and defence experts, explained how carbon monoxide binds to one’s haemoglobin and that stops the oxygen from holding onto the haemoglobin of a person. And in turn, because oxygen cannot stick to the haemoglobin of the blood, so oxygen cannot be taken by the blood to various organs and eventually the lack of oxygen, a person will die. There is no dispute there. Everyone agreed that is what happened.

Now, there is a disagreement though as to what had been termed ‘extra haemoglobin effect’ plays a part on how the body was affected by the toxicity of the carbon monoxide. The prosecution camp, if I can call it that, you know, the experts on the prosecution side, they subscribe to the theory, that is, the toxic effect of carbon monoxide is not just limited to a decrease in the blood oxygen delivery capacity, but there is also this extra haemoglobin effect.

But the defence side, basically, just Professor Buckley because the other professor did not really discuss about this extra haemoglobin effect. The defence side, Professor Buckley, does not subscribe to this theory, on the basis that this theory was proposed by people who are supporters of what is called the hyperbaric oxygen treatment. They are the people who said, ‘Apart from the blood, there is all this other tissue damage’ and so on and so forth, so the hyperbaric oxygen forcing the oxygen into the body would help heal the patient faster, better.

Those are the people who say that there is this extra haemoglobin effect, though it is not proven, but Professor Buckley said, ‘I am not subscribing to it now because it has not been proven’. So basically, you have this difference of the different theories, but at the end of the day, it may not be of much difference to your deliberation whether there is extra haemoglobin effect or not. It is a matter for you.” (Emphasis added)

205.The Judge continued to summarise the evidence and the issues in relation to the applicant’s research[188]:

“Now, all experts for the prosecution said the elimination half-life of carbon monoxide in humans had been known for decades. You remember the elimination half-life. That is, how long does it take for a body to get half of the carbon monoxide out of the body? That is what elimination half-life is. All of them say that has been known for years, decades. You remember Professor Kwong said, ‘When I was a student I already know. Yes, we all know. There is nothing to experiment. You do not have to do a test to determine the half-life, elimination half-life.’ Of course, you remember in the protocol that is one of the aims that the defendant wanted to do. So that is what they said.

They also say they do not see any clinical benefit, I emphasise clinical benefit, to be obtained from studying the elimination half-life of carbon monoxide in a rabbit during the oxygen therapy. They do not see treating a rabbit the result would be useful to humans clinically in a clinical setting, in hospital, saving patients. That is what they said.

They all confirmed that when dealing with carbon monoxide patients or a poisoned patient which is an emergency situation, the standard procedure is to use 100 per cent oxygen to save the patient’s life with or without pressure. That is either normal pressure or with added pressure to push the oxygen in. Either way, it is 100 per cent oxygen. They would not use anything less than 100. There is no question of using 80 per cent oxygen or 60 per cent oxygen. That is their evidence, standard, 100 per cent.

So their evidence is even though pure oxygen, 100 per cent oxygen is by itself toxic because we do not have 100 per cent oxygen around the air that we breathe in, even though it had toxic effect, the toxic effect is small. The oxygen, pure oxygen toxic effect is small when compared with the toxic effect of carbon monoxide. So the risk of damage to a patient suffering from carbon monoxide poisoning from being given 100 per cent oxygen, less risk. It is a fair trade-off to save the patient’s life because the risk is so small. That is basically what they are saying to you.

So what they say is that the study on the impact of giving less than 100 per cent oxygen will have no benefit clinically because doctors would not try it. You try to save life, you give 100 per cent so what is the point of testing 60 per cent on a rabbit? That is basically what they are saying.

But the defence said differently. The defence experts said that the half-life elimination is still only within a range, different people, age, gender and so on. It is only a range. It is not very clear so there is still a point to study the elimination half-life and this study would be of scientific value. Further, as pure oxygen is harmful to humans, a study to determine the effect of a lesser concentration of oxygen, 60 per cent in this case proposed by the defendant when compared with the standard administration of 100 per cent oxygen, will be of scientific value.

That is what the defence said. It is useful. It is of use scientifically to study whether it is useful, whether it is beneficial to use a lesser amount of oxygen in treating because of the harm produced by giving a patient 100 per cent pure oxygen, so that is the difference there. You will have to decide whether you agree with either side or none. Obviously you cannot agree with both because they are conflicting.

Now, the prosecution experts also are critical of the methodology to be adopted; that is, drawing blood out and then infuse the blood with carbon monoxide and reinject it back into the animal. The prosecution said they all make this point that when people suffer from carbon monoxide poisoning they inhale. They do not get the carbon monoxide injected into their blood. So what is the point of studying a model when blood was taken out, infused with carbon monoxide and then directly reinjected back? That is the one point.

Another point is the experts said when you take 20 per cent, one-fifth of the blood out from the animal, the animal will be in shock. So effects of the shock will be there and so when the carbon monoxide infused blood was reinjected into the rabbit and then you study the data, the data is confusing. You do not know which part is from shock, which part is from the effect of carbon monoxide, so, useless. That is what the prosecution said.

Professor Michael Irwin also pointed to the study by this person called Goldbaum back in the seventies and told you that this particular research methodology, that is, drawing blood out, infuse with carbon monoxide, reinject it in the animal, is not new. It had been done before although not with rabbits, with dogs.

Now, Professor Buckley, for the defence, when asked whether the model can be transferred clinically, that is, you study about the rabbit with the result transferred to humans, can you just relate that way? He said the study may be looking at something other than a clinical model. So basically, Professor Buckley is not saying that it is a clinical model. It is studying something other than being used clinically. He said he thinks that the model can provide some interesting insight which might tell one how carboxyhaemoglobin was dealt with by the body and he thought the defendant was studying half of the problem.

Now, Professor Buckley also said that, because he was asked under cross-examination by the prosecutor, the chance of this model being clinically relevant is zilch, right, none. He was not prepared to say that. He said it was unknown. That is his view. He can only agree that the benefit -- the chance of this model being clinically relevant is unknown. So when Professor Buckley was asked, well, given that the universal ingestion of carbon monoxide was through inhalation, he agreed that people inhaled. Whether it is a forest fire or whatever, you inhale and get poisoning by the carbon monoxide; so he was asked whether the study of injection of carbon monoxide into the blood of an animal is of any use, his reply was that, ‘The use is not immediate but there might be an insight as a result.’ In other words, Professor Buckley was not prepared to say that the defendant’s research had no clinical relevance. He was not prepared to say that. He did not say there was. He is not saying there was clinical reference. He did not. But he was not prepared to say it had no clinical relevance whatsoever.

The other expert, Dr Wang, on the other hand agreed. He accepted that this method of blood out, carbon monoxide in, blood in, had no direct use clinically, under cross-examination. But he was of the view that the defendant was trying to establish an acute carbon monoxide poisoning model, not to do clinically. That was his view.

Now, Professor Buckley was of the view that the high concentration of oxygen might theoretically increase or decrease tissue injury, so there is a good argument for further academic research on the optimal timing and concentration of oxygen in carbon monoxide poisoning. Now, I place the emphasis on the word ‘academic’ as opposed to ‘clinical’. So basically, Professor Buckley was saying that academically the study can be useful. All the prosecution experts say clinically it is of no use, so bear that in mind.

Now, you will recall Professor Kwong actually said more than once when he was giving evidence before you anyone can research anything. Whatever you want to research of course you can research it because it is objective, but when it comes to whether a study would be of benefit to others, whether your research, you know, would be published and a recognised publication, it had to go through a procedure of peer review. Your peers will look at it and what would they look at? They do not look at subjectively what you wanted to do. They look at it objectively; is it of any use? After your test are you going to benefit people? Is it useful clinically? So that is what Professor Kwong was telling you.

Now, about the Goldbaum study, I think it is called G‑O‑L‑D‑B‑A‑U‑M, it is a German word I think. I find it hard to pronounce. Now, interestingly, Professor Irwin of course was the one who mentioned it, and Professor Buckley agreed that Goldbaum had done similar studies before with dogs. But Professor Wang insisted that the defendant’s methodology of withdrawing blood from the animal, the rabbit in this case, infuse it with carbon monoxide and reinject it into the animal, had never been done before as far as he knew, and he called it original, remember? In his report, he said it was original.

So when he was giving evidence, he was referred to a reference he used himself in the report. In fact, it is item 6, the Roderick study that was published in 2015 in August. Although you have not been given the article, you did not see it, but you heard evidence from Professor Irwin that in this article by Roderick, Goldbaum was mentioned. You remember Professor Irwin made the reference to say that, well, if the defendant had read Roderick, he would have known about Goldbaum, the study.

He was being cross-examined to say hey, hang on, that was published in August 2015. This incident happened in May 2015 and the study was earlier. How could he have read the Roderick, and Professor Irwin agreed and withdraw -- he cannot say that, but the point is Professor Buckley agreed that in this Roderick report, Goldbaum was mentioned.

So when the expert, Professor Wang, said, ‘No, never mentioned’, remember he was actually given the article. He was going through it, because the prosecutor said, ‘Well, show me where’, and he looked at it, ‘Yes, I can’t find it’. He never referred to the actual article itself, so of course you may think that Professor Irwin is lying to you or Professor Buckley is lying to you, but those two were saying, yes, in the Roderick report Goldbaum was mentioned.

What did Goldbaum do? He tested it on dogs, two different groups on dogs, and yet, Professor Wang professed that he had read the article because it was asked first, ‘Have you read that article?’ ‘Yes, I have’. ‘In that article that was mentioned?’ ‘No, I can’t find it’, so you will have to decide how much you can rely on Professor Wang when he said it is an original idea, no one had ever done it before. It is a matter for you.

There is also criticism on how crude or rudimentary or preliminary the protocols are. You heard everyone had commented on that, the prosecution side particularly. Now, Professor Joynt, the head of department, he told you that if he had seen this he would have sent it back, no way that he would have passed it to the Ethics Committee. He would not have passed it. It was just too crude, too rudimentary. Even Professor Buckley agreed that if he had been the head of department, he had been sitting on the Ethics Committee, these protocols come to him he would have sent it back, because there are still question marks around it. It is such a draft form. So those were definitely not -- the protocols were not in a form fit to be submitted, quite clearly.

Well, what did Professor Wang say? He praised the protocols, said, well, original, well planned, fully informed. Why there is such a big difference? You form your own view. Was it an original proposal? Was it well planned and fully informed? So how do these opinions assist you in deciding whether the research, the experiment, was a sham or not?

You look at the defendant’s CV, very impressive CV, there is no question about it, and you hear evidence from people like Dr Leung. He is a specialist in obstetrics and gynaecology. This doctor had the same medical interests as the defendant. The defendant and Dr Leung’s main interest is on pregnant women, delivery of babies, caesarean, unborn babies. So you can see the defendant did not do any research on this carbon monoxide before. He had done, no doubt, research on oxygen therapy or what is called supplementary oxygen, but that is all in relation to saving pregnant women who were giving birth, caesarean, and so on and so forth. There is nothing about studying carbon monoxide poisoned patients.

So you can see with reference to the CV was this proposed research outside what the defendant would normally do, outside his interests, it is a matter for you. Professor Joynt, his own supervisor or head of department, told you in the box that he had looked at the CV, looked at of course what the defendant had been doing because, after all, the defendant is in his department. He saw no link, no direct link between carbon monoxide poisoning with any of the interests that the defendant had shown on the CV. So that is on the prosecution side.

On the defence side, the suggestion to you is that carbon monoxide was used to produce a state of hypoxia, to test the different concentration of oxygen. The research was not aiming at carbon monoxide poisoning itself. So from the defendant’s CV you can see that he was doing a lot of research on supplementary oxygen. Was it something that you expect the defendant to do, a matter for you.

The state of the protocol, again, can also assist you in this area. The protocols had been described as rudimentary, preliminary, crude. You have firm evidence from Professor Joynt and Professor Buckley saying they would not have passed these protocols. They would not have vetted. So what you have to consider is when there is a germ of an idea, you start to develop the idea, so the protocol was still in a very, very initial form, draft form, and these protocols had not been submitted to the head of department. There is no dispute about that.

So what did the defendant do with this germ of an idea with the protocol still in draft form?  He already arranged for the two experiments.  Why, you may think?  What is the urgency?  Why go and test on these two rabbits?  Now, was it because the defendant wanted to do trial runs before the actual experiment later and making use of animals, according to the defence term which Professor Irwin did not like, ‘recycled rabbits’, but the idea I think you can get it is: was it the case that the defendant thought? ‘Well, these rabbits had already gone through other experiments.  They are going to be destroyed in any event, so make use of that window before they get killed.  Let me just try this out and see if it works to get the blood out, if it works’, and so on and so forth, so is that the reason why the experiment was done both on the 13th and on the 20th?  It is a matter for you.” 

206.At this point of her summing-up on the subject of the applicant’s research, the Judge turned to address the issue of reinjection of the blood into the rabbits[189]:

“Now, another aspect of this area is what was actually done? I have already outlined to you that according to Dick Chow, the carbon monoxide infused blood was not reinjected into the rabbit at all. Even Miss Tam described a totally different situation of the animal inhaling carbon monoxide and then had the blood tested. So what was the purpose of those two experiments then? Was it just to test how could one infuse the blood with carbon monoxide? Is that why it was testing or is it now 100 per cent with those blood or 80 per cent? We just do not know. There is no evidence on that.

Now, as to what was being measured, you remember you heard Professor Wang, or Dr Wang, saying that he read the results because there was the rise and sudden drop of carbon monoxide. It had to be blood. That is from the rabbit after the carbon monoxide injected blood had gone into the rabbit. You remember he was very clear about that. Of course he was not there. He gave actually two reasons in fact. He tried to say the defendant told him, but I stopped him, remember? He said, ‘The defendant said’. I said, ‘No, you cannot tell us what the defendant said’, so the other reason is he looked at the printout of the oximeter.

Now, why did I stop the professor from saying what the defendant told him? The reason is because the defendant did not come into the witness-box to give evidence saying what happened during those two experiments how the blood was put in or something like that. So Dr Wang not being present, he cannot tell you what some other people had told him unless there is evidence to support that. That is why I stopped him. We have got a term for it. It is called hearsay, 傳聞証供, but that is the reason.

But I will also remind you when you go through the expert report of Professor Wang, on page 13, there is a paragraph under the table listing the 12 people. He said, ‘These people knew about the experiment.’ That is what he said. I will just read out to you now, no need to look at that right now. He said, ‘Confirmed by Professor Khaw...’, that is of course the defendant, ‘...at least three times the rabbit was ventilated and the CO saturated blood was injected so there was no CO inhalation’, that particular line.

So what Professor Wang was trying to tell you that he was told by the defendant that on three occasions blood was drawn out, infused with carbon monoxide and put back into the rabbit, not by inhalation. I have already explained to you, you have to ignore that part of the evidence totally because Professor Wang was not there. He said he was told by the defendant. The defendant did not come up to tell you there were three times it had been reinjected, so just ignore that part of the evidence.

Then you will have to decide, on the one hand you have Dick Chow, his evidence, on the 13th no mention of blood being injected back in the rabbit, on the 20th positive evidence that blood was not injected. Saline was injected. He was there. That is what he said. Then you have Professor Wang looking at the printout to say, ‘Wait, according to this, it must have been reinjected’. You decide whose evidence you are going to accept.

Now, in your consideration of whether the research was a sham, you may also wish to consider the previous research application submitted by the defendant. There are P62 and 63, the very thick bundle that were produced and Professor Joynt said that, yes, he had seen them before. Those were the proper applications, remember? So look at that to compare them with the protocol now, but of course bearing in mind that the protocol, 113A, B and C, could well be just a draft form. It has not gone to the formal application stage yet.

You may also wish to consider whether basic information, such as had similar research been done before? Did you do any research to see, yes, it had been done on dogs so this is not the first time? And if so, should there be justification in the protocol and say, ‘Hey, I am intending to do this drawing out from the rabbit. I am aware that it had been done on dogs before, but why do I want to do it now?’ the justification of doing it, should that basic information be in this protocol already or should it not be? That might assist you. It is a matter for you.

Now, whether you come to the conclusion that the research was a sham in order for the defendant to get hold of carbon monoxide, or that there was a genuine or may be a genuine research and then he obtained the carbon monoxide, may have a bearing on your final decision at the end of the day what was his intention.  …”

207.The Judge, at the conclusion of her summing-up, reminded the jury of the defence case that the applicant’s research was not a sham.  She said[190]:

“Now, the defence had suggested to you not only that there was no motive and the experiment was not a sham, there was no plan to get hold of carbon monoxide. If so, he would not have left the paper trail and you have to consider that because there was clearly quite a lot of email and WhatsApp messages which indicated that he was getting carbon monoxide, he had used carbon monoxide. So consider that when you consider whether the prosecution had proved that it was the defendant who had put the yoga ball inside the Mini Cooper.

Of course, another aspect that you have to consider too is that very morning, on 22 May before he left for the Chinese University to attend the students’ presentation; I will not bring you there now, but the WhatsApp between him and Dick Chow also mentioned carbon monoxide.  So would he have been bringing the attention of carbon monoxide to himself on the very day that he planned to kill his wife?  That is a matter for you.”

208.Mr Grounds complained that the Judge was selective in her overview of Professor Joynt’s evidence[191].  He noted that the Judge only mentioned that the protocols outlining the applicant’s research had not been submitted to the University’s Ethics Committee; that the protocols were too rudimentary and had they been submitted to him, he would have sent them back; and that he saw no direct link between carbon monoxide poisoning and the applicant’s particular speciality. 

209.Mr Grounds submitted that the Judge failed to mention Professor Joynt’s evidence that reperfusion injury, the inflammatory response which occurs when oxygen is restored to tissue after hypoxia, could in theory be reduced by treatment with less than 100% oxygen and that at some time in the future it might be shown that administering 100% oxygen may not be the safest way of treating a patient with carbon monoxide poisoning.

210.Professor Joynt’s references to reperfusion injury and treating a patient for carbon monoxide poisoning with 100% oxygen arose in the context of his general criticisms of the applicant’s proposed research, including the procedure of injecting an experimental animal with carbon monoxide, rather than breathing it in.  He was particularly critical, and this was the focus of his evidence, of the rudimentary nature of the applicant’s proposed research and his failure to follow the strict procedures for approval of a research project of this kind.  Leading counsel for the applicant did not mention these matters in his closing address, nor did he complain about the Judge’s overview of the expert evidence relating to the applicant’s research in her summing-up.  As already noted in the quoted passages of the Judge’s summing-up, she did provide a summary of the expert evidence in respect of the use of 100% oxygen for patients suffering from carbon monoxide poisoning[192]

211.As pointed out by the respondent, the Judge gave a very comprehensive account of the competing evidence concerning the applicant’s proposed research. 

212.A key aspect of the prosecution case as to the genuineness of the applicant’s research was that the applicant had not submitted his research proposal to Professor Joynt.  He testified that the applicant had not followed the procedures required; moreover, he was critical of the nature of the research proposal the applicant had produced[193]. It was also the case that the applicant’s expert, Professor Buckley, a clinical pharmacologist and toxicologist, expressed his disapproval of the form of the proposal submitted by the applicant[194].

213.As explained by the respondent, the Judge directed the jury that there was a disagreement between the prosecution and the defence on the issue of “extra haemoglobin effect”.  She quoted Professor Buckley who testified that he did not subscribe to it because it had not been proven.

The reinjection of the blood

214.Mr Grounds complained about the Judge’s comments as to whether or not blood was reinjected into the rabbits.  It is contended that if reinjection took place it gave greater legitimacy to the applicant’s experiments.

215.Mr Grounds submitted that leading counsel for the applicant argued that the carbon monoxide oximeter readings for the two experiments revealed a percentage increase that was followed by a decrease of the carboxyhaemoglobin, which in the opinion of Professor Wang showed that the blood had been reinjected into the rabbits.

216.We will also need to address this ground in some detail to show the point made by Mr Grounds about reinjection was unsupported at trial and unsustainable upon appeal.

217.Mr Bruce addressed the jury on the genuineness of the applicant’s research and the issue of reinjection as follows[195]:

“Now, Professor Wang Jiangyu testified for the defence.  You’ll remember him.  I think he was the last witness. Professor Wang has no track record in carbon monoxide research itself.  He is not a practising doctor.  He has a CV which extends over many pages.  It’s here somewhere.  However, members of the jury, it isn’t quantity that matters.  It’s quality and insight.  He said that the method used in the animal research protocol is a new method and it’s a pilot experiment that may produce interesting results.  Professor Wang was unable to say how “interesting” would ever translate into anything clinically useful.

He agreed that it was only an animal study.  There’s no mention of clinical value.  There’s no mention of the problems with the study in its present form.  Dr Wang’s opinion was that the proposal was original, well-planned, a good protocol.  His standard of assessment is one I suggest to you that you would be entitled to question.  He and the accused had done some papers in the past.  You will recall in the course of his evidence, Professor Wang told you there was some debate about using 100 per cent oxygen as a means of treatment.  In this regard, he pointed to a very recent article in The Lancet, a highly regarded publication, saying that the application of 100 per cent oxygen therapy might produce some bad results.  That was true as far as it went.  But what is clear, as Professor Wang finally conceded, the article was only talking about a variety of medical conditions and none of them included carbon monoxide poisoning.

Although, it was outside his professional expertise or right on the border of his professional expertise, he accepted that the standard therapy for treating carbon monoxide poisoning when someone presented at hospital was the application of 100 per cent oxygen.  I think we heard that many, many times.

Professor Wang told you in his report, which he backed up in his testimony, that apart from protocol development, multiple preparation works were included for pilot experiments.  It involved many personnel, organisations, agents, etc.  And you remember that table of the 12 people mentioned in Professor Wang’s report and we saw the clerical staff, the salesman from Hong Kong Oxygen, that leaves 10, and a multitude of other people.  And I’m not sure I know what point he was trying to make, but if it was the basis that everybody knew about what this experiment was trying to do, no way, that’s just silly. 

Now, he said he got his information about that from statements and the solicitors for the accused.  I hope he doesn’t do that when he does his research.  He isn’t also very good at reading carbon monoxide meters either or the results of it.  Could we go to D15, his report, please?  That’s in the defence bundle.  And I’d invite your attention to page 13 and go over the page to 14.  Now, the problem with all of this is that the plain meaning of what was said here was that he thought those readings included reinjection of the carbon monoxide blood into the body of the rabbit.  Middle of the page 13, experiments:

‘The experiments on 13 May and 20 May 2015, on both dates, a rabbit was anaesthetised...(reads) So there was no carbon monoxide inhalation.’

And then let’s go over the page to the graphs.  You’ll remember this was him saying this showed the reinjection. Here’s the problem, as you and I both know, there was no reinjection on 13 May or 20 May.  So how he could interpret these graphs, the figures that were given from the printouts of the device that measures carboxyhaemoglobin, is something you may want to ask yourself.  This is an independent expert, a researcher, an expert, and he didn’t even get the basics right.  I suggest to you that might tell you something about the quality of his evidence.  I submit to you that the assessment of Dr Wang that the accused’s research proposal was good is simply unconvincing.”

218.In his closing speech, Mr McCoy addressed the issue of the genuineness of the applicant’s research and focused on Professor Wang’s evidence that the carbon monoxide oximeter readings showed the blood had been reinjected[196]:

“What’s hypoxia? You have learnt about this. It’s the cell death. What causes cell death? Many things. Something that will promote cell death is carbon monoxide. Carbon monoxide will lead to carboxyhaemoglobin that leads to apoxia. Remember Professor Wang, the last witness. He had done a PhD, I think from Australia, or maybe it was Chinese University, on foetal hypoxia, hypoxia cell death of the foetus inside the mother. I mean, if anyone knows about hypoxia, it’s Professor Wong. Professor Wong spoke very highly of the defendant and I will ask you to examine what he said about the defendant’s experiments in due course because yesterday the prosecutor made a big point and the big point was this, quote, talking about the defendant’s experiments, my learned friend said yesterday, ‘Here is the problem, there was no reinjection.’ Remember this point? ‘He has been saying it’s a reinjection model, but there was no reinjection. Ha ha, gotcha.’ Well, unfortunately, I’ve got him and I’ll show you.

Can we just have a look at Professor Wang’s evidence? That is Exhibit D15. I’ll just give you a minute to turn it up, Exhibit D15, Professor Wang. This is the man with two PhD’s. Some people are greedy for knowledge. And if you come to page 14, please, with the two graphs. Members of the jury, let me just now introduce this point so I can make it clearly. Yesterday the prosecution said this. He said Dick Chow never said that the blood was reinjected. That’s true, he never said it. Dick Chow said saline had been put in and then the prosecutor never asked whether there was any further step. So, there’s no evidence whether there was or wasn’t reinjection. But Professor Wang said there is reinjection. And then my learned friend yesterday quite unfairly, I think, and would ask you to consider, tried to trash Professor Wang by saying, ‘Professor Wang was just saying anything.’ But Professor Wang had done what the others, Professor Kwong and Professor Irwin, had never done. He looked at the oximeter readings. That’s the machine that automatically records the levels in either the patient, but in this case the rabbit.

Now, just have a look, please, at page 14.  This is so simple.  See those two graphs.  See the line of little diamonds that basically almost follow a flat line.  You can see those?  The top one and there’s another one at the bottom.  What are they showing?  Well, look at the left-hand axis.  They are showing 10 or maybe nine, you might think, per cent carboxyhaemoglobin.  I’m sorry, how did they get 10 per cent carboxyhaemoglobin in the rabbit?  Was it smoking a cigar or did it, in fact, simply get the reinjection?  Duh.  They haven’t looked at the evidence.  The 10 per cent concentration is because there had been reinjection of the carbon monoxide.”

219.Dr Dick Chow testified about the procedures adopted in the two experiments on 13 and 20 May 2015.  He said that the applicant had obtained a quotation for the purchase of carbon monoxide on 10 June 2014.  This was well before the applicant had suggested the proposed research to anyone.  He later purchased the carbon monoxide initially arranged by the applicant for the experiments in May 2015[197].  He referred to the procedure for the experiments as set out in a document entitled ‘Short-Term Experimentation’ under section D of the form[198].

220.He said the experiment on 13 May involved one rabbit, but there was some difficulty getting the blood out of the animal and saturating the blood with carbon monoxide.  Eventually they were only able to extract 10 millilitres of blood (instead of 20 ml) and settled on a method of saturating the blood with carbon monoxide by putting the gas directly into the syringe containing the blood.  They then tested the blood concentration of the carbon monoxide using the oximeter.  The experiment ended at that point[199].

221.The experiment on 20 May also involved one rabbit.  They followed the same procedure as before and only extracted 10 millilitres of blood “because again this is a rehearsal or trial, so we didn’t -- it’s not actually performing the actual experiment, so we are still testing out the protocol.”[200] He further explained[201]:

“A. So it’s kind of like preparing for the actual experiment, because we may need to -- we will need to inject the blood, the carbon-monoxide-saturated blood back to the rabbit later. So -- but his time we didn’t inject this saturated -- carbon-monoxide-saturated blood back to the rabbit. We just tried it with saline and to make sure the cannulation is not blocked.

COURT: So on this occasion, the rabbit was not injected with blood that is saturated …

A. No.

COURT: … with carbon monoxide.

A. Yes, we did not.”

222.In cross-examination, Dr Dick Chow was not challenged at all in respect of the procedures adopted for the two experiments; or that no reinjection took place on those two occasions[202].

223.Ms Tam was the laboratory assistant involved in the experiments who was responsible for the carbon monoxide oximeter analysis. Her evidence was read out pursuant to section 65B of the Criminal Procedure Ordinance, Cap 221.  Insofar as is material, she stated that at the beginning of the experiment on 13 May, the applicant and Dr Dick Chow let the rabbit inhale carbon monoxide and drew the blood from the rabbit.  The blood sample was passed to her and her colleague for analysis by the carbon monoxide oximeter.  This procedure was conducted repeatedly and the result was printed by the carbon monoxide oximeter for each analysis.

224.On 20 May, the procedure of the experiment was similar to the previous one.  The applicant and Dr Dick Chow let the rabbit inhale carbon monoxide and drew blood from the rabbit.  She received the blood sample for analysis by the CO oximeter and handed the results to the applicant.  She did not mention that there had been reinjection of the blood back in the rabbit[203].

225.Ms Tam’s evidence was slightly at variance with that of Dr Dick Chow as to how the carbon monoxide was infused into the blood, but they were both consistent in making no mention of reinjection having taken place.

226.Professor Wang’s opinion that reinjection took place in both experiments was solely based on his interpretation of the CO oximeter readings, although he seemed also to have been influenced by what he had been told by the applicant.  His evidence on this issue was as follows[204]:

“Q. So you believe that the process of the two experiments on 13 and 20 May involved reinjection of the blood into the rabbits?

A. There are two facts. I repeatedly asked Professor Khaw.

COURT: I’m afraid whatever you asked, Professor, cannot be evidence.

A. (Witness answers in English) Okay.  Okay. (Interpreted evidence continues) Based on the changes shown, I believe that it was put back into the body.”

The timeline defence

227.The term ‘timeline defence’ was a description used on appeal but was not one used at trial[205]. One of the many suggestions made by the defence was the timeframe in which the applicant had an opportunity to place the yoga ball in the boot of the Mini and remove its stopper, and the timeframe for the level of carbon monoxide in the Mini to result in the death of Madam Wong and Lily by the time and at the location where the vehicle was found.  This submission, to the extent it was made at trial, was in very general terms and with various qualifications because of the inexact nature of the evidence.

228.It is argued by Mr Grounds that the defence case in relation to the carbon monoxide concentration was that for Madam Wong to have been overcome so soon after departing from her home, the yoga ball must have been placed in the car shortly before she set off, which would rule out the applicant as the person responsible for putting the yoga ball in the car. 

229.Mr Grounds pointed to Mr McCoy’s closing speech to the jury when this defence was explained to them.  In order to evaluate the argument being advanced by Mr Grounds, it is necessary to refer to the actual comments of leading counsel, on which he relies.  It will be recalled that Madam Wong and Lily got into the car at about 2:00 pm and Madam Wong then drove the car for a distance of about 1.6 kilometres when she and Lily were overcome by the carbon monoxide released from the yoga ball.  Mr McCoy stated[206]:

“… But actually when it happened determines who it cannot have been. When it happened determines who it cannot have been, because when I show you the very low levels of CO that existed in this car, how do you know they’re low?

Well, because we’ve got Dr Foo’s table and, hang on, Buckley made it very clear, ‘You walk in, one or two breaths, you’re dead if it’s high.’ They got in the car, presumably, you know, they buckled up, reversed out, down the road, along, turned right, 1.6 K, that is the act of somebody who might have gone, ‘Gosh it’s a bit stuffy in here. I’ve got a bit of a headache or feel a bit flushed.’ You know, that’s life. We all feel like that sometimes. Most people just turn the window down. But Hong Kong people go, ‘The air is dirtier outside than inside, so we leave it up.’ Some people do that. If she’d opened the window, it would have just all flushed out. Put the aircon on, it would have flushed out, it would have recircled out, but it’s on recirculation so it’s going round and round.

So, somebody is driving 1.6 K, when she parks, Siew Fing, she put the handbrake on, put it into P, park. These are all actions of a thinking mind. I mean, you can imagine it. She must be feeling increasingly dozy. You know, if you drive a long way in some countries, the way that people wake is put the window down and turn the radio up. In Hong Kong, that’s not always the solution. But she’s driven that distance, so what does that prove? It’s not a one or two breath high concentration thing, it’s very low. It’s low enough to be able to drive and we’ll look at Dr Foo’s numbers. But, more importantly, we’ll look at Dr Wong’s numbers and Dr Li’s numbers.

… The number is low because remember the flushing effect of opening the door. Now, you can’t get it right. This can’t be precise. But we know from Dr Wong’s table, opening the doors 89 per cent late in the piece. Earlier in the piece, 40 per cent. So, you know, opening the doors has a massive flushing effect and opening two, we know this in our own homes with opening windows to get the flow, that increases it even more. So, you’ve got natural leakage. When did that start? Well, when the ball went in. When did that start? No idea. Got natural leakage, we’ll come back to that.

We’ve got the doors opening and remember the boot is a door for this purpose, and we’ll come back to that. And we’re going to have to work out just how you would, if you know how dangerous carbon monoxide is - and he does, that’s their case - how would you, knowing how dangerous carbon monoxide was, how would you actually squeeze it out in the boot or put it in the boot, because their case can’t be that before he went to the Mini that he knew the ball wouldn’t fit in the boot. I mean, what is their case? Sometime in that window, broad daylight. The maid got it wrong about the rain, she got everything wrong, of course. You’ve got the conclusive Royal Observatory. There was no rain until 3 o’clock, okay? She got that just wrong. The windscreen wipers were on in the car. Does that suggest they were still awake and conscious, albeit fading, at 3 o’clock in the car? Aha.

And, remember, they don’t die on the spot. It’s slow. It’s unconscious. It’s a silent killer. When you are unconscious, automatic breathing until the time that CO gets you. You continue to breathe. So, the time of death can only be in a band. You can’t point it out. If the window wipers are on and it’s a sunny day, it’s either put on by mistake or it was put on when it was raining and the rain starts at three. So, they might have been drowsy and not well, who knows. I just -- I don’t know where that point goes, but you need to think, perhaps, about that.

230.Later in his closing speech[207], Mr McCoy returned to this issue:

You’ll remember Dr Wong had to redo some calculations because the instruments in the Government Chemist weren’t calibrated correctly.  But the difference between the two experts, prosecution and defence, is not really material for this purpose.  So, five minutes, you’re up to the early 3,000 level.  That’s the sort of point, members of the jury, you should perhaps keep in your mind.

Now, let’s think about leakage from a vehicle.  Now, you cannot - cannot - exactly determine the leakage of this Mini or any car.  You can’t do it because there are variables.  It depends on too many unknowns and too many things.  But what is an absolute fact is that there is significant -- but I’m not quantifying it - significant leakage from every car.  And you probably can work it out.  It’s, well, common sense.  The more expensive the car, you may think, the better precision, the higher engineering, the less leakage.  I mean, I’m not here to advertise for Lexus or Benz, but they are quality vehicles of reputation and a Mini is a cheap little car and this one had done 90,000 kilometres on Hong Kong roads.  That’s not an airtight vehicle on any view on the matter.

So, whatever you put into the car, there’s just going to be an unknown and unquantifiable leak.  We do know under 10 per cent pressure it’s 50 litres at least a minute, but this is not under pressure, this is ordinary atmosphere.  It’s just you can’t quantify that amount.  But there’s leakage, that’s why cars don’t float.  They go to the bottom of the sea because they’re not watertight. In fact, they’re not designed to be watertight because of airbags and things like that.

Now, the second form of leakage that always occurs, and this is proved by the government chemist and Dr Li, is to open the doors and the boot is a door, as I said earlier.  And the third form of leakage is when the vehicle is driven.  You’ve got Dr Wong’s figures.  So, it’s one plus two plus three.  It’s natural leakage plus doors opening, plus driving leakage.  That’s what you’ve got to bear in mind.  But any attempt to give a real number to these would be misleading.  It’s just not possible because they’re so specific in every reasonable sense.

But what we do have are the numbers and the tables from the prosecution expert and the defendant’s expert and when it drops 89 per cent -- but I want to remind you that took after five hours, but in a five-minute period looking at Dr Wong’s table, five-minute period.  If you go to 300 divide by 60, that’s five hours, it’s 2966.  Open the door and it drops to 319, not 2319, not 1319, to 319.  It drops 89 per cent and you’ve seen Dr Li’s graph which chartered what Dr Wong had done.  Even earlier, after 30 minutes, simply opening the door once drops it 40 per cent.  It simply drops it 40 per cent.

So, you can understand trying to work out when the ball, on the prosecution’s case, must have gone into the car, is actually very important.  And you are now going to understand there are two ways -- two perfectly reasonably ways of looking at this.  The prosecution really are saying this.  Ball goes in and there’s leakage, natural, and the doors and the boot and the driving, and that’s why the number was low.  That’s why they could drive because you can’t drive if it’s higher, you’ll be unconscious.  Sorry, the other way of looking at it is the ball only went in just before they drove, not suicide.  I’m going to come to this.

Just look at the five-minute thing. After five minutes, everybody agrees it’s either 3409 or Dr Li, slightly less, at 3100.  Open the door, massive drop, drive, further massive drop.  It is entirely consistent on this evidence from the government chemist that the ball was only put in the car shortly before they drove, which means, if that’s right, it can’t be the defendant because he’s on the other side of Sha Tin teaching 250 students.”  (Emphasis added)

231.It can be seen from leading counsel’s speech on this point that it was heavily qualified, and not in the concrete terms in which it is now presented before us.  As will become apparent, counsel on appeal are, to a significant extent, arguing a completely new and different case.

232.It is complained that the Judge did not remind the jury of the defence case in respect of the timeline argument, and without giving guidance to the jury as to the conclusions open to them from Dr Wong’s tables as to concentration levels, nor from the tables of Dr Foo and Professor Kwong as to the consequences of different levels of exposure; and without properly reminding the jury of Professor Li’s experiments and his criticisms of Dr Wong’s evidence.

233.The argument presented in respect of the timeline defence vacillated between accepting Dr Wong’s evidence at its highest and rejecting his evidence because it was unreliable or inaccurate.  It is complained that the jury should have been directed to ignore the results of Dr Wong’s experiment as to fresh air being drawn into the car by assuming that the car engine was on and that the ventilation was set on recirculation.  As leading counsel for the applicant pointed out in his closing address, there were a number of variables and unknowns in relation to what were the exact circumstances at the relevant time or times[208]. The Judge gave a lengthy and comprehensive overview of the expert evidence, particularly the evidence of Dr Wong and Professor Li.  In our view, the criticisms levelled at the Judge in this regard are unfounded, and amount to a selective view and novel analysis of the expert evidence. 

234.It was accepted by Mr Grounds that the Judge dealt at great length with Dr Wong’s and Professor Li’s evidence but complaint was still made that the jury was left without proper guidance as to the implications of Dr Wong’s tables and the carbon monoxide exposure tables. These tables were all before the jury and the Judge did direct them to use the tables as appropriate to see the effect that could be expected from the concentration levels of carbon monoxide[209].

235.Mr Grounds has carried out a detailed analysis of the carbon monoxide level in the car over a time period, which ignores, as was appreciated at trial, that such analysis was subject to a number of variables and unknowns. 

236.The relevant passage of the Judge’s summing-up on the timeline defence was extremely comprehensive and addressed the relevant issues raised by the parties[210].

“Now, I will remind you the evidence about the tests inside the car for the concentration, the dissipation of the carbon monoxide from the yoga ball and the natural leakage of the car. Now, I will, as I told you, give you more detail about what Dr Wong said because you do not have the report in front of you, so bear with me.

Now, Dr Wong, I will tell you his background first because you also need to know on what basis he is giving evidence as an expert. Now, he is a graduate of the Royal Society of Chemistry, RSC UK, Master of Science in analytical chemistry and instrumental analysis. It is Loughborough University UK. He has got a PhD in analytical instrumentation at King’s College University of London. He is a member of the Royal Society of Chemistry and a chartered chemist. That is UK as well. He is also a chartered scientist. He had been employed as a scientist by the government since 1992. He has been accepted by court as an expert and he had been accepted by me as an expert to give evidence.

So his evidence is on 28 May, that is six days after the incident which happened on the 22nd, he examined the Mini Cooper in the Tai Lam Chung vehicle examination centre. He said all the doors, windows and the tailgate, that is the backdoor, he called it the tailgate, were closed except the offside window because it had been broken, so it was taped and I think you can see from the photograph how it was taped.

So he tested that one and his test was to see if there was any leakage of carbon monoxide into the cabin. So he sat on the passenger seat and then drove the Mini Cooper around the compound and he place carbon monoxide monitors around the headrest and no leakage was found. This is from the car, from the engine itself. I think you already heard evidence there was nothing mechanically wrong with the car but he did the test anyway and there was no leakage from the car.

Then he had the Mini Cooper delivered to the government laboratory where he worked to conduct further tests on the same day. Now, he told you he filled the inside of the car with white smoke to check if there is any leakage, visually you can see the leakage, and there was no significant leakage except where the broken window with tape was. That test was done with the engine idling and the car stationary, so engine on but idling.

The next test that he did, he placed the Mini Cooper inside a vehicle cage and the engine was idling and then he monitored the carbon monoxide level both inside and outside the car. He read some -- outside the carbon monoxide was 10 ppm, but the maximum reading outside was 45 ppm because it was towed around for a while so different readings, but there was none inside the car. Again, there was just no suggestion there was any leakage from the vehicle itself, no suggestion that the CO, the carbon monoxide came from the engine. So even though of course Dr Wong had done these tests, they were not challenged at all. They were not really an important part of the case.

The next test he did there is some dispute about this. Now, he did another test, this is still on 28 May. This is the test with the burning incense. He put burning incense inside the car and the incense generated smoke and then he visually checked to see if any of the incense smoke came out. Now, he actually used the term “saturated”. He burned the incense enough to fill the whole cabin with smoke and then he checked any leakage. He found that only from the gap of that broken offside window some smoke came out. No smoke came out anywhere.

When he was cross-examined on this, he said he did not count how many sticks of incense he used, just a whole bunch that he propped up inside the cabin. He also did not measure how much smoke. He could just visually see that the cabin is filled. He accepted that the smoke, the incense smoke would be hot because he burned the incense, so the incense smoke would be hot. But he said that by the time the smoke came out that you can see from the window, it would not be that hot anymore. That is his evidence.

He also agreed that the Chinese incense, the component are mainly of wood dust, so he had not studied the size of particle created by burning wood. He had no idea. He did not study about it, but he accepted that the size of the carbon monoxide molecule would be significantly smaller than the particles created by burning incense. But he did not agree that the tests he conducted with the burning incense were poorly designed because that was the criticism from the defence expert. He did not agree that it was poorly designed or lacked scientific validity.

He said he did not agree that the only legitimate way to test leakage from the vehicle is the pressure test. You remember the other Dr Li did the pressure test. He did not agree that it is the only legitimate test but he agreed that that would be a good test; the pressure test would be a good test. He did not do a pressure test of course. Now, he said for his purpose just to see whether there is any leakage from the Mini Cooper so burning incense would already be a good enough test, whether burning incense or using a pressure test is still the same purpose, equal validity. That is what he told you. So that was what was done in May.

Then about six months later, on 26 November, he examined the yoga ball. That is P17. You remember he told you he tested there was no carbon monoxide inside the ball. He told you how he squeezed some of the air out and used bag and testing and the sample bag was only about 200 millilitres. The small sample was good enough for him to test. He said there was no sign of carbon monoxide inside the yoga ball.

He also measured the volume of the yoga ball and he measured the outside of the ball by fully inflating the yoga ball with compressed air, so visually he filled it all up with compressed air and then he measured the height of the ball and he then -- how did he do it because there was no plug, remember? He described he had something called a pipette, like a pipe, a small pipe, a pipette, and that he just fashioned it from his laboratory to make a plug and plugged the yoga ball after he filled it up, so the ball then maintained the compressed air inside the ball.

He did not measure the inside of the yoga ball. In other words, he did not measure the thickness of the material so that he can get the inside volume of the yoga ball. He just measured whatever is outside to determine with the equation to the volume of the whole yoga ball, and because the diameter, not the radius, the diameter of the whole yoga ball was 58 cm so he worked out the volume was just over 102 litres.

Then he did a test to see how quickly or how slowly the yoga ball deflates. What he wanted to test, he told you more than once, was to see whether it will be very quick like a balloon or very slowly like a soufflé. I think you all have soufflé before and you know the consistency of soufflé. So that was his purpose of testing. He prepared a table which you have, and that is Exhibit P128, at page 2178, in table 4, how the time when he started the test, for how long and what height was the ball to determine the rate of air coming out of the yoga ball.

On this occasion, of course the plug was taken out. The pipette was taken out otherwise the air cannot come out, so that was what he did and he did not use carbon monoxide on this occasion, only air, compressed air. He was of the view that whether using compressed air or using carbon monoxide he expected the two would be comparable. He also said that the density of carbon monoxide is very slightly lighter than air, not too much of a difference.

Now, this Dr Wong was asked whether a suitable gas to use would be nitrogen because nitrogen and carbon monoxide had the same density. Dr Wong’s answer was not to the extent that we need to observe. He did not test with nitrogen. He said for what he needed to observe there was no need to use nitrogen. He also agreed that when the yoga ball deflates it will not keep its spherical shape. He agreed with that. So that particular table, because it was every five minutes what happened, right, with the height measured from the polar region, if you remember.

So it was put to him that that table was very misleading because if one looked at the table, it would appear that it was every five minutes’ even interval. The ball would go down sort of like evenly, gradually, but the ball is not one dimensional. The ball is three dimensional, so the table gave one the idea it looks like it is a one-dimensional item. Remember he was asked about that. He reiterated, he said that again the test was only trying to show to the court that the ball did not deflate like a balloon. It deflated like a soufflé, so that was his reason of doing the test that way and having the table that way. That was what happened in November 2015.

Now, almost a year later after the incident, on 16 May 2016, he examined four yoga balls. These were the other yoga balls that were found inside the defendant’s house and none of them contained carbon monoxide. Then four days later, on 20 May 2016, that is the time when he was supplied with that big red canister of carbon monoxide and the yoga ball, that is P17, he had to test the varied yoga ball with the carbon monoxide from the canister.

What he did was, he told you, he filled the whole ball up full with carbon monoxide and then he tried to put it inside the yellow Mini Cooper and found that he could not close the tailgate, so he let some of the carbon monoxide out. Visually he did not measure visually about one-quarter out, so the ball left behind will be about three-quarters full of carbon monoxide. Then he found that he could close the tailgate and the flap. You all have seen the Mini Cooper. You could see there is a flap. The flap was on top of the ball but not pressing hard, no pushing the ball you know to force the release. Then he left it like that and then of course he removed the stopper first to make sure that the carbon monoxide can come out.

At that time the Mini Cooper was on a towing platform being towed around, engine was on, air-conditioning was off, the ventilation was on, so it was towed around. The result of the test was shown in table form. This is page 2179 of the same Exhibit P128. This time after half an hour, the plug had been removed. Half an hour later, the reading he recorded was 5-8-0-0, 5,800 ppm. At this stage then both sides, because you remember the Mini Cooper only had two doors, passenger side and driver’s side, both doors were opened for 30 seconds and then closed again. Then the reading dropped from 5,800 to 3,480, quite a big drop, about 40 per cent. Then after another six minutes the reading then dropped to 1,360 and further, after 27 minutes, down to 310. It is on that table. He told you to have the table to help you.

Dr Wong said that, yes, there are variables that would affect how much quantity of carbon monoxide was inside the cabin; for instance, door opening, that clearly affected it, and whether the vehicle is in motion or not and also there is the natural leakage rate from the car, naturally that would leak out. So these would all affect how much was inside the car as the concentration.

Now, on 20 October, that is almost six months later, the first test with the yoga ball for just over half an hour was done on 20 May. Then on 20 October he did another test again with the same yoga ball. This is P17. This time it is a much longer test, over five hours. You remember there is another table for it because the first one was only a short test, just over half an hour. This time he decided to do more than five hours. So this time the ball was still okay, because you remember the ball burst later but at this time, the P17 was still intact.

Again he filled the ball with carbon monoxide from the same canister. The fullness, he said, may be a bit more than three-quarters this time. Again, he was only doing it visually gauging it. He filled it as full as he can but then the yoga ball can still fit into the Mini Cooper and then he can close the tailgate or the door. He had the photograph. That is photo 2 of Exhibit P128A. You can see how the ball fitted into the yellow Mini Cooper.

Now, on this occasion, he said the plug was provided by the police and what he did was he put two uncalibrated monitors inside the cabin on the dashboard. Why did he use two uncalibrated ones and not a calibrated one? The reason was because a calibrated monitor could only measure carbon monoxide accurately under 1,000 ppm. The carbon monoxide to be released from the yoga ball would be more than 1,000 ppm and an uncalibrated monitor could measure carbon monoxide up to 9,990 ppm, so he chose the uncalibrated ones.

What he did then was that afterwards he checked the two uncalibrated ones against a calibrated monitor, like checking two watches; one is a synchronised one and one is not. Then he can tell the uncalibrated one is how many whatever is out to do the adjustment. Also the other thing he did is, because there were two uncalibrated monitors, so apart from doing the adjustment, he also take an average of the two, not just either one, so that is how he did this test.

Now, you can see that from this table after the doors, both doors were open for 30 seconds, that 300 minutes, that is five hours, the reading dropped from 2,966 to 319. That is an 89 per cent drop, a huge drop. Then the reading was further dropped down to 86, at 310 minutes. So that means at 300 you can see it was 2,966. The door opened for 30 seconds, closed, and then at 305, five minutes later, it was 319, and another five minutes later, at 310, it was 86. So the reading was not immediately after the door was closed, after 300 minutes after the closing, opening and closing, they still wait for five minutes before they do the reading. But in any event, the main point, of course, is once both doors were open there was a huge leakage of the gas, the carbon monoxide inside the vehicle.

You remember there is another table yet. This one, what Dr Wong tried to show is the reading between five minutes to 170 minutes because in the big one that I have just mentioned for the five hours, between five minutes and 170 minutes he did not give all the details, so he gave you another table just to list out all those details. There is a graph, at page 2181. Now, you remember in that graph he indicated that the concentration of carbon monoxide between 15 and 170 minutes exceeds 7,000 ppm, almost like a plateau, remember, that part of the graph. Dr Wong told you that the actual reading was in fact beyond what the uncalibrated monitor could measure. That is beyond 9,990 ppm. He called it overloaded. So he did the correction with the calibrated one and worked out that the ppm, the concentration exceeded 7,000 ppm.

Now, he was also asked about the difference between the very first reading after five minutes. One uncalibrated monitor said 4,549. The other one says 2,269, quite a big difference, remember, so he was asked why would there be such a big difference. They are both there in the car five minutes. He thought, he cannot be sure of course, he thought it could be because of the position of the yoga ball and at that time the gas coming out had not saturated the whole cabin. That is the equilibrium had not been reached, so maybe that is the reason why one registered more, one registered less. That is the explanation he gave. He said after that then the two are more consistent with one another. After five minutes, the reading did not vary that much. He was of the view that the errors were well within range and satisfactorily comparable, so that was what was done on 20 October.

Then almost two months later, on 16 December, there was a further experiment with the yoga ball. Now, the experiment -- the yoga ball at P17, and another yoga ball, you remember the blue yoga ball. The idea is just to see whether the plugs are interchangeable and he found that they are. You take one and plug the other, they all work and no leakage and so on, so that was done on that day.

Then the next test was in 2017, March 17. He was asked to do one more test with P17, the yoga ball, the actual yoga ball that was found inside the Mini Cooper, again with carbon monoxide to do another simulation. He told you he filled the yoga ball up about three-quarters, visually checking three-quarters, and put it in the Mini Cooper just like before and plugged the plug. But he found out that three hours later when he opened the car door the yoga ball had burst. Remember there was a tear there, so the test did not work because you do not have a proper yoga ball anymore. He said he thought of using another yoga ball to do the test, but he decided against it, so the test was aborted on the last occasion.

So that basically was what Dr Wong told you he had done and the table is there to assist you of the various tests. So, of course, the importance in this case is not his test that there was nothing, no carbon monoxide coming out from the engine. The important test that he had done was whether there was any leakage from the car, from the cabin of the car, what was the dissipation rate of the yoga ball, that is the carbon monoxide coming out from the yoga ball, and how much concentration there was in the cabin of the car at different time intervals. So those were his evidence.

Now, I remind you what Professor Li told you. Professor Li’s report was actually given to you. That is D13. Now, because you have the actual report in front of you, so I will not go into detail. You can read that for yourself, but I can tell you that Professor Li did not do the test with the actual ball. You remember he never touched P17. Instead, he used three new yoga balls. They were produced in court and that he said it was of the same size and material as the yoga ball in question, D14A, B and C.

He used the blue Mini Cooper to do the test. He did not do the test on the yellow Mini Cooper and you have seen the blue Mini Cooper of course when we all went to the Police Training School to look at the yellow Mini Cooper. The blue one was next-door, so we had the chance to see both. Of course, you have the Admitted Facts that they were made in the same year, same model. Even from the mileage they seem to have done similar mileage as well.

So Professor Li gave you the internal volume of the yoga ball. It is 100.58 litres, so slightly less than the 102 that Dr Wong gave you, but of course, that is after Professor Li had deducted the thickness of the material to get the internal volume of the yoga ball. Now, in his experiment he did not just gauge this three-quarters or whatever. He actually tested. He found that the yoga ball with volume larger than 65 litres will not fit in. Of course, if you filled it all up it would be 100.58 litres, but he found that if it is bigger than 65 litres then it will not fit into the car and have the door closed.

He also told you how he tested it. He filled it with the air and then checked how much had gone in, in order to decide how much was inside the yoga ball. He also told you that there is a theoretical maximum concentration; theoretical in the sense that assuming the car is it had no leakage whatsoever. If that is the case, because from the documents coming with the car he already knew that the volume, the internal volume of the car, of the Mini Cooper, was 2,306 litres. So he worked out that the theoretical maximum concentration of carbon monoxide inside an airtight Mini Cooper was 29,000 ppm, but of course he told you no car is airtight, including the Mini Cooper, so no way that it would be 29,000 ppm inside the Mini Cooper because there would be the leakage.

About the deflation rate of the yoga ball I have already reminded you a little bit. Professor Li was pretty critical of Dr Wong’s method because he said it was misleading. The data gave one the impression it was one-dimensional when it was three. Also, he said that the table prepared by Dr Wong gave one the impression that it was a constant rate of deflation when it cannot be because it is a sphere. It does not deflate steadily at the same rate all along, so that was his criticism. Professor Li also told you that his experiment with the new yoga ball shows that the yoga ball could not be deflated totally flat on its own because of the material because it is rigid. It just cannot be totally deflated unless, you know, there is passive pressure from outside.

He told you he used nitrogen to do the test because he did not have any carbon monoxide to do the test. He used nitrogen, but nitrogen is proper to be used because nitrogen and carbon monoxide have the same density. So he filled up two balls with nitrogen up to 65 litres and he tested the two balls one at a time, so he tested one first and tested the second one going through the same procedure. So each ball has its stopper removed for five minutes and then it was refilled to measure how much had leaked out. He said to you that the amount of nitrogen needed to refill the yoga ball after the gas had been let out for five minutes would indicate how much had leaked out, so that is a more scientific method.

He had a table 3 in his report showing the deflation over time and the theoretical concentration of ppm. That is assuming the car was airtight. It is on the table itself. It is theoretical. Not the actual one. Actual one has to take into account the natural leakage. He observed that the rate of deflation decreased with time because between five minutes -- at five minutes it was 7 litres and yet 120 to 180 minutes was 4.5 litres. So the longer the time is, the less it deflates. It is not constant. You can see that from the table. The ball deflated faster initially and slower later.

In his report, on page 9, he used a diagram to show the experiment done by Dr Wong on 20 May and on 20 October 2016; you remember different colours that he used. Now, he also observed that the variables that may change the level of carbon monoxide concentration include the opening of the car doors and leakage from the cabin of the vehicle.

He also observed another variable and that is because the two tests were done on two different days. One is sort of summer or closer to the hotter time in May, the other one in October, cooler time. Remember he was telling you that the weather may make a difference if, say, the car is under the hot sun the air will be lighter and more air would leak out and the door open, or with opening and closing the door it had a fan effect, you remember that. So to him these two tests also there is another variable there because of the weather.

He, Professor Li, pointed out that with the opening of the doors there is a huge drop. For the first one, 40 per cent drop, so it was just half an hour or so test, the other one over five hours it was 89 per cent drop, so it is quite clear that any opening and closing of the door substantially reduced the carbon monoxide concentration inside the car.

Now, about leakage now, Professor Li said that looking at Dr Wong’s test, clearly there is leakage. From the visual test with the smoke and the burning incense clearly, there is leakage and he agrees that there is leakage because no car is airtight. But he told you that in an experiment he conducted there was a massive leakage of almost 50 litres per minute. That is on page 10 of his report. In this test, he used compressed oxygen and released at approximately 50 litres per minute inside the blue Mini Cooper and then he monitored the oxygen concentration and the pressure inside the cabin. He had a graph, on page 12, showing the theoretical pressure increase, in blue, and the actual recorded pressure increase in red and showing that the cabin was not airtight.

He also did a smoke test. He did not use incense. He used smoke generated from something called Speedtox, which is an insecticide fumigator, to generate the smoke. He said when he did the test, there was no trace of smoke visibly leaking from any part of the vehicle, none at all. He did not see any.

Now, Professor Li criticised Dr Wong using the incense, I think I have told you, because his opinion was that the incense the main components were dust and particles created by wood dust is much larger than the molecules of carbon monoxide. So the fact that you do not see any smoke coming out at other parts of the Mini Cooper, say, for instance, apart from the broken window, not from the bottom, not from any doors, does not mean that carbon monoxide was not leaking out from those places. That is the point he was trying to make because the molecules of carbon monoxide is so much smaller.

He actually told you that if you compare the particles of carbon monoxide and the particles generated by the smoke from burning incense, the former is 50,000 times smaller, 50,000. The particles of carbon monoxide is 50,000 times smaller than those particles generated by the wood dust from the incense; so significantly smaller. So that was his point that you cannot see does not mean that did not happen. So basically, Professor Li criticised Dr Wong using the burning of the incense saying that it is not scientifically good enough because of the big difference between the particles created by the smoke from the incense and from the particles of the carbon monoxide itself.

Now, he was asked when you compare the leakage of gas and smoke can it be done empirically? Can you measure it? He said, yes, you can measure it but he did not do it. I mean that is under cross-examination, remember. He said you can measure the difference empirically between these two but he did not do it.

Now, he also told you that he had done a test of putting a fully-inflated yoga ball into the cabin of the blue Mini Cooper. He tried on the passenger side and he also tried on the driver’s side; on the driver’s side, very, very difficult. On the passenger side, still difficult but because there is the yoga ball he can press it, so he pushed it in, it could go in. But he also agreed with you that it was not in the report, so although he had done the test for whatever reason, he did not put it in the report, so that is what he said. That is what he told you about these tests.

Now, under cross-examination, he was asked that the balls he used were new, whereas the yoga ball in question, P17, is not new. So whether there may be difference in their elasticity, meaning the new ball is more elastic and the old ball is less elastic and he agreed that they may not be the same. The elasticity may not be the same because he used three new balls.

Of course, he was criticising Dr Wong’s use of the polar region. Remember he said that, one-dimensional, you do not just measure the height of the ball. During cross-examination, he was asked to look at the ball done by the picture in Dr Wong’s report how the ball had fallen down. He agreed that, when you look at it, still it is more the polar region that deflated more although there is still some deflation on the side. So he did agree that you can visually see it more top to bottom, the polar region, rather than on the side. He also agreed that the deflation was somewhat like a soufflé.

Professor Li agreed that if the air pressure inside the cabin does not increase by whatever is coming out from the yoga ball, then there is no pressure to force the air out, no difference in pressure, no pressure to force the air out. In his experiment using compressed oxygen, it was only at a 10 degree -- 10 per cent increase of pressure that he recorded this 50 litre per minute leakage. So in other words, if there is no added pressure there will not be a 50 litre per minute leak from the Mini Cooper.

Professor Li agreed that, according to Dr Wong’s experiment, between 15 to 170 minutes the concentration was greater than 7,000 ppm and when asked whether there was not much chance of it going down from this 7,000 ppm, he said it was probably so, agree it probably will not go down much from there. In re-examination, that is when he was asked questions to clarify his answers in cross-examination, he gave you a very vivid example of the difference between particles of smoke generated from burning incense and carbon monoxide. His example was a human being and a dinosaur, remember, very different, so 50,000 times larger. So when you think about it, remember a dinosaur as opposed to a human being.

So, of course, it is a matter for you whether you decide to accept Dr Wong’s evidence or Professor Li’s evidence in respect of these tests. But I would like to point out that when Professor Li did the smoke test with the insecticidal fumigator, no smoke was seen visibly escaping from the blue Mini Cooper at all, none. When Dr Wong tested with incense smoke, although there is no evidence of the particles in difference between incense smoke and the fumigator, we do not know what is the particle coming off a fumigator, but he saw smoke escaping from the broken offside window. The window that was broken it was taped but clearly not taped good enough otherwise there should not be anything coming out.

Then since both the yellow and the blue Mini Cooper were of the same model, made in the same year and had done similar mileage, you may think that you can draw the inference that smoke would not have been leaking out from the yellow Mini Cooper had the window not been broken. It was intact. It will be the same as the blue; no smoke would have come out. You can consider that.

Now, Professor Li did not do any tests with the carbon monoxide concentration. Dr Wong did. The concentration level of carbon monoxide recorded by the two uncalibrated monitors was adjusted by Dr Wong when he checked the readings with the calibrated monitor and he took the average. That is what he told you in his evidence. Such concentration would have been the concentration minus whatever natural leakage was from the yellow Mini Cooper. Whatever was recorded by the monitor, it must be the concentration minus what was already leaking out because it was then measuring whatever was inside the cabin.

You have to bear in mind at the time of Dr Wong’s test the window of the Mini Cooper was already broken. So at the time before the car was found stopped and later police came and what not, the car was not broken. The car was intact. It was broken by the sergeant, so the concentration inside would have been higher because it would not have leaked out from the broken window. Of course, there is no evidence of how high, but you can see that it must be higher even with the natural leakage from the yellow Mini Cooper. So bear that in mind when you consider the case.

So just to repeat, if carbon monoxide was leaking out of the yoga ball into the cabin of the yellow Mini Cooper when the offside window was intact, was not broken and was not patched up, less amount of carbon monoxide would have leaked out and the concentration level inside the Mini Cooper would have been higher than what was shown by Dr Wong’s evidence, his report.”

237.Having dealt with the expert evidence and related issues, the Judge went on to deal with the question of who put the yoga ball in the boot of the Mini, and when it could have been done.  She noted that there was no evidence as to when the yoga ball was actually placed inside the boot, and then took the jury through relevant factual matters for them to consider[211].

“Now, you will have to consider if it was indeed the defendant who had placed the yoga ball with the carbon monoxide inside the Mini Cooper. The window of time that allowed him to do that would be between the time that Madam Wong had returned after dropping the kids to school and before he left for the Chinese U to attend the presentation, so it is not a very long period of time. Now, since Madam Wong was seen admiring flowers in the garden and playing with her mobile phone in the sitting-room before she went upstairs about 10 o’clock then you may think it is unlikely to be before 10 o’clock while she was still in the garden or while she was in the sitting-room, so you may infer that it is more likely to be after 10 o’clock, but again, it is a matter for you.

Now, I suggest you to use the tables prepared by Dr Wong as a guide of the time of the dissipation of the yoga -- the gas or whatever, the carbon monoxide from the yoga ball for just about half an hour and for five hours.  Use the tables to assist you to see what kind of concentration you can expect, bearing in mind of course it would not be exactly the same because the window of the car was not broken then and the dissipation may not be exactly the same even though it is the same ball.  But use that to gauge if it had been at 10 o’clock, what, it had been 10.15, 10.20, something like this so that you can gauge for yourself what kind of concentration you can expect and then with the opening of the car door, of course that would affect it.  Now, from the two tests done by Dr Wong, one for just over half an hour, one for over five hours, there is no question about it that the opening and closing the door resulted in a huge drop.  I mean 40 per cent is already a huge drop, 89 is even bigger.  So how long the door was remained opened and then closed obviously would have an important part to play here.”

238.There is no justification in the criticisms of the Judge’s lengthy outline of the evidence of Dr Wong and Professor Li.  She specifically mentioned the two tests that had been carried out by Dr Wong and the differences between them, as well as the various factors and circumstances that had to be taken into account when considering this evidence.  She suggested to the jury that they use Dr Wong’s tables to consider when the yoga ball would have been placed in the car, noting that the opening and closing of the car doors would have caused a “huge drop” in the concentration of carbon monoxide present and that the period of time the doors of the car remained open was important because of the possible dissipation of gas[212].

239.In contrast, Mr Grounds submitted that the Judge should have directed the jury to ignore Dr Wong’s test results of gas concentration in the car, although at times he relied on them in his arguments.

240.The fact is the evidence in respect of the gas concentration and dissipation was before the jury and the Judge comprehensively summarised that evidence and appropriately identified the issues that had been raised by the parties.  The Judge also reminded the jury of the timeframe and opportunity for the applicant to have placed the yoga ball in the car.  It has to be remembered that the evidence concerning the gas concentration and dissipation of carbon monoxide in the car was subject to a number of variables and qualifications and was part of a matrix of evidence that the jury were asked to consider.  As noted by Mr Bruce, the undisputed evidence was that Madam Wong with Lily got into the car and managed to drive for a distance of 1.6 kilometres before she lost consciousness, both exhibiting high levels of carbon monoxide in their blood.

The missing stopper

241.It is complained that the Judge misdirected the jury as to whether the stopper was missing on 22 May 2015.  The applicant made three points.  First, the Judge did not appropriately address the possible errors and unreliability of the exhibits officer as well as the possibility that he may have been mistaken when he stated that he had not found the stopper. Secondly, the Judge omitted to direct the jury as to the possibility that the stopper had been mishandled or misplaced in the six months which had elapsed before the exhibits officer searched the Mini.  Thirdly, the Judge erred in directing that if the jury were sure the exhibits officer had been truthful then Lily could be ruled out as the person who had put the yoga ball in the Mini as she had no reason to throw the stopper away.  It is contended the same direction should have been given in respect of the applicant.

242.The last point complained that the Judge failed to direct the jury that if there was no stopper there was no reason for the applicant to throw it away.  This presupposed the applicant put the yoga ball in the boot of the Mini and removed the stopper.  This argument was contrary to the defence case at trial.  If the applicant had done so, then it would have supported the prosecution case against him that he put the yoga ball in the car and intentionally released the carbon monoxide.  In such circumstances, it would be argued that he threw away the stopper so as to draw attention away from the yoga ball as the source of the carbon monoxide.  Sensibly, leading counsel did not make a point of this before the jury, although it does highlight the problem when a new team of lawyers take on a case on appeal and with creative fervour come up with a point which was not run at trial.  As to the body of the complaint under this ground, there is no substance in it.  The Judge reminded the jury of the very things that the applicant has complained she did not do.  She reminded them of the defence case and told them that they should consider the evidence of the exhibits officer and whether his account of not finding the stopper to the yoga ball was reliable.  She also mentioned the matter of the exhibits officer mixing up the exhibit numbers and other items[213]

243.At the conclusion of her summing-up, the Judge gave a comprehensive overview of the defence case, including the defence argument about the missing stopper.  She said[214]:

“So when you consider this scenario suggested by the defence you may also wish to consider the evidence of the exhibit officer.  That is DPC, or Detective Police Constable 7021, about not finding any yoga ball stopper or plug inside the Mini Cooper, whether that piece of evidence is reliable.  You heard how this officer had mixed up exhibit numbers.  He had to be recalled.  He got the number of the two car keys mixed up, so it affected the reading because one key, of course, the last recorded use was on the 22nd.  The other one was the night before, about 9-something. So you will have to consider how reliable is this officer when he said, “I searched the whole car and I can’t find any plug or stopper to the yoga ball.”

Now, when you consider, of course it is a matter for you, but you may also like to think about when someone says, ‘Well, there are two items and I mixed these two up’, is it the same as, ‘There was something.  I said I haven’t found it’, or ‘There wasn’t anything and I said it was there’, if someone is unreliable would it be unreliable to that extent? That basically is what I am saying, but it is for you to decide.

Now, why I suggest that it may be of relevance, because if you are sure that the officer was correct that he searched through the whole car and there was no yoga ball stopper or plug found, you may consider then if Lily had been using the yoga ball just to try to get the gas out to kill insects, what happened to the plug?  I mean there was no reason for her to throw away the plug.  I mean would it not be just there because she was only trying to get the gas out? 

So if you are sure that the exhibit officer was telling you the truth that there was no plug, then whoever had put the yoga ball in there had removed the plug totally.  There is just no plug at all to stop the gas from coming out and even afterwards, there would not be any plug to show you that originally there was a plug to this yoga ball before it became deflated.  So that might help you when you consider who would have -- who had the -- who had done it actually, who put the yoga ball in the Mini Cooper.

Now, I have already reminded you the state of the yoga ball as shown earlier from photographs 4 and 5, so I will not repeat that now.  But if you come to the conclusion that the ball was not folded, then it might again assist you that it was only naturally deflated.  It is not folded, because if someone had folded it, from the evidence if you accept that the ball does not flatten totally, it had to be done passively then the folding action would have squeezed some more gas out.  That is why I suggest you to have a look at the photograph.”

244.It is plainly apparent that the judge did address the issue of the missing stopper detailing the evidence and emphasising the defence submissions on the matter.

Conclusion

245.In the course of this lengthy judgment, we have seen fit to recite extensive passages from the summing-up.  We have done this not only to show how many of the grounds are fully answered in the Judge’s summing-up itself, but also to demonstrate, with respect, how thorough and meticulous the Judge was in addressing the myriad of issues which were raised at trial. 

246.It follows from our detailed consideration of the evidence and the multitude of the grounds of appeal advanced that in our judgment the applicant had a thoroughly fair trial; moreover, there was cogent and compelling evidence to justify his convictions on both counts of murder. His application for leave to appeal against conviction is refused and his appeal accordingly dismissed.

(Jeremy Poon)
Chief Judge
of the High Court
(Andrew Macrae)
Vice President
(Kevin Zervos)
Justice of Appeal

Mr Andrew Bruce SC, counsel on fiat, and Ms Christal Chan, ADPP (Ag), of the Department of Justice, for the Respondent

Mr Christopher Grounds, Mr Andrew Bullett and Mr Franco Kuan, instructed by CL & Co Solicitors, for the Applicant


[1] Appeal Bundle, pp A60H-O and 61N-68K.

[2] Appeal Bundle, p A70N-O.

[3] Appeal Bundle, p C479N.

[4] Appeal Bundle, pp C479T-480A.

[5] Appeal Bundle, p C479T.

[6] Appeal Bundle, p C480H-L.

[7] Appeal Bundle, p 31, at [47]-[48].

[8] Appeal Bundle, p A140P-V.

[9] Appeal Bundle, p A33.

[10] Appeal Bundle, p A93A-C.

[11] Appeal Bundle, p A93K-M.

[12] Appeal Bundle, pp A92T-94F.

[13] Appeal Bundle, p A95A-P.

[14] Appeal Bundle, p A116O-R.

[15] Appeal Bundle, pp A90C-92N.

[16] Appeal Bundle, p A113G-I.

[17] Appeal Bundle, p A113.

[18] Appeal Bundle, pp A146S-147A-E.

[19] Appeal Bundle, p A147E-F.

[20] Appeal Bundle, p A147I-L.

[21] Appeal Bundle, p 147Q-U.

[22]  Appeal Bundle, p A148M-P.

[23] Appeal Bundle, p A149A-I.

[24] May May; Ms Sarah Niu, a classmate in Renaissance College; Ms Eunice Li Lok Ting, a school friend; Ms Lesley Ann Harrison and Mr Trevor John Newman, teachers from Renaissance College; Ms Foong and Dr Cheong, family friends; and Madam Chen Li, a neighbour.

[25] Appeal Bundle, p A151J-L.

[26] Appeal Bundle, p A151S.

[27] Appeal Bundle, pp A152P-153D.

[28] Appeal Bundle, p A157N-Q.

[29] Appeal Bundle, p A85K-L.

[30] Appeal Bundle, p A153H-M.

[31] Appeal Bundle, p A153M-O.

[32] In addition to these witnesses, the prosecution also relied on the evidence of Professor Qin Ling (PW8), Dr Dick Chow and Ms Tam Tsui Ping (PW23).

[33] The time taken for a body to get half of the carbon monoxide out of the body.

[34] Appeal Bundle, p A168I-N.

[35] Appeal Bundle, p A167H-L.

[36] Appeal Bundle, p A172Q-T.

[37] Appeal Bundle, p A169B-G.

[38] Appeal Bundle, p A169N-O.

[39] Appeal Bundle, p A170A-E.

[40] Appeal Bundle, p B727. 

[41] Appeal Bundle, p B715C-G.

[42] Appeal Bundle, p A169S-U.

[43] Appeal Bundle, p B722M-R.

[44] Appeal Bundle, pp B714N-715A.

[45] Appeal Bundle, p A160Q-T.

[46] Appeal Bundle, pp A161U-A162B.

[47] Appeal Bundle, p A172L-N.

[48] Appeal Bundle, p A171Q-S.

[49] Appeal Bundle, p A171R-T.

[50] Appeal Bundle, pp 507R-508D.

[51] Appeal Bundle, p 537G-M.

[52] Appeal Bundle, pp C544-547.  Ms Tam’s witness statement was admitted into evidence pursuant to section 65B of the Criminal Procedure Ordinance, Cap 221.  She stated that on each occasion they let the rabbit inhale carbon monoxide and drew its blood.  She made no reference that the blood was infused back into the rabbit.

[53] Appeal Bundle, pp A163-164 and A75S-U and A76L-N.

[54] Appeal Bundle, p A75P-R.

[55] Appeal Bundle, A172M-O.

[56] Appeal Bundle, p C209. 

[57] Appeal Bundle, p A155E-J.

[58] Appeal Bundle, p A77E-G.

[59] Appeal Bundle, p A68N.

[60] Appeal Bundle, pp A54P-R and B538L.

[61] Appeal Bundle, p A117E-F.

[62] Appeal Bundle, p A155P-U.

[63] Appeal Bundle, p A156A-E.

[64] Appeal Bundle, p A156K-L.

[65] Appeal Bundle, p A157B-C.

[66] Appeal Bundle, p A157H-M.

[67] Appeal Bundle, p A69B-C.

[68] Appeal Bundle, p A179D-E.

[69] Appeal Bundle, p A179J-L.

[70] Appeal Bundle, p A176S-T.

[71] Appeal Bundle, p A177A-D.

[72] Exhibit D1.

[73] Appeal Bundle, p A203N-U.

[74] Appeal Bundle, p A204J-L.

[75] There was no issue with the mechanics of the car and it was an admitted fact the carbon monoxide did not come from the car: see Admitted Facts, at [15], Appeal Bundle, p A19.

[76] This test was done over the course of 5 hours, using the Mini and a yoga ball filled ¾ full with carbon monoxide.  Another test was done on March 17, 2017, but when Dr Wong opened the door after 3 hours he found the yoga ball had burst.

[77] Appeal Bundle, p A194M-P.

[78] Appeal Bundle, p A188S-T.

[79] Appeal Bundle, p A110E-F.

[80] Appeal Bundle, p A197F-H.

[81] Appeal Bundle, p A190Q-T.

[82] Appeal Bundle, p A197J-Q.

[83] See Admitted Facts, at [49], at Appeal Bundle, p A32.

[84] See Admitted Facts, at [50], at Appeal Bundle, p A32.

[85] Appeal Bundle, p A198D-F.

[86] Appeal Bundle, p A198F-I.

[87] Amended Perfected Grounds of Appeal Against Conviction filed on 23 January, 2020, signed by Mr Christopher Grounds and Mr Andrew Bullett.

[88] Appeal Bundle, pp A127N-A129M, A133D-H, A136E and A137I-T.

[89] Appeal Bundle, p A129A-I.

[90] Appeal Bundle, p A130B-C, M-Q, and T.

[91] Appeal Bundle, p A128A-I.

[92] Appeal Bundle, pp A196C-206H.

[93] Appeal Bundle, pp A158T-176A. 

[94] Appeal Bundle, pp A124L, A145L-O, and A205N-206B. 

[95] Appeal Bundle, p A205I-M.

[96] Ground 11.

[97] Appeal Bundle, p A143I. 

[98] Appeal Bundle, pp A125D-H, 143I-J and 208A-211T. 

[99] Appeal Bundle, p A114S-T. 

[100] Appeal Bundle, p A113H-I.

[101] R v Henderson and Others [2010] 2 Cr App 24.

[102] Appeal Bundle, pp A178J-179K.

[103] Appeal Bundle, pp A124S-125C and A202B-204O.

[104] Appeal Bundle, pp A202B-206P. 

[105] Appeal Bundle, p A152C-K.

[106] Appeal Bundle, pp A177B-178N.

[107] Appeal Bundle, pp A131T-135P. 

[108] Appeal Bundle, p A133M-Q.

[109] Applicant’s Reply, dated 30 November 2020, at para 4.

[110] Applicant’s Reply, dated 30 November 2020, at para 5.

[111] There were two keys to the Mini.  Exhibit 2 was the key found in the Mini at the time of the deaths, Exhibit 11 was the key recovered by DSgt 16404 in the house on 13 July 2015, which was handed over to the exhibits officer.  See Appeal Bundle, pp C571F-C572C and C573T-C574E.

[112] Appeal Bundle, p A107R-S.

[113] Appeal Bundle, pp A16-A17, at [6]-[9].

[114] Appeal Bundle, p A205A-M.

[115] Appeal Bundle, p A205I-M.

[116] Appeal Bundle, p A160B-L.

[117] Appeal Bundle, pp A125T-A126E.

[118] HKSAR v Apelete (No 1) [2019] 5 HKLRD 574, at [69].

[119] HKSAR v Abedin Minhajaul [2022] 2 HKC 205.

[120] HKSAR v Chung Chun Fung and Another (Unrep., CACC 28/2020, 3 December 2021), at [35] and [36].

[121] Roach v The Queen (2019) 344 FLR 429, at para 195.

[122] Written Submissions of the Applicant, at [14].

[123] Appeal Bundle, pp C153-C154, at entries 2312-2322.

[124] Appeal Bundle, pp C154-C156, at entries 2328-2348.

[125] Appeal Bundle, p A91S-U.

[126] Appeal Bundle, p A154J-K.

[127] Appeal Bundle, at [6].

[128] Appeal Bundle, p A157C-M.

[129] Appeal Bundle, p A95R; p A103G.

[130] Appeal Bundle, p A96A.

[131] Appeal Bundle, p A95T.

[132] Appeal Bundle, pp A95U-A96A.

[133] Appeal Bundle, p A103G.

[134] Appeal Bundle, p A103H; p A103P.

[135] Appeal Bundle, p A103P.

[136] Appeal Bundle, p A103I.

[137] Appeal Bundle, p A103K; p A114T.

[138] Appeal Bundle, p A103R.

[139] Appeal Bundle, p A103J.

[140] Appeal Bundle, p A103T; p A104H; p A104J.

[141] Appeal Bundle, p A103L.

[142] Appeal Bundle, p A114S-U.

[143] Rice v Connolly [1966] 2 QB 414.

[144] Ibid., p 419E-F.

[145] Ibid., p 420A-B.

[146] Boyes v MacLeod 1997 SLT 1348.

[147] The two accused were brothers with the surname Boyes.

[148] Ibid., p 1349B-E.

[149] Lee Fuk Hing v HKSAR (2004) 7 HKCFAR 600.

[150] Ibid., at [55].

[151] Ibid., at [56].

[152] HKSAR v Ata Asaf (2016) 19 HKCFAR 225.

[153] In evidence, the appellant said that the apparatus was outside the window under an air-conditioning unit.

[154] Ibid., at [28] and [29].

[155] Ibid., at [10] and [11].

[156] Ibid., at [12].

[157] Respondent’s further written submissions, dated 10 February 2021, at [4].

[158] Reply to letter dated 28 January 2021 from Macrae VP, dated 10 February 2021, at [8].

[159] Appeal Bundle, p A133T.

[160] Appeal Bundle, p A133P-Q.

[161] Appeal Bundle, p A126D.

[162] Appeal Bundle, pp A128L-Q; p A129M.

[163] Appeal Bundle, p A129F-H.

[164] Appeal Bundle, p A129K-M.

[165] R v P [2008] 2 Cr App R 6.

[166] Ground 5.

[167] Ground 2.

[168] Ground 6 and Ground 17(a), (b), (c)(i), (ii) and (iii), (d) and (e).

[169] Ground 7(a) and (b) and Ground 16(a), (b), (c), (d), (e) and (f)(i) and (ii).

[170] Ground 8.

[171] Ground 11(a), (b), and (c).

[172] Appeal Bundle, pp A127O-128I.

[173] Ground 2(a). 

[174] Appeal Bundle, p A160C-N.

[175] R v Turner [1975] QB 834, at 841.

[176]  R v Kai-Whitewind [2005] EWCA Crim 1092, at [89].

[177]  R v Brennan [2014] EWCA Crim 2387, at [44].

[178] Appeal Bundle, p A125C-N.

[179] Appeal Bundle, pp A158T-159M.

[180] Appeal Bundle, pp A159Q-160C.

[181] See para VIII of the Amended Perfected Grounds of Appeal, which is set out in para 124 of this Judgment.

[182] Appeal Bundle, pp A96D-100C.

[183] Appeal Bundle, pp A165N-175S.

[184] Appeal Bundle, pp A160P-162B.

[185] Appeal Bundle, p A162C-P.

[186] Appeal Bundle, pp A162Q-165N.

[187] Appeal Bundle, pp A165O-166Q.

[188] Appeal Bundle, pp A166R-173N.

[189] Appeal Bundle, pp A173O-175S.

[190] Appeal Bundle, pp A205N-206B

[191] This addresses Ground 8.

[192] Appeal Bundle, pp A166R-168H.

[193] Appeal Bundle, p A162H-M. 

[194] Appeal Bundle, pp A171R-178A.

[195] Appeal Bundle, pp A66I-67Q.

[196] Appeal Bundle, pp A96M-97L.

[197] Appeal Bundle, pp C509R-511M.

[198] Appeal Bundle, pp C542-543, Exhibit 124E.

[199] Appeal Bundle, pp C522K-526O.

[200] Appeal Bundle, p C530D-E.

[201] Appeal Bundle, pp C530T-531F.

[202] Appeal Bundle, pp C536F-540B.

[203] Appeal Bundle, pp C545-546, at [5] and [6].

[204] Appeal Bundle, p B725H-L.

[205] This addresses Ground 7.

[206] Appeal Bundle, pp A104M-105T.

[207] Appeal Bundle, pp A111B-112K.

[208] Ibid.  See the italicised passages.

[209] Appeal Bundle, pp A198J-P and 200B-J.

[210] Appeal Bundle, pp A179T-194P.

[211] Appeal Bundle, p A198C-S.

[212] Appeal Bundle, p A198Q-T.

[213] Appeal Bundle, pp A204P-205N.

[214] Appeal Bundle, pp A203O-204T.

Other Judgments in This Case

Further hearings and rulings under CACC 309/2018