- Notable Unreported Decision
SUPREME COURT OF QUEENSLAND
R v Lang  QCA 289
CA No 294 of 2017
SC No 273 of 2017
Court of Appeal
Appeal against Conviction
Supreme Court at Brisbane – Date of Conviction: 5 December 2017 (Flanagan J)
6 December 2019
6 August 2019
Gotterson and Philippides and McMurdo JJA
CRIMINAL LAW – APPEAL AND NEW TRIAL – CRIMINAL LAW – APPEAL AND NEW TRIAL – MISCARRIAGE OF JUSTICE – PARTICULAR CIRCUMSTANCES AMOUNTING TO MISCARRIAGE – IMPROPER ADMISSION OR REJECTION OF EVIDENCE – where the appellant was found guilty of murder – where the deceased died as a result of stab wounds to her abdomen – where the appellant and the deceased were the only persons in the deceased’s apartment before she was found dead – where it was common ground at trial that the deceased’s death could only have been a result of homicide by the appellant or suicide by the deceased – where the prosecution adduced evidence from the deceased’s psychiatrist that, according to a peer-reviewed article, the suicide rate of females in Australia between 1988 and 2007 was 3.96 deaths per 100,000 population and that 3.76 per cent of those females who suicided used a sharp implement as the means – where that psychiatrist also agreed with a proposition from the prosecutor that, as the rate of suicide amongst people with bipolar was fifteen times that of the general population, the suicide rate of females in Australia was approximately 60 deaths per 100,000 – where the prosecutor in their closing submitted that the jury should reject the hypothesis of suicide because, amongst other things, the statistical evidence showed that suicide by cutting was rare – where the appellant’s counsel at trial did not object to the statistical evidence or the prosecutor’s use of it – where the appellant contended on appeal the statistical evidence was irrelevant and misleading and its use was prejudicial – whether the statistical evidence was irrelevant and misleading – whether the admission of irrelevant and misleading statistical evidence about the incidence of suicide and types of suicidal acts together with the prosecutor’s reliance upon that evidence occasioned a miscarriage of justice
R M O’Gorman with D M Caruana for the appellant (pro bono)
D L Meredith for the respondent
Fisher Dore Lawyers for the appellant (pro bono)
Director of Public Prosecutions (Queensland) for the respondent
THE COURT: On 5 December 2017, the 12th day of his trial in the Supreme Court at Brisbane, the appellant, Thomas Chris Lang, was found guilty of the murder of Maureen Lois Boyce (“the deceased”). The count on which he had been indicted alleged that the murder occurred on or about 22 October 2015 at Kangaroo Point and that it was a domestic violence offence.
The appellant was sentenced on the same day to life imprisonment. A period of some 773 days of pre-sentence custody was declared to be time served under the sentence.
On 12 December 2017, the appellant filed a notice of appeal against his conviction. At the hearing of the appeal, leave was granted to the appellant to substitute for the grounds stated in it, a single ground of appeal, namely, that the learned trial judge erred by admitting evidence of the suicide rates of Australian women. This evidence is detailed later in these reasons.
Circumstances of the alleged offending
In October 2015, the appellant, who was a medical practitioner, and the deceased were in a relationship and were living at her apartment on level 20, Macleay Towers, Kangaroo Point which was accessible only by a secured lift. Sometime before 6.30 pm on 21 October 2015, the appellant and the deceased went out. A showing of the apartment by a real estate agent was scheduled to take place that evening between 6.30 pm and 7 pm. In due course, the appellant and the deceased returned to the apartment. They were the only persons in it during that evening.
The deceased died as a result of stab wounds to her abdomen. The appellant twice telephoned emergency services to report her death at about 5.30 am on 22 October 2015. When police and ambulance officers arrived at the scene, the deceased’s body was found in her bed with a knife protruding from her abdomen. In one of the telephone calls and when police arrived at the apartment, the appellant said that he believed that the deceased had committed suicide, she having become distraught over a number of matters during the preceding evening.
It was common ground at the trial that the deceased’s death could only have been a result of homicide by the appellant or suicide by the deceased. There was no evidence suggesting the involvement of a third party. The appellant’s version precluded defences of accident or acting in self-defence in the event of homicide.
The prosecution adduced evidence in support of a motive for the appellant to have sought revenge against the deceased. The appellant suspected that she was romantically interested in one “Kenneth”. That evening he checked her mobile phone and discovered that she had been communicating with Kenneth. It was to be inferred that the appellant became enraged, threw her phone over the balcony and then stabbed her to death.
Two forensic pathologists gave evidence at the trial, Dr B B Ong called in the prosecution case and Dr J Duflou called in the defence case. A number of factors including the direction of blade tracks, the absence of blood on the deceased’s dominant hand, that the stabbing was through a sheet and the absence of any hesitation marks caused Dr Ong to favour the involvement of a second party in the stabbing, but not conclusively so.
Dr Duflou also thought that the observable factors were not conclusive either way on the issue. It was uncommon, he said, for a knife to be left behind in the event of a homicide by the infliction of knife wounds. However, he acknowledged that where two individuals only were involved, removal of a knife would tend to indicate a homicide.
Thus, on the evidence of the forensic pathologists, the reasonable possibility that the death was the result of suicide was not excluded. In those circumstances, evidence of the deceased’s mental health, its history and state at the time of her death, was highly relevant to whether the death was the result of a homicide or a suicide. We now turn to that evidence.
Evidence as to the deceased’s mental health
Dr M J Spelman, consultant psychiatrist, was called in the prosecution case on the fourth day of the trial. He was the deceased’s treating psychiatrist. She had first consulted him in 2001. Dr Spelman gave evidence that he diagnosed the deceased as having either a type 1 or type 2 bipolar disorder and a borderline personality disorder. In a bipolar disorder, the patient could exhibit changes in mood both towards depression and towards mood elevation. With type 1, during the elevated mood the patient would be manic and might need hospitalisation whereas for type 2, the patient would be hypomanic but not need hospitalisation and would have longer depressed periods.
Dr Spelman testified that the deceased had expressed suicidal ideation to him on quite a number of occasions during the period that he had treated her. That occurred when she was in a depressed state. The more severely depressed she was, the more preoccupied she was with suicidal thoughts. She would say that if she was going to suicide, she would jump off the building in which she was living.
Dr Spelman gave evidence that the deceased attended a consultation with him at the Belmont Hospital on 20 October 2015. In his opinion, she was “reasonably settled”. It was not evident to him that she was either depressed or manic. She did not tell him she was feeling suicidal. He did not believe that she was at any risk of taking her life at that time. Next, Dr Spelman was asked about medications taken by the deceased as they related to mood and about her health otherwise.
At that point in the evidence in chief, the prosecutor turned to the subject of methods of suicide. He asked Dr Spelman whether people who suffer from bipolar disorder are more likely to commit suicide. The answer was “Yes”. Dr Spelman added that it is generally accepted that patients with bipolar disorder have a five per cent risk over a lifetime of killing themselves which is higher than the prevalence in the general population “of the order of one in 10,000”.
When the prosecutor asked about figures for means of suicide, defence counsel objected on the basis that he had not been given any material to indicate that the topic would be raised. Argument ensued in the absence of the witness and the jury. The prosecutor said that he understood that Dr Spelman would give evidence that there was no correlation between bipolar disorder and any particular means of suicide. The learned trial judge indicated that he would allow that evidence. He said that if Dr Spelman was to be asked about the incidence in general of suicide by cutting, then he should reduce any opinion he had to that effect to writing supported by the relevant literature and that it should be provided to the defence.
The prosecutor then sought a brief adjournment in order to speak to Dr Spelman. When the hearing resumed, the prosecutor informed the learned trial judge that Dr Spelman would give evidence that there was no particular correlation between bipolar disorder and the means by which suicide was effected. However, Dr Spelman had informed the prosecutor that he had a paper based on a long-term study on the means of suicide adopted by persons which, he understood, showed that the rate of suicide for women is considerably lower than for men and that it is more prevalent for women to commit suicide by overdose than by a violent means such as cutting or shooting. By oversight, Dr Spelman had not mentioned the paper when he was questioned about relevant literature by the appellant’s legal representative at the committal. On the footing that defence counsel would be given a short statement of the evidence that the witness would give concerning means of suicide, continuation of Dr Spelman’s oral testimony was stood over to the following day.
When Dr Spelman was recalled, the following exchange occurred between him and the prosecutor:
“Now, have you had reference to a study of suicide in Australia, a report – a paper of analysis of rates and methods of suicide between 2000 – sorry, 1988 and 2007?---I do, yes.
Yes. Now, is that a peer-reviewed article? Or is it from a peer-reviewed journal?---As I understand it is, but I don’t know. It’s from the Medical Journal of Australia - - -
All right?--- - - - which I – I assume so.
All right. And what is the rate of suicide amongst – as of 2007, using that – those – was that – are these are conclusions using that 20-year study?---That’s correct.
All right. What’s – as of 2007, what was the suicide rate of males in Australia?---The suicide rate of males in Australia in 2007 was 13.9 deaths per 100,000 members of the population.
And what about females?---It was 3.96 deaths per 100,000 population.”
Dr Spelman went on to give evidence that, according to the paper, hanging was the most common means for Australian women to commit suicide, followed by poisoning including overdosing, gassing, jumping from heights, shooting and drowning. He gave a percentage for each of those means. Significantly, Dr Spelman went on to say that 3.76 per cent of the women who suicided had used a sharp implement as the means. That percentage included stabbing by the instrument but it was not broken down to distinguish between “cutting” or “slicing and stabbing”. The paper was not tendered as documentary evidence in the trial.
In cross examination, Dr Spelman affirmed that the statistics of which he had given evidence, confirmed that women do in fact commit suicide using a sharp blade, by cutting or stabbing. He clarified that the paper from which he had drawn the statistics did not make reference to whether the individuals who suicided had a mental illness or not.
Defence counsel established that Dr Spelman knew of a standard diagnostic manual for psychiatrists known as “DSM”. Dr Spelman agreed that a statement in the manual that the lifetime risk of suicide of individuals with bipolar disorder was of the order of five per cent. He accepted that that risk was 15 times that for the general population and that “certainly the lifetime risk of suicide of individuals with bipolar disorder is substantially raised”. Dr Spelman was examined in some detail about the following passage in DSM-5 which defence counsel read into the record:
“The lifetime risk of suicide in individuals with bipolar disorder is estimated to be at least 15 times that of the general population. In fact, bipolar disorder may account for one quarter of all completed suicides. A past history of suicide attempt and a per cent days spent depressed in the past year are associated with a greater risk of suicide attempts or completions.”
Dr Spelman was re-examined. At the commencement of the re-examination the following exchange occurred:
“(Prosecutor): Yes. Just dealing with the most recent of your evidence, was the effect of what that that DSM-IV entry was that – was it that people with bipolar disorder were 15 times more likely than the general populace to commit suicide?---That’s what it said, and I have got [indistinct] to disputed [indistinct]
Right. Now, the statistic you gave us before for suicide rate in Australia was for – for females was roughly – was 3.96, but roughly four in 100,000?---That was what was reported in that study - - -
Yes?--- - - - in 2007, yes.
So 15 times that is 60 in 100,000. Just in rough figures?---Ye – that’s – that – that’s – if you multiply four - - -
Right, yes?--- - - - by 15, you get 60. It’s a - - -
Yes?--- - - - complicated – I’m not sure how that actually is – specifically translates.
Right?---Yes, because that – that – that rate is still very low.
That’s still very low. That’s 60 in 100,000 or six in 10,000?---That’s right.
(Prosecutor): So it’s still quite rare?---If we use the – if we use the – the figures from DSM-V - - -
Yes?--- - - - and try and then match them against the – the popula – the – the direct figures from that study, that gives you a figure of six – 60 in 100,000. That seems – that still seems low to me.”
In his address, defence counsel referred to his cross examination of Dr Spelman. He read from the transcript the affirmation and clarification given by Dr Spelman to which we have referred and also Dr Spelman’s answers given with respect to DSM-V.
The prosecutor in his address reminded the jury that Dr Spelman’s evidence was that the deceased had often spoken of suicide but had never mentioned using knives. During the long periods of depression she had suffered, she had not committed suicide. In the course of addressing factors which the prosecutor submitted would cause the jury to reject suicide as an explanation for the death, he said that “suicides are rare and knifing, cutting are rare amongst them” and that the deceased had no history of cutting.
During the summing up, the learned trial judge instructed the jury as follows:
“Dr Spelman’s opinion was that, on 20 October 2015, he did not believe that Maureen Boyce was at risk of taking her own life at that time.
To the contrary, there were reasons why she would not. According to Dr Spelman, while Maureen Boyce had talked about suicide previously, she clearly had endured many years of being mentally ill and depressed, and had not suicided. She was looking forward to the birth of her granddaughter in February of the following year, and she was concerned for her husband’s health. Dr Spelman had no recollection of Maureen Boyce ever expressing thoughts of self-harm by a knife or sharp implement. He opined that the suicide rate of females in Australia between 1988 and 2007, was 3.96 deaths per 100,000 population. The most common form of suicide was hanging, followed by poisoning, including overdose and carbon monoxide poisoning, then jumping from a height, shooting, drowning.
The percentage of those females who have suicided by using a sharp implement was 3.76 per cent. That percentage does not, however, draw any distinction in statistics between cutting or slicing and stabbing. These figures would include people with mental health problems. Dr Spelman did not recall ever being told about the 2009 suicide attempt. In cross-examination he accepted that, if the deceased was unable to reduce financial stress to enable Graham Boyce to return to Brisbane, she would be at a somewhat greater risk of suicide. He also agreed in cross-examination, with a study that suggested the lifetime risk of suicide in individuals with bipolar disorder, which was suffered by the deceased, is estimated to be at least 15 times that of the general population.”
His Honour then reminded the jury that defence counsel had emphasised that Dr Spelman had agreed that the lifetime risk of suicide is higher for persons who suffer from bipolar disorder.
There was no objection to these directions. Nor was there a request for redirection on those matters.
Before turning to the ground of appeal, we note that the Crown also adduced evidence of the physical layout of the scene and from members of the deceased’s family and a friend of hers concerning recent interactions that they had had with her which were relevant to her mood at the time. It is unnecessary to elaborate upon that evidence for present purposes. No issue is raised as to its relevance.
The ground of appeal
In oral submissions, lead counsel for the appellant who, with her junior, were not defence counsel at trial and who appeared pro bono in the appeal, identified as the evidence the subject of the appeal, Dr Spelman’s evidence of the general rate of stabbing suicides against Australian women. It was “irrelevant”, “highly misleading” and “occasioned further unfairness to (the appellant)”.
The ground of appeal as formulated attributes error to the learned trial judge in admitting the subject evidence. Presumably, that formulation was intended to engage the ground in s 668E(1) of the Criminal Code (Qld) that there had been a wrong decision on a question of law by his Honour in that he had wrongly decided that the evidence was admissible. However, there was no objection to the admissibility of the subject evidence and hence his Honour was not called upon to rule whether, as a matter of law, the evidence was admissible or not.
In the circumstances, we think it is appropriate to treat the ground of appeal as being that the admission of irrelevant statistical evidence about the incidence of suicide, and various types of suicidal acts, together with the prosecutor’s reliance upon that evidence, occasioned a miscarriage of justice in that it is reasonably possible that the same may have affected the verdict.
The appellant submitted that the subject evidence was irrelevant and hence inadmissible. That was because, even considered with other evidence, it was not capable of assisting the jury to determine whether there was a reasonable possibility that the deceased had suicided by stabbing. In support of this submission, reliance was placed on the decision of the Court of Appeal in the United Kingdom on reference in R v Clark.
The evidence was prejudicial to the defence, it was submitted, because the jury having heard it, might have been more inclined to discount the possibility of death by suicide than was warranted by the admissible evidence in the case. The jury was encouraged to rely on the evidence when the prosecutor in his address spoke of suicides being rare and of knifing and cutting being rare amongst them.
Further, the learned trial judge had repeated the subject evidence in the course of outlining reasons why, in Dr Spelman’s opinion, the deceased was not at risk of taking her own life at the time. Moreover, the jury were not warned against reasoning in a way that is sometimes called “the prosecutor’s fallacy”, namely, that because suicide is rare in the general population and stabbing is rare as a means of suicide, the deceased is more likely to have died by homicide than by suicide.
Counsel for the appellant further submitted that the omission by defence counsel to object to the subject evidence could not have been because of a forensic decision on his part to exploit it. The omission must have been inadvertent.
Lastly, we note that in oral submissions, counsel maintained as a proposition that statistical rates of any events which may or may not be offences ought never be admitted into a criminal trial to prove that it was more or less likely to have occurred in a particular instance.That proposition was not developed in written submissions by either party. Nor is it embraced by the ground of appeal which is limited to the subject evidence.
The respondent submitted that the subject evidence given by Dr Spelman was relevant and admissible. From his initial contact with emergency services, the appellant had sought to characterise the death as suicide. The defence had foreshadowed as early as the committal that it would pursue an argument that because the deceased had a bipolar disorder, she was more likely to die by suicide than a member of the general population. That argument was, in fact, pursued in evidence adduced during the cross examination as to the lifetime risk being at least 15 times that of the general population and in defence counsel’s address.
The respondent submitted that evidence of that kind rendered admissible evidence which addressed the frequency with which particular means of suicide had been adopted by women. Further, the respondent submitted, these circumstances distinguished the case from that in Clark. The distinction was illustrated by observations in another decision of the Court of Appeal in the United Kingdom on appeal in R v Cannings. Here, the jury was not left with statistical information alone as conclusive of the issue of cause of death.
At no point was the jury invited to reason that suicide is rare and therefore it must have been murder, or that stabbing is a rare means of suicide for women, therefore it must have been murder. The statistical evidence was one factor amongst many others on which a rejection of suicide was urged in the prosecution case.
Before discussing the significance or otherwise of this evidence, something should be said about that part of it which was given in re-examination. Dr Spelman’s evidence, in relating to the jury what was contained in the published paper, was that there was a rate of suicide of 3.96 deaths per 100,000. Although this was not explained to the jury, the apparent effect of that statistic was that this was, on average, the number of females who had suicided per year over the period of the study, for every 100,000 of the female population. On that basis, it represented that from a population of, say, 10 million females in Australia, there had been about 400 suicides per year. However, the prosecutor’s suggestion to Dr Spelman, with which he agreed, was that if the rate of suicide by persons with bipolar disorder was 15 times higher than that of the general population, it followed that the rate of suicide was about 60 per 100,000, or in our example, about 6,000 females from a population of 10 million. The prosecutor’s suggestion, and Dr Spelman’s acceptance of it, was incorrect, because the group of 400 in our example would include those who had a bipolar disorder. That error, however, was not the real difficulty with this evidence.
The real difficulty was that the statistical evidence, upon which the prosecution relied, even when considered with the other evidence, had no significance. For present purposes, let it be accepted that the evidence indicated that a female (regardless of age, ethnicity, socioeconomic level or other considerations) was very unlikely to commit suicide, even if suffering from a bipolar disorder. Taken at its highest, the evidence might provide some indication of the improbability that a female, with bipolar disorder, would eventually take her own life. Even for that purpose, the evidence was incomplete and inexact. But it may be accepted that the evidence indicated that a death, by a self-inflicted knife wound, would be a rare occurrence, even for a woman with bipolar disorder.
But in this case, something rare did occur. A death caused by a knife wound is itself a rare occurrence, whether the wound is self-inflicted or otherwise. This event having occurred, there were two hypotheses which had to be considered by the jury. The prosecution case was that it was very rare for a woman, even with bipolar disorder, to kill herself, particularly by stabbing herself, which, whilst not of itself proving the fact, made it more probable that she had been murdered. The fallacy in that argument is that the evidence signified nothing about the relative likelihood of the alternative hypothesis, namely that she was murdered.
In this Court, counsel for the appellant sought to adduce new evidence, in the form of statistics of rates of homicide, for the purpose of demonstrating that, in effect, death by homicide is even rarer than death by suicide. It is not necessary to admit that evidence in order to understand the insignificance of the statistical evidence which was admitted at the trial.
In a case where the jury had to consider two alternative hypotheses, evidence which was said to indicate the likelihood of one hypothesis could mean nothing without evidence of the likelihood of the alternative. It was not as if there was evidence that, within a reliable sample of cases where death had been caused by a knife wound, the knife had been used by the deceased in so few cases that, considered with the other evidence, the possibility that it occurred here was almost incredible. Consequently, there was no logical way to use this evidence to add to the prosecution case.
For the respondent, it is argued that at no point was the jury invited to reason, from this evidence alone, that the appellant was guilty. It is submitted, correctly, that in the way in which the prosecution presented its case to the jury, this statistical evidence was only part of its case. However, as we have said, this evidence had no probative value, when considered with the other evidence. This was not statistical evidence which, taken by itself, would be insignificant, but which would be significant when put with other evidence.
It is not submitted for the appellant that, absent the statistical evidence, the evidence was insufficient to make it open to the jury to convict. The appellant’s argument is that the misuse of the statistical evidence may have denied the appellant an acquittal, so that there was thereby a miscarriage of justice, requiring the conviction to be set aside and a new trial ordered. In our conclusion, that argument should be accepted. The jury may have been persuaded by the evidence, especially when it came from a professional witness who was referring to a published paper in the Medical Journal of Australia. It is true that some of this evidence was adduced by defence counsel, and that no application was made to exclude the evidence or for a direction to be given to the jury that it was to be ignored. But those things do not mean that there was no miscarriage of justice from the real possibility that irrelevant but misleading evidence was used to convict the defendant.
We would order that the application to adduce evidence be refused, the appeal be allowed, the conviction be quashed and the appellant be re-tried.
 AB 1214-1216.
 Appeal Transcript (“AT”) 1-2 ll19-29. This was the new ground stated in the Appellant’s Outline of Submissions (“AOS”) paragraph 1.
 Transcript of 000 phone call at 5.33 am: AB 958 l50 – AB 959 l10.
 AB 252 Tr 4-4 ll35-40.
 Ibid ll44-47.
 AB 253 Tr 4-5 ll3-27.
 AB 254 Tr 4-6 ll44-46.
 AB 255 Tr 4-7 ll1-3.
 Ibid ll16-17.
 AB 261 Tr 4-13 l39 – AB 262 Tr 4-14 l8.
 AB 266 Tr 4-18 ll36-46.
 AB 267 Tr 4-19 ll1-15.
 AB 270 Tr 4-22 ll11-18.
 AB 271 Tr 4-23 l39 – AB 272 Tr 4-24 l14.
 AB 351 Tr 5-12 ll9-26.
 Ibid l28 – AB 352 Tr 5-13 l5.
 AB 352 Tr 5-13 ll7-11. This and other percentages had been calculated by Dr Spelman from figures for each means of suicide by women that were given in the paper.
 AB 373 Tr 5-34 ll35-45.
 AB 374 Tr 5-35 ll12-15.
 AB 375 Tr 5-36 ll1-39.
 AB 375 Tr 5-36 l26 – AB 377 Tr 5-38 l43.
 AB 378 Tr 5-39 l42 – AB 379 Tr 5-40 l33.
 AB 729 l9 – AB 732 l17.
 AB 753 ll27-31.
 AB 755 ll2-3.
 AB 833 l42 – AB 834 l19.
 Ibid ll32-33.
 AT 1-9 ll7-13.
 Dhanhoa v The Queen  HCA 40; (2003) 217 CLR 1 per McHugh and Gummow JJ at .
  EWCA Crim 1020.
 AOS paragraphs 18-20.
 Ibid paragraphs 21, 22.
 AT 1-18 l9 – AT 1-19 l4.
 AT 1-39 ll21-24.
 AT 1-48 l45 – AT 1-49 l26.
  EWCA Crim 1.
 AT 1-53 ll18-27.
 For an example of that kind, see R v Doheny and Adams  1 Cr App R 369 at 372-373 per Phillips LJ.
 Dhanhoa v The Queen  HCA 40; (2003) 217 CLR 1.
- Published Case Name:
R v Lang
- Shortened Case Name:
R v Lang
 QCA 289
Gotterson JA, Philippides JA, McMurdo JA
06 Dec 2019
- White Star Case:
No Litigation History