Court of Criminal Appeal: Doyle CJ, Prior and Gray JJ
DOYLE CJ: I would dismiss the appeal. I agree with the reasons given by Gray J.
I wish to add some remarks in relation to the submission by Mr Wells QC, counsel for the appellant, that the Judge erred in failing to warn the jury against the danger of misusing the statistical evidence in this case. The statistical evidence to which I refer is the evidence interpreting or explaining the significance of the matches found in the DNA profile obtained from the material found on the deceased’s blouse, and the DNA profile of the appellant. That statistical evidence was expressed in the form of a likelihood ratio and in the form of a match probability.
I add these remarks because of the practical importance of the issues raised by the submission. The issues are likely to occur in other trials.
I do not propose to summarise the relevant facts, the issues at trial, or the manner in which the prosecution case and the defence case were put at trial. All of these are adequately dealt with by Gray J.
The submission for the appellant
Mr Wells submitted that a warning against the misuse of the statistical evidence should have been given for a number of reasons. As I recall his argument, the terms of the required warning were not spelt out by him. I take it that the type of warning which he submits should have been given flows from the circumstances said to give rise to the need for the warning.
Mr Wells submitted that a warning was required as a matter of law when the prosecution relies on a statistical calculation, expressed as a mathematical probability, to interpret or to explain the significance of evidence about a match between DNA material linked to the commission of a crime (I will call this the incriminating DNA) and DNA material obtained from an accused. He submitted in the alternative that in most such cases, and in this case in particular, such a warning will be required as a matter of practice to discharge the judge’s duty to give the jury such directions as are required to ensure that there is a fair trial and to avoid a miscarriage of justice.
The fundamental reason for requiring the warning is said to be that without it the jury will give undue weight to the evidence of a match between the incriminating DNA and the accused’s DNA. They are likely to do so because of the impact on them of the interpretative statistical evidence. Implicit in the submission is the acknowledgement that the danger arises when the likelihood ratio or match probability (see Gray J at [66] - [67] are calculated and expressed as a ratio that is sufficiently high or low to suggest to the ordinary person something like certainty, or something sufficiently high or low to be what one might call distractingly impressive. It…
Grounds of appeal
[14] The appellant contends that the primary judge erred in finding that the respondent had discharged the onus of establishing the matters required under the relevant provision.
[15] It was submitted that his Honour failed to give adequate weight to the evidence of the second witness, and that the findings of fact were not reasonably open on the material before the Court.
[16] The respondent, for its part, submits that no error is disclosed, and that the conclusions reached were well within the range of outcomes available to the primary judge.
Consideration
[17] It is convenient to deal first with the question of construction, as the resolution of that issue informs the remaining grounds advanced by the appellant.
[18] For the reasons that follow, we are of the view that the appeal should be allowed in part and the matter remitted for further determination.
Continue reading R v Karger
Sign in free to access the full text
Details
AGLC
R v Karger [2002] SASC 294
Case
[2002] SASC 294
Decision Date
CaseChat Overview and Summary
Damon John Karger was convicted by a jury of the murder of Kerryn Jean Ostendorf and appeals against that conviction. The appeal concerns the trial judge’s failure to give a specific warning to the jury about the use of statistical evidence, specifically the likelihood ratio in relation to the DNA evidence. The statistical evidence showed that the likelihood of the DNA match if the appellant was the source was greater than 90 billion times more likely than if an unknown, unrelated person was the source. The appellant’s submission was that the trial judge should have directed the jury that the statistical evidence did not indicate the probability that the appellant was the source of the incriminating DNA and that the jury should not treat the statistical evidence as proof of guilt. The court found that the trial judge was not required to give such a warning. The statistical evidence, while strong, was not direct evidence of the appellant’s guilt. The jury should consider the statistical evidence in light of all the evidence in the case, and the trial judge was not in error in failing to give the proposed warning. The appeal was dismissed.
Orders
Orders of the court
Full text does not contain this section.
Background
Background to the litigation
Full text does not contain this section.
Evidence
Evidence Before The Court
A further or alternative reason for requiring a warning was that if it is not given there is a real danger of the jury committing the so-called “prosecutor’s fallacy”: see Gray J at [183] - [185] Doheny and Adams [1997] 1 Crim App R 369 at 372-373 and R v GK [2001] NSWCCA 413; (2001) 53 NSWLR 317 at 328-329 Mason P. The risk is that the jury will reason that the evidence of the likelihood ratio or match probability expresses the probability that the incriminating DNA was the DNA of the accused. Properly understood, it does no such thing. Even worse is the risk that the jury will treat the likelihood ratio or match probability as expressing the probability that the accused is guilty.The statistical evidence The evidence is summarised by Gray J at [66] - [68]. Mr Pearman’s evidence about the likelihood ratio was as follows:“The likelihood ratio is becoming a more commonly used form of presenting not only DNA evidence but other forms of forensic evidence. It is based on the fact that, in a Court situation, generally, the Court is faced with two competing scenarios, or points of view; the prosecution point of view that the accused has left the crime stain, and the defence point of view, or hypothesis, which is that some other unrelated person has left the stain. The calculation we do doesn’t estimate the probability that the person left the stain, but what we can do is estimate the probability that the crime stain and the reference sample match, if the accused left it, or if someone else left it. You can then weigh up those two competing probabilities in what is called the likelihood ratio…for this stain, we would say that the crime stain is greater than 90 billion times more likely to match the profile of the accused if he left it, rather than if an unknown, unrelated person left it.”Dr Buckleton ’s evidence about the matter was as follows:“[The figure, reported as a ‘frequency’, being] the chance that another person [as well as the appellant] would possess this profile is approximately one in 90 billion…[reported as a likelihood ratio] it is just the other way up. So the evidence is about 90 billion times more likely if Mr. Karger donated a sample than if a random person donated the sample.” Clearly enough, that evidence must be used with care. For the purposes of expressing the likelihood ratio, two factual hypotheses are compared. One is the prosecution hypothesis that the appellant was the source of the bloodstain and of the incriminating DNA. The other hypothesis is the defence hypothesis that an unrelated person is the source of the bloodstain and of the incriminating DNA. The statistical evidence calculates the probability of the occurrence, in either case, of the match found between the incriminating DNA and that of the appellant, and expresses the competing probabilities as a ratio. The statistical evidence interpreting the significance of the DNA match is not evidence of the probability that the appellant was the source of the incriminating DNA. To so regard it would be to make an error. However, the statistical evidence interpreting the DNA match is expert evidence that the jury could use in deciding whether it was satisfied beyond reasonable doubt that the appellant was the source of the incriminating DNA. The statistical evidence is undeniably strong evidence pointing to a conclusion that the accused was the source of the incriminating DNA, but is not direct evidence of that fact. And, as is obvious, the statistical evidence must be considered in the light of other evidence in the case.
Decision
Reasons for decision
PRIOR J
(dissent)
GRAY J
For all those reasons I do not accept the submission that the Judge erred in failing to warn the jury against the misuse of the statistical evidence.Conclusion For those reasons, and the reasons given by Gray J, I would dismiss the appeal. I agree with the reasons given by Gray J and with the remarks added by the Chief Justice. Damon John Karger the appellant was charged with the murder of Kerryn Jean Ostendorf. He was convicted on the verdict of a jury. This is an appeal against the conviction. The Crown Case - an outline General Mrs Ostendorf (“the deceased”), aged 41 years, died at her home in Campbelltown in the early hours of Saturday 17 January 1998. She had been out the previous night and drove the last of her friends home at approximately 3.45 am. Given the proximity of her home to that address she arrived home shortly after 4.00 am. The Crown case was that the deceased was murdered during the course of a sexual attack soon after she returned home. The deceased’s body was found by her daughter Rachel Rowlands soon after 7.30 pm on Saturday 17 January 1998. She was lying face down on her bed. She had been strangled with a camisole. Most of the clothing she had worn on the Friday night lay on the floor near the bed. Her knickers, skirt, bra and blouse were cut and torn. The deceased had bruising to her arms and legs, bruising to her eyes and mouth and grazes on her knees. Dr Ross James the examining pathologist gave evidence about the injuries sustained and the cause of death. On examination he observed that the deceased’s anus was markedly dilated or enlarged. This was inconsistent with normal post mortem appearance but could not be described as an injury. Dr James’ opinion was that penetration by an object could not be excluded as the cause. If penetration had occurred he could not say whether it was before or after death. The deceased was security conscious. There were no signs of forced entry to her home. There were no signs of disturbance other than in her bedroom. There were no signs of theft other than it appeared that her purse had been emptied. It was the Crown case that the deceased knew her attacker. The appellant, aged 28 years had known the deceased for some years through her daughter, Mrs Rowlands. In more recent times, the appellant had visited the deceased when he returned to Adelaide on army leave. In January 1998 the appellant was staying with his parents at Paradise. This was a three minute drive or a ten minute walk from the deceased’s home. On the night of Friday16 January 1998 and in the early hours of Saturday 17 January 1998 the appellant made nine unanswered telephone calls to the deceased. These calls were at 11.05 pm, 11.57 pm, 12.02 am, 12.12 am, 12.21 am, 12.32 am, 3.47 am, 3.50 am and 3.52 am. The appellant did not attempt to telephone the deceased after 3.52 am. She died shortly after that time. Her body was not found until soon after 7.30pm on Saturday 17 January 1998 and the media did not disclose the fact of her death until the Sunday afternoon.
Ratio Decidendi
Legal Principle Established
Established by:
GRAY J
Mr Pearman’s evidence about the likelihood ratio was as follows:“The likelihood ratio is becoming a more commonly used form of presenting not only DNA evidence but other forms of forensic evidence. It is based on the fact that, in a Court situation, generally, the Court is faced with two competing scenarios, or points of view; the prosecution point of view that the accused has left the crime stain, and the defence point of view, or hypothesis, which is that some other unrelated person has left the stain. The calculation we do doesn’t estimate the probability that the person left the stain, but what we can do is estimate the probability that the crime stain and the reference sample match, if the accused left it, or if someone else left it. You can then weigh up those two competing probabilities in what is called the likelihood ratio…for this stain, we would say that the crime stain is greater than 90 billion times more likely to match the profile of the accused if he left it, rather than if an unknown, unrelated person left it.”Dr Buckleton ’s evidence about the matter was as follows:“[The figure, reported as a ‘frequency’, being] the chance that another person [as well as the appellant] would possess this profile is approximately one in 90 billion…[reported as a likelihood ratio] it is just the other way up. So the evidence is about 90 billion times more likely if Mr. Karger donated a sample than if a random person donated the sample.” Clearly enough, that evidence must be used with care. For the purposes of expressing the likelihood ratio, two factual hypotheses are compared. One is the prosecution hypothesis that the appellant was the source of the bloodstain and of the incriminating DNA. The other hypothesis is the defence hypothesis that an unrelated person is the source of the bloodstain and of the incriminating DNA. The statistical evidence calculates the probability of the occurrence, in either case, of the match found between the incriminating DNA and that of the appellant, and expresses the competing probabilities as a ratio. The statistical evidence interpreting the significance of the DNA match is not evidence of the probability that the appellant was the source of the incriminating DNA. To so regard it would be to make an error. However, the statistical evidence interpreting the DNA match is expert evidence that the jury could use in deciding whether it was satisfied beyond reasonable doubt that the appellant was the source of the incriminating DNA. The statistical evidence is undeniably strong evidence pointing to a conclusion that the accused was the source of the incriminating DNA, but is not direct evidence of that fact. And, as is obvious, the statistical evidence must be considered in the light of other evidence in the case. It is necessary for the jury to appreciate these points if they are to make proper use of the statistical evidence. It is also necessary to avoid the so-called prosecutor’s fallacy, if the evidence is to be properly used. Reasoning in the form of that fallacy involves, as I understand it, both the error just identified and an approach which would treat the statistical evidence as proof of guilt, rather than as evidence from which a conclusion on a matter which might in turn point to guilt could be reached.The use to be made of statistical evidence