Woody Carrigan, R (on the application of) v Secretary of State for Justice

This is a case about when, in law, the Secretary of State for Justice (“SSJ”) can reject a Parole Board recommendation to transfer an indeterminate sentence prisoner (“ISP”) to open conditions. The Claimant is aged 46. On 3 October 2006 at Leeds Crown Court, under his given name Dean Wood, he was sentenced to life imprisonment having pleaded guilty to murder. His minimum term tariff expired on 8 February 2019. On 22 June 2022, a Panel of the Parole Board held an oral hearing to consider his suitability for (a) release on licence or alternatively (b) transfer to open conditions. Panel directions for release on licence are binding and the Board is the primary decision-maker. Panel recommendations for transfer to open conditions are advisory and the SSJ is the primary decision-maker. Release decisions and transfer decisions both involve questions of risk-assessment and risk-management. The SSJ puts forward subject-matter experts to write reports for the Panel. Other witnesses can give evidence. Witnesses can appear to give oral evidence, answer questions and respond to information and opinions from one another.

In this case the Panel was David Gravells (independent chair) and Fiona Ainsworth (Psychologist member). The Panel received reports and heard oral evidence from the Claimant and from these expert witnesses: Zoe Wright (Prison Psychologist); Debbie Hogan (Prison Registered Mental Nurse); Sally Morris (Independent Psychologist); Donna Brown (Community Offender Manager); and Mr. Rob Jowsey (Prison Offender Manager). The Claimant was legally represented. The Board issued a 20-page Determination on 25 July 2022. It decided not to direct release. It went on to decide to recommend transfer to open conditions.

The question whether to follow that recommendation of transfer to open conditions passed to the SSJ. It was dealt with by Julia Whyte (Head of Parole Eligible Casework within the Public Protection Group of His Majesty’s Prison and Probation Service) and ultimately Gordon Davison (Head of the Public Protection Group). In an Open Recommendation Proforma document, Ms Whyte set out a two-page summary of the views which the Panel had received from Ms Brown, Mr Jowsey, Ms Wright and Ms Morris; together with the Panel’s views. She went on to set out her two-page reasoned recommendation (7 March 2023), that the Board’s recommended transfer to open conditions be rejected. The decision maker was Mr Davison. He approved Ms Whyte’s recommendation, on 10 March 2023, by endorsing the Proforma and adding his own observations. A 6-page Decision Letter was drawn up, dated 29 March 2023. It set out the SSJ’s reasoned decision to reject the Panel’s recommendation. The Decision Letter set out background facts, the Statutory Directions, the Panel’s conclusions, the law and policy background, and then the outcome of the SSJ’s decision with reasons. The issue which I have to decide is whether the SSJ’s decision was lawful.

SSJ rejection of a Parole Board recommendation to transfer an ISP to open conditions has been given careful consideration in a long series of judicial review cases. In R (Sneddon) v SSJ [2023] EWHC 3303 (Admin) [2024] 1 WLR 1894, I referred (at §10) to a line of ten cases to which the parties in that case had drawn my attention. In the present case, further cases were drawn to my attention. Some were decided before Sneddon; some decided after it. Two post-dated the substantive hearing and were the subject of written submissions. I am, as always, very grateful to the legal teams for the assistance given to the Court.

The upshot is that I have in the present case had the following portfolio of cases to consider. R (Banfield) v SSJ [2007] EWHC 2605 (Admin) (10.10.07, Jackson J); R (Hindawi) v SSJ [2011] EWHC 830 (Admin) (1.4.11, DC); R (Adetoro) v SSJ [2012] EWHC 2576 (Admin) (26.9.12, HHJ Gilbart QC); R (Wilmot) v SSJ [2012] EWHC 3139 (Admin) (9.11.12, King J); R (Gilbert) v SSJ [2015] EWCA Civ 802 (23.7.15, CA); R (Kumar) v SSJ [2019] EWHC 444 (Admin) [2019] 4 WLR 47 (28.2.19, Andrews J); R (John) v SSJ [2021] EWHC 1606 (Admin) (14.6.21, Heather Williams QC); R (Stephens) v SSJ [2021] EWHC 3257 (Admin) (2.12.21, Whipple J); R (Oakley No.1) v SSJ [2022] EWHC 2602 (Admin) [2023] 1 WLR 751 (17.10.22, Chamberlain J); R (Wynne) v SSJ [2023] EWHC 1111 (Admin) (11.5.23, Steyn J); R (Green) v SSJ [2023] EWHC 1211 (Admin) (22.5.23, Sir Ross Cranston); R (Zenshen) v SSJ [2023] EWHC 2279 (Admin) (15.9.23, Dexter Dias KC); R (McKoy) v SSJ [2023] EWHC 3047 (Admin) (1.12.23, UTJ Elizabeth Cooke); R (Overton) v SSJ [2023] EWHC 3071 (Admin) (7.12.23, Eyre J); R (Sneddon) v SSJ [2023] EWHC 3303 (Admin) [2024] 1 WLR 1894 (21.12.23); R (Oakley No.2) v SSJ [2024] EWHC 292 (Admin) (14.2.24, HHJ Keyser KC); R (Cain) v SSJ [2024] EWHC 426 (Admin) (29.2.24, Calver J); R (Uddin) v SSJ [2024] EWHC 696 (Admin) (27.3.24, HHJ Walden-Smith); and R (McPhee) v SSJ [2024] EWHC 1247 (Admin) (23.5.24, HHJ Keyser KC). I can also add these two more recent cases: R (Valentine) v SSJ [2024] EWHC 1534 (Admin) (20.6.24, HHJ Carmel Wall); and R (Hahn) v SSJ [2024] EWHC 1559 (Admin) (24.6.24, Eyre J). In recent cases this Court has been made aware of the fact that appeals in Sneddon and Oakley (No.2) mean those cases are heading for authoritative resolution at higher judicial altitude.

By s.239(6) of the Criminal Justice Act 2003, Parliament empowered the SSJ to give the Parole Board statutory “directions” as to the matters to be taken into account by the Board in discharging the Board’s statutory functions. Those functions include the Board’s statutory duty (s.239(2)) of advising the SSJ with respect to any matter, to do with the early release or recall of prisoners, which has been referred by the SSJ to the Board. That is the advisory function which the Board discharges when it makes a recommendation about transfer to open conditions. By s.239(6), Parliament required the SSJ, in giving statutory directions, to have regard to (a) the need to protect the public from serious harm from offenders and (b) the desirability of preventing the commission by them of further offences and of securing their rehabilitation. Different statutory directions have been issued by the SSJ at different times.

The relevant Statutory Directions for the purposes of the present case are agreed to be those which the SSJ issued, pursuant to s.239(6), on 28 June 2022. The document in which they are found includes the following at §§1-3 (I have included letters in square brackets to replace bullet points in the original):

As well as issuing statutory directions, the SSJ issues policy guidance. Different policy guidance has been issued by the SSJ at different times. The relevant Policy Guidance for the purposes of the present case is agreed to be the version of the Generic Parole Process Policy Framework which was issued by the SSJ on 12 October 2022. It includes the following at §§5.8.2-5.8.3 and 5.8.6 (again, with letters in square brackets replacing bullet points):

The Board has a duty (2022 Statutory Directions §2(i)) to consider “all information before it, including any written or oral evidence obtained by the Board”. The same duty appeared in the August 2004 statutory directions (at §4) and the April 2015 statutory directions (at §6): see Adetoro §19 and Oakley No.2 §10. The same duty also appears in the August 2023 statutory directions (§2(i)), which will have featured in Hahn.

Courts sometimes invent shorthand terms in judgments, with the aim of decluttering the judgment and helping it to flow. In this judgment, I am going to use the shorthand terms “the Progress Duty” and “the Essential Criterion”.

The Board has a duty to consider whether an ISP has made sufficient progress in addressing and reducing risk to a level consistent with protecting the public from harm in open conditions. This is “the Progress Duty”, as I am calling it. Here it is again from the 2022 Statutory Directions §2(ii):

The Progress Duty is not framed as one of the “criteria” for the Board to recommend transfer to open conditions (2022 Statutory Directions §2[iii]). Nor is it described as one of the criteria for the SSJ accepting a Board recommendation (2022 Statutory Directions §1; 2022 Policy Guidance §5.8.2). It is framed as a duty on the Board when carrying out the assessment leading to the recommendation. And it must be relevant to the SSJ’s decision whether to accept or reject that recommendation. I will address how I think this works at §§21-22 below.

In August 2023 statutory directions (relevant in Hahn), the Progress Duty now stands (at §2(ii)), as one of two criteria for the Board to recommend transfer to open conditions. Under those August 2023 statutory directions (§1), and Policy Guidance from July 2023 (see Valentine §87), the Progress Duty stands as one of three criteria for the SSJ to accept a Board recommendation. The Essential Criterion (§14 below) has dropped out of the picture in these later instruments.

One of the two “criteria” which needed to be met before the Board can recommend a transfer to open conditions (2022 Statutory Directions §2(iii)[b] and §3), and one of the three criteria for the SSJ to accept a recommendation (2022 Statutory Directions §1[b]; 2022 Policy Guidance §5.8.2[b]) is that:

It is clear, in my judgment, that the Essential Criterion is referable to the virtues and utility of a period in open conditions. It – and those virtues and that utility – are squarely focused on final progress towards release. That is the informing (“to inform future decisions”) and the preparing (“to prepare”) to which the Essential Criterion refers. Under the Essential Criterion, the period in open conditions has to be considered “essential” so as “to inform future decisions about release”; and it has to be considered “essential” so as “to prepare for possible release”.

From August 2004 the statutory directions (§1) had said this (see Adetoro §§16, 19):

In my judgment, it is important to appreciate that the Essential Criterion in the 2022 Statutory Directions (§§1[b], 2(iii)[b]) and the 2022 Policy Guidance (§5.8.2[b]) has to be read and understood as subject to three qualifications. These solve a problem about the direction from which the Essential Criterion is being approached.

The first qualification is that “essential” in the Essential Criterion must mean “essential at this time”. The basis for this interpretation is straightforward. The very function and purpose of the Essential Criterion is in informing an assessment by the Board, and then by the SSJ. The assessment is as to whether – at this time – to transfer the ISP to open conditions. This is the same idea as “at this time” when that phrase featured explicitly in the criterion (“a wholly persuasive case for transferring the prisoner to open conditions at this time”) for rejecting Board recommendations, within the Policy Guidance from January 2020 (see Oakley No.1 at §22) to July 2022. It is reflected in the Proforma, which speaks of a move to open conditions being “essential at this time”. I think “at this time” is obviously implicit in the Essential Criterion.

I have referred to the direction from which the Essential Criterion is approached. Mr Grandison has explained that release of an ISP direct from closed conditions, without a period in open conditions, can be possible in some cases: see Gilbert at §56. But it is frequently possible to say, with confidence, that an ISP will not be able to be released without first having had a period in open conditions. That makes a period in open conditions a presently-identifiable necessary precondition to any release. If you are the Board, approaching the Essential Criterion from the direction of considering release, you could say with confidence that a period in open conditions is a presently-identifiable necessary precondition to any release. That assessment could be the basis of your decision not to direct release. You would be saying that a period in open conditions is presently assessed, for this ISP, as being “essential”. That was the sense of “essential” which was being used between 2004 and 2015 (§16 above). This is why as Eyre J has explained, read ‘literally’, the Essential Criterion would already be met: see Overton at §33. The importance, in applying the Essential Criterion, of essential being “at this time” is that the question is being approached from the other direction. It is not release, but progression within the custodial setting towards eventual release. Yes, a period in open conditions is a presently-identifiable necessary precondition to release for this ISP. But is it essential “at this time”? Is the ISP ready for it? Is it safe?

The second qualification is that the assessment – of “essential at this time” – is informed by considering all the information. The basis for this is straightforward. The duty to consider all the information is mandated by the 2022 Statutory Directions (§7 above), without itself standing as a criterion. It must have traction. Its plain function and purpose must be to inform those matters which do stand as criteria.

This is the third qualification. Once again, the basis for this is straightforward. The Progress Duty is mandated by the 2022 Statutory Directions (§7 above), without itself standing as a criterion. It too must have traction. Like the duty to consider all the information, its function and purpose must be to inform those matters which do stand as criteria. It informs the narrower question of whether the ISP is “low risk of abscond”. And it informs the wider criterion of whether transfer of the ISP is essential “at this time”. There is an obvious link between a period in open conditions being essential “at this time” and being satisfied that the ISP has made sufficient progress in addressing and reducing risk to a level consistent with protecting the public from harm in open conditions. Being so satisfied is the function of the Progress Duty.

All of which means there is a principled fusion. It is no more than being faithful to the structure, wording and purpose of the 2022 Statutory Directions and 2022 Policy Guidance. It means that, properly understood and applied, this is what the Essential Criterion means (the encapsulation is mine):

The crucial point is this. The Board as adviser, and the SSJ as decision-maker, would be able to say that the Essential Criterion is not satisfied because of insufficient progress to a safe risk level. As Eyre J put it in Overton at §34, in addressing the stage reached in the ISP’s progress and development under the Essential Criterion, is it necessary to consider whether the ISP has reached a stage such that “the level of risk which he or she poses can safely be managed in the open estate”. I agree. Progress to a safe risk level is ‘baked in’ to the Essential Criterion, whoever is applying it. This means the Progress Duty has full traction in the criteria applied by the Board. That full traction is important, because there is more to progress to a safe risk level than abscond risk (2022 Statutory Directions §2(iii)[a]). It means concerns relating to progress to a safe risk level can be relied on by the SSJ by reference to the Essential Criterion (Statutory Directions §1[b]; Policy Guidance §5.8.2[b]). It means those concerns do not need invocation of the ‘undermining public confidence’ criterion (Statutory Directions §1[c]; Policy Guidance §5.8.2[c]).

This has a practical consequence. It means that, when the Panel or the SSJ are talking about whether a period in closed conditions is “essential”, they are likely really to be talking about whether a period in open conditions is, “at this time”, safe in terms of the Progress Duty: sufficiency of progress to a safe risk level. That is an entirely lawful approach to the Essential Criterion. It directly engages questions of risk-assessment and risk-management. It is, in my judgment, the essence of the present case.

I have encapsulated the correct meaning of the Essential Criterion (§22 above). Where does that leave the relevance of further work in closed conditions? Plainly, if there is a further step in closed conditions whose completion is necessary to show progress to a safe risk level before transfer to open conditions, then the Essential Criterion can lawfully be considered not to be satisfied.

Mr Grandison says the point goes further. He submits that if there is a further step which can “just as well” be taken in closed and open conditions, then – as a matter of legally correct interpretation – the Essential Criterion cannot be met. In support of that contention, he relies on these observations of Calver J (Cain at §65; cited in Uddin at §44 and noted in Hahn at §26):

In order to understand the contention being made, it is helpful to see how it plays out in the present case. Here, everyone agreed that what was needed was a further period of mental health stabilisation, before any release. The Panel thought – by reference to the Progress Duty – that there had been progress to a safe risk level and the pre-release mental health stabilisation could safely take place in open conditions in parallel with the virtues and advantages of a period of testing in open conditions, that being an identified precondition to any release. The contention, as I see it, is that a further period of mental health stabilisation could take place “just as well” either in closed conditions or in open conditions. That engages the observations in Cain. It means the Essential Criterion could not, on its correct interpretation, be met. It follows that the Panel made an error of law, so this is a “deficiency” case (Sneddon §28(5)) which means the claim must fail.

In my judgment, this does not fit with what was actually decided in the present case. The Panel recognised the virtues and advantages of a period in open conditions, which could not be achieved “just as well” in closed conditions. The period of mental health stabilisation was not “further work”, still less “further work on the prisoner’s offending behaviour”. The axis of disagreement was all about whether there had been progress to a safe risk level and the pre-release mental health stabilisation could safely take place in open conditions. One expert witness thought it could not be; and the SSJ rejected the Panel’s recommendation in light of those concerns. This case is about the Progress Duty. It is about “whether the prisoner has reached a stage such that the level of risk that he … poses can safely be managed in the open estate”. The SSJ had concerns on that question. The SSJ did not say that, even if that stage had been reached, mental health stabilisation should take place in closed conditions, denying the Claimant the parallel virtues and advantages of a period in open conditions. The argument based on Cain §65 cannot succeed because it does not fit with what was decided. As in Hahn (see §26), the difference between Overton §34 and Cain §65 is not material to the present case.

Since the observations in Cain at §65 have been raised, and since Mr Grandison has submitted that they have “paramount importance” in understanding the Essential Criterion, I will identify the legal position as I see it. (1) I have sought to encapsulate the correct meaning of the Essential Criterion (see §22 above). The questions that then arise are really questions of reasonable “application”, rather than questions of objectively correct “interpretation”. All questions of reasonable application, using the phrase which Calver J (Cain §65) and Eyre J have emphasised (Hahn §26), are questions to be answered “depending always on the particular facts of the case”. (2) It may be that a period in open conditions is not, for the particular ISP, a presently-identifiable necessary precondition to release. In those circumstances, where there is necessary work available in closed conditions, the Essential Criterion may well reasonably be regarded as unmet. That may readily fit with Calver J’s description of the ISP being “in closed conditions on a Progression regime”. (3) Or it may be that necessary work on the ISP’s offending behaviour is itself a prior step, before the virtues and advantages of a period in open conditions can be meaningfully beneficial. In such a case, a decision not to transfer to open conditions until after the further work which “can be done just as well in closed conditions” may well be a reasonable application of the Essential Condition, even though the ISP has already reached a stage such that the level of risk that they pose can safely be managed in the open estate. (4) Much more challenging is a scenario with all of these features: (a) a period in open conditions is presently-identifiable necessary precondition to release; (b) the virtues and advantages of testing in open conditions would be immediately meaningfully beneficial; (c) in parallel to those virtues and advantages there is some step (whether or not it is “work”) which “can be done just as well in closed conditions”; and (d) the ISP has already reached a stage such that the level of risk that they pose can safely be managed in the open estate. In such a scenario, I have difficulty seeing how a refusal to transfer would be a reasonable application of the Essential Criterion. In the present case, the disagreement was about whether feature (d) is present.

In their list of agreed issues, Counsel identified this as an issue: Whether “essential” means that open conditions is the only option for progression or whether open conditions can still be deemed essential even if there are arguably ‘other’ routes that might lead to progression. As I have explained, this issue does not fit with the disposition of the present case. But, if it mattered, my answer would be this. The reasonable application of the Essential Criterion depends on the particular facts and circumstances of the individual case. The identification of a step which can “just as well” be taken in closed conditions does not, necessarily and of itself, mean that that the Essential Criterion is unmet.

Mr Buckley submits that it was unreasonable for the SSJ to decide not to accept the Panel’s recommendation of transfer to open conditions. He says, properly understood, the question on which the SSJ departed from the Panel’s assessment (a) was a question of significant or particular advantage requiring “very good reason” (Oakley No.1 §§48, 51) or (b) was an “other” question requiring “good reason” (Sneddon §28(7)). That is so, notwithstanding the 2022 recasting of the statutory directions and policy guidance (cf. Oakley No.1 §§21-22; Sneddon §6). On analysis of the SSJ’s decision, there was no “very good reason” or “good reason”. That is the argument. To address it, I am going to start by analysing the SSJ’s reasoned decision, in the light of the Panel’s reasoned recommendation. I will take the key topics in the case. This exercise will enable me to consider the nature of the disagreement and to characterise the reasoning for departing from the Panel’s assessment. I will deal, at the end of the judgment, with some questions of law and some observations as to the state of the law.

This was a first key topic. On it, I have detected no disagreement. As the Panel’s Determination explained, Mr Jowsey’s evidence identified a period in open conditions as:

This was a second key topic. It engaged one of the criteria for the Board (2022 Statutory Directions §2(iii)[a]) and for the SSJ (§1[a]). Here too, there is no discernible disagreement. As the Panel explained in the Determination: Mr Jowsey “did not think there would be any problems around abscond”; Ms Wright “did not think Mr Carrigan was an abscond risk”; Ms Morris “did not think he was an abscond risk”; and Ms Brown “did not think there would be an abscond risk”. Everybody agreed that there was no abscond-risk problem. The Panel could “find no reason to disagree”. The SSJ’s Decision Letter did not disagree. The SSJ’s Proforma recorded that there were “no noted concerns regarding his level of risk of abscond”; and “no concerns around … abscond”.

This was a third key topic. Again, there was no discernible disagreement. The Panel described in the Determination: the programmes which the Claimant had completed; Ms Wright’s recognition that the Claimant had “completed a range of offending behaviour programmes”, that he had “attended interventions to support him in desisting from future alcohol misuse”, and that there was no “core risk reduction work to complete”; Ms Morris’s recognition that the Claimant had “completed core risk reduction work”; Ms Brown’s recognition that the Claimant had “completed all the sentence planning objectives” and had “embraced all offence focused work”; and the Panel’s acceptance that “there is no further core risk reduction work to be done”. The SSJ’s Decision Letter recognised that the Claimant had demonstrated positive progress regarding work and that “there is no further core risk reduction work to be completed”. The SSJ’s Proforma recorded “no further work to complete”.

This was a fourth key topic. Again, there was no discernible disagreement. The Panel’s Determination recorded that Ms Brown had assessed the Claimant as presenting a “medium” probability of future proven offending using OGRS3 (Offender Group Reconviction Scale Version 3), a “low” probability of future proven general offending using OGP (General Offending Predictor), and a “low” probability of future proven violent offending using OVP (Violent Offending Predictor); and that Ms Brown had assessed, in light of the nature of the index offence (the murder which the Claimant had committed), that if he were to reoffend he would present a “high” risk of serious harm to the public, a “low” risk of serious harm to known adults, and a “medium” risk to children. The reasoning in the SSJ’s Decision Letter and Proforma involved no deviation.

This was a fifth key topic. It too involved no discernible disagreement. The Panel explained in the Determination: that Mr Jowsey’s evidence described the Claimant having developed good communication skills and being very motivated to approach people if he was struggling; that the Claimant in his oral evidence was candid and frank, was very insightful about his own risks, and had made creditable and considerable efforts to understand the implications of a move to open conditions; that Ms Wright’s evidence described the Claimant as having an understanding of and good insight into his mental health conditions; that Ms Morris described the Claimant as reflective, balanced and consistent; and that Ms Brown’s evidence described the Claimant as having demonstrated a largely positive attitude, with an apparent insight into his risk areas, who had developed his motivation to remain offence-free. The Determination went on to describe the Claimant’s acute awareness of his own responsibility for maintaining medication, his full understanding of risks, his determination to seek help, his understanding of the challenges and implications of moving to open conditions, and his clear future plans and motivation to begin his gradual resettlement into the community. The reasoning in the SSJ’s Decision Letter and Proforma involved no deviation on this topic.

This was a sixth key topic. Again, it involved no discernible disagreement. The Panel explained in the Determination: that the context of the murder committed by the Claimant included his suffering visual and auditory hallucinations, having been admitted to hospital earlier that same day because of mental health problems; that Mr Jowsey’s evidence described the Claimant’s last period of ACCT (assessment, care in custody and teamwork), which closed in March 2021, as having followed an episode of paranoia and hallucinations; that the Claimant had three January 2022 negative behaviour entries, all of which were linked to a decline in mental health rather than any deliberately obstructive behaviour; that there had been a similar cluster of warnings following a July 2020 return to HMP Wakefield from the PIPE (psychologically informed planned environment) at HMP Frankland; and that the Claimant’s custodial behaviour had been generally good except at times when his medication had been an issue out of his control. The SSJ’s Decision Letter and Proforma referred to all of these aspects, emphasising relevant custodial behaviour and making clear its link to mental health, and pointing out that the July 2020 return involved the presentation of being unstable in mental health. Nobody was materially disagreeing about the link between behaviour and mental health; nor about when relevant behaviour had taken place; nor about how it linked to mental health.

This was a seventh key topic. It is the topic which was at the heart of the case. This can be seen from the following features, which the SSJ recorded in the Decision Letter, in explaining the outcome and why the Panel’s recommendation was not being followed:

The Panel – which included a psychologist member – had the opportunity at an oral hearing to probe and evaluate the opposing views of the witnesses, including the two expert psychologists who gave evidence. Ms Morris, the Independent Psychologist, supported transfer to open conditions. Ms Wright, the Prison Psychologist, did not; but rather supported a further period in closed conditions. Ms Hogan, the mental health nurse, gave no positive view either way. Mr Jowsey (the POM) and Ms Brown (the COM) both supported transfer to open conditions. Three of four witnesses who expressed a view thought there had been progress to a safe level for open conditions. Nobody expressed the view that mental health deterioration and destabilisation would not be a significant backward step. The Panel referred to gaps which it saw in the information available to Ms Wright when she wrote her report. Against that is the fact that, having participated through the hearing at which all information and evidence was ventilated, Ms Wright had all the information when she reviewed the position and wrote her addendum report, and maintained her view.

Ms Wright was concerned about the prospect and implications of transfer to open conditions, at this time, in terms of mental health stability and destabilisation. No witness expressed the view that there could be no mental health destabilisation. Ms Brown recognised “a possible risk of destabilisation”, but did not think that “violence would be likely”. The Panel – clearly by reference to the Progress Duty – identified points which in its assessment “evidenced” that there had been “sufficient progress during the sentence in addressing and reducing risk to a level consistent with protecting the public from harm”. The Panel did not say it had assessed, with the benefit of the due process of an oral hearing, that there was a low risk of destabilisation. What it said was that there seemed to have been a very considerable movement in the Claimant’s ability to maintain the stability of his mental health, and that he has – in many respects – a positive attitude and insight, the key relevance of which was in reacting to triggers and seeking help and support in the context of deterioration. This was the point emphasised by the witnesses who supported transfer to open conditions. Ms Wright was not convinced that attitude and insight in responding to triggers was an answer, in light of the nature and duration of the period to date of greater stability. The mental health review team’s concerns, ventilated through Ms Hogan, were about mental health decline, including impairment in the ability to comply with restrictions. Whether or not mental health deterioration and instability would mean triggers, which would or would not be picked up, there remained the point that it would be a setback. It would put the Claimant further away from achieving release. Conversely, a further period of mental stability would be a positive benefit. This was part of Ms Wright’s thinking. It was part of the SSJ’s thinking.

In my judgment, once there is a clear understanding as to where the actual disagreements were and were not, we can see that the evidenced prospect of mental health deterioration (not contested), with the behavioural risks (on which legitimate expert views differed) and the prospect of a setback (not contested) justified as reasonable the SSJ taking a different view from that of the Panel, by reference to the Progress Duty. The question is not whether the Panel’s view was legitimate. The only questions are about whether the SSJ had a reasonable basis for the decision to depart from the Board’s assessment. I can see no particular conclusion or proposition on which the Panel had a particular advantage over the SSJ, to engage any principled requirement of a “very good reason” (Oakley No.1 §§49, 52). The highest it could be put, in my judgment, is that there needed to be “good reason” for departing from the Panel’s assessment (Sneddon §28(7)). In my judgment, the SSJ has demonstrated good reason. That makes the SSJ’s decision lawful. The position is a fortiori, if all the SSJ needs is reasoned disagreement (Sneddon §30).

What I have said so far is derived from the SSJ’s Decision Letter, read in context. But there is reinforcement to be derived from the observations of the decision-maker, Mr Davison, endorsed in the Proforma. In accepting the reasoned recommendation of Ms Whyte, Mr Davison wrote this:

It was not argued in the present case that the Proforma is inadmissible as evidence which could provide illumination as to the SSJ’s reasons (cf. Valentine §112). The status of a proforma has been a controversial topic (see Valentine §§97-100). My views are as follows: (1) There is an important discipline in the public interest, with candour and procedural due process implications, in a reasoned decision document being drawn up to reflect the substance of the decision and its reasoning process. (2) Where a contemporaneous reasons document exists, whose disclosure did not accompany the reasoned decision document, the judicial review court will proceed with caution. The Court may want to know whether the letter that was drawn up was approved by the decision-maker, or whether it was merely derived from a reasoned recommendation accepted by – and/or contemporaneous reasoning of – the decision-maker. (3) Illumination derived from such an undisclosed contemporaneous reasons document is not the same as retrospective amplification (Wynne §55). It is not necessarily improper, at least on a reasonableness review, to derive assistance from a contemporaneous reasons document which did not accompany and was not fully reflected in the reasoned decision document drawn up to reflect its substance. This cuts both ways: it may help a claimant show why the decision was unreasonable; or it may help the SSJ show why it was reasonable. But none of that detracts from the need for the reasoned decision document to be fully and fairly reflective of the underlying contemporaneous reasoned decision document. The Court could grant judicial review on the basis that the reasons communicated were legally inadequate, or that non-disclosure of true reasons was procedurally unfair. And there could be costs implications.

As the line of cases shows (§5 above), a lot of water has flowed under the High Court bridge since October 2022, when Chamberlain J reasoned in Oakley (No.1) (at §49) that “many questions” addressed by the Board about “the level of risk posed by a prisoner” fall within a category where “very good reason” needs to be shown by the SSJ for taking a different view. And there are many cases in this area, before and after December 2023, when I reasoned in Sneddon (at §28(7)) that “other questions” addressed by the Board require a “good reason” to be shown by the SSJ for taking a different view, and I could not accept the “reasoned disagreement” analysis (at §§30-31). The submissions in the present case involved a return to those themes, in the light of the case-law as a whole. I will be transparent about where I think the law currently stands. I do so, recognising in particular two key points which have been emphasised in the cases.

A first key point is about the need for proper regard to the SSJ’s risk-expertise. Whether the analysis in Sneddon at §§28-31 fails to have proper regard to the SSJ’s risk-expertise raises a question (see Oakley (No.2) at §21) on which views have differed (see Uddin §38; but cf. Valentine §95). In the present case, Mr Grandison takes issue with the general idea that the Board has an “expertise in assessing the risk posed by individual prisoners” with “legally significant institutional and due process advantages over the SSJ” (Sneddon §28(2)). The SSJ claims a primacy of risk-expertise in the area of risk-management (see McKoy §45; McPhee §23) and in the area of risk-assessment; or at least a relevant risk-expertise with the Board in both of those areas (Banfield §29; Gilbert §71; Overton §27; Zenshen §80; Uddin §45). Mr Grandison submits that to require a “very good reason” or a “good reason” for rejecting an assessment of the Board on any “risk” question, whether relating to risk-assessment or risk-management, overlooks the SSJ’s primacy or parity of risk-specialism, which was recognised by Jackson J in Banfield at §29 (and Sales LJ in Gilbert at §71) when speaking of the SSJ and “the expertise of his department”.

A second key point is about the legal significance of Policy-Reorientation. The various emanations of Policy Guidance from 2012 to January 2020 identified narrow criteria for SSJ rejection of a Board recommendation: (i) this was the shape of PSI 36/2012 §6.5 (Sneddon §7); (ii) it was the shape of PSI 22/2015 §6.4 which included “not … a wholly persuasive case” (Sneddon §8). The versions of the Policy Guidance in January 2020 and August 2021 also identified criteria for SSJ rejection of a Board recommendation (including “not … a wholly persuasive case”) (Sneddon §§6, 9). The Policy-Reorientation (see Hahn §29) came on 21 July 2022, since which time the Policy Guidance has stated criteria for acceptance of a Board recommendation. These have throughout included low abscond-risk, to which has been added: (i) the Essential Criterion and not undermining public confidence in July 2022 (Overton §8) and again in October 2022 (Zenshen §29); and then (ii) the Progress Duty and “wholly persuasive case” in July 2023 (Valentine §87) and August 2023 (Hahn §11). The same Policy-Reorientation is found in the statutory directions since 28 June 2022 (§7 above) and 1 August 2023. The statutory directions are significant in having what has been called “statutory legal effect” (Gilbert §53). Mr Grandison relies on the reversal of the ‘default position’ and submits that, even if were correct to say that a “good reason” was needed for a departure from a Board assessment before the Policy-Reorientation (Sneddon), that is not correct after it (Overton): see Hahn at §§30-31. Mr Buckley says it is possible to identify “overarching principles” which are independent of which iteration of policy guidance is being considered.

To understand what is happening in the case-law in this area, I think it may be worth standing back:

I encapsulated (Sneddon at §30) the SSJ’s argument on what I called substitutionary disagreement (§27). I called this analysis “reasoned disagreement”. It remains the SSJ’s position – at least so far as any risk question is concerned – and Mr Grandison has again embraced it. Here it is again:

If what I have called “reasoned disagreement” were the law, for all risk-assessment or risk-management questions assessed by the Board, then the analysis in Oakley (No.1) will prove to have been wrong. The judgment in Oakley (No.1) reasoned that many risk questions addressed by the Board require “very good reasons” if the SSJ is to depart from that assessment, consistently with the duty to act reasonably. If reasoned disagreement is sometimes the law, for some (or all) risk-assessment or risk-management questions assessed by the Board, then the analysis in Sneddon will prove to have been wrong. That is because the judgment in Sneddon accepted that ‘other’ risk questions addressed by the Board require “good reasons” if the SSJ is to depart from the Board’s assessment, consistently with the duty to act reasonably.

I have identified the point about risk-expertise (§45 above). In Wilmot in 2012 (at §46) it was observed that the courts had “recognised” the “expertise” which resides in the Board “in particular on the assessment of risk”. In that case, the SSJ described the then Policy Guidance as itself being based on a “recognition” of “the parole board’s experience and expertise in the risk assessment of prisoners” (§52). In Banfield in 2007 the “fact” had been identified – as a relevant consideration – that “the Parole Board has particular expertise in assessing the risk posed by individual prisoners” (Banfield §28(1)); and so was the “benefit” to the SSJ of “the expertise of his department” (§29), in the context of whether the SSJ had or could claim “any superior expertise” (§26). In Hindawi in 2011 it had been emphasised – in the context of release recommendations – that the Board “is expert in the assessment of risk” as well as “immunised from external pressures”, operates like a court in sifting and analysing the evidence (§50); and the SSJ had “some expertise” but not “superior expertise” (§51). In Harris in 2014, it was observed that the Board “is expert” and “in many cases performs a judicial function” (§29), as well as seeing “the witnesses” at any oral hearing (§30). Then in Gilbert in 2015, the Court of Appeal rejected the contention that the SSJ was obliged to refer to the Board any ‘transfer to open conditions’ case where risk may be in issue (§§69-70), explaining that the SSJ, his department and his agencies “are also experts” in risk-management “including assessing prisoner risk” (§71). These cases and observations (2007-2015) provide the setting for the various ways in which, at that time and thereafter, the Courts have charted what I have seen as a “middle way” between treating the Board’s recommendation as binding and allowing departure by substitutionary disagreement (Sneddon §27). That includes the idea that the SSJ must have “due regard” to the Board’s “particular expertise” (Kumar §54; John §43) and “sufficient regard” to the advantages of an oral hearing on the Board “in its assessment of the relevant risks” (Kumar §55; Wynne §47).

In Oakley (No.1), Chamberlain J used the phrase “appropriate respect” to describe what public law reasonableness requires of the SSJ’s, when departing from questions involving the exercise of judgment requiring the balancing of private and public interests (§51), including the question which constitutes the Progress Duty (§50). Chamberlain J distinguished the “appropriate respect” class of question from the “very good reasons” class of question. The idea of “appropriate respect” was itself linked to the Board’s “expertise and experience” (§50). The primary judgment was for the SSJ, and the Board had no “presumptive priority” (§50). This idea of “appropriate respect” for Board assessments on what I have called “other questions” (Sneddon §28(7)) was derived from Hindawi §60, where it was specifically linked to the Board’s “assessment of risk” (Oakley (No.1) §27). In cases where, like Oakley (No.1) and Sneddon, the relevant statutory directions and Policy Guidance preceded the Policy-Reorientation (§46 above), “appropriate respect” featured in Harris (at §§19, 41); Kumar (at §§8, 12, 63); John (at §40); Wynne (at §57); Green (at §§42iii, 48); McKoy (at §44); Oakley (No.2) (at §17); and Uddin (at §§40, 45). In cases where the relevant statutory directions and Policy Guidance postdated the Policy-Reorientation, “appropriate respect” featured in Zenshen (at §38); Overton (at §§29-30); Cain (at §67); Valentine (at §§105, 109); Hahn (at §§22, 27, 31). In its most basic form, the reasonableness duty does not require a duty of “appropriate respect”. But a reasonableness review, whose nature and intensity are context-specific, can.

In Hindawi, the Court said (at §52) that it was self-evident that the SSJ “should and would accord weight” to the Board’s recommendation, but that the weight which the SSJ “should” accord must depend on the matters in issue, type of hearing, findings and nature of the assessment of risk. Given the use of “should”, I have described this idea as “required” weight (Sneddon §28(3)). Other relevant observations in cases involving statutory directions and policy guidance prior to the Policy-Reorientation (§46 above) are Adetoro (at §55); Green (at §§34, 42i); and Oakley (No.2) (at §11). In cases about statutory directions and policy guidance after the Policy-Reorientation, there is Zenshen (at §32); McPhee (at §8); and Valentine (at §90i). McPhee (at §8) says the SSJ “must” give the Board’s recommendation “appropriate weight”. In its most basic form, the reasonableness duty does not require that a public authority “should” accord “weight” to a legally relevant consideration. But a reasonableness review, whose nature and intensity are context-specific, can.

In Kumar, in explaining why the impugned 2020 policy guidance was lawful, the Court emphasised (at §57) that the policy guidance was not permitting “substitution” for the Panel’s views as an “expert body”, without “justification”. Also in Kumar, the basis for the SSJ’s rejection of the Board’s recommendation was both “clearly explained” (ie. clarity) and also “objectively justified” (§63). I saw this idea of “justification” with the “reasonable basis” for rejection of a Board finding (Sneddon §28(4)). Another case about statutory directions and policy guidance prior to the Policy-Reorientation (§46 above) is Oakley (No.2) which speaks (at §35) of “sufficient justification”. In the cases about statutory directions and policy guidance after the Policy-Reorientation, the idea has been taken up of a “required justification” and “cogent … justification” (McPhee at §§8, 23, 28), where the “basis for departure” by the SSJ must be “rational” and “properly justified” (Overton at §25; Oakley (No.2) at §19; Hahn at §22). The idea of “justification” for the departure has continued, with references to “proper justification”. In its most basic form, the reasonableness duty does not speak of “justification” for rejecting a finding. But a reasonableness review, whose nature and intensity are context-specific, can.

The case-law has increasingly referred to the idea of the SSJ “engaging” with the Board’s assessment of a question relating to risk. There are references to “proper” engagement, “genuine” engagement and “adequate” engagement. This is a heavy feature in cases about statutory directions and policy guidance after the Policy-Reorientation (§46 above): see Zenshen at §25; Overton at §§25, 26, 29, 31, 38, 39 and 41; McPhee at §23; Valentine at §110; and Hahn at §§22, 32, 38, 41, 42, 46. The idea of “engaging” also featured in cases about statutory directions and policy guidance before the Policy-Reorientation: see Oakley (No.1) at §57; Green at §49; Oakley (No.2) at §19; and Uddin at §48. The idea of “genuine engagement” has been linked to the idea of a “basis for departure” which is “properly justified”: see Overton at §25; Oakley (No.2) at §19; Hahn at §22. In its most basic form, the reasonableness duty – and the duty to give legally adequate reasons – do not involve “engagement”; or “proper” engagement, “genuine” engagement and “adequate” engagement. But a reasonableness review, whose nature and intensity are context-specific, can.

One possibility is that “appropriate respect”, “required weight”, “justification” are all unhelpful glosses and what is apt is a basic, stripped-back reasonableness review with legally adequate reasons. The Board’s recommendation would become a legal relevancy, the decision whether to follow or depart from the Board’s assessment would be a principal controversial issue, and that is what “engagement” means. That would be “reasoned disagreement”. Another possibility is that the reasonableness review, whose nature and intensity is context-specific, involves something of substance here; beyond the Board recommendation being a legal relevancy, and the decision whether to follow or depart from the Board’s assessment being a principal controversial issue. It engages “the nature and quality of the reasons required” (Oakley (No.1) at §51).

I have suggested that this can be thought of, straightforwardly, as the need for a “good reason” (Sneddon §28(7)). In Valentine, the view was expressed (at §95) that this was not inconsistent with the Court of Appeal’s analysis of risk questions and risk-expertise in Gilbert. That is my view too. When Sales LJ spoke for the Court of Appeal in Gilbert, he considered (at §73) “the distribution of responsibility” between the SSJ and the Board “as contemplated by the statute”; and he considered Banfield. I see, as what Sales LJ identified, as a straightforward idea – in the context of risk questions – that the SSJ could depart from the Board’s advice for “good reason”. He said this:

It is right of course to recognise that the SSJ has risk-specialism, with the advantage of “the expertise of his department” (Banfield §29); that the SSJ is and remains the primary decision-maker; and that the focus is on whether the SSJ’s decision was reasonable. But the Board has specialist decision-makers, with the institutional advantages of independence, and the rigours of due process. Yes, the SSJ has relevant risk experts. But the SSJ puts relevant risk experts forward, to give expert risk-related evidence, to the Board; in a process where that evidence – emanating from the “expertise” of the “department” – can be considered and tested. This is a significant combination of circumstances involving institutional and due process advantages (Sneddon at §28(2)), when the Board has advised, before the SSJ makes the final decision. I do not for my part think that “very good reason”, for questions of significant advantage (Sneddon §28(6)), and “good reason” for “other questions” in light of the Board’s general position of legally significant advantage (Sneddon §28(2)(7)), impose unrealistic or inapt standards. In particular, “good reason” orientates the SSJ to focus on the Board’s assessment and articulate a reasoned basis of substance for the disagreement. No more; but no less.

That leaves the Policy-Reorientation point (§46 above). In Hahn, Eyre J has analysed Sneddon and Overton, suggesting that Sneddon is attributable to the position prior to the Policy-Reorientation. My view is as follows. First, I have no difficulty with the analysis in Overton, since it does not embrace reasoned disagreement (Sneddon §30), and because the repeated references to “engagement” are clearly linked to a point of substance (the phrase used is “properly justified”): see Overton at §24; Hahn at §22. Secondly, I think what the cases in this area reflect are a firm principle of context-specific reasonableness, not simply the current formulation in Policy Guidance. In Kumar, new policy guidance was lawful, because it left intact (and “reflected”) the appropriate respect for the Board’s fact-findings as well as its role and expertise and its assessment of risk (§§12, 63). I have identified (§§51-53 above) how the various ideas found in the case-law are in cases before and after the Policy-Reorientation. I accept of course that the SSJ enjoys a latitude in the shaping of statutory directions and policy guidance. I would not, however, accept that the public law significance – for the SSJ’s duty to act reasonably – of the Board’s assessments is no more and no less than whatever policy or statutory directions have been promulgated by the SSJ. If policy guidance said the SSJ could ‘disagree’ with the Board, public law would I think read-in “provided that the SSJ acts reasonably”. If policy guidance said the SSJ could ‘disagree’ with the Board on a question of credibility after an oral hearing, public law would I think read-in “provided that the SSJ acts reasonably” from which it would insist on a Hindawi good reason (Kumar §55). I do not think, in conferring the statutory power to make statutory directions, Parliament is to be taken as permitting abrogation of the basic common law duties of reasonableness and legally adequate reasons. None of the changes in the statutory directions or policy guidance have, moreover, purported to abrogate those duties. I see in the relatively modest threshold of a “good reason” is a bedrock contextual minimum that the common law would require of the SSJ, when departing from a question assessed by the Board with its legally significant institutional and due process advantages, including on questions of risk where the SSJ has been able to ensure that the MOJ’s relevant risk experts have given evidence to the Board.

The claim fails because there was a “good reason” for departure from the risk-assessment and risk-management question assessed by the Board, and nothing more was needed. If less were needed, because “reasoned disagreement” suffices in law, then the claim would necessarily also have failed. As to consequential matters, the parties are agreed as to the appropriate order, in light of this judgment circulated in draft: (1) the substantive claim for judicial review is refused; (2) the Claimant is to pay the SSJ’s costs, not to be enforced without further order from the Senior Courts Cost Office under reg.16 of the Civil Legal Aid (Costs) Regulations 2013; and (3) there shall be a detailed assessment of the Claimant’s publicly funded costs.