Para No
Appeal [1]
Declaration of inconsistency [2]
Extension of time [3]
Issues on appeal [4]
Mr McDonnell’s background [9]
Statutory scheme [12]
Was compliance with s 107E of the Parole Act a
prerequisite to the making of an ESO? [15]
Did the failure to provide Mr McDonnell the opportunity
to participate in the preparation of the health
assessor’s report invalidate the process? [19]
Factual background [20]
Was there a failure to comply with the Bill of Rights? [36]
Did the failure to provide for Mr McDonnell’s
participation mean the requirements for a report
were not met? [47]
Did the health assessors’ reports fail to specify
s 107F(2) criteria? [54]
Did the High Court fail to apply s 107H(3)? [58]
Was Mr McDonnell an eligible offender? [63]
Jurisdiction: conclusion [67]
What is the standard of proof? [69]
(a) “Satisfied” [71]
(b) “Likely” [76]
Was the nine year term of the ESO justified?
The High Court decision [86]
Submissions for Mr McDonnell [92]
Section 107(5) [95]
(a) Level of risk [100]
(b) Seriousness of harm [106]
(c) Duration of risk [107]
Overall assessment [112]
Should a declaration of inconsistency be made? [114]
Result [132]

Appeal

[1] On 19 May 2008, Baragwanath J made an extended supervision order (ESO) for a term of nine years in relation to the appellant, Mr McDonnell: Chief Executive of the Department of Corrections v McDonnell HC AK CRI-2005-404-239. Mr McDonnell appeals against the order under s 107R(1) of the Parole Act 2002. Part 13 of the Crimes Act 1961 applies to the appeal as if it were an appeal against sentence: s 107R(2).

Declaration of inconsistency

[2] In addition, the appellant seeks a declaration that the provisions in Part 1A of the Parole Act providing for ESOs are inconsistent with the New Zealand Bill of Rights Act 1990 (the Bill of Rights). That application is pursued only if this Court determines that McMcDonnell is eligible to have an ESO imposed on him, and if the High Court decision to make the ESO is upheld.

Extension of time

[3] The notice of appeal was filed almost two months after the expiry of the period in which an appeal must be brought. The respondent (the Chief Executive) accepts that the delay has not caused prejudice and does not oppose an extension of time being granted. We therefore extend the time for appealing to the date on which the appeal was accepted for filing in this Court.

Issues on appeal

[4] The challenge to the issue of the ESO in this case is directed both at the jurisdiction to issue an ESO and the merits of the decision to do so.
[5] The jurisdictional challenge raises the following issues:

(a) Whether the obligation to assess eligible offenders under s 107E of the Parole Act is a prerequisite to an application for, or the making of, an ESO;

(b) Whether the failure of a health assessor to provide an opportunity for the offender to participate in the health assessment process:

(i) breaches any provision of the Bill of Rights;

(ii) means there is no health assessor’s report for the purposes of s 107F(2);

(c) Whether the health assessor’s report failed to deal with factors which s 107F(2) requires to be addressed;
(d) Whether there was a failure to comply with s 107H(3);
(e) Whether Mr McDonnell was an eligible offender.
[6] The merits decision raises the following issues:

(a) Whether the Chief Executive is required to prove beyond reasonable doubt that an offender is likely to commit relevant offences under s 107I(2);

(b) Whether the evidence before the High Court was sufficient to found the decision to make an ESO for a term of nine years.
[7] The notice of appeal and the written submissions foreshadowed a challenge to the merits of the decision to make an ESO at all. However, at the hearing, Mr Bott, who conducted this aspect of the appeal for the appellant, confined his challenge on the merits to an argument that the term of the ESO was too long, rather than arguing that an ESO should not have been made. We record that, having considered the evidence before the High Court and in light of our conclusion on the standard of proof, we have no doubt that an ESO was rightly made in this case.
[8] Before turning to the issues identified above, we briefly summarise the factual background and the relevant statutory provisions.

Mr McDonnell’s background

[9] Mr McDonnell’s background is summarised in the judgment under appeal as follows:

[11] The respondent is nearly 47. The pre-sentence report recorded that he had no significant previous convictions. There was brilliance and psychiatric instability among his siblings – one is a scientist who has won international prizes, another suffers from schizophrenia. He left university without graduating and after three years as a Seminarian obtained a number of one or two year positions in sales and marketing areas. In about 1992 he began to pursue professionally a longstanding interest in what he called “hypnotism, mind control, self-esteem, self-improvement”. He conducted public seminars on hypnotism. At the time of his offending he was living in rented accommodation and later lived in hotels. He lived beyond his means. He described himself as bisexual, not in any permanent relationship and without dependants. Cannabis and alcohol had been a feature of his social life. He described himself as being disowned by his family. The probation officer described him as presentable, positive and apparently highly intelligent. He also described him as personable, persuasive and predatory.

[12] Following conviction by a High Court jury, on 12 December 1995 the respondent was sentenced by this Court on eight counts to a series of concurrent terms of imprisonment. The longest sentence was of 10 years on one count of sexual violation by anal intercourse of a 15 year old boy. On five other counts of sexual violation he received terms of 7-8 years and [on] two counts of indecent assault he received three year terms, all concurrent. The boys were between 14 and 16 at the time of the offending.

[13] In his sentencing remarks Williams J described the respondent as reasonably well-educated, reasonably intelligent and capable of fulfilling a useful part in society had it not been for the activities for which he was to be sentenced. He was said to be very skilled in hypnosis, a talent which he had distorted to commit the offences.

[14] The Judge drew attention to the need to protect strangers, as the victims had been. He described as aggravating factors the use of drugs supplied to lessen the victim’s resistance, namely alcohol, cannabis and amyl nitrate in various combinations, and referred to the extent and grave character of the offending. He emphasised the gravity of use by an expert hypnotist of that skill to facilitate the offences. Also of relevance to the present application is the Judge’s description as an important aggravating factor that the respondent committed offences of indecent assault and sexual violation even after he had been charged with earlier offending. He stated (at 4):

The psychiatric assessment says that you have strongly held beliefs in your own capacity relating to hypnosis, freedom of expression, and your sexual preferences, and strongly held beliefs that you have been treated unfairly in the Court. You disagree strongly with views held by others with regard to your being considered a threat to young people, especially boys. The psychiatrist comments that you have some considerable difficulties in understanding or accepting anyone else’s views. There are echoes of that in other material which I have before me. Even the probation office says you have a lack of appreciation of the damage you have done to these young men.

[15] Dr Wilson said of the offending:

In terms of offence planning...there was evidence of Mr McDonnell planning and gaining access [to] victims by grooming in his offending[,] namely by use of drugs, taking the victim to a place where he had control over them and seeking out potential victims by gaining easy access to them through friendship.

The circumstances described by Williams J bear out that account.

[10] In June 2002, one month before the appellant’s release date, the Chief Executive applied to the Parole Board for an order under s 107 of the Parole Act that the appellant serve the full term of his sentence. On 20 March 2003, the Board made a s 107 order on the grounds that the appellant was likely, if released, to commit a specified offence before the expiry of his full sentence. As required (s 107(6)), the Board reviewed its order regularly (at least once every six months), each time deciding not to revoke it.
[11] Prior to the appellant’s scheduled release from prison on 18 July 2005, the Chief Executive applied for an ESO to be imposed on the appellant. It is this application that was dealt with in the judgment under appeal.

Statutory scheme

[12] The statutory regime for ESOs is set out in Part 1A of the Parole Act, which is summarised in two earlier decisions of this Court, Belcher v Chief Executive of the Department of Corrections [2007] 1 NZLR 507 at [3] – [8] (Belcher No 1) and R v Peta [2007] 2 NZLR 627 at [5] – [14]. We adopt those summaries but do not replicate them in this judgment.
[13] Part 1A is broken into a number of sub-parts. The first, containing ss 107A – 107E is headed “Preliminary”. The second, headed “Application for, and making of extended supervision orders”, contains ss 107F – 107IA. It is with the latter sub-part that the present application is principally concerned. Other sub-parts are headed “Nature of extended supervision order”, “Cancellation, extension, variation, and suspension of extended supervision orders or conditions”, “Appeal and reviews”, “Miscellaneous provisions”, and “Transitional measures”. The location of sections in different sub-parts is important in analysing the effect of one provision, s 107E, which is in issue in this case.
[14] We now turn to the issues relating to jurisdiction set out in [5] above.

Was compliance with s 107E of the Parole Act a prerequisite to the making of an ESO?

[15] Mr Ellis argued that there was nothing before the High Court to establish that the Chief Executive had complied with s 107E, which he argued was a prerequisite to making an ESO application. He argued that the Chief Executive had to give the offender a chance to be heard before making the assessment required by s 107E.
[16] Section 107E provides:

107E Obligation to assess eligible offenders

The chief executive must ensure that, before an eligible offender is released from detention, the offender is assessed to determine the likelihood that the offender will commit any of the relevant offences referred to in section 107B(2) after release.

[17] This argument was not made in the High Court, was not referred to in the notice of appeal to this Court and was not mentioned in the written submissions filed in advance of the hearing. Counsel for the Chief Executive, Ms Edwards, responded to the argument orally, without having had the opportunity to prepare. Perhaps for this reason neither counsel referred us to R v Brown [2005] NZCA 303; (2005) 22 CRNZ 233 in which this Court rejected an argument along the same lines. In that case, the Court said:

[42] We note that s 107E is in the Preliminary sub-part of Part 1A of the Act dealing with ESOs. It does not form part of the sub-part dealing with applications for and the making of ESOs. As we read the section, it imposes an obligation upon the [Chief Executive] to ensure all eligible offenders are assessed prior to their release for detention. It does no more. It does not require the assessment to be by a health assessor. It is for the purpose of an evaluation as to whether the offender is likely to commit a relevant offence after release. The process can only be for the purpose of the protection of the public. The section does not require the offender to be advised of the assessment.

[43] Nor is the section relevant to the application process. Nowhere in s 107F and following is there any reference to compliance with s 107E. If the legislature had intended that the s 107E assessment was relevant to the application, it undoubtedly would have made that clear. There is no basis for implying such a requirement.

[44] We have no doubt that the legislature intended it to be obligatory for the [Chief Executive] to assess eligible offenders before their release. However, any failure to do so is not a matter that on the face of it has any consequences whatever in respect of an application for an ESO.

[45] We specifically reject the submission for the respondent that it is implicit in s 107E that the offender be informed of the assessment to enable the offender to avoid jeopardy. It is significant that there is no requirement upon the [Chief Executive] under s 107G that a copy of any assessment obtained by the [Chief Executive] under s 107E should be made available to the offender. In any event the offender is not in jeopardy in respect of an ESO until there is an application for an ESO when the offender has an opportunity to meet the case for the [Chief Executive].

[46] The provisions of ss 107E [sic but apparently intended to be s 107F] to 107L inclusive are effectively a code in respect of the application for and the making of ESOs. Except to the extent that it may be necessary for the purpose of their interpretation to refer to provisions outside them there is no need to look outside them. There is nothing to suggest that the requirements in respect of the application for and the making of ESOs encompass material not within that sub-part of the Act. Certainly there is nothing to suggest that an offender can rely upon a provision such as s 107E, which is not in that sub-part. That is particularly so when s 107E gives the offender no rights whatever.

[47] We are therefore of the view that the Judge was wrong in considering that compliance with s 107E was a necessary requisite of the application for an ESO. Section 107E does not form part of the ESO application procedure.

[18] We agree.

Did the failure to provide Mr McDonnell the opportunity to participate in the preparation of the health assessor’s report invalidate the process?

[19] Section 107F(2) requires that any application for an ESO must be accompanied by a report by a health assessor, addressing specified factors. The present application was accordingly supported by a report dated 31 May 2005 prepared by Dr Ronnie Zuessman, a psychologist with the Department of Corrections. Subsequently, another Corrections psychologist, Dr Nick Wilson, prepared reports for the Court. Neither interviewed Mr McDonnell. Mr Ellis argued that their failure to do so, or to provide a proper opportunity for Mr McDonnell to participate in the preparation of their reports, was a fatal flaw in the process and deprived the Court of jurisdiction to make an ESO. To evaluate that argument, we need to trace the factual background in some detail.

Factual background

[20] We start with the circumstances leading to the preparation of Dr Zuessman’s report. The only information before the Court about what happened is the account which appears at the start of Dr Zuessman’s report. That account records that the appellant was originally advised that an “assessment appointment” was scheduled for 12 May 2005. However, at the appellant’s request, the appointment was rescheduled for 24 May, in order to give him time to consult his lawyer. Upon arriving at the appointed time, the appellant was informed of the purpose of the interview, to which he responded angrily and indicated that he did not trust Dr Zuessman. He refused to participate in the interview, stating that he realised his refusal, and the reasons for it, would be considered by the “Board”; he also refused to read or sign a two page consent form that was given to him. He then left the office and went back into the corridor. At this point, Dr Zuessman called out to inform the appellant that the report would be considered by a court, not the Parole Board. The appellant responded angrily and left.
[21] Dr Zuessman did not attempt to arrange any further interviews with the appellant and his report was therefore based entirely on “collateral sources”.
[22] Dr Paul Barrett, an expert engaged by the appellant, produced a report dated 10 March 2006 in which he evaluated Dr Zuessman’s report, particularly its usage of actuarial risk assessment tools.
[23] In mid-2005, Dr Zuessman resigned from his position with Corrections and took up a job in Australia. (At the time Dr Zuessman was asked to prepare his report, Corrections were aware of his plans to leave shortly thereafter.) This meant he would not be available for cross-examination at the hearing of the ESO application. Two subsequent reports in relation to the appellant, dated 4 May 2006 and 20 November 2006, were therefore prepared by Dr Wilson.
[24] Dr Wilson’s first report (4 May) responded to Dr Barrett’s report. More importantly, his second report (20 November) was prepared to replace Dr Zuessman’s assessment report, both because Dr Zuessman had since left New Zealand and due to the considerable time that had passed since that assessment. In compiling this report, Dr Wilson did not conduct (or even request) an assessment interview with the appellant. This meant the report was based entirely on collateral sources and previous psychological assessments of the appellant. Dr Wilson acknowledged that this limited the assessment in certain respects; in particular, it meant “it was not possible to accurately assess acute risk or protective factors that may be present” for the appellant.
[25] In the High Court, Mr Bott argued that the unavailability of Dr Zuessman for cross-examination was a fatal flaw that provided a jurisdictional bar to the issuing of an ESO. This was because Dr Zuessman’s affidavit had accompanied the application for the ESO. Baragwanath J rejected that argument: he found that the provisions of Part 1A must be interpreted to promote rather than defeat the statutory purpose expressed in s 107I(1) (protection of a community) and that the procedures had to accommodate cases where the original assessor died or became too ill to be cross-examined. He said that the availability of Dr Wilson’s affidavits, and his availability for cross-examination, meant that the health assessment relied on by the Chief Executive could be properly challenged.
[26] We agree. However, the Judge’s conclusion on this point was not challenged on appeal. Rather, a different argument was made. The essence of the argument is that both Dr Zuessman and Dr Wilson made their assessments without the involvement of the appellant, and the appellant was not given a reasonable opportunity to put his position to the report writers for consideration.
[27] Mr Ellis, who conducted this aspect of the appeal for the appellant, said that the original health assessment conducted by Dr Zuessman had been done by stealth or ambush. He said that the appellant had not been informed of the purpose of the interview before it was due to take place and that he had not been given the consent form prior to the beginning of the interview, which meant that he did not have an opportunity to consider the form or to take legal advice on it.
[28] A copy of the consent form presented to Mr McDonnell by Dr Zuessman was provided to us by counsel for the Chief Executive. It is two pages long, and requests the offender’s “informed consent to participate in assessment interviews with the health assessor to better inform his/her health assessment report”. It then records the following:

(a) that the offender understands the matters which the health assessment report must address (those specified in s 107F(2));

(b) the sources on which the health assessment report will be based;
(c) the risk assessment measures that will be used;
(d) the fact that participation in the interviews and psychological testing is voluntary;
(e) the fact that refusal to co-operate can be taken into account by the Court, but if so, the offender’s reasons for this refusal must also be taken into account (this is a reference to s 107H(3));
(f) the fact that an ESO may be made by the Court and that, if an order is made, the Parole Board may impose special conditions; and
(g) the fact that the report may later be accessed by Corrections staff, including for future reports to the Parole Board, and by any other health assessor who is preparing an additional report.
[29] As Dr Zuessman recorded, the appellant did not sign the form and refused to take part in the interview. A Corrections officer noted on the form:

Mr McDonnell refused to participate in the health assessment, did not want to speak with the psychologist, and refused to read the consent form.

[30] There was no evidence from Mr McDonnell before the High Court (indeed, he was not present in the High Court, in breach of the requirement of s 107G(4)) and no evidence in this Court either. The only evidence we have of what occurred is in the report prepared by Dr Zuessman, which records the events as described in [20] above.
[31] Dr Wilson treated the attitude taken by Mr McDonnell at the time of Dr Zuessman’s attempt to engage him in the interview as an indication that there was no point asking whether the appellant wanted to participate in an interview and assessment, in terms of the consent form described earlier.
[32] Dr Wilson said in evidence that he regretted not seeking to interview the appellant, but considered that his reasons for not doing so were valid. He was aware that the appellant had previously persistently refused contact with psychological staff and considered that a further request might put undue pressure on the appellant given the stage that the proceedings had reached by the time Dr Wilson had been asked to provide his assessment. (Mr McDonnell had refused to participate in interviews in 2002, 2004 (twice) and earlier in 2005.)
[33] Mr Ellis submitted that the failure to provide a proper opportunity for Mr McDonnell to consent to the assessment interview and to co-operate in the preparation of the health assessment report meant that neither Dr Zuessman nor Dr Wilson conducted a lawful assessment of the appellant. He said such an assessment was a prerequisite to the issuing of an ESO because s 107F(2) requires applications to “be accompanied by a report by a health assessor” addressing the matters set out in that subsection.
[34] Mr Ellis argued that for an assessment report to be valid, it must comply with the Bill of Rights, the International Covenant on Civil and Political Rights (ICCPR) and the common law. He said it failed to do so because it did not comply with the statutory provisions, did not comply with a number of provisions of the Bill of Rights (ss 24(c), 24(d), 25(b), 25(c) and 27) and their ICCPR equivalents, and involved a breach of natural justice. He said that the appellant should have been warned that anything he said could incriminate him, and informed that he was entitled to a lawyer.
[35] We will deal with the arguments relating to the Bill of Rights first.

Was there a failure to comply with the Bill of Rights?

[36] Mr Ellis said that the rights guaranteed by ss 24 and 25 applied to Mr McDonnell. Those sections apply to “everyone who is charged with an offence”. The nature of the rights show that they are intended as protections for a person who is charged with an offence and is to face trial. Mr Ellis said in his written submission that s 24 applied “as the charge is continuing with the prospect of a new penalty”. He did not otherwise elaborate on the application of ss 24 and 25 to a person who has been convicted and sentenced and is about to be released from prison having served his or her term of imprisonment, but is the subject of an application for an ESO.
[37] In Belcher No 1, this Court concluded at [49] that the imposition through the criminal justice system of significant restrictions (including detention) on offenders (by the imposition of an ESO) in response to criminal behaviour amounts to punishment and thus engages ss 25 and 26 of the Bill of Rights. The context of that finding was the allegation that the imposition of an ESO on an offender who had been sentenced prior to Part 1A of the Parole Act being enacted amounted to a breach of the protection in s 25(g) (the right to the benefit of a lesser penalty where the penalty has been varied between the commission of the offence and sentencing) and the right to be free from being convicted for an offence for which the offender had already been tried or punished (s 26(2)).
[38] The essence of Mr Ellis’ argument is that when an ESO is applied for, this should be treated as analogous with a new criminal charge being made, so that all of the rights in ss 24 and 25 apply to the offender. In particular, he referred to s 24(c) (the right to consult and instruct a lawyer), s 24(d) (the right to adequate time and facilities to prepare a defence), s 25(b) (the right to be tried without undue delay) and s 25(c) (the right to be presumed innocent until proved guilty according to law). However, instead of s 25(b), his argument appeared to be directed at s 25(d) (the right not to be compelled to be a witness) and we will deal with it on that basis.
[39] We do not consider it appropriate to treat an application for an ESO as being analogous with the bringing of a fresh charge against the offender. For example, it makes no sense to say that the right to be presumed innocent (of the offence which makes the offender eligible for the making of an ESO) applies to an offender who has been through a trial process and has been proved guilty according to law. A number of the other rights guaranteed by s 24 are equally inapplicable, such as the right to trial by jury (s 25(e)). We see the ESO process as analogous with the sentencing process which follows conviction, so that the rights guaranteed by ss 24 and 25 which apply in relation to sentencing apply equally to the ESO process. However, rights which are applicable to persons facing charges who have not yet been convicted, but which cease to be of relevance once a finding of guilt has been made according to law and a conviction has been entered, are not re-ignited when an ESO application is made.
[40] Thus, ss 24(d) and 25(c) are inapplicable to the ESO process. On the other hand, as this Court found in Belcher No 1, the rights referred to in ss 25(g) and 26(2) apply, but for the reasons given in Belcher No 1 these have been overridden by Part 1A. Of course, an offender who is the subject of an ESO application has the same right to legal representation as a prisoner facing sentence. Mr McDonnell exercised that right in this case and was represented in the High Court. That does not, however, mean that he was entitled to be represented at the interview with the health assessor: that interview is not the equivalent of a police interview of an offender before or soon after a charge is laid. He was, however, entitled to advice about the consequences of his consenting or refusing to consent to an interview with the health assessor.
[41] Although Mr McDonnell had been given notice of the proposed interview with Dr Zuessman, he was not given a copy of the proposed consent form prior to the interview date. Dr Zuessman did not have the chance to explain the purpose of the interview to him or to talk about the proposed consent form with him, and there was no opportunity to address Mr McDonnell’s desire for legal advice (because Mr McDonnell disengaged from the process effectively before it had begun). As the High Court Judge put aside Dr Zuessman’s report and relied only on Dr Wilson’s reports, the events that happened in relation to the proposed interview with Dr Zuessman are of little practical significance. But we observe that health assessors need to allow offenders to obtain legal advice on the issue of whether to consent to an interview, and the process they follow should allow for this.
[42] As stated earlier, there was no attempt to allow Mr McDonnell to participate in an interview with Dr Wilson, and we will revert to that issue later.
[43] We do not consider that s 25(d) applies in the present situation, for the reasons given (in a slightly different context) in Burke v Superintendent of Wellington Prison [2003] 3 NZLR 206 at [27] – [41] (HC) and R v Jones [1994] 2 SCR 229. Even if it had applied, no breach would have occurred in this case. Mr McDonnell did not participate in any part of the process and so made no incriminatory statement. If he had participated, it would have been because he consented, not because he was compelled.
[44] Mr Ellis also argued that an offender who is the subject of an ESO application has the right to the observance of the principles of natural justice by the health assessor who is responsible for preparing the assessment report. He said this requirement was not complied with by either Dr Zuessman or Dr Wilson.
[45] Section 27(1) gives a right to the observance of the principles of natural justice “by any tribunal or other public authority which has the power to make a determination in respect of that person’s rights, obligations, or interests”. Clearly, the Court which has the task of determining whether an ESO should or should not be made is required to comply with s 27(1). There was no suggestion that the Court had failed to do so in this case. Mr McDonnell was represented at the High Court hearing, the hearing was fair and public, he engaged his own expert witness and his counsel cross-examined the Chief Executive’s expert witness. He could have given evidence himself if he had wished to do so.
[46] We do not see the earlier stage of the process, where a health assessor is seeking to interview the offender in the course of preparing the health assessment required by s 107F(2), as engaging s 27(1). The health assessor is not determining any rights, obligations or interests of the offender, but rather is preparing a report for the body that will make such determination (the sentencing court). As this Court said in Barr v Chief Executive of the Department of Corrections CA60/06 20 November 2006 at [32], the decision to impose an ESO is a decision for the judge, not the health assessor. Section 27(1) did not apply to the manner in which Dr Zuessman and Dr Wilson prepared the health assessment reports that were submitted to the Court, but rather to the whole process leading to the decision as to whether the ESO should be made. We are satisfied that s 27(1) was complied with in this case.

Did the failure to provide for Mr McDonnell’s participation mean the requirements for a report were not met?

[47] Mr Ellis argued that the failure to provide Mr McDonnell with the opportunity to participate in an interview with Dr Zuessman or Dr Wilson meant there was no health assessor’s report before the Court as required by s 107F(2), because a report prepared in such circumstances was not a proper report.
[48] Dr Wilson acknowledged in his evidence-in-chief that an interview should be provided for if possible. He said:

It’s always preferred as a psychologist to be able to interview the person being assessed to be able to provide some information about their behaviour while interviewed and to be able to put information from collateral sources to them to gain their response and finally an interview enables protective information to be assessed and more immediate information on their functioning to be assessed.

[49] We agree, for the following reasons:

(a) The Court is required to assess dynamic risk factors: Peta at [30]. An individualised assessment of these factors is required: Peta at [52].

(b) As these factors change over time, the health assessor’s report should be as up-to-date as possible: Belcher No 1 at [90].
(c) An interview with the offender may increase the accuracy of the assessment of the level of risk and the ways in which it can be managed: Beech, Fisher and Thornton “Risk Assessment of Sex Offenders” (2003) 34(4) Professional Psychology: Research and Practice 339 (cited by Dr Wilson in his report in this case).
(d) Allowing offenders to participate in assessment processes of this kind is best practice: see Brookbanks and Simpson (eds) Psychiatry and the Law (2007) at 271, where the authors say (in relation to reports for consideration of preventive detention, but equally applicable to ESO reports):

At the outset, then, the prudent report writer should inform the subject of the purpose of the report; the nature of the assessment; the agency for which the report is written; the consequences of all disclosures made and the possibility that the information in the report may be made available to a wider audience. The subject should have the time and space to consider such matters and to consult legal counsel if he or she wishes.

(Emphasis added)

[50] In the present case, Mr McDonnell was not given a fair opportunity to consider whether to participate in the interview with Dr Zuessman. The original assessment was therefore conducted in his absence and without his consent. This is not, however, a situation where protected communications to a health assessor were disclosed without consent: R v D [2003] 1 NZLR 41 (CA). Instead, a statutorily required report was produced, albeit without giving the appellant fair opportunity to participate in its production. The situation in this case is therefore more similar to that which pertained in Belcher No 1, where this Court commented:

[98] Likewise, we do not accept that the appellant had a right of veto over choice of the psychologist who was to prepare the report. The statutory scheme assumes that reports will be provided and it would make a mockery of that scheme if an offender could derail the process by refusing to see the (or any) psychologist chosen by the chief executive to provide the necessary report. We likewise do not accept that the absence of consent on the part of the offender and associated unwillingness to engage in the process means that it is unethical for a psychologist to provide a statutorily required report. In all of this, it is important to recognise that the decision to impose an ESO rests with the Court and that it is open to the offender to place other evidence before the Court, as the appellant did in this case.

(Emphasis added)

[51] It is also clear (and, indeed, Dr Wilson himself acknowledged this) that Dr Wilson should not have assumed that Mr McDonnell would refuse to participate in the process. While Mr Ellis accepted that the assumption was probably correct in fact, that does not alter the fact that it was for Mr McDonnell to make the decision about participation in the process, not Dr Wilson.
[52] But we do not see these failures as fatal to the process for the reasons given in Belcher No 1. We need to consider the fairness of the process as a whole. If Mr McDonnell had considered that his lack of participation in the report process made the reports more unfavourable to him than they otherwise would have been, he could have corrected this at the hearing. He did, in fact, call an expert witness, but not for that purpose. Nor did he give evidence. In fact, he did not even attend the hearing. That was the opportunity to put before the Court evidence to counterbalance the views of the health assessors.
[53] We are satisfied that the failure to involve Mr McDonnell in the preparation of the reports did not render those reports invalid or remove the Court’s jurisdiction to make an ESO. Procedural flaws in the process leading to the preparation of a report will, however, be relevant to the weight given by the judge to the report.

Did the health assessors’ reports fail to specify s 107F(2) criteria?

[54] Mr Ellis argued that Dr Zuessman’s and Dr Wilson’s reports failed to deal with some of the specific criteria in s 107F(2), and therefore did not amount to health assessors’ reports at all for the purposes of that section. In particular, he said that the reports did not address s 107F(2)(a) (nature of any likely future sexual offending) or s 107F(2)(c) (predilection and proclivity for sexual offending). He also argued that Dr Wilson’s report dealt inadequately with Mr McDonnell’s ability to control his sexual impulses (s 107F(2)(b)).
[55] That submission must be evaluated in light of the decision in Grieve v Chief Executive of the Department of Corrections (2005) 22 CRNZ 20 at [14] and [15], in which this Court held that some leeway should be given to report writers as to the way they deal with the s 107F(2) factors. The Court held that a report will be sufficient if the judge is able to discern the health assessor’s view on the relevant factors, even if only by inference. It rejected an argument that the report needed to be conclusionary on each matter, because the judge is required to make his or her own assessment on all matters relevant to the application.
[56] We are satisfied that Dr Zuessman’s report (at 6) expressly addressed the nature of Mr McDonnell’s likely future sexual offending, and that the discussion of Mr McDonnell’s dynamic risk factors at pages 5 – 6 provide sufficient guidance about his predilection and proclivity for sexual offending to satisfy s 107F(c).
[57] We are also satisfied that Dr Wilson addressed all of the factors set out in s 107F(2). At [48] – [52] of his report he expressly dealt with each mandatory factor in turn. In particular, at [49] he dealt with Mr McDonnell’s ability to control his sexual impulses, noting that his offending has been persistent and predatory and remains untreated. Earlier in the report, Dr Wilson noted that sexual self-regulation was identified as a problematic risk factor for the appellant (at [42]). Evidence of the appellant displaying impulsive behaviour was also highlighted at [45], namely his being found under the influence of drugs, with an antisocial associate, in an area after release known to be frequented by under-age prostitutes.

Did the High Court fail to apply s 107H(3)?

[58] Section 107H(3) provides:

At any hearing, the court is entitled to take into account the fact that an offender refused to co-operate with the preparation of the health assessor’s report required under s 107F(2), but it must also take into account any reasons the offender gives for refusal to co-operate with the preparation of the health assessor’s report.

[59] Mr Ellis said that the Judge had failed to consider why Mr McDonnell refused to be interviewed by Dr Zuessman or why there was no co-operation with Dr Wilson. He said this was contrary to s 107H(3) and to natural justice.
[60] In his judgment, Baragwanath J did not consider the propriety of Dr Wilson not having invited the appellant to attend an interview. The Judge did, however, accept at [62] that, “in accordance with his Bill of Rights entitlement, no inference adverse to the [appellant] may be drawn from the election not to be interviewed or to give evidence.”
[61] This conclusion was obviously favourable to the appellant, although it is perhaps difficult to reconcile with s 107H(3) itself, which seems to permit the drawing of adverse inferences from a refusal to co-operate with the preparation of a heath assessor’s report.
[62] We agree with Ms Edwards that the obligation imposed by s 107H(3) comes into play only if the Court does, in fact, take into account the fact that the offender has refused to co-operate. As Baragwanath J did not do so in this case, there was no failure to comply with s 107H(3) and no breach of natural justice.

Was Mr McDonnell an eligible offender?

[63] Part 1A of the Parole Act provides for the imposition of ESOs on offenders who are “eligible offenders” as defined in s 107C. There is no doubt that Mr McDonnell meets the criteria for eligibility set out in s 107C(1). In his case, the legislation applies retrospectively, because it was not in force when he was originally sentenced for his offences. The retrospective application of the legislation is specifically referred to in s 107C(2), which provides:

To avoid doubt, and to confirm the retrospective application of this provision, despite any enactment or rule of law, an offender may be an eligible offender (including a transitional eligible offender as defined in s 107Y) even if he or she committed a relevant offence, was most recently convicted, or became subject to release conditions or detention conditions, before this Part came into force.

[64] In Belcher No 1 at [53] – [56], this Court rejected an argument made on behalf of Mr Belcher that Part 1A should be read down so that it applies only where the offender consents to it being applied retrospectively. Mr Ellis made a variation of that argument in this case. He said that s 107C(2) says that an offender “may” be an eligible offender notwithstanding retrospectivity. He said that this left open the possibility that the offender may not be an eligible offender. In order to make sense of the provision, he said, it was necessary to apply the logic of s 4 of the Criminal Justice Act 1985, which prohibited a court from imposing a sentence on an offender that it could not have imposed at the time of the offence, except with the offender’s consent. In this case Mr McDonnell did not consent to the imposition of an ESO. He said that the meaning of s 4 of the Criminal Justice Act should apply, notwithstanding its repeal, because the Supreme Court in R v Mist [2006] 3 NZLR 145 found that s 6 of the Sentencing Act (which mirrors s 25(g) of the Bill of Rights) should be interpreted as having the same meaning as s 4 of the Criminal Justice Act.
[65] Arguments based on those provisions were rejected in Belcher No 1 by a Full Court of this Court, and the application for leave to appeal against that decision to the Supreme Court was dismissed. In those circumstances, there is no basis for us to revisit this aspect of the Belcher No 1 decision. As the Court said in Belcher No 1 at [56], the reality is that there can be absolutely no room for doubt that the intention of the Legislature in enacting Part 1A was that it should apply retrospectively.
[66] Mr McDonnell meets the requirements to be an eligible offender and the Judge was right to treat him as such.

Jurisdiction: conclusion

[67] The arguments in support of the contention that there was no jurisdiction to make an ESO fail.
[68] We now turn to the arguments on the merits.

What is the standard of proof?

[69] Section 107I(2) gives the Court power to make an ESO if it is “satisfied... that the offender is likely to commit any... relevant offences... on ceasing to be an eligible offender”.
[70] This raises two separate but related issues:

(a) To what degree must the Court be “satisfied”?

(b) How “likely” must it be that the offender may re-offend?

(a) “Satisfied”

[71] In the High Court, Baragwanath J held at [33] that “satisfied” means that the Crown must “establish beyond reasonable doubt that, given its actual and potential constraints on the liberty of the subject, an ESO is proportionate to the risk presented”. Thus, in this context, satisfied was interpreted as meaning “beyond reasonable doubt”.
[72] Baragwanath J’s approach differs from that taken by a full court of the High Court in Chief Executive of the Department of Corrections v McIntosh HC CHCH CRI-2004-409-162 8 December 2004 at [20] – [21]. In that case John Hansen and Panckhurst JJ adopted the meaning given to “satisfied” in the context of preventive detention, as set out in R v Leitch [1998] 1 NZLR 420 at 428 (CA):

The need to be “satisfied” calls for the exercise of judgment by the sentencing Court. It is inapt to import notions of the burden of proof and of setting a particular standard, eg beyond reasonable doubt. As this Court said in R v White (David) [1988] 1 NZLR 264 at p 268 with reference to s 75(2) [of the Criminal Justice Act 1985], “The phrase ‘is satisfied’ means simply ‘makes up its mind’ and is indicative of a state where the Court on the evidence comes to a judicial decision. There is no need or justification for adding any adverbial qualification...”.

[73] Mr Bott supported the High Court Judge’s decision on this aspect of the case. He said the standard applying to preventive detention cases should not be applied in the ESO context because an ESO is retrospective and therefore should be imposed only when the Court has a high degree of assurance that it is necessary.
[74] Ms Edwards supported the adoption of the Leitch formulation, as the full Court did in McIntosh. She said this Court in R v D at [30] had pointed out that Parliament had used the term “satisfied” in s 87(2)(c) of the Sentencing Act (which replaced the provision in issue in Leitch) and must be taken to have approved Leitch in that regard. She said the use of the same term in s 107I(2), given the similarities between ESOs and preventive detention, could be seen as indicating Parliament’s intention that the Leitch approach should be applied in ESO cases.
[75] We accept Ms Edwards’ submission. The fact that ESOs are a penalty and the proceedings are therefore criminal does not require the criminal standard of proof that applies in criminal trials to be applied. If it did, Leitch would obviously be wrong. We conclude that “satisfied” must be given the same meaning as it was given in Leitch and R v D. We agree with the full Court’s decision in McIntosh in that regard.

(b) “Likely”

[76] The second term, likely, was interpreted by this Court in Belcher No 1. The Court held at [11] that “likely” refers to a risk of relevant offending that is “both real and ongoing and one that cannot sensibly be ignored having regard to the nature and gravity of the likely re-offending”.
[77] Mr Bott submitted that the law of the State of Victoria provided a model for the correct interpretation of “likely”. In Victoria, the power to make extended supervision orders (of up to 15 years in duration) is found in the Serious Sex Offenders Monitoring Act 2005.
[78] The Victorian Court of Appeal recently considered the statutory conditions precedent to the imposition of an extended supervision order, issuing its reasons on 18 December 2008: RJE v Secretary to the Department of Justice [2008] VSCA 265. That decision considered the meaning of s 11 of the Victorian Act, which relevantly provided:

11. When may a court make an extended supervision order?

(1) A court may only make an extended supervision order in respect of an offender if it is satisfied, to a high degree of probability, that the offender is likely to commit a relevant offence if released in the community on completion of the service of any custodial sentence that he or she is serving, or was serving at the time at which the application was made, and not made subject to an extended supervision order.

(2) The Secretary has the onus of proving the existence of the likelihood referred to in subsection (1).

...

[79] In particular, the Court considered: (1) the meaning of the requirement that the court be “satisfied, to a high degree of probability”; and (2) the meaning of the phrase “likely to commit a relevant offence”. In relation to the first point, the Court considered that this related to the requisite standard of proof. It concluded that the standard of proof imposed was sui generis, but appeared inclined (at [25]) to the view that it was in excess of the balance of probabilities and required the judge “to feel a high degree of satisfaction”.
[80] In relation to the second point, the Court reached the conclusion that the Act required proof that the offender was more likely than not to commit a relevant offence. It did so by applying ordinary principles of statutory construction, without resort to the Charter of Human Rights and Responsibilities Act 2006 (Vic). In reaching this conclusion, the Court departed from its decision in TSL v Secretary to the Department of Justice (2006) 14 VR 109, where it held that likely meant “a high degree of probability” but not necessarily more than 50 per cent. The Court considered that the TSL decision was “clearly wrong”.
[81] Mr Bott urged us to follow the RJE decision and to overrule Belcher No 1 on this point.
[82] Following the release of the RJE decision, the Serious Sex Offenders Monitoring Act was amended. New subs (2A) and (2B) were added to s 11, providing that:

(2A) For the purposes of subsection (1), an offender is likely to commit a relevant offence if there is a risk of the offender committing a relevant offence and that risk is both real and ongoing and cannot sensibly be ignored having regard to the nature and gravity of the possible offending.

(2B) For the avoidance of doubt, subsection (1) permits a determination that an offender is likely to commit a relevant offence on the basis of a lower threshold than a threshold of more likely than not.

[83] Clearly, s 11(2A) and (2B) are a response to the RJE decision. That is apparent from the remarks made by the Minister for Corrections at the amendment’s second reading: (3 February 2009) Victorian Parliamentary Debates (Legislative Assembly) at 34.
[84] The Victorian amendment explicitly adopted (in subs (2A)) the definition of “likely” set out by this Court in Belcher No 1. The Minister, in his statement on the compatibility of the Bill with the Victorian Charter ((3 February 2009) Victorian Parliamentary Debates (Legislative Assembly) at 32, said that the Belcher No 1 formulation had given the New Zealand courts:

... the flexibility to make extended supervision orders in circumstances where the offender is clinically assessed at a moderate level of risk, but the seriousness of harm that might be caused if sexual offending occurred cannot be ignored....

[85] We do not consider the Victorian developments provide any reason for this Court to revisit Belcher No 1. We reject Mr Bott’s submission that we should do so.

Was the nine year term of the ESO justified?

The High Court decision

[86] After holding that it had been proved beyond reasonable doubt that the risk posed by the appellant required an ESO, Baragwanath J concluded (at [132]) that the term of the order should be at or near the ten year maximum provided by s 107I(4). It is clear the Judge considered the extensive reasoning for the decision to make an ESO justified that term, because he did not separately identify his reasons for his conclusion about the appropriate term. The Judge reduced the otherwise appropriate ten year term to nine years to take account of the fact that the process had taken over a year to complete. The Judge also noted that the appellant would be approaching 60 years old by the time an ESO expired, and ageing tends to lower the risk of sexual offending.
[87] Baragwanath J’s decision in respect of the ESO application largely depended on the weight to be given to the expert evidence of Dr Wilson and Dr Barrett. This, in turn, required an assessment of the actuarial tools underlying Dr Wilson’s assessment report and the extent to which those tools could reasonably be used to predict the risk posed by Mr McDonnell.
[88] Dr Wilson relied on the following actuarial instruments:

(a) Psychopathy Checklist: Screening Version (PCL:SV);

(b) Violence Risk Scale: Sex Offender Version (VRS:SO) – Deviancy Sub Scale;

(c) Risk of Conviction x Risk of Imprisonment (RoC*Roi);

(d) Static-99;

(e) Automated Sexual Recidivism Scale (ASRS) (formerly known as Static Automated Scoring (Static-AS));

(f) STABLE-2000.

[89] Each instrument was discussed at some length by the Judge. We summarise that discussion below:

(a) The PCL:SV provides a reliable and quick way of assessing the presence of psychopathy in an individual, although it does tend towards over-prediction. Psychopathy is a robust predictor of future violent reoffending. Mr McDonnell’s score of 21 out of a possible 24 illustrated the presence of a high degree of psychopathic behaviours (both interpersonal and affective deficits as well as antisocial behaviour and lifestyle) and placed him within a group of offenders with a 65 per cent chance of reimprisonment within five years (although not necessarily for sexual offending). Dr Barrett pointed out that Mr McDonnell’s true score may be as low as 19, but acknowledged that this would not reduce the statistical risk he poses.

(b) The VRS:SO measures sexual deviance, being arousal related to either sexual activity with children or coercive sex with non-consenting adults. Mr McDonnell was assessed as having a very high score on this measure, indicating untreated sexual deviancy, which, in combination with psychopathic behaviour, is strongly predictive of sexual recidivism. The Judge treated Mr McDonnell’s PCL:SV and VRS:SO results as of “particular relevance” (at [119]).
(c) The RoC*RoI assessed Mr McDonnell as being of a “moderate risk of serious recidivism within five years of release”, based on static factors associated with his criminal history. This measure is not, however, concerned specifically with predicting sexual recidivism. Its main value, according to counsel for the Chief Executive, was that it showed Mr McDonnell’s antisocial orientation, which is generally treated as associated with sexual recidivism. However, the Judge accepted Dr Barrett’s evidence that the risk of recidivism attributable to this factor alone is only 1.5 per cent. For these reasons, the Judge treated the RoC*RoI as virtually irrelevant.
(d) The ASRS is a shortened version of the Static-99 assessment. It is based on static risk factors, and is used to screen sexual offenders under consideration for release. Mr McDonnell’s score indicated an 11 per cent risk of him sexually reoffending over a ten year period.
(e) According to the full Static-99 assessment (which considers three additional variables to the ASRS), Mr McDonnell’s cumulative risk of sexual recidivism was 33 per cent at five years and 38 per cent at ten years post-release. This result was, however, at least partly based on the inclusion of sexual offending for which Mr McDonnell was charged but not convicted. The Judge held that such material did not have a solid enough evidential base and should have been excluded from the analysis. For this reason, the Judge excluded the appellant’s Static-99 score from his assessment.
(f) The Stable-2000 assessment is based on dynamic risk factors and is used to distinguish between sexual recidivists and non-recidivists. Mr McDonnell was assessed as of high risk (scoring 10 out of a possible 12), which equates to an 18.7 per cent risk of sexual recidivism over a three year follow up period. In particular, the Judge noted that Mr McDonnell exhibited: an absence of positive social influences; intimacy deficits; a lack of sexual self-regulation; attitudes supportive of sexual assault; limited co-operation with supervision; and a lack of general self-regulation.
(g) The Judge finally considered Dr Wilson’s structured professional judgment, an assessment based on a combination of empirical data and his own experience in which the results of all the various actuarial instruments and clinically relevant information are brought together. In Dr Wilson’s professional opinion, Mr McDonnell posed a high risk of committing a further relevant sexual offence while in the community. The Judge noted that the legitimacy and value of this approach to risk assessments is a matter of dispute among specialists, but concluded that Dr Wilson’s expertise was entitled to be given weight.
[90] Baragwanath J then turned to consider whether this evidence established beyond reasonable doubt the existence of a risk that warranted the imposition of an ESO. He began by making a number of observations on Mr McDonnell, describing him (at [125]) as an intelligent, egocentric psychopath with no real appreciation of either the harm done to his victims or the significance of his conduct. This lack of self-awareness, in the Judge’s opinion, showed that the appellant has not learned (and is unlikely to learn) that his attitude towards young people (viewing them as “legitimate game”) is inappropriate and needs to be altered. The Judge also noted that the offences for which Mr McDonnell was convicted occurred despite him having earlier been charged (but not convicted) in relation to similar behaviour.
[91] The Judge then considered the results given by the various actuarial instruments. In particular, he noted (at [126]) that the appellant’s scores on the PCL:SV and VRS:SO together provided cause for grave concern. The lower predictions of risk given by the ASRS and RoC*RoI were not regarded as inconsistent with this assessment. Dr Wilson’s structured clinical judgment on the appellant’s overall risk was then considered. The Judge described (at [130]) the reasons given by Dr Wilson for his conclusion that the appellant was at a high risk of sexually reoffending against boys as “intellectually compelling”. He concluded that an ESO was required and that the gravity of the risk posed justified a term at or near the ten year maximum.

Submissions for Mr McDonnell

[92] Mr Bott criticised the judgment as not clearly showing how the Judge came to be satisfied beyond reasonable doubt that an ESO near the maximum term was necessary and that probability statements would have assisted in following how the Judge reached this conclusion.
[93] Mr Bott argued that a term of no greater than five years was appropriate. He said that the risk posed by Mr McDonnell was not especially great, particularly because he does not have a history of sexual offending (other than the qualifying offences) and has not reoffended following his release in 2005. Further, given the element of retrospective punishment inherent in the imposition of an ESO, Mr Bott urged caution when setting the term of the order.
[94] Mr Bott then went through the probabilities given by the various actuarial tools of Mr McDonnell reoffending. He submitted that none of them established a sexual recidivism risk of greater than 50 per cent. (The greatest probability – the 65 per cent chance of reimprisonment within a five year period (see [89](c) above) given by the RoC*RoI – Mr Bott pointed out, related to general recidivism.) In effect, he argued that the Judge was wrong to consider that the lower probabilities given by the other tools, supplemented by Dr Wilson’s structured clinical judgment, justified the imposition of an ESO for a nine year term.

Section 107I(5)

[95] The test for determining the term of an ESO is set out in s 107I(5), which provides:

(5) The term of the order must be the minimum period required for the purposes of the safety of the community in light of—

(a) the level of risk posed by the offender; and

(b) the seriousness of the harm that might be caused to victims; and

[96] The criteria set out in s 107I(5) are not identical to those that must be considered by the judge when deciding whether or not to impose an ESO: ss 107F(2) and 107I(2). A sentencing judge should therefore give separate consideration to the appropriate term of an ESO: Peta at [61].
[97] When assessing the propriety of an ESO’s term, this Court will require the appellant to meet the same standard as that demanded in ordinary sentence appeals, namely establishing that the length of the term imposed is manifestly excessive (or manifestly inadequate), provided the term must be the minimum required for the purposes of the safety of the community. This is consistent with the characterisation of ESO appeals in s 107R(2): see [1] above. We see that as a more appropriate analogy than the analogy with bail appeals adopted in R v Poutawa [2007] NZCA 206 at [10] – [14].
[98] The Judge gave only brief reasons in relation to the term of the order and Mr Bott concentrated his submissions on the merits on the term of the ESO rather than on whether it should have been made at all. In light of those factors, we will consider the submissions in some detail. We should say, however, that we do not expect such detail in ESO decisions (or appeals from such decisions) where the reasons for making the ESO have engaged with the statutory criteria and the factors relating to the term of the ESO raise broadly similar issues.
[99] We now turn to the three criteria in s 107(I)(5).

(a) Level of risk

[100] The first factor is the level of risk posed by Mr McDonnell. This requires an assessment of his likelihood of committing further sexual offences (falling within the meaning of “relevant offence” in s 107B) against children or young persons. Each of the actuarial tools discussed above supplies different information relevant to this assessment.
[101] The ASRS and Static-99 assessments are of particular relevance, given they assess a person’s risk of sexually reoffending. The former is used in order to screen sexual offenders, and is recognised as providing an initial, rather conservative, indication of risk. It places the appellant in a group of offenders with a known probability of sexually reoffending of 11 per cent over a ten year period.
[102] Static-99 includes three variables that are omitted for logistical reasons from the ASRS. It places the appellant in the medium-high risk group, which has a 33 per cent probability of reoffending over a five year period, rising to 38 per cent (cumulatively) over ten years. The instrument requires the inclusion of charges for sexual offending that were not proceeded with or that did not result in convictions (although they are given less weight than actual convictions). Dr Wilson noted that there was evidence in probation reports that the appellant was, prior to the index offences, charged with (but not convicted for) sexual offending. This information increased the appellant’s assessed level of risk.
[103] Baragwanath J held (at [81]) that a court should not consider a Static-99 score that takes into account the presence of prior charges of, as opposed to convictions for, sexual offending. The Judged noted that an expert opinion must be confined to facts proved, and that the presumption of innocence should apply to allegations of past criminal conduct. We respectfully disagree: prior charges are statistically significant and their use is required by the Static-99. We do not see the presumption of innocence as bearing on the statistical propriety of including given variables in an actuarial risk assessment, particularly in this case where the instrument gives reduced weight to charges compared with convictions. There was an adequate evidential foundation for the existence of prior charges against the appellant and, in the absence of any challenge to this, we see nothing inappropriate in them contributing to his overall Static-99 score.
[104] Supplementing these figures is the appellant’s score on the Stable-2000, which puts him at an 18.7 per cent risk of sexual recidivism over a three year follow up period. Further, and importantly, the combined presence of sexual deviancy and psychopathy in the appellant significantly increase his risk of sexually reoffending.
[105] Based on these actuarial tools and his own clinical opinion, Dr Wilson assessed the appellant as at a high risk of sexual recidivism. We agree with this assessment, essentially for the same reasons as Baragwanath J, but also having regard to the appellant’s Static-99 score.

(b) Seriousness of harm

[106] The second factor to be considered is the seriousness of the harm that might be caused to victims. In his evidence, Dr Wilson explained that the age and gender of an offender’s victims remain relatively stable. Accordingly, he said that if the appellant were to reoffend, his victims would likely be post-pubescent boys under the age of 16 years. Based on the appellant’s past offending, Dr Wilson suggested, and we agree, that any reoffending is likely to be of a serious nature, involving indecent assault or rape: see also Peta at [10] and [45]. Such offending would likely cause serious harm to the victims, in the same way that the appellant’s past offences seriously affected the lives of each of the three complainants.

(c) Duration of risk

[107] Finally, the likely duration of the risk posed by the appellant must be considered. This is the most challenging of the required factors to assess. It requires a prediction of the way in which the risk currently presented by the appellant will behave over the next ten years.
[108] In his report, Dr Wilson noted that “[r]esearch indicates that the risk of reoffending for individuals with Mr McDonnell’s assessed risk level continues over an extended period” and that this risk level “is unlikely to reduce over a short to medium time period”. These statements are perhaps unsurprising, given many of the actuarial risk assessment tools used to assess the appellant rely solely on static, historical variables.
[109] Of most relevance to assessing the duration (and stability) of the risk posed by the appellant is the presence of dynamic risk factors. In particular, Dr Wilson emphasised both the absence of any indication that the appellant has developed internal management of his risk and the lack of insight he has shown into his offending. These two risk factors are specifically addressed in validated treatment programmes for sex offenders. The appellant has not, however, sought to receive any form of treatment, instead choosing to deny his offending.
[110] Dr Wilson also emphasised the presence of both psychopathy and sexual deviancy in the appellant. He said that these two factors, in conjunction, are predictive of sexual recidivism: their interaction has a multiplicative, rather than additive, effect in heightening risk (see also the discussion in Peta at [39] – [43]). Without management, these factors are not believed to change over time. Indeed, people assessed as having psychopathic behaviours are generally resistant to treatment. Nothing in the appellant’s recent history suggests that he is an exception to this general proposition: there is no evidence that he has sought or undertaken treatment either in prison or after his release. Also concerning is the appellant’s express desire to return to practising hypnosis.
[111] Accordingly, the appellant’s psychopathy and untreated sexual deviancy, combined with his failure to take any steps to ameliorate these conditions, give cause for concern. There is no reason to conclude that the risk currently posed by the appellant will reduce over the short to medium term.

Overall assessment

[112] Each of the three statutory factors indicate cause for concern about the risk posed by the appellant. The minimum term appropriate for the purposes of the safety of the community depends on the appellant’s behaviour over the short to medium term. The appellant is a sexually deviant psychopath, who has been assessed by a number of statistical tools and clinicians as falling within a medium to high risk group of sexual offenders. Any reoffending is likely to be of a serious sexual nature and risks causing significant harm to vulnerable individuals. No evidence to the contrary has been called. Moreover, the appellant has consistently denied his offending, has not participated in any form of treatment and has been generally uncooperative with efforts to manage his risk. There is a complete absence of any evidence to suggest that these factors are likely to change.
[113] We therefore conclude that a term of nine years was properly within range as being the minimum period appropriate for the safety of the community.

Should a declaration of inconsistency be made?

[114] In Belcher No 1, the appellant alleged that Part 1A of the Parole Act was inconsistent with the Bill of Rights, and sought a declaration to this effect.
[115] The Court expressed the view that Part 1A was inconsistent with the Bill of Rights, primarily because the Crown had not sought to establish that the legislation was a justified limitation in terms of s 5. Because the point had not been the subject of full argument, the Court decided to reserve the question whether it should make a declaration of inconsistency. It envisaged holding a full hearing on this issue at a later date. In particular, argument was invited (at [59]) on whether courts have jurisdiction to make declarations of inconsistency, generally, and in criminal proceedings, specifically.
[116] Following that decision, but before the further hearing in Belcher, the Supreme Court issued its judgment in Taunoa v Attorney-General [2006] NZSC 95, in which the appellants (and cross-respondents) sought, inter alia, a declaration that the Prisoners’ and Victims’ Claims Act 2005 is inconsistent with the Bill of Rights.
[117] The Supreme Court made the following comments:

[6] The alternative relief sought is a declaration of inconsistency. It is plain that this is beyond the jurisdiction of the Court in the circumstances of the present cross-appeal. The subject matter of the application has never been before the High Court or the Court of Appeal. The appellants and cross-respondent are therefore seeking to bring a new case directly to this Court. The Supreme Court is established by statute and has no original jurisdiction. Under s 25(1) of the Supreme Court Act 2003, to which Mr Ellis directed attention, the Court has wide general powers on an appeal “in a proceeding that has been heard in a New Zealand court” to make “any order, or grant any relief, that could have been made or granted by that court”, that is, by the Court appealed from. No relief in the form now sought could have been granted in the Court of Appeal for the obvious reason that no issue relating to the Act was before it. This Court also has under s 25(1) all the powers of the Court of Appeal even if the proceeding has not been heard in that Court. But, again, as the issue was never before the High Court, the Court of Appeal would have lacked any power to hear it, for it too has no originating jurisdiction.

[7] Mr Ellis’s broad submission was that the Act was, in his words, unjustified in a free and democratic society, and seriously offensive to international human rights law and to the rule of law itself. He therefore urged us to assume an extraordinary jurisdiction to consider and determine his alternative application, as we understood him, effectively out of necessity. He appeared to be contending that otherwise his clients would never be able to endeavour to obtain relief against the operation of the Act.

[8] We do not agree. If some relief of the character sought might be available from the courts, on which we express no opinion, the appropriate place for the proceeding to be commenced is in the High Court which, unlike the appeal courts, has a general jurisdiction.

[118] The Supreme Court’s judgment was issued on 13 November 2006. On 13 March 2007 this Court held the further hearing in relation to Mr Belcher’s appeal: [2007] NZCA 174 (Belcher No 2). Because of the Supreme Court’s judgment in Taunoa, that hearing was restricted to whether there was, in Mr Belcher’s case, jurisdiction to grant a declaration of inconsistency. The Court considered that the comments made in Taunoa were applicable to Mr Belcher’s situation. It held that the effect of Taunoa was that the Court could not engage with an application to grant a declaration of inconsistency unless such an application had first been made to the High Court: to so do would amount to the Court exercising an originating jurisdiction that it did not possess (at [12]).
[119] The Court (with the exception of Hammond J, who reserved his position on these matters) went on to consider generally the availability of declarations of inconsistency in criminal proceedings. It observed at [13]:

...it is unheard of for the courts hearing criminal cases to grant what is truly civil relief, for instance Baigent damages. The reasons for this are obvious. Criminal procedures, as laid down by statute, are appropriate for the determination of criminal proceedings but not for the granting of civil relief. This is so in terms of who the parties are, who determines questions of fact (in serious cases juries and not judges), pre-trial procedures (which do not include mutual discovery) and appeal rights (which are usually narrowly expressed and would not provide for appeal rights in relations to the granting or refusing of declarations of inconsistency). As well, there is the ever present risk of the criminal process being diverted into collateral issues.

[120] The Court also pointed out at [14] that District Court judges do not have jurisdiction to grant declarations of inconsistency in civil proceedings. It concluded its discussion in the following way:

[15] This is not to say that the consistency or otherwise of legislation with the New Zealand Bill of Rights Act is irrelevant in criminal cases. Indeed the contrary is obviously the case, as exemplified by R v Hansen [2007] NZSC 7. It may well be necessary for courts exercising criminal jurisdiction at all levels of the hierarchy [of] courts either to make a finding of inconsistency (as in Hansen) or perhaps to assume inconsistency where the Crown has chosen not to develop an evidential basis for a s 5 justification (as in the present case).

[16] So if a declaration of inconsistency is available (on which we express no opinion), it should be sought in a civil proceeding commenced in the High Court, a course which we think is consistent with [8] of the Taunoa judgment.

[121] Mr Belcher’s application for a declaration of inconsistency was therefore dismissed for want of jurisdiction. He then applied for leave to appeal to the Supreme Court against both decisions of this Court.
[122] The Supreme Court dismissed Mr Belcher’s application for leave to appeal (which was supported by the Chief Executive): [2007] NZSC 54 (Belcher SC). It did, however, comment on the availability of declarations of inconsistency and in doing so found this Court’s interpretation of Taunoa to be erroneous:

[6] As to the decision to decline a formal declaration of inconsistency: assuming, without deciding, that a declaration may be available in a criminal proceeding, we consider that it was entirely appropriate for the Court of Appeal to leave the matter in essentially the same way as it was subsequently left by the majority of this Court in R v Hansen where the inconsistency was described in the reasons for judgment but no declaration was made. It is also of some moment in the present case that no issue concerning s 5 was required to be determined in the necessary course of interpreting the legislation and resolving questions between the parties. A response in the form of a declaration was quite unnecessary.

[7] We should, however, before leaving this matter correct the Court of Appeal's erroneous view that, if a declaration of inconsistency had been otherwise available and appropriate, there would nevertheless have been in this case the same jurisdictional difficulty as existed in Taunoa v Attorney-General. The difficulty in that case arose from an unusual situation in which the asserted cause of action in respect of which relief was being sought in this Court could not have existed when the proceeding was before the High Court since the Act of Parliament in respect of which relief was sought had not been enacted until the Taunoa proceeding was already before the Court of Appeal.

This Court said:

No relief in the form now sought could have been granted in the Court of Appeal for the obvious reason that no issue relating to the Act was before it. This Court also has under s 25(1) all the powers of the Court of Appeal even if the proceeding has not been heard in that Court. But, again, as the issue was never before the High Court, the Court of Appeal would have lacked any power to hear it, for it too has no originating jurisdiction.

[8] In the present proceeding the issue, namely the application of the relevant aspect of the Parole Act to Mr Belcher's circumstances, was before the High Court. Relief in the form of a declaration could, if otherwise available, have been granted by that Court and accordingly the Court of Appeal was not precluded from considering that question. It would not thereby have been exercising an originating jurisdiction. In this respect the Court of Appeal fell into error but, for the reasons given above, that does not require the granting of leave to appeal to this Court.

[123] In summary, the position after the Belcher litigation is:

(a) No decision has yet been made by the Supreme Court on whether declarations of inconsistency are available in criminal proceedings. However, this Court has indicated (in an obiter comment) that they are not; a separate civil proceeding is required: see [119] above. The Supreme Court did not give leave to appeal from that decision and did not contradict that comment (though it did contradict another aspect of the decision).

(b) The preferred approach to identifying inconsistencies is to do so in the reasons for judgment, without issuing a formal declaration.

(c) A declaration will be unnecessary where s 5 of the Bill of Rights does not need to be considered in order to determine the questions at issue between the parties.
(d) There is no jurisdictional bar to the Court of Appeal granting a declaration of inconsistency where such a declaration was not first sought in the High Court but was technically available.
[124] Notwithstanding this Court’s decision in Belcher No 2, no civil proceeding seeking a declaration was commenced in the High Court. Counsel for Mr McDonnell had indicated well before the case came to a hearing that he intended to seek a declaration. On 22 November 2006, Cooper J issued a minute in which he noted that intention. He also recorded that counsel were agreed that it would be sensible and efficient if the application were dealt with at the same time as the ESO application was heard. He directed that the civil proceeding seeking a declaration should be filed as soon as convenient.
[125] No application was ever filed. Counsel for the Chief Executive followed up matters on numerous occasions, without success. On 15 May 2007, the day before the High Court hearing was to commence, Mr Bott filed and served further submissions in which a “Hansen declaration” was sought.
[126] At the High Court hearing, Baragwanath J noted the decisions of this Court in Belcher No 2 requiring a separate civil proceeding and in Belcher No 1 that ESO applications were criminal proceedings. He said this meant he was bound not to entertain the originally intimated submission by Mr Bott for a declaration of inconsistency, and that Mr Bott had withdrawn that submission.
[127] In his judgment, Baragwanath J dealt with the matter as follows:

[10] While accepting that this Court must apply the statutory ESO regime, Mr Bott sought a declaration that the regime is inconsistent with the New Zealand Bill of Rights Act 1990. Since the oral hearing the Supreme Court in [Belcher SC]... has determined the point, which requires no further comment.

[128] The conduct of this aspect of the case has not been any better in this Court. The notice of appeal filed on 13 August 2008 recorded as one of the grounds of appeal:

The High Court made an error of law in declining to make a declaration of inconsistency between the ESO legislation and the New Zealand Bill of Rights Act 1990 as per Hansen v R, Supreme Court of New Zealand, 20 February 2007.

[129] The first indication which the Court or the Chief Executive had of the basis for the argument that a declaration of inconsistency should have been made, or indeed that the remedy sought was a declaration rather than a statement of inconsistency of the kind made in Hansen, came when submissions were filed on behalf of the appellant on 20 February 2009. As a result, all counsel were agreed that, if we were to deal with the application for a declaration, it would be necessary to adjourn the hearing to allow for a response to be prepared. It was also clear that it would not be appropriate for the Chief Executive to be left to respond to such an application: service on the Attorney-General may be the appropriate way of resolving that issue.
[130] More fundamentally, however, this Court is now placed in substantially the same position as the Court found itself in the Belcher litigation. It is being asked to deal with the issues arising from the declaration application without the benefit of a High Court decision. Counsel for the Chief Executive indicated that the Crown would wish to adduce evidence supporting its contention that the ESO regime can be demonstrably justified in a free and democratic society (s 5 of the Bill of Rights), and, if cross-examination were required, that would have to take place before this Court.
[131] This Court’s decision in Belcher No 2 clearly stated that declarations of inconsistency could not be issued in criminal proceedings, and a separate civil application was required. The Chief Executive would not be the appropriate defendant in such proceedings. Efforts were made to facilitate a civil application being made in the High Court but this did not occur. Although the Supreme Court’s decision declining leave to appeal in Belcher SC includes the statement “assuming, without deciding, that a declaration may be available in a criminal proceeding”, the Supreme Court has not overruled the decision of the full Court in Belcher No 2. Indeed, the Supreme Court expressed the view that it was unnecessary to make a declaration in the Belcher case and that the way the matter was left in this Court was “entirely appropriate”. No civil application was made in this case. In those circumstances we reiterate the position set out in Belcher No 2 and decline the application for a declaration of inconsistency.

Result

[132] An extension of time for appealing is granted, but the appeal is dismissed. The application for a declaration of inconsistency is dismissed.

Solicitors:
Crown Law Office, Wellington for Respondent

NZLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.nzlii.org/nz/cases/NZCA/2009/352.html