REGINA v GALLI

• 44 -IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF CRIMINAL APPEAL

60586/00 SPIGELMAN CJ SULLY J ADAMS J

Wednesday 12 December 2001 REGINA v Juan Carlos GALLI Facts The Appellant was convicted by a jury of having sexual intercourse with Hannah Lucy Sodersten without her consent in contravention of s61J of the Crimes Act 1900. The circumstances of aggravation were that Ms Sodersten had a serious intellectual disability and was under the care of the Appellant who worked at the house where Ms Sodersten lived.

The Appellant was sentenced by his Honour Acting Judge Stewart to imprisonment for seven years, with a non-parole period of five years and three months, the sentence commencing on 6 July 2000.

The Crown case was circumstantial. One aspect of the Crown's case on which it placed particular reliance was evidence of DNA testing carried out by Mr Goetz, a forensic biologist. The Appellant submitted that the trial miscarried because references to the DNA evidence in a record of interview were improperly admitted into evidence. The Appellant also submitted that his Honour failed to give adequate directions to the jury and erred in the ones he did give. Leave to appeal against sentence was also sought. The Crown submitted that even if errors had occurred, a conviction was inevitable and that the proviso in s6(1) of the Criminal Appeal Act 1912 should be applied.

Held Per Spigelman CJ, Sully and Adams JJ agreeing

A Inadmissible evidence 1 The references in the second of the Appellant's records of interview to the Relative Chance of Paternity were inadmissible. R v Taylor (NSWCCA, 18 April 1995, unreported), R v GK [2001] NSWCCA 413 referred to. Although at the trial no objection was taken to this evidence going before the jury, leave under Rule 4 should be granted.

B Directions to the Jury i Use to be made of the DNA evidence 2 It would have been desirable for the trial judge to direct the jury concerning the approach to be taken the statistics involved in the DNA evidence. R v Duke (1979) 1 ACrimR 39, R v Milat (1996) 87 ACrimR 446 referred to. However, the failure to do so in the case was not an error of law and did not constitute a miscarriage of justice.

3 His Honour did not commit the "Prosecutor's Fallacy". However, in the circumstances of this case, it would have been desirable for the trial judge to have warned the jury about the impermissibility of this line of reasoning. R v GK (supra) referred to.

4 His Honour erred in referring to the Relative Chance of Paternity in his summing-up. No such evidence of Mr Goetz was before the jury and, in any case, such evidence was inadmissible. R v GK (supra) referred to.

ii Circumstantial evidence 5 His Honour referred to the Crown's onus on a number of occasions in the summing-up. There was no reversal of that onus suggested by his Honour in his directions regarding circumstantial evidence.

C Application of the proviso 6 The errors committed at the trial were not of so fundamental a character as to lead to the conclusion that there was hardly a trial at all. Wilde v The Queen (1988) 164 CLR 365, Glennon v The Queen (1993-1994) 179 CLR 1, Green v The Queen (1996-1997) 191 CLR 334 referred to.

7 In the circumstances of the present case, the number of persons who could have committed the crime was restricted to an exceptionally unusual degree. Given the force of the unchallenged evidence regarding the Paternity Index a conviction was inevitable and it cannot be said that the Appellant lost a chance of acquittal. The proviso should be applied. R v Dohney and Adams [1997] 1 CrAppR 369, Wilde v The Queen (supra), Glennon v The Queen (supra), Gilbert v The Queen (2000) 201 CLR 414, KRM v The Queen (2001) 75 ALJR 550; [2001] HCA 11, Grey v The Queen [2001] HCA 65 referred to.

D Sentence 8 The sentence was within the range which his Honour could legitimately impose. There was no error which could justify the intervention of this Court.

Orders Appeal against conviction dismissed. Leave to appeal against sentence granted. Appeal against sentence dismissed.

IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF CRIMINAL APPEAL

60586/00 SPIGELMAN CJ SULLY J ADAMS J

Wednesday 12 December 2001 REGINA v Juan Carlos GALLI

JUDGMENT 1 SPIGELMAN CJ: The Appellant was indicted before his Honour Acting Judge Stewart on a charge against s61J of the Crimes Act 1900, that between 3 April 1998 and 18 April 1998, at Eleanora Heights, he had sexual intercourse with Hannah Lucy Sodersten without her consent. The circumstances of aggravation were that at the time of the alleged offence, Ms Sodersten had a serious intellectual disability and was under the care of the Appellant who worked at the house where Ms Sodersten was confined. 2 The Appellant pleaded not guilty and was tried before his Honour and a jury. On 6 July 2000 the jury returned with a verdict of guilty. On 25 August 2000, the Appellant was sentenced to imprisonment for seven years, with a non-parole period of five years and three months, the sentence commencing on 6 July 2000. 3 At the time of the commission of the offence, Ms Sodersten was 26 years old and a resident of a group home operated by the Department of Community Services ("DOCS") at Elanora Heights. The Appellant was employed by DOCS as a carer at the home. Ms Sodersten was one of four people in the group home for whose care the Appellant had responsibility. The nature of Ms Sodersten's disability is such that she cannot speak or make signs to communicate in any meaningful way. 4 The evidence given by Dr Edwards for the Crown was that there had been penile penetration of Ms Sodersten as a result of which she fell pregnant during the period between 3 April 1998 and 18 April 1998. The fact that Ms Sodersten had had sexual intercourse was only confirmed on 29 May 1998, after it was found that she was nine weeks pregnant. Subsequently, the foetus was aborted. Ms Sodersten was never able to identify who was responsible for the act of sexual intercourse, nor could she give evidence in relation to that issue at the trial. 5 The Appellant started working at the home on a permanent basis in April 1993. Part of his duties required him to administer medication to the residents. Ms Sodersten had been on the contraceptive pill and the Appellant had administered that to her. He was aware that she ceased taking the pill at the end of 1997. The Appellant also was aware that she menstruated, as he changed her nappies. 6 The Appellant voluntarily gave blood samples and participated in two records of interview. No admission was made by him. The Crown case was entirely circumstantial. The first aspect of the Crown's case was the opportunity for the Appellant to have intercourse with Ms Sodersten. The allegation was that the Appellant was one of three men employed by DOCS who had access to Ms Sodersten in the relevant time period. The Appellant did not challenge this and, in fact, gave evidence that he was on night duty during the relevant period in the home, alone with the victim and other residents. 7 The second aspect of the Crown's case was the evidence given by Mr Goetz, a forensic biologist, who referred to the DNA testing carried out on the blood sample given by the Appellant and the two other male employees of DOCS, Mr Phillips and Mr Burcell, who had access to the victim over the relevant period. These samples were compared with a DNA analysis of Ms Sodersten's blood and a DNA analysis of the foetal tissue. Mr Goetz gave evidence that the tests eliminated Mr Phillips and Mr Burcell as possible biological fathers of the foetus carried by Ms Sodersten, but established the Appellant as a possible father. Further tests were conducted. According to Mr Goetz, it was 2.4 million times more likely that the Appellant was the father, rather than a person taken at random from the community. 8 No blood sample was taken from the two male patients (one now deceased) who lived with Ms Sodersten in the house. One patient suffered cerebral palsy, was profoundly intellectually disabled and was confined to a wheelchair. The other had Down's syndrome, was profoundly intellectually disabled and was blind. The house was staffed on a 24 hour basis. The patients were subject to close supervision, both at night and during the day. The bottom half of the doors of the two male resident patients were locked at night. 9 There was no issue as to consent. It was uncontested that the victim was intellectually disabled and did not have the capacity to consent. The Appellant defended the case on a limited basis founded on three issues. The first was the assertion by the Appellant of his innocence. The Appellant said that he volunteered to give blood samples because he had "had nothing to hide" and was confident that the tests would eliminate him as a possible offender. The second concerned the possibility of "statistical error" in relation to the evidence of DNA matching undertaken by Mr Goetz. The third matter relied on by the Appellant was an assertion of good character, conceded by the informant. A number of people were called to give positive character evidence for the Appellant.

The DNA Evidence 10 DNA tests were conducted by Mr Robert Goetz, the officer in charge of the Forensic Biology Laboratory in the Division of Analytical Laboratories of the Institute of Clinical Pathology and Medical Research. He prepared two reports dated 19 October 1998 and 16 November 1998. He gave evidence at the trial, first on a voir dire and, after his Honour determined that the evidence was admissible, he gave evidence before the jury. 11 In his first report, Mr Goetz said that DNA testing had been carried out on blood from the victim and from the foetus on the one hand, and each of the Appellant and the other two employees, Keith Phillips and Ian Burrell, on the other hand. The report was a three-page document of which the third page was an annexure indicating the results of the tests in tabular form. The first two pages was the written report which, relevantly, contained the following passage: "The results show that Keith PHILLIPS or Ian BURRELL cannot be the father of the foetus. The results show that Juan GALLI could be the biological father of the foetus. Statistical analysis of the results shows the following: Paternity Index 172:1 Relative Chance of Paternity 99.4% Relative Chance of Non-Paternity 0.6%". 12 It is significant to note that only the table outlining the results of the tests was tendered before the jury. The report quoted above was not tendered before the jury. Mr Goetz, however, gave oral evidence before the jury explaining the results in the table to the effect which I will outline below. 13 The second report by Mr Goetz provided further information of comparison between the samples from the victim and foetus with the blood of the Appellant. The results were set out in a table. The report to which the table was annexed concluded: "The results show that Juan GALLI could be the biological father of the foetus. Statistical analysis of the results shows the following: Paternity Index = 14330:1 Relative Chance of Paternity = greater than 99.99% Relative Chance of Non-Paternity = less than 0.01% When these results are combined with those in my previous report (dated 19.10.98) the statistics are as follows: Paternity Index = 2,400,000:1 Relative Chance of Paternity = greater than 99.9999% Relative Chance of Non-Paternity = less than 0.00001%." 14 As with the previous report, it was only the table that went to the jury, not the report. However, the reports were used by the police in a record of interview that was before the jury. 15 The Appellant consented to two recorded records of interview, the first on 9 June 1998 and the second on 25 November 1998. He made no admissions and maintained his innocence. During the course of the second ERISP the two reports from Mr Goetz - that were not tendered - were put to him in terms and he was asked to comment on them. This ERISP was played at the trial to the jury immediately before the closing of the Crown case. 16 The following questions by the police and the Appellant's answers appear in the transcript of the video which was supplied to the Court: "Q30 Yes, just where it says, The results show. A And ah, the second one, the second line? Q31 Yeah. A The results show that Keith Phillips or Ian Burrell cannot be the father of the foetus. Q32 And can you just read the - - - A The results show that Juan Galli could be the biological father of the foetus ….. sorry my reading, analysis of the results show the following. Q33 And can you tell me what those results show? A Which one, that from the top - - - Q34 Yeah. A - - - or I don't understand what ….. Q35 ….. I'll just tell you on this third page. This is showing that you could be the biological father - - - A Ah hmm. Q35 - - - of the foetus, and this is the statistics that show it. It says, The paternity index is a hundred and seventy two to one, which means you could be - - - A Mmm. Q35 - - - if there's a hundred and seventy two people, then you could be one of those. A Mmm. Q36 And the relative chance of paternity is ninety nine point four per cent and the relative chance of non-paternity is nought point six per cent. A ….. I don't understand those numbers anyway. Q37 O.K. I'll just show you - - - DETECTIVE SENIOR CONSTABLE HUNT Q38 Should I explain those, what those statistics mean. PLAINCLOTHES SENIOR CONSTABLE GILL Q39 Well it's saying, you understand that it shows that Keith Phillips and Ian Burrell cannot be the father. A Well, it say so very clear in there, yes. Q40 And that there's a chance that you can be the father. A Yeah, mmm. Q41 And it's saying that ninety nine point four per cent, that's - - - A That I, that I could be the father. Q41 - - - how much of a chance that you are the father. A - - - that I could be the father of that child. Q42 Yes, of Hannah's child. DETECTIVE SENIOR CONSTABLE HUNT Q43 What do you say about that? A Well, as I say before I've got nothing to do with it. I got nothing, nothing to do with it. PLAINCLOTHES SENIOR CONSTABLE GILL Q44 Sorry, and that as dated the 19th of the 10th,1998. There was some further tests done on your sample of blood alone. A Ah hmm. Q45 And this is the second test results. I'll just read it for you. I says the results show that Juan Galli could be the biological father of the foetus. Statistical analysis of the results show the following: The paternity index says that in fourteen thousand three hundred and thirty people, that you could be one of those. So the results have gone up higher. A Mmm. Q46 So there's an even bigger chance now that you are the father. Do you understand that. A (NO AUDIBLE REPLY) Q47 It's also saying the relative chance of paternity is greater than ninety nine point ninety nine percent. So there's less than nought point nought one per cent that you aren't the father. It also goes on to show the paternity index, sorry, then they were done again - - - A Yeah. Q48 - - - these are combined with these ones over here. That you were the only one in two million, four hundred thousand - - - A Ah hmm. Q49 - - - that you could be the father. And it shows that ninety nine point nine nine nine per cent of chance that you are the paternal father. DETECTIVE SENIOR CONSTABLE HUNT Q50 [It] says, Greater than. PLAINCLOTHES SENIOR CONSTABLE GILL Q51 And there is less than nought point nought nought nought one per cent that you aren't the father. What can you, have you got anything to say about these results? A ….. there's got to be something wrong. I haven't, as I said, I haven't done anything to her, nothing at all, I've been, as I said, I, I've been with them for over six, six years with her. Q52 And - - - A Never dream of or thought of doing anything to her. … PLAINCLOTHES SENIOR CONSTABLE GILL Q69 Is there anything else you wish to say about this matter, Carlos? A Well, I would say it again and over and over again, I got nothing to do with her pregnancy. I'm not that kind of person to, or take advantage of anyone. So that's all I got, I got to say. I've got nothing to do with her pregnancy. What, and the tests, what the tests are shown it look like I, I am the one, but I say there's got to be something wrong in there. I haven't done anything to her. Q70 So you see there's no other reason for the results showing that you are the father. Can you think of any reason? A Well, it's got to be something wrong in there." 17 This video containing these questions and answers was played to the jury immediately before the closing of the Crown case. 18 As I have indicated above, Mr Goetz gave evidence on a voir dire. It is not necessary to set out the whole of the evidence given. It is sufficient to note that in some respects Mr Goetz made observations during the course of the voir dire which he did not repeat before the jury. 19 When the Crown Prosecutor tendered the table annexed to the second report, it was tendered even on the voir dire as a one-page document, without the covering report. However, in the course of the voir dire the following evidence was given: "Q It is the combination of these two lots of results referring to your paternity index that you then establish the probability of 99.9999 per cent, is that right? A That's correct. HIS HONOUR: That's not shown here though, is it? CROWN PROSECUTOR: No your Honour, I haven't tendered that in the documentation." 20 Subsequently, in the course of cross-examination on the voir dire the following occurred: "Q So what it comes down to at the end of the day is you are saying in terms of the probabilities the real question that you are answering is what is the chance of taking somebody from the population at large and having them match this particular combination of alleles? A That's correct. Q That's not the same thing as saying that it's 99.9999 per cent certain that Mr Galli is the father, is it? A No, that's not what I'm saying, I am saying the relative chance of paternity is 99.999 per cent, I am not saying that he is 99.999 per cent likely to be the father. HIS HONOUR: What does the relative chance of paternity mean though? A It's just another way of re-wording the probability of paternity, the paternity index. If I had a paternity index of 100 to 1 that would be a relative chance of paternity of 99 per cent. It's just another way of rewording it, it really flows very much from the Family Law Act which specifies the way they want the reporting done and that is the way they want the reporting done in that case. Q This isn't a Family Law Court, this is a criminal court and the Crown has an obligation if it is to prove the case against this accused of proving it beyond reasonable doubt. Can you equate that in any way to this probability of paternity test. A No I prefer actually to use the paternity index, the 100 to 1 or 1000 to 1 or 2.4 million to 1 rather than the relative chance of paternity. I think that can sometimes be a bit misleading." 21 The reference to the relative chance of paternity and the figure of 99.9999 per cent during the course of the voir dire was not repeated in the evidence given by Mr Goetz before the jury. This may have been because of Mr Goetz's opinion that it could be misleading. 22 In his evidence before the jury Mr Goetz said: "… the quality of the DNA was very good. As a forensic biologist we're used to dealing with samples that are extremely degraded, samples that have been outside in the sun for weeks if not months, samples that are minute, samples the size of pinheads. In this case the blood samples, I had a number of mls of blood and with the foetus I had a sample that was of extremely good quality. There was no problem getting the DNA in sufficient quantities from all the samples I analysed." 23 Mr Goetz gave evidence to the jury about the nature of DNA and the conduct of tests for DNA. He gave evidence about testing for paternity by identifying the DNA of the mother and of the foetus, which could then identify what the foetus would have inherited from the father. 24 The two tables of results, being the annexures to Mr Goetz's reports of 19 October and 16 November 1998, were identified and tendered. As noted above, the covering reports were not tendered. 25 Mr Goetz referred to the first table of results of DNA testing, being the table attached to his report of 19 October 1998. He told he jury: "From my testing I could exclude Keith Phillips and Ian Burrell from paternity so they could not be the father of the foetus, however, I was unable to exclude Juan Galli from paternity." 26 Mr Goetz referred to the second table, being that annexed to his report of 16 November 1998 as containing the results of further DNA testing of samples of the blood from the victim, the foetus and the Appellant. 27 Mr Goetz explained to the jury that the columns of the tables comprised different DNA markers. He noted the presence of particular alleles in the foetus which were not present in the blood sample from the mother. They must have come from the father. He explained that the absence of such alleles in the case of both Mr Phillips and Mr Burrell enabled him to exclude them as possible fathers. The presence of the alleles in the case of the Appellant enabled him to state that the Appellant could be the biological father. 28 With respect to each DNA marker referred to in each column of the tables, the last line item in each column was a computation of a "Paternity Index". Mr Goetz explained this to the jury as follows: "Paternity Index is a calculation that is made where I have not been able to exclude someone. If I can't exclude him, exclude the possible father I need to know the significance of the inclusions. What is the probability that Mr Galli could be the biological father as compared to anybody else taken at random in the population. And to do that I take two hypothesis (sic) to calculate my paternity index. My first hypothesis is what is the likelihood that Juan Galli is the biological father and my second hypothesis is what is the likelihood that Juan Galli is not the biological father but someone else taken at random in the population is." 29 Mr Goetz then explained the significance of the figures in the table with respect to the issue of whether the Appellant was the father. He indicated that the paternity index for each DNA marker was based on the combination of the probability that he would have passed on the relevant allele to the foetus and the frequency of the occurrence of that particular allele in the general population, based on a specific database. 30 He also gave the following evidence: "Q As a combination of each of those paternity indexes for each particular DNA marker, at that point in time did you come to a combined total paternity index at that stage? A Yes I did. Q What was the result of that combined paternity index? A The combined paternity index is calculated by multiplying those numbers at the bottom together, the bottom of that column, bottom of that table together. If you multiply all those numbers together and you come up with a combined paternity index of 172 to 1. So using the 10 different DNA markers in that table, it's 172 times more likely that Juan Galli is the father of the foetus than anybody else taken at random from the population." 31 Mr Goetz went on to explain the further testing that he had done to generate he second table, with the application of the Profiler Plus system. He said that he had looked at each marker for both the foetus and the mother and then by deduction identified what was passed on to the foetus by its biological father. He testified that in each of nine DNA markers, the Appellant possessed the allele which was inherited by the foetus from its father. Mr Goetz calculated the paternity index for each of these markers. He then gave the following evidence to the jury: "Q In regard to this further testing, did you come to a further refined paternity index? A Using the profiler plus system, the combined paternity index was 14,330 to 1. HIS HONOUR: Q What does that mean? A Using profiler plus alone Mr Galli is 14,330 times more likely to be the father of the foetus than anybody else taken at random in the population. CROWN PROSECUTOR: Q So the results of the first tests were, you told us, a 172 times more likely Mr Galli was the father of the foetus than a person taken at random from the population. A That's correct. Q In regard to the second test the figure is Mr Galli was 14,330 times more likely to be the father than a person taken at random from the general population? A That's correct. Q Because these are different DNA markers and different indicators are you able to then combine those results? A Yes I am. Q Once you combine those results from a total of 18 DNA markers … what was your result there? A The combined paternity index now is 2.4 million to 1 which means that Mr Galli is 2.4 million times more likely to be the biological father of the foetus than anybody else taken at random in the population."

The Summing Up 32 In the course of his summing up, his Honour instructed the jury that the Crown case was based on circumstantial evidence. He said: "Circumstantial evidence, members of the jury, is not necessarily any less reliable than direct evidence, such as the evidence of an eye witness. Indeed, in some cases, it can be more convincing. However, before you can find an accused person guilty of a crime on the basis of circumstantial evidence, you must be satisfied that such a finding is not only reasonable but that it is the only reasonable finding to make. It follows that, if there is another finding which is reasonably open, that is, if there is another reasonable explanation which is consistent with innocence, it is your duty to find the accused not guilty, and, of course, when you consider circumstantial evidence, you do so in the light of all the other material in the case, including, of course, the material placed before you by the accused." 33 With respect to the DNA evidence, his Honour instructed the jury: "The Crown's case is that he wasn't the only male person who had access to her without anyone else who could talk about it being present. The other men are Phillips and Burrell. They were, according to Mr Goetz's evidence, eliminated as possibilities by Mr Goetz's first test, and the Crown puts it to you as part of the Crown's case that they were, in fact, eliminated." 34 His Honour outlined the expertise of Mr Goetz and his evidence about the way he conducted the tests and added: "All three men who had the opportunity to be alone with Hannah Sodersten were tested. They all had blood samples taken. The Crown put to you that Mr Goetz conducted scientific tests. These tests were not done by flight-by nights. It is a recognised area of science and Mr Goetz told you about DNA." 35 After referring to some of the detail of the expert evidence, his Honour continued: "The very first marker excluded Phillips and Burrell. There was no 1.2 in their test. The accused had 1.2 so he was a possible says Mr Goetz. He had an allele that the foetus had and he concluded that there was a 172 to one likelihood he was the father, when compared with a likelihood of a person taken at random from the community having the same markers and allele as the accused. … Now the Crown concedes that if that was where it ended it may well be that it would not be open to you to find beyond reasonable doubt that the accused was the father of that child., but Mr Goetz did not stop there, he went on to do further testing. He prepared charts or tables which are exhibit C and D. You will have them with you in the jury room. The second chart, exhibit D, indicates that Mr Goetz told you, that eight further independent tests were taken. In each the accused possessed the same allele as the foetus. These additional tests showed that the chances of a person chosen at random from the community having the same as the accused had risen from 172 to one to 14,330 to one. That is that it was 14,330 times more likely that it was the accused than anyone taken at random from the community, and he told you that when you combined the two tests, it became 2,400,000 times more likely that it was the accused rather than someone taken at random from the community." 36 His Honour also directed the jury: "In Goetz's view it is 2.4 million times more likely that it is the accused who is the father than a person taken at random from the community. He bases that, members of the jury, on this 99.9999 basis. You will remember him talking about, if it is only 95 per cent certain it goes down to 800,000 to one, but up to seven-odd million to one, and Mr Goetz takes the view, and it is his opinion and he expresses confidence in it, that the 2.4 million statistic is correct." 37 There was no point in Mr Goetz's evidence before the jury at which he used the figure of "99.9999 per cent". The figure of 95% was used once in the course of his cross-examination. The cross-examination was directed to the issue of the validity of the data bases used for the purposes of computing the frequency of the occurrence of a particular DNA marker in the general population. The cross-examiner introduced the concept of "confidence levels" which Mr Goetz identified as a margin of error based upon the sampling size of the respective databases. The following passage appears in the cross-examination: "Q Just in relation to these confidence levels you arrived at a figure which you have given in your evidence in chief without any reference to any confidence level? A That's correct. Q I understand though that in any statistical analysis that sort of figure presumably is simply a figure that falls within the range of confidence levels? A That's the middle range I have given you, I haven't given you an up confidence level or a lower confidence level, it's my best estimate. Q What would your confidence levels be in relation to -- A From my experience in dealing with hundreds and thousands of these calculations the 95 per cent confidence level which is the traditional one used usually drops the statistic usually one third one way and takes it up three times the other way. In other words, in this case you'd expect it to be approximately between 840,000 to 1 up to 7.2 million to 1. Q Focusing as I want to do on the lower range, that's 840,00 to 1? A That's correct." 38 In his summing-up his Honour has referred to the "95 per cent confidence level" in terms of "95 per cent certain" His Honour also referred to Mr Goetz "basing" the 2.4 million to one figure "on this 99.9999 basis", a figure Mr Goetz had not used before the jury. This figure had been mentioned only in the record of interview.

Grounds of Appeal 39 The Appellant has raised five grounds of appeal against his conviction. These are: "1. The Appellant received an unfair trial in that the Crown led inadmissible evidence from the police as to the questioning of the Appellant regarding his explanation for the DNA results. 2. The trial judge failed to give a proper and/or adequate direction to the jury regarding circumstantial evidence as it related to the evidence of the DNA. 3. The trial judge erred in failing to re-direct the jury as to the use to be made by the jury of the evidence of the DNA. 4. The verdict cannot be supported having regard to the evidence. 5. Generally, there has been a miscarriage of justice." 40 The Appellant also sought leave to appeal from the sentence imposed on him. 41 Each of Grounds 1, 2 and 3 turn on the evidence relating to the DNA results, either in terms of inadmissibility (Ground 1) or defects in the trial judge's directions (Grounds 2 and 3). It was not suggested that Grounds 4 and 5 have any basis independent of the earlier grounds. 42 The evidence about the DNA results from Mr Goetz is of considerable force. The Crown submits that its case was of overwhelming strength. If appealable error has occurred, the Crown submits that the Court should apply the proviso in s6(1) of the Criminal Appeal Act 1912.

The Paternity Index 43 The admissibility of the figures in the Paternity Index was conceded by the Appellant in the present case. The use of such figures was expressly approved by this Court in R v GK [2001] NSWCCA 413 at [57] and [100]. The primary sample, sometimes referred to in the cases and literature as the "crime stain", consisted of DNA tests of the foetus and of the mother from which was deduced the contribution to the foetus from the father. The use of such testing for the purposes of establishing paternity is well established. The computation and use of the paternity index is not subject to any challenge in the present case. 44 It is of central significance in the present case that the number of persons who could have had sexual intercourse with the victim was exceptionally limited. The particular circumstances of the victim were such that only three carers and two male patients were possible suspects. This is an unusually small number of suspects. Two of the suspects were themselves eliminated by DNA testing. The Appellant tested positive in all relevant respects. The other two suspects, being the two male patients, were not tested at all. On the basis of the figure of 2.4 million to one, there are three or four males in the Australian community who may have the DNA profile of the father as suggested by the DNA tests. The possibility that one of the other two or three men in Australia who may have the DNA profile suggested in the testing, would be one or other of the two patients, is so slight that the DNA testing carried particular force on the facts of the present case. 45 DNA test results identify information in the form of two alleles at each locus on the DNA strand that has been targeted. In the present case, at each of the loci targeted, there was an identity between the alleles inherited by the foetus from its biological father and those of the Appellant. The significance for the jury of this information depends upon scientific information which would not normally be known to a jury. It also depends on computation which a jury could not undertake. This is, accordingly, an appropriate subject matter for expert evidence. 46 The first level of expert evidence identifies a question of probability for the jury. With respect to each allele identified as existing in both the Appellant sample, and in the biological father: what is its frequency of occurrence in the general population? The more frequent is the occurrence, the less the coincidence points to the Appellant as the father. The degree to which this is so, however, is a matter on which a jury may require assistance. 47 A second level of expert evidence is also required in order to combine the results with respect to each allele in a manner which the jury could not do unassisted. There is a generally accepted scientific technique for combining the results to take into account, and to compute, the improbability of the co-existence in both the biological father and the Appellant of the number of different alleles that had been identified. A computation is required in order to explain the combined effect of the co-existence of all the alleles. 48 Unless evidence can be given in a numerical form, a jury is unlikely to appreciate the probative value of the DNA tests. As Callaway JA said in R v Noll [1999] 3 VR 704 at 708-709: "It is the statistical step in the reasoning that gives DNA evidence its probative value." 49 Where, as here, there is a well-established and widely recognised scientific basis for conducting the relevant computations, it is appropriate that the jury be assisted in this manner. 50 The danger in evidence of this character is that its very precision and concreteness suggests an exactness which a statistical distribution does not have. There will be circumstances in which this may mislead the jury and lead it to give the evidence greater weight than it ought be given. 51 As King CJ put it in R v Duke (1979) 1 ACrimR 39 at 41: "There may be unusual cases in which the judge has reason to fear that the jury will be overawed by the scientific garb in which the evidence is presented and will attach greater weight to it than it is capable of bearing." 52 The Appellant submitted that in this case it was essential that directions should have been given by his Honour to "balance the concreteness of the scientific evidence offered by the Crown". Neither before his Honour nor in this Court, was the nature of the direction that ought to have been given in this regard specified. The one "fact" that could balance the "concreteness" of the evidence was that on the figures themselves there were three or four persons in Australia who may have shared this DNA profile. I do not see that a direction of that character would have been of substantial assistance to the Appellant. His Honour fully summarised he submissions made for the Appellant at the trial. 53 In R v Milat (1996) 87 ACrimR 446 Hunt CJ at CL said at 452: "The jury in every DNA case should, of course, be directed not to approach the issue of chance upon any strictly mathematical basis." 54 Counsel for the Appellant submitted that a direction of that character was required in this case. 55 It may be desirable to warn a jury in this way as a general rule, in order to balance the concreteness and precision of the statistical calculations. Findings of fact in both civil and criminal cases require commonsense judgment and the tribunal of fact is required to reach a level of actual persuasion on the whole of the evidence. This does not involve a mechanical application of probabilities. (See the authorities collected in Seltsam Pty Ltd v McGuiness (2000) 49 NSWLR 262 at [136]. See also GK (supra) at [21]-[22]; the Honourable Justice D Hodgson "A Lawyer Looks at Bayes' Theorem" Australian Law Journal forthcoming and R v Adams [1996] 2 CrAppR 467 at 481 per Rose LJ.) 56 Where, as here, the mathematical computations were stark, and evidence capable of counterbalancing the numbers was limited, it would have been desirable for the trial judge to indicate to the jury that the task before them was not one capable of being reduced to a mathematical computation. However, in view of the limited range of issues in dispute, I do not believe his Honour's failure to give such a direction was an error of law or constituted a miscarriage of justice. The case for the Appellant including his denials on oath, his expectation that DNA testing would clear him and the strong character evidence were all emphasised in the summing-up. Leave is required under Rule 4. The matter is not, in the circumstances, of sufficient significance to justify leave. 57 The Appellant submitted that the trial judge had, in substance, reversed the onus of proof in his directions on circumstantial evidence. This direction is fully set out above. Immediately after giving the jury the direction that they must be satisfied that the finding of guilt on circumstantial evidence must be "the only reasonable finding to make", his Honour went on to say "if there is another finding which is reasonably open, that is if there is another reasonable explanation which is consistent with innocence, it is your duty to find the accused not guilty". 58 The Appellant submitted that the reference to "another reasonable explanation" carried the suggestion that it was a matter for the Appellant to adduce evidence capable of constituting such an "explanation". This submission should be rejected. 59 His Honour referred on a number of occasions to the onus that the Crown had and, on more than one occasion, emphasised that the Appellant had no onus of any character and was not obliged to give evidence, although he had done so. I do not believe in this context that a suggestion of a shift in the onus can reasonably be inferred. All that his Honour was doing, in the passage to which the Court's attention was directed, was to state in the obverse the primary proposition on which he directed the jury, that they had to be satisfied, in the context of a circumstantial case, that the finding of guilt was "the only reasonable finding to make". 60 The Appellant did not rely on a failure by the trial judge to expressly direct attention to the fact that the two male patients had not been tested. Although there was evidence indicating a lack of opportunity and physical difficulty for any suggestion that one or other of those patients could be responsible for the pregnancy, nevertheless the evidence was not such as to remove any reasonable possibility, as distinct from probability, that one of them could have been responsible. 61 In his directions to the jury his Honour did not identify either of the fellow patients as even a possibility. Indeed, in his directions to the jury he dismissed them as possibilities, no doubt by reason of the way the case had been run before him. 62 In his summing-up his Honour said: "The Crown puts it to you, as part of its case, that we are not talking about a stranger here, we are talking about somebody who knew Hannah Sodersten and who had unsupervised access to her." (Emphasis added) The reference to "unsupervised access" is, in the circumstances, intended to exclude the two patients. 63 To similar effect was his Honour's observation, in the context of outlining the evidence of Mr Goetz: "All three men who had the opportunity to be alone with Hannah Sodersten were tested. They all had blood samples taken." (Emphasis added) Again the reference to "the opportunity to be alone" was directed to excluding the relevance of either of the other patients. 64 It does not appear that the defence submissions at trial made any reference to the possibility of one of the other patients being responsible. No doubt his Honour proceeded to make the remarks in the summing-up on this basis. 65 Similarly, in this Court the Appellant did not identify either of the references which I have italicised in pars [62] to [63] hereof as constituting a misdirection. It appears that the trial proceeded on the assumption that neither of the other patients was a real chance. No doubt that reflected the practical reality of the situation. 66 The Crown was not able to refer to evidence which would exclude the possibility of impregnation by one of the patients. There was evidence of supervision, but nothing to suggest that such supervision was infallible. There was evidence of physical separation and some degree of physical impossibility, e.g. the bottom half of each door being locked at night and one of the patients needing a wheelchair. Complete physical impossibility was not established. 67 This is a factor which I will keep in mind when considering the application of the proviso.

Relative Chance of Paternity Percentages 68 Since the trial in the present case, this Court has considered the admissibility of DNA evidence in R v GK (supra). In that case a trial judge had ruled that the Crown was entitled to lead evidence that DNA testing showed that the accused could be the father but that there should be no statistics put before the jury. The issue came before this Court pursuant to a question for determination under s5A(2) of the Criminal Appeal Act 1912. Mr Goetz was one of two witnesses called by the Crown. The statistics in issue in that case were in the same form as Mr Goetz's reports in the present case. There were statistics for each of a Paternity Index and for a Relative Chance of Paternity. This Court held in GK that the former were admissible but the latter were not admissible. The Crown does not challenge that finding and has proceeded to make submissions in the present case on the basis that statistics for a Relative Chance of Paternity are not admissible. 69 Mason P set out reasons for admitting the Paternity Index figures at [24] to [58]. He agreed at [60] with the reasons given by Sully J for excluding the Relative Chance of Paternity percentage, with one additional reason. Sully J also found that the Paternity Index figures should be admitted. His Honour said: "[99] I am more ambivalent about the percentage quantifications of the Relative Chance of Paternity. Mr Goetz himself volunteered that he thought that the Paternity Index was the easier of the two statistical measures to understand, and I myself agree with that view. The bare percentage figures are so extraordinarily close to 100 per cent that there must be, at least, a real risk that jurors, even sensible jurors properly directed, would think that the difference between the percentage figures and 100 per cent was so miniscule that there could be no possibility that GK was not the father; when it is scientifically uncontroversial that the one thing that DNA testing cannot do is to exclude absolutely any possibility, however statistically trifling, to the contrary." 70 Sully J indicated that he would have excluded these percentage figures on the basis of each of s135 and s137 of the Evidence Act: "[100] … because of the combined effect of the considerations: [1] that the residual risk of unfairness deriving from the subliminal impact of the raw percentage figures would have been, to me, both real and unacceptable; and [2] that the admission of the Paternity Index figures would have given the Crown every fair opportunity of persuading the jury to accept the Crown contention that there was no real possibility that GK was not the father." 71 Mason P at [60] added to this reasoning the consideration referred to by Mr Goetz in evidence in that case, that the relative chance of parentage percentage was more complicated to explain. 72 The Crown accepted the reasoning in GK. Specifically, the Crown accepted that such evidence was capable of being misleading or confusing within s135(b) of the Evidence Act 1995. The figures put to the Court in the present case by the Crown were that a Paternity Index of 50 to one would convert into a relative chance of paternity of about 98 percent. This was a submission, not based on expert evidence in the case. However, if a figure of 98 percent were put to a jury, it is likely that many jurors would regard that as very significant evidence pointing to the accused, even though the Paternity Index ratio was very low, so that numerous persons in the general community could share the DNA profile. 73 By accepting the authority of GK, the Crown conceded that that part of the record of interview in the present case which made reference to the percentage figure of the Relative Chance of Paternity was inadmissible. 74 I have quoted above the passage from the summing-up where the trial judge referred to the computation that the odds were 2.4 million to one times more likely that Mr Galli was the father than a person taken at random from the general population, as being put "on this 99.9999 basis". Mr Goetz had given no such evidence before the jury, although he had on the voir dire. The only reference to "99.9999" percent was in the course of the record of interview. The Appellant relied upon this as a material misdirection. The Crown submitted that the reference in the trial judge's summing-up was merely an "unfortunate lapse" and that it was "a quite fleeting reference" and "almost a throw-away line". 75 There is nothing intrinsically wrong about a police officer putting before a suspect, evidence which suggests that the suspect's account could not be true. (See e.g. Errey v The Queen [2001] WASCA 75 at [55].) Nevertheless, leading such evidence cannot be justified if the effect is to place inadmissible material before the jury. (See R v Taylor (NSWCCA, 18 April 1995, unreported) at p12.) That was the effect here. 76 No objection was taken to the admission of the record of interview. Counsel who appeared for the Appellant at the trial may well have taken the view that it was in the Appellant's interests to have before the jury his client's unhesitating denials when first confronted with the scientific evidence against him. No evidence from trial counsel was led in this Court. However, Counsel at trial did not have the benefit of the decision in GK. 77 Leave is required under Rule 4. I would grant leave because the result was to place significant inadmissible evidence before the jury. 78 By reason of his Honour's admission of that part of the record of interview which referred to the percentage of the Relative Chance of Paternity, and also by reason of his Honour's use of that percentage in his summing-up to the jury, Ground 1 and, in part, Ground 2 in the Notice of Appeal must be upheld. Subject to the application of the proviso, the Court should allow the appeal and order a new trial. In order to assess the application of the proviso, it is necessary to consider the other issues raised in the appeal.

The Prosecutor's Fallacy 79 The evidence in the case and his Honour's directions to the jury, were expressed in terms of the combined effect of the various DNA markers tested, to the affect that it was 2.4 million times more likely that it was the accused who was the father than a person taken at random from the community. It was submitted that his Honour had committed what has come to be known as the "Prosecutor's Fallacy" in his directions to the jury. In my opinion this submission should be rejected. 80 The Prosecutor's fallacy has been variously described. (See R v Dohney and Adams [1997] 1 CrAppR 369 at 372-3; Balding and Donnelly "The Prosecutor's Fallacy in DNA Evidence" [1994] CrimLR 711 at 715 and 716). 81 The nature of the evidence given in the present case was an answer to a question in the following form: "What was the probability of the accused having the … DNA of the father compared to a person taken at random?" 82 The answer to this question in this case was 2.4 million to one. 83 The question for the jury was: "What was the probability of the accused being the father?" 84 One means of committing the Prosecutor's Fallacy is a reasoning process which treats the answer to the first question as if it was an answer to the second question, i.e. that the probability that he was the father was 2.4 million to one. That is not a permissible form of reasoning. 85 Given the size of the male population in Australia, on the basis of a probability of 2.4 million to one, there would be three or four males in Australia who share the DNA profile of the father of the foetus. A statement in the form that the "odds are 2.4 million to 1 that the accused is the offender" or that the "odds are 1 in 2.4 million that the accused is innocent", overlooks the number of people who could have committed the offence. 86 In terms of Bayes' theorem, the argument forgets that there are "prior odds", to which the likelihood ratio from the DNA tests must be applied, before computing the "posterior odds". Although Bayes' theorem is not appropriate to be applied to determine guilt or innocence (see references in par [55] above) it nevertheless represents a logical reasoning process. (See Evett et al "DNA profiling: A discussion of issues relating to the reporting of very small match probabilities" [2000] CrimLR 341 at 355.) The difficulty with its application is that most evidence in a criminal trial cannot be reduced to odds. 87 I have set out above the relevant extracts from his Honour's summing-up. In my opinion, at no point in his summing-up did his Honour commit the Prosecutor's Fallacy. Each of his references was in the form of the first question. This was a form approved in GK at [33]-[34]. 88 The Appellant also submitted that this was a case in which a warning to the jury against committing the Prosecutor's Fallacy should have been given. In GK Mason P held that on the facts of that case it would have been appropriate to give a warning of this character. His Honour said: "[59] It follows that the paternity index figures should not have been withheld from the jury pursuant to s137 of the Evidence Act . They should have gone to the jury accompanied by appropriate directions emphasising the need to avoid the prosecutor's fallacy. The appeal is not the proper vehicle to formulate a model direction in that regard. I content myself with a suggestion that it would be desirable if, at the time the judge tells the jury that evidence of paternity has been expressed as a high or very high probability, the jury should be reminded then and there that the evaluation of that evidence is a matter for them in light of the totality of the expert and non-expert evidence." 89 I do not understand Mason P to hold that a direction about the Prosecutor's Fallacy is necessary in all cases in which DNA evidence is led. His Honour indicated that it was required in that case. Trial judges are already required to give numerous directions and it is by no means clear that in all circumstances a direction concerning the Prosecutor's Fallacy would assist the jury. As McHugh J noted in KRM v The Queen (2001) 75 ALJR 550; [2001] HCA 11: "[37] The more directions and warning juries are given the more likely it is that they will forget or misinterpret some directions or warnings." 90 The question is whether or not in the circumstances of a particular case juries should be assisted by a warning not to engage in an impermissible form of reasoning. 91 In many cases a jury would be particularly assisted by a direction as to the use of probability by means of examples that are more likely to be within their own experience. Such a direction would usually avoid the need for an express warning against one particular kind of impermissible reasoning. 92 A good example of this kind of assistance is to be found in a direction to the jury in the earlier years of fingerprint evidence, when the issues raised were similar to those that have been raised in recent times with respect to DNA evidence. In R v Krausch (1913) 32 NZLR 1229, Chapman J summed-up to the jury in the following manner at 1230-1231: "In comparing prints of two individuals you may chance to find a close agreement as to a particular point, but the whole weight of scientific testimony shows that even this is rare. If the chances are five to one against this - and they are said to be much more than five - then when you come to half a dozen such points what may in popular language be termed the odds against two being alike have gone up with great rapidity, and beyond this go up with enormous strides. I need not take you into arithmetic: I prefer to ask you to deal with the matter as you would deal with any other question of probability that arises in daily life. Take, for instance, this set of facts which came before this Court in a country town some years ago. A little girl had made a complaint against a man of having interfered with her. She had described him to the police, and on the following day the accused was pointed out by her and arrested. He produced family witnesses who tried to prove that he was somewhere else at the date of the occurrence. He was a tall, elderly man who wore a grey suit. The girl stated that he had taken a penny from a yellow purse with a brown lining, and had offered it to her. When she refused this he produced a mouth-trumpet from his pocket and offered her that. When arrested the accused had a grey suit on; in one pocket was a brown purse of a common type with a yellow lining, and in another a mouth-trumpet. I pointed out to the jury that there were, so to speak, five marks of identification besides the girl's general recognition - namely, (1) a tall man, (2) elderly, (3) with a grey suit, (4) with a purse showing those two colours not described with absolute accuracy, and (5) a mouth-trumpet. I further pointed out that there were about twelve thousand inhabitants in the district, that they might discard all women and children, and all too young to be called "elderly," and all those who could not be described by a child as tall. I then put it to them: Suppose the police had on that day arrested and searched every man not excluded from the description, how many do you think would be found carrying a purse showing those two colours and a mouth-trumpet?

You are entitled to treat the matter as one of probability, or, to use a popular phrase, as a question of odds. Now, you are in a position to treat this question in the same way in which that jury was allowed to treat and did treat that question; but if you consider the evidence you have heard from the experts, and the known scientific facts, you will have to bear in mind that a grey suit or a purse or a mouth-trumpet may be readily acquired or lost, while a particular mark in a particular place in a finger-print is unalterable, though of course every finger-print has many marks.

But, after all, are we dealing with anything absolutely new or, strange? It has long been known that every hair in a man's head has its own place, and if you could make an accurate map of a few square inches of a man's scalp, fixing the place of each hair by means of fine lines of latitude and longitude, and you were then to try and find another man presenting exactly the same features in this respect, you would, I have no doubt, find it extremely difficult if not impossible. Such a fact when established has no value as a means of identification because it is not practicable to use it, but it may be and probably is as firmly established a biological fact as those relating to finger-marks." 93 His Honour concluded the summing up with the following illustration: "I take from my pocket this bunch of keys. You see that they are of various sorts and sizes, made to fit a considerable variety of locks. Suppose you were now to arrest every man in New Zealand who possessed a bunch of keys, can you suppose for a moment that you would find another man whose keys correspond in type and detail with mine to the extent of, say, one-quarter of their number? You might occasionally find one or even two which corresponded, but would you expect to find more? The proposed illustration may at first sight appear to go too far, but if the accepted scientific facts respecting finger-marks are true facts, that illustration presents a fairly close analogy." 94 As noted above the logical problem with the Prosecutor's Fallacy, in its most usual form, is that it ignores the number of persons who may have committed the offence by purporting to devise a statement as to the odds that the accused is the father from what the DNA tests indicate is the probability that the accused is the father, rather than the person taken at random from the community. Whether or not a jury is likely to be mislead into this form of reasoning depends on an assessment of the whole of the evidence and of the summing-up. In the present case, the trial judge did not provide any assistance to the jury by way of relating technical issues of probability to matters within their own experience. 95 Mr Goetz's evidence and the trial judge's summing-up reiterated, on a number of occasions, the precise nature of the evidence in the correct form. There was, however, one matter which may have suggested an inappropriate form of reasoning to the jury. That matter is the content of the record of interview. 96 I have quoted the relevant extract from the record of interview above. To reiterate the police officer said: "That you were the only one in two million, four hundred thousand … that you could be the father". Further, in relation to the relative chance of paternity percentages, the police officer said: "It's also saying the relevant chance of paternity is greater than ninety nine point ninety nine percent. So there's less than nought point nought one per cent that you aren't the father".

and "And it shows that ninety nine point nine nine nine per cent of chance that you are the paternal father". 97 The questioning, which was in any event inadmissible, did commit the Prosecutor's Fallacy. It came from a police officer, rather than an expert or the trial judge and was, accordingly, of limited significance. Nevertheless, in the absence of any direction about the assessment of probabilities generally, it would have been desirable for the trial judge to have warned the jury about the impermissibility of this line of reasoning.

DNA Evidence and the Proviso 98 The Courts have approached DNA evidence with caution. However, that caution is naturally abating as experience with the use of such evidence has grown. 99 A similar process occurred less than a century ago when fingerprint evidence was comparatively new. Fingerprinting was adopted in England for criminal identification only in 1901, a decade after Sir Francis Galton's book "Fingerprints" was published. (See Dodds "Identification by Fingerprints" (1986) 18 Australian Journal of Forensic Sciences 136.) 100 In R v Parker [1912] VLR 152, the prisoner had been convicted of breaking into a warehouse and stealing, from a safe. A fingerprint was left on a bottle that was in the warehouse. The fingerprint matched the fingerprint of the prisoner. The trial judge reserved a question for the Supreme Court. The question was: "When the only evidence against an accused person depends upon the resemblance between fingerprints, as in this case, whether such evidence is sufficient to support a conviction?"

101 By majority, the Full Court of the Supreme Court of Victoria affirmed the conviction. Madden CJ was in dissent. The Chief Justice said at 156: "The finger-print was found here on a ginger-beer bottle, and the Crown did not call the people through whose hands the bottle had gone, the man who had delivered it at the shop, or the woman who sold it to the boy. The finger-print might have been made by one of them. Or it might conceivably have been handled by the prisoner somewhere before it came to the owner of the house which was broken into. And apart from that, there is the general danger of saying that these prints may be relied on as exclusive evidence, when other evidence might have been called, but was not, which might have thrown light on the matter, one way or the other."

102 The Chief Justice, reflecting the judicial caution we have witnessed in recent years about DNA evidence, said at 153-155; "I regret that I am unable in this case to agree with my learned brethren in holding that the resemblance between the finger-prints on the bottle found in the room where the robbery was committed and the prisoner's finger-prints, standing alone, is sufficient evidence to warrant the conviction of the prisoner. The extreme danger of arriving at such a conclusion warrants me in not deferring to their opinions. We are asked to accept the theory that the correspondence between two sets of finger-prints is conclusive evidence of the identity of the person who made those prints as an established scientific fact, standing on the same basis as the proposition of Euclid or other matters vouched for by science and universally accepted as proved. If this finger-print theory were generally recognized by scientific men as standing on this basis, there would be no more to be said. It is said that the markings on the fingers of any individual retain their special characteristics from the cradle to the grave, and also that the markings on the fingers of no two individuals are the same, so that absolute correspondence between a finger-print and the markings on a man's hand is unmistakable evidence that he is the person who made such print. My difficulty arises from the fact that the subject of finger-prints has not been sufficiently studied to enable these propositions to be laid down as scientific facts. Finger-prints have been studied by Monsieur Bertillon in France from an anthropometrical point of view, and by Sir Francis Galton and a few others, doubtless highly intelligent persons, from the standpoint of mere observers. But the matter has not been investigated by scientists generally so that we can say that the propositions relied on by the Crown are accepted scientific facts. … The jury are told that the markings on a man's fingers never alter, and are different in every man. The jurymen know nothing in the world about that matter; they strike nothing like it in their daily lives. This is that kind of evidence that is particularly dangerous, for it carries with it a savour of mystery, as in this case the detective swears that no two men's markings are alike, and it is assumed not only that that is true, but that there is some mysterious brand implanted on man's hand for some definite purpose of characterizing him physically." 103 The majority in Parker were Hodges and Cussen JJ. Hodges J said at 157-158: "Now, supposing it were proved or admitted that the finger-print on the article stolen corresponded in every respect with the mark of the peculiar finger of the man charged with the theft, and supposing there were corroboration of that evidence by other witnesses - as is this case there were two of them - swearing positively to the identity of the impression on the article stolen with the impression of the accused's finger, it seems to me it would be the very safest possible evidence of identity that could be submitted to the jury, more trustworthy than the momentary glance that some individual might have had of the accused's face. I do not think it is necessary to say that there could not be any other finger-mark in the world like it, and I agree with what the learned Chief Justice has said as to the inadmissibility of that piece of evidence. In the same way, if a child who sees the accused's face for a moment swears positively to the person, she cannot say there is no other person in the world with a face like that." His Honour also said at 158: "In my opinion, it may be the safest of all evidence, as it does not depend upon the impressions caused by a momentary glance, but the impression is put on record, and the jury can see and judge for themselves as to the identity of the finger-marks, and the expert be merely a help to enable the jury to use the evidence of their own eyes."

105 The High Court refused an ex parte application by the prisoner, Parker, for special leave to appeal against the decision of the Full Court of the Supreme Court of Victoria (Parker v The King (1912) 14 CLR 681). Griffith CJ said at 682-683: "We think that leave must be refused. We are asked to allow the point to be argued whether, when evidence of finger prints is the only evidence of identity, it is sufficient to support a conviction. Leave is asked in the hope that the rule may be laid down that it is not. Signatures have been accepted as evidence of identity as long as they have been used. The fact of the individuality of the corrugations of the skin on the fingers of the human hand is now so generally recognized as to require very little, if any, evidence of it, although it seems to be still the practice to offer some expert evidence on the point. A finger print is therefore in reality an unforgeable signature. That is now recognized in a large part of the world and in some parts, has I think, been recognized for many centuries. It is certainly now generally recognized in England and other parts of the British Dominions. If that is so, there is in this case evidence that the prisoner's signature was found in the place which was broken into, and was found under such circumstances that it could only have been impressed at the time when the crime was committed. It is impossible under those circumstances to say there was no evidence to go to the jury. 106 The use of DNA evidence appears to have reached a similar stage to fingerprint evidence in about 1912. Like fingerprint evidence, it is of particular power, both to establish innocence and guilt.

Application of the Proviso 107 The application of the proviso in s6(1) of the Criminal Appeal Act 1912 involves the formulation of a judgment on which minds frequently differ. Courts of Criminal Appeal are often divided in its application. The High Court, often by majority, frequently comes to a different conclusion from a court of criminal appeal. 108 The errors committed in the present case are not, either singly or cumulatively, of so fundamental a character as to lead to the conclusion that there was hardly a trial at all. (See Wilde v The Queen (1988) 164 CLR 365 at 373; Glennon v The Queen (1993-1994) 179 CLR 1 at 8; Green v The Queen (1996-1997) 191 CLR 334 at 346-347 and c/f 371-372.) 109 The issue in the present case is whether a conviction was inevitable notwithstanding the errors committed, so that it cannot be said that the Appellant lost a chance of acquittal. (See Wilde v The Queen (supra) at 374; Glennon v The Queen (supra) at 8-9; Gilbert v The Queen (2000) 201 CLR 414 at [19]-[20]; [52]; [85]-[86]; KRM v The Queen (supra) at [72] and [123]-[129]; Grey v The Queen [2001] HCA 65 at [26]-[27]; [55]-[56].) 110 I do not attribute significance to the fact that the appellant was confronted in the record of interview with scientific evidence which he was unable to explain. In this regard, the Appellant's clear denials may well have worked to his advantage, as trial counsel could have properly determined. Nevertheless, it is wrong in principle for a police officer in the course of an examination to press too strongly on an interviewee scientific interpretations and conclusions in a manner suggestive of entrapment. I have in mind particularly the question in the record of interview in the present case: "So you see there's no other reason for the results showing that you are the father. Can you think of any reason?" In another case the unfairness implicit in such a line of examination may lead to a different result. 111 The errors I have identified relate to a risk that the jury may have been confused about, or been unable to properly assess the weight of the DNA evidence. This was, of course, the critical evidence in the case. That fact tends against the application of the proviso but is not, in my opinion, determinative in the present case. 112 Controversial issues have arisen in other cases about the weight that ought be given to DNA evidence and to the statistical representation in a single figure of the significance of that evidence. In this case neither at trial nor on appeal has there been any suggestion, either by way of submission or evidence on the part of the Appellant, which has, in the particular context of this case, cast any doubt about the paternity index and its significance in the trial. Nothing has called into question the conclusion that the odds were that it was 2,400,000 more likely that it was the accused who was the father than a person taken at random from the community. 113 The qualities of the samples in this case were uniformly high, unlike the frequently degraded and difficult to extract DNA from "crime stains" in other cases. There was no suggestion in this case about inefficient or inappropriate police conduct in obtaining and handling samples. (See e.g. R v Lisoff [1999] NSWCCA 364). Nor was there any suggestion of any doubt about the integrity of the samples by reason of laboratory procedures. Nor was any issue raised about the validity of the size of the data bases used, as discussed in R v Pantoja (1996) 88 ACrimR 554. (In the retrial there was no challenge to the validity of the data bases. See R v Pantoja (No 2) (NSWCCA, 5 November 1998, unreported).) Nor was there any issue about the validity of the system used to conduct the testing, relevantly, in the case of the second group of tests conducted in the present case, the profiler plus system. (See R v Karger [2001] SASC 264; R v Gallagher [2001] NSWSC 462). 114 It is of central significance that the circumstances of the present case restrict to an exceptionally unusual degree the number of persons who could possibly have committed the crime. (In terms of Bayes' theorem the prior odds based on this measurable factor only were very low.) This affects each of the errors I have identified. 115 The Relative Chance of Paternity percentage of 99.9999 percent should not have been before the jury. The particular circumstances of this case are that two out of five possible suspects had been scientifically eliminated and the possibility that two of the remaining three could have committed the act was low. In these circumstances, the Paternity Index comes closer to the degree of certainty suggested in the high percentage computed for Relative Chance of Paternity. Similarly, the difference between the statement of odds in the permissible form and a statement which commits the Prosecutor's Fallacy, becomes smaller. In the case of each of these errors, the extent to which the jury could have been misled was substantially attenuated by the particular circumstances of the case. 116 This distinction was appreciated by the Court of Appeal in R v Doheny and Adams (supra) where the Court stated the Prosecutor's Fallacy in the form of a syllogism at 372-373: "1 Only one person in a million will have a DNA profile which matches that of the crime stain. 2 The defendant has a DNA profile which matches the crime stain. 3 Ergo there is a million to one probability that the defendant left the crime stain and is guilty of the crime." The Court went on to say at 373: "Taking our example, the prosecutor's fallacy can be simply demonstrated. If one person in a million has a DNA profile which matches that obtained from the crime stain, then the suspect will be 1 of perhaps 26 men in the United Kingdom who share that characteristic. If no fact is known about the Defendant, other than that he was in the United Kingdom at the time of the crime , the DNA evidence tells us no more than that there is a statistical probability that he was the criminal of 1 in 26. The significance of the DNA evidence will depend critically upon what else is known about the suspect. If he has a convincing alibi at the other end of England at the time of the crime, it will appear highly improbable that he could have been responsible for the crime, despite his matching DNA profile. If, however, he was near the scene of the crime when it was committed, or has been identified as a suspect because of other evidence which suggests that he may have been responsible for the crime, the DNA evidence becomes very significant. The possibility that two of the only 26 men in the United Kingdom with the matching DNA should have been in the vicinity of the crime will seem almost incredible and a comparatively slight nexus between the defendant and the crime, independent of the DNA, is likely to suffice to present an overall picture to the jury that satisfies them of the defendant's guilt." (Emphasis added) 117 The same reasoning applies with even greater force in the present case. By reason of the particular circumstances of the victim in this case, there were only five men who had access to her. Two were ruled out by DNA tests. Two did not have DNA tests conducted but were both disabled persons, one in a wheelchair and the other blind, both with difficulties of access and subject to supervision. The possibility that one of these patients was one of the two or three persons in Australia other than the Appellant with the DNA profile of the father is, in my opinion, not a credible hypothesis. 118 I have not found this judgment an easy one. I have, however, concluded that in the particular circumstances of this case, the unchallenged Paternity Index was of such force that a conviction was inevitable. The proviso should be applied and the appeal against conviction dismissed.

Sentence 119 The submissions in this regard were briefly put. The Appellant submitted that the sentence was outside the appropriate range. Specifically, it was submitted that the Appellant's age, the strong evidence of good character, and the strong prospects of rehabilitation were such that the usual ratio should be varied to reduce the non-parole period. 120 The maximum penalty for the offence was twenty years imprisonment. His Honour imposed a sentence of seven years with a non-parole period of five years and three months. The objective seriousness of the offence was high. His Honour took into account all the relevant factors, including the subjective circumstances to which the appellant made particular reference. 121 In my opinion, the sentence was within the range which his Honour could legitimately impose. I detect no error which could justify the intervention of this Court. I would grant leave to appeal against sentence but dismiss the appeal.

122 SULLY J: I agree with the Chief Justice. 123 ADAMS J: I agree. **********

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.