Plaintiff v Defendant

3 Because of late developments in the proceedings, which will be discussed shortly, it was not possible for me to publish detailed written reasons before the CDO expired. On 19 December 2023, I made orders replacing the CDO with an ESO. At the time of making the orders I gave extensive oral reasons for decision, and indicated that I would publish full written reasons at a later date. These are those written reasons.

4 In order to understand the issues in dispute, it is convenient to begin with a discussion of the background to these proceedings.

5 Between mid-2004 and late 2005, Mr Benbrika was the leader of a terrorist organisation, whose members were preparing to commit a violent terrorist act in order to advance Islamic ideology through the pursuit of violent jihad.

6 In September 2008, Mr Benbrika was found guilty by a jury of intentionally being a member, and directing the activities, of a terrorist organisation. In February 2009, Mr Benbrika received a total effective sentence of 15 years' imprisonment, with a non-parole period of 12 years.[1]

7 Mr Benbrika was not granted parole. His sentence expired on 5 November 2020. From then until 19 December 2023, he remained in prison in the following circumstances.

8 CDOs and ESOs are both "post-sentence orders" that may be sought on application by the "AFP Minister", who is defined in s 100.1 of the Code as the minister administering the Australian Federal Police Act 1979 (Cth). At most times relevant to this case, the Minister for Home Affairs was the AFP Minister. Since 1 July 2022, the A-G has been the AFP Minister.

9 Shortly before Mr Benbrika's sentence expired, the Minister for Home Affairs applied to this court for a CDO.

10 After making a couple of interim CDOs, on 24 December 2020 Tinney J made a CDO for a period of 3 years, ending on 23 December 2023.[2] The effect of the interim and final CDOs was to commit Mr Benbrika to detention in a prison for the period of the respective orders.

11 Tinney J made the various CDOs because he was satisfied to a high degree of probability that Mr Benbrika posed an unacceptable risk of committing a relevant offence, and there was no less restrictive measure available that would be effective to prevent that risk.

12 In order to come to those conclusions, Tinney J was required to consider competing expert evidence regarding Mr Benbrika's risk of committing a serious offence if he was released into the community. The Minister for Home Affairs' two expert witnesses both used an assessment tool or instrument known as the Violent Extremism Risk Assessment - Version 2 Revised ("VERA-2R"). The Minister for Home Affairs was highly critical of Mr Benbrika's expert for criticising VERA-2R's validity. On the basis of the evidence before him, Tinney J accepted the validity of VERA-2R and the assessments made using it.

13 On 9 November 2021, the Court of Appeal dismissed Mr Benbrika's appeal against the making of the CDO.[3]

14 The AFP Minister is required to apply for a review of a CDO every 12 months.[4]

15 The first review hearing was listed to begin on 8 December 2021. In the lead up to the listed date, it became apparent that the Code was about to be amended, to introduce ESOs as a less restrictive type of post-sentence order than CDOs. Accordingly, I formally began hearing the review in December 2021, to ensure that it began within the required period. By agreement, I then adjourned it off part-heard, to enable the parties to file further material addressing the possibility of an ESO being ordered instead of the CDO. Substantive argument took place in January and February 2022.

16 The underlying validity of VERA-2R was not disputed before me, because its validity had already been upheld by Tinney J and the Court of Appeal. Instead, Mr Benbrika's lawyers challenged the manner in which VERA-2R had been applied by the Minister for Home Affairs' experts in assessing Mr Benbrika's risk of offending. On 21 April 2022, I affirmed the CDO on its current terms.[5]

17 At a directions hearing on 12 May 2022, the Minister for Home Affairs foreshadowed that the next proceeding would either be an application for annual review of the CDO, or an application for a new CDO (in which case, a review would not be necessary). The Minister agreed to inform the court and Mr Benbrika by 1 November 2022 whether the application would be for a review or a new CDO. Provisional hearing dates in June and September 2023 were set, in order to cover either possibility.

18 On 28 October 2022, the A-G notified the court that he would be making an application for a new post-sentence order, rather than a review of the current CDO. At a mention on 15 November 2022, the A-G said he had not decided whether to apply for an ESO or a further CDO; in either event, he anticipated filing his application by 21 April 2023.

19 Shortly after that November mention, Mr Benbrika discovered that there had been some serious non-disclosure of important evidence by the Minister for Home Affairs in the earlier proceedings.

20 In June 2020, Dr Emily Corner and Dr Helen Taylor from the Australian National University produced a 150-page report headed "Testing the reliability, validity and equity of terrorism risk assessment instruments" ("the Corner report"). The Corner report had been commissioned by the Department of Home Affairs ("Home Affairs").

21 The stated purpose of the Corner report was to undertake a holistic and impartial analysis of both VERA-2R and a risk assessment instrument called Radar, to demonstrate the extent to which they accurately classified offenders, or overestimated or underestimated the risk they posed. The authors also set out to provide the most comprehensive overview of the state of the empirical knowledge of the causes of radicalisation and terrorism to date.

22 The findings of the Corner report included the following:

(a) The theoretical and empirical evidence base cited in both instruments' user manuals and supporting documentation was of poor quality, and was predominantly composed of theoretical assertions, secondary citations of literature reviews, and media articles.

(b) Neither instrument followed a true "structured professional judgment approach", as claimed by the authors; rather, they were both "SPJ-lite".

(c) Of the 1,500 variables found in the empirical evidence to be statistically associated with movements towards radicalisation and violent extremism, VERA-2R covered only 14.2% of them.

(d) VERA-2R offered poor predictive validity. Radar's superior predictive validity resulted from assessors not relying on the instrument to make decisions, and instead using their own knowledge and experience.

(e) Without a strong theoretical and empirical basis for factor inclusion, it was not reasonable to anticipate that the instruments were able to predict their specified risk with "anything other than chance".

23 Post-sentence orders, such as CDOs and ESOs, are exceptional. As a general rule, our legal system punishes people by sentencing them for crimes they have committed. Its purpose is not to detain people to prevent a crime that they may or may not commit in the future. However, parliaments have introduced preventative detention regimes for certain types of very serious offenders, including terrorism offenders. Such regimes are the exception, not the norm. Courts have repeatedly said that the fundamental liberty of a person should not be constrained to any greater extent than is strictly warranted by the particular preventative detention legislation.

24 Preventative detention legislation generally contains strict safeguards to ensure that such exceptional powers are exercised appropriately. An application for a CDO or ESO must include "a copy of any material in the possession of the applicant ... that would reasonably be regarded as supporting a finding that the order ... should not be made."[6] The High Court has noted that this requirement to disclose exculpatory material of which the AFP Minister is aware is one of the statutory safeguards in Div 105A of the Code.[7]

25 The Corner report was clearly a document which should have been disclosed in the CDO and first review proceedings. The Minister for Home Affairs did not disclose the contents or existence of the Corner report to Mr Benbrika or the court, even though the underlying validity of VERA-2R was an absolutely fundamental issue in dispute in the earlier proceedings. The A-G now concedes that the non-disclosure of the Corner report was a serious breach of the disclosure obligation imposed by the Code.

26 Later in these reasons, I will discuss the various explanations given for non-disclosure. It is sufficient for present purposes to note that the Corner report was withheld from Mr Benbrika and the court in all of the earlier proceedings.

27 Another important safeguard is the existence of the Independent National Security Legislation Monitor ("the INSLM"), whose role is to independently review the operation, effectiveness and implications of national security and counterterrorism laws, and to consider whether they contain appropriate protections for individual rights, and remain necessary and proportionate to relevant threats.

28 During 2022, the then INSLM, Grant Donaldson SC, conducted a review into the operation, effectiveness and implications of Division 105A of the Code. In the course of his review, the INSLM became aware from Dr Corner's webpage that she had completed a research project for Home Affairs, concerning the testing of risk assessment tools. He used his statutory powers[8] to require Home Affairs to produce the Corner report to him.

29 The Corner report was initially provided to the INSLM with a notation from Home Affairs that it contained "operationally sensitive information". After examining the Corner report, the INSLM correctly concluded that there was nothing in the report that could properly be described as operationally sensitive.

30 The Corner report was first publicly referred to at an INSLM hearing on 21 November 2022. The INSLM was highly critical of the Minister's failure to produce the report in the Benbrika proceedings.[9] The INSLM's strong, targeted criticisms still did not prompt either Home Affairs or the A-G's Department to produce the Corner report to Mr Benbrika, or to inform him or this court of its existence.

31 Mr Benbrika's lawyers first learned of the Corner report from media coverage of the INSLM public hearings in November 2022. A redacted copy of the Corner report was only provided to Mr Benbrika's lawyers on 6 December 2022, after they specifically requested a copy from the A-G's Department.

32 On 13 December 2022, Mr Benbrika commenced proceeding S ECI 2022 5118 ("the second review proceeding"), seeking a review of the CDO arising out of the non-disclosure of the Corner report.

33 At the first directions hearing, on 15 December 2022, I ordered the A-G to provide an unredacted copy of the Corner report to Mr Benbrika's lawyers. I also fixed the review for hearing on 1 June 2023, and made a timetable for the delivery of disclosure material, evidence and court documents well in advance of the trial date.

34 On 6 March 2023, the A-G commenced proceeding S ECI 2023 855 ("the ESO proceeding"), seeking an order that this court make an ESO in place of the CDO.

35 Thereafter, the second review proceeding and the ESO proceeding continued alongside each other, relying on the same materials, and listed for hearing together. The parties accepted that the CDO should be replaced by an ESO, on the basis of Mr Benbrika's reduced risk.

36 For many months, the parties prepared evidence and court documents addressing the question of what ESO conditions should be imposed, on the assumption that Mr Benbrika would be living with his family in the community. Shortly before the trial, that fundamental assumption changed in the following circumstances.

37 In November 2020, Mr Benbrika's Australian citizenship was cancelled by the Minister for Home Affairs.[10] Upon the cancellation of his citizenship, Mr Benbrika was taken to have been granted an "ex-citizen visa".[11]

38 On 12 May 2023, the Minister for Home Affairs determined to cancel Mr Benbrika's ex-citizen visa.[12] By virtue of s 189(1) of the Migration Act, the effect of cancelling Mr Benbrika's ex-citizen visa was that he immediately became an "unlawful non-citizen", who must be detained in immigration detention.

39 At various times since his citizenship was cancelled, including during the hearing of the first review proceeding, the Minister for Home Affairs had considered the possibility of cancelling Mr Benbrika's ex-citizen visa. However, no steps were taken to do so until a few weeks before the trial of these two proceedings was due to start. No explanation has been given to Mr Benbrika or the court as to why the cancellation decision was made at that particular time.

40 There is a wide range of places where a person in immigration detention may be held, including a detention centre, a prison or remand centre, a police station, or any other place (including a private residence) approved by the Minister for Home Affairs.

41 On 15 May 2023, a delegate of the Minister for Home Affairs decided that, at the end of his CDO, Mr Benbrika would be detained in immigration detention in a correctional setting. As a result of negotiations between Home Affairs and Corrective Services NSW, it was subsequently agreed that Mr Benbrika would be held in immigration detention in the Long Bay Correctional Centre, a prison in NSW.

(a) The A-G's own expert had assessed that Mr Benbrika's risk could be satisfactorily managed with him living in the community under a suitable ESO; and

(b) The A-G himself had issued a proceeding seeking orders that Mr Benbrika be able to live in the community under supervision,

Home Affairs determined not to approve any immigration placement that would have kept Mr Benbrika in Victoria (where his family and lawyers are located), or in any form of immigration detention less secure than a prison.

43 The last-minute decision to place Mr Benbrika in Long Bay if an ESO was ordered obviously required a substantial change to the A-G's case and the parties' preparation for trial. Further draft ESO conditions were prepared, and six further affidavits were delivered by the A-G the day before the trial was due to start.

44 In the days leading up to the trial, further disclosure material continued to be produced by the A-G. The late disclosure material included three more expert reports that were critical of VERA-2R, and which had never previously been produced to Mr Benbrika. Mr Benbrika did not want to seek an adjournment of the trial, as I would be unavailable to hear the trial before September, if it did not proceed on the scheduled dates. However, the volume and complexity of the last-minute material that Mr Benbrika and the court had to read and deal with was totally unacceptable, and does not reflect well on the Commonwealth, which is supposed to act as a "model litigant".

45 After the evidence had been completed, the parties requested, and were granted, time to prepare written submissions. The A-G continued to serve additional disclosure material even after the close of evidence, including yet another expert report that was critical of VERA-2R.

46 The fact that all of the critical expert reports apart from the Corner report were disclosed so late in the proceedings prevented the court from being able to examine the reasons for their earlier non-disclosure.

47 On 14 June 2023, the High Court heard Mr Benbrika's appeal against the cancellation of his citizenship, and reserved its decision.

48 Also on 14 June 2023, Mr Benbrika issued a judicial review proceeding in the Federal Court, challenging the validity of the decision to cancel his ex-citizen visa.

49 As at the end of the trial of these proceedings, it was unknown when the High Court's decision might be handed down, or when the Federal Court proceeding might be heard and determined. Unless and until Mr Benbrika succeeded in either of those proceedings, the evidence was that he would be held in immigration detention in Long Bay after the end of his CDO.

50 One of the matters that was hotly contested at trial was the scope of any ESO that the court might make. The A-G argued that the ESO conditions should be limited to the situation in which Mr Benbrika would be held in immigration detention in Long Bay. Mr Benbrika argued that the ESO conditions should also address the alternative situation in which he would be living in the community, if he ceased to be in immigration detention.

51 By the time of trial, the parties had provided their evidence and written submissions addressing both the community and custodial situations. Mr Benbrika cross-examined on, and made oral and written submissions addressing, both possible situations. However, the A-G chose not to make oral submissions, or file closing written submissions, addressing the conditions that would apply in the community; instead, he chose to focus on the proposed conditions in custody. The A-G expressly acknowledged that, if I ruled against him on the scope of the conditions that the court should make, I would go on to determine both custodial and community conditions on the basis of the material that had already been provided by the parties.

52 In the course of initially preparing my reasons for decision, I had come to the conclusion that I would only set conditions that would apply if Mr Benbrika was in custody. That was not because of any constitutional or statutory limitation on the breadth of the court's power to make an ESO addressing future matters. However, because of the uncertainty as to:

(a) Mr Benbrika's prospects of success in the High or Federal Court proceedings; and

I decided that I could not be satisfied as to the necessity and appropriateness of the proposed community conditions at some unknown future time. It was clear on the evidence led at trial that, upon the expiry of the CDO, Mr Benbrika would be placed in immigration detention in Long Bay. If Mr Benbrika subsequently succeeded in either the High or Federal Courts at a future date, a number of important matters (such as his risk assessment, his treatment needs, or his particular family circumstances) might have changed by that time.

53 On 1 November 2023, before I had published my reasons, the High Court upheld Mr Benbrika's challenge to the cancellation of his citizenship.[13] The immediate effect of that decision was that Mr Benbrika would no longer be held in immigration detention when his CDO was replaced with an ESO. Instead, he would live in the community under an ESO, as both parties originally envisaged when these two proceedings were commenced. Given that my draft reasons had not considered the community conditions at all, the effect of the High Court decision was to require substantial revision of my draft reasons.

54 On 2 November 2023, I informed the parties that I would need to revise my draft reasons, to address the consequences of the High Court decision. Unfortunately, I began a jury trial in another proceeding shortly thereafter, which delayed the preparation of written reasons addressing the community conditions.

55 As previously indicated, on 19 December 2023, I made final orders and gave detailed oral reasons for my decision.

56 There is no dispute that the formal requirements of ss 105A.3(1) and 105A.7A(1)(a) have been met in this case, so I will say nothing further about those matters. Instead, I will focus on the areas of dispute.

57 The court can make an ESO if, after having considered the matters specified in s 105A.6B of the Code, the court is satisfied on the balance of probabilities that: the offender poses an unacceptable risk of committing a serious Part 5.3 offence; and each of the conditions, and the combined effect of all of the conditions in the ESO, are reasonably necessary, and reasonably appropriate and adapted, for the purpose of protecting the community from that unacceptable risk.[14]

58 The AFP Minister bears the onus of satisfying the court of the existence of an unacceptable risk, before either a CDO or an ESO can be made.[15] However, the AFP Minister must establish such a risk to a higher standard of proof for a CDO (to a high degree of probability) than an ESO (on the balance of probabilities).

59 Mr Benbrika did not formally concede that he posed an unacceptable risk of committing a serious Part 5.3 offence, and therefore did not consent to an ESO being made. However, his counsel said that Mr Benbrika "accepted the reality" that it was open to the court to find that he did pose such a risk, and to make an ESO.[16] Given Mr Benbrika's formal position, it was still necessary for me to address the issue of "unacceptable risk", albeit in considerably less detail than was required in the first review proceeding.

60 The parties disagreed as to the duration of any ESO that the court might make. The A-G argued that the order should be made for three years. Mr Benbrika argued that it should be made for no more than one year.

61 Many of the proposed conditions of the ESO were not in dispute, or were only disputed to a minor extent. However, Mr Benbrika argued that some of the ESO conditions proposed by the A-G exceeded what was reasonably necessary, and reasonably appropriate and adapted, for the purpose of protecting the community from any unacceptable risk that the court found that he posed. Mr Benbrika also sought to have some additional conditions imposed.

62 Before an ESO could be ordered, the court had to be satisfied on the balance of probabilities, on the basis of admissible evidence, that Mr Benbrika posed an unacceptable risk of committing a serious Part 5.3 offence.

63 The concept of "unacceptable risk" is relevant to a number of provisions in the Code. However, it is not defined in the Code.

64 I discussed the relevant authorities in paragraphs [48]-[52] of the first review reasons. For present purposes, it is sufficient to note the following.

65 The concept of unacceptable risk is a flexible one, which is calibrated to the nature and degree of the risk, so it can be adapted to the particular case. Whether a risk is unacceptable requires consideration of both the degree of likelihood of the risk eventuating, and the seriousness of the consequences if it does.[17]

66 The risk of offending must carry a threat of harm to members of the community that is sufficiently serious as to make the risk of the commission of the offence unacceptable to the court.[18]

67 In deciding whether Mr Benbrika posed an unacceptable risk, the court must have regard to the matters listed in s 105A.6B of the Code ("the specified matters"). The specified matters will be considered shortly.

68 The case against Mr Benbrika was primarily a circumstantial one. In considering the specified matters, a court is required to examine and consider the effect of all of the evidence as a whole, rather than take a piecemeal approach. It does not matter whether or not one piece of circumstantial evidence is sufficient to establish to the requisite standard that the offender poses an unacceptable risk; it is the effect of the whole of the evidence that is important.[19]

69 The A-G did not suggest that Mr Benbrika would commit a so-called "lone wolf" terrorist attack. Rather, the A-G argued that the general nature of the risk posed was that he would, if released into the community, again radicalise others, or encourage them to engage in acts of religious-inspired violence. The A-G said, for example, that Mr Benbrika might play a supportive role within, or on the edge of, a terrorist organisation. This is an important point to bear in mind, when assessing the reasonableness of some of the conditions that were sought by the A-G.

70 The A-G argued that the risk would be of committing one or more of the following serious Part 5.3 offences: providing training connected with terrorist acts (s 101.2); directing the activities of a terrorist organisation (s 102.2); intentional membership of a terrorist organisation (s 102.3); recruiting for a terrorist organisation (s 102.4); or providing or participating in training involving a terrorist organisation (s 102.5).

71 Whichever serious Part 5.3 offences are being considered, it is sufficient for the court to consider and identify the risk that a person poses, without linking the risk to a specific offence, as long as the risk identified is a risk of committing one or more of those offences. It is not necessary to identify with precision the specific offences that Mr Benbrika might commit; nor is it necessary to say with respect to each such offence, considered in isolation, that there is an unacceptable risk of him committing that offence. Such an analysis would be unrealistic.[20]

72 At the time of the original CDO proceeding in late 2020, the Minister's primary expert, Dr Chelsey Dewson, assessed Mr Benbrika's overall risk of committing a relevant offence as "high". By the time of the first review proceeding in late 2021, she had reduced that to "moderate-high." In the current proceedings, Dr Dewson reduced her overall risk assessment to "moderate-low".

73 Mr Benbrika's expert, Dr Anne Speckhard, assessed Mr Benbrika's current risk of committing a relevant offence as considerably lower than that.

74 I will discuss the competing risk assessment evidence shortly. However, the A-G did not dispute that Mr Benbrika has been making substantial progress towards deradicalisation in recent times. Notwithstanding that progress, the A-G argued that the very serious nature of the possible offending and its consequences were such that the court should be satisfied that the risk was still "unacceptable".

75 In assessing whether there is an unacceptable risk, the court must have regard to the matters specified in s 105A.6B of the Code. Some of the specified matters were not disputed in these proceedings, others were contested.

76 Where there had been no change to the specified matter since the first review, I have relied on the relevant parts in the first review reasons, without necessarily specifically referring to them in these reasons.

The object of Division 105A (s 105A.6B(1)(a))

77 The only stated object of the Division is to "protect the community" from serious Part 5.3 offences, by providing that terrorist offenders who pose an unacceptable risk of committing such offences are subject to a CDO or ESO (s 105A.1).

Mr Benbrika's prior terrorist offending (s 105A.6B(1)(g))

78 The court must have regard to Mr Benbrika's convictions for, and findings of guilt made in relation to, any serious Part 5.3 offences.

79 Mr Benbrika's convictions for intentionally being a member of, and intentionally directing the activities of, a terrorist organisation, are the only relevant convictions for the purpose of this subsection.

80 The details of Mr Benbrika's offending have been summarised at length before, including in paragraphs [69]-[70] of the first review reasons, and it is not necessary to repeat them here.

81 The sentencing judge, Bongiorno J, found that there was no credible evidence that Mr Benbrika had been planning an attack on any specific target. Notwithstanding that clear judicial finding, some journalists persist in reporting the false allegation that Mr Benbrika was planning to blow up iconic Australian landmarks, such as the MCG or Crown Casino. In fact, Bongiorno J described the witness who alleged that Mr Benbrika wanted to blow up such targets as "a liar, a cheat, and a fraudster of significant accomplishment," someone whose evidence was "untrue and designed purely to serve his own ends."[21] Mr Benbrika and his co-accused were arrested long before any plans had been substantially developed, or any specific target identified.

The views of the sentencing court (s 105A.6B(1)(h))

82 The court must have regard to the views of the sentencing court at the time of imposing sentence. It is clear from the sentencing remarks that Bongiorno J regarded Mr Benbrika's offending as very serious. The key findings made by Bongiorno J were summarised at paragraphs [73]-[78] of the first review reasons.

83 For the purpose of these proceedings, the focus must be on Mr Benbrika's current risk. The legislation specifically requires the court to have regard to past offending, and the remarks of the sentencing judge, as part of the overall risk assessment. How much weight should be given to such historical matters will no doubt vary from case to case, depending on the individual circumstances.

Compliance with legal obligations (s 105A.6B(1)(f))

84 The court must have regard to the extent of an offender's compliance with any obligations that applied while the offender was on parole, or subject to a CDO or control order.

85 For the reasons discussed in paragraphs [443]-[445] of the first review reasons, this specified matter was not relevant to Mr Benbrika.

Participation in treatment or rehabilitation programs (s 105A.6B(1)(e))

86 The court must have regard to any treatment or rehabilitation programs in which Mr Benbrika has had an opportunity to participate, and the level of his participation in any such programs.

87 Since the first review, Mr Benbrika had continued his engagement with Corrections Victoria psychologists. Between 4 November 2021 and 15 December 2022, Mr Benbrika attended 16 psychological sessions. Two other sessions were cancelled by Corrections Victoria, and 17 sessions were cancelled by Mr Benbrika (nine of which were because Mr Benbrika failed or refused to attend or engage).

88 At the end of May 2022, Mr Benbrika was allocated a new clinician; it was mostly after this that his attendance became sporadic. His refusal to attend sessions usually occurred when he was frustrated with prison staff, or had received an adverse outcome. However, Corrections psychologists reported that when Mr Benbrika did attend sessions, he generally engaged well.

89 His unwillingness to engage with the new Corrections psychologist was of less concern than it might otherwise have been because, in December 2022, Mr Benbrika also began seeing an experienced private psychologist, Dr Rachel MacKenzie. Sessions ranged from weekly to once every three weeks and, at the time of giving evidence, Dr MacKenzie had conducted 11 one-hour sessions with Mr Benbrika. Dr MacKenzie reported that Mr Benbrika engaged openly in the sessions, and was receptive to strategies she proposed.

90 Mr Benbrika also continued to participate in the Community Integration Support Program ("CISP"). Between 17 November 2021 and 30 January 2023, Mr Benbrika attended 43 CISP sessions. He refused to attend on three occasions. CISP records show that Mr Benbrika engaged fully in his sessions, asked questions of the mentors, and was open to challenging his core beliefs.

91 In March 2023, Mr Benbrika engaged in five sessions, totalling 18 hours, with Dr Speckhard and an Islamic scholar from the International Centre for the Study of Violent Extremism ("the Sheikh"). Those sessions included testing Mr Benbrika with challenging hypothetical scenarios, to see how he applied his ideological beliefs. He engaged fully, and responded positively to the Sheikh's mentorship.

92 Both sides' experts agreed that Mr Benbrika had made considerable progress in deradicalisation over the previous year, since the first review. However, the process of deradicalisation was incomplete, and there remained outstanding goals that he needed to address.

93 The A-G accepted that Mr Benbrika had taken steps towards renouncing his previous ideology and, at the time of the trial, did not support an ideology that justified the use of violence.

Assessments by relevant experts (s 105A.6B(1)(b) and (c))

94 The court must have regard to any assessment received from a "relevant expert" as to the risk of the offender committing a serious Part 5.3 offence, and the level of the offender's participation in the assessment.

95 A "relevant expert" means any registered doctor, psychologist, psychiatrist or other expert "who is competent to assess the risk of a terrorist offender committing a serious Part 5.3 offence".[22]

96 For the purposes of these two proceedings, Mr Benbrika was assessed by two relevant experts, one appointed by each side.

97 The A-G's expert was Dr Chelsey Dewson, a registered psychologist. She also gave evidence for the Minister for Home Affairs in the CDO and first review proceedings. Dr Dewson used the same general approach as in previous proceedings, namely a structured professional judgment ("SPJ") approach, in which she drew on and interpreted a number of sources of information applying the VERA-2R guidelines.

98 Mr Benbrika's expert was Dr Anne Speckhard, a professor of psychiatry. She had not previously given evidence in relation to Mr Benbrika. She used the clinical judgment approach, rather than the SPJ approach.

99 The Code does not specify a preferred risk assessment approach. In particular, the Code does not require that a relevant expert use a formal risk assessment tool, in order to carry out a risk assessment. But if an expert does choose to use such an instrument, it is self-evident that it should be empirically based and have some underlying validity. The AFP Minister's statutory obligation to disclose exculpatory material clearly includes disclosing expert reports that cast doubt about the validity of a risk assessment tool used by the Minister's own expert. That is why the non-disclosure by the AFP Minister of the Corner report, and the four other expert reports critical of VERA-2R, is of such great concern to the court.

100 Dr Dewson's opinion was that Mr Benbrika was a "moderate-low" risk of engaging in a relevant terrorism offence. She said that he had made positive progress in CISP over the previous year, and had demonstrated a willingness to challenge his core beliefs more meaningfully than in the past. She noted that the CISP mentors had placed greater confidence in his recent progress.

101 While Dr Dewson accepted that Mr Benbrika did not currently support an ideology that justified the use of violence, she regarded that as "highly dynamic" and requiring future monitoring.

102 In assessing Mr Benbrika as presenting a "moderate-low" risk of violent extremism, Dr Dewson summarised what she regarded as the main risk factors as follows:

The enduring risk factors include his associations; rejections of some elements of a democratic society; grievances towards perceived injustices (associated with his enduring detention and treatment in custody); cognitive inflexibility; poor insight; and poor coping/emotional dysregulation. Some historical risk factors may present as static-like risk factors, such as exposure to extremist material and participation in informal weapons training.

103 Dr Dewson thought that the most likely offence scenario would involve Mr Benbrika playing a supportive role within a terrorist organisation, or on the periphery of such an organisation.

104 Dr Speckhard did not use VERA-2R labels, such as "low" or "moderate-low". Asked to give a one-word summary of Mr Benbrika's risk of engaging in a relevant terrorist offence, she described it as "miniscule". If Mr Benbrika remained mentored, Dr Speckhard said she was 100 percent confident that he would not revert to jihadist thought patterns. If Mr Benbrika was left unmentored, there would be some risk to consider, but it would be very low.

105 Dr Speckhard's opinion was that Mr Benbrika did not present a danger to Australian society, or a danger of radicalising others. The main reasons for her forming that opinion were as follows. Mr Benbrika deeply regretted his previous views, and was totally convinced that he lived under a contract or covenant which prevented him from breaking Australian laws, bringing harm to Australians, or encouraging others to do so. Mr Benbrika was totally convinced that he had to follow the covenant and that extremist actions, violent actions, and militant jihadist actions were prohibited to him. His past experiences in custody have shown that he would not be re-radicalised in response to people who insulted him. Mr Benbrika was also terrified of the prospect of returning to prison, and would not do anything to put that in jeopardy. Dr Speckhard and the Sheikh both found Mr Benbrika to be honest, teachable, responsive to strong Islamic scholarship and guidance, and a willing and enthusiastic pupil.

106 Although both experts agreed that Mr Benbrika had made considerable deradicalisation progress in the past year, there were some matters about which their professional assessments of him differed. For example, Dr Dewson regarded Mr Benbrika as being "cognitively inflexible", whereas Dr Speckhard found him to be "completely thoughtful, flexible and responsive." Dr Dewson thought that Mr Benbrika had made "minimal advancements with emotional regulation", whereas Dr Speckhard recognised that as being a cultural matter, in which Mr Benbrika behaved in a culturally normative manner for an Arab person.

107 Dr Speckhard recognised that Mr Benbrika still had some key vulnerabilities that tipped him into being a militant jihadist inciter and a group leader. Those vulnerabilities include the fact that he needs followers and a sense of importance and respect, and that he is emotionally volatile.

108 Although there were many things about which the two experts agreed, they adopted different approaches to the risk assessment task. Dr Dewson utilised VERA-2R in her SPJ approach, whereas Dr Speckhard used the clinical judgment approach. It was not the task of this court to undertake an abstract consideration of the relative strengths and weaknesses of the different possible risk assessment approaches (SPJ, actuarial risk assessment, and clinical judgment), and to decide which should be preferred in the case of violent extremists. My task was to evaluate the evidence of Dr Speckhard and Dr Dewson, to the extent that it was necessary to resolve any differences in their opinions about the risk that Mr Benbrika presented.

109 It is clear from the Corner report (and the other four expert reports) that there are serious problems in using VERA-2R to assess a person's risk of violent terrorist offending. Paragraph [22] of these reasons listed what the Corner report said were the most significant problems with VERA-2R; it concluded that "without a strong theoretical and empirical basis for factor inclusion, it was not reasonable to anticipate that [VERA-2R and Radar] were able to predict their specified risk with 'anything other than chance'."

110 The AIC report, upon which the A-G relied, did not satisfactorily address the fundamental problems identified in the Corner report. The fact that the AIC report concluded that, "By simple elimination, it appears that the VERA-2R is the better of the available risk assessments for use", does not establish its validity; it effectively says no more than that it appears to be the best of a bad bunch.

111 VERA-2R openly acknowledges that it does not have predictive validity. As the Corner report noted, VERA-2R is extremely poor at predicting outcomes, in discriminating between those who will go on to commit an offence, and those who will not. That is of considerable concern, because it is a fundamental part of the court's task to assess the risk that Mr Benbrika will commit a relevant terrorist offence in the future.

112 Both experts agreed that an SPJ tool such as VERA-2R may have some utility, even without having predictive validity. Dr Speckhard accepted that a checklist might be useful for a relatively inexperienced or non-expert practitioner with little idea of terrorism. Dr Dewson said that an SPJ tool can provide a standardised, manualised template that can be replicated, and which allows practitioners to review and understand each other's assessments. Those may well be desirable outcomes from a policy perspective. However, if the tool being applied consistently is bad, then the results will be consistently bad. Furthermore, such considerations do not guarantee the correctness of the tool's output, and are irrelevant to the court's risk assessment task under s 105A.7A of the Code.

113 In fact, Dr Dewson agreed with many of the criticisms in the Corner report. However, she tried to distance herself to some extent from the problems, by minimising the significance of VERA-2R to her assessment, and stressing the "professional judgment" part of her opinion.

114 When a practitioner uses the VERA-2R tool, they turn their mind to each of the 34 core factors and 11 additional risk indicators in the VERA-2R checklist, forming a view about how they apply, and then using those factors and any other relevant information to form a final judgment about risk. A checklist of factors in an SPJ tool tells the practitioner to think about particular things. An empirically-sound SPJ tool should direct the practitioner to think about the things that are most relevant to risk, and not direct them to think about things that are not relevant to risk. The Corner report noted that assessors who used VERA-2R only found 9 out of the 45 indicators to be frequently useful. Further, the Corner report found that VERA-2R covered less than 15% of the variables found in the empirical evidence to be statistically associated with movements towards radicalisation and violent extremism.

115 Even where there is empirical support for the inclusion of a factor, the fact that a practitioner is thinking about a factor because a checklist says to do so does not improve the quality of the opinion. For example, both experts considered Mr Benbrika's ideology. Dr Dewson did so because she was applying VERA-2R, and ideology is one of the factors in the tool's checklist. Dr Speckhard considered ideology because her study, training and experience informed her that ideology is relevant to risk assessment.

116 Dr Speckhard is an internationally recognised expert in the field of terrorism studies. She is an adjunct professor of psychiatry at the Georgetown University Medical School, and the founder of the International Centre for the Study of Violent Extremism in the United States. She has frequently been consulted by, or provided training to, governments and security and police forces from around the world, including the US Departments of Justice, Defence and Homeland Security, the UK Home Office, and the Australian Federal Police. She has also been consulted by the developers of SPJ tools. The list of her publications, appointments and consultancies in the field runs to more than 80 pages, and demonstrates the length, depth and breadth of her expertise in relation to violent extremism.

117 In coming to her risk assessment, Dr Speckhard relied upon her extensive experience studying and working in the violent extremist field for around 40 years. Dr Speckhard had interviewed more than 800 terrorists, violent extremists, their family members and supporters from around the world.

118 I was satisfied that Dr Speckhard's expertise extended to Islamic extremism in Australia, even if that were to be regarded as a discrete area of specialised knowledge (which I did not accept, for the reasons articulated by Dr Speckhard).

119 The A-G objected to several other parts of Dr Speckhard's report. I was satisfied that the objections were without substance, for the reasons set out in paragraphs [149] to [165] of Mr Benbrika's closing submissions. I will not spend any further time discussing them here, for two reasons. First, the objections did not detract from the force and validity of Dr Speckhard's opinions. Secondly, my decision on risk assessment does not require a detailed ruling on the objections.

120 Dr Speckhard has vastly more experience than Dr Dewson, who only obtained her doctorate in 2021, has only worked in the extremism field since 2016, and had assessed less than 20 offenders under high-risk offender legislation.

121 Dr Speckhard worked with the Sheikh, who is an expert in Islamic theology. Dr Dewson had no assistance from an Islamic expert, and acknowledged that she had no expertise in that area herself.

122 Dr Speckhard strongly believed that Mr Benbrika had made considerable progress in deradicalisation during the 18 hours that she and the Sheikh spent with him in March 2023. That was more than the total time that Dr Dewson had ever spent with Mr Benbrika, being a total of seven hours over three years.

123 As the A-G pointed out, Dr Dewson had the benefit of seeing Mr Benbrika's trajectory over a number of years. That was a benefit that Dr Speckhard did not have. However, Dr Dewson's observation that Mr Benbrika's trajectory had been a positive one, in which he had been moving away from his previous violent ideology, was not a contentious one.

124 Dr Dewson had not spoken to Mr Benbrika at all since his deradicalisation work with Dr Speckhard and the Sheikh. Dr Dewson acknowledged that was a limitation on her most recent assessment. It is unknown what Dr Dewson would have assessed Mr Benbrika's risk to be, if she had assessed him since his intensive work in March 2023.

Reports relating to community management (s 105A.6B(1)(d))

125 The court is required to have regard to any reports relating to the extent to which the offender can reasonably and practicably be managed in the community, that have been prepared by a person competent to assess such matters.

126 The A-G's written submissions only relied upon the tenth Dewson report under this heading. Dr Dewson's evidence has been discussed elsewhere in these reasons.

Any other relevant information (s 105A.6B(1)(i))

127 At the end of the list of specified matters is a provision that the court must consider "any other information" as to the risk of the offender committing a serious Part 5.3 offence.

128 In contrast to the first review, the A-G did not raise any additional things for consideration under this specified matter.

129 It was undisputed that Mr Benbrika had made substantial progress in deradicalisation since the first review and, at the time of trial, did not support an ideology that justified the use of violence. That was a relatively recent development in his ideology, which still remained potentially fragile. He still had unmet needs in terms of his deradicalisation.

130 If Mr Benbrika continues to participate in further deradicalisation work under the ESO, particularly under the mentorship of the Sheikh, I accepted that his risk of committing a relevant offence would be minimal. However, if he fails to engage appropriately in further deradicalisation work, or re-engages with radical jihadists, there would be a risk of him reverting back to his previous ideology, which would increase his risk of committing a relevant offence.

131 At the end of the day, it was not necessary for me to make a finding as to which expert's evidence should be preferred. Even on Dr Speckhard's evidence, Mr Benbrika posed an unacceptable risk of radicalising others to commit acts of religiously motivated extremism. Given the recency of Mr Benbrika's change in ideology, I accepted that the serious nature of the possible offending and its consequences were such that the relevant risk was still unacceptable as at December 2023.

132 However, the A-G rightly accepted that Mr Benbrika no longer needed to be retained in a prison on a CDO; any risk that he still presented was capable of being managed by him living in the community under a comprehensive CDO.

133 The court can only make an ESO if it is satisfied that each of the conditions, and the combined effect of the conditions, to be imposed is "reasonably necessary, and reasonably appropriate and adapted", for the purpose of protecting the community from the unacceptable risk that Mr Benbrika presents. I will use the expressions "reasonable" or "reasonableness" as shorthand. That expression imposes a requirement of proportionality.[23] The court must ensure that the conditions do not impose a greater degree of restraint than is required.[24]

134 In determining whether there is a sufficient connection between a condition and the relevant risk of offending, the court should consider:

(a) whether the condition addresses a means of controlling possible risk factors related to the type of offending in question;

(b) whether the condition is designed to avoid situations where the offender's rehabilitation or reintegration into the community might be compromised; or

(c) in some cases where the condition does not directly relate to the offender's risk, whether it promotes the efficacy of other conditions that do.[25]

135 Section 105A.7B sets out more than five pages of general and specific conditions that may be included in an ESO. It is not necessary to consider them in any detail, as there was no dispute that the A-G's proposed conditions were of a sort that is permitted by the provision. The dispute in this case was as to whether particular conditions proposed by the A-G were reasonably necessary, and reasonably appropriate and adapted to Mr Benbrika's situation.

136 In making that decision, the court must take into account, as a paramount consideration, the object of Division 105A, namely, the protection of the community from serious Part 5.3 offences (s 105A.7A(2)).

137 Nevertheless, in determining the reasonableness of any of the A-G's proposed conditions, it must be remembered that breach of an ESO condition is a criminal offence, carrying a maximum penalty of 5 years' imprisonment.[26]

138 In addition to Dr Dewson, the A-G also relied upon evidence from Brooke Hartigan and Nicholas Read, about whose evidence I make the following observations.

139 Ms Hartigan is the First Assistant Secretary of the Security and Counter-Terrorism Division of the A-G's Department. She provided affidavits dated 27 February and 21 April 2023 (which addressed the proposed conditions if Mr Benbrika was living in the community), and an affidavit dated 31 May 2023 (which addressed the proposed conditions if he was detained in Long Bay). She was also cross-examined at trial. Much of her evidence was simply an expression of her personal opinion as to the reasonableness of the proposed conditions, and why she believed they would diminish risk. In particular, her May affidavit consisted of little more than a mantra-like repetition of the statutory test in respect of the various conditions.

140 Unsurprisingly, Mr Benbrika objected to the admissibility of much of her evidence, on the grounds set out in a document dated 1 June 2023. In particular, Mr Benbrika argued that there is no field of specialised expert knowledge concerned with identifying what types of conditions are reasonably necessary, or reasonably appropriate and adapted, to deal with a given level of risk. Rather, he argued that whether the statutory test is satisfied is a question of judgment, weight and balance, for the court alone to determine.

141 Ultimately, it was not necessary for me to rule on the objections, as the parties agreed that I should treat Ms Hartigan's opinions on the proposed conditions as submissions, not evidence. However, I make the following observations in relation to Ms Hartigan's evidence, in order to ensure that a more suitably qualified witness is put forward in any future proceeding.

142 At the time of trial, Ms Hartigan had only been in her current position for less than one year. Her current role included oversight of the High Risk Terrorist Offenders Branch, within the Security and Counter-Terrorism Division. Within the previous five years, she had held senior public service positions in various government departments, including in the Department of Education, Skills and Employment, the Department of Social Services, and a position in education and research at the Australia Embassy in Beijing.

143 I accept that Ms Hartigan is an experienced public service manager. However, she has no formal education, practical training or experience (other than in her current role) in risk assessment or dealing with terrorist offenders. That did not stop her from expressing her opinions, in extremely confident terms, as to why all of the A-G's proposed conditions were necessary to contain Mr Benbrika's risk of offending. Overall, Ms Hartigan was an unimpressive witness.

144 Evidence as to the proposed conditions was also given by Nicholas Read, who held the rank of Commander, Counter Terrorism Investigations East, within the AFP. He provided affidavits dated 6 March and 23 April 2023 (which addressed the proposed conditions if Mr Benbrika was living in the community), and an affidavit dated 31 May 2023 (which addressed the proposed conditions if Mr Benbrika was detained in Long Bay). Commander Read was also cross-examined at trial.

145 Unlike Ms Hartigan, Commander Read is very experienced in working in counter terrorism, and no attack was made on his expertise in the area.

146 No objection was taken to the admissibility of the first part of his March affidavit, which discussed his understanding of the threat of religiously motivated violent extremism in Australia. However, I note that much of the information in his affidavit about extremism related to historical, rather than current, matters.

147 Objection was taken to the opinions expressed in Commander Read's May affidavit about the reasonableness of the proposed conditions; they involved numerous recitations of the statutory test in a mantra-like way. As with Ms Hartigan's evidence, the parties agreed that the court should treat Commander Read's opinions as to the reasonableness of the conditions as submissions, not evidence, so I will say nothing further about it.

148 Since the High Court decision, it was no longer necessary for me to consider the proposed conditions that would have applied if Mr Benbrika was held in immigration detention in Long Bay.

149 The A-G's proposed conditions for a community-based ESO were set out in the schedule attached to the proposed minute of final orders, dated 27 February 2023. There were 28 different conditions, covering a wide range of matters. They were further refined by the A-G in the course of written and oral submissions.

150 Mr Benbrika agreed to most of the proposed conditions, either in their original form, or with modifications, but objected altogether to a few conditions. He also proposed some additional conditions.

151 Unless the precise words of a proposed condition were critical to my consideration of the reasonableness of the condition, I have not set out the proposed condition verbatim in my reasons. Instead, I have focussed on the substance of the proposed condition. The final ESO and conditions are attached as an appendix to these reasons.

152 At the back of the conditions is a section headed "Interpretation", which contains a number of definitions that apply in the conditions. The following two are worth noting here, as they are relevant to a number of the conditions:

(a) "AFP Superintendent" means a member of the AFP performing the duties of a Superintendent within the Counter Terrorism and Special Investigations Command, and whose contact details are provided to Mr Benbrika at the time of the ESO coming into force, and any other member of the AFP acting under the direction of or on behalf of the AFP Superintendent; and

(b) "Branch Manager" means an employee of the A-G's Department performing the duties of Assistant Secretary with responsibility for high risk terrorist offenders, and whose contact details are provided to Mr Benbrika at the time of the ESO coming into force, and any other employee of the A-G's Department acting under the direction of or on behalf of the Branch Manager.

153 There was no dispute as to conditions 1.1 and 1.2. That is to say, Mr Benbrika must reside at a specified address, which he cannot change without the approval of the AFP Superintendent, and must be at the specified premises during a prescribed curfew period each night.

154 The A-G also proposed a condition that would have enabled the AFP Superintendent to visit and enter the specified premises, for the purpose of ensuring compliance with the curfew.

155 It was reasonable to provide a mechanism to enable the AFP to ensure compliance with the curfew. However, it was not reasonable to empower the AFP to enter and search a residential home, which is also occupied by persons other than Mr Benbrika, in the middle of the night, simply to check that Mr Benbrika is complying with the curfew. The desired result could be achieved by adopting the usual practice that is used whenever a curfew is imposed as a condition of bail, or a community correction order, or the like, namely by requiring Mr Benbrika to present himself at the front door of the specified premises during the curfew period, when required to do so. I drafted condition 1.3 accordingly.

156 There was no objection to the A-G's proposed case management condition. It requires Mr Benbrika to report to the Branch Manager, and participate in any assessment or review of his case plan by the Branch Manager.

Mental health assessment and treatment condition (condition 3)

157 The parties agreed that Mr Benbrika should continue to receive psychological treatment with Dr MacKenzie, with whom he has formed a positive therapeutic relationship. She is prepared to continue treating him.

158 The parties also agreed that if treatment with Dr MacKenzie ceased, Mr Benbrika should be assessed to determine whether he is suitable for further psychological treatment. If he is assessed as suitable, then he should undertake that further treatment.

159 The position was slightly different in relation to psychiatric treatment. At the time of making the ESO, Mr Benbrika was not currently seeing a psychiatrist, and had not been assessed as having any unmet psychiatric treatment needs.

160 Mr Benbrika did not object to undertaking a psychiatric assessment, and treatment if recommended by a psychiatrist, but wished to have a reasonableness requirement imposed. That is because psychiatry is a discipline in which seriously invasive treatment (such as electro-convulsive therapy, or medications with adverse side-effects) could theoretically be recommended. I agreed that there should be scope for Mr Benbrika to decline to undertake any unreasonably recommended psychiatric treatment.

161 There was no dispute that the Commonwealth should pay for all mental health assessments and treatments.

162 The only other dispute related to the wording of a proposed condition requiring Mr Benbrika to attend all psychological and psychiatric appointments. The A-G wanted the condition to be that Mr Benbrika must attend all mental health treatment appointments, unless he had a reasonable excuse. Mr Benbrika wanted the condition to be that he must not fail to attend three or more mental health treatment appointments in a row, unless he had a reasonable excuse. I accepted Mr Benbrika's proposal, for the following reasons.

163 It is not uncommon for persons undergoing mental health treatment to face various hurdles in the course of their treatment, particularly if they are confronting challenging personal issues as part of the treatment. Occasional non-attendance at an appointment is not necessarily a barrier to making therapeutic progress. Mr Benbrika has faced such challenges from time to time whilst in custody, but has generally persevered and made progress with his treatment.

164 The primary management of Mr Benbrika's mental health treatment should be left as far as possible to experienced mental health practitioners, not public servants in the A-G's Department. It is not desirable to create a situation in which the Branch Manager can be demanding an explanation every single time Mr Benbrika misses the occasional appointment.

165 The mental health practitioners who treat Mr Benbrika will be subject to the information-sharing provisions contained in condition 5, which will provide community protection in the event that Mr Benbrika disengages from treatment, or there is a relevant deterioration in his mental state or a concern for the safety of him or others.

166 In recent times, Mr Benbrika has been making good progress through his participation in deradicalisation programs, both with CISP and the Sheikh. The parties agreed that he should continue to participate in such programs. There was no dispute about the substance of the appropriate conditions.

167 By the time of the trial, Mr Benbrika had started participating in such a program with the Sheikh. The A-G accepted that the Sheikh was a suitable person to conduct such a program in the future. The A-G also accepted that the Commonwealth should pay for Mr Benbrika's ongoing participation in the program run by the International Centre for the Study of Violent Extremism.

168 Mr Benbrika also agreed to continue to participate in the CISP program, or an alternative community-based program (if CISP ceased to be made available to him).

Reasonable directions and information sharing requirements (condition 5)

169 There was no dispute about condition 5.1, which requires Mr Benbrika to provide certain information to the AFP Superintendent.

170 There was also general agreement about condition 5.2, which requires Mr Benbrika to provide certain information to the Branch Manager. However, I agreed with Mr Benbrika that he should not be required to provide the Branch Manager with information about the substance of his mental health treatment, for the following reasons.

171 Mr Benbrika agreed to the general sharing of information between persons involved in his supervision (condition 5.3(a)). I accepted a modified information-sharing regime in the case of treating mental health practitioners.

172 Confidentiality is critical for the maintenance of a therapeutic relationship, especially where mental health issues are concerned. For psychological or psychiatric treatment to be effective, the patient must be able to speak freely and confidentially, and in a relationship of trust. I accepted Dr MacKenzie's opinion that the proposed condition as drafted by the A-G would be likely to have a detrimental effect on the quality and effectiveness of the psychological treatment provided to Mr Benbrika. The same would apply to any psychiatric treatment.

173 I drafted condition 5.3(b) so as to impose information-sharing requirements in respect of mental health practitioners, which will enable the relevant authorities to monitor Mr Benbrika's compliance with the ESO, and to protect community safety, whilst at the same time preserving his therapeutic relationships. The substance of condition 5.3(b) is similar to the reporting obligations that apply, and work effectively, in the case of serious violent offenders and serious sex offenders.

174 The A-G also proposed three other conditions under this general topic of directions, which I did not impose for the following reasons.

175 Two of the proposed conditions sought to impose a general obligation on Mr Benbrika to comply with "any reasonable direction" given by the AFP Superintendent or the Branch Manager, in relation to certain specified matters.

176 I set ESO conditions that granted an extensive and detailed range of specific powers to the AFP Superintendent and the Branch Manager, the reasonableness and appropriateness of which had been rigorously tested through a transparent court hearing. If there is to be a substantial change to the powers of the AFP Superintendent or the Branch Manager, it should be done through a similar process.

177 Parliament has given the power to make post-sentence orders, and to set their conditions, to the courts, not the AFP or the A-G's Department. The proposed power to allow them to give "reasonable directions" would have enabled the AFP Superintendent or Branch Manager to unilaterally rewrite the ESO conditions.

178 The AFP have demonstrated a degree of rigid thinking, and a focus on historical rather than current matters, when it comes to Mr Benbrika. The AFP is also understandably concerned about operational issues (including resource allocation), that may not be relevant to a consideration of the court's power under the Code. The AFP's view of what is reasonable frequently differs from what the court regards as reasonable. Two examples will suffice for present purposes.

179 In the first review reasons, I rejected as utterly unfounded the AFP assertion that Mr Benbrika was likely to have a "sophisticated understanding of police surveillance methodology", which he acquired prior to and during his criminal trial. After discussing the matter in the first review reasons, I concluded that:

This argument rests on the unproven, and highly unlikely, assumption that police methodology and technology have remained static over the last 17 years. There is simply no evidence to support the Minister's argument that Mr Benbrika has any knowledge of current police methodology and technology.

180 In the current proceedings, Commander Read did not seek to remedy that deficiency by adducing any evidence as to police methodology at relevant times. Instead, he simply repeated the same opinion and assertion expressed by the relevant AFP witness in the first review proceeding.

181 The second example is this. In the first review proceeding, it was clear that the AFP would like to prohibit Mr Benbrika from having any contact with two of his sons, Bakr and Oussama. After considering the very limited evidence about those sons, and the possible effect that such a ban may have on Mr Benbrika, in my first review reasons, I concluded as follows:

If I were otherwise minded to order an ESO, ... such an extraordinary prohibition should be decided by the court, not left as a matter of unfettered discretion for the AFP Superintendent to decide.

182 Nothing more was put before the court in these proceedings that might justify giving the AFP Superintendent the power to prevent Mr Benbrika from associating with any of his family members. Accordingly, I inserted condition 6.3 to make it clear that any such condition may only be imposed by a court variation, made on cogent, admissible evidence.

183 The A-G had also sought a condition that Mr Benbrika must notify the AFP Superintendent in writing three calendar days prior to his attendance at any place of public religious worship, service or celebration, or at any private prayer group. That was unreasonable for a number of reasons.

184 In general terms, it was an unjustifiable limitation on the right of Mr Benbrika to freely practise his religion, and to attend community events such as weddings and funerals.

185 In so far as it sought to deal with a "private prayer group", the A-G's proposed condition would have prevented Mr Benbrika from praying with his own family in his own home, or from engaging in spontaneous prayer with friends outside his home.

186 The A-G's proposed condition was potentially very onerous. It was also not reasonable, having regard to the combined effect of all the other ESO conditions, which include the following. There are conditions that prevent Mr Benbrika from associating with inappropriate people, or engaging in prohibited activities. Mr Benbrika will be required to wear and maintain an electronic monitoring device, so the AFP will know where he is at all times. The AFP will also be given extensive monitoring powers through the numerous conditions dealing with mobile phones, electronic communications, and the like. In those circumstances, the A-G's proposed notice condition was not reasonable.

187 The A-G proposed an extensive list of persons or classes of persons with whom Mr Benbrika must not communicate or associate. The prohibited persons included certain named terrorists, as well as unnamed people who fall within specified categories (including prisoners, persons subject to terrorism supervision orders, and persons resident in specified countries where terrorist activity is widespread).

188 For the most part, there was no objection to the form of the conditions proposed by the A-G. However, I made the following amendments to the A-G's proposed form.

189 At the start of condition 6.1, I made it clear that the prohibition is on Mr Benbrika "knowingly**"** communicating or associating with the prohibited persons. That is to avoid possible criminal consequences flowing from an inadvertent communication or association with such a person. For example, Mr Benbrika does not know all of the people listed in condition 6.1(e), and may not know additional people who may be specified in the future by the AFP Superintendent under condition 6.2; there is therefore the potential for inadvertent contact with such people. Or he could inadvertently communicate with a prohibited person if they were within earshot of a non-prohibited person's speakerphone. It would be unreasonable for criminal consequences to flow in such circumstances.

190 The A-G also suggested that the words "directly or indirectly" be added after "communicate or associate with" at the start of condition 6.1. It is not reasonable to have criminal consequences flow from a concept as vague and uncertain as "indirect communication or association." Accordingly, I replaced the suggested words with "or cause another person to communicate on your behalf with".

191 I also narrowed conditions 6.1 (b) and (c) to persons who have been charged with, or committed of, terrorist-related offences. The A-G's proposed condition would have prevented Mr Benbrika from associating with somebody who was on bail or community release for even the most trivial, non-terrorist related offending, such as shoplifting or a driving offence. There was no evidence that such a sweeping ban was necessary to reduce Mr Benbrika's risk of offending.

192 Condition 6.2 empowers the AFP Superintendent to give written notice of other prohibited persons or classes of prohibited person, where the AFP Superintendent suspects on reasonable grounds that Mr Benbrika's communication or association with that person or class of persons would present an unacceptable risk to the public. That is a reasonable condition.

193 However, for the reason previously given, I included in condition 6.3 a carve out to the AFP Superintendent's powers in respect of Mr Benbrika's immediate family.

194 There was no dispute about condition 7, which prohibits Mr Benbrika from starting employment, or participating in any volunteering, training or education programs, without the prior written approval of the AFP Superintendent.

195 The A-G sought extensive conditions restricting the activities in which Mr Benbrika can engage, some of which were agreed, some were opposed.

196 Condition 8.1 prevents Mr Benbrika from joining clubs, organisations or groups, or forming any corporate or business entities, without the prior written approval of the AFP Superintendent. The only changes I made to condition 8.1 were ones of form, not substance, to make the condition clearer.

197 Condition 8.2 prevents Mr Benbrika from engaging in prohibited activities (such as making a public statement, publishing material, or teaching or mentoring), in relation to a long list of prohibited matters. The prohibited matters include a wide range of matters relating to terrorism and violence.

198 In so far as the A-G had sought to prohibit Mr Benbrika from engaging in the prohibited activities in relation to any subject matter at all, I did not regard that as reasonable. For example, the A-G's proposal would have prevented Mr Benbrika from making a speech at a family wedding or birthday celebration. However, it is reasonable to prohibit him from engaging in the prohibited activities in relation to the prohibited matters.

199 Mr Benbrika proposed that there be an exception to the condition, to permit him to publicly condemn, reject, denounce or disapprove of any of the prohibited matters. I rejected his proposal for the following reasons.

200 First, opinions may well differ as to whether a particular statement is approving or disapproving of a prohibited matter. It is undesirable to impose a condition which is so open to dispute as to whether or not it has been breached.

201 Secondly, Mr Benbrika's process of deradicalisation is ongoing. He should be able to privately discuss the prohibited matters in the course of any deradicalisation program. However, for the duration of this ESO, Mr Benbrika should not be participating in any public discussion about the prohibited matters, which is what condition 8 is largely concerned with. Whether such an exception would be appropriate in any future ESO would depend on the evidence at that time.

202 Condition 9 places limits on Mr Benbrika's ability to engage in financial transactions, including banking, donations, payments and the like. It was largely uncontested.

203 Condition 10 prohibits Mr Benbrika from visiting various places, including prisons, the homes of prohibited persons, airports and ports. The condition was largely uncontested.

204 I accepted Mr Benbrika's proposal that the words "except in the course of transit to another destination via a direct route" should be included at the end of conditions 10.1(c) and (d). Given the location of Mr Benbrika's home, the contents of the electronic monitoring condition (condition 24), and the nature of his risk of offending, the inclusion of the transit exception is reasonable.

205 There was no dispute about the terms of condition 11, which prohibits Mr Benbrika from leaving Victoria or obtaining travel documents without the AFP Superintendent's approval.

206 Condition 12.1 prohibits Mr Benbrika from intentionally doing various things, including possessing, copying, recording or distributing in hard copy or electronic form documents or electronic media that relate to "further prohibited matters". The further prohibited matters themselves are set out in condition 12.2, and are largely uncontentious.

207 I accepted Mr Benbrika's suggested narrowing of condition 12.2(l), having regard to the broad concept of martyrdom in Islamic thought.

208 I also removed the reference to "any method of assault or murder by any means" from the list of further prohibited matters. Such matters are regularly reported in contexts that have nothing to do with terrorism. For example, it is almost impossible to watch the evening news, or read a newspaper, without seeing a story about an assault or murder.

209 The parties agreed in principle that there should be an exception for publications or broadcasts in the mainstream media, and for prohibited material contained in court documents, but disagreed about the terms of the exception. I drafted condition 12.3 so as to permit Mr Benbrika to access material broadcast on or published in mainstream media in Australia, NZ, the USA or the UK, and to access mainstream religious texts. The original version proposed by the A-G was unreasonably restrictive, and likely to lead to the criminalisation of innocuous conduct which is unrelated to the risk of offending.

210 There was no dispute about condition 13, which governs Mr Benbrika's use of mail services.

Telecommunications and computers (conditions 14 to 19)

211 The next six conditions impose extensive restrictions on Mr Benbrika's use of certain technologies, and were all largely undisputed.

212 Conditions 14 and 15 deal with mobile and other phones. Condition 16 deals with internet services, condition 17 with email, condition 18 with computers and other electronic devices, and condition 19 with electronic programs and platforms.

213 I added the word "intentionally" in relevant places in each of the conditions, to avoid criminalising behaviour that may be completely inadvertent.

214 The only other changes I made to any of those conditions were matters of form not substance.

215 Condition 20 grants the AFP Superintendent extensive power to conduct searches to monitor Mr Benbrika's compliance with the ESO, particularly his use of phones and electronic devices. The condition was undisputed.

216 Condition 21 prohibits Mr Benbrika from purchasing, possessing or using any one of a large list of weapons or things that might be used as weapons. There was no objection to the substance of the A-G's proposed condition.

217 Mr Benbrika submitted that, instead of referring to individual sections in relevant legislation (such as the definition sections in the Firearms Act 1996, or the Control of Weapons Act 1990), it would be better to specify every item falling within the relevant definitions. I did not accept that submission, for several reasons. First, it is not the drafting approach that has been adopted with references to legislation in other conditions. Secondly, if the list of prohibited items in the Firearms Act or Control of Weapons Act were to change, then Mr Benbrika's obligation would change accordingly.

218 Condition 22 places restrictions on Mr Benbrika's use of motor vehicles, and was not disputed.

219 Condition 23 prohibits Mr Benbrika from changing his name or identification documents without the approval of the AFP Superintendent. It was not disputed.

220 Condition 24 imposes extensive electronic monitoring conditions on Mr Benbrika. It was not disputed.

Exclusion for mental health treatment and disengagement programs (condition 25)

221 Condition 25 was suggested by Mr Benbrika. It is a sensible provision that will enable Mr Benbrika to discuss what would otherwise be prohibited matters in the course of his mental health treatment or disengagement programs. Discussing and challenging potentially extremist views is an important part of both the therapeutic and disengagement processes.

222 Condition 26 was another sensible proposal by Mr Benbrika. It provides a procedure by which the AFP Superintendent can go through all of Mr Benbrika's existing books and other documents, to determine which ones can and cannot be retained by him in order to comply with the ESO conditions.

223 Conditions 27 and 28 empower the AFP Superintendent to exempt Mr Benbrika from various ESO conditions, and set out the procedures to be followed. Condition 27 deals with exemptions sought by Mr Benbrika. Condition 28 deals with exemptions granted by the AFP Superintendent without a request from Mr Benbrika.

224 No logical reason was given for the A-G's proposal that the AFP Superintendent have more extensive powers in respect of condition 28 than condition 27, so I drafted the conditions accordingly.

225 Condition 29 sets out the process for Mr Benbrika to obtain approval from the AFP Superintendent for various matters under the ESO conditions. It was not disputed.

226 Condition 30 was inserted at the suggestion of Mr Benbrika. It provides a process for review by the court of decisions made by the AFP Superintendent in relation to significant matters under the ESO conditions. A review cannot be sought in respect of the exercise of some of the more minor powers granted under the ESO. But I was satisfied that condition 30 provides an appropriate mechanism for judicial oversight of some of the more important powers, to ensure that the AFP Superintendent is not effectively rewriting the ESO conditions through the approval, direction and notice powers.

227 At the end of the ESO conditions are some statutory notices and definitions, which were not disputed.

228 There were two further proposed conditions that I did not include; one was sought by the A-G, and one by Mr Benbrika.

229 The A-G sought a condition that Mr Benbrika not permit any person other than 13 named members of his family to enter and remain at the specified premises. The A-G also proposed that the AFP Superintendent be given the power to approve other persons or classes of persons who could enter the family home, under the approval process in condition 29.

230 I agreed with Mr Benbrika that the residential visitors condition was not reasonably necessary, nor appropriate and adapted, to prevent an unacceptable risk, for the following reasons.

231 First, Mr Benbrika's wife and four of his children (three of whom were over 18, and one of whom was 17) live in the family home. His other three children are all married, and have children of their own. Mr Benbrika has five grandchildren, only two of whom were listed in the A-G's permitted visitor list. Why his other grandchildren (two of whom were under 12 months old), or any future grandchildren, should not be permitted to visit the family home was not explained by the A-G.

232 Secondly, the proposed condition would have effectively prevented Mr Benbrika's wife and children from having visitors to their own home. It could criminalise a situation where one of their friends dropped around unannounced. The proposed condition would create an unacceptable risk of potentially criminalising conduct that has no connection whatsoever with Mr Benbrika's risk of offending.

233 Thirdly, the proposed condition was unreasonable in a household that comprises a number of adults, all of whom would feel entitled to have their own friends and family visit them. Both sides' experts regard Mr Benbrika's home environment as a protective factor against further offending. It is important that ESO conditions do not have a seriously adverse effect on that protective factor, for example, by causing interpersonal difficulties or unreasonable restrictions on other family members who also live in the family home.

234 Fourthly, the A-G proposed that nobody apart from the 13 named family members could enter and remain in the family home without the AFP Superintendent's prior written approval. The proposed condition would have prevented visits from professional persons such as lawyers or health care visitors. It would have prevented delivery persons and tradespeople from entering the house.

235 The A-G's suggestion that such people could be permitted to visit with the prior written approval of the AFP Superintendent was unreasonable and impracticable. A formal written approval process would not be appropriate to deal with urgent repairs, or even ordinary day-to-day situations (for example, a professional visitor or delivery person may need to change the date of an appointment or delivery, or the name of the actual service provider, at short notice).

236 Finally, the A-G's approach outside the family home was to prevent Mr Benbrika from associating with particular people and groups of people, rather than the reverse approach of permitting association only with specified people. No good reason was put forward by the A-G for not adopting the same approach inside and outside Mr Benbrika's house. The prohibited associations condition is drafted broadly enough to preclude Mr Benbrika from associating with prohibited persons both inside and outside his family home. The AFP have the capacity to monitor who enters and exits the Benbrika family home. If any of the prohibited persons visits Mr Benbrika at home, the AFP have the means to ascertain that.

237 For those reasons, I concluded that the proposed residential visitors condition was not reasonable to prevent unacceptable risk. That was particularly so when regard is had to the combined effect of all the other conditions that I imposed.

Conduct that would, but for this clause, constitute contravention of an ESO condition, including any direction, shall not be a contravention of the order or direction where the conduct engaged in by Mr Benbrika is or was minor, insignificant or inconsequential.

239 At first blush, such a provision might seem attractive, to avoid criminalising minor breaches of conditions.

240 However, the expressions "minor, insignificant or inconsequential" are imprecise. The expressions are also inherently evaluative, and necessarily require the drawing of conclusions that may be highly contestable. Because non-compliance with an ESO condition attracts criminal liability, an ESO must be certain, and understood by those who are subject to it, as well as those who enforce it.

241 By imposing the various requirements of "intention" or "knowledge" that I have inserted into many of the ESO conditions, I have sought to prevent criminal consequences flowing from inadvertent breaches.

242 For those reasons, I was not prepared to make the proposed contravention condition sought by Mr Benbrika.

243 Most of the ESO conditions were not disputed at all, or were only the subject of minor dispute.

244 There were a few conditions, or parts of conditions, proposed by the A-G that I did not allow, on the ground that they were not reasonably necessary or reasonably appropriate and adapted for the purpose of protecting the community from the unacceptable risk that Mr Benbrika presented. They were not reasonable even if the unacceptable risk was regarded as "moderate-low" (as assessed by Dr Dewson) rather than something lower than that (as assessed by Dr Speckhard).

245 The next matter I will consider is the serious, and extremely concerning, matter of non-disclosure of exculpatory material by the AFP Minister.

246 The A-G filed a number of affidavits in relation to the non-disclosure of the Corner report. Before I discuss their contents, I will make some observations about the adequacy of the affidavit evidence.

247 The affidavits of the AGS lawyers and counsel satisfactorily explained the deponents' personal involvement in, and knowledge of, relevant matters, and acknowledged how things could, and should, have been done differently. I have no doubt that it must have been uncomfortable for those particular deponents to publicly admit any mistakes they may have made, and I accept their apologies.

248 Emily Nance, Deputy Chief Solicitor at AGS, spoke to two former AGS lawyers who had worked on the CDO proceeding, but had since left the AGS. Ms Nance informed the court of their recollections in an affidavit dated 3 February 2023.

249 In his affidavit dated 6 February 2023, Michael Kingston, the Australian Government Solicitor, acknowledged that the non-disclosure of the Corner report was an error, as it was clearly a document that needed to be disclosed. Although Mr Kingston had no involvement in the Benbrika proceedings, his position that, as AGS, he took responsibility for the non-disclosure, was an honourable one for him to take. However, he does not, and cannot, address Home Affairs' critical role in the non-disclosure, which has not been satisfactorily explained to the court.

250 Home Affairs took a completely different approach to giving evidence. No affidavit was filed on behalf of anybody in Home Affairs who had personal knowledge of any of the relevant meetings, communications or decisions. Instead, affidavits dated 6 February and 24 April 2023 were affirmed by Stephen Webber, First Assistant Secretary, Disputes and Corporate Law Division in Home Affairs. Mr Webber first became aware of the Corner report in November 2022, and had no involvement in, or personal knowledge of, the non-disclosure issue.

251 Mr Webber's first affidavit spent many pages describing the documentary searches that had been undertaken, including listing page after page of individual keyword search terms that had been used. He also asked three senior Home Affairs officers, who have no apparent relevance to the non-disclosure issue, about their knowledge of the Corner report. In his second affidavit, he listed more than a dozen people who attended a critical meeting on 18 August 2020, at which he claimed that the AGS gave legal advice that the Corner report did not need to be disclosed. There is no document which records who within Home Affairs made the decision not to disclose, or the reasons for that decision.

252 Mr Webber offered an explanation for how the non-disclosure occurred that was based entirely on assumptions he made, based on his examination of departmental records. There is clearly more to the story than the documents disclose. But Mr Webber did not speak to a single current or former Home Affairs officer who might have been able to tell him what actually happened. I do not accept Mr Webber's apology on behalf of Home Affairs, based as it is on grossly inadequate enquiries, and the bald assertion that non-disclosure was simply a "system failure".

253 Given that Home Affairs has been less than transparent in explaining why the Corner report was not disclosed, and there has been no explanation given as to why the four other expert reports were not disclosed, my ability to come to certain conclusions about the non-disclosure is limited to some extent. However, there is sufficient evidence to make the conclusions and observations that I make below.

254 The Corner report was commissioned by Home Affairs. Work on the Corner report began during 2018. Various Home Affairs officers were involved in briefing and communicating with the authors during the report's preparation.

255 By late 2019, Home Affairs was well aware that the ANU study "could potentially cause concern", because it was likely to identify significant problems with the VERA-2R and RADAR tools. The Home Affairs officers who were aware of that at the time included John Cianchi, a senior executive within the Centre for Violent Extremism ("CVE"), as well as internal Home Affairs lawyers, including Annabel Knott (Senior Legal Officer, National Security Legal Section), and Logan Tudor, all of whom will feature in the subsequent discussion about non-disclosure.

256 CVE received a draft of the report in mid-May 2020.

257 As soon as Home Affairs saw the draft, they realised that it was indeed highly critical of VERA-2R and Radar, and could present serious problems for the Minister in post-sentence order applications. Considerable effort went into deciding how to deal with the Corner report, and to minimise the clear problems it presented to the continuing use of VERA-2R, which was described in internal documents as the department's "preferred structured professional judgement tool."

258 Home Affairs officers met with Dr Corner in late May or early June, to discuss the department's concerns about the draft report.

259 On 12 June 2020, Alexander Engel, the Director of the CVE, wrote to the authors, asking them to make a number of changes to the draft report. Mr Engel said that the suggestions were intended to "contextualise" things, "so that the finding can be better understood by readers without a research background." After stating that Home Affairs valued the academics' independence, and was not seeking to amend their findings, Mr Engel went on to make to a number of suggestions for rewriting the report in a way more favourable to Home Affairs. The following examples are sufficient to illustrate the nature of the department's concerns:

Executive summary. This is the section of the report most readers will visit. We would be grateful if you would revisit the writing tone. Some decision makers might misinterpret your findings as they are currently expressed and conclude that the Australian Government should dispense with both tools, leaving law enforcement and correctional agencies with only operational experience.

Dot points 4 and 5, represent important elements of your research findings, but we are concerned that the finding as they are expressed suggests the instruments are not fit for purpose. We invite you to reconsider how to express your finding here and in the body of the report ...

Dot point 7 as it is currently worded might lead to the rejection of both instruments in Australia, as decision makers may misunderstand your research parameters. We request this finding be clarified ...

260 Mr Engel's comments demonstrated a clear awareness within Home Affairs of the serious problems that findings in the Corner report could pose for the continued use of VERA-2R and Radar in court proceedings.

261 The authors of the Corner report made a number of amendments to the report at the department's request, and delivered the amended final report on 25 June 2020. On 29 July, Mr Engel thanked Dr Corner for making the changes, and said that Home Affairs now accepted the report as final.

262 The final Corner report was received three months before the CDO proceeding was commenced, and six months before the trial in that proceeding.

263 Because of the damaging nature of the Corner report, it seems that a decision was made to limit the people to whom the draft or final Corner reports would be provided. The top of every page of the report said, "The content of this report may be shared within agencies but is not to be reproduced or distributed further." Why and when that note was added is not clear.

264 It was also decided not to inform the experts retained by the Minister in the various proceedings against Mr Benbrika of the contents of the report, or to provide them with a copy of it. Had the experts been aware of the report, they would have been obliged to disclose its existence to the court, under the Supreme Court's expert witness code of conduct.

265 The Corner report was not widely circulated within Home Affairs itself, and was treated as a sensitive document when it was sent to people, as the following discussion shows.

266 Dr Corner had sent Dr Cianchi, Mr Engel and two other Home Affairs officers a copy of what she described as "the amended final report" on 25 June. For reasons that have not been explained, Dr Cianchi was still telling internal departmental lawyers in July 2020 that Home Affairs only had a draft report, and the final was expected "in a couple of months".

267 By mid-2020, internal Home Affairs lawyers, AGS lawyers, and counsel retained on behalf of the Minister, were all well aware that the operation and reliability of the VERA-2R tool could be subject to challenge in the proposed CDO proceeding, and were discussing strategies to defend it. They were concerned about problems with the VERA-2R tool, even before they learned of the criticisms in the Corner report.

268 By early July, some internal Home Affairs lawyers were aware that the department had received the Corner report, and that it was not favourable to Home Affairs. That led Ms Knott to note in an email dated 8 July to Home Affairs lawyers Logan Tudor and Kerrie Cowan that "It may be that we don't want to know about it... it's something to consider."

269 On 9 July, Ms Knott write to Dr Cianchi, asking for a copy of what she understood was still a draft of the report. Ms Cowan and Mr Tudor were copied in to Ms Knott's email.

270 A few minutes later, Ms Cowan emailed Ms Knott in the following terms:

It occurs to me that this draft may be information held by the Cth which goes against the order being made which will have to be included in the application. I do not think this changes whether John C gives it to us or not as it is held by the Department - unless of course it is held by John in his personal capacity? Can you just follow up with John if the report is the Department's or his personal document. If it is the Department's I think we cannot guarantee that will not be used in the application due to its exculpatory nature ...

271 Ms Cowan was the Principal Legal Officer in the National Security Legal section of Home Affairs, and Ms Knott's superior. Ms Cowan's immediate instinct that the report was exculpatory, and might be disclosable, was absolutely correct. Her suggestion that a lengthy academic report, commissioned and paid for by Home Affairs, might somehow be treated as being Dr Cianchi's personal document, demonstrates just how damaging she understood the Corner report would be to Home Affairs if it had to be disclosed to Mr Benbrika.

272 About an hour after receiving Ms Cowan's email, Ms Knott emailed Rachel Deane, the most senior AGS lawyer working on preparation for the CDO proceeding. Ms Knott replied to a number of queries about VERA-2R that Ms Deane has raised the previous day. Ms Knott mentioned the draft Corner report, and noted that Ms Cowan had "raised the possibility that this draft may contain exculpatory material (for disclosure)".

273 On 10 July, Dr Cianchi sent Ms Knott a copy of what he described as "the draft report" she had requested. Even though he had personally received the final report almost a fortnight earlier, Dr Cianchi told Ms Knott that "we expect the final by the end of July." He also expressed his concern that "a reader might misinterpret the findings if they do not understand social research methods and writing, or take the findings out of context." He then told Ms Knott that the report was a "confidential and sensitive product prepared for Home Affairs", and asked her not to share it beyond her internal colleagues and the AGS.

274 Mr Cianchi's email, with the attached Corner report, was forwarded to Ms Deane and a number of other Home Affairs personnel a few hours later. Permission was later given to provide the Corner report to counsel as well.

275 On 12 July, Ms Deane emailed Mr Tudor and Ms Knott about the Corner report. Her comments included the following:

I am concerned that this paper identifies some vulnerabilities in the VERA-2R particularly the view expressed that the tool is unable to predict the specified risk with anything other than chance. I am concerned not because I think that this makes the report "disclosable" to Mr Benbrika. More fundamentally, I am concerned to ensure that DHA, AGS and the counsel team are aware of these vulnerabilities and that we do what we can to ensure that in preparing our case we do not expose ourselves by ignoring those vulnerabilities.

276 On Monday 13 July, there was a regular weekly teleconference between Home Affairs and AGS lawyers working on the Benbrika matter. There was discussion about the Corner report, and how to deal with its various criticisms of VERA-2R.

277 Nobody from Home Affairs who was present at the meeting has given their account of the meeting. However, the minutes of the meeting recorded the following in relation to the possible disclosure of the Corner report (the typographical and grammatical errors are in the original):

Possible the report may be exculpatory information: AGS can think about that. AGS thinks exculpatory information of actual eviedence/matieal/evidence of government, versus a reaserach paper which is critical of a risk assessment tool, AGS does not think falls in that body. Will think about it ..."would reasonable be considered going against order being made"... AGS does not think the paper in itself is readoably be reviewed why an order should not be made. We want to be armed, if not includd in statement of facts for why order be made, sort position why not included in that statement. WE don't have strong views, other than if decide not to include it, thing it through, and be prepared to respond to criticaism. Possible an FOI/summpoena may capture this and may be disclosred. WE can't assume will remain unknown. We could er on side of caution and disclose it. We underatsnd CV team controlling if is/is not made public. If line ball call if needs to be disclosed, if readonble aruments to not discloused, may want to ere onnot to support policy team's requiremts in terms of broader CVE program. A few consideration whether proactive to disclose or not.

278 Anna Lord was the member of counsel who was primarily involved in relation to briefing the Minister's expert witnesses on risk assessment. She first learned of the existence of the Corner report in a telephone call with Ms Deane on 13 July, after the teleconference. Shortly after the phone call, Ms Deane sent a copy of the Corner report to Ms Lord, under cover of a lengthy email which summarised some of its contents. The final paragraph of the email said:

Finally, the client has asked whether this research report is "disclosable" on the basis that it might reasonably be seen as supporting a finding that the CDO should not be made. I don't myself think that this report falls into that category, but we should think about that further.

279 Ms Lord had no recollection or record of reading the email or the Corner report on the 13th.

280 Andrew Berger, senior counsel for the Minister, was one of several people who were copied into Ms Deane's email to Ms Lord. IT records show that Mr Berger did open the email, but he had no recollection of opening or reading the email or the attachment. Given that the lengthy email was not addressed to him, and the sheer volume of emails he was receiving about the Benbrika matter at the time, it is understandable that the documents would not have registered with him. The Corner report was never raised again with Mr Berger, and he was never asked to advise in relation to its disclosure.

281 On 14 July, Ms Deane emailed Ms Lord a "draft explanation of the VERA-2R tool", prepared by somebody in Home Affairs, to assist in briefing counsel and advising the court about VERA-2R.

282 The draft explanation document arrived shortly before a teleconference on 14 July, between Ms Deane, Ms Lord, Dr Cianchi and other Home Affairs officers. Most of the discussion involved Dr Cianchi briefing Ms Dean and Ms Lord on how to respond to possible criticisms of VERA-2R. Ms Knott's file note of the teleconference recorded Ms Deane as having said towards the end of the teleconference:

Obligation - to discover anything in our possession which might reasonably be viewed as going against the making of a CDO.

Initial impression - doubt academic research paper would fall into this category - but it's something we will discuss with Counsel and advise the Dept as to whether it's something that's discoverable.

283 Ms Lord recalled participating in a conference with Dr Cianchi about a report concerning VERA-2R, at a relatively early stage of her involvement in the matter. However, she had no independent recollection of the conference, or what was discussed at it. There is no evidence that the Corner report was ever raised again with Ms Lord after that teleconference. Like Mr Berger, Ms Lord was never asked to advise (formally or informally) in relation to the disclosure of the Corner report.

284 Running task lists and agendas were prepared for discussion at the weekly teleconferences between the AGS lawyers and Home Affairs lawyers involved in preparing the CDO proceeding. Items were added to the documents from time to time. Items were struck through, but left on the documents, once they had been attended to. In July and August, the agendas continued to include the following:

AGS/Counsel consideration and advice to Dept on whether ANU paper is exculpatory. To be considered at full counsel meeting/counsel closer review of doc.

285 Given how potentially damaging the Corner report was to Home Affairs, and the serious consequences of making an incorrect decision about its disclosability, it is unsurprising that the department and AGS proposed to obtain counsel's advice on whether or not it had to be disclosed. That would have been an obvious course for any responsible litigant - especially one with an obligation to act as a model litigant - to take. Counsel's advice was never sought.

286 Alternatively, a responsible model litigant might have requested formal written advice from the AGS or their own in-house lawyers. Home Affairs did not do that either. Instead, Mr Webber asserted that Home Affairs decided to act on Ms Deane's informal, unconsidered, oral opinion that the Corner report was not disclosable. Nobody in Home Affairs with actual knowledge of such a highly unusual decision has explained when, why or by whom the decision was made not to disclose the Corner report.

287 There was a regular weekly teleconference on 18 August. The task list for discussion at the teleconference included the following item: "ANU review paper on VERA2R - view on whether this is exculpatory [AGS]." No explanation has been given for removing the numerous earlier references to seeking counsel advice.

288 Later that very same day, Ms Knott updated the item in the task list as follows:

18 Aug - AGS confirmed previous advice to Dept that ANU paper is not exculpatory. Referred to previous discussions with John Cianchi.

289 According to the second Webber affidavit, 12 named people and one unnamed person dialled into the teleconference that day. They included Ms Cowan, Ms Knott, Ms Deane, and other Home Affairs and AGS lawyers.

290 The only one of those 13 people who has gone on oath as to what happened at the meeting is Ms Deane. She had no independent recollection of the meeting.

291 The AGS file note of the 18 August meeting did not record there being any discussion at all about the disclosability of the Corner report.

292 For the purpose of explaining the non-disclosure to this court, Home Affairs provided the AGS with a copy of the amended task list referred to above, which said that "AGS confirmed previous advice to Dept that ANU paper is not exculpatory." Ms Deane's evidence in relation to that amended task list was:

I have no recollection of confirming that advice on or around 18 August 2020 and I am informed that AGS has no record of me having considered the issue at that time or of confirming previous advice about the issue.

293 The Nance affidavit deposed that two of the other AGS lawyers who attended the 18 August meeting, Elle Addams and Rebecca Verdon, have no recollection of ever attending any meeting at which the Corner report was discussed.

294 Logan Tudor was the Home Affairs officer who usually took detailed notes of the weekly teleconferences with AGS. He was unwell on 18 August, and did not attend the teleconference. According to paragraph [11] of the second Webber affidavit, nobody else from Home Affairs took "a detailed file note" of what occurred at the teleconference. Whether there are any "less than detailed" notes has not been disclosed.

295 Mr Webber deposed that he knew who attended the meeting, and that no detailed file note was taken, because Ms Cowan told him so. Mr Webber apparently communicated with Ms Cowan about those administrative matters, but not about anything of substance, such as: what was discussed at the 18 August meeting, or when and why did Home Affairs decide not to disclose the Corner report?

296 Ms Cowan was the most senior Home Affairs lawyer present at the 18 August meeting. Her immediate impression that the Corner report was exculpatory and disclosable was absolutely correct. Ms Cowan still works in a senior legal position in Home Affairs; there was no suggestion she was unavailable to swear an affidavit in these proceedings. If it was in fact the case that Home Affairs decided not to disclose the Corner report because of oral advice received from Ms Deane at that meeting, it would have been very simple for Ms Cowan (or Ms Knott, or any of the many other Home Affairs officers who attended the meeting) to have provided an affidavit to that effect. Alternatively, they could have informed Mr Webber of their recollections, so that he could depose on a hearsay basis. That did not happen.

297 Instead, affidavits were provided by somebody with no personal knowledge about anything relevant. The Webber affidavits are deafening in their silence as to what, if anything, any of the relevant Home Affairs officers remember about the 18 August meeting, or about the decision not to disclose the Corner report. I infer from that silence either that:

(a) Mr Webber deliberately chose not to ask them for their recollections; or

(b) Mr Webber has asked some or all of them, and their answers do not assist Home Affairs.

298 After conceding that both of those inferences were open to be drawn, senior counsel for the A-G submitted that there was a third possible inference, namely, that none of the Home Affairs officers could remember the relevant events. I reject that submission. Had that been the case, it would have been extremely advantageous to Home Affairs (who could then have simply asserted the accuracy of the amended task list without fear of contradiction), and there would have been no sensible reason for Mr Webber not to have put that in his affidavits.

299 I do not accept the proffered "explanation" or "apology" from Mr Webber on behalf of Home Affairs. They are both utterly disingenuous, in circumstances where Mr Webber (or others in the department) has deliberately chosen not to put a proper explanation before the court.

300 Not only was the Corner report not produced to Mr Benbrika in the CDO, CDO appeal or first review proceedings, it was not even produced to him after the INSLM started asking difficult questions about its non-production during 2022.

301 Between September 2020 and November 2021, Richard Feakes, First Assistant Secretary and Deputy Counter-Terrorism Coordinator, affirmed five affidavits in the CDO and first review proceedings, including three affidavits which asserted that the AFP Minister had fully complied with the disclosure obligation. Mr Feakes has deposed in affidavits dated 3 and 6 February 2023, filed in these proceedings, that he first became aware of the Corner report in 2022. The very person whose role it was to verify on oath that all exculpatory material had been disclosed to Mr Benbrika seems to have been kept in the dark about the existence of the Corner report (and all of the other critical expert reports that were finally disclosed in the current proceedings).

302 Mr Benbrika urged me to make a number of adverse findings against various Home Affairs lawyers, and to conclude that the Corner report was improperly withheld as a result of a deliberate Home Affairs decision.

303 Mr Webber's assertion that Home Affairs decided not to disclose a highly damaging report because of informal, unconsidered, oral advice, given by one AGS lawyer at the 18 August teleconference, which was never subsequently mentioned in any file note, email or letter, seems highly improbable for so many reasons.

304 However, given the state of the evidence before me, I am not in a position to make the adverse findings suggested by Mr Benbrika. Nor was it necessary for me to do so, in order to decide the disputed issues in these proceedings. However, I will be referring the matter to the current INSLM, for him to investigate further, should he think it appropriate to do so.

305 One week before the trial in these proceedings was due to start, the A-G disclosed the following three reports, all of which were prepared by Victoria University:

(a) A report entitled "Validation of National CVE & CT Tools: RADAR and ANZCTC Counter-Terrorism Persons of Interest Prioritisation Tool". The report was prepared for Victoria Police, and is dated May 2020;

(b) A report entitled "Validation of the RADAR Countering Violent Extremism (CVE) Intervention Tool". The report was prepared for Victoria Police and the Department of Justice and Community Safety (Vic), and is dated November 2020; and

(c) A report entitled "Revisiting Risk: An Assessment of How Terrorism Subjects Make the Transition to Violence". The report was prepared for Victoria Police and the Australia-New Zealand Counter-Terrorism Committee, and is also dated November 2020.

306 Each of the VU reports compared VERA-2R with other violent extremism assessment tools, to assess the validity of those other tools. They noted that VERA-2R captured many features of violent extremism that were captured in other tools, but commented that there was a lack of published results and peer-reviewed evidence on the validity and reliability of VERA-2R.

307 Specific criticisms of VERA-2R included that the revisions to VERA had made it ambitious in its attempts to assess all aspects of violent extremism. The first VU report noted that in attempting to be versatile in its use across ideology, gender and age, the tool may have reduced its clarity and utility. To assess individuals would require such a rich source of detail to accurately address each indicator that it may be impractical and inefficient. However, the report noted that VERA-2R did explicitly consider protective factors, which was supported by literature. The second and third VU reports noted that VERA-2R included some items that could be marked positive for multiple reasons, potentially hiding patterns between items and reducing internal validity. They noted that VERA-2R needed to be more rigorously analysed for reliability and validity.

308 After the close of evidence, the A-G produced a further report entitled "Countering Violent Extremism: Critical Review of Risk Assessment Tools" ("the DST report"). The DST report was prepared by the Joint and Operations Analysis Division of the Defence Science and Technology Group within the Department of Defence, and is dated December 2019.

309 The DST report contained a comparative analysis of six risk assessment tools for violent extremists, including VERA-2R. After noting that VERA-2R had experienced comparatively more validation efforts than its counterparts, DST described the tool as "largely a-theoretical" and noted that "the authors' various claims of validity and reliability are not supported by published results and peer-reviewed evidence."

310 The four further expert reports are clearly exculpatory. Unlike the Corner report, Home Affairs was not the commissioning party for those four reports. That does not mean that the Minister for Home Affairs did not have a duty to disclose the reports. Even if Home Affairs had received the relevant reports on a confidential basis, they should have been disclosed in the earlier proceedings, subject to any public interest immunity claim that might have been made by Victoria Police, the Department of Defence, or any other interested party.

311 It is clear from the evidence that at least two of the VU reports and the DST report were in the possession of the Minister for Home Affairs by November 2020, prior to the trial in the CDO proceeding. They were clearly disclosable in the CDO and first review proceedings, on the same basis as the Corner report.

312 The remaining VU report would also have been disclosable, if it was in the possession of the Minister at the time of those earlier proceedings.

313 Because those four reports were produced to Mr Benbrika and the court so late in the current proceedings, there was no opportunity to examine when, or in what circumstances, they came into the possession of Home Affairs, or why they were not disclosed in the earlier proceedings.

314 It is totally unacceptable that those further reports were produced so late in the current proceedings, in circumstances where the non-disclosure of expert material critical of VERA-2R was clearly of significant concern to both Mr Benbrika and the court, and was raised regularly at directions hearings.

315 The statutory requirement that the AFP Minister disclose exculpatory material is a fundamental safeguard to ensure the protection of individual liberty under what is very unusual and draconian legislation.

316 What happened in this case should never have happened, and should not be repeated in the case of Mr Benbrika or any other person the subject of a post-sentence order application.

317 The non-disclosure of the various expert reports amounts to a serious interference with the administration of justice.

318 Mr Benbrika did not argue that the AFP Minister's breach of the statutory disclosure obligation should disentitle him from obtaining a post-sentence order at all. The non-disclosure has also not affected the contents of any of the ESO conditions that I ordered. However, it had some relevance to the length of ESO that I ordered, for the reasons discussed below.

319 I am going to refer the non-disclosure of the Corner report and the four other expert reports to the current INSLM, and provide him with a copy of these reasons and all of the relevant evidence. It will be for him to decide what, if any, further investigations he wants to make, to get to the bottom of why the various expert reports were not disclosed to Mr Benbrika.

320 It seems from the INSLM report that the AFP Minister has engaged in similar non-disclosure in relation to other terrorist offenders. The wider implications of the AFP Minister's conduct in terrorist proceedings are a matter more appropriately considered by the INSLM than by a court hearing an individual application.

321 I accept the accounts offered by the AGS solicitors and counsel retained on behalf of the AFP Minister as to their respective roles in the non-disclosure. In each case, they have adequately explained what happened and why, and have demonstrated an understanding of what they should do differently in such a situation in the future. I see no basis for referring any of them to a professional disciplinary body.

322 Many of the Home Affairs officers involved in the non-disclosure are lawyers and officers of this court. Ms Cowan and Ms Knott are both senior lawyers, who have not exactly demonstrated professional excellence in their handling of the Corner report. However, given the totally inadequate evidence put forward by Home Affairs, and the likelihood that there are other senior officers in the department who played a significant role in the decision to withhold the various reports from Mr Benbrika, it would not be fair for me to make any findings of wrongdoing in relation to specific departmental lawyers. If any of them have been involved in improperly withholding exculpatory material, that may have professional disciplinary consequences for them. However, I do not propose to make any referral to the relevant professional disciplinary body, given the incomplete evidence before me. Whether the INSLM chooses to make any such reference after making further enquiries will be for him to decide.

323 The final dispute between the parties related to the duration of the ESO. The A-G argued that the ESO should be for three years, Mr Benbrika argued that it should be for no more than one year.

324 I determined to make the ESO for 1 year, for the following reasons.

325 The first (and primary) reason related to the issue of risk assessment. This will be the first time that Mr Benbrika will have lived in the community in almost 20 years. Both he and the community have changed in various ways over the past two decades. Much of the expert assessment of how Mr Benbrika will behave out of prison was necessarily based on predictions that may or may not come to pass. Dr Speckhard believed that if Mr Benbrika continued to make good progress in his deradicalisation work with the Sheikh over the next 12 months, that may have a significant impact on whether the ESO should be continued at all and, if so, on what terms. Dr Dewson was less optimistic about what a difference 12 months will make. I generally preferred Dr Speckhard's evidence to that of Dr Dewson. However, only time will tell what Mr Benbrika's risk is at the end of his first year out of prison. In the meantime, the community will be protected by the strict ESO conditions. The A-G did not persuade me that it was necessary to make the ESO for three years.

326 The second reason arose out of the AFP Minister's non-disclosure of exculpatory material. There is no doubt that the administration of justice has been adversely affected by the non-disclosure of the Corner report, and the four other expert reports critical of VERA-2R. The accepted validity of VERA-2R was absolutely fundamental to the decisions to impose and continue the CDO. It is not possible for me to say now what the outcome of the CDO, CDO appeal and first review proceedings would have been had the five expert reports been disclosed in them. However, the earlier proceedings have all been tainted to varying degrees by the non-disclosure.

327 In addition, the late provision of so much relevant material, so late in the two current proceedings, prevented this court from properly considering the underlying validity of VERA-2R. However, because of the limited nature of the dispute between the parties about the proposed ESO conditions, and the evidence before me, I was able to set appropriate conditions without having to make any findings as to VERA-2R's underlying validity.

328 Whether there will be greater dispute as to the imposition or conditions of any future ESO is presently unknown. It may be that the judge at any future hearing will need to make such validity findings.

329 It would have been unfair for me to have imposed an extremely onerous ESO for three years, in the circumstances. Had I done so, the AFP Minister would have continued to have the benefit of the improper non-disclosure, and Mr Benbrika would have borne the onus of persuading a future judge to reduce the ESO's strictness or duration. It will be fairer if the judge hearing the matter at the end of 2024 starts with a clean slate when considering what, if any, ESO conditions would be appropriate at that time, having regard to all relevant material.

330 Finally, given the Code requirement for annual reviews of post-sentence orders, there is going to need to be another substantial court hearing in 12 months' time in any event. That proceeding will involve further expert assessments, disclosure of exculpatory material, provision of affidavits and submissions. Affidavits which have been filed in the current or earlier proceedings can continue to be referred to, if they are relevant. The parties will need to perform essentially the same amount of work, irrespective of whether the hearing at the end of 2024 is an application for a fresh ESO or a review of the current ESO. Requiring the AFP Minister to make an application for a further ESO should be no more burdensome from a financial or resource perspective than a review of an existing ESO.

331 For all of those reasons, I ordered that the ESO be in force for one year, expiring on 19 December 2024.

332 Appended to these reasons is a copy of the ESO and ESO conditions that I made, excluding confidential annexure B.

[1] R v Benbrika & Ors [2009] VSC 21, Bongiorno J ("the sentencing remarks").

[2] Minister for Home Affairs v Benbrika [2020] VSC 888.

[3] Benbrika v Minister for Home Affairs [2021] VSCA 303 ("the CoA reasons").

[5] Minister for Home Affairs v Benbrika [2022] VSC R169 ("the first review reasons").

[7] Minister for Home Affairs v Benbrika (2021) 272 CLR 68, [12].

[8] Pursuant to the Independent National Security Monitor Act 2010 (Cth), s 24.

[9] The INSLM repeated those criticisms in his final report dated 3 March 2023. He also noted that VERA-2R had not been developed for use in making decisions about whether someone should be detained.

[10] Pursuant to s 36D(1) of the Australian Citizenship Act 2007 (Cth).

[11] Pursuant to s 35(3) of the Migration Act 1958 (Cth) ("the Migration Act").

[12] Pursuant to s 501(3) of the Migration Act.

[13] Benbrika v Minister for Home Affairs & Anor [2023] HCA 33.

[16] Benbrika opening submissions, [14]-[16].

[17] Nigro v Secretary to the Department of Justice [2013] VSCA 213; (2013) 41 VR 359, [6], [165], [167].

[18] Minister for Home Affairs v Benbrika [2021] HCA 4; (2021) 95 ALJR 166, [47] (Kiefel CJ, Bell, Keane and Steward JJ).

[20] CoA reasons, [73]-[74], [79].

[21] Sentencing remarks, [40], [42].

[23] Attorney-General (Cth) v Sa'adat Khan (No 2) ("Sa'adat Khan") [2022] VSC 687, [41]; Thomas v Mowbray [2007] HCA 33; (2007) 233 CLR 307, [19], Gleeson CJ.

[24] Thomas v Mowbray, op cit, [22], Gleeson CJ.

[25] Sa'adat Khan, [43], citing New South Wales v BP (No 2) [2019] NSWSC 806, [11].