JD v The Attorney-General of the Northern Territory

CITATION:JD v The Attorney-General of the Northern Territory [2020] NTCA 11
PARTIES:JD

v
THE ATTORNEY-GENERAL OF THE NORTHERN TERRITORY

TITLE OF COURT:  COURT OF APPEAL OF THE NORTHERN TERRITORY
JURISDICTION:  CIVIL APPEAL from the SUPREME COURT exercising Territory jurisdiction
FILE NO:AP 10 of 2019 (21518615)
DELIVERED:  31 July 2020
HEARING DATES:  9-10 June 2020
JUDGMENT OF:  Blokland & Hiley JJ, Mildren AJ
CATCHWORDS:

APPEALS – Nature of appeal - Serious Sex Offenders Act 2013 (NT) – Appeal from decision of the Supreme Court confirming the making of a final continuing detention order – Where the appeal is expressed in s 104 of the Serious Sex Offenders Act 2013 (NT) to be by way of review of the original decision – Appeal requires demonstration of error or that the decision was unreasonable or plainly unjust.
SERIOUS SEX OFFENDERS – Continuing detention orders – Standard of proof – Court must be satisfied to “a high degree of probability” that person is a serious danger to the community – Standard of proof well above the civil standard.
SERIOUS SEX OFFENDERS – Unacceptable risk that the person will commit a serious sex offence unless he or she is in custody or subject to a supervision order – Determination of risk – Balancing exercise required to consider, amongst other things, the nature of the risk and the likelihood of the risk coming to fruition on the one hand and the serious consequences for the offender on the other – Nature of the risk includes the commission of a serious sexual offence with serious consequences for the victim.
SERIOUS SEX OFFENDERS – Evidence in relation to whether there is an unacceptable risk that the person will commit a serious sex offence unless he or she is in custody or subject to a supervision order – Use and weight of psychiatric risk assessment tools known as Static 99, RSVP and HCR-20 where the person is an indigenous person – Structured professional judgement used by psychiatrist partly based upon psychiatric risk assessment tools.
SERIOUS SEX OFFENDERS – Supervision order – Whether the Commissioner of Correctional Services has a statutory obligation to provide accommodation outside a Correctional Centre suitable for supervision of a person – Whether the failure to provide such accommodation precludes the making of a continuing detention order.
Crimes (High Risk Offenders) Act 2006 (NSW) s 9(2), 17(2), 17(3)
Dangerous Prisoners (Sexual Offenders) Act 2003 (Qld) ss 13(3), 13(5)(a)
Dangerous Sexual Offenders Act 2006 (WA) s 7(2)
Interpretation Act 1978 (NT) s 62B(2)
Sentencing (Crime of Murder) and Parole Reform Act 2003 (NT)
Sentencing Act 1995 (NT) s 71
Serious Offenders Act 2018 (VIC)
Serious Sex Offenders (Detention and Supervision) Act 2009 (VIC) s 9
Serious Sex Offenders Act 2013 (NT) ss 4, 7(1), 6, 9, 14, 23, 25, 26, 27, 30, 31,63,65,71, 79, 80, 82, 83,88, 95, 102, 104, 106, 107
Supreme Court Act 1979 (NT)
Allesch v Maunz (2000) 203 CLR 172, Attorney General of the Northern Territory v EE [2013] NTSC 68, Attorney-General (Qld) v Francis [2006] QCA 324, Attorney-General for the State of Queensland v Sybenga [2009] QCA 382, Attorney-General of the Northern Territory v JD (No 4) [2019] NTSC 82, Attorney-General of the Northern Territory v JD (No 2) [2016] NTSC 12, Attorney-General of the Northern Territory v JD [2015] NTSC 28, Attorney-General of the Northern Territory v JD [2017] NTSC 48, Briginshaw v Briginshaw (1938) 60 CLR 336, Builders Licensing Board v Sperway Constructions (Syd) Pty Ltd (1976) 135 CLR 616, CDJ v VAJ (No 2) (1998) 197 CLR 172, Coal and Allied Operations Pty Limited v Australian Industrial Relations Commission (2000) 203 CLR 194, Director of Public Prosecutions (WA) v GTR (2008) 38 WAR 307, Director of Public Prosecutions (WA) v Mangolamara [2007] WASC 71, Director of Public Prosecutions (WA) v Pindan [No 3] [2014] WASC 95, Dwyer v Calco (2008) 234 CLR 124, Fardon v Attorney-General (Qld) (2004) 223 CLR 575, House v The King (1936) 55 CLR 499, Leach v The Queen (2005) 16 NTLR 117, Nigro v Secretary to the Department of Justice (2013) 41 VR 259, Railways (NSW) v Agalianos (1955) 92 CLR 390, Re H (Minors) [1996] AC 563, Slaveski v Smith (2012) 34 VR 206, The State of Queensland v DBJ [2017] QSC 302, Tourism Holdings Australia Pty Ltd v Commissioner of Taxes (NT) (2005) 15 NTLR 80, Woods v Director of Public Prosecutions (WA) (2008) 38 WAR 217, referred to
REPRESENTATION:
Counsel:
Appellant:M Thomas and M Jehne
Respondent:  T Anderson
Solicitors:
Appellant:Northern Territory Legal Aid Commission
Respondent:  Solicitor for the Northern Territory
Judgment category classification:    B
Number of pages:  66

IN THE COURT OF APPEAL
OF THE NORTHERN TERRITORY
OF AUSTRALIA
AT DARWIN
JD v The Attorney-General of the Northern Territory [2020] NTCA 11

No. AP 10 of 2019 (21518615)
BETWEEN:

JD

Appellant

AND:

THE ATTORNEY-GENERAL OF THE NORTHERN TERRITORY

Respondent

CORAM: BLOKLAND, HILEY JJ AND MILDREN AJ
REASONS FOR JUDGMENT
(Delivered 31 July 2020)
THE COURT:
Introduction

The appellant has appealed against the order made by the Court (Barr J) on 1 February 2019 (the Order) confirming the Final Continuing Detention Order made by the Court on 3 December 2015 (the FCDO). The Order was made following a review brought by the Attorney-General on 6 June 2018 pursuant to s 65(1)(b) of the Serious Sex Offenders Act 2013 (NT) (the Act).[1]
The FCDO was made under s 31 of the Act and was the subject of detailed reasons for decision in Attorney-General of the Northern Territory v JD (No 2).[2] Prior to that, the Court had made an interim continuing detention order on 27 November 2015 pursuant to s 30 of the Act.[3] The FCDO was reviewed on various dates from 8 December 2016 and was continued by an order made on 19 June 2017 pursuant to s 71(1)(c) of the Act.[4] The review the subject of this appeal was the second review of the FCDO. A third review is scheduled to take place on 15-16 July 2020.
The appellant contends that the primary judge should have revoked the Continuing Detention Order (CDO) under s 71(1)(a) of the Act, or alternatively revoked the CDO and made a final supervision order (FSO) under s 71(1)(b).
Relevant legislation
The Act was enacted in 2013. It is in a similar, but not identical form, to legislation in other States.[5] The overall regime set up by the Act is described in more detail in Attorney General of the Northern Territory v EE[6] at [6] – [7] and [18]; JD (No 1) at [3] – [7] and [8] – [9]; and JD (No 2) at [2] – [9].
Subject to exceptions not relevant for present purposes,[7] s 23 of the Act enables the Attorney-General to apply to the Supreme Court for a final continuing detention order or final supervision order in relation to a “qualifying offender”. The Supreme Court is then required to conduct a preliminary hearing to determine “whether the matters alleged in the application would, if proved, satisfy the Court that the qualifying offender is a serious danger to the community.”[8] If the Court decides that it would be satisfied, it must set a date for the hearing of the application[9] and make a medical assessment order[10] naming two medical experts.[11] Once the Court sets a hearing date under s 25(2)(a) of the Act, s 26 requires victim submissions about the offender to be sought under s 83 and s 27 requires a supervision report about the offender to be prepared by the Commissioner of Correctional Services under s 88.
On the hearing of an application made under s 23, the Court may make a final continuing detention order or a final supervision order “if satisfied that the offender is a serious danger to the community.”[12] By that time, the Court should have available to it victim submissions sought under s 83, a supervision report from the Commissioner of Correctional Services under s 88 and two medical reports containing the information required by s 79(5) of the Act.
Where the Court makes a final continuing detention order, it may state in the order a review period for the purposes of s 65.[13] Section 65 requires the Attorney-General to apply to the Court for a review of a final continuing detention order before the expiry of the review period.
Once an application for review is made, a date is set for the hearing of the application,[14] and victim submissions under s 83 and a supervision report from the Commissioner of Correctional Services under s 88 are to be sought.[15] The Court may also make a medical assessment order: that is, an order seeking a further medical report under s 79.[16]
A medical assessment order is an order that the person concerned must submit to being examined by 1 or 2 medical experts named in the order.[17] The medical expert is then required to provide a medical report containing an “opinion of the likelihood of the person committing another serious sex offence if he or she is not detained in custody or subject to a supervision order, the reasons for that opinion, and the extent to which the person cooperated with the expert.”[18] The resulting medical report is admissible as evidence in relation to a matter mentioned in ss 6(2), 9 or 14.[19] A “medical expert” is defined to mean a “health practitioner registered to practice in the medical profession in the recognised specialty of psychiatry, or in a health profession prescribed by regulation.”[20]
Section 71 states what orders are to be made on a review sought under s 65.
71   Orders on review
(1)   On hearing an application made under section 65 or 66, the Supreme Court must do one of the following:
(a)revoke the continuing detention order;
(b)revoke the continuing detention order and make a final supervision order in relation to the detainee;
(c)confirm the continuing detention order.
(2)   The Court must not make an order under subsection(1)(b) or (c) unless satisfied that the detainee is still a serious danger to the community.
(3)   …
The Attorney-General has the onus of satisfying the Court that it is appropriate to confirm the continuing detention order or to make a final supervision order.[21]
As we have stressed by use of underlining above, an important issue concerns the satisfaction of the Court that the person in respect of whom a detention order or supervision order is sought or to be continued “is” or “is still” “a serious danger to the community”. The Act expressly deals with this concept, sets out particular matters for the Court to consider, and refers to onus of proof and standard of proof.
Sections 6 and 7 provide as follows:
6     Serious danger to the community
(1)   A person is a serious danger to the community if there is an unacceptable risk that he or she will commit a serious sex offence unless he or she is in custody or subject to a supervision order.
(2)   In deciding whether a person is a serious danger to the community, a court must have regard to the following:
(a)the likelihood of the person committing another serious sex offence;
(b)the impact of serious sex offences committed, or likely to be committed by the person on:
(i)victims of those offences and the victims’ families; and
(ii)members of the community generally;
(c)the need to protect people from those impacts.
7     Standard and onus of proof
(1)   A court must not decide that a person is a serious danger to the community unless it is satisfied, to a high degree of probability, that there is acceptable and cogent evidence of sufficient weight to justify the decision.
(2)   The Attorney-General has the onus of satisfying the court that the person is a serious danger to the community.
Section 9 of the Act sets out particular matters to which a Court must have regard when making, confirming or revoking a continuing detention order. Section 14 imposes similar requirements when a Court is deciding whether to make, amend or revoke a supervision order. Section 9 provides as follows:
9     Considerations when making, confirming or revoking continuing detention order
(1)   In deciding whether to make, confirm or revoke a continuing detention order in relation to a person, a court must have regard to the following:
(a)as the paramount consideration – the need to protect:
(i)victims of serious sex offences committed, or likely to be committed, by the person; and
(ii)the victims' families; and
(iii)members of the community generally;
(b)as a secondary consideration – the desirability of providing rehabilitation, care and treatment for the person.
(2)   In considering the need for protection mentioned in subsection (1)(a), the court must have regard to the following:
(a)the likelihood of the person committing another serious sex offence;
(b)whether adequate protection could reasonably be provided by making a supervision order in relation to the person.
Part 9 of the Act is headed “Procedure”. Division 2 of Part 9 is headed “General matters” and includes s 95:
95 Standard of proof
(1) The standard of proof as to whether a person is a serious danger to the community is as mentioned in section 7(1).
(2) The standard of proof for all other matters under this Act is the balance of probabilities.
Nature of the Appeal
Part 10 of the Act provides for appeals against decisions made under the Act to the Court of Appeal. The right to appeal is conferred by s 102(1). Section 102(2) provides that “the appeal may be on matter of fact, law or fact and law.”
Section 104 deals with the nature of the appeal. It provides:
(1) The appeal is to be by way of review of the original decision, not a rehearing.
(2) However, the Court of Appeal may receive further evidence as it considers appropriate.
Section 106 provides:
(1)   After hearing an appeal, the Court of Appeal must do one of the following:
(a)confirm the original decision;
(b)vary the original decision;
(c)set aside the original decision;
(d)set aside the original decision and replace it with the Court of Appeal’s own decision;
(e)set aside the original decision and remit the matter back to the Supreme Court for reconsideration.
(2)   If it remits a matter under subsection (1)(e), the Court of Appeal may give directions to the Supreme Court as it considers appropriate.
(3)   The Court of Appeal may make any ancillary order it considers appropriate.
Section 107 empowers the Court to make interim continuing detention or supervision orders where the matter has been remitted under s 106(1)(e).
These provisions are intended to apply to appeals under the Act in substitution for the general provisions conferring a right of appeal from the decisions of the Supreme Court exercising its civil jurisdiction contained in Part III of the Supreme Court Act 1979 (NT). However, they are only intended to apply to the extent that the provisions of the Act and Part III of the Supreme Court Act 1979 differ.
The nature of an appeal depends upon the terms of the statute or statutes conferring the right. Generally speaking, leaving aside interlocutory appeals and appeals limited to questions of law only, there are four kinds of appeal.[22] The first type of appeal is an appeal strictu sensu, where the issue is whether the judgment below was right on the material before the trial court.[23] The Court in such a case can only give the decision which should have been given at first instance.[24] Its function is to decide whether the decision in question was right or wrong on the evidence and the law as it stood when that decision was given. The Court cannot receive further evidence and its powers are limited to setting aside the decision under appeal, and, if it be appropriate, to substitute the decision that should have been made at first instance,[25] although it may also have a power of remittal. Because the Court is empowered by s 104(2) of the Act to hear further evidence, this appeal is unlikely to be an appeal of that nature.
A second type of appeal is an appeal de novo. In that case, the matter is heard afresh and a decision is given on the evidence presented at that hearing. Error does not have to be established. The parties may begin again and call such evidence as they wish, or they may agree to accept the evidence given at first instance and also call further evidence.[26] The reference in s 104(1) that the appeal is by way of review and is not by way of rehearing might suggest that this is the case here.
A third type of appeal is an appeal by way of rehearing. In such a case, the appellant must show that the decision below is vitiated by error, whether the error is one of law, or fact, or mixed fact and law, or discretionary error.[27] In such a case, the Court determines the appeal by reference to the law as at the date of the hearing of the appeal.[28] Because s 104(1) states that the appeal is not by way of rehearing, it seems that the appeal in this case is not this type of appeal.
A fourth type of appeal is an appeal by way of rehearing with a power to receive further evidence. In cases of this kind, if further evidence has been admitted, the Court will decide the case on the basis of the evidence then before the Court. The principal purpose of the power to receive further evidence is to facilitate the avoidance of errors which cannot otherwise be remedied by the application of conventional appellate procedure.[29] If no further evidence has been admitted, and the law has not changed, the Court will decide the appeal on whether or not error has been made at first instance.[30] It may be that an appeal to this Court is an appeal of this kind, notwithstanding s 104(1) of the Act.
Because appeals are creatures of statute, the classification of appeals is not rigidly defined, and the nature of the appeal depends upon the true construction of the Act.[31]
One possibility is that the reference in s 104(1) of the Act to the “review” not being “a rehearing” is to make it clear that the “review” was not a rehearing de novo. If this is so, it would not necessarily mean that an appeal by way of rehearing with a power to receive further evidence if required was out of the question.
The powers of the Court as set out in s 106 of the Act provide significant support for this view. Section 106 refers to the Court confirming, varying, or setting aside the original decision. This indicates that the original decision is the focus of the appeal. Further, although the Court may substitute its own decision for the original decision, it also has a power to remit to the Supreme Court, which is a further indication that the power is intended to be exercised in the ordinary case upon the establishment of error by the appellant.
Another factor supporting this view, is that the appellant may be the Attorney-General appealing an order dismissing the original application or revoking a continuing detention order or supervision order. It would seem unlikely that the legislature would have intended to give the Attorney-General a second right of hearing absent the establishment of error by the trial judge. The fact that the appeal “may be on a matter of fact, law or fact and law” strongly suggests that error must be established.
The conclusion reached is that error must be established, whether the error be an error of fact, or law, or both.
As was pointed out by the Victorian Court of Appeal in Nigro,[32] the determination of the question of whether an offender poses an unacceptable risk of committing a serious sex offence unless he or she is in custody or subject to a continuing supervision order is “so much a question of value judgment based on a matrix of fact and degree” that this Court ought not to interfere unless it is plainly wrong. In that case, the Court of Appeal held that the appeal was against a discretionary judgment with the consequence that, except where additional evidence is received, the principles in House v The King[33] applied with respect to the County Court’s ultimate decision upon an application for a supervision order; but if the ground of appeal is directed to the County Court’s determination of the intermediate question of whether the offender poses an “unacceptable risk”, the test to be applied is simply whether that conclusion was plainly wrong or wholly erroneous.[34]

There are some differences in the legislative provisions relating to the powers of the County Court of Victoria and the Supreme Court of the Northern Territory in deciding whether or not to make a continuing supervision order. Under the Repealed VIC Act, the Court could either make or refuse to make an order, even if the Court found that the offender posed an unacceptable risk of committing a ‘relevant offence’ if a supervision order was not made: see s 9 (7) of the Repealed VIC Act. Under the NT Act, there is no such discretion expressed in so many words. What s 31(1) provides is:
On hearing an application made under section 23, the Supreme Court may make a final continuing detention order or final supervision order in relation to the qualifying offender if satisfied that the qualifying offender is a serious danger to the community.
In our opinion, despite the word “may”, the Court would be bound to make one kind of order or the other once the conclusion was reached that the offender was or remained a serious danger to the community. It is difficult to see how it could make an order dismissing the application in those circumstances. If the nature of the error is such that the decision is unreasonable or plainly unjust, the Court can interfere even if no specific error can be identified.[35]
Grounds of appeal
The grounds of appeal pursued are set out in the “Additional Grounds of Appeal” and are as follows:
1. The learned Trial Judge erred in law in confirming the continuing detention order pursuant to s 71(1)(c) of the Serious Sex Offenders Act
Particulars
The Learned Judge erred in:
(a)Having made a determination that JD was still a serious danger to the community, his Honour failed, in making that decision, to have regard to the criteria necessary to arrive at such a determination, namely whether the evidence satisfied those matters referred to at section 6(2) of the Act (in particular the likelihood of the person committing another serious sexual offence, and the impact of [a] serious sex offence committed, or likely to be committed, by the person upon victims of those offences, their families and members of the community generally).
(b)Having made a determination that JD was still a serious danger to the community, his Honour failed to address whether the admissible evidence satisfied the standard of proof specified in section 7(1) of the Act, namely, that a court must not decide that a person is a serious danger to the community unless it is satisfied to a high degree of probability that there is acceptable and cogent evidence of sufficient weight to justify the decision.
(b1)The trial Judge erred in law by misapplying the “unacceptable risk” test and improperly considering the reasonable practicability of supervision when determining whether JD was a “serious danger to the community”.
(c)Having made a determination that JD was still a serious danger to the community, his Honour failed to refer to section 7(2) of the Act, which states that the Attorney-General bore the onus of proof.
(d)Failing to make a finding that the evidence of Dr Sullivan satisfied section 6 and 7 of the Act and in particular what part or parts of Dr Sullivan’s evidence provided the foundation for this, despite his Honour noting (at paragraph 12 of his judgment) that prediction of re-offending risk is imprecise, with a significant rate of false positive and false negative results.
(e)Failing to give little or no weight to the risk assessment made by Dr Sullivan due to the utilisation by him of three psychiatric assessment instruments, namely the Static 99 actuarial instrument, the RSVP (Risk for Sexual Violence Protocol) instrument and the HCR-20 V3, (a violence risk assessment tool), all of which had no proved validation in predicting the risk of the commission of serious sexual offences by particular persons.
(f)Failing to give little or no weight to the risk assessment made by Dr Sullivan due to the utilisation by him of the three psychiatric risk assessment instruments referred to above, which were not devised for Australian indigenous persons and had no proved validation to Australian indigenous persons, in particular indigenous persons from remote communities (such as JD).
(g)Failed to take into account, in combination with each other, the following factors as significantly reducing the risk of the commission of a serious sexual offence by JD, namely:
(i)JD was/is suffering from significant ill-health (including having suffered a heart attack) and continue to suffer from multi-faceted, chronic, health problems,
(ii)JD was of relatively advanced age,
(iii)JD was not diagnosed as being mentally ill,
(iv)JD was not diagnosed as having paedophilic tendencies,
(v)There was no similarity between the two incidents of sexual offending on JD’s antecedents,
(vi)There was a gap of 17 years between the two incidents of sexual offending,
(vii)The two incidents of sexual offending represented less than 2% of total offences on JD’s criminal record,
(viii)A lapse of time of 15 years occurred since the last sexual offence took place, (JD having spent that entire time in prison),
(ix)The absence of any planning or concealment in regard to the final sexual offence.
(h)Failing to refer to Dr Sullivan’s evidence that the effects of declining ill-health were unknown, in determining whether the evidence met the evidentiary requirement specified in section 7 of the Act.
(i)Failing to take into account as matters that reduce the risk of sexual re-offending, the combination of the ageing and declining physical health.
(j)Failing to take into account that the risk assessment made by Dr Sullivan was affected by JD’s long term incarceration in an institutional setting, which meant that there was a limited opportunity to demonstrate lowered dynamic risk and consequently a greater emphasis placed upon static risk factors.
(k)Failing to find that Dr Sullivan’s capacity to engage in a meaningful clinical exercise with JD was subject to significant limitations due to the fact that it was conducted on one occasion only on audio-visual link and for 45 minutes.
(l)Declining to find that a consequence of Professor Boer’s concerns in relation to using unvalidated instruments had a direct impact upon the standard of evidence to be evaluated under section 7 of the Act.
(2) The Learned Trial Judge’s decision to confirm the continuing detention order occurred in circumstances where the absence of accommodation for JD outside of Darwin Correctional Centre precluded the possibility of the making of a supervision order, thereby resulting in the Serious Offenders Act operating punitively in regard to JD, which was contrary to the intent of the Act, with the consequence that there was no lawful basis for the Learned Trial Judge to make an order of any kind under the Act.
Relevant history of this matter
The history of this matter is set out in JD (No 1) at [8] and [11] and JD (No 2) at [1] and [10].
In short, the Attorney-General initiated these proceedings under the Act on 28 April 2015 by applying for a final continuing detention order. On 12 May 2015 the Court (Mildren AJ) conducted a preliminary hearing under s 25 and concluded that the matters alleged in the application would, if proved, satisfy the Court that, unless the appellant was in custody or subject to a supervision order, there was a very high degree of probability of further sexual offending of a serious nature[36] and thus the appellant was a serious danger to the community.[37] The Court set a date for the hearing of the application and made a medical assessment order pursuant to s 25(2), and made an interim continuing detention order pending determination of the application pursuant to s 30. The application was heard by the Court (Barr J) between 23 July and 3 December 2015, when the Court made the FCDO. Barr J delivered his reasons for decision on 29 February 2016 (JD (No 2)).
The appellant’s history of offending was set out by Barr J in JD (No 2).[38] This included:
(a)an aggravated sexual assault of a six year old boy and an aggravated assault of another six-year-old boy on 16 January 1988, for which he was sentenced to 5 years imprisonment with a non-parole period of two and a half years;[39]
(b)having sexual intercourse with a female without her consent and aggravated assault on the same female on 29 March 2005, for which he was sentenced to 10 years imprisonment with a non-parole period of 7 years;[40] and
(c)numerous offences between 16 July 1990 (when he was released from prison) and 2005.These included offences of violence (including violence against females) for which he spent a total of about 8 years in custody[41], and numerous incidences of bad behaviour in prison from 2005 which included threats to kill prison officers and numerous threats of violence including threats to rape some of the female prison officers.[42]
When the Court made the interim supervision order on 12 May 2015 the Court had a range of materials that included two reports from Dr Raeside, a psychiatrist, and a neuropsychological report from Ms Vidula Garde, a clinical and forensic psychologist. Amongst other things, the evidence was to the effect that the appellant had borderline intellectual disability, extremely low range ability to speak and comprehend English, gross deficit in audiology and visual attention, possible memory impairment, difficulties with problem-solving, behavioural disinhibition consistent with frontal lobe brain damage, inability to correctly identify facial emotion and interpersonal distress, history of binge drinking, and possible long history of illicit drug use.[43] Dr Raeside expressed the opinion that the appellant did not have a mental illness which could be treated to reduce risk factors. In addition, the appellant had a mixed personality disorder with antisocial, paranoid and narcissistic traits. No meaningful rehabilitation had been able to occur in prison, and there was a very high risk of further offending, both violent and non-violent as well as sexual offending.[44]
In the course of hearing the application and conducting the two subsequent reviews, Barr J had available to him reports from medical experts and reports from the Commissioner of Correctional Services, some of which attached reports from relevant health professionals.[45] The medical experts in relation to the application included Dr Michael Beech and Dr Lester Walton, both specialist psychiatrists.[46] In relation to the first review, the medical experts included Dr Danny Sullivan, a consultant forensic psychiatrist and Dr Olav Nielssen, a psychiatrist.[47] In relation to the second review, the medical experts included Dr Sullivan and Professor Douglas Boer, a registered clinical psychologist who holds a PhD in psychology.[48] On each occasion the Court was also assisted by counsel, Mr Thomas on behalf of JD, and Mr Anderson on behalf of the Attorney-General.
Ground 1(b) - Standard of Proof
In JD (No 2), Barr J held that s 7(1) of the Act did not impose a standard of proof to a high degree of probability that a person is a serious danger to the public, and that the civil standard of proof applies to the Attorney-General to prove its case that the relevant person is a serious danger to the community.[49] In arriving at that decision, his Honour referred to the observations of Mildren J in relation to s 71 of the Sentencing Act 1995 (NT) in Leach v TheQueen.[50]
Despite there being no appeal from that decision, the appellant now complains that his Honour was incorrect, and that when his Honour reviewed the matter in the decision appealed from (JD (No 4)), the error was not corrected, so that his Honour’s finding that the appellant was still a serious danger to the community is in error. Counsel for the respondent conceded that, if his Honour’s decision in JD (No 2) as to the burden of proof was incorrect, his Honour’s finding in JD (No 4) was also in error.
The starting point is a consideration of the relevant statutory provisions, particularly those which have been set out above in paragraphs [13] and [15] above. On one reading of s 7(1) of the Act, it is open to conclude that the standard of proof required is the civil standard, provided that the court is satisfied, to a high degree of probability, that the evidence relied upon to reach the conclusion that the person is still a serious danger to the community is of sufficient weight and cogency to reach that decision. The alternative construction is that not only must the evidence be of such weight and cogency to reach that decision, the Court must be satisfied to a high degree of probability that the person is a serious danger to the community. That is, the standard of proof as to whether the person is a serious danger to the community is not the civil standard, but some higher standard, approaching, but less than, proof beyond reasonable doubt. In our opinion, the answer to that question is that the standard of proof required is not the ordinary civil standard. Rather, it is that the Court must be satisfied to a high degree of probability that the person is a serious danger to the community.
This conclusion is consistent with conclusions expressed by other courts in relation to similar provisions in analogous legislation. It is also consistent with s 95 of the Act. Section 95(1) specifically provides that “the standard of proof as to whether a person is a serious danger to the community is as mentioned in s 7 (1).” Section 95(2) provides that “the standard of proof for all other matters under this Act is the balance of probabilities.” Counsel for the respondent contended that s 95(1) does not assist because it does no more than refer back to s 7(1). We disagree. Read as a whole, s 95 attempts to make a clear distinction between the standard of proof necessary when determining whether a person is a serious danger to the community, and the standard of proof sufficient for all other matters under the Act. It is difficult to see how, in the light of s 95, the standard of proof as to whether a person is a serious danger to the community is on the balance of probabilities. Otherwise it would leave s 95(1) with no work to do and it would mean that the words “other matters” in s 95(2) would not have been necessary.
The alternative constructions noted in [41] above are very similar to those considered in some detail by the Court of Appeal of Victoria in Nigro. That case concerned s 9 of the Repealed VIC Act which provided:
(1) The court may make a supervision order in respect of an eligible offender only if the court is satisfied that the offender poses an unacceptable risk of committing a relevant offence if a supervision order is not made and the offender is in the community.
(2) On hearing the application, the court may decide that it is satisfied as required by subsection (1) only if it is satisfied-
(a)by acceptable, cogent evidence; and
(b)to a high degree of probability-
that the evidence is of sufficient weight to justify the decision.
Section 9(2) of the Repealed VIC Act, although structured differently, is somewhat similar to s 7(1) of the NT Act. It is also similar to corresponding provisions elsewhere such as s 7(2) of the WA Act, ss 9(2), 17(2) and (3) of the NSW Act, s 13(3) of the QLD Act and to s 71 of the Sentencing Act 1995 (NT).
Putting aside the differences in structure it is noted that the ultimate issue under s 9(2) of the Repealed VIC Act is the court’s satisfaction that the offender “poses an unacceptable risk of committing a relevant offence”, whereas under s 7(1) of the NT Act, the ultimate issue is the court’s satisfaction that the person “is a serious danger to the community.” As with the Repealed VIC Act, the relevant issue under the Western Australian and New South Wales legislation is the court’s satisfaction of unacceptable risk. However s 13 of the QLD Act, like s 7(1) of the NT Act, refers to satisfaction that the prisoner is a serious danger to the community.
At [139] of Nigro their Honours (Redlich, Osborn and Priest JJA) said:
The terms of ss 9(2), 17(2) and (3) of the Crimes (Serious Sex Offenders) Act 2006 (NSW), like the Western Australian provisions, state that the court may not make the relevant order unless satisfied “to a high degree of probability that the offender poses an unacceptable risk.” The standard of satisfaction relates to the ultimate issue, rather than the quality of the evidence supporting that issue. Like its Western Australian counterparts, those provisions have consistently been construed as requiring a standard of proof higher than the civil standard but lower than the criminal standard of proof of unacceptable risk. …
Their Honours proceeded to consider the alternative possible constructions of s 9(2) of the Victorian Act concerning the standard of proof. At paras [141] – [142]:
[141] The question of construction that arises is whether the standard of proof in s 9(2) relates to the ultimate issue of unacceptable risk or only to the level of persuasion that the evidence must attain which will support the ultimate conclusion. Section 9(2) is drafted in a most unsatisfactory form and is not free from difficulty … The reference twice to “evidence” is problematic as is the application of a standard of high degree of probability to the sufficiency of the “weight” of the evidence. The second limb of the requirement is capable of supporting more than one construction. One construction is that the court must be satisfied to a high degree of probability that the weight of the evidence establishes an unacceptable risk. Another is that the court is satisfied to a high degree of probability of the evidence establishing the foundational facts forming the basis of the decision. The assessment of unacceptable risk, being evaluative and predictive, necessarily involves inferential reasoning from circumstantial evidence. The requirement that the evidence be of sufficient weight to justify the decision might be thought to direct attention to the combined probative weight of the evidence as distinct from the degree of satisfaction of individual circumstances. Such weight can only be assessed by reference to the degree that it is probative of the ultimate fact in issue, namely whether an offender poses an unacceptable risk of committing a relevant offence if a supervision order is not made and the offender is in the community.
[142] On this view, because s 9(2) requires the court to satisfy itself not simply to a high degree of probability as to the reliability of the evidence which form the basis of the relevant inference, but as to whether it is satisfied to a high degree of probability that the evidence is of sufficient weight to justify the decision, the subsection ensures that the court will be satisfied to a high degree of probability of the ultimate fact in issue.
The Court then discussed a range of matters, including certain observations of members of the High Court in Fardon v Attorney-General (Qld)[51] which they said supported the conclusion that the standard of proof to a high degree of probability applied to the finding of unacceptable risk.[52] Their Honours noted that the concept that facts may sometimes be required to be proved to a higher standard than the ultimate question is not a novel one, referring to Re H (Minors)[53] as one example. They also noted that the conceptual and logical difficulties arising from the way the terms in s 9(2) were employed were such that the words of s 9(2) were capable of bearing more than one construction.
Their Honours said, at [154]:
If the words of the statute bore only … one clear meaning we would be required to give it that meaning. But the matter of construction is not without difficulty, partly because, as we have set out above, there are conceptual and logical difficulties arising from the way in which terms have been employed. The words of Dixon CJ in Commissioner for Railways (NSW) v Agalianos[54] are here apposite that “the context, the general purpose and policy of a provision in its consistency and fairness are surer guides to its meaning than the logic with which it is constructed”. As we have said, the words of s 9(2) are capable of bearing more than one construction. One such meaning would attract the principle of legality - the presumption that the legislature does not intend to interfere with common law rights and freedoms except by clear and unequivocal language. As there is a constructional choice open, the principle requires that the provision be construed so as to avoid or minimise its encroachment upon rights and freedoms at common law. We must give it the meaning which best accords with ensuring that the enjoyment of the human rights in question are not defeated or diminished, so long as that meaning is consistent with both the grammatical meaning and apparent purpose of the enactment.[55]

The Court concluded, in para [156]:
In assessing whether there is an unacceptable risk, the court must therefore identify those matters which establish the risk and those that bear upon whether the risk is unacceptable. The judge must consider whether or not those matters have been proved by acceptable cogent evidence. The judge must be satisfied by the evidence to a high degree of probability that there is an unacceptable risk. That involves a standard well above the civil standard and approaching the criminal standard.
The analysis by the Court of Appeal of Victoria in Nigro is consistent with the interpretation of the provisions of s 7(1) of the Act which we have arrived at.
This view of the standard of proof required is supported by the Minister’s Second Reading Speech, where he said:
The bill contains a range of safeguards to prevent the powers being abused. These include the need for two independent medical assessments, a comprehensive review process, and an extremely high burden of proof on the applicant…
For the court to make either a continuing detention order or supervision order it must be satisfied to a high degree of probability that the offender is a serious danger to the community.[56]
Counsel for the respondent submitted that although the Second Reading Speech assisted the appellant’s case, there was no ambiguity in s 7(1) such as would engage the provisions of s 62B of the Interpretation Act 1978 (NT). We do not accept this submission. As we have said, s 7(1) seems capable of two different constructions. Indeed the construction contended for by the respondent is similar to that expressed in some first instance decisions in Queensland[57] and by Mildren J in Leach.
Section 62B(2) of the Interpretation Act 1978 (NT) permits the Court to consider various extrinsic materials when interpreting a provision of an Act. This includes a “statement given by a Minister … on introducing the Bill”[58], and “any relevant material in any official record of debates in the Legislative Assembly”[59]. The Court is entitled to have regard to this material “to confirm that the meaning is the ordinary meaning conveyed by the text, taking into account its context in the Act and the purpose and object underlying the Act”[60] or “to determine the meaning of the provision when … the provision is ambiguous or obscure.”[61] Whilst we accept that primacy must be given to the text, and that the Minister’s speech cannot be used to displace the text of the statute, in our opinion the Minister’s speech is consistent with the meaning which we have arrived at.
Counsel for the respondent also submitted that, although structured differently, s 7(1) is not materially different to s 71 of the Sentencing Act1995 (NT), discussed briefly by Mildren J in Leach.[62] Section 71 of the Sentencing Act 1995 was not directly relevant to the question in issue in that case, which concerned the construction to be given to s 19(5) of the Sentencing (Crime of Murder) and Parole Reform Act 2003 (NT). In that case, and in the case of s 71 of the Sentencing Act 1995, there was no provision such as is found in s 95 of the Serious Sex Offenders Act2013 which threw any further light on the meaning to be given to the relevant provision so far as concerned the standard of proof. In fact, the particular provisions being considered in Leach had nothing to say about the standard of proof or the standard of the evidence required to arrive at the relevant decision.
The Court of Appeal of Victoria in Nigro went on to consider whether what is called the Briginshaw principle[63] applies to the quality of the evidence. The Court said, at [161]:
Whatever the standard of proof is under the Act, s 9(2)(a) is undoubtedly intended to reflect the Briginshaw principle. It is formulated in recognition of the gravity of the consequences flowing from the making of a supervision order. To borrow from Neat Holdings, it requires that “clear or cogent or strict proof” is necessary to justify the making of a supervision order. Adapting the observations of Dixon J in Briginshaw, the finding ‘should not be produced by inexact proofs, indefinite testimony or indirect inferences.’ The proofs must survive careful scrutiny and appear precise and not loose and inexact.
So far as the quality of the evidence required by s 7 of the Act is concerned, we agree that the wording of s 7(1) reflects that concept, save and except for the fact that, unlike the standard of proof required by the Briginshaw test, the standard of proof is higher than the ordinary civil standard for the reasons we have explained above.
It follows from our conclusion that s 7(1) of the Act requires the court to be satisfied to a high degree of probability that the person is a serious danger to the community the Court was in error in applying the civil standard of proof. Accordingly this ground is made out.
Ground 1(a) and the first part of ground 1(b1)
Ground 1(a) complains that his Honour, in reaching his ultimate conclusion that the appellant was still a serious danger to the community, failed to have regard to the criteria referred to in s 6(2) of the Act. In his submissions concerning ground 1(b1) counsel for JD contended that the judge erred in misconceiving and misapplying the tests in s 6(1) and s 7(1) in two respects: first, “in relation to the s 6(1) test, failing to conduct a balancing exercise in order to determine whether the risk that JD would commit a serious sex offence was unacceptable”; second, conflating the s 6(1) test with the s 71(1) [sic] test by “having regard to evidence from the Commissioner of Correctional Services to the effect that supervision was not reasonably practicable, in order to find that JD represented an unacceptable risk of committing a serious sex offence.”[64] We shall discuss the second alleged error later in these reasons.
Counsel submitted that the concept of “unacceptable risk” involved a balancing exercise. Counsel referred to Director of Public Prosecutions (WA) v GTR[65] where Steytler P and Buss JA said:
The word “unacceptable” necessarily connotes a balancing exercise, requiring the Court to have regard, amongst other things, for the nature of the risk (the commission of a serious sexual offence, with serious consequences for the victim) and the likelihood of the risk coming to fruition, on the one hand, and the serious consequences for the offender, on the other, if an order is made…
Similar conclusions were reached by the Victorian Court of Appeal in Nigro[66] and by the Supreme Court of Queensland in The State of Queensland v DBJ.[67] In addition, the passage from the decision in GTR quoted above was accepted by Barr J in JD (No 2).[68] We agree that the test is not met by any risk that the appellant may commit a further serious sexual offence. Clearly some risks can be acceptable consistently with the adequate protection of the community.
It was put that neither the nature of the risk nor the consequences for the appellant or victims were analysed by the learned Judge. It was further put that although his Honour stated that there was an unacceptable risk, no details were provided, nor was a finding made regarding the likelihood of the appellant committing a serious sex offence.
As noted previously, on hearing an application to review a final continuing detention order under s 65, the Court can only confirm that order if it is satisfied that the detainee “is still a danger to the community”: see s 71(2). The burden of proof remains on the Attorney-General to prove that the detainee is still a serious danger to the community (s 72). Whether or not a person is “still” a serious danger to the community refers back to s 6 which in turn refers back to s 7 as to the burden and standard of proof. Consequently, the Court is required to have regard to the factors mentioned in s 6(2). In addition, s 9 requires that the Court must have regard to the matters in s 9, which means that the Court must give paramount consideration to the need to protect the victims of serious sexual offences committed or likely to be committed by the offender, including the victims’ families and members of the community generally. In this case, there was no information before the Court relating to any concerns which the victims or their families may have. Rather, the focus was on the need to protect the community generally. A secondary consideration is the desirability of providing rehabilitation, care and treatment to the offender: see s 9(2).
Counsel for the respondent argued that the findings of the learned Judge that the appellant was still a serious danger to the community had to be read in the light of the court’s previous decisions in JD (No 2) and JD (No 3). Although not stated in so many words, the submission must be in part based upon the word “still”: in other words, bearing in mind that the appellant had twice previously been found to be a serious danger to the community if released, one interpretation of the respondent’s position is that all it had to prove was that nothing had changed since the last review; the same factors which led to the earlier findings were still present and had not ameliorated since the last review.
In JD (No 2) the Court had the benefit of detailed reports and evidence from two psychiatrists, Dr Beech and Dr Walton. Both placed the appellant in the high risk category of future sexual offending if released into the community. Barr J concluded that he was satisfied to a high degree of probability that the evidence of Dr Beech was cogent and of sufficient weight for him to rely on it in concluding that there was an unacceptable risk that the appellant would commit a serious sex offence unless he were in custody.[69] Indeed, Dr Beech’s evidence went so far as to state that the risk in JD’s case was “acute, that is something that is likely to happen within a fairly short time after his release.”[70] His Honour also referred with apparent approval to the opinion of Dr Walton that “by any sensible standard, JD is … in a high – very high category of risk of reoffending generally and re-offending specifically in relation to a serious sex offence.”[71]
In arriving at his conclusion that the risk was unacceptable, his Honour explained that he took into account a range of matters which his Honour listed in detail.[72] Further, his Honour went so far as to say, that:
Even if, contrary to my interpretation in para [6] above, the standard of proof of unacceptable risk of commission of a serious sexual offence were to “a high degree of probability”, I would still have been so satisfied on all the evidence.
In JD (No 3)[73] his Honour referred to a number of findings which he had made in JD (No 2) about JD’s borderline intellectual functioning, low level intellect, and significant deficits in executive functioning and memory, which affected his judgment and insight. His Honour also referred to JD’s severe anti-social disorder, which appeared to strongly persist, and to his paranoid and antagonistic behaviour while a serving prisoner. He then referred to Dr Sullivan’s detailed description of JD’s personality disorder in his report of 22 February 2017. This included his opinion, consistent with the opinion of all three of the other psychiatrists who had examined JD in recent times, to the effect that JD did not suffer psychosis and did not present with evidence of a diagnosable mental illness.
His Honour then dealt in some detail with the appellant’s participation in rehabilitation programs, noting that although he had made very good progress as a result of participation in the programs, he had started from a low base and still had a long way to go.[74] His Honour accepted the opinion of Dr Sullivan that the appellant was still at high risk of committing another serious sex offence if not detained in custody or subject to a supervision order.[75]
His Honour then dealt with the psychiatric evidence relating to the assessment of the risk in more detail. This included the evidence of Dr Olav Nielssen called by JD. The main thrust of Dr Nielssen’s evidence was, as his Honour put it, “that psychiatrists have very limited ability to predict the future behaviour of any individual, because the science of behaviour prediction is weak.” Dr Nielssen was critical of various tests used by Dr Sullivan in arriving at his assessment of the risk of reoffending. His Honour accepted that there were short-comings if the assessment was based on the tests alone. He observed that Dr Sullivan used the tests as part of what he called “structured professional judgment” which on the evidence was current best practice. His Honour again referred to the need for the risk to be “unacceptable” and for the Court to engage in a balancing exercise having regard to the nature of the risk and the likelihood of the risk being realised, as well as the consequences for the offender who may be detained or subject to an onerous supervision regime without having committed any further offence. His Honour referred back to para [141] of his previous judgment in JD (No 2) and to his acceptance there of the passage from GTR which we have quoted above at [60].[76] The focus of the judgment then moved to consider whether the community could be adequately protected by making a supervision order. His Honour concluded that effective management and supervision of JD was not reasonably practicable at that time, but that if he continued to make progress, it was foreseeable that the need to protect potential victims and the general community could be met by a final supervision order rather than a continuing detention order.[77] There was no appeal from that decision.
In the judgment appealed from, JD (No 4), the Court accepted the evidence of Dr Sullivan that the risk had not changed since his previous report. In paragraph [49] of his report of 1 August 2018[78], Dr Sullivan said:
The risk profile is essentially unchanged. This reflects the lack of discriminant capacity when static factors place [JD] into the high risk category but there is limited opportunity to demonstrate lowered dynamic risk in a correctional setting. [JD] is very institutionalised.
Later is his report, Dr Sullivan said (at paragraph 54):
[JD] would be at high risk of committing another serious sex offence if not detained in custody or subject to a supervision order. This is based on historical factors, although his recent behaviour shows marked improvement compared with earlier years of his incarceration. I consider that this risk could potentially be managed in the community if [JD] had stable and supported accommodation and a standard range of monitoring conditions. The risk of further offending would be associated with alcohol use.
His Honour also reached the view that Dr Sullivan had formed an unjustifiably favourable impression of JD’s conduct during the period under review.[79] Dr Sullivan had only referred to one episode in which JD had been aggressive and threatening. However there were a number of other incidents. These included JD becoming agitated and angry when spoken to by a clinical program facilitator about his withdrawal from the sexual offender treatment program (STOP) sessions, making statements that he had assaulted female staff members in the past, making threats to kill the judge, making a threat to kill a prison officer on his release, making threats of harm towards custodial staff, and assaulting a fellow prisoner with whom JD was said to be in a relationship which had ended. His Honour also noted the view expressed by Acting Commissioner Steer that those various incidents demonstrated “a marked increase in the number and severity of the breaches over this review period as compared to the previous review period.”[80]
His Honour then referred to Dr Sullivan’s conclusions at paragraph [54] of his report, and turned to consider whether or not it would be possible to provide JD with the level of direct supervision required to meet the paramount considerations set out in s 63(2) of the Act.[81] The evidence from Commissioner McNairn and his predecessor was that that was not reasonably practical at this stage, but that, provided that JD engaged appropriately in a Transition Plan, he could be ready for a supervision order within 12 to 18 months from October 2018.
Although there is no direct reference in the judgment appealed from to the balancing exercise required to be undertaken, it is clear from his Honour’s two previous judgments that his Honour was well aware of what was required before making a finding adverse to the appellant. In those circumstances, we do not think it was necessary for his Honour to repeat the factors that led to his original decision, bearing in mind that the factors relevant to the assessment of the risk had not changed.
The same conclusion is reached about the failure of the learned Judge to refer to the requirements of s 7(1) as to the quality of the evidence. It is evident that the learned Judge accepted the evidence of Dr Sullivan as well as the other witnesses called by the respondent. It is clear that His Honour was well aware of the standard that the evidence had to meet, as his Honour referred to that in both JD (No 2) at [142] and in JD (No 3) at [2]. There was in JD (No 4) no contrary opinion from Professor Boer, or anyone else, that the appellant was not an unacceptable risk.
The thrust of Professor Boer’s evidence was to criticize the validity of the use of certain assessment tools which he said were not cross-validated with a member of a unique cultural group. This meant that errors could be made. But, as his Honour found, Dr Sullivan did not rely only on these tools, but on what he described as structured clinical judgment tools. In re-examination Professor Boer referred to these (structured clinical judgment tools) as being “driven by the international empirical literature.” He also said that “it is much more likely that the RSVP (one of the tools used by Dr Sullivan) is going to be valid with indigenous offenders than say, Static 99” (a statistical tool with significant shortcomings, which Dr Sullivan acknowledged). Further, Professor Boer acknowledged that non-validated structured professional judgment instruments should be used in preference to unstructured clinical judgments. Whether in the light of those criticisms his Honour ought to have been so satisfied is a matter dealt with separately below.
We would dismiss ground 1(a) and the first part of ground 1(b1).
Ground 1(b1)
The second part of ground 1(b1) was that the learned Judge erred in having regard to the evidence of the Commissioner of Correctional Services that supervision was not reasonably practical when determining whether JD was a serious danger to the community.[82] Counsel for JD contended that his Honour “erred in conflating the test in s 6(1) with the test in s 71(1) [sic].”[83]
Mr Thomas contended that the legislation provided for a two staged process.[84] The first stage was to decide whether or not the Attorney-General had satisfied the Court that the detainee was still a serious danger to the community.[85] The second stage was to determine if the Attorney-General had proved that it was appropriate to confirm the continuing detention order or to make a final supervision order.[86]
Counsel contended that:
The conflation of these two tests dilutes the protection for individual liberty. An analysis which proceeds to determine whether a risk is unacceptable by considering if there are reasonably practical measures to mitigate against the risk substantially lowers the bar. A much broader range of risks will be unacceptable.[87]
Counsel contended that the error of reasoning is reflected in his Honour’s concluding remarks, at JD (No 4) at [54]:
In the circumstances, I considered that it was necessary to confirm the continuing detention order pursuant to s 71(1)(c) of the Act. I was satisfied that the respondent was still a serious danger to the community. There remained an unacceptable risk that he would commit a serious sex offence unless he were in custody. I did not consider that effective and appropriate management and supervision of the respondent in the community would be reasonably practicable.
(underlining added by us)

Counsel contended that the sentence underlined by us elides the two statutory tests. By considering whether JD was a serious danger to the community by reference to the practicability of measures available to mitigate that risk, his Honour change the Act’s two staged analysis into one.[88]
We accept that the first question must be whether the Attorney-General has proved that the appellant is still a serious danger to the community. If the answer to that question is yes, it is only after that, that the Court is required to consider whether or not to confirm the continuing detention order, or to revoke that order and make a final supervision order. However, we do not consider that his Honour conflated the two tests, as the appellant has submitted. His Honour clearly found that the appellant was still a serious danger to the community.[89] His Honour then considered whether or not the appellant should be made the subject of a final supervision order or a continuing detention order.
The relevant criteria for deciding whether or not to make a final supervision order (as opposed to a continuing detention order) are set out in s 14(2) and (3) of the Act. Importantly, s 14(2) requires the Court to have regard to the need to protect the community as a paramount consideration. Section 14(3) requires the Court to consider whether adequate protection could only reasonably be provided by making a continuing detention order. Section 14(2) requires the Court to have regard to whether it will be reasonably practical for the Commissioner of Correctional Services to ensure that the detainee is appropriately managed and supervised as mentioned in s 63.
The Court heard evidence from Acting Commissioner Steer, Commissioner McNairn and Dr Sullivan which included the following matters: the appellant had a number of behavioural breaches whilst in prison suggesting a fairly low frustration threshold or difficulty regulating negative emotion in time of stress; whether the risk could be potentially managed in the community if the appellant had stable and supported accommodation and a standard range of monitoring conditions; that Community Corrections staff were not trained in the same physical restraint and self-defence techniques utilised by custodial officers so as to properly deal with aggressive and violent behaviour and resolve and de-escalate violent confrontations; the appellant had not completed necessary rehabilitation programs to lessen the risk of his sexual re-offending; no appropriate accommodation options had been identified by either Community Corrections or by the appellant; the appellant’s wish to live at One Mile Dam with his alleged promised wife was not feasible because his alleged promised wife did not live there, that she was not his promised wife and had no intention of living with him; and that his relatively low level of cognitive functioning would affect his ability to fully understand and comply with the terms of a final supervision order.
The Court also heard evidence from an occupational therapist concerning the assistance JD would need if released under such an order, and that a seven-staged Transition Plan had been prepared by Corrections which would enable him to be placed under supervision when completed. He had thus far only completed stage 1 of the Plan and was then progressing through Stage 2.
His Honour considered that effective and appropriate management and supervision in the community was not reasonably practicable to provide that level of direct supervision required to meet the paramount consideration set out in s 63(2) of the Act. His Honour did not conflate the two tests as alleged. We would dismiss this ground of appeal.
Ground 2
This ground asserts that the learned Judge’s decision to confirm the continuing detention order occurred in circumstances where the absence of accommodation outside Darwin Correctional Centre precluded the possibility of making a supervision order thereby resulting in the Act operating punitively, contrary to the intent of the Act, with the consequence that there was no lawful basis for the Judge to make an order of any kind under the Act.
Counsel for the appellant referred to the evidence of Commissioner McNairn that the Department of Correctional Services “was not resourced for a structured setting outside” of the prison which was “one of the problems” with JD being adequately supervised outside.[90] There was evidence that the prison had cottages at the Darwin Correctional Precinct which counsel suggested could be used on a trial basis to see how JD would manage with cooking, shopping, managing his electronic bracelet etc. However, Commissioner McNairn said that he did not have the staff to adequately manage the appellant in that setting.[91]
His Honour referred to the Cottage option[92]:
Acting Commissioner Steer, like his predecessors, had considered the possibility of [JD] living in one of the cottages at the Darwin Correctional Precinct. His concern was that it would not be possible to provide the level of support implicit in Dr Sullivan’s reference to “supported accommodation.” The Cottages (as they are called) are not actively supervised by prison officers. There would be a risk of conflict and social problems between [JD] and other supervisees living there, a significant risk given [JD’s] volatility and history of violent behaviour towards prison staff and other prisoners. Moreover, [JD] himself always rejected the idea of living in the Cottages and would become frustrated and agitated when the possibility of living there was discussed with him. He constantly repeated that he wanted to live at One Mile Dam with his promised wife.
Counsel referred to comments of his Honour during oral submissions where his Honour referred to the lack of accommodation and appropriate resources for people like the appellant who arguably should not be in the prison environment, and how in the absence of acceptable residential arrangements, this remains a significant obstacle to the appellant’s release. His Honour wondered whether in these circumstances he could impose a supervision order when there were no resources available to make that a practical and viable option.[93]
Mr Thomas contended that the Commissioner of Correctional Services “has a statutory obligation to provide a community-based alternative to prison (where supervision is deemed reasonably practicable)”[94] and concluded his written submissions concerning ground 2 by contending that the Commissioner was in breach of “his statutory obligation to provide accommodation in which the appellant may be appropriately managed and supervised.”[95] Counsel did not point to any part of the Act or refer to any relevant case law that supported those important contentions. The only reference to case law was to dicta of Hasluck J in Director of Public Prosecutions (WA) v Mangolamara[96] which related to the court’s exercise of its powers when making orders bearing upon the liberty of the subject, as distinct from the existence or otherwise of statutory obligations upon the Executive. We reject this contention.
We were referred to a number of cases where courts have had to deal with the difficulties in accommodating a detainee outside a prison. It is not necessary to consider all of them: they were adequately reviewed by Corboy J in Director of Public Prosecutions(WA) v Pindan [No 3].[97] Suffice it to say that there is no authority to support the proposition that the Court can refuse to make an order for the continuing detention of a detainee once a finding has been made that the detainee is a serious danger to the community, merely because the protection of the community could be met by making a supervision order if adequate resources were in place but are not.
Section 71 of the Act does not contemplate the making of such an order, which would not be in accordance with the principal object of the Act, namely to provide for the protection of the community. The rehabilitation, care and treatment of the detainee is a secondary consideration. Further, although the Court can impose conditions, the type of conditions contemplated by ss 18 and 19 relate to conditions which the detainee must comply with and do not include conditions of a type which cannot be met by the Executive. Such an idea is contrary to the requirement of the Court under s 14(3)(b) to consider whether it will be reasonably practical for the Commissioner to ensure that the person is appropriately managed and supervised as mentioned in s 63.
Counsel referred to some observations of the Queensland Court of Appeal in Attorney-General (Qld) v Francis[98] that there might be some circumstances where the Court would make a supervision order if the detainee had clear and pressing prospects of rehabilitation, in circumstances where further treatment is necessary to ensure adequate protection of the community but that treatment is not available in detention. The Court then went on to say:
It is possible, too, that the view taken by Gummow J in Fardon v Attorney-General for Queensland supports the argument that executive government repudiation of the preventative objects of the Act in a particular case (as, for example, by the refusal of any treatment to a prisoner clearly capable of, and amenable to rehabilitation) could lead the court to refuse to make any order at all. If it were to appear to the court that any further detention would be truly punitive in character, and, thus, contrary to the intention of the legislation, there would be no basis for the court to make an order of any kind under the Act. The conditions of further restraint upon the detainee’s liberty would be out of character with the intention of the legislature that such restraint is preventative.
However, the QLD Act is differently structured in one important respect to the NT Act in that s 13(5)(a) of the QLD Act identified three purposes for which an order may be made: control of the dangerous prisoner, care for the dangerous prisoner, or treatment of the dangerous prisoner. As the Court pointed out in Francis[99], the words “control, care and treatment” must be read disjunctively:
This disjunctive reading suggests that there may be cases where the basis for an order may be, either
·The control of an incorrigible offender, or
·The care of an offender whose propensities endanger the offender as well as others, or
·The treatment of an offender with a view to rehabilitation.
It will often be the case that more than one of these considerations will inform the making or an order.
As noted previously at [94], the primary object of the Act, as set out in the Objects of the Act, s 3, is to enhance the protection and safety of victims of serious sex offences and of the community generally by allowing for the control, by continued detention or supervised release, of offenders who have committed serious sex offences and pose a serious danger to the community. The secondary object is to provide for the continuing rehabilitation, care and treatment of those offenders. There is no power under the Act to order the detention of a prisoner merely for the purpose of rehabilitation if the person is no longer a serious danger to the community. As Corboy J noted in Pindan [No 3][100]:
It appears that the Court of Appeal had in mind circumstances in which a continuing detention order had been made solely for the treatment of an offender and not because the offender also required a degree of control that could only be provided by further detention.
We reject the appellant’s contentions. In any event, the authorities referred to in Pindon [No 3], and in the decision of the Queensland Court of Appeal in Attorney-General for the State of Queensland v Sybenga[101] support the conclusion that the Court has no power to force the Executive to make facilities available if there are insufficient resources for this to occur. Furthermore, the Commissioner had put in place a Treatment Plan which it was hoped would enable the appellant to be released under a supervision order in due course. In our opinion there was no error in making the continuing detention order on this ground, and we would dismiss this ground of appeal.
The remaining grounds
We reject ground 1(c). Counsel for the appellant did not say why it was necessary for his Honour to refer to s 7(2) of the Act. Nor did he point to any part of his Honour’s reasons from which it might be inferred that the Attorney-General did not bear the onus of proof. Consistently with his approach in JD (No 2) and JD (No 3) his Honour proceeded on the assumption that the Attorney-General bore the onus of proof.

[100]The remaining grounds essentially attack the finding that the appellant is still a serious danger to the community. It is convenient to deal with them together.
[101]In summary, the appellant complains that the Court failed to give little or no weight to the utilization by Dr Sullivan of three psychiatric risk assessment tools all of which had no validation in predicting the risk of the appellant in the commission of serious sexual offences in the future, and failed to take into account a number of factors which reduce the risk of the appellant committing another serious sexual offence. These grounds all relate to matters relevant to the assessment of whether or not there was an unacceptable risk that the appellant “will commit a serious sex offence unless he is in custody or subject to a supervision order”: see s 6.
[102]To the extent that it is alleged that the learned Judge failed to take into account factors which reduced the risk, the short answer is that the Judge did take some of them into account. Counsel referred to the following matters:
(a)The appellant’s health issues: these were referred to at paragraph [30] of the judgment, where his Honour referred to
Dr Sullivan‘s opinion that [JD’s] “overall sexual offending risk” remained in the high range. One risk-mitigation factor, referred to by Dr Sullivan, was [JD’s] overall declining physical health, which “increasingly suggests a degree of frailty which may act as a protective factor were he to return to the community.” However, the possible effects of declining physical health on sexual behaviours, about which Dr Sullivan speculated, were unknown.
The appellant’s health issues were not such as strongly suggest to the Court that he was no longer physically capable of sexual activity or no longer likely to be interested in sexual activity. Further, the opinion of Dr Beech which his Honour accepted in JD (No 2)[102] was that the appellant’s heart condition “has not resulted in his becoming frail or weak to the extent that he is not capable of committing a serious sex offence.”
(b)The appellant’s relatively advanced age: age and state of health are factors incorporated in the structured clinical judgment used by Dr Sullivan.[103] JD was born on 1 July 1969 and thus was aged 50 at the time of the review judgment, and 49 at the time he was interviewed by Dr Sullivan. Although no particular mention of his age was made in his Honour’s reasons, the evidence of Dr Sullivan did not support a finding that his age reduced the risk in his case. In essence, Dr Sullivan accepted that “there is a reduction of sexual re-offending which is more prominent at the age of 40 in advanced populations. And that number increases until only a small number of people are still offending in their sixties.”[104] However, he went on to say that he could not tell which of those people will be offenders and which will be non-offenders.[105] It is not clear what was meant by “advanced populations” (but we presume he meant advanced in age).
(c)The appellant was not diagnosed as being mentally ill. His Honour referred to that at paragraph [9] of the judgment.

[103]A number of other matters were put as not having been the subject of specific mention in his Honour’s reasons:
(a)The appellant was not diagnosed as having paedophilic tendencies. There was no suggestion of this by Dr Sullivan or by any of the other psychiatrists who have examined the appellant. It is not remarkable that no mention is made of it.
(b)There was no similarity between the two incidents of sexual offending. Again, it was not suggested that there was any similarity.
(c)There was a gap of 17 years between the two incidents of sexual offending. Although not mentioned specifically, his Honour was well aware of this. His Honour dealt with the facts of the previous offences in some detail in JD (No 2) at [28] - [53]. There was no evidence that this gap reduced the relevant risk. Dr Sullivan was cross-examined about this and said that there are is a range of risk factors relating to the causation of sexual offending. The two most important were deviant sexual arousal and social attitudes. In the appellant’s case, the past sexual offending is not as relevant as that fact that he has a “dense offending history.”[106] It was then put to him that the fact that he had this extensive history did not necessarily help to provide reliable guidance on whether he was likely to commit a serious sexual offence. Dr Sullivan replied that both Static-99 and RSVP, based on the research literature, provide an evidence-based association [of a dense based offending history] with re-offending.[107] Significantly, Dr Sullivan did not accept that the gap of 17 years in the offending was of any significance.
(d)The two incidents of sexual offending represented less than 2% of total offences on JD’s criminal record. Even if that is so, there was no evidence that this was of particular relevance to the assessment of the relevant risk. On the contrary, the evidence of the appellant’s prior offending history, and Dr Sullivan’s evidence about this referred to above, inferred that this factor was of no significance.
(e)The absence of any planning or concealment in regard to the final sexual offence. There was no evidence that this affected the assessment of the risk.
(f)Failing to take into account that the risk assessment was affected by the appellant’s long-term incarceration in an institutional setting which meant that there was limited opportunity to demonstrate lowered dynamic risk and consequently a great emphasis placed upon static risk factors. His Honour referred to this in his judgment at [13]. That is a factor which applies in all cases where a person has been incarcerated for a long period of time. It does not mean that the risk is in any way lessened. What was important in gauging the assessment of the risk were factors such as the appellant’s behaviour in prison, whether he had undertaken any courses made available to him to reduce the risk, such as alcohol rehabilitation, sex offender rehabilitation programs, anger management and the like, and whether he had been given any opportunities in prison to show a change of attitude. The findings of his Honour were that he remained threatening and aggressive, that he was angry when spoken to by a clinical program facilitator about his withdrawal from the sexual offender treatment program, that he had threatened staff on a number of occasions when he got out of prison, and that there were “a marked increase in the number and severity of the breaches over the review period.” These were all matters referred to by his Honour in the assessment of the risk.
(g)Failing to find that Dr Sullivan’s capacity to engage in a meaningful clinical exercise with the appellant was subject to significant limitations due to the fact that his recent interview with JD was conducted on one occasion only on audio-visual link and in 45 minutes. It was put that the limited conference conducted with a man with borderline intellectual disability in a language which was not his first language without the use of an interpreter meant that “heightened attention must have been placed upon the actuarial and SPJ instruments used by Dr Sullivan.” However, as his Honour observed, Dr Sullivan was familiar with the appellant having met with him in person in February 2017 for the purpose of an earlier court-ordered assessment.[108] When cross-examined on this subject, Dr Sullivan explained why he considered this more recent attendance was adequate for the purpose of preparing his report. In summary, he said that whilst it is preferable to use an interpreter, he is experienced at interviewing cognitively impaired offenders and that he uses a range of strategies to ensure that the offender understands and that are aimed to ensure that the offender’s response meets the purposes of the question. He said that he ended the interview only when he was not getting any further useful information despite repeated efforts to ask different questions.[109] His Honour noted that Professor Boer also interviewed the appellant without an interpreter and stated that the appellant “strove to make himself understood and responded well to requests for clarification.”[110]

[104]Professor Boer was asked to comment on whether “the structured professional judgement” used by Dr Sullivan, “based on a 45 minute AVL interview with [JD], in the absence of an interpreter, had any reliable basis regarding making a risk assessment under the Act.” Professor Boer wrote[111]:
I am not qualified to question or evaluate Dr Sullivan’s expertise. In my experience, he is an excellent clinician. I note that it was his second assessment … that was performed by Audio Video Link. He had already met [JD] for an in-person interview for the 22 February 2017 report. “Reliability”, when it comes to assessments, relates to the degree of consistency across assessor (or across assessments, for example in Dr Sullivan’s two reports). Dr Sullivan’s reports are certainly consistent (the risk profile for the RSVP and the HCR V3 were, in Dr Sullivan’s words, “essentially unchanged”). I am sure that Dr Sullivan would have noted it in his report if he felt [JD] was unwilling to be interviewed by AVL or if Dr Sullivan had any doubt regarding the clinical utility of the information thereby gained. It would seem, on the surface, that the 45 minute interview was considered valid for the purposes of the assessment.
…
Ultimately, it is up to the Court to judge whether the 45 minute AVL interview, without an interpreter, would be a reliable basis for making a risk assessment under the Act. However, it would seem to me that Dr Sullivan gained consistent, i.e., reliable, information via that process.

[105]We do not consider that there is any substance in the submissions referred to in [98] and [99] above.
[106]A major argument put forward by Mr Thomas, both before Barr J and this Court, related to Dr Sullivan’s utilisation of the three psychiatric risk assessment tools: known as Static 99, RSVP and HCR-20 V3. Counsel contended that his Honour erred in failing to give little or no weight to the risk assessment made by Dr Sullivan due to his use of those tools.[112]
[107]The appellant called evidence from Professor Douglas Boer, a registered clinical psychologist.[113] Professor Boer’s report states that:
[JD] was referred to [him] for a psychological assessment … to provide an opinion on the validity and reliability of past risk assessments completed on [JD], with a focus on available risk assessment methodology used in [JD’s] case. Mr. Thomas [counsel for JD] provided a list of questions regarding “risk assessment” and a list of “formal questions” for my response …
Please note that I was not asked to provide an additional risk assessment in [JD’s] case or to evaluate his progress in treatment. …[114]

[108]Professor Boer stated that there are three main types of risk assessment:[115]
(a)“Unstructured professional judgment” (UPJ): ie essentially a risk assessment based on the clinician’s experience. He said that the research literature showed that risk levels attained using UPJ have the lowest level of predictive validity in both sexual and non-sexual violence risk assessment.
(b)Actuarial risk assessment instruments (ARAIs). These are tools comprised of “items that derived, weighted, and combined to yield an estimated likelihood of some form of future violence within a particular period of time with a known group of recidivists and non-recidivists and then applied to similar sorts of offenders.”
(c)Structured professional judgment tools (SPJs). These are “tools in which items are directly derived from the scientific (clinical and research) literature and are then either prospectively or retrospectively validated with known offender groups (usually by raters blind to recidivism).”

Professor Boer stated that there is good evidence that both ARAI and SPJ methods attain significant levels of predictive accuracy and that this is “best practice.”
[109]As to the actuarial test STATIC 99 used by Dr Sullivan, Professor Boer stated that Static-99, but not the updated version of it (Static-99R), was “possibly valid with indigenous offenders over a relatively brief period (about 3 years post-release.)”.[116] However, he stated that there were no well validated tools for the other tests utilised by Dr Sullivan.[117] Essentially, the reason for this was that these tests had not been validated with a unique cultural group, viz Aboriginal people. Consequently:
The resulting scores may be indicative of a risk level, but not of membership in a percentage likelihood grouping from the original, but very different culturally, validation study (eg the original Canadian and UK data). While likely better that UPJ, using an actuarial tool in an invalidated manner may also yield very problematic results - either portraying offenders as low risk when they are higher risk (putting the public at risk) or vice versa (possibly keeping low risk offenders in custody unnecessarily).[118]

[110]Professor Boer also expressed the view that tests such as the HCR-20 V3, which was a test designed for the assessment of risk for future physical violence, is not logically part of an assessment for risk for future sexual violence.[119]
[111]Professor Boer was asked to state whether, if it were to be accepted that risk assessments could be made, JD satisfies the criteria in the Act such that he ought to be considered a serious danger to the community in that there is an unacceptable risk that he will commit a serious sexual offence unless incarcerated or subject to a supervision order. He stated:
In my view, this is entirely a matter for the Court. It is my professional opinion that there is no empirical basis (e.g. based on a risk assessment test) for concluding [JD], or any other Indigenous sexual or non-sexual violent offender, is conclusively of a particular risk level. Hence, it is difficult to definitively conclude an estimate of his likelihood of committing another serious sex offence.[120]
(underlining added by us)

[112]Professor Boer’s evidence assumed that a test may not be valid for a different cultural group, without explaining why, other than it had been the subject of criticism in the literature and in some court decisions. The learned Judge asked him about this, to which he replied, in relation to Static-99:
Well, if offence- if arrest rates in general and being fined and convicted, it those rates are different across cultural groups within Australia, then that one item for example received a legal consequence more than three times, one of the items on the Static-99, might preferentially or the opposite thereof see Indigenous offenders receiving a score there more often than a non-indigenous offender. So if there is a higher rate of conviction or appearing in front of a court, Indigenous offenders might score higher on that one item for example. If they are offending in remote communities where virtually no victims are strangers, then some of those items could be differently scored.[121]

[113]In cross-examination Professor Boer was asked whether he was saying that the use of those tools and tests would not permit an expert like Dr Sullivan to express an opinion as to the likelihood of risk of reoffending, that the Court could place any proper weight on. He said:
… the psychiatrists can provide a knowledgeable opinion regarding risk. What … some - Dr Sullivan for example, are doing [are] using instruments that have proven validity and reliability in other settings for other groups, [sic] other data in this setting where they have not got that proven validity and reliability. It might be best practice in general, but when they are not validated, that is not best practice. Their estimate of likelihood may still be accurate. I cannot say otherwise. I do not know. So I think we’re really - how the court interpret[s] likelihood of that person committing another serious offence is entirely up to the court. All I am saying is that these measures are not – have not got proven validity for that purpose.[122]

[114]In his Honour’s reasons, his Honour dealt with these issues at considerable length.[123] It cannot be said that he failed to take Professor Boer’s evidence into account. His Honour noted that Dr Sullivan accepted that Static-99 did not enable identification of a particular person within a group of sex offenders who may be likely to offend within a nominated period. All that it did was simply to place a person within a risk category.[124] He also acknowledge the shortcomings of the RSVP instrument, specifically that it had not been validated in its application to Indigenous offenders. However, his Honour accepted Dr Sullivan’s opinion that RSVP was “one of the most robust predictors of sexual re-offending risk.”[125] In relation to HCR-20, his Honour referred to Dr Sullivan’s evidence that although it had not been proven to apply to Indigenous offenders generally or to those living in Aboriginal communities, it had been broadly used in a wide range of people in many countries around the world and that “most people regard it as having a reasonably broad application to a range of identified variables which appear to be correlated [to] a range of different populations.[126] His Honour concluded that he “failed to see why it would not be relevant to Aboriginal populations.”[127]
[115]In essence, his Honour concluded that it was not inappropriate for Dr Sullivan to have utilised HCR-20, RSVP or Static-99 as structured professional judgment instruments in the assessment of an individual’s risk of future sexual offending.[128] His Honour noted that Dr Sullivan used HCR-20 as a control for the RSVP instrument to ensure that RSVP did not simply capture a constellation of behaviours which disadvantage JD. He also used HCR-20 to identify unmet needs and appropriate interventions for medical management in the community.[129] As to Professor Boer’s opinion, his Honour referred to the fact that he said the “structured clinical judgment tools were driven by the international empirical literature, and that it was much more likely that the RSVP is going to be valid with Indigenous offenders than say, Static 99.” His Honour went on to say that, if judgments have to be made, Professor Boer had said that “non-validated structured professional judgment instruments should be used in preference to unstructured clinical judgments.”[130]
[116]Reference was made to an article tendered by the respondent authored by Alfred Allan, Catherine L Parry and others on Assessing the Risk of Australian Sexual Offenders Reoffending.[131] The article suggests that the reasons why the use of non-validated instruments to assess Indigenous Australians was problematical was because “people from this population group are significantly over-represented in the criminal justice system”[132], and they are also more likely to be re-convicted.[133] It is difficult to see how these factors, even if true, are likely to skew the results with the result that the risk factor is increased. Whilst it is a notorious fact that Aboriginal offenders are overly represented in the prison statistics in the Northern Territory, there was no statistical information available to the Court to show that this translated to an over-representation of Aboriginal offenders who have committed serious sex offences. The same applies to the re-conviction rate.
[117]Elsewhere the learned authors noted that a study by Smallbone and Rallings in 2013 found that
Indigenous offenders scored significantly higher than their non-Indigenous counterparts on the Static-99 and the Static-99-R and were over-represented in the higher risk categories and under-represented in the lower risk categories. Table 2 shows that the short term prediction accuracy of Static-99-R was low and Static-99 moderate for Indigenous sexual offenders. The Smallbone and Rallings paper, however, does not reveal that the Static-99s false-positive range for cut-off scores of 1 to 6 ranged from 94.4% to 83.1% because the editor and the reviewers instructed the authors not to report them.[134]

[118]Under the heading “Predictor variables that could be considered” the authors noted:
Our review of the quantitative published literature identified no variable other than those reported by Allan and colleagues (Allan & Dawson, 2002; Allen et al 2006) with strong enough support to justify inclusion in an instrument to assess the risk of Indigenous sexual recidivism. These authors found that poor coping skills, defined as maladaptive coping strategies (e.g. alcohol use), was a significant factor in distinguishing Indigenous Australian sexual recidivists from non-recidivists. They also found that unfeasible release plans (e.g. returning to their home community where the victim resides or where drug and alcohol use are prevalent) and unrealistic long-term goals (e.g. doing work which they did not have qualifications for) were strongly associated with recidivism among WA Indigenous male sexual reoffenders.[135]

[119]These observations are hardly surprising, and, if anything, suggest that when these factors are considered in assessing the risk in this case, (as they were by Dr Sullivan and by the learned Judge) that this strengthens rather than diminishes the weight to be given to Dr Sullivan’s opinion.
[120]His Honour observed that the authors of that article concluded that:
for the foreseeable future practitioners will have to continue using non-validated instruments. However, they stated an important caveat: the need for practitioners to disclose the limitations of their assessments, to enable courts to determine whether or not to admit and/or give weight to their evidence and opinions.[136]

[121]Although the report of Dr Sullivan did not disclose these limitations, he made clear in his evidence what those limitations were, and what use he made of those tests.
[122]There was no evidence that an assessment made using the instruments utilised by Dr Sullivan as part of his structured clinical judgment was likely to be false based on any empirical data other than Professor Boer’s opinion that some at least of the assessment tools used were not valid only because they had not been validated for use on an Aboriginal population. This assumes, rather than proves, that the cultural differences likely to exist between remote Aboriginal populations and populations elsewhere are such as are likely to bear adversely, and to the detriment of the appellant, on the evaluation of the risk. The learned Judge was alive to these issues and based his assessment of the risk largely on the opinion of Dr Sullivan. In the absence of any compelling evidence to the contrary, we do not think that it can be said that his Honour erred. We would dismiss this ground of appeal.
Conclusion and orders
[123]We have found that the learned Judge was in error in relation to his view as to the standard of proof in this case. That error is likely to have been perpetuated in dealing with the review. For that reason, the appeal must succeed in relation to that issue. Otherwise the appeal is dismissed.
[124]Since we heard this appeal Barr J has conducted another review and made orders under s 71(1)(b) of the Act revoking the continuing detention order and making a final supervision order in relation to JD.
[125]Accordingly we make the following orders:
1.Ground 1(b) set out in the “Additional Grounds of Appeal” is upheld.
2.The remaining grounds set out in the “Additional Grounds of Appeal” are dismissed.
3.The appeal is allowed.
4.The order of the Supreme Court made on 1 February 2019 is set aside.
5.The matter is remitted back to the Supreme Court for reconsideration.
6.We will hear the parties as to costs.

------------------------------

[1] See Attorney-General of the Northern Territory v JD (No 4) [2019] NTSC 82 (JD (No 4)).

[2] [2016] NTSC 12 (JD (No 2)).

[3] Attorney-General of the Northern Territory v JD [2015] NTSC 28, Mildren AJ (JD (No 1)).

[4] Attorney-General of the Northern Territory v JD [2017] NTSC 48, Barr J (JD (No 3)).

[5] See Dangerous Sexual Offenders Act 2006 (WA) (the WA Act); Dangerous Prisoners (Sexual Offenders) Act 2003 (Qld) (the QLD Act); Crimes (High Risk Offenders) Act 2006 (NSW) (the NSW Act); Serious Offenders Act 2018 (VIC) (the VIC Act) which repealed and replaced the Serious Sex Offenders (Detention and Supervision) Act 2009 (VIC) (the Repealed VIC Act).

[6] [2013] NTSC 68 (EE).

[7] See s 23(2) of the Act.

[8] Section 25(1) of the Act.

[9] Section 25(2)(a) of the Act.

[10] Sections 79-82 of the Act.

[11] Section 25(2)(b) of the Act.

[12] Section 31(1) of the Act.

[13] Section 31(2) of the Act.

[14] Section 67 of the Act.

[15] Sections 68 and 69 of the Act.

[16] Section 70(2) of the Act.

[17] Section 79 of the Act.

[18] Section 79 of the Act.

[19] Section 82 of the Act.

[20] Section 4 of the Act. There are no regulations made under the Act.

[21] Section 72 of the Act.

[22] Builders Licensing Board v Sperway Constructions (Syd) Pty Ltd [1976] HCA 62; 135 CLR 616 (Sperway) at 621 [4] per Mason J; Dwyer v Calco [2008] HCA 13; 234 CLR 124 at 128 [2].

[23] Sperway per Mason J at 621 [4]; Dwyer v Calco (2008) 234 CLR 124 at 128 [2].

[24] Allesch v Maunz [2000] HCA 40; 203 CLR 172 at 180-181 [23].

[25] Coal and Allied Operations Pty Limited v Australian Industrial Relations Commission [2000] HCA 47; 203 CLR 194 at 203 [12] per Gleeson CJ, Gaudron and Hayne JJ.

[26] Tourism Holdings Australia Pty Ltd v Commissioner ofTaxes (NT) [2005] NTCA 3; 15 NTLR 80 at [10] – [11] per Martin (BR) CJ.

[27] Coal and Allied Operation Pty Limited v Australian Industrial Relations Commission (2000) 203 CLR 194 at 203-204 [13].

[28] Sperway at 620; Allesch v Maunz (2000) 203 CLR 172 at 181 [23].

[29] CDJ v VAJ (No 2) [1998] HCA 76; 197 CLR 172; Nigro v Secretary to the Department of Justice [2013] VSCA 213; 41 VR 259 (Nigro) at [23].

[30] Allesch v Maunz (2000) 203 CLR 172 at 181 [23].

[31] Nigro at [29] citing Dwyer v Calco (2008) 234 CLR 124 at 128 [2].

[32] Nigro at [41].

[33] [1936] HCA 40; 55 CLR 499.

[34] Nigro at [13], [41], [55] and [64].

[35] House v The King (1936) 55 CLR 499 at 505.

[36] JD (No 1) at [17].

[37] JD (No 1) at [17] – [18] and [20] – [21].

[38] JD (No 2) at [19] – [118].

[39] JD (No 2) at [28] – [41].

[40] JD (No 2) at [48] – [52].

[41] JD (No 2) at [42] – [47].

[42] JD (No 2) at [54] – [118].

[43] JD (No 1) at [11].

[44] JD (No 1 [12].

[45] In relation to the reports from the Commissioner of Correctional Services see JD (No 2) at [144] – [156], JD (No 3) at [6] – [7], [11] – [14] and [34], and JD (No 4) at [26] – [27] and [35] – [53].

[46] See discussion in JD (No 2) at [119] – [137].

[47] See discussion in JD (No 3) at [9] – [10] and [15] – [33].

[48] See discussion in JD (No 4) at [7] – [34].

[49] JD (No 2) at [6]

[50][2005] NTCCA 18; 16 NTLR 117 (Leach) at [14].

[51] [2004] HCA 46; 223 CLR 575.

[52] Nigro at [149] and [151].

[53] [1996] AC 563.

[54] [1955] HCA 27; 92 CLR 390.

[55] Slaveski v Smith [2012] VSCA 25; 34 VR 206 at [24].

[56] Hansard, Debates, Thursday 14 February 2013, p 1053.

[57] See cases noted in footnote 156 and 157 to [144] in Nigro.

[58] Interpretation Act 1978 (NT) s 62(2)(f).

[59] Interpretation Act 1978 (NT) s 62(2)(h).

[60] Interpretation Act 1978 (NT) s 62B(1)(a).

[61] Interpretation Act 1978 (NT) s 62B(1)(b)(i).

[62] at [14].

[63] Referring to Briginshaw v Briginshaw [1938] HCA 34; 60 CLR 336 at 362.

[64] Written submissions titled “Additional Particular (b1) – Unacceptable Risk” provided to the Court on 8 June 2020 (Particular 1(b1) Submissions).

[65] [2008] WASCA 187; 38 WAR 307 (GTR) at 159 [27].

[66] citing Woods v Director of Public Prosecutions (WA) [2008] WASCA 188; 38 WAR 217 at 241 per Steytler and Buss JA.

[67] [2017] QSC 302 at [12]-[13] per Bowskill J.

[68] At [141].

[69] At [132].

[70] Cited at [129].

[71] Cited at [136].

[72] At [138]-[142].

[73] At [8]-[10].

[74] At [11]-[14]; [18]- [21].

[75] At [22].

[76] At [33].

[77] At [35-[38].

[78] Appeal Book (AB) at 717

[79] At [25] – [29].

[80] At [27].

[81] These are the same as those set out in ss 9(1) and 14(2).

[82] Particular 1(b1) Submissions [4].

[83] Particular 1(b1) Submissions [1].

[84] Particular 1(b1) Submissions [5].

[85] Referring to ss 6(1) and 71(2) of the Act.

[86] Referring to s 72 of the Act.

[87] Particular 1(b1) Submissions [6].

[88] Particular 1(b1) Submissions [7].

[89] See the second sentence in [54] quoted in [77] above.

[90] AB at 606.

[91] AB at 607.

[92] JD (No 4) at [40].

[93] AB at 648-649.

[94] Plaintiff's Written Submissions filed 28 May 2020 at [47].

[95] Ibid [50].

[96] [2007] WASC 71; 169 A Crim R 379.

[97] [2014] WASC 95 (Pindan [No 3]).

[98] [2006] QCA 324; [2007] 1 Qd R 396 (Francis) at [30]-[31] per Keane and Holmes JJA and Dutney J.

[99] At [28]-[29].

[100]At [83].

[101][2009] QCA 382 at [23].

[102] At [140].

[103] See JD (No 4) at [11].

[104] AB at 531.

[105] AB at 531-532.

[106] AB at 526.

[107] AB at 527.

[108] JD (No 4) at [7]-[8].

[109] AB at 574.

[110] JD (No 4) at [23].

[111] Exhibit R5, AB at 906.

[112] See grounds 1(e) and (f).

[113] Professor Boer is not “a medical expert” within the meaning of s 4 of the Act. No objection was taken to the admissibility of his evidence.

[114] Exhibit R5, AB at 899.

[115] Exhibit R5, AB at 903-4

[116] Exhibit R5, AB at 905.

[117] Exhibit R5, AB at 905.

[118] Exhibit R5, AB at 904.

[119] Exhibit R5, AB at 907.

[120] Exhibit R5, AB at 907.

[121] AB at 628.

[122] AB at 635.

[123] JD (No 4) at [11] - [24].

[124] JD (No 4) at [15].

[125] JD (No 4) at [15].

[126] JD (No 4) at [15].

[127] JD (No 4) at [20].

[128] JD (No 4) at [21].

[129] JD (No 4) at [21].

[130] JD (No 4) at [24].

[131] Exhibit A6, AB at 916-937.

[132] Exhibit A6, AB at 918.

[133] Exhibit A6, AB at 918.

[134] Exhibit A6, AB at 923. Presumably this was because the results were considered by the editor and reviewers to be unreliable.

[135] Exhibit A6 AB 929. These three factors are referred to elsewhere in the article as the so-called “3-Predictor model”: AB 922. This is not to be confused with the assessment instruments used. Apart from Static-99, the article does not specifically discuss the other assessment instruments used by Dr Sullivan, although reference is made to a finding by Blaxell J in Director of Public Prosecutions (WA) v Moolarvie [2008] WASC 37 at [59] that RSVP (which Dr Sullivan did use) is “a relatively sophisticated instrument which is carefully structured and seems to address all conceivable risk factors that may be relevant. … I have come to the view that a clinical assessment which is partially based upon the correct application of RSVP will have added weight.” No criticism was made of that finding in the article.

[136] JD (No 4) at [22].

[2020] NTCA 11

31 July 2020

Key Legal Topics

Areas of Law

Criminal Law

Legal Concepts

Appeal

Criminal Liability

Serious Sex Offenders Act 2013 (NT)

Continuing Detention Order

Standard of Proof

Public Protection