R v Gum - [2007] SASC 311

SUPREME COURT OF SOUTH AUSTRALIA
(Court of Criminal Appeal)
R v GUM
[2007] SASC 311
Judgment of The Court of Criminal Appeal
(The Honourable Justice Duggan, The Honourable Justice Vanstone and The Honourable Justice Kelly)
22 August 2007
CRIMINAL LAW - APPEAL AND NEW TRIAL AND INQUIRY AFTER CONVICTION - APPEAL AND NEW TRIAL - PARTICULAR GROUNDS - MISDIRECTION AND NON-DIRECTION
CRIMINAL LAW - EVIDENCE - SIMILAR FACTS - RELEVANCE - SEXUAL OFFENCES
Appellant convicted after trial by jury for two counts of rape - prosecution counsel in closing address referred to similarities between the two offences - whether trial judge was required to give a propensity warning - discussion of principles relating to propensity evidence - Held: no risk of impermissible reasoning by jury, direction not required in order to avoid miscarriage of justice.

CRIMINAL LAW - EVIDENCE - MATTERS RELATING TO PROOF
Appellant argued chain of evidence relating to DNA samples and analysis was not conclusively established - Held: sufficient evidence of link to enable relevant conclusion to be drawn.
Evidence Act 1929 (SA) s 45a and s 45b, referred to.
Dimitriou v Samuels (1975) 10 SASR 331; Pfennig v The Queen (1995) 182 CLR 461; Makin v Attorney-General (NSW) [1894] AC 57; Thompson v The Queen (1989) 169 CLR 1; Director of Public Prosecutions v Boardman [1975] AC 421; Director of Public Prosecutions v P [1991] 2 AC 447; Moorov v H M Advocate (1930) JC 68; Hoch v The Queen (1988) 165 CLR 292; Martin v Osborne (1936) 55 CLR 367; R v Perry (1982) 150 CLR 580; Phillips v The Queen (2006) 225 CLR 303; R v Armstrong (1990) 54 SASR 207; R v Straffen [1952] 2 All ER 657; The Queen v Schlaefer (1984) 37 SASR 207; R v J (No 2) [1998] 3 VR 602; KRM v The Queen (2000) 206 CLR 221, applied.

R v GUM
[2007] SASC 311
Court of Criminal Appeal: Duggan, Vanstone and Kelly JJ

DUGGAN J. The facts of this matter are set out in the judgment of Vanstone J.
I agree with the conclusion and reasons of Vanstone J in relation to the first ground of appeal which raised the issue of the DNA chain of evidence.
The particular concern expressed by counsel for the appellant in relation to the other ground of appeal arose out of comments made by the prosecutor in her closing address to the jury concerning similarities between the alleged rapes on the two victims. The prosecutor said:
Whoever raped these women, you might think, has watched them. He must have. Both women were alone in their homes at the time of the rapes: A because she lived alone, and B because her housemate was in Bali. Both women, you might think, had parked their cars in full view so that you would know if and when they had arrived home. Both women, on the nights that they were raped, arrived home late, about 10.30, 11 o'clock. He must have waited for them, you might think. That, you might think, is the relevance ultimately of that faecal material found out the back of A's house, and the cigarettes there as well, because A, told you, didn't she, during her evidence, that her rapist's hands smelled of cigarette smoke. Those butts and the faecal material are both found at the back of her flat where the bathroom window was. It's a matter for you, ultimately, but you might think that that is indicating to you that her rapist has waited for her, perhaps smoked some cigarettes, perhaps needed to relieve himself. It doesn't identify the accused as being her rapist, it just shows you, you might think, that her rapist has waited, watched and waited. In both cases, the rapist gains entry to the houses via a window: A's bathroom window and ultimately, you might think, B's bedroom window. Both rapes occurred in the early hours of the morning. The rapist in both cases is crouched down by the bed watching his victim. The rapist in both cases tries to quieten his victim: in A's case, he puts his hand over her mouth and, in B’s case, he goes to put a pillow over her face. The rapist of both these women tells them to roll over. The rapist of both these women then vaginally rapes them with his penis and ejaculates. The rapes are within months of each other in 1991, the rapes are both on the western side of town. Both women describe their assailant as tall with short, dark hair.
Now would all of that evidence alone be enough to identify to you beyond reasonable doubt who raped these women? No. All it does is show us that the rapist of both women might be the same person because of the similarities. It’s not enough to identify who that person is, but enough to show you it might be the same person.
It was argued that, in the light of these submissions, the trial judge was required to instruct the jury against impermissible reasoning in the use of the evidence. In particular, it was said that the jury should have been given a warning against propensity reasoning.
In my view, there was no risk of impermissible reasoning resulting from the submissions made by the prosecutor. The submissions did not go beyond suggesting that the similarities provided circumstantial evidence that the same person, whoever it was, might have committed both offences. It was not suggested that the jury could argue from one offence to the other on the basis of propensity reasoning.
I express no opinion on whether the evidence could have provided the basis for striking similarity or underlying unity reasoning. The prosecutor did not suggest to the jury that it could be used for either purpose and no instructions on such use were given to the jury.
VANSTONE J:
Introduction
David Christopher Gum appeals against his conviction for two counts of rape after a trial by jury in the District Court. Each offence was committed upon a woman in her home by an intruder on occasions about fifteen weeks apart. In each instance the sole evidence of identification was in the form of DNA evidence. The answer to the charges consisted only of testing aspects of that evidence: highlighting what were said to be deficiencies in the storage and analysis of the various samples and in proving the chain of evidence relating to them.
There was no objection to the joinder of the two charges on one information and no application for separate trials.
The appellant raises two grounds. First he submits that a miscarriage of justice resulted from the failure of the learned trial judge to direct the jury as to how evidence relating to one count could and could not be used in respect of the other. Then, he argues that because of the failure to conclusively link the DNA analysis of the reference sample of the appellant with the profile used for comparison with the samples from the “scenes”, the jury could not be satisfied beyond reasonable doubt that the appellant was the rapist.
Background
The offences occurred in March and June 1991 at Fulham and West Lakes Shore respectively. In neither case did the victims (whom I shall call “A” and “B”) see enough of the intruder to be capable of identifying him. Rather, the descriptions given were very general. There were parallels between them. For instance, A gave evidence that her attacker was “mid 20s”, “quite tall” – over six feet, had short, “darker coloured hair” and was clean shaven, while B described the man who raped her as in his “early 20s”, six feet tall, having short dark hair and clean shaven. On the other hand, A said he had “not a large build – solid” while B said her attacker was “lean, a bit toned”.
Although the prosecution was put to proof as to all aspects of the case, defence counsel did not challenge the evidence that each woman was raped.
The prosecution called a number of witnesses who were involved in various ways with gathering the DNA evidence. The source of the DNA said to match that of the appellant was – in the case of A – a swab from her groin where the attacker was said to have ejaculated and, in B’s case, a high vaginal swab. The profile developed from the sample in the instance of A was attributable to only one person, being a male. There was correspondence with the appellant’s reference sample at each hypervariable region examined. In the case of B the profile was a mixture. Even though Mr Sobieraj, the forensic scientist who gave evidence of the results, did not have a separate reference sample from B, he was able to distinguish between the profile attributable to the major component of the mixture – being the attacker’s – and the balance. That was because Mr Sobieraj was able to generate the DNA profile of B from her “epithelial cell fraction”. Therefore, notwithstanding that the sample from B contained a mixed profile, the calculation of a “likelihood ratio” by reference to the DNA derived from the appellant was not on that account any different from the calculation for A’s attacker.
In two of the hypervariable regions examined there was insufficient DNA to report the result. However this appears, on the evidence, to have made no discernable difference to the likelihood ratio.
The result was that the DNA profile from each victim was said to be greater than one billion times more likely to match the reference DNA profile of the appellant if he were the true source of the DNA, as against if the source were an unknown, unrelated man.
Chain of evidence
On the appeal the principal focus of the attack on this evidence concerned the reference sample attributed by Mr Sobieraj to the appellant. It was argued that there was no sufficient link between the results of the forensic scientist, Ms Toop-Mitchell, who analysed a buccal sample from the appellant to extract DNA, and the ultimate evidence given by Mr Sobieraj.
Ms Toop-Mitchell did not give evidence at trial but her statement was read to the jury as one which had “been agreed between counsel”. The judge instructed the jury that it could regard the statement as equivalent to oral evidence. The statement set out qualifications and experience of the deponent. It then recorded that Ms Toop-Mitchell had examined and analysed exhibits relating to “David Christopher Gum, defendant”. It referred to a report detailing such examination and analysis and to the Forensic Science Centre (now Forensic Science South Australia (“FSSA”)) case number of 0403012. That report was apparently annexed to the statement and incorporated into it. It too was read to the jury. It purported to be a report “on the examination of items in two matters involving David Christopher Gum, FSSA case number 0403012 …”. The report related that the witness had undertaken DNA “extraction on a reference buccal sample from David Gum which was given the item number “DCG” at FSSA”. It then described two further analytical processes said to have been performed on 19 and 20 July 2004. It recorded that “the results of these examinations/analysis I describe in the report prepared by Thomas Christopher Sobieraj also identified as Forensic Science South Australia case number 0403012 and dated 3 August 2004”.
When Mr Sobieraj came to give evidence he produced documents described as “login receipts”. These purported to relate to the receipt at the FSSA of the biological samples taken from each of the victims. Each was identified by a separate FSSA case number. Mr Sobieraj also produced a document headed “Table of DNA Results” for the “FSSA case number 0403012”. That document collected the results of DNA analysis of the reference sample attributed to the appellant and numbered 0403012, together with the “crime scene” samples referring to the victims by name and giving the reference numbers matching those on the login receipts. That document was tendered without there being objection by counsel then acting for Mr Gum.
As I understand the first part of this argument it is that because the part of Ms Toop-Mitchell’s report which contained the DNA profile of the appellant was not read in court, there was no sufficient link between her work and Mr Sobieraj’s use of a DNA profile said to relate to the appellant.
In my view that argument can be swiftly rebutted. First the “Table of DNA Results” went into evidence, as I have said, without objection. Insofar as Mr Sobieraj could not give first-hand evidence of its contents, it went in, presumably, either as a business record of the FSSA or as a statement of fact or opinion, under s 45a and s 45b Evidence Act 1929 (SA) respectively. Therefore the document itself amounted to proof that the reference swab from David Christopher Gum, bearing the identification number 0403012, gave the results as set out in the chart. Even more, there was circumstantial evidence to link those results with the analysis made by Ms Toop-Mitchell, inasmuch as she too referred to the case number I have just mentioned – which other evidence established was unique to a particular case within the FSSA – and she also referred to her results being described in a report of Mr Sobieraj bearing the same number.
But there was a further aspect to this argument. Before the jury, counsel for the appellant had Mr Sobieraj produce from his file a document referred to as an electropherogram, said to have been produced by another member of staff, Andrew Donnelly, in respect of the reference sample of the appellant. It was dated 21 July 2004. It was admitted into evidence. The defence relied on the difference between the date of that document as against Ms Toop-Mitchell’s last analysis of 20 July 2004.
I do not think the tender of this document changed the position as I have already outlined it. True, this document did not bear either of the dates mentioned by Ms Toop-Mitchell. However, there was no evidence as to how the date on the electropherogram should be interpreted; that is whether it related to an examination on that day, or merely to a collation or printing on that day. I consider that this document did not in any way detract from the proof offered to the jury in the course of the prosecution case.
I acknowledge that the “chain of evidence” on the appellant’s reference sample could have been more perfectly proved. It could readily have been done by admissions of specific facts making reference to the results of Ms Toop-Mitchell’s analysis or to the passing on of specific results to Mr Sobieraj. However, there is no rule that requires conclusive proof of a chain of evidence any more than of any other fact in issue. Whether a jury is prepared to draw relevant inferences is for it to say: see Dimitriou v Samuels (1975) 10 SASR 331 per Bray CJ. Consequently I consider that there was sufficient proof before the jury on the basis of which it could conclude that Mr Sobieraj’s opinions were based on the analysis of the appellant’s reference sample.
I would add this observation. I mentioned that there was no objection to the various documents going into evidence. But also, there was no objection to Mr Sobieraj giving his opinions based (partly) on the use of Ms Toop-Mitchell’s results. If there were some deficit in the evidence underpinning his opinions, then objection should have been taken before he expressed them. In fact this argument was put to the judge in the form of a submission of no case to answer. In my view the opportunity to raise it had passed.
In my opinion the trial judge was correct to find that there was a case to answer and no miscarriage of justice has been caused on account of this matter.
Directions
The appellant did not argue against the proposition that the counts were cross-admissible. I consider they plainly were. Rather, he submitted that in the absence of any direction as to the proper and improper use of the evidence of the one count in respect of the other, the jury could have engaged in an impermissible chain of reasoning. He suggested that reasoning by way of propensity would have been impermissible.
In support of this complaint he pointed to the use made by the prosecutor in her final address of the similarities in the crimes. He submitted that if the jury was satisfied of the accused’s guilt on count 1 – where the DNA result was straightforward – then it might have reasoned from that conclusion to one of guilt of count 2 by using those similarities, but without the assistance of accompanying directions by the judge.
In order to address this complaint it is necessary to determine on what basis the evidence of each count was admissible in proof of the other. This necessitates reference to the principles relating to propensity or similar fact evidence. Acknowledging that there is a difference in those terms, I shall use the descriptor “propensity evidence” in the course of this discussion.
The general principle is that “evidence that reveals that the accused is a person of bad character is not admissible if it proves no more than that he or she has a general disposition or propensity to commit crime or crime of a particular kind”: Pfennig v The Queen (1995) 182 CLR 461, per McHugh J at 512, citing Makin v Attorney-General (NSW) [1894] AC 57. Propensity evidence may be relevant because of the light it throws on any of a number of issues in a case. For example, it might assist in proving identity or intention, or in disproving accident or mistake or innocent association: Thompson v The Queen (1989) 169 CLR 1, 16.
It might achieve its purpose by reason of it bearing striking similarities with the offence for which the accused is on trial. But restricting its admissibility to cases of striking similarity gives too much weight to a particular manner of stating the principle and such expressions are not definitive: Director of Public Prosecutions v Boardman [1975] AC 421, 452-453; Director of Public Prosecutions v P [1991] 2 AC 447, 460-461; Pfennig at 478-484. It may show a pattern or system or underlying unity or inextricable connexion when viewed alongside the other conduct alleged: Moorov v H.M. Advocate (1930) J.C. 68, 73-74 per the Lord Justice-General (Lord Clyde); Thompson per Deane J at 32 and Gaudron J at 39-40. But the common thread running through all such descriptors is that the evidence must raise the improbability of the events having occurred other than as alleged by the prosecution: Hoch v The Queen (1988) 165 CLR 292, 295. The utility of such evidence has been described as allowing for “admeasuring the probability or improbability of a fact or event in issue … given the fact or facts sought to be adduced”: Martin v Osborne (1936) 55 CLR 367, 385.
The sole criterion for its admission is the strength of its probative force: Hoch at 294, rather than any judgement that one or more of the above labels is apt to fit it. The degree of probative force required has been described as such that to exclude the evidence would be “an affront to common sense”: Boardman at 456; and as such that it “clearly transcends its merely prejudicial effect”: R v Perry (1982) 150 CLR 580, 609 per Brennan J; Phillips v The Queen (2006) 225 CLR 303, at 320 per the Court. The test though, in this country, is that its probative value or cogency must be such that “if accepted, it bears no reasonable explanation other than the inculpation of the accused person in the offence charged”: Hoch at 294; Pfennig at 483; Phillips at 323.
I turn then to consider what was the potential use in this case of each count in respect of the other.
Plainly the evidence that DNA matching the appellant’s was found at each “scene” was extremely potent. The jury would have been entitled to view this evidence, standing alone, as sufficient proof of either count. That common factor was in my view sufficient to provide the “inextricable connection” or “underlying unity” referred to in Thompson’s case; that latter also being relevant in the local case of R v Armstrong (1990) 54 SASR 207. I would also have been prepared to conclude, if it were necessary to do so, that the finding of matching DNA about the bodies of the two victims would amount to a striking similarity. But in the end, the question remains: what use could be made by the jury of the evidence on one count in proof of the other?

The answer to that question is, in my mind, that the jury could use the broad match of DNA profiles at each scene and indeed the parallels between the physical description of the offender in each instance to resolve any doubts they might otherwise have entertained about the integrity of the samples taken from each woman, the storage of those samples prior to delivery to the FSSA, the chain of evidence, the storage of the samples over a long period of time after delivery and any doubt arising from the fact that the profile detected in the case of B was a mixed one. They could further use the broad similarity between the appellant’s appearance and descriptions of the offenders and indeed the proximity of time and place of the crimes as some additional evidence confirming the statistical evidence of the unlikelihood that any other person could have been the attacker.
Such uses of the evidence would not necessitate any direction to the effect that until either count was found proved, evidence relating to it could not be used in respect of the other count. Rather, in reasoning in this way the jury would have been taking an overview of the whole of the evidence, just as the juries were entitled to do in, for example, Martin v Osborne and Armstrong and indeed Makin.
The scope for sequential reasoning such as that available to the jury in R v Straffen [1952] 2 All ER 657, Pfennig and Perry was extremely limited. This was not a case where once the jury had concluded that the accused person had a particular propensity they could go on to resolve the issue of identity by reference to their conclusion. Here, as I said, the issue really came down to whether the jury could rely on the integrity of the DNA evidence. Propensity was demonstrated only as an incident of acceptance of the DNA evidence and (as almost inevitably followed) guilt.
Indeed it was precisely in this way that the prosecutor suggested the jury could use the broad similarities in the physical description and behaviour of the two attackers. She pointed out to the jury that although the prosecution “rises and falls with the DNA evidence” the other evidence in the case – which she analysed – tended to relegate to the realm of theory any suggested confusion in the handling or analysis of the various samples. She pointed to the unlikelihood of contamination or mistake having played a role when identical DNA profiles were arrived at from two nearby rapes bearing broad similarities and of those profiles matching the appellant’s, unless the appellant was the rapist.
If these were the true issues in the case, then I cannot see how a so-called “propensity direction” could have assisted the jury. I take a propensity direction to be a direction to the effect that the jury should not reason from a finding of guilt on one count that the accused is the sort of person who rapes women and is therefore likely to be guilty of the other count. As I have said, this was not the sort of case where the jury could have reasoned from one count to the other, except insofar as the coincidence of DNA results could have engendered confidence in their provenance. It was not suggested by anyone that either crime could be proved without reliance on the DNA profile referable to that crime.
There is no universal rule that in every case where the jury hears evidence suggesting that the accused has committed other crimes of a similar type a propensity warning is necessary to guard against a miscarriage of justice: The Queen v Schlaefer (1984) 37 SASR 207; R v J (No 2) [1998] 3 VR 602 per Callaway JA, Winnecke P and Charles JA agreeing; KRM v The Queen (2000) 206 CLR 221 per McHugh J at 233-237, per Gummow and Callinan JJ at 246-247, per Kirby J at 259-261 and per Hayne J at 264.
Whilst the judge could not have been criticised for giving such a warning, I do not consider that one was necessary in order to avoid a perceptible miscarriage of justice.
Conclusion
In the end it seems to me that the issues in the trial were stark and in the directions of the judge the jury had sufficient assistance as to how to approach them.
Neither ground of appeal is made out.
I would dismiss the appeal.
KELLY J. I agree with the reasons of Vanstone J.