R v McNeill

Dr Melton's second report 11 Dr Melton said that her second report was intended to provide a shorter explanation of the mitochondrial DNA analysis contained in her first report. However, it did not replace or supplement that first report. The conclusion in her second report was stated in essentially the same terms. She said: "The analysis cannot exclude Janelle Patton and her maternal relatives as donors of the two hairs tested as 2576Q1 and 2576Q4/2576Q5."

Dr Melton's email of 13 December 2006 12 It was clear from the evidence given by Detective Sergeant Robert Peters and Ms Brooks on the first voir dire (see R v McNeill (Ruling No 1) [2007] NFSC 2) that they had difficulty in understanding Dr Melton's opinion regarding the mitochondrial DNA. That difficulty appears to have been shared by the Crown which sent a request for clarification to Dr Melton, on 12 December 2006. In an email dated 13 December 2006, Dr Melton set out her responses to the specific questions posed by the Crown. 13 Among the questions asked, the Crown sought an explanation of the meaning of the term "mixed site", and its causes. That was provided. The Crown then asked whether Q4 and Q5 were in fact different portions of the same hair, or the same portion of hair tested twice. Dr Melton replied that these were different portions of the same hair. The Crown next asked for an explanation of the mechanism whereby two pieces of the same strand of hair showed different mixed sites. Dr Melton said that this was probably because the hair had been damaged or degraded, presumably by environmental factors. 14 Importantly, the Crown then asked the following question: "Does the fact that they show different mixed sites in any way detract from our ability to say that they have the same profile on [sic] Janelle Patton's mitochondrial DNA (mtDNA)?" 15 That question met with the following response: "I cannot emphasize enough that you must be very careful in stating that these are Janelle Patton's hairs. I would be able only to state "she and her maternal relatives are not excluded as donors of the hair" although it is highly tempting to be more conclusive than that based on all my experience with this kind of testing. The presence of these mixtures keeps me from being more adamant. It's difficult, I know, and you have my sympathy. Everything points to these being her hairs, but you would not want to risk an appellate decision overturning the case if I were to "overinterept" based on all the biological likelihoods!" 16 Dr Melton was asked to explain what she meant by a "consensus profile". She said that she used the terms "basic" and "consensus" interchangeably, in order to deal with ambiguous DNA sequences that seemed to share many similarities, but included an occasional site that did not fit. She explained that this variation might be the product of some form of contamination. 17 She was asked whether the fact that Q4 had the same nucleotide results, at the eight nucleotide positions, as the reference blood sample meant that this could be called a "match" despite the presence of additional mixed sites. She said that mitochondrial DNA experts did not use terms like "match" because juries might read more into that language than they should. She said experts in her field used the terminology "cannot exclude". 18 Dr Melton was asked to elaborate upon the nature of the database that she had searched. She explained that it was composed of samples most appropriate to North American populations and included a mixture of sequences from Caucasians, Africans, Asians, Hispanics, and Native Americans. 19 When asked whether she could give any probability as to the likelihood of a given profile occurring in the "general profile", Dr Melton said that she did not do any statistical calculations when mixtures were present. She said that there were similarities between the hairs and the blood sample that would not allow her to exclude the hairs as coming from Janelle Patton, but essentially she could go no further than this.

The voir dire 20 Mr Garling submitted that opinion evidence presented in this form was irrelevant, and therefore inadmissible. He referred to s 56(2) of the Evidence Act 2004 (NI) ("Evidence Act 2004"), and relied upon the fact that Dr Melton had used a North American database which had not been shown to be applicable to the population of Norfolk Island. 21 Alternatively, Mr Garling submitted that Dr Melton's evidence should be excluded in the exercise of the Court's discretion under s 135(a) and (b) of the Evidence Act 2004. He initially submitted that the probative value of her evidence was substantially outweighed by the danger that it might be unfairly prejudicial to the accused. However, he subsequently clarified that submission and indicated that by "prejudicial" he meant the danger separately identified in s 135(b) that the evidence might be misleading or confusing. 22 Because I had great difficulty in understanding Dr Melton's two reports, and her replies by email to the questions posed by the Crown, I was uncertain what evidence she would actually give if she were called to testify before the jury. Mr Howard SC informed me that he had spoken to Dr Melton after her arrival on Norfolk Island, and he understood that she could "clarify in more detail" the conclusions set out in her various written comments. I therefore determined that there should be a voir dire. 23 In the course of that voir dire Dr Melton first indicated that she had had the opportunity, since arriving on Norfolk Island, to read reports prepared by Dr John Buckleton and Susan Vintiner in connection with this trial. She had also had the opportunity to read an article by Dr Buckleton concerning specific genetic testing that had been done on the Norfolk Island community in relation to cardiovascular disease. She said that in the past she had studied the mitochondrial DNA profiles of Polynesian populations. She was also familiar with the mitochondrial DNA profiles common to, and expected of, various racial groups including Africans, Asians and Caucasians. 24 Dr Melton next explained that she had used a "consensus profile" which showed the common findings at particular nucleotide positions. She stated that the profiles obtained for hairs Q1, Q4 and Q5 had many things in common. However, they also had some slightly different mixed sites. Dr Melton said there could be two explanations for the different mixed sites. The first would be contamination, and the second would be that the mitochondrial DNA had been damaged in the time after the hairs were initially deposited. In her view, the second explanation for the mixed sites was to be preferred. 25 Dr Melton agreed that there was no mitochondrial DNA database in Australia or New Zealand. It was for that reason that she had used the North American database. She explained that given what she knew of the genetic make-up of the Australian and New Zealand population, she would have expected similar levels of diversity in the Caucasian population as were found in North America and Europe. 26 She said that she detected either an Asian or African element in the mitochondrial DNA profiles that she obtained. She said that she would not have expected to have obtained such a profile from many members of the Norfolk Island community. In her own words, Dr Melton said (at Tx 21): "We don't want to overstate the constraints of what we know, but my anticipation is, based on the fact that when you sample any population you have high diversity, that we would not expect to find this profile at extremely high frequency. We cannot put a statistic on it because we don't have a representative database, but we would say that these three hairs have the same type as Janelle Patton and, as in all DNA testing, the language "cannot exclude" is used - " 27 When asked to clarify what she meant by the term "cannot exclude", Dr Melton continued (at Tx 21): "Well it means that there is a pool of individuals that has the type that Janelle Patton has, including her maternal relatives, her known maternal relatives, and in theory people who are not apparent maternal relatives somewhere in the population. If you think about your own maternal lineage, if you go back five or six generations, you don't know who your maternal relatives are. There could be in theory people in the world with your mitochondrial DNA type. The way we place a context on that is by understanding what databases of what mitochondrial DNA types look lie [sic]. Are there pools of types in various populations. Are there types that are unique to one population and not to others. And so we say she's not excluded. By that we mean there is a pool of people that has her type." 28 Dr Melton was then asked the size of the pool. She said it was "very small". When asked what that meant, she replied that generally speaking, mitochondrial DNA produced frequency results that were very low, of the order of one to two percent or less. When asked why she persistently spoke in terms of an inability to exclude, rather than saying that it was highly likely that the hairs came from the same maternal line, she replied that this was historically the way in which mitochondrial DNA results were reported. 29 Dr Melton said that mitochondrial DNA was not, and never would be, as discriminating as nuclear DNA. However, she added that it was more discriminating than blood typing. She declined to be drawn into any statistical evaluation, insisting that all that she could say was that "some percentage of the population" would be expected to have this profile. She said there were two reasons why she would not proffer a statistic. The first was that she had no specific database applicable to Norfolk Island, although she said other available databases all had approximately the same high diversity make-up. The second was that the hairs that she examined had mixed sites, and mitochondrial DNA experts did not do statistics on mixtures. 30 It is fair to say that, at various points during the course of her evidence on the voir dire, Dr Melton was prepared to say that the particular profile that she had discerned was not likely to be common. She agreed that she was basing that opinion upon her knowledge of the historical genetic make-up of "this part of the world". She said that she did not believe that she would find this profile in "20, 30 and 40 per cent of the population". However, when pressed as to whether she could or would go further if permitted to give evidence before the jury, she returned to the mantra that she would only be able to state that Janelle Patton and her maternal relatives could not be excluded as donors of hairs Q1, Q4 and Q5. She said (at Tx 27): "That is the appropriate terminology for mitochondrial DNA. I'm very careful to explain to juries that we can never say for sure these are her hairs, but that these hairs do have the profile that she has and that this is how many times we saw this profile in our database and this is the ethnicity determination of these profiles, and that's basically what I would say." 31 Dr Melton agreed, under cross-examination, that she had never used terms such as "highly unlikely" when dealing with mitochondrial DNA that contained mixed sites. She said that in such cases she would say no more than that she "could not exclude". She said that she would be prepared to tell the jury that Ms Patton had rather unusual ethnicity (there being some Asian or African influence in her mitochondrial DNA). However, she repeated that she would only tell the jury that these were the ethnicities associated with this profile and "as in all mitochondrial DNA cases … allow the jury to make their own conclusions regarding how they would like to interpret that". 32 Dr Melton added (at Tx 31): "All I can do is answer questions to the best of my knowledge and honestly, and the terminology of mitochondrial DNA testing here and everywhere I'm aware of is that you cannot exclude, and then to provide a context for that failure to exclude. So that's what I intend to do." 33 Her evidence continued as follows (at Tx 32-34): "A. I would say that we cannot place a statistical likelihood on this because we do not have currently a representative database for this part of the world. However, the profile is the same as Janelle Patton's profile, the consensus profile is, and therefore we cannot exclude, and that is the nature of all mitochondrial DNA evidence. Q. Would you say it's 99 per cent likely to be hers? A. That's not the way mitochondrial evidence is given. Q. Would you say it's 60 per cent likely to be hers? A. I cannot say that. Q. Would you say it's 50 per cent likely to be hers? A. No. We'd never give evidence that way in mitochondrial DNA testing. Q. I am just testing what you can and cannot say, doctor. The next question I want to ask you is the expression "she and her maternal relatives", a common expression in mitochondrial DNA for the reasons that you explained, that it's passed down the maternal line? A. Yes. Q. How many people are we talking about in that group - she and her maternal relatives? A. I don't know. Q. It could be thousands? A. Unlikely for the known maternal relatives to be thousands, although it's possible. I don't know. Q. Can you give me a maximum number it might be? A. No. Q. Could you give me a minimum - I suppose you can give me a minimum number: a mother and daughter would be the minimum, I suppose? A. That's correct, and obviously we would consider all the other factors in the case as well, as mitochondrial DNA never stands alone, it stands in addition to other information in the case. Q. Doctor, you are only giving the bit about mitochondrial DNA? A. That's correct. Q. We can leave the rest of the case to the jury. A. Correct. Q. What I want to know is if I asked you how many people are in the group of Ms Patton and her maternal relatives, you couldn't give a number? A. No. Q. All you can say is it is potentially a large number, but you don't know? A. Or potentially a very small number. Q. You just have no idea? A. I do not know. Q. Is it right to understand that the same mitochondrial DNA can go back many generations? A. Many generations, yes. Q. Is there a limit, to your knowledge, on the number of generations it can go back? A. No. Q. So it is an unlimited number of generations? A. Correct. Q. There are some studies that have gone back, to your knowledge, at least 24 or 25 generations, aren't there? A. That's correct. Q. Used in that term by scientists, is a generation thought to be 20 years or -- A. Approximately 20 to 25 years. Q. Twenty to 25 years? A. Yes. Q. So if we're talking of 25 generations, somewhere between 500 and 600 years; is that right? A. Yes." 34 For completeness, I should add that Dr Melton was re-examined and then recalled, at the request of the Crown, in an effort to elicit from her a more precise answer to the question of what she would say to the jury if allowed to give evidence. Her evidence did not take the matter further. She insisted that she would be forced to say that she could not exclude the possibility that the hairs in question came from Janelle Patton, or her maternal relatives, using the same terminology as she did in all mitochondrial cases. She said that she could not give any indication of how frequently she would expect to find a profile such as that discerned in the context of Norfolk Island. She said (at Tx 48) that she could not give a number of any kind, but would say that it would "not be likely to see this profile endemic on the island in high frequency". 35 Despite Mr Howard's best efforts, Dr Melton persisted in using language of this kind when asked specifically what she would say to the jury if her evidence was admitted.

The competing submissions 36 After Dr Melton had given evidence on the voir dire, Mr Garling submitted that any opinions that she might express were still unfathomable. He repeated his submission that her opinion, couched as it was in the language of "inability to exclude", was irrelevant. He also repeated his submission that her evidence should be rejected in the exercise of the Court's discretion under s 135(b) because it would be likely to confuse a jury, and therefore be unfairly prejudicial to the accused. 37 Before Dr Melton was called, Mr Howard had submitted that the jury would understand from her evidence that mitochondrial DNA profiles were far more discriminating than blood types, and that the common profile across hairs Q1, Q4 and Q5 and the blood sample taken from Janelle Patton indicated a close alignment to a maternal line back in the distant past which was either African or Asian. He acknowledged that Dr Melton could not say how many generations back that African or Asian element may have been introduced, and that it could go back as far as "time immemorial". He submitted that Dr Melton knew enough about the genetic make-up of the population of Norfolk Island based upon what she had read of Dr Buckleton's work. 38 After hearing her evidence on the voir dire, Mr Howard submitted that the jury would not be misled or confused by anything that Dr Melton might say. He referred to R v Lisoff [1999] NSWCCA 364 in which the New South Wales Court of Criminal Appeal held that the fact that scientific evidence was complex did not mean that to leave to such evidence to a jury would result in unfair prejudice within the meaning of s 137 of the Evidence Act 1995 (NSW).

Relevant Legal Principles 39 Section 135 of the Evidence Act 2004 provides as follows: "The court may refuse to admit evidence if its probative value is substantially outweighed by the danger that the evidence might: (a) be unfairly prejudicial to a party; or (b) be misleading or confusing; or (c) cause or result in undue waste of time." 40 This section replicates the same provision in both the Evidence Act 1995 (Cth) and the Evidence Act 1995 (NSW). There has, however, been limited judicial consideration of the term "misleading or confusing" in the context of s 135(b) of either of those Acts. 41 In R v BD (1997) 94 A Crim R 131, Hunt CJ at CL observed (at 139) that the meaning of the term was "self-evident". 42 In its report which led to the enactment of the uniform evidence law, the Australian Law Reform Commission explained that s 135(b) might apply where there was a danger that a court might incorrectly assess the weight of a particular piece of evidence. Stephen Odgers, in Uniform Evidence Law (7th ed, Lawbook Co, 2006) pp 623-624, provides a number of examples of cases where s 135(b) was sought to be invoked. In the context of jury trials see: R v Singh-Bal (1997) 92 A Crim R 397 at 403-405 per Hunt CJ at CL; and Reading v Australian Broadcasting Corporation [2003] NSWSC 716 at [26]-[33]. In judge alone proceedings see: Hughes Aircraft Systems International v Airservices Australia (1997) 80 FCR 276; and Re GHI (a Protected Person) [2005] NSWSC 466 at [8]. 43 Mr Garling did not rely upon s 137 of the Evidence Act 2004, which provides that in a criminal proceeding the court must refuse to admit evidence adduced by the prosecutor if its probative value is outweighed by the danger of unfair prejudice to the defendant (for a discussion of s 137 see R v McNeill (Ruling No 1) at [295]-[304]). However, he did submit that by reason of the misleading and confusing nature of Dr Melton's evidence, s 135(a) would also be triggered, and the Court should, in the exercise of its discretion, exclude her evidence. 44 Evidence is not unfairly prejudicial to a party merely because it tends to damage that party's case or support the case of its opponent. Unfair prejudice means that the trier of fact may use the evidence to make a decision on an improper basis by, for example, giving that evidence more weight than it should be given. In a criminal trial, for evidence to be "unfairly prejudicial" there must be a real risk that it will be misused by the jury in some unfair way: see generally R v BD per Hunt CJ at CL at 139. 45 Accordingly, there is likely to be some overlap between s 135(a) and (b). Evidence which is potentially misleading and confusing may also, thereby, be unfairly prejudicial to an accused. Arguably, such evidence can trigger the operation of s 137 as well as s 135.

Conclusion 46 I was not persuaded by Mr Garling's submission that Dr Melton's evidence was irrelevant. Properly explained, and properly understood, it had the capacity to rationally affect (directly or indirectly) the assessment of the probability of the existence of a fact in issue in this proceeding: Evidence Act 2004 s 55(1). Dr Melton's lack of familiarity with the ethnic make-up of the population of Norfolk Island did not render her evidence inadmissible, but was rather a matter that went to its weight. 47 In my view, however, Mr Garling's submissions regarding s 135 were well founded. Dr Melton's two initial reports were all but incomprehensible. Regrettably she continued to express herself, throughout the voir dire, in a manner that a jury would have had great difficulty in understanding. There was, therefore, a danger that some members of the jury might have ascribed to her evidence a particular meaning which it was not intended to have. The problem lay with the manner in which she sought to convey her opinions, adhering strictly to a series of scientific conventions which would have little real significance for anyone outside her narrow scientific community. 48 Dr Melton was understandably conservative and cautious when it came to expressing her opinion as to the significance of the apparent match between the mitochondrial DNA profiles that she detected from the questioned hairs, and the profile taken from Janelle Patton's blood sample. Indeed, she repeatedly declined to express any opinion in statistical terms, and eschewed the use of the term "match". She fluctuated between occasionally being prepared to say that the chances of anyone selected at random from the general population of Norfolk Island having the same mitochondrial DNA profile as the hairs found in the boot of the Honda Civic were low, to resisting even that description. In my view there was a serious risk that one or more members of the jury would have been misled or confused into thinking that her evidence actually went beyond what it did. 49 I am perfectly satisfied that Dr Melton is a highly qualified expert, and that she is a truthful and honest witness. I accept that mitochondrial DNA can be highly probative, when it comes to providing a link between a hair sample and a particular individual, if that evidence is presented in a manner that, speaking realistically, a jury can understand. Regrettably, Dr Melton's presentation of her opinions, couched in the language that she insisted upon using, would have been of little real utility in this case. Having expertise in a particular discipline is not enough. An expert must also be able to express his or her opinions in a manner that is comprehensible, and able to be understood by the trier of fact: see generally Australian Retailers Association v Reserve Bank of Australia (2005) 148 FCR 446 at [480]-[485]. 50 In the particular circumstances of this case, involving as they did mixed sites, a Norfolk Island population, and the use of an unproven database, the danger of permitting evidence to be given in this form outweighed its potentially probative value. 51 It was for these reasons that I exercised my discretion pursuant to s 135 of the Evidence Act 2004, and refused to admit the evidence of Dr Melton. I certify that the preceding fifty-one (51) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Chief Justice Weinberg.

Counsel for the Applicant: Mr D Howard SC and Mr R Holdsworth

Solicitor for the Applicant: Crown Counsel for Norfolk Island

Counsel for the Respondent: Mr P Garling SC and Mr J Morris

Solicitor for the Respondent: McIntyres Solicitors