SUPREME COURT OF QUEENSLAND

1. [1]
   The applicant, Cameron Andrew Stewart, seeks an order pursuant to s 590AA Criminal Code for the exclusion of DNA evidence. 

2. [2]
   The applicant and his co-accused, Vladimir Garcia, are charged with the murder of Alexander Davie and his wife Suzanne Davie on or about 3 May 2009 on the Gold Coast.  Mr Davie was a security guard.  His body was found on 4 May 2009 in a toilet cubicle at commercial premises at the Gold Coast for which he had security responsibilities.  The deceased was dressed in his work uniform, which included a robust utility (or accoutrement) belt.  He was also wearing a watch.  His mobile phone was found in the toilet.  The body of Mrs Davie was found in her home at Robina.

3. [3]
   There were two places where relevant identifiable DNA was found:

1. (a)
   Item 53 – from accoutrement pouch on the utility belt worn by Mr Davie;

2. (b)
   Item 56 – from watch on left wrist of Mr Davie.

4. [4]
   The case against the applicant is a circumstantial one. As outlined in the submissions on behalf of the applicant, it is apprehended that the prosecution case against him is as follows:

1. (a)
   The applicant had previously worked for Mr Davie in his security business, and had known Mrs Davie.  He had worked with Mr Davie for approximately three years until 3 April 2006.

2. (b)
   There had been a falling out between the applicant and Mr Davie, who alleged that the applicant and his co-accused had stolen property belonging to him, including firearms.

3. (c)
   The applicant had previously made threats to kill Mr Davie.

4. (d)
   The applicant and his co-accused were on the Gold Coast around the time of the killings.

5. (e)
   The body of Mr Davie was found with the hands tied up with zip ties.  Zip ties in the form of handcuffs were subsequently found in the possession of the applicant in Victoria. 

6. (f)
   Partial DNA profiles were obtained from the body of Mr Davie; item 53, from the empty accoutrement pouch on the utility belt that was apparently of an appropriate size and shape to hold a mobile phone, and item 56, from a watch on Mr Davie’s left wrist. 

5. [5]
   In respect of the DNA evidence, Ms Caunt’s report dated 7 August 2009 states:

“The DNA profiles obtained from these samples indicate the presence of DNA from two contributors in approximately equal proportions.  Since the samples are said to be from items that belong to [the male deceased] then it is reasonable to assume that his DNA would be present.  Once [the male deceased’s] DNA profile has been ‘subtracted’ from these mixed DNA profiles incomplete DNA profiles remain.

The corresponding components of both remaining partial DNA profiles match each other and match the corresponding components of the DNA profile of [the applicant].  Therefore this DNA could have come from him or from some other person whose corresponding components also match.

It is estimated, based on Queensland Caucasian data, that the probability of these remaining partial DNA profiles occurring if this DNA had come from someone other than, and unrelated to [the applicant] is approximately:

• 1 in 140 (Item 56);
• 1 in 880 million (Item 53).”

6. [6]
   No objection is taken to the partial profile obtained from the pouch on the utility belt (item 53). The only challenge to the DNA evidence is in respect of item 56, the partial profile obtained from the watch on the left wrist of Mr Davie.  As stated, in respect of that partial profile, Ms Caunt provided a statistical estimate based on Queensland Caucasian data, that the probability of that partial DNA profile occurring if the DNA had come from someone other than, and unrelated to the applicant, was approximately 1 in 140. 

The applicant’s submissions

7. [7]
   Objection was taken to evidence of the watch DNA profile and its statistical significance, its low figure of probability being contrasted with the statistical estimate provided for item 53 (1 in 880 million).

8. [8]
   While reference was made to Ms Caunt’s evidence at the committal hearing (at R3-15, 3-16), that DNA is a very stable molecule and can persist for any number of years, in the right conditions,^[1] it was accepted for the purposes of this application that there was evidence that the deceased’s watch was cleaned in April 2006. In the light of that evidence, it was not argued that the watch DNA could have been deposited at a time before April 2006 when the applicant and the deceased worked together. Accordingly, counsel did not press an argument that any lack of probative value of the watch DNA was accentuated by the fact that the DNA could have existed for some time before the murder of Mr Davie.

9. [9]
   As to the probative value of DNA evidence, counsel for the applicant referred to the following comments made in R v Noll [1999] 3 VR 704 by Callaway JA (with whom Phillips CJ and Ormiston JA agreed) at 708-709:

“It is the statistical step in the reasoning that gives DNA evidence its probative value.  In People v Barney at 742 it was called ‘the pivotal element of DNA analysis’ ..... without careful examination, such evidence may appear to be more probative than it is: see R v Pantoja at 561-2 and R v Humphrey (1999) 72 SASR 558 at 567-8.”

10. [10]
    Counsel also placed reliance on statements of Kellam JA (with whom Vincent JA and Whelan AJA agreed) in R v Rye (2007) 178 A Crim R 345 at 355:

“Clearly the evidence of DNA analysis is helpful to a jury in only one of two circumstances.  The first of those is, that it is obviously highly relevant if the analysis excludes a person from having been a contributor to the DNA in question  …   The second circumstance is when the DNA analysis does not exclude a particular person as being a contributor to the DNA material in question.  The question then to be determined is, ‘What statistical probability is there, based upon valid and appropriate statistical data bases, that persons other than the contributor may have contributed to the DNA?’  The answer to this question may reveal that the DNA evidence is highly probative or it may reveal that the evidence is limited indeed.” 

11. [11]
    The applicant submitted that in respect of item 56, the statistical probability of the DNA coming from someone other than, and unrelated to the applicant, being approximately 1 in 140 was such that it did not have the probative value as discussed in the cases above.

12. [12]
    It was accepted that, in many circumstantial cases, a singular piece of evidence may not have any value of itself, yet when viewed together with other evidence, its probative value and admissibility can be seen. However, counsel for the applicant argued that there was no other evidence or circumstances relied upon that overcame the lack of probative value in the particular DNA evidence in question.

13. [13]
    Further, it was argued that there was a very grave risk of undue prejudice to the applicant if the evidence was admitted. In that regard, counsel referred to R v Lucas (1991) 55 A Crim R 361 at 368, where Hampel J stated:

“Caution must be exercised as a scientific appearance of expert evidence may be overwhelming.  This is particularly the case when the evidence sought to be adduced is of such an esoteric character that there is no real basis on which a non expert jury can evaluate it independently of the experts.”

14. [14]
    It was submitted that in the present case in respect of item 56, there was a considerable, if not great, potential for the evidence to be misused or given too much weight: R v Gardner & Coates [2003] VSC 153 at [20]. Counsel for the applicant submitted that the statistical relevance of the watch DNA evidence was slight and that it could not have any significance in a circumstantial case. Moreover, any probative value it had was outweighed by the prejudicial effect of its admission. The prejudicial consequences were that the DNA statistical evidence found on the pouch (1 in 880,000,000) was likely to overwhelm a separate and rational consideration of the partial DNA profile on the watch and result in it being accorded greater probative value than it warranted. In those circumstances, it was submitted that the proper exercise of judicial discretion required the exclusion of the DNA results of item 56: see R v Hasler, ex parte Attorney-General [1987] 1 Qd R 239.

The respondent’s submissions

15. [15]
    The respondent disputed the contention that the statistical discrimination power of the figure of 1:140 made the prejudicial effect of the evidence greater than its probative value. It was said that the applicant’s argument drew a false comparison with the sorts of numbers typically generated in a full profile (where the random match probability was typically in the realm of billions to one) as though those types of figures represented a necessary condition precedent for admissibility. It was submitted that figures far less discriminatory than 1:140 were common in the era before DNA, but that did not mean the evidence was inadmissible.

16. [16]
    The starting point was to observe that DNA evidence was a species of evidence that tended to prove identity. An analogy was sought to be made with eye witness evidence. It was submitted that typical issues that arise in that area of the law involve distinguishing between “identifications”, for example, where a witness says the equivalent of “the man in the photo shown to me was the man I saw committing the offence” and the evidence less conclusive than that, such as where a witness says “the offender was short, heavyset, blonde, bearded and wearing a checked shirt and jeans”. Such evidence was typically paired with evidence that the applicant fitted that description at the relevant time. It was said that the DNA evidence the subject of the present objection may be thought of as cognate with evidence of the latter sort. (The jurisprudence relating to identification cases was largely concerned with jury directions based on human fallibility in performing identifications; such issues did not arise here.) Such evidence was almost always admissible (although if it was the only evidence linking an accused with the offence, it might be insufficient to carry the case for the Crown).

17. [17]
    Counsel referred to the analysis of principle in Festa (2001) 208 CLR 593 by Gleeson CJ at [9]-[23] with whom Hayne J agreed at [216]. After distinguishing questions as to the admissibility of evidence from those as to whether the totality of the evidence in a case is sufficient to sustain a jury’s verdict or questions as to warnings that need to be given to a jury about the use that may be made of the evidence, Gleeson CJ addressed the discretion to exclude evidence on the Christie principle. His Honour explained the principle in restrictive terms at [14], noting that if evidence

“… is of some, albeit slight, probative value, then it is admissible unless some principle of exclusion comes into play to justify withholding it from a jury’s consideration. It is not enough to say that it is ‘weak’, and , as already mentioned, whether it is weak might depend on what use is made of it.”

18. [18]
    His Honour elaborated further at [22] observing that there are two principal dangers associated with certain identification evidence:

“The first kind of risk concerns the probative value of the evidence.  The second is a risk that the jury will draw an inference about a fact which, even if true, would ordinarily be excluded from evidence.  In that connection, some care is needed in the use of the term ‘prejudice’.  Where it is present, a risk of the second kind is clearly a risk of unfair prejudice.  It is a risk that a fact will be suggested which is of a kind that is ordinarily excluded from evidence in the interests of fairness to an accused.  But prejudice does not arise simply from the tendency of admissible evidence to inculpate an accused.  It is unfair prejudice that is in question.  Where evidence is relevant and of some probative value, prejudice might arise because of a danger that a jury may use the evidence in some manner that goes beyond the probative value it may properly be given.  If there is relevant prejudice of that kind, it lies in the risk of improper use of the evidence, not in the inculpatory consequences of its proper use.  If it were otherwise, probative value would itself be prejudice.  All admissible evidence which supports a prosecution case is prejudicial to an accused in a colloquial sense; but that is not the sense in which the term is used in the context of admissibility.” (footnotes omitted)

19. [19]
    McHugh J made the following pertinent remarks at [51] concerning the issue of weak identification evidence:

“But the weakness of relevant evidence is not a ground for its exclusion.  It is only when the probative value of evidence is outweighed by its prejudicial effect that the Crown can be deprived of the use of relevant but weak evidence.  And evidence is not prejudicial merely because it strengthens the prosecution case.  It is prejudicial only when the jury are likely to give the evidence more weight than it deserves or when the nature or content of the evidence may inflame the jury or divert the jurors from their task.”

20. [20]
    McHugh dealt with the Christie discretion at [63]-[67], noting that the judicial discretion to exclude evidence in criminal cases applies to circumstantial identification evidence as much as it does to positive identification evidence and that, when “a trial judge is asked to exclude circumstantial identification evidence on the ground of unfairness, the judge must examine its probative value and its prejudicial effect”.

21. [21]
    Counsel for the respondent submitted that in the light of the analysis in Festa, there was no identified prejudice to the applicant from the admission of the evidence beyond its probative value.  Assertions that it was “weak” were only relative, and in any event not to the point. Nor was it conceded that the evidence of a random match probability of 1:140 was “weak”; it was argued that it was, for example, much stronger than identification evidence such as that the offender “wore a red shirt”.  It was said that typically jurors must assess the weight of such evidence that is not conclusive of identity and without the benefit of statistical assistance, but that did not mean that where statistical evidence was available, the process was in principle any different, or that there was some numerical threshold below which evidence became admissible.

22. [22]
    A jury’s obligation also required it to assess the evidence in context against the background of other evidence. It was an example of the “Defence Lawyers’ Fallacy”, it was said, to imply that if there were a million people on the Gold Coast, a random match probability of 1:140 implied that there were some 7,000 people as well as the applicant who could equally have committed the offence. The answer to such assertions was that not all those people have the same probability of being guilty: see Butler, Forensic DNA Typing, p 500-501.

23. [23]
    Counsel submitted that it was not necessary in order to determine the question of admissibility to embark upon a consideration of the way in which all the evidence aligned against the applicant – that was a function for the jury. Nevertheless, counsel pointed out that it was of assistance to understand that there was other relevant evidence, without being exhaustive. In the present case, in assessing the significance of the watch DNA, the jury had the benefit of the DNA found on the pouch on the utility belt located nearby. In this regard it was said that, if an attempt was made to provide an alternative explanation for how the DNA was located on the pouch, “the jury would have the watch DNA by which to assess the likelihood of that alternative explanation – the two pieces of evidence mutually self-support”. Similarly, there was evidence of hostility on the part of the applicant towards the deceased, and evidence of his presence on the Gold Coast at the relevant time (given that he lived out of Queensland at the relevant time).  There was also evidence of his former employment by the two deceased in circumstances where he might expect there to be money to be had if a robbery were to be conducted.

24. [24]
    In short, it was contended that there was no prejudice in the present case in the sense in which that term is properly used as identified in Festa, since when seen in the light of all the other evidence, any potential prejudicial effect of the DNA evidence in question was derived from its probative force.  Accordingly, the watch DNA evidence should not be excluded in the exercise of the discretion.

Determination

25. [25]
    While it may not necessarily be the case that the watch DNA evidence has no probative value because of its low statistical probability, its probative value is very limited. Its probative value is said to extend inferentially to the dating of the pouch DNA. As to its probative value in terms of inferentially linking the applicant to the scene and the deceased after 2006 when the applicant had no legitimate occasion for contact with the deceased, that is problematic. This is because, as the respondent’s counsel conceded, the inference that the partial DNA profile on the watch was deposited at the same time as the DNA profile on the pouch would require the jury to be satisfied that the watch DNA belonged to the applicant. Given the very low statistical probability of the watch DNA, its probative value in that regard remains very limited.

26. [26]
    On the other hand, the prejudicial consequences of admitting the evidence of the watch DNA, given its low probative value, would be substantial. There is a danger that the jury would fail to keep in mind the low statistical probability of the watch DNA because the consideration of it would be overwhelmed by the high statistical probability evidence of the pouch DNA, particularly given the proximity of the location of the two DNA profiles. Although there exists other (non-DNA related) circumstantial evidence which may permissibly be used to bolster the probability that the partial DNA profile on the watch was deposited by the applicant, the presence of the two DNA profiles in such close vicinity (a connection which is emphasised on the prosecution case in terms of the timing of the relevant deposits) is likely to colour the jury’s consideration of them such that they are likely to fail to consider and weigh the two distinct statistical probabilities separately.

27. [27]
    This prejudicial effect is compounded because of the significance that the prosecution seeks to place on this evidence in terms of dating the DNA deposits. Indeed, that the respondent seeks to use the watch DNA as part of an argument that it should be inferred that both DNA samples were deposited at the same time serves to highlight the prejudice attended by admission of the watch DNA. There is a risk not only that the statistically low probability of the watch DNA will be overlooked by a failure to keep separate the high statistical probability of the pouch DNA, but also that that will occur in circumstances where the watch DNA will be used to inferentially date the pouch DNA and thereby bolster that DNA evidence. Prejudice is likely to arise from the manner in which the jury will be asked to use the watch DNA in drawing inferences in respect of the pouch DNA, including as to the timing of the deposit of both DNA. There is, therefore, the real danger that a jury may use the watch DNA in a manner that goes beyond the probative value it may properly be given and accord it “more weight than it deserves” to use the words of McHugh J in Festa.  Such a high prejudicial effect does outweigh the low probative value of the partial DNA profile on the watch.

28. [28]
    I do not consider that jury directions can effectively counteract the prejudicial effect of admitting the low probative evidence of the watch DNA. In those circumstances, the application should be granted and the discretion should be exercised to exclude the evidence of the partial DNA profile located on the watch of the deceased.

Order

29. [29]
    It is ordered that the evidence of the watch DNA should be excluded.