Risky business: Predicting recidivism

Psychiatry, Psychology and Law

Volumn 17, Issue 1, 2010, Pages 88-120

  Glazebrook, Justice Susan ^a  

^a New Zealand Court of Appeal ^*  (New Zealand)

Author keywords

Expert evidence; Health assessors' reports; Risk assessment tools; Risk of recidivism

Indexed keywords

EID: 75649135842     PISSN: 13218719     EISSN: None     Source Type: Journal    
DOI: 10.1080/13218710903040421     Document Type: Article

References (305)

1. This paper was presented at the conference of the Australian Psychological Society College of Forensic Psychologists "Advancing Forensic Psychology", held in Melbourne from 25-28 February 2009. I acknowledge the invaluable assistance of the Court of Appeal law clerks, Jane Standage and Natasha Caldwell, in the research for and writing of this paper. Any errors remain my own and the views expressed are my own and are not to be taken as the views of the Court of Appeal or the New Zealand judiciary.
2. E Beecher-Monas, Evaluating Scientific Evidence: An Interdisciplinary Framework for Intellectual Due Process (Cambridge University Press, Cambridge 2007) 155-56.
3. Public concern is often aroused by high profile tragic cases of reoffending which appear to have been preventable had a number of people addressed what are seen to have been known risks. One suspects that in many of these cases hindsight bias exaggerates the foreseeability of events, but this does not make the fears any less real.
4. These tools join a large body of other risk assessment tools. A Google search undertaken at the end of November 2008 showed a wide variety of such tools. The first tool for assessing the risk of serious and violent sexual offending was number 22 on the list. Those earlier on the list included a US Federal Aviation Administration tool to assess flight risks. Bioprotection New Zealand contributed a number of weed risk assessment tools. There was an ecological risk assessment tool used by the United States Department of Energy. The Federal Deposit Insurance Corporation had a tool relating to information security. New Zealand Aid had a conflict risk tool for use in development aid decisions. There was a risk assessment tool for estimating the ten year risk of having a heart attack and finally a tool to assess the risk of fusarium mycotoxin (associated with fungal infection in cereal crops) in the United Kingdom.
5. K McMaster and A Wells, 'From Uncertainty to Certainty - is it Possible?' in K. McMaster and L. Bakker (eds), Will They DoIt Again?: Assessing and Managing Risk (HallMcMaster and Associates, Christchurch 2006) 7 at 11.
6. B McSherry, P Keyzer and A Freiberg, Preventive Detention for 'Dangerous' Offenders in Australia: A Critical Analysis and Proposals for Policy Development Report to the Criminology Research Council (December 2006) at [1.2].
7. D Riley, 'From Mysticism to Science: An Overview of Risk Prediction', in K. McMaster and L. Bakker (eds), Will They Do It Again? Assessing and Managing Risk (Hall McMaster and Associates, Christchurch 2006) 24 at 35.
8. [2007] 2 NZLR 627.
9. See BE Harcourt, Against Prediction: Profiling Policing and Punishing in an Actuarial Age (University of Chicago Press, Chicago 2007) for a criticism of the use of actuarial methods in the criminal justice system, including with regard to policing.
10. For example, s 7 of the Parole Act 2002 (NZ) provides that the paramount consideration for the Parole Board is the safety of the community. Accordingly, the Court of Appeal has emphasised that the safety of the community, rather than aims of deterrence, should be the paramount consideration for the Parole Board when considering applications for release on parole or home detention.
11. See Reid v The New Zealand Parole Board (2006) 22 CRNZ 743 at [38].
12. The assessment of risk of offending is also relevant in certain civil contexts - for example in parenting decisions.
13. Although, if remand time before trial is likely to be long then this may be a matter that may need rethinking. This is particularly the case as conditions on remand (in some New Zealand prisons at least) may not be ideal, with long lock-down periods and no access to programmes.
14. Section 7 of the Bail Act 2000 (NZ) provides that a defendant is entitled to bail as of right when: charged with an offence that is not punishable by imprisonment; charged with an offence for which the maximum punishment is less than three years imprisonment, unless that offence relates to assault on a child or by a male on a female; charged with an offence under a number of listed provisions in the Crimes Act 1961.
15. Section 7(5) Bail Act 2000 (NZ).
16. Note the threshold of mere "risk" was introduced pursuant to the Bail Amendment Act 2008 (NZ), which was introduced and passed under urgency under the National-led government on 9 December 2008. The provision was amended from the higher threshold of a "real and significant risk", a provision introduced by the previous Labour-led government.
17. 13. See s 8(1) of the Bail Act 2000 (NZ). There are a number of other matters, such as the seriousness of the offence and the likelihood of conviction, that, under s 8(2)(5) the court may take into account. There are also more restrictive provisions relating to bail in certain circumstances where higher risk categories of offenders, such as repeat violent offenders, are concerned - see ss 9-12 of the Bail Act 2000 (NZ).
18. See generally Australian Institute of Criminology, Review of Bail Legislation, accessed 16 February 2009.
19. Bail Act 1992 (ACT) s 22 (1)(c)(ii).
20. Bail Act 1978 (NSW) s 32 (1)(b)(iv).
21. Bail Act 1982 (NT) s 24(1)(c)(iii).
22. Bail Act 1980 (Qld) s 16 (1)(a)(ii)(A).
23. Bail Act 1985 (SA) s 10 (1)(b)(ii).
24. See the High Court of Australia decision in Veen v The Queen (No 2) (1988) 164 CLR 465 at 472-475.
25. For more detail see B McSherry, P Keyzer, and A Freiberg, above n 4 at [2.1]-[2.24].
26. For the New Zealand position see R v Ward [1976] 1 NZLR 588 (CA), s 7(1)(g) of the Sentencing Act 2002 and the discussion in Hall's Sentencing at I.3.4.
27. See B McSherry, P Keyzer and A Freiberg, ibid at [2.3.2].
28. See s 87(2) of the Sentencing Act 2002 (NZ). On 1 July 2002 the Sentencing Act 2002 (NZ) replaced the Criminal Justice Act 1985 (NZ), which had previously governed preventive detention in New Zealand. The Sentencing Act 2002 made significant changes by increasing the number of qualifying sexual and violent offences, lowering the age of eligibility from 21 years to 18 years at the time of the offence, and removing the requirement that an offender must have previously been convicted of a specific offence.
29. 19. These provisions can include some measure of detention. Under s 16(h) of Victoria's Serious Sex Offenders Monitoring Act 2005, amended in 2008 also to include sex Offenders with adult victims, there a condition, equivalent to a home detention option, that the offender be electronically monitored.
30. For more detail on these regimes and a discussion of the Queensland case law highlighting issues with the regime see B McSherry, P Keyzer and A Freiberg (n 4) at [4.3]-[4.5] and [6.3]-[6.4].
31. One of the main issues seems to have been inadequate access to treatment and lack of coordinated implementation of plans for rehabilitation. There have also been concerns about procedural fairness aspects with insuficient time for prisoners to prepare for hearings, and concerns about the availability and funding of legal representation and expert witnesses. In New Zealand, personal information about child sex Offenders can be provided by the Department of Corrections to the Ministry of Social Development, the New Zealand Police, Housing New Zealand Corporation, and the Department of Child, Youth and Family Services pursuant to s 182A-E Corrections Act 2004 (NZ).There is currently no public register of sexual offenders in New Zealand. However, in both 1996 and 2004, former Act MP Deborah Coddington published a Paedophile and Sex Offender Index for New Zealand, and the Sensible Sentencing Trust has established a paedophile and sex offender database.
32. Significantly, in Brown v Attorney General [2006] DCR 630 (NZ), the District Court ordered a $25,000 award of damages to a convicted sex offender after finding that the publication and distribution of a flyer, publishing both a photograph and the address of the offender, amounted to an invasion of his privacy.
33. See s 107I(1) of the Parole Act 2002 (NZ).
34. See s 107I(2) and s 107F(2) of the Parole Act 2002 (NZ) for the factors that must be considered.
35. See also Grieve v Chief Executive of the Department of Corrections (2005) 22 CRNZ 20 (CA) at [14] and [25].
36. See Belcher v Chief Executive of the Department of Corrections [2007] 1 NZLR 507 (CA) [11].
37. See ss 107I(4) and (5) of the Parole Act 2002 (NZ).
38. Belcher (n 23) at [108].
39. The New Zealand Parole Board is responsible for imposing appropriate restrictions on the offenders who are subject to ESOs. At their most draconian, the conditions can amount to home detention and a requirement that the offender be accompanied and monitored for up to 24 hours a day by a person approved by the Chief Executive of the Department of Corrections can be imposed under the legislation. This requirement can, however, only apply within the flrst 12 months of the orders.
40. See s 107J(1)(b), s 107K(2) of the Parole Act 2000 (NZ)
41. and see s 107K(3)(b) of the Parole Act 2000 (NZ).
42. J Vess, 'Preparing Practitioners for Assessing and Managing Risk' in K. McMaster and L. Bakker (eds) Will They Do At again? (n 4) 173.
43. International Covenant of Civil and Political Rights 1966.
44. It must be emphasised that under the statutory provisions, rights can be limited in certain circumstances. Section 5 of the New Zealand Bill of Rights Act 1990 provides that the rights and freedoms contained in the Bill of Rights may be subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. Both s 7(2) of the Charter of Human Rights and Responsibilities Act 2006(Vic) and s 28(1) Human Rights Act 2004 (ACT) provide that a human right may be subject under law only to such reasonable limits as can be demonstrably justified in a free and democratic society.
45. See New Zealand Bill of Rights Act 1990.
46. See also s 22(1) Human Rights Act 2004 (ACT); s 25(1) Charter of Human Rights and Responsibilities Act 2006 (Vic).
47. McSherry, Keyzer and Freiberg (n 4) [7.2.1].
48. The principle, aside from being a fundamental common law principle, is also enshrined in Art 14(1) of the ICCPR, and more particularly for our purposes, in 21(1) of the Human Rights Act 2004 (ACT); s 24(1) of the Charter of Human Rights and Responsibilities Act 2006 (Vic) and s 25(a) of the New Zealand Bill of Rights Act 1990.
49. This is the principle that sentencingmust be proportional to the offence of which the person has been convicted.
50. McSherry, Keyzer and Freiberg (n 4) [7.2.2].
51. Orders such as supervision orders are more fiexible than detentionmeasures and can be tailored to be more consistent with proportionality. For instance, a convicted sex offender may be prevented fromhaving employment in an environment with children, but not prevented from working in other areas.
52. McSherry, Keyzer and Freiberg (n 4) [7.2.3].
53. Article 14(7) of the ICCPR states that no one shall be liable to be tried or punished again for an offence for which he or she has already been convicted in accordance with the law.
54. See s 25(d) New Zealand Bill of Rights Act 1990;
55. s 22(2) Human Rights Act 2004 (ACT);
56. s 25(2)(k) Charter of Human Rights and Responsibilities Act 2006 (Vic); Art 14(3)(g) ICCPR.
57. See s 19 New Zealand Bill of Rights Act 1990; s 8 Human Rights Act 2004 (ACT); s 8 Charter of Human Rights and Responsibilities Act 2006 (Vic); Art 26 ICCPR. There also exists a range of federal antidiscrimination statutes in Australia. See generally Age Discrimination Act 2004 (Aus); Racial Discrimination Act 1975 (Aus); Sex Discrimination Act 1984 (Aus); Disability Discrimination Act 1992 (Aus).
58. P Marshall, 'An Analysis of Preventive Detention for Serious Offenders' (2007) 13 Auckland U L Rev 116, 127.
59. See generally C Slobogin, Minding Justice (Harvard University Press, Cambridge Massachusetts 2006) 103.
60. N Morris, 'Incapacitation Within Limits' in A. von Hirsch and A. Ashworth (eds), Principled Sentencing (2nd edn, Hart Publishing, Oxford 1998) 107.
61. For discussion of general international human rights obligations see M Freeman and G Van Ert, International Human Rights Law (Irwin Law, Toronto 2004) 24-85.
62. See also International Law Commission Draft Articles on State Responsibility accessed 24 February 2009.
63. For a discussion of state responsibilitywith regards to torture see: Advisory Council of Jurists of the Asia Pacific Forum of National Human Rights Institutions, Reference on Torture: Final Report (2005) 169-72 accessed 25 February 2009.
64. (1988) 164 CLR 465.
65. The Australian High Court is Australia's court of final appeal. In New Zealand, the High Court is the equivalent of the state Supreme Courts in Australia. The New Zealand Court of Appeal is the intermediate appellate court, with the Supreme Court as the highest New Zealand court, replacing the Privy Council in 2004.
66. Veen (n 38) 486.
67. For discussion of the decision see B McSherry, 'Indefinite and Preventative Detention Legislation: From Caution to Open Door' (2005) 29 Crim L.J 94, 96.
68. (1996) 189 CLR 51.
69. Ibid 97 and 121-22 respectively.
70. Here, the judges were referring to the power to impose an indefinite sentence at the time a crime had been committed. However, Toohey, Gaudron, McHugh and Gummow JJ, all delivering majority judgments, held that s 5 of the Community Protection Act 1994 (NSW), which enabled the Supreme Court of New South Wales to provide for the preventative detention of an offender post sentence compromised the integrity of the judicial system because it obliged the Supreme Court of New South Wales to exercise a non-judicial function.
71. R v D [2003] 1 NZLR 41 (CA) at [26]-[31]. Emphasis was placed on the legislative history surrounding the enactment of the Sentencing Act 2002, the fact that the consistency of the sentence of preventative detention with the ICCPR was considered in the Select Committee process, and that the Justice and Electoral Committee was satisfied the provisions complied with the ICCPR.
72. Emphasis was placed on the legislative history surrounding the enactment of the Sentencing Act 2002, the fact that the consistency of the sentence of preventative detention with the ICCPR was considered in the Select Committee process, and that the Justice and Electoral Committee was satisfied the provisions complied with the ICCPR.
73. HRC, Communication No 1090/2002, UN Doc CCPR/C/79/D/1090/2002 (15 December 2003).
74. Ibid 7.3.
75. Note, however, there were a variety of differing opinions delivered in the Communication.
76. (2004) 223 CLR 575.
77. Ibid [24], [109], [217]-[219].
78. Ibid [97], [225].
79. Belcher (n 23).
80. Ibid [49].
81. It must be noted, however, that the Court declined to consider whether such a breach could be justified under s 5 of the Act, which, as noted above at n 28, provides that the rights and freedoms contained in the Bill of Rights may be subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.
82. See ibid [59].
83. Section 4 of theNewZealand Bill of Rights Act 1990 provides that no court shall hold any enactment to be impliedly repealed or revoked, or decline to apply any provision of the enactment by reason of the fact that it is inconsistent with the New Zealand Bill of Rights 1990. For the correct statutory interpretation of the provisions in the Act see R v Hansen [2007] 3 NZLR 1 (SC).
84. For example the Dangerous Sexual Offenders Act 2006 (WA), s 7 requires "acceptable and cogent evidence" "to a high degree of probability". Under the Sentencing Act 1991, s 18B (Vic), the court must be satisfied to a high degree of probability that the offender is a serious danger to the community. Under the Tasmanian Sentencing Act 1997: indefinite sentences should be confined to exceptional cases and there must be cogent evidence that the accused is a constant danger to the community. The Dangerous Prisoners (Sexual Offenders) Act 2003 (Qld), s 13 states that the court must be satisfied to a high degree of probability that there is an unacceptable risk that the prisoner will commit a serious sexual offence if released or if released or if released without a supervision order. The High Court of Australia in Fardon (n 46) emphasised the importance of the statutory requirement for acceptable and cogent evidence that was provided by the Act.
85. See (n 48).
86. The Sentencing Act 2002, s 87(2)(c) (NZ) provides that when imposing a sentence of preventative detention the court must be satisfied that the person is likely to commit another qualifying sexual or violent offence.
87. The Court of Appeal in R v Leitch [1998]1 NZLR 420 at 428 held that the term "satisfied", also used in the previous legislation, called for the exercise of judgement by the sentencing court, and that it was inapt to import notions of a burden of proof, and setting of a particular standard, such as reasonable doubt, into the requirement. Under s 107I(2) Parole Act 2002 a sentencing court is able to make an extended supervision order if satisfied that the offender is likely to commit any of the relevant offences.
88. In Chief Executive of the Department of Corrections v McDonnell HC AK CRI 2005-404-000239 at [33] Baragwanath J for the High Court noted that the Crown must establish beyond reasonable doubt that, given the actual and potential constraints on the liberty of the subject that occur with the imposition of an ESO, the ESO is necessary and proportionate to the risk presented. This has not yet been commented upon by the Court of Appeal or Supreme Court, but the decision is currently under appeal to the New Zealand Court of Appeal.
89. As is suggested by M Wood and J Oglo., 'Victoria's Serious Sex Offenders Monitoring Act 2005: Implications for the Accuracy of Sex Offender Risk Assessment' (2006) 13(2) Psychiatry, Psychology and Law 182, 188.
90. Beecher-Monas (n 2) 139-140.
91. McSherry, Keyzer, and Freiberg (n 4) [1.3] (citing J Monahan, The Clinical Prediction of Violent Behaviour (US Department of Health and Human Services, 1981).
92. GB Melton, et al., Psychological Evaluations for the Courts: A Handbook for Mental Health Professionals and Lawyers (3rd edn Guilford Press, New York 2007) 306, suggests that these error figures may be too high as they may overestimate the percentage of false positive statistics. Some clinicians too were clearly better than others at predicting outcomes.
93. McMaster and Wells (n 4) 12.
94. Melton (n 55) 301.
95. MA Campbell, S French and P Gendreau, 'Assessing the Utility of Risk Assessment Tools and Personality Measures in the Prediction of Violent Recidivism for Adult Offenders' (2007) Public Safety 1 http://www.publicsafety.gc.ca/res/cor/rep/_./risk-tools-2007-04-eng.pdf accessed 19 February 2009.
96. Melton (n 55) 308.
97. See I Freckelton and H Selby, Expert Evidence: Law, Practice, Procedure and Advocacy (4th edn, Thomson, Sydney 2009).
98. See Wood and Oglo. (n 53) 187.
99. As noted in D Crighton and G Towl, Psychology in Prisons (2nd edn Wiley-Blackwell, Chichester 2008) 110, in the United Kingdom, where individuals had been assessed as being "high risk" under the Violence Risk Assessment Guide (VRAG), 55% of these individuals went on to behave violently as opposed to only 18% in the low-risk group. Studies of unstructured clinical assessments suggest lower levels of accuracy. In one study 39% of those labeled medium or high risk went on to engage in violence whereas26%of those rated as low risk committed a violent offence.
100. Carlisle Report into the Parole System in England and Wales, Command 532 (1988).
101. See also S Shute, 'Parole and Risk Assessment' in N. Padfield (ed), Who to Release: Parole, Fairness and Criminal Justice (Willan Publishing, Cullompton 2007) 32-33.
102. Peta (n 5) [30].
103. CC Mercardo and J Oglo., 'Risk and the Preventive Detention of Sex Offenders in Australia and the United States' (2007) 30 International Journal of Law and Psychiatry 49, 55.
104. Campbell, French and Gendreau (n 58) 4.
105. Peta (n 5) [16].
106. Ibid [81].
107. Crighton and Towl (n 60) 119-25.
108. Peta (n 5) [29];
109. Beecher-Monas (n 2) 159.
110. Wood and Ogloff (n 53) 189.
111. Mercado and Ogloff (n 63) referring to FS Berlin,
112. NW Galbreath, B Geary and G McGlone, 'The Use of Actuarials at Civil Commitment Hearings to Predict the Likelihood of Future Sexual Violence' (2003) 15 Sexual Abuse: A Journal of Research and Treatment 377-82.
113. Freckelton and Selby (n 59).
114. Melton (n 55) 12;
115. RA Duff, 'Dangerousness and Citizenship' in A Ashworth and M Wasik (eds), Fundamentals of Sentencing Theory (Clarendon Press, Oxford 1998) 141, 156.
116. Wood and Ogloff (n 53) 190.
117. Shute (n 61) 27.
118. McSherry, Keyzer and Freiberg (n 4) [6.2.1] and [6.2.3].
119. D Mossman, 'Analysing the Performance of Risk Assessment Instruments: A Response to Vrieze and Grove' (2008) 32 Law & Hum Behav 279.
120. See for example D Dawson, 'Risk of Violence Assessment: Aboriginal Offenders and the Assumption of Homogeneity' in Best Practice Interventions in Corrections for Indigenous People Conference (Australian Institute of Criminology, Adelaide 1999)
121. and L Snowball and D Weatherburn, 'Theories of Indigenous Violence: A Preliminary Empirical Assessment' (2008) 41 ANZJ Crim 216-235.
122. The predictive validity of the ASRS in New Zealand, as discussed at n 170, was established by applying it to all offenders released from prison in New Zealand after serving a sentence for a sexual offence in the years 1987 and 1992-2005 (a total of 5,789 cases) and evaluated in terms of predicting sex offender recidivism.
123. See generally A Skelton, D Riley, D Wales and J Vess, 'Assessing Risk for Sexual Offenders in New Zealand: Development and Validation of a Computer-scored Risk Measure' (2006) 12 Journal of Sexual Aggression 277.
124. Crighton and Towl (n 60) 111.
125. Ibid 120.
126. The SONAR was developed from data derived from a meta-analytical literature review and information obtained from direct field research in Canada and the United States. SONAR assesses both stable dynamic factors that contribute to risk, but change slowly over time, and acute factors that may be present for a short period of time and which also contribute to individual sexual offending. For a list of the stable and acute factors to be assessed under SONAR,
127. see Peta (n 5) at [32]-[34].
128. Mercado and Ogloff (n 63) 55.
129. Riley (n 4) 33.
130. McMaster and Wells (n 4) 12.
131. Melton (n 55) 301.
132. Evidence Act 2006 (NZ) s 59.
133. R v D [2003] 1 NZLR 41 (CA). The case was concerned with s 33 of the old Evidence Amendment Act (No 2) 1980 (NZ) written in similar terms to s 59 of the current Act.
134. See Adams on Criminal Law - Evidence (Brookers, Wellington 2009) EA 59.01.
135. Different Australian states have different legislative provisions surrounding the admissibility of communications made to a health professional.
136. See generally A Ligertwood, Australian Evidence (4th edn Lexis Nexis, New South Wales 2004).
137. McSherry, Keyzer, and Freiberg (n 4) [6.2.1].
138. Melton (n 55) 14.
139. J Monahan, 'A Jurisprudence of Risk Assessment: Forecasting Harm Among Prisoners, Predators and Patients' (2006) 92 Va.L.Rev 391, 434.
140. Note the case of Victor Saldono 530 U.S 1212, 120 S.Ct 2214, 147 L.Ed. 2d 246 where an expert testified that Mr Saldono possessed many risk factors, including his Hispanic ethnicity. Eventually the death sentence was set aside (by consent in the end) at Supreme Court level.
141. ndeed, ProfessorMonahan (n 84) 428 goes so far as to suggest that past criminal behaviour is the only scientifically valid risk factor for violence that unambiguously implicates blameworthiness and therefore the only one that should feature in sentencing.
142. McSherry, Keyzer and Freiberg (n 4) [1.3].
143. Riley (n 4) 29.
144. McSherry, Keyzer and Freiberg (n 4) [1.3].
145. For instance, the Serious Sex Offenders Monitoring Act 2005 (Vic) sets out a comprehensive list of crimes that are deemed to be serious.90 These include non-contact acts of indecent exposure, possession of child pornography, loitering near schools where the victims are under 18 years. They do not include the same crimes committed on a victim who is over 18. The tools used may not, however, make the same type of differentiation.
146. See generally Wood and Ogloff (n 53) 184.
147. This applies often to sexual offending. For instance the SVR-20 defines violence as "actual, attempted, or threatened sexual contact with a person who is non-consenting or unable to give consent" and Static-99 defines sexual violence in a limited manner which would not include pornography crimes. On the other hand,
148. see Mercado and Ogloff (n 63) 56 referring to
149. SD Hart, 'Actuarial Risk Assessment: Commentary on Berlin et al' (2003) 15 Sexual Abuse 383.
150. Melton (n 55) 309-12.
151. Ibid 50.
152. McGarry v The Queen (2001) 207 CLR 121 at 131 (HCA).
153. McSherry, Keyzer and Freiberg (n 4) [6.2.1].
154. R v Peta [2007] 2 NZLR 627 at [50]-[51].
155. Beecher-Monas (n 2) 162.
156. Riley (n 4) 30.
157. See Crighton and Towl (n 60) 111
158. and K Gelb, Recidivism of Sex Offenders Research Paper (2007) Sentencing Advisory Council Victoria, 4.
159. See McDonnell (n 52) [33].
160. Quoted in Marshall (n 36) 129
161. see generally A von Hirsh, 'The Problem of False Positives' in von Hirsch and Ashworth (n 36).
162. See D Ruschena, 'Determining Dangerousness: Whatever Happened to the Rules of Evidence?' (2003) 10(1) Psychiatry, Psychology and Law 122, 123
163. and McSherry, Keyzer and Freiberg (n 4) [1.2].
164. McSherry, Keyzer and Freiberg (n 4) at [1.3].
165. Section 107F(2)(a) Parole Act 2002 (NZ) provides that an application for an ESO must be accompanied by a report by a health assessor, while s 88(1)(b) Sentencing Act 2002 (NZ) provides that a sentence of preventative detention must not be imposed unless the court has considered reports from at least two appropriate health assessors about the likelihood of the offender committing a further qualifying sexual or violent offence. The requirement for the courts to rely on medical evidence is also found in a variety of Australian statutes. To give a few examples, s 18B(2) Sentencing Act 1991 (Vic) provides that the court in making an order of indefinite sentence must consider medical, psychiatric and other relevant reports. Section 13(4) of the Dangerous Persons (Sexual Offenders Act) 2003 (Qld) provides that when the court is ordering a sentence of preventative detention that it must have regard to psychiatric reports and other medical assessments, while ss 6-7 of the Serious Sex Offenders Monitoring Act 2005 (Vic) provides that an application for an ESO must be accompanied by an assessment report by a psychologist, psychiatrist or specialist health provider.
166. See discussion above n 55. Indeed in other areas judges have been shown to be just as fallible in making judgments as the general population:
167. see C Guthrie, J Rachlinski and AJ Wistrich, 'Blinking on the Bench: Hon Judges Decide Cases' (2007) Cornell L. Review 101.
168. For a recent discussion see Accident Compensation Corporation v Ambros [2007] NZCA 304 at [65]-[76].
169. Peta (n 5) [29].
170. Mercado and Ogloff (n 63) 57.
171. Peta (n 5) and also Barr v Chief Executive of Department of Corrections CA60/06 20 November 2006 at [32].
172. Wood and Ogloff (n 53) 195.
173. Beecher-Monas (n 2) 4.
174. LM Levett and MB Kovera, 'The Effectiveness of Opposing Expert Witnesses for Educating Jurors about Unreliable Expert Evidence' (2008) 32 Law Hum Behav 363, 363.
175. Beecher-Monas (n 2) 4.
176. McDonnell (n 52) [44].
177. L Loeb, 'Forensic Testimony: What Judges Want' (2003) 24 Pace L. Rev 211, 211-212.
178. I Freckelton, P Reddy, and H Selby, Australian Judicial Perspectives on Expert Evidence: An Empirical Study (Australian Institute of Judicial Administration 1999).
179. The response rate was 50%, however, the survey was only directed to those judges with trial experience, therefore those who had spent their entire judicial career in the appellate courts were ineligible. Therefore the response rate was approximately 60%.
180. Freckelton, Reddy and Selby, ibid 3. The other types of expert evidence that were of concern were science and psychology.
181. Question 2.10 stated: "From the following list please circle the three factors which you consider to be the most persuasive when an expert is giving oral evidence. (a) their appearance 0.57% (b) clarity of explanation 27.48% (c) educational qualifications 1.84% (d) prior experience in the field 22.38% (e) prior experience as an expert witness 3.12% (f) familiarity with the facts 18.84% (g) impartiality 25.64% (h) publications 0.14%."
182. Peta (n 5) [27].
183. Ibid [27].
184. For instance in Curran v Parole Board HC CHCH CRI 2003-409-110 5 February 2004 at [16] the Court held that for an ESO to be imposed there must be a real and substantial risk of reoffending that is significantly greater than the risk in relation to inmates generally, given the substantially greater baseline risk of recidivism among serving prisoners.
185. Melton (n 55) 14-15.
186. As they point out, even where correlations are strong scientifically they may be inherently misleading because of the diffculty most people have in processing basis rates. See also R v Aymes [2005] 2 NZLR 376 at [114], [133]-[135] (CA).
187. Beecher-Monas (n 2) 142.
188. Ibid 141.
189. When doing so it must be remembered that there are differing degrees of statistical competence among the judiciary. It is probably fair to say that, as a general rule, judges, like lawyers, have a tendency towards being mathematically challenged.
190. Peta (n 5) [27].
191. Beecher-Monas (n 2) 160-61.
192. McDonnell (n 52) [45].
193. Peta (n 5) [52].
194. Ibid [52].
195. Ibid [52].
196. Skelton, Riley, Wales and Vess (n 73) at 284.
197. Baragwanath J in McDonnell (n 52) [122].
198. For example, in Peta (n 5) the relevant offending for the purposes of the statute was reoffending against children rather than sexual reoffending generally - see at [23]. Originally the reoffending rates given to the District Court in Peta related to sexual offending generally and not to sexual offending against children. The same appears to have been the case in Barr v Chief Executive of the Department of Corrections CA60/06 20 November 2006 (see at [8] of that decision). The figures given in Belcher (n 23) at [80] and [93] for sexual reoffending against children (43% at 10 years) were not in fact for sexual reoffending against children, but for sexual reoffending generally. That there was this disconnect with the statutory criteria does not seem to have been explained to the relevant courts involved.
199. 133. Mercado and Oglo. (n 63) 56.
200. See also Loeb (n 115) 211.
201. Melton (n 55) 48.
202. British Psychological Society, Psychologists as Expert Witnesses: Guidelines and Procedure for England and Wales (British Pyschological Society Expert Witnesses Working Party, Leicester 2007) [3.5].
203. Ibid [3.5].
204. Melton (n 55) 580.
205. A comment by SL Brodsky and NG Poythress (in the United States context) quoted in Melton (n 55) 579.
206. Melton (n 55) 11.
207. 140. Riley (n 4) 31.
208. Beecher-Monas (n 2) 162. Guideline 2.2 of the Guidelines for Expert Witnesses in Proceedings in the Federal Court of Australia Practice Direction 5 May 2008 states that all assumptions of fact made by the expert should be clearly and fully stated. "Code of Conduct for expert witnesses" Schedule 4 of the New Zealand High Court Rules at [3(d)].
209. Beecher-Monas (n 2) 162.
210. 143. British Psychological Society Report (n 135) [1.5]. "Expert witnesses have a duty to provide the Court with the necessary criteria so that the Court may be able to evaluate the basis of the expert's opinions and conclusions."
211. Department of Corrections v McDonnell (n 52) [35].
212. Vess (n 26) 175.
213. See also Adams on Criminal Law (n 81) PA107F.03.
214. Peta (n 5) [53].
215. For a recent discussion of the making of sentencing decisions see H Bennett and GA Broe, 'Judicial Neurobiology, Makrarian Synthesis and Emotion: How Can the Human Brain Make Sentencing Decisions?' (2007) 31 Crim LJ 75 and SL Sporer and J Goodman-Delahunty, 'Disparities in Sentencing Decisions' in M. Oswald et al. (eds), Social Psychology of Punishment of Crime (forthcoming 2009).
216. Peta (n 5) [29].
217. Riley (n 4) 34
218. and Beecher-Monas (n 2) 166.
219. McSherry, Keyzer and Freiberg (n 4) [1.3]. Guideline 2.9 of the Guidelines for Expert Witnesses in Proceedings in the Federal Court of Australia Practice Direction 5 May 2008 states that "if an expert's opinion is not fully researched because the expert considers that insufficient data are available, or for any other reason, this must be stated with an indication that the opinion is no more than a provisional one. Where an expert witness who has prepared a report believes that it may be incomplete or inaccurate without some qualification, that qualification must be stated in the report." "Code of Conduct for expert witnesses" Schedule 4 New Zealand High Court Rules at [4]-[5].
220. Melton (n 55) 11.
221. Ibid 11.
222. 151. Mercado and Oglo. (n 63) 52.
223. Riley (n 4) 33.
224. Wood and Oglo. (n 53) 184
225. Wood and Oglo. also make the point that sexual offenders do not have a higher level of reoffence rates than other criminal and violent offenders.
226. Mercado and Oglo., ibid (n 63) 53.
227. Wood and Oglo. (n 53) 184.
228. See also Gelb (n 99) 31.
229. 154. Peta (n 5) [25].
230. Mercado and Oglo. (n 63) 53.
231. Ibid 53.
232. Ibid 53.
233. However, studies have found that offenders who have demonstrated an offending pattern involving penetrative sexual assaults and the use of abduction or force would be those presenting the highest risk of such serious offending in the future.
234. See Peta (n 5) [45].
235. Ibid 54.
236. For a discussion on treatment approaches that have shown some success see Melton (n 55) 294.
237. and Gelb (n 99) 33-40.
238. In R v Vincent HC AK CRI 2006-044-285 30 August 2007 it was held that ten-monthold reports were not sufficiently recent.
239. Riley (n 4) 34.
240. McSherry, Keyzer and Freiberg (n 4) [8.3] and [6.2.1].
241. A Kapardis, Psychology and Law: A Critical Introduction (2nd edn Cambridge University Press, Cambridge 2003) 12.
242. British Psychological Society Report (n 135) [4.6].
243. Wood and Oglo. (n 53) at 193.
244. Freckelton, Reddy and Selby (n 116) 3. This survey was of Australian judges' views on expert evidence.
245. Schedule 4 of the New Zealand High Court Rules "Code of Conduct for expert witnesses" at [1]. Guideline 1.1 of the Guidelines for Expert Witnesses in Proceedings in the Federal Court of Australia Practice Direction 5May 2008. British Psychological Society Report (n 135) [1.5].
246. Vess (n 26) 164-65.
247. Peta (n 5) [58].
248. The Court had warned judges in previous cases against merely rubber-stamping health assessors' reports. It said that what is required is a careful assessment of all the historical and current factors, along with expert opinions of others, bearing in mind that an ESO can have substantial ongoing impact on an offender who has already completed his or her sentence. Every case requires individual judgement and assessment, and jurisdiction to make an ESO arises only if there is a real and ongoing risk of committing relevant sexual offences.
249. See Barr v The Chief Executive of the Department of Corrections CA60/06 20 November 2006 at [32].
250. and also Grieve (n 22) [15].
251. 169. Peta (n 5) at [57].
252. The ASRS is a shortened version of the Static-99 measure which was developed in Canada using the technique of metaanalysis. This allowed a large number of research studies to be aggregated statistically to gain a more objective assessment of the relationship between predictive variables and the event which is to be predicted. In New Zealand because three of the ten items in the Static-99 were not routinely captured in the New Zealand Criminal History Database a different instrument needed to be developed. These three items were "stranger victim", "unrelated victim", and "never lived with a lover for two years". Another difference between Static-99 and ASRS is that, under ASRS, prior sentencing dates are counted whereas in Canada previous charges are counted.
253. Discussed (n 76).
254. Peta (n 5) [57].
255. 173. Ibid [59].
256. Section 107I(5) of the Parole Act 2002 (NZ) provides that that the term of anESO must be the minimum period required for the purposes of the safety of the community in light of: the level of risk posed by the offender; the seriousness of the harm that might be caused to the victims; and the likely duration of the risk.
257. Peta (n 5) [61].
258. This was inconsistent with the Court of Appeal's approach in Belcher, see discussion (n 23).
259. 176. This was changed on 26 June 2008 pursuant to the enactment of the Crimes Amendment Act (No 2) 2008 (2008 No 37) (NZ). Under that Act, s 385(3)(c) was inserted into the Crimes Act 1961 (NZ). Section 385(3)(c) of the Crimes Act 1961 (NZ) provides that the Court of Appeal or Supreme Court may remit the case to the court that imposed the sentence with a direction that such court quash the sentence and replace it with another sentence warranted in law (whether more or less severe); or vary, within the limits warranted in law, the sentence or any part of it or any condition imposed in it.
260. Peta (n 5) [3].
261. Ibid [62].
262. Ibid [63].
263. (2005) 22 CRNZ 233 [53].
264. Peta (n 5) [64].
265. Ibid [65].
266. Mr Riley for the Crown, the director of the Psychological Service of the Department of Corrections, explainedthat current practice when preparing health assessment reports is for Department of Corrections psychologists to undertake a semi-structured clinical assessment by using the factors in the SONAR as anchors to provide a form of checklist of known risk factors. Reference to these factors further informs the assessment of risk, the initial baseline being established by the use of ASRS as a screening instrument. Mr Riley accepted that this meant that there was the potential for variation amongst Corrections Department sta. in their assessment of the risk factors contained in the SONAR instrument. The Department had, however, only been able to arrange formal training in this measure from August 2006.
267. Peta (n 5) [67].
268. 185. Ibid [67].
269. Ibid [70].
270. Ibid [71].
271. Ibid [72].
272. Ibid [74].
273. Ibid [16].
274. Ibid [30].
275. 192. Ibid [50].
276. See generally HA Miller, 'A Dynamic Assessment of Offender Risk, Needs and Strengths in a Sample of Prerelease General Offenders' (2006) 24 Behav Sci Law 767 at 770.
277. Peta (n 5) at [40].
278. Ibid [88].
279. Ibid [87].
280. Ibid [90].
281. Ibid [91].
282. Ibid [92].
283. Ibid [93].
284. Ibid [94].
285. Section 107F(2)(a) Parole Act 2002 (NZ).
286. Peta (n 5) [95].
287. Section 107F(2)(b) Parole Act 2002 (NZ).
288. Peta (n 5) [96].
289. 205. Section 107F(2)(c) Parole Act 2002 (NZ).
290. Peta (n 5) [97].
291. 207. Ibid [98].
292. Ibid [99].
293. Section 107F(2)(d) Parole Act 2002 (NZ).
294. Peta (n 5) [100].
295. Ibid [101].
296. RK Hanson and K Morton-Bourgon, 'Predictors of Sexual Recidivism: An Updated Meta-Analysis' (2004) (Public Safety and Emergency Preparedness Canada's Website) accessed 19 February 2007.
297. Peta (n 5) [102].
298. It was clear that Mr Peta could not be called a model citizen. As a juvenile, Mr Peta had accumulated an extensive list of criminal convictions, beginning with an aggravated robbery at the age of 14. As an adult, Mr Peta had been convicted of a variety of offences, including theft, assault, possession of a pistol, unlawful interference with a motor vehicle, escape from custody, and possession of cannabis.
299. Applying the test established by the Court of Appeal in Belcher (n 23) [11].
300. Peta (n 5) [103].
301. Ibid [105].
302. Ibid [104].
303. Mercado and Oglo. (n 63) at 56.
304. Ibid 57.
305. Melton (n 55) 577.