R v Bowie (No 1)

1. [1]

   Roxlyn Bowie (“Roxlyn”) disappeared on 5 June 1982. At the time, Roxlyn lived at 93 Euroka Street, Walgett, with her husband, John Bowie (“the accused”), and their two children, a daughter aged six years old and a son who was nearly two years old. She has not been seen or heard from since.

2. [2]

   On 5 October 2019, the accused was arrested and charged with the murder of Roxlyn. The accused pleaded not guilty to the one count of murder on the indictment, an offence contrary to s 18 of the Crimes Act 1900 (NSW) (“Crimes Act”). The trial was listed to commence on 26 September 2022.

3. [3]

   The Crown case is wholly circumstantial. It is alleged that the accused killed Roxlyn sometime between about 7.00pm and about 11.00pm, on 5 June 1982. Although the Crown does not have to establish to any requisite standard the manner in which the accused disposed of Roxlyn’s body, the primary case theory is that the accused disposed of her body by feeding her to pigs.

4. [4]

   On the first day of the trial, the Crown made an application, pursuant to s 79(1) of the Evidence Act 1995 (NSW) (“Evidence Act”), to adduce portions of a report prepared by Sergeant Lauren Atwood (“Sergeant Atwood”), dated 20 May 2021, as expert evidence.

5. [5]

   Objection was taken to the evidence upon the following grounds:

   1. (1)

      The evidence is not relevant, pursuant to s 55 of the Evidence Act;

   2. (2)

      Alternatively, the evidence is inadmissible, pursuant to s 79(1) of the Evidence Act; and

   3. (3)

      Lastly, in the event the Crown satisfied the requirements of s 79(1) of the Evidence Act, the probative value of the evidence is outweighed by the danger of unfair prejudice and, therefore, should be excluded, pursuant to s 137 of the Evidence Act.

6. [6]

   Roxlyn was born in 1951. She met the accused in 1968. They married in August 1971.

7. [7]

   The accused and Roxlyn had three children as a result of their marriage: Brenda, born in 1975; Charlene, born in 1977; and Warren, born in 1980. Charlene tragically died a few days after her birth.

8. [8]

   Whilst employed by NSW Ambulance, the accused was a self-confessed ‘womaniser’ and a heavy drinker. Roxlyn, the accused, and their children moved to Walgett in March 1978. They resided at 93 Euroka Street, Walgett.

9. [9]

   At about 5.00pm, on 5 June 1982, the accused finished his shift at the Ambulance Station, and returned home for a short time. The accused informed Roxlyn that he was going drinking at the Imperial Hotel in Walgett. Roxlyn told the accused that if he went out drinking, she would not be there when he returned home. The accused attended the Walgett RSL Club at about 9.00pm. He returned home at about 11.00pm and discovered that Roxlyn was not there.

10. [10]

    The Crown case is that on 5 June 1982, the accused, by a deliberate act or acts, caused the death of his wife and, at the time he committed the act or acts, had an intention to kill her. The evidence does not establish the manner in which Roxlyn was killed, or the precise mechanism used.

11. [11]

    In 1982, the accused had an interest in a “piggery” located at the Walgett Meat Works, which was a short distance out of town. There appears to be no dispute that he was responsible for feeding the pigs on numerous occasions.

12. [12]

    Over the years that followed Roxlyn’s disappearance, the accused is alleged to have made statements to various persons about pigs “not leaving evidence”. Those representations are relied upon by the Crown as admissions that the accused disposed of Roxlyn’s body. They constitute part of the circumstantial case to establish that Roxlyn is dead and that the accused caused her death by way of a deliberate act or acts. In summary, those representations are as follow:

Summary of Expert Report Produced by Lauren Atwood

1. [13]

   The Crown relies upon the Preliminary Report: ‘A casework study: The effect of the porcine digestive process on animal carcasses and human teeth’, prepared by Sergeant Atwood. Sergeant Atwood was not called on the voir dire.

2. [14]

   It must be noted at the outset that there is no published peer reviewed scientific research that deals with the topic of what, if anything, would be left behind if human remains were ingested by pigs. Sergeant Atwood stated that:

3. [15]

   The Report, in summary, examined the effect of the domestic porcine digestive process on animal carcasses and human teeth. In particular, Sergeant Atwood sought to answer the following questions:

   1. (1)

      Would pigs consume a human body?

   2. (2)

      Would pigs consume flesh, bone and human teeth?

   3. (3)

      What would be left behind (if anything) if they consumed the above?

4. [16]

   Based on observations made during the experiment (with deference to the limited sample size), the findings relating to the research questions were as follows:

   1. (1)

      The pigs utilised in the study “immediately and enthusiastically fed on a slurry of cooked animal carcass, wheat and water”. Changes to the type of carcass (porcine vs kangaroo) did not impact the feeding behaviours of the pigs. The pigs fed on and consumed soft feed (what/meat) and hard feed (bones and teeth);

   2. (2)

      The pigs immediately and enthusiastically fed on a ‘wounded’ porcine carcass, with preference to feed on the flesh accessible through the wound site;

   3. (3)

      The pigs did not immediately and enthusiastically feed on an intact porcine carcass. The pigs waited nine days before feeding on the decomposed carcass. The pigs fed on a consumed soft feed (flesh, organs) and hard feed (skull);

   4. (4)

      It is possible to find components of an animal carcass/human teeth both uneaten from the pig enclosure, and post-digestion in the faeces of the pigs. Only 29% of all human teeth fed to the pigs were recovered (either eaten or uneaten) during the study. The remaining 71% of human teeth were not recovered;

   5. (5)

      A total of 81% of the recovered human teeth from the study were deemed suitable for identification by a forensic odontologist; and

   6. (6)

      The opinions of forensic biologists and forensic archaeologists are pending.

5. [17]

   Sergeant Atwood concluded that the findings of the experiment, whilst comprising a small sample size, indicated that pigs will feed on, and are capable of digesting, flesh, teeth, and bones effectively to the point that identifiable remains may be difficult to recover due to their size. However, intact, identifiable remnants (such as teeth, or large tooth fragments) do survive the porcine digestive process.

6. [18]

   Ethical considerations limited the way in which the experiment was conducted. Porcine and kangaroo carcasses, as opposed to human carcasses, were used in the experiment. The reference made to the capacity of pigs to digest flesh and bones related to their capacity to digest porcine and kangaroo flesh and bones. Only the teeth used in the experiment were human teeth, and 29% of all human teeth fed to the pigs were recovered.

7. [19]

   Some information was provided to Sergeant Atwood by way of background. The information included that the accused fed the pigs a mixture of kangaroo carcasses, wheat, and water (also referred to as “slurry”). This slurry was contained in a 44-gallon drum and cooked on a campfire overnight. I pause to note that there is a complete absence of evidence that the accused, before feeding Roxlyn’s body to the pigs, engaged in a process of placing the corpse in a 44-gallon drum and cooking it on a campfire.

8. [20]

   Further, the background information provided to Sergeant Atwood was erroneous insofar as it referred to the accused keeping a herd of approximately 40 to 50 wild pigs. The pigs kept by the accused were domestic pigs. It is unclear whether much turns on this in terms of the accuracy of the experiment. The experiment conducted by Sergeant Atwood examined the effect of the domestic porcine digestive processes, not those of wild pigs.

9. [21]

   The experiment conducted by Sergeant Atwood involved two 44-gallon drums, which were cut in half and altered to become suitable containers to hold a wood fire. Drum 1 contained 45kg of kangaroo bone cut-offs, 5.2kg of wheat chaff and water added to cover. Drum 2 contain 30kg of dismembered porcine carcass, 4.4kg of wheat and water added to cover. Human teeth were individually contained within small, labelled cotton bags. The cotton bags were placed inside a larger mesh bag, which was subsequently submerged in Drum 2.

10. [22]

    Thermocouples were placed both within the “fire drums” and the “feed drums”. Lids were placed on drums and weighted with a brick. Fires were ignited and monitored. Thermocouples recorded data from 29 minutes post ignition to 18 hours into the cook.

11. [23]

    Fires were stoked and fuel was added intermittently over a period of eight hours. Fire was left to cool naturally after that. The fire burned for 23 hours before feed was removed. Thereafter, the feed was portioned out and vacuum sealed and kept refrigerated until use. The five feed stages are set out at page 14 of Sergeant Atwood’s Report.

12. [24]

    On 13 October 2020, the pig enclosure was excavated and searched for remaining bones and teeth. The top 5cm of the ground was removed and wet sifted. Suspected bones and teeth were collected, documented, and stored.

13. [25]

    As indicated above, opinions from forensic biologists and forensic archaeologists are pending. Those opinions were not before me during the hearing of this application. Furthermore, the findings of this experiment comprised of an unidentified and small sample size.

Consideration

1. [26]

   The accused submitted that the evidence is not relevant because the findings giving rise to the opinion expressed are based upon an experiment that involved kangaroo and porcine carcasses, as opposed to human cadavers. The evidence is, therefore, not capable of rationally affecting, directly or indirectly, the assessment of the probability of the existence of a fact in issue in the proceedings, namely, whether the accused fed Roxlyn’s body to pigs.

2. [27]

   There is no evidence in the Crown case that establishes any basis to conclude that because pigs will consume kangaroo or porcine flesh, they are, therefore, more likely to be able to consume human flesh.

3. [28]

   In answer to this submission, and during oral argument, the Crown pointed to p 256 of the paginated bundle on the voir dire, where Sergeant Atwood stated:

4. [29]

   However, the Crown also accepted that a review of the literature revealed that there has been no peer reviewed research into this topic. It is necessary to set out the exchange during oral argument, because, ultimately, the Crown did not rely upon the statement that “this is a commonly accepted substitute in anatomical, medical and forensic research”, in support of the argument:

5. [30]

   There is no evidence on this application to support a contention that the use of kangaroo and porcine carcasses was an appropriate or accepted substitute for the purpose of the experiment conducted in this case. Put another way, no evidence was adduced to support the proposition that the opinions expressed in the report were capable of supporting an inference that pigs could eat human flesh and bones without leaving a trace.

6. [31]

   Section 55 of the Evidence Act provides:

7. [32]

   In Dasreef Pty Ltd v Hawchar (2011) 243 CLR 588; [2011] HCA 21, French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ said at [31]:

8. [33]

   In Honeysett v The Queen (2014) 253 CLR 122 (“Honeysett”); [2014] HCA 29, French CJ, Kiefel, Bell, Gageler and Keane JJ stated (at [25]):

9. [34]

   Here, the Crown relies upon the report of Sergeant Atwood to support the contention that after killing Roxlyn, the accused disposed of her body by feeding it to pigs, thereby leaving no trace. Although the Crown does not have to establish the method of disposal, the Crown relies upon this possible method to explain the fact that the investigating authorities have been unable to find any trace of Roxlyn’s remains. Further, the Crown relies upon the report to rebut the dismissal of the representations made by the accused (as set out above) as mere fantasy or storytelling.

10. [35]

    In the absence of further material, I am unable to draw the inference that pigs eat human flesh and bone without leaving a trace. Such an inference is simply not available from the findings flowing from the single experiment conducted by Sergeant Atwood. Without more, the opinion expressed that “pigs will feed on and are capable of digesting porcine and kangaroo flesh and bones effectively to the point that identifiable remains may be difficult to recover due to size”, is not relevant to any fact in issue in the trial. It follows that this opinion is not admissible pursuant to s 55 of the Evidence Act.

11. [36]

    The opinion that “pigs will feed on and are capable of digesting human teeth effectively to the point that identifiable remains may be difficult to recover due to size”, is relevant to a fact in issue. However, for the reasons set out below, I find that each of the opinions expressed by Sergeant Atwood is not admissible

12. [37]

    In written submissions relied upon by the Crown, it was contended that the opinions expressed by Sergeant Atwood are based on specialised knowledge. The Crown stated the following: “Sergeant Atwood has specialised knowledge based on her experience as a scientific researcher and her specific study into the questions posed of her means that she is now and ‘expert’ in this field”. [2]

13. [38]

    The reasoning appears to be grounded in the fact that Sergeant Atwood has qualifications as a researcher and “she conducted a study into the questions posed to her by the OIC”. [3] The Crown relies upon Sergeant Atwood’s specialised knowledge (as a scientific researcher); her experience (conducting this sole experiment); and her study (noting that there is no peer reviewed literature into the specific area about which she was asked to give an opinion), to conclude that the opinions expressed are admissible as an exception to the opinion rule.

14. [39]

    That reasoning is wholly misconceived for reasons that I will set out in due course. Before I set out those reasons, I will briefly summarise the submissions made on behalf of the accused. The accused focused on the lack of evidence to establish that there is a field of expertise on the topic. Furthermore, Sergeant Atwood’s conclusions were drawn from a single experiment. The opinions are not based upon a field of study, but rather are the results of one experiment. Accordingly, the opinion is not wholly or substantially based on specialised knowledge.

15. [40]

    Pursuant to s 76 of the Evidence Act, evidence of an opinion is not admissible to prove the existence of a fact about the existence of which the opinion was expressed (“the opinion rule”).

16. [41]

    Section 79 of the Evidence Act provides a specific exception to the opinion rule. It states:

17. [42]

    There is no dispute that Sergeant Atwood has extensive training and experience as a scientific researcher. However, upon drilling down, and in an effort to identify with some clarity the specialised knowledge upon which the opinion is said to be wholly or substantially based, it was conceded by the Crown, during oral argument, that there is no “prior” area of specialised knowledge. It is necessary to set out the relevant exchange in full:

18. [43]

    A failure to demonstrate that an opinion expressed by a witness is founded on the witness’s specialised knowledge based on their training, study or experience is a matter that goes to the admissibility of the evidence, not its weight: Honeysett at [42].

19. [44]

    As for what is meant by specialised knowledge, in Honeysett, the Court stated at [23] (footnotes omitted):

20. [45]

    In relation to what constitutes a field of expertise, the Australian Law Reform Commission (“ALRC”) in Evidence (Interim) [1985] ALRC 26 stated, at 743:

21. [46]

    Section 79 does not refer to a field of expertise, but rather “specialised knowledge based on training, study or experience”. The question as to whether specialised knowledge requires an independent means of gauging the reliability and validity of an opinion based on that knowledge, was not decided in Honeysett. Instead, the Court held that “in light of the concession [by the respondent] … Professor Henneberg’s specialised knowledge was confined to anatomy, the appeal does not provide the occasion to consider the appellants larger challenge respecting the requirement of an independent means of validation before an opinion may be found to be based on ‘specialised knowledge’”: at [42].

22. [47]

    The decision by the Court in Honeysett not to resolve the issue was referred to in Xie v The Queen [2021] NSWCCA 1, where the Court said at [297]-[301]:

23. [48]

    In Tuite v The Queen [2015] VSCA 148 (“Tuite”), the Victorian Court of Appeal, dismissing the appeal, held that the language of s 79(1) of the Evidence Act 2008 (Vic) left no room for reading in a test of evidentiary reliability as a condition of admissibility. [7] Section 79(1) contains its own specification of the requisite foundation of the witness’s ‘knowledge’, namely, that the knowledge must be based on the person’s training, study or experience.

24. [49]

    The accused was charged with aggravated burglary, rape, indecent assault and intentionally causing injury. Expert opinion evidence was to be called at his trial about the analysis of DNA samples from the crime scene and what was said to be the similarities between those samples and a DNA sample provided by the accused following an unrelated conviction. The DNA evidence was to be presented in the usual form of a likelihood ratio. Those ratios had been calculated using a recently developed analytical method known as STRmix, which was introduced into Victoria in March 2013. At a pre-trial hearing, the accused challenged the admissibility of the proposed DNA evidence on the ground that the new methodology was not sufficiently reliable for use in criminal trials, was largely untested, and had not been generally accepted by the forensic science community.

25. [50]

    The Court in Tuite concluded, at [77]:

26. [51]

    In dismissing the appeal, the Court held that the question was whether the judge’s decision was reasonably open, not whether it was correct. Although, the Court held that the language of s 79(1) left no room for reading in a test of evidentiary reliability as a condition of admissibility.

27. [52]

    The applicant in that case acknowledged that the only appellate court which had directly addressed the question of construction concluded that reliability fell outside the scope of s 79(1). In R v Tang (2006) 65 NSWLR 681 (“Tang”), Spigelman CJ (with whom Simpson and Adams JJ agreed) said:

28. [53]

    The Court also referred to the decision in R v McIntyre [2001] NSWSC 311 (“McIntyre”), where Bell J (sitting as the trial judge) ruled that she would not permit the voir dire to extend to a challenge based upon a contention that the Profiler Plus system was not reliable. In her Honour’s view, at [14]:

29. [54]

    The Court held that the conclusion reached by the New South Wales courts – first in McIntyre and then in Tang – was correct. In coming to that conclusion, the Court also relied on the observations made by Gleeson CJ in HG (1999) 197 CLR 414 at 427 [40]: “it is the language of s79(1) which has to be applied”, acknowledging that the High Court has repeatedly emphasised that statutory interpretation begins and ends with the words which Parliament has used.

30. [55]

    The Court concluded, at [72]:

31. [56]

    What is to be made of the requirement that an inference must be drawn “on good grounds”? If, for the purposes of s 79(1), the requirement does not mean ‘reliability’ or ‘validation’, does it simply mean that the opinion is specialised because it is based on a person’s training, study or experience? If, for instance, evidence is sought to be adduced by a party from an astrologer to explain how the movement of the stars and planets may have impacted upon an offender’s behaviour, is that admissible under s 79(1)? The astrologer has had years of experience in reading astrological charts. Is that sufficient?

32. [57]

    It is not appropriate, or necessary, in this judgment to comment on the correctness or otherwise of the approach taken to date to the application of s 79(1) of the Evidence Act. The cases of Tuite, Tang, McIntyre, and Xie are binding. However, determining admissibility under s79(1) should, at the very least, involve a rigorous assessment of the scope and limits of the asserted specialised knowledge. Professor Gary Edmond of the University of New South Wales, together with other academics and scientists, have long argued that:

33. [58]

    In Tuite, although the Court held that s 79(1) of the Evidence Act left no room for reading in a test of evidentiary reliability as a condition of admissibility, the Court strongly emphasised that the touchstone of reliability for scientific evidence must be trustworthiness, and trustworthiness depends on validation.

34. [59]

    Rigorous assessment of evidentiary reliability when expert opinion evidence is proposed to be called, is a matter of first importance to the integrity and fairness of the criminal justice system. The question of the reliability of opinion evidence fell to be determined not under s 79(1), but rather as part of the assessment which the Court undertook for the purposes of s 137 of the Evidence Act: see see Tuite at [10] and [82].

35. [60]

    Far from excluding an assessment of reliability when considering the admissibility of expert evidence, the Court in Tuite went to great lengths to emphasise the importance of a vigorous assessment of evidentiary reliability when expert opinion evidence was to be called. The Court held that the touchstone of reliability for scientific evidence was trustworthiness, which depended on validation. Ideally, there should be proof of both in-house validation, and independent external validation: see [101]-[102].

36. [61]

    Of course, the Court was deciding the issue prior to the decision of IMM v The Queen (2016) 257 CLR 300; [2016] HCA 14 (“IMM”), in which the High Court held [at 54]:

37. [62]

    If the assessment of reliability is outside the scope of s 79(1), and not a relevant consideration when assessing probative value under s 137, is there any room for a trial judge to assess reliability when determining the admissibility of expert evidence?

38. [63]

    IMM left open the possibility that an assessment of the “reliability’ of evidence may be permissible as part of the enquiry into the “danger of unfair prejudice” under s 137: at [57].

39. [64]

    It seems to me that this is particularly so in a case where the subject evidence is asserted to be “expert evidence”, that is, an opinion based on specialised knowledge. I say “particularly so”, because of the caution expressed by his Honour Hidden J in Morgan v The Queen (2011) 215 A Crim R 33, at [145], of the dangers of the “white coat effect”.

40. [65]

    As Dawson J stated in Murphy v The Queen (1989) 107 CLR 94; [1989] HCA 28 at [216]:

41. [66]

    As the Supreme Court of Canada said in R v Trochym [2007] 1 SCR 239, at 260 [27]: “Evidence that is not sufficiently reliable is likely to undermine the fundamental fairness of the criminal process”. The dangers of ‘junk science’ was pointed out by an earlier decision of the Supreme Court of Canada in R v Mohan [1994] 2 SCR 9, at 21:

42. [67]

    The question of the reliability of an opinion can, in my view, be a relevant consideration when determining the danger of unfair prejudice under s 137 of the Evidence Act.

Conclusion

1. [68]

   Returning then to the evidence the subject of this application. As indicated above, I am not persuaded that evidence that the pigs used in the experiment will feed on, and are capable of digesting flesh and bones of porcine and kangaroo carcasses, is relevant to any issue in this trial. It is, therefore, inadmissible, pursuant to s 55 of the Evidence Act.

2. [69]

   Alternatively, the Crown has failed to identify with some precision the area of specialised knowledge upon which the opinion is wholly are substantially based. Indeed, the Crown has conceded that there is no “prior” area of specialised knowledge; the witness is a “pioneer” in the field. I am not satisfied that an area of specialised knowledge has been identified. I am therefore not persuaded that Sergeant Atwood’s opinion is wholly or substantially based on an area of specialised knowledge.

3. [70]

   Had I been satisfied that this evidence was relevant and admissible pursuant to section 79(1) of the Evidence Act, I would have excluded it pursuant to s 137 of the Evidence Act. The probative value of the evidence, based as it is on the results of one experiment, is outweighed by the danger of unfair prejudice, namely, the danger that the jury will misuse the evidence, or give it disproportionate weight.

4. [71]

   The evidence from Sergeant Atwood that pigs will feed on human teeth effectively to the point that identifiable remains may be difficult to recover due to size, is relevant to a fact in issue, namely, the disposal of Roxlyn’s body without leaving a trace. However, that evidence goes to the capacity of pigs to digest human teeth, as distinct from human flesh and bone. Although relevant, its probative value is slight.

5. [72]

   The evidence is, however, inadmissible under s 79(1) of the Evidence Act. The Crown has failed to establish an area of specialised knowledge. There is no peer reviewed literature from which any conclusion can be drawn as to the comparability of the findings from this experiment (conducted with porcine and kangaroo carcasses) and the issue of whether pigs can feed on human flesh and bones effectively to the point that identifiable remains may be difficult to recover. Put another way, I am not satisfied that the opinion expressed is wholly or substantially based on an area of specialised knowledge. Accordingly, the evidence is not admissible.