The Secretary of State for Justice, R (on the application of) v The Parole Board for England and Wales

This is my judgment on the expedited rolled-up hearing of the claim by the Secretary of State for Justice (“the SoS”) for judicial review. The decision under challenge is that of the defendant, the Parole Board of England and Wales (“the Parole Board”) of 28 January 2026, refusing to set aside its earlier direction to release the interested party, Mr Zenshen (“the set aside decision”).

Statements (attaching relevant documents) were filed by Mr Gordon Davison, Public Protection Group Director for His Majesty’s Prison and Probation Service (“HMPPS”), in support of the claim, and, in response, by Mr Simon Creighton, the consultant with conduct of this matter at Mr Zenshen’s solicitors. All parties lodged skeleton arguments, and an application was filed on behalf of the SoS (on a “precautionary basis”), seeking an extension of time for the claim.

All parties were represented at the hearing, albeit the Parole Board did not seek to add to the points made in its skeleton argument. Further written submissions were received from the SoS and the Parole Board on 22 May 2026 concerning the position if the claim were to be allowed; responsive submissions on this point were filed on 26 May 2026 by those acting for Mr Zenshen.

Prior to the index offence, Mr Zenshen (formerly Mr Reginald Wilson) had a number of convictions for offences that included theft, burglary, attempted robbery, arson, wounding with intent and unlawful wounding. In 1991, he was sentenced to life imprisonment for murder. The facts of the murder provide important context to the decision under challenge, and to the challenge itself.

On 3 February 1990, Mr Zenshen (then Mr Wilson) was 25. His chosen victim – a complete stranger, who lived alone - was a highly respected consultant dermatologist and consultant palaeopathologist. By a ruse demonstrating both planning and premeditation, Mr Zenshen - armed with a heavy blunt instrument (most likely a hammer), wrapped inside a plastic carrier bag, and carrying a rope - was able to gain access into his victim’s house. Once in, using the blunt instrument (still in the carrier bag), Mr Zenshen struck his victim down, inflicted a series of heavy blows to his head, dragged him into the study (using a rope to avoid any forensic link), and then inflicted further blows. Some 17 blows were struck to the back, sides and front of the victim’s head, fracturing the vault of his skull, and resulting in a fatal brain injury.

After the killing, Mr Zenshen scoured the house and stole a wallet and pocket watch. A medieval skull that had been in the house, which showed violent damage to the frontal area, was also found to be missing; later the trial judge (Potts J) noted the coincidence of injury between that skull and the injuries inflicted by Mr Zenshen.

During the investigation into the killing, Mr Zenshen sent an anonymous letter to the police. Making various remarks about the offence, he said it “gave me a buzz better than smack”, and that the police were “barking up the wrong tree with theft ... more like fun”. When ultimately arrested and interviewed, Mr Zenshen denied any responsibility for the murder.

After Mr Zenshen’s subsequent conviction at trial, on 25 July 1991 Potts J sentenced Mr Zenshen to life imprisonment. In his report to the Home Secretary (then standard for life sentence cases), Potts J referred to the fact that, when arrested, Mr Zenshen’s house had been searched and a concealed cache of arms was found (including a sawn-off shotgun, cartridges and cross bows; prior to his trial, Mr Zenshen pleaded guilty to possession of a firearm and 73 cartridges), along with a plan to kill police officers, and a notebook containing a list of police officers and their addresses. In subsequent legal representations (in May 2006), Mr Zenshen agreed that when the offence was committed he was “preoccupied with a hatred of authority and ... had some form of loose and relatively unformed idea that by committing this murder he would be brought into direct and physical conflict with the police. It was this conflict that he was seeking to precipitate by committing the murder”. I further note that, while on remand, Mr Zenshen was alleged to have tried to stab a police officer and a charge of attempted murder was ordered to lie on the file.

In his report, Potts J also referenced psychiatric reports provided to the court, which diagnosed Mr Zenshen as having a severe psychopathic disorder, representing a grave and immediate danger to the public and, at that time, considered to be untreatable. Finding Mr Zenshen to be “an exceptionally dangerous man”, Potts J recommended that his life sentence “ought to be served in its entirety”. On 16 December 1994, Mr Zenshen was notified that the Home Secretary had determined that the requirements of retribution and deterrence could only be satisfied by his remaining in prison for the whole of his life. This view was endorsed on appeal by the then Lord Chief Justice, Lord Lane.

Following changes introduced by the Criminal Justice Act 2003, Mr Zenshen applied for a review of his whole life order, and, in a judgment of 16 May 2008, Tugendhat J revised the sentence, substituting a minimum term of 18 years. That decision was the subject of challenge by way of Attorney-General reference (No 38 of 2008 [2008] EWCA Crim 2122), which resulted in the Court of Appeal (Sir Igor Judge (PQBD), Penry Davey and Simon JJ) re-assessing the sentence, such that the minimum term was stated to be one of 30 years’ imprisonment. In giving the judgment of the CACD, Sir Igor Judge observed:

“36.

... We have no doubt that from the very outset the offender intended to kill his victim and that every one of the blows he inflicted was struck with that intent. In the end he did kill him. He had ample opportunity to desist from the attack. That opportunity was not taken and as his letter to the police officers investigating the murder clearly demonstrated, in our judgment it is beyond doubt the offender gloated over this killing, preening himself with pleasure at the horrific consequences to the victim of his attack. It was indeed, as the judge told him, a most appalling murder and in our judgment it was a most appalling sadistic murder or a most appalling murder involving sadistic conduct. It was not motivated by an intent to gain, but aggravated by the fact that after the killing the offender made off with some of his property, including the skull. The horrific scene which greeted the victim's 16-year-old daughter (which we shall deliberately not describe) has blighted the rest of her life and the continuing impact on each member of this family is movingly, but so far as possible, objectively described in the statements which we have read. ...”

Following his imprisonment for the index offence, and during the early part of his sentence, Mr Zenshen accumulated 21 adjudications for assaults/violence within the prison. The evidence before me suggests he remained an extremely violent man for many years, such that he was confined in solitary prison conditions until around 2004 due to the risk he presented, albeit he had not demonstrated overt violence against staff or prisoners since 1999. According to the Offender Assessment System (“OASys”) assessment undertaken in June 2025, it was not until around 2011 that Mr Zenshen began to engage with the prison regime; thereafter he was considered to have progressed extremely well. Between 2013 and 2014, Mr Zenshen engaged in the Violence Reduction Programme, and then completed a further 12-months in a Psychologically Informed Planned Environment; he achieved a downgrade to category B and completed the RESOLVE programme, before undertaking the Therapeutic Community Programme, spending three years in treatment between 2018 and 2021, during which he moved to category C conditions.

In July 2021, the 30-year minimum term expired. In 2022, the Parole Board recommended Mr Zenshen be transferred to open conditions (prior to then he had remained in the closed estate), although that recommendation was resisted by the SoS. In November 2023, following a successful judicial review of the SoS’s refusal to follow the Parole Board’s recommendation ([2023] EWHC 2279 (Admin)), Mr Zenshen was moved to open conditions. He started supervised release on temporary licence (“ROTL”) on 2 April 2024, and unescorted ROTL on 12 June 2024. Between October 2024 and July 2025, Mr Zenshen secured employment in a food-processing plant but found the work too physically onerous and, after seeking advice from his Prison Offender Manager (“POM”), secured more suitable employment at a warehouse, having obtained a reference from his previous employer.

On 11 October 2024, the Public Protection Casework Section (“PPCS”) – a unit within HMPPS responsible for managing parole reviews and licence conditions for serious offenders - referred Mr Zenshen’s case to the Parole Board.

On 21 October 2025, the Parole Board held an oral hearing to consider Mr Zenshen’s release; at that hearing, a dossier of materials, prepared by the PPCS, was considered, along with representations from the SoS (represented at the hearing), who opposed Mr Zenshen’s release. As well as hearing from Mr Zenshen, the Parole Board heard evidence from Ms Joukhi (POM), Mr Taylor (Community Offender Manager (“COM”)), and Mr Woulfe (HMPPS psychologist), each of whom recommended Mr Zenshen’s release.

By letter dated 30 October 2025, the Parole Board gave a direction for Mr Zenshen’s release (“the release decision”). The release decision was accompanied by detailed bespoke licence conditions and was issued to Mr Zenshen and all relevant parties on 4 November 2025, although, pursuant to rule 25(2) of the Parole Board Rules 2019, it could not become final until a period of 21 days had elapsed and any application for reconsideration determined.

In explaining its reasons for the release decision, the Parole Board set out the details of the murder, and referred to Mr Zenshen’s plea in respect of the firearms offences and to his earlier convictions, observing:

“Your pattern of offending indicated that you were resistant to authority and that you had seen criminal behaviour as a routine and acceptable part of your way of life”.

The release decision also relayed the history of Mr Zenshen’s time in custody, and summarised the evidence provided to the Parole Board, including the OASys risk assessment, concluding:

“3.4

... Based on the content of the dossier and oral evidence, the risk of re-offending in a manner likely to cause serious harm is fairly described as not imminent, if released on licence. This is because the panel agreed with all three professionals that warning signs associated with substance misuse, disengagement, and/or a significant change in his presentation and demeanour were likely to be identified under a suitable risk management plan and monitoring/supervision.”

The Parole Board explained it had found that:

“3.7

... Mr Zenshen has demonstrated evidence of internal protective factors associated with effective emotional regulation, insight into previous triggers/risk factors, effective relationships with close professionals and strong life goals.”

Acknowledging there were limited external protective factors, it stated that:

“3.8

... The panel considered that release will enable him to build additional protective factors associated with stable employment, suitable long-term accommodation and a broader support network in his chosen resettlement area.”

Having referred to the points made by the SoS (opposing Mr Zenshen’s release), the Parole Board said it found the objections were:

“4.8

... at odds with the evidence, including the persuasive evidence of the three professional witnesses, all of whom unreservedly supported release on licence.”

The Parole Board made clear Mr Zenshen’s release would be subject to initial placement at an approved premises (“AP”). Considering whether further psychological intervention might be required, it concluded:

“4.11

There was no evidence that Mr Zenshen requires a further psychological intervention but, if support is required, Mr Taylor confirmed he will be eligible for a referral to the local IIRMS [Intensive Intervention and Risk Management Service]. Although voluntary, Mr Zenshen has shown himself positively responsive to offers of support when required.”

Acknowledging:

“the sheer gravity and brutality of the index offence as well as the violence before that and the violence he went on to subject prison staff to in the years that followed.”

The Parole Board stated it had completed its own “independent dynamic assessment of risk”, placing weight on the following factors:

“4.14

...

. the absence of evidence that any core risk reduction work remains outstanding,

. sustained evidence of a pro-social attitude, meaningful engagement with professionals and an absence of violence/aggression,

. the absence of any ongoing problematic traits,

. more than adequate evidence of relevant and robust testing in open conditions ...,

. adequate protective factors including a sister, reasonable prospects for employment and funds available for private rented accommodation if required,

. more than adequate evidence of Mr Zenshen’s insight, coping skills and motivation/wherewithal to lead a prosocial life,

. the likelihood that warning signs associated with a prolonged and significant deterioration in his presentation will be identified under monitoring and supervision, even in the long-term,

. notwithstanding the objections from the Secretary of State, all three professional witnesses supported release under the proposed risk management plan without reservation.”

and concluding:

“4.15

Taking all these factors into account, together with the nature of the proposed risk management plan, ... the risk could be effectively managed in the community and ... it is no longer necessary for the protection of the public that Mr Zenshen is confined in prison.”

On 25 November 2025, the SoS made an application for reconsideration under rule 28(1) of the Parole Board Rules, submitting that the release decision was irrational because Mr Zenshen was insufficiently tested in the community, with limited exposure to stressors that might activate identified areas of risk and problematic personality traits. On 8 January 2026, the Parole Board dismissed the application; that had the effect of rendering the release decision final.

Meanwhile, on 3 December 2025, while working on ROTL, Mr Zenshen was caught attempting to steal £30 of items (toiletries and some toys) from his employer. His placement was terminated and Mr Zenshen reported the incident to his COM and POM.

On 4 December 2025, Mr Zenshen’s COM informed the PPCS of the attempted theft, describing this as:

“a significant change of circumstances which I am hoping you can pass on to the Parole Board please, as this may well effect their decision”

As it is not possible to make a set aside application until a release decision is final - which would only be after the determination of any reconsideration application - the view taken at that stage was that the PPCS should await the outcome of the reconsideration application but, if that was unsuccessful, an application should then be made to set aside the release decision.

On the same day, Mr Zenshen was the subject of a prison review, which led to his privileges being removed, he was downgraded from ‘enhanced’ to ‘standard’ on the prison’s incentives and earned privileges scheme, and his ROTL was suspended; Mr Zenshen was not, however, considered to be an abscond risk and he was assessed as remaining suitable for open conditions, albeit he was moved into a residential unit. In January 2026, Mr Zenshen’s enhanced status was reinstated and he again became eligible for ROTL.

On 5 December 2025, a meeting took place between Mr Zenshen, his COM and POM to discuss the theft incident. The COM subsequently provided his views to the PPCS in an email of 11 December 2025, explaining:

“Mr Zenshen was asked why he would jeopardise all the hard work he has put in over the last 17 yrs. He struggled to offer a rational explanation stating he has gone over in his head and has put it down to a moment of madness.

He was asked if he self-sabotaged his release. He purported to want release desperately, values his freedom and has not enjoyed one day in prison, so he did not commit the theft for this reason.

In exploring this, it is evident he took a calculated risk that he would not be caught, explaining that in the time he had been employed at the Range, he had only been stopped on several occasions by store security. He advised he saw a box of toys opened and thought he could get his great niece some Christmas presents. He reported the value of the goods (which included Dove deodorant) was about £30 however, he accepts it is not the value but the principle of the matter.

He decided to commit the Theft despite having approximately £6000 in savings. …

Mr Zenshen presented as extremely remorseful stating that he feels he has let everyone down who were working so hard to support him and pushing for his release. …

It is acknowledged that Mr Zenshen’s behaviour does not indicate an increase in risk of harm however, one needs to ponder why he would undertake such an act, considering he was possibly on the verge of release. Mr Zenshen is fully aware that the Parole Board will be informed of the situation.”

On 18 December 2025, in email correspondence with the Multi-Agency Public Protection Arrangements (“MAPPA”) coordinator, Mr Singh Here, the COM commented that the theft amounted to:

“... a significant change which is likely, in my opinion, to affect release being confirmed for 5 January ...”

In a later exchange with Mr Singh Here, on 5 January 2026, the COM observed:

“It still does not make any sense why he undertook his actions and you are not the only one to ask whether this is self-sabotage however, both his POM and I are of the opinion it isn’t.

I am sure the Parole Board, PPCS and MAPPA representatives will all have additional questions as to Mr Zenshen’s thought process which led to him making the decision to try and steal goods from his employer.

Has he done this before? What would have been next if he had got away with it? Why he would do this with all the years of hard work and progress behind him?”

On 6 January 2026, Mr Singh Here responded:

“... there are a ra[n]ge of unanswered queries and if RZ responses going forward do not tally then the dilemma for all will be framed around his ability and willingness to apply lessons learnt during all the work completed in custody in a consistent and effective manner once/if in the general community.”

An initial draft application had been prepared as early as 16 December 2025, and, when the Parole Board’s reconsideration decision was received on 8 January 2026, a final decision was made that day to submit an application to set aside the release decision, on the basis of new information, amounting to “a change in circumstances relating to the prisoner that occurred after the direction was given”. The application made clear the change in circumstances relied on was Mr Zenshen’s theft from his employer, and the full text of the COM’s email of 11 December 2025 was set out. In the light of this new information, the application asked that the Parole Board:

“... consider its implications for the assessment already made that [Mr Zenshen’s] case meets the requirements of the public protection release test”

stating:

“... The Secretary of State views with considerable concern that Mr Zenshen took such action, to steal items of low value, despite having significant savings and being so close to being released after nearly 35 years in prison. In light of this development, the Secretary of State respectfully requests the Parole Board sets aside its decision of 21 October and should decide again whether Mr Zenshen remains suitable for release.”

On 28 January 2026 the Parole Board (HH Peter Jones, sitting as a single member: “the decision maker”) refused the application to set aside. The letter communicating the set aside decision acknowledged:

“9.

The application for set aside submits that the decision for release would not have been made if new information had been available when the Panel made the decision to direct [Mr Zenshen’s] release.”

Having set out the relevant background, and the applicable legal test, it went on to explain the reasons for refusing the application, as follows (the SoS being referred to as “the Applicant”, Mr Zenshen as “the Respondent”):

“24.

Crucially, the professional judgement is that this behaviour does not indicate an increase in his risk of causing harm and it is not argued on behalf of the Applicant that there has been a change of circumstances such that the new information affects the risk management assessment.

25.

Nevertheless, in consequence of this new information, the Applicant submits that the release decision should be set aside.

26.

I am satisfied that this is new information which was not available to the panel on 21 October 2025 and, therefore, the further question for me is whether, had it been available, the panel would not have given a decision for release.

27.

The Respondent has not demonstrated overt violence against staff or prisoners since 1999 and, during his many years in prison, he has evidenced significant, sustained improvement in behaviour, has achieved all sentence objectives, engaged with professionals, and completed many treatments, programmes and risk-reduction interventions.

28.

On all relevant criteria he is assessed as posing a low risk of further offending and the prison psychologist assessed there to be a low risk of violence if he is released with low imminence.

29.

The panel found that the Respondent’s risk of reoffending in a manner likely to cause serious harm was not imminent, that he had been properly tested in the community, where he attends a religious centre regularly, and that his risks could be effectively managed in the community on the basis of a robust risk management plan.

30.

The application for set aside must therefore be considered against the background of these findings and the written and oral evidence before the panel and, in my view, crucially, the way in which the Respondent's behaviour has been dealt with by the prison authorities, in particular, the decision not to remove him from open prison conditions and the professional judgement that his behaviour does not indicate an increase in risk of harm.

31.

I do not criticise the decision to bring this new information to the attention of the Board and, as the COM notes, it will be necessary “to ponder” the Respondent's motivation for his behaviour which will no doubt be further considered with him while under supervision in the community, but, in all the circumstances, I am very far from satisfied that, had the panel been aware of this incident, it would not have given a direction for release.

32.

Given this finding it is not necessary for me to consider the question of the interests of justice.”

On 28 January 2026, the PPCS wrote to the Parole Board, saying para 24 of the set aside decision did not properly reflect the SoS’s position. By email of 29 January 2026 the Parole Board responded that it had “no power to amend this decision and are not of the view that the slip rule under rule 30 of the Rules should be utilised in this case”; it considered that it was functus.

During January and February 2026, the Ministry of Justice received correspondence from members of parliament raising concerns about Mr Zenshen’s release. Mr Davison has explained that officials and ministers reflected on the government's position, including the possibility of issuing judicial review proceedings, and steps were put in place to consider the appropriate course of action, which included personal consideration by the SoS.

On 20 February 2026, a MAPPA meeting took place. Noting Mr Zenshen’s release was scheduled for 11 March 2026, there was a discussion about the arrangements for this. Providing an update on what was said to be “significant involvement” from government ministers, Mr Davison referenced the power to refer release decisions to the High Court, but said that the provisional decision in Mr Zenshen’s case “was released too early for that power to be in effect”, further noting that, although judicial review might be a possibility, “there is no indication that ministers are going to do that”. Mr Davison also expressed the view that the attempted theft was “a troubling incident”, which needed to be “factored into the plan to manage [Mr Zenshen’s] risk upon release”.

On 23 February 2026, Mr Davison was notified that the SoS wished to discuss this case further with his special advisers, and a meeting took place on 5 March 2026, at which Mr Davison was instructed to seek a judicial review. Instructions were provided to the Government Legal Department and counsel the following day, and leading counsel was instructed on 8 March 2026.

On 11 March 2026, pursuant to the direction of the Parole Board, Mr Zenshen (now 61) was released from prison on life licence. Since his release, pursuant to his licence conditions, Mr Zenshen has resided at an AP in the community supervised by the probation service. It is not reported that any issues have arisen since his release. The evidence of Mr Creighton states:

“10.

I have been in regular contact with Mr Zenshen since his release to keep him informed of these proceedings and to take his instructions. He has informed me that he has found the transition from prison to the community to be very complex but that he feels he has coped with the challenges. ...

11.

Mr Zenshen has informed me that the proceedings have added an additional layer of stress to his release but that he has sought to manage this by speaking with his community probation team and the staff at his AP. Mr Zenshen has made it clear to me that he is fully committed to engaging with the legal process and fully accepts that this is a necessary consequence of his life licence.”

On 10 March 2026, a pre-action protocol letter was sent to the Parole Board and Mr Zenshen. An initial reply on behalf of Mr Zenshen, on 13 March 2026, made clear any application would be resisted, highlighting that the SoS’s delay (and Mr Zenshen’s release) raised issues going to the substance of the claim and to any relief, and questioning whether the SoS had exhausted all other remedies. The Parole Board also responded on 13 March 2026, stating it would take a neutral stance on the merits of the claim.

The SoS’s claim for judicial review was issued on 19 March 2026, accompanied by the statement of facts and grounds, and an application for expedition and a rolled-up hearing; Mr Davison’s statement was filed at the same time. Acknowledging that “in an ideal world” the decision to bring proceedings would have been taken more quickly, Mr Davison explains:

“52.

It is important to emphasise that a decision to seek a judicial review [of] a decision of the Parole Board is a significant step, which has only been taken on a very small number of occasions. It is a decision which was not taken lightly. As soon as the decision was taken, the claim has progressed very promptly ...”

Mr Davison has also stated that:

“42.

... The delay in taking the decision to proceed with judicial review was a key factor informing the decision not to apply for interim relief seeking to stay Mr Zenshen’s release”.

On 30 March 2026, Mould J ordered that this matter be expedited and listed for a rolled-up hearing, giving directions to enable that to take place.

An acknowledgement of service was filed by Mr Zenshen on 8 April 2026, with detailed grounds of resistance following on 23 April 2026. The Parole Board filed its acknowledgement of service on 10 April 2026, along with a legal position statement; it filed detailed grounds of resistance on 24 April 2026.

In R (Pearce) v Parole Board [2023] UKSC 13, [2023] AC 807, the Supreme Court described the Parole Board in the following terms:

“3.

The Board is a statutory body, in being since 1967 and presently established under section 239 of the Criminal Justice Act 2003 (“CJA 2003”). ... [I]t now has statutory responsibilities for itself making the decision about early release, that is to say release on licence sooner than the end of the court’s sentence. The Secretary of State (now of Justice) is obliged to follow any directions for release which it may give. In so doing, the Board acts judicially and as a body independent of the executive. It is properly treated as a court ...”

Although a court, in carrying out its statutory responsibilities, the Parole Board does not adopt an adversarial process; it is, rather, essentially inquisitorial, albeit with the participation of interested parties, such as the prisoner and (if he wishes) the SoS: Pearce paras 11, 57. As the Divisional Court explained in R (DSD) v Parole Board [2018] EWHC 694 (Admin), [2019] QB 285:

“117.

The evaluation of risk, central to the Parole Board’s judicial function, is in part inquisitorial. It is fully entitled, indeed obliged, to undertake a proactive role in examining all the available evidence and the submissions advanced, and it is not bound to accept the Secretary of State’s approach. The individual members of a panel, through their training and experience, possess or have acquired particular skills and expertise in the complex realm of risk assessment.”

By section 28 of the Crime (Sentences) Act 1997 (“the 1997 Act”), it is provided that, in the case of a prisoner who is serving a life sentence in respect of whom a minimum term order has been made, and who has served that minimum term:

“(5)

As soon as- ... (b) the Parole Board has directed his release under this section, it shall be the duty of the Secretary of State to release him on licence.”

Section 28ZA(5) sets out mandatory considerations which the Parole Board must take into account when applying the statutory release test, as follows:

“ (a) the nature and seriousness of the offence in respect of which the relevant sentence was imposed;

(b)

the nature and seriousness of any other offence for which the prisoner has at any time been convicted;

(c)

the conduct of the prisoner while serving the relevant sentence (whether in prison or on licence);

(d)

the risk that the prisoner would commit a further offence (whether or not specified in Schedule 18B to the Criminal Justice Act 2003) if no longer confined;

(e)

the risk that, if released on licence, the prisoner would fail to comply with one or more licence conditions;

(f)

any evidence of the effectiveness in reducing the risk the prisoner poses to the public of any treatment, education or training the prisoner has received or participated in while serving the relevant sentence;

(g)

any submissions made by or on behalf of the prisoner or the Secretary of State (whether or not on a matter mentioned in paragraphs (a) to (f)).”

The statutory release test has been considered on various occasions by the courts. Relevant to the assessment of risk, it has been made clear that, although the risk must be “substantial”, this can mean no more than that it is “not merely perceptible or minimal”: R v Parole Board, ex p Bradley [1991] 1 WLR 134; moreover, the test does not entail a balancing exercise, its goal is simply to safeguard the public from the danger posed by the prisoner: R (King) v Parole Board [2016] EWCA Civ 51, [2016] 1 WLR 1947 at para 31.

Proceedings before the Parole Board are governed by the Parole Board Rules 2019 (made under powers conferred by the Criminal Justice Act 2003 (“CJA 2003”)).

Under the heading “Case management and directions”, rule 6 provides a panel chair or duty member with wide-ranging case management powers, and expressly provides that:

“(11)

The panel chair or duty member may adjourn or defer the proceedings to obtain further information or for such other purpose as they consider appropriate.”

Rule 25 relates to decisions made by a panel of the Parole Board at an oral hearing; it makes clear that where such a decision is eligible for reconsideration under rule 28 it “is provisional” (rule 25(2)), only becoming final if no application for reconsideration is made within the relevant time period.

In cases such as the present (involving a prisoner serving an indeterminate sentence), rule 28 provides for a decision of the Parole Board to be reconsidered on the grounds that it contains an error of law, is irrational, or is procedurally unfair. An application for reconsideration must be made (and served on the other party) no later than 21 days after the decision in issue is sent to the parties.

Rule 28A provides that, on application by a party or on initiation by the Parole Board chair, the Parole Board may set aside a final decision made (relevantly) pursuant to rule 25(1). By rule 28A(5) it is made clear that:

“an application or initiation to set aside a decision ... must be made before the prisoner is released.”

If it is directed that a final decision should be set aside, by rule 28A(8) it is provided that a direction must also be made that the case should be:

“(a)

decided again on the papers by the previous panel or a new panel ..., or (b) decided again at an oral hearing by the previous panel or a new panel ...”

Pending the determination of an application to set aside, rule 28A(10) provides:

“Any requirement on the Secretary of State to give effect to a Parole Board direction to release a prisoner ... is suspended ...”

The power enabling the Parole Board to issue rules governing the making of a set aside decision is derived from section 239(5) CJA 2003; further provision in this regard is then made by subsections 239(5A)-(5C) CJA 2003, which make clear (see subsection (5C)) that this:

“(a)

may not confer power on the Board to set aside a direction for the release of a prisoner at any time when the prisoner has already been released pursuant to that direction, ...”

It is the SoS’s position (as set out in his post-hearing submissions) that, should this claim be allowed, and the set aside decision be quashed, the effect will be to restore the position to that before the making of the set aside decision, that is: there will be an outstanding set aside application made by the SoS prior to Mr Zenshen’s release, which will mean the requirement on the SoS to give effect to the Parole Board’s direction to release will remain suspended (rule 28A(10)) and, therefore, that Mr Zenshen will be required to return to custody pending the Parole Board’s (re-)determination of the set aside application. The position is not, the SoS submits, affected by section 239(5C) CJA 2003; it is the SoS’s case that quashing the set aside decision would mean that the direction to release in Mr Zenshen’s case would remain suspended such that it could not be said that Mr Zenshen had been released “pursuant to that direction”.

Responding to that argument, those acting for Mr Zenshen disagree, reasoning as follows: (i) the suspension of the release duty imposed by section 28(5) of the 1997 Act ceased to apply when the set aside decision was made on 28 January 2026; (ii) on 11 March 2026, the SoS, giving effect to the duty imposed by section 28(5), released Mr Zenshen from custody, recognising that the Parole Board’s release decision was no longer suspended; (iii) releasing a prisoner from custody in this context must be understood as a one-off act (not least given the importance of clarity and finality); (iv) that interpretation of rule 28A was consistent with the enabling provision under section 239(5C) of the CJA 2003; and (v) were the Parole Board to purport to again consider the set aside application, it would – as a matter of fact and on any ordinary understanding of the position – be doing so when Mr Zenshen had “already been released pursuant to the release direction”, something the Parole Board has no jurisdiction to do.

Returning to the way in which the Parole Board is to approach an application or initiation to set aside, by rule 28A(2) it is provided that such an application or initiation is to be considered on the papers. Rule 28A(3) states that a final decision may be set aside:

“... if— (a) it is in the interests of justice to do so; and (b) one or more of the conditions in paragraph (4) are satisfied.”

The conditions specified by rule 28A(4) are (relevantly) that:

“… (b) the decision maker is satisfied that a direction given by the Board for the release of a prisoner would not have been given if — (i) information that was not available to the Board when the direction was given had been so available, or (ii) a change in circumstances relating to the prisoner that occurred after the direction was given, had occurred before it was given.”

In R (Umar Jones) v The Parole Board of England and Wales and Secretary of State for Justice [2025] EWHC 3216 (Admin), Andrew Kinnier KC (sitting as a Deputy Judge of the High Court) was required to consider the nature of the test to be applied by the Parole Board on a set aside application under rule 28A. In that case, the application to set aside the Parole Board’s direction for Mr Jones’ release was made on the basis of seven “gists” from security intelligence that had not been available when the direction had been given. The material in question was, however, contested by Mr Jones, who made clear he denied the primary facts on which the allegations relied. In allowing the application to set aside, the decision-maker – expressing no view as to the reliability of the allegations made - found that, had this intelligence been before the panel, a direction for Mr Jones’ release would not have been made.

Allowing Mr Jones’ claim, Mr Kinnier KC concluded that, while the decision maker had correctly stated the test under rule 28A(4), he had failed to apply it. As Mr Kinnier explained:

“41.

... As a matter of grammar, the phrase “would not have been given” [in rule 28A(4)(b)(i)] represents the past conditional tense. It is a clear and straightforward phrase: it describes a situation in which the final decision would not have been made had certain information been available to the panel when it decided to direct [Mr Jones’] release. The phrase is not qualified by probability or likelihood.

42.

In order to form a view as to whether the final decision ‘would not have been given’ the decision-maker is required to consider the evidence relied upon in support of the set aside application in the context of the evidence before the panel which made the final decision. That exercise requires a sufficient consideration of its nature, relevance, reliability and cogency to allow the decision-maker to decide whether the information is such that the direction to release a prisoner would not have been given. What is sufficient will depend upon the facts of each case and, in particular, the nature, relevance, reliability and cogency of the evidence relied upon in support of an application to set aside a final decision.”

In Mr Jones’ case, as Mr Kinnier observed, although the decision maker had summarised the nature of the allegations, had noted Mr Jones’ strong denial and had emphasised the fundamental importance of fairness in fact-finding:

“44.

... he did not express a view on the reliability of the allegations sufficient to satisfy himself that had the new information been available, the direction to release would not have been given.”

That was particularly important in Mr Jones’ case, given his denial of the primary facts on which the allegations relied, putting those allegations squarely in issue; as Mr Kinnier continued:

“... In those circumstances, the judicial member could not have been satisfied that the direction for [Mr Jones’] release would not have been given without considering the reliability, credibility or accuracy of the allegations sufficient for the purposes of r. 28A. On the face of the decision, he did not do so. Instead, the judicial member effectively treated the question whether the allegations were reliable, credible or accurate as a matter for the new hearing which he directed should be held.”

Although unnecessary for his decision on Mr Jones’ case (upheld on the first ground of claim), Mr Kinnier went on to consider the further grounds of challenge. By his second ground, Mr Jones had complained that the Parole Board had breached its Tameside duty (SoS for Education and Science v Tameside MBC [1977] AC 1014) to obtain further information; Mr Kinnier considered that was a valid objection as it would not have been onerous for the decision maker to have asked for the reliability ratings of the intelligence material in issue, which would have given him the information necessary to make a lawful decision under rule 28A. Had, however, a lawful decision been reached that a release direction would not have been given had the new information been available, Mr Kinnier rejected Mr Jones’ third ground, that finality of judicial decision-making would mean that the interests of justice should weigh in favour of declining to set aside the original direction.

Pursuant to section 32ZAA of the 1997 Act, in specified cases, it is now open to the SoS to direct the referral of certain release decisions of the Parole Board to the High Court. Section 32ZAA provides:

“(1)

This section applies where- (a) a prisoner is serving a life sentence in respect of [a specified offence, including murder], and (b) the Parole Board directs the prisoner’s release under section 28(5) ....

(2)

The Secretary of State may direct the Parole Board to refer the prisoner’s case to the High Court if the Secretary of State considers that- (a) the release of the prisoner would be likely to undermine public confidence in the parole system, and (b) if the case were referred, the High Court might not be satisfied that it is no longer necessary for the protection of the public that the prisoner should be confined ...

(3)

The requirement for the Secretary of State to give effect to the Parole Board’s direction to release the prisoner is suspended- (a) during such period, beginning with the day on which the direction is given, as the Secretary of State reasonably requires to determine whether to direct the Parole Board to refer the prisoner’s case to the High Court under this section, and (b) if the Secretary of State gives such a direction, pending determination of the reference ....

(4)

Where the Secretary of State gives a direction under subsection (2), the Secretary of State must notify the prisoner of the direction and the reasons for giving it.

(5)

This section applies in relation to a prisoner whose sentence was imposed before, as well as after, this section comes into force.

(6)

But nothing in this section affects the duty of the Secretary of State to release a prisoner whose release has been directed by the Parole Board before this section comes into force.”

The referral power thus provided by section 32ZAA was added to the 1997 Act by section 6(1) of the Victims and Prisoners Act 2024, which was brought into force on 31 December 2025 (regulation 2 Victims and Prisoners Act 2004 (Commencement No.9) Regulation 2025/1325).

The view taken by the SoS (as expressed by Mr Davison at the MAPPA meeting on 20 February 2026) is that the referral power could not be exercised in Mr Zenshen’s case because the Parole Board’s direction for his release pre-dated its coming into force (see subsection 32ZAA(6)). Those acting for Mr Zenshen disagree, arguing that section 32ZAA is concerned with final (and binding, see R (Huxtable) v SoS for Justice [2021] EWCA Civ 1394, [2022] 1 WLR 813) decisions of the Parole Board, which give rise to the SoS’s duty to release the prisoner in question; in this case, the release decision was not final until after the determination of the reconsideration application, on 8 January 2026, which was after section 32ZAA had been brought into force.

The real issue between the SoS and Mr Zenshen in this regard is as to whether the referral power provided the SoS with a suitable alternative remedy, which he failed to use, such that I should refuse to entertain the application for judicial review (see the observations at para 50 Re McAleenon’s Application for Judicial Review [2024] UKSC 31, [2025] AC 1362). This is a point to which I return below but at this stage it is sufficient to note (per the Supreme Court in McAleenon): (1) judicial review by the High Court is ordinarily a remedy of last resort, ensuring the rule of law is respected where no other procedure is suitable to achieve that objective (R (Glencore Energy UK Ltd) v Revenue and Customs Comrs [2017] EWCA Civ 1716 at para 55, [2017] 4 WLR 213); (2) if other means of redress are conveniently and effectively available, they ought ordinarily to be used before resort to judicial review (Kay v Lambeth LBC [2006] UKHL 10 at para 30, [2006] 2 AC 465; R (Watch Tower Bible & Tract Society of Britain) v Charity Commission [2016] EWCA Civ 154 at para 19, [2016] 1 WLR 2625); (3) where Parliament has enacted a statutory scheme for appeals in respect of certain decisions, that will, in ordinary circumstances, be regarded as a suitable alternative remedy in relation to such decisions, which ought to be pursued rather than a judicial review (Glencore Energy, paras 55-58; Watch Tower Bible & Tract Society, para 19); and (4) where, however, there is no statutory right of appeal, but, instead, different forms of legal proceeding (one of which is judicial review), each giving rise to different issues and their own associated risks and costs, a claimant might be afforded greater latitude in deciding which claim to bring: the question whether they have a suitable alternative remedy will fall to be addressed by reference to the type of claim they have chosen to bring and what relief they have sought (McAleenon, paras 52-55).

Referral applications under section 32ZAA are to be made by way of a modified CPR Part 8 claim. Assuming that it had been open to the SoS to utilise section 32ZAA, the powers of the High Court on such a referral are specified by section 32ZAC, as follows:

“(1)

On a referral of a prisoner’s case under section 32ZAA ..., the High Court- (a) must, if satisfied that it is no longer necessary for the protection of the public that the prisoner should be confined, make an order requiring the Secretary of State to give effect to the Parole Board’s direction to release the prisoner on licence; (b) otherwise, must make an order quashing the Parole Board’s direction to release the prisoner on licence.

...

(3)

An order under subsection (1)(b) ... has effect as if the prisoner’s case were disposed of by the Parole Board on the date on which the order was made.”

The Ministry of Justice has published non-statutory guidance relating to the “Applicability of the Secretary of State Referral Power”. In that guidance, considering how the referral power might be used, it is observed that:

“13.

The reconsideration mechanism is available to ask the Parole Board to review a provisional release decision for irrationality, errors of law or procedure. In cases where there are legal flaws in a decision, the Secretary of State would likely apply for reconsideration than referral (though there may be cases where referral is considered more appropriate).”

Further explaining:

“14.

In order to refer a case, the Secretary of State must have reason to believe that the Parole Board has reached the ‘wrong’ conclusion on the risk of a prisoner, to the extent that the release test is not met. ... there must be sufficient objective evidence of the Secretary of State finding a difference of view on the prisoner’s risk which could lead to a different conclusion on the release test, indicated or supported by a number of features of the decision. ....”

For completeness, I also note that, pursuant to section 32(1) of the 1997 Act, the SoS has the power to revoke the licence of a prisoner and recall them to custody. I mention this solely because in R (McCourt) v Parole Board [2020] EWHC 433 (Admin), the Divisional Court, at para 20, expressed the view that recall to prison might be a potential remedy were Mrs McCourt to succeed in her challenge to a release decision when the prisoner had already been released.

The SoS’s power of recall is exercised in accordance with the Recall, Review and Re-Release of Recalled Prisoners Policy Framework and is itself a matter that has been the subject of consideration in the case-law. This is not, however, a power the SoS has sought to exercise in Mr Zenshen’s case (nor is it suggested that he ought to have done); if he were to do so, Mr Zenshen’s case would then be referred to the Parole Board (see section 32(4) of the 1997 Act), which would then proceed to make a new release decision (it is not suggested that this would simply be a decision as to whether or not to set-aside the earlier release decision).

It is right to record that the specialist expertise of the Parole Board in its evaluation of complex risks is not in dispute (and see DSD at para 117, supra). This has led the courts on numerous occasions to emphasise the need for caution when judicial intervention is sought in relation to the exercise of judgement in this “specialist domain” (see DSD at paras 118-119, and the cases there cited). This is, moreover, a caution that, it has been made clear, extends to the procedure the Parole Board might choose to adopt in order to arrive at its decision (R (James & ors) v SoS for Justice (Parole Board intervening) [2010] 1 AC 553 per Lord Judge CJ at para 134, [2009] UKHL 22).

The current claim is pursued on grounds of both process and outcome rationality. As Chamberlain J observed in KP v Secretary of State for Foreign, Commonwealth and Development Affairs [2025] EWHC 370 (Admin):

“55…rationality is the standard by which the common law measures the conduct of a public decision-maker where there has been no infringement of a legal right, no misdirection of law and no procedural unfairness. It encompasses both the process of reasoning by which a decision is reached (sometimes referred to as “process rationality”) and the outcome (“outcome rationality”) ...”

Process rationality “focuses on the decision-making process” (per Lady Hale in Braganza v BP Shipping Ltd [2015] UKSC 17 at para 24, [2015] 1 WLR 1661). It allows that the court may intervene where the decision-maker has:

“…taken into account matters which they ought not to take into account, or conversely, have refused to take into account or neglected to take into account matters which they ought to take into account.”per Lord Greene MR Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223 at pp 233-234

But it is not limited to that; as Chamberlain J explained in KP:

“56.

... In addition, the process of reasoning should contain no logical error or critical gap. This is the type of irrationality Sedley J was describing when he spoke of a decision that “does not add up – in which, in other words, there is an error of reasoning which robs the decision of logic”: R v Parliamentary Commissioner for Administration ex p. Balchin [1998] 1 PLR 1, [13]. In similar vein, Saini J said that the court should ask, “does the conclusion follow from the evidence or is there an unexplained evidential gap or leap in reasoning which fails to justify the conclusion?”: R (Wells) v Parole Board [2019] EWHC 2710 (Admin), at [33].”

Where it is contended that there has been a failure to take into account relevant considerations, a claimant must show those matters are expressly or impliedly identified in the legislation or policy as required to be taken into account, or:

“... on the facts of the case, they were ‘so obviously material’ as to require direct consideration”: R (Samuel Smith Old Brewery (Tadcaster)) v North Yorkshire Council [2020] 3 All ER 527, para 32.

Thus, in general terms, it will be for the decision-maker to decide what is relevant or irrelevant, albeit that decision will be subject to rationality review: R (Friends of the Earth Ltd) v SoS Transport [2020] UKSC 52 at paras 116-121, [2021] 2 All ER 967. And, similarly, a Tameside duty - requiring that decision makers ask the right questions, and take reasonable steps to inform themselves with the relevant information to answer those questions - recognises it is the responsibility of the decision maker to evaluate the material available, and for them (and not the court) to decide upon the manner and intensity of the enquiry to be undertaken; the court should intervene only if no reasonable decision maker could have been satisfied on the basis of the enquiries made that they possessed all the information necessary for their decision: R (Balajigari) v SoS for the Home Department [2019] EWCA Civ 673 para 70, [2019] 1 WLR 4647; R (Campaign Against Arms Trade) v Secretary of State for International Trade [2019] EWCA Civ 1020, [2019] 1 WLR 5765 paras 58-59.

Outcome rationality, on the other hand, is concerned with whether – even where the process of reasoning leading to the challenged decision is not materially flawed – the outcome is “so unreasonable that no reasonable authority could ever have come to it” (Wednesbury) or, “in simpler and less question-begging terms, outside the “range of reasonable decisions open to a decision-maker” (Boddington v British Transport Police [1999] 2 AC 143, 175)”: KP para 57.

It is not in dispute that a claim for judicial review must be filed:

“(a)

promptly; and (b) in an event not later than 3 months after the grounds to make the claim first arose”: CPR 54.5(1)

Moreover, as Lady Black observed in Mauritius Shipping Corporation v Employment Relations Tribunal [2019] UKPC 42, at para 8, [2020] 1 All ER 844 (there referring to substantially the same requirement under Mauritian law):

“The primary requirement is that the application be made promptly. The requirement that it be made in any event within three months is a “longstop”. It is well established ... that, depending on the circumstances of the case, an application made within three months may not qualify as having been made promptly, in which case, leave to apply for judicial review may be refused.”

The need for speed and promptness in judicial review claims is not disputed (see per Carr LJ in R (GLP) v SoS Health [2022] EWCA Civ 355 at para 38, [2022] 1 WLR 2339); the point is usually put in terms of the need for public authorities to have certainty as to the validity of their decisions and actions (GLP at para 39; R (Law Society of England and Wales) v Legal Services Commission [2010] EWHC 2550 (Admin), [2011] Costs LR Online 57 at para 116), but it is also recognised that prejudice to the rights of third parties will be a relevant consideration (see per Lord Lloyd-Jones Maharaj v National Energy Corporation of Trinidad and Tobago [2019] UKPC 5 at para 37, [2019] 1 WLR 983). As always, context is crucial: given the nature of the decision under challenge, has the claimant proceeded with reasonable promptness? As Lord Sales emphasised in R (TN (Vietnam)) v Secretary of State for the Home Department [2021] 1 WLR 4902:

“89.

... the expectation is that a claimant will proceed with their case with reasonable promptness after they become aware or could reasonably be expected to have become aware of the grounds for the claim. ... Claimants who delay without good reason may not be granted permission to proceed.”

And, in this regard, section 31(6) of the Senior Courts Act 1981 provides that:

“Where the High Court considers that there has been undue delay in making an application for judicial review, the court may refuse to grant- (a) leave to the making of the application, or (b) any relief sought on the application, if it considers that the granting of the relief sought would be likely to cause substantial hardship to, or substantially prejudice the rights of, any person or would be detrimental to good administration.”

An extension of time for bringing judicial review proceedings may be made pursuant to CPR 3.1(2)(a). On such an application, the approach of the court will be the same as for an application for relief from sanctions under CPR 3.9: applying “the same rigour”, having regard to “the seriousness and significance of the failure to apply within time, the reason for that failure and all the circumstances of the case” (per Lewis LJ R (Barot) v SSHD [2026] EWCA Civ 218 at paras 26 and 29). And at para 38 Maharaj it was emphasised that:

“... the statutory test is not one of good reason for delay but the broader test of good reason for extending time. This will be likely to bring in many considerations beyond those relevant to an objectively good reason for the delay, including the importance of the issues, the prospect of success, the presence or absence of prejudice or detriment to good administration, and the public interest.”

As explained at para 6.4.4.2 of the Administrative Court Guide:

“In considering whether to grant an extension of time, the Court must first determine the date from which the relevant time period started to run so that the period of delay can be calculated correctly. The Court will then consider all the circumstances, including whether an adequate explanation has been given for the delay, the importance of the issues, the prospects of success and whether an extension will cause substantial hardship or prejudice to the defendant or any other party or be detrimental to good administration.”

The SoS contends that the Parole Board’s set aside decision was irrational: (1) as a matter of process, and (2) in outcome.

In advancing his process irrationality challenge, the SoS submits that the Parole Board: (i) failed to consider the key factors said to justify release in the release decision and to consider the impact of the new information (the attempted theft and Mr Zenshen’s lack of explanation for it) on those factors, as required by rule 28A(4)(b) (Jones para 42), which amounted both to a failure to take into account obviously material considerations and gave rise to a “critical gap” (KP para 56) in its reasoning; (ii) had (notwithstanding its inquisitorial function) failed to consider/make inquiries in relation to Mr Zenshen’s motivation for the theft, suggesting this was a matter the probation service could investigate following his release, which amounted to a breach of its Tameside duty, a failure to take into account obviously relevant information, and a critical gap in its reasoning (as the purpose of the statutory release test was to safeguard the public, potential risks identifiable from the new information could not be thus delegated); (iii) stated that “crucially” the SoS had not argued “there has been a change of circumstances such that the new information affects the risk management assessment”, when that was inaccurate; (iv) failed to take into account that many of Mr Zenshen’s previous convictions were for theft, robbery and burglary, wrongly assuming that the risk assessment would not have changed; and (v) failed to take into account the relevance of the new information to the likelihood that Mr Zenshen would comply with his licence conditions.

As for the outcome irrationality challenge, the SoS further contends that the set aside decision was Wednesbury irrational. In this regard, it is submitted that, if the theft incident had occurred shortly before the oral hearing on 21 October 2025, the Parole Board would have either not directed release, on the basis that a further period of testing in the community was required, or would have exercised its powers under rule 6(11) of the Parole Board Rules and postponed its decision to obtain further assessments from the relevant witnesses, including, in particular, the HMPPS psychologist.

In keeping with established case law (see R (Stokes) v Gwent Magistrates Court [2001] EWHC Admin 569), the Parole Board (widely recognised to be acting as a court when making release decisions) adopts a neutral stance, seeking to provide assistance and information to the court, not to contest the proceedings.

Adopting this position, should relief be granted as sought by the SoS, the Parole Board has raised the following points: (1) so far as the Parole Board is aware, section 239(5C) CJA 2003 has not been the subject of consideration by the courts, but its present understanding is that it has no statutory power to set aside a direction for the release of a prisoner after that prisoner has been released pursuant to that direction; and (2) to the extent that the set aside decision may nevertheless be re-made, the court might wish to exercise caution in directing that take place within a specific time period (see per Lane J at para 77 R (Bousfield) v Parole Board [2021] EWHC 3160 (Admin)).

Mr Zenshen resists the claim, contending that permission to apply for judicial review should be refused/the claim should be dismissed on its merits. Before addressing the merits, however, those acting for Mr Zenshen raise three preliminary objections: (1) the time point; (2) the question of suitable alternative remedy; and (3) whether the claim is now futile.

First, in the context of decisions impacting on liberty (clearly intended to be resolved, if at all possible, prior to release), Mr Zenshen contends that the SoS’s seven-week delay meant the claim was not brought “promptly” and was out of time. The undue delay meant relief sought by the SoS would cause substantial hardship to Mr Zenshen and prejudice his rights, not least given the significant legal and practical issues concerning the quashing of a decision not to set aside a release decision (and the question whether the decision could be re-made post release). The SoS had failed to engage with these issues prior to being prompted to do so by the court at the oral hearing. The clarification provided made clear the prejudice to Mr Zenshen was such that, even if the court considered there was “good reason” for the delay (not accepted), it should nevertheless refuse to grant any extension of time. Second, it is said the court should decline to entertain the application for judicial review when the SoS had failed to use the suitable alternative remedy provided by the power of referral to the High Court under section 32ZAA of the 1997 Act; Mr Zenshen contends it is wrong to consider this was not an available remedy because the release decision pre-dated section 32ZAA coming into force. Third, Mr Zenshen submits the claim is futile, because it ultimately seeks further testing in circumstances where that has now taken place: Mr Zenshen has now lived in the community for over two months and it is inconceivable that a Parole Board, given this evidence of sustained successful risk management, would do other than maintain the release.

Turning to the merits, Mr Zenshen makes the general point that an irrationality challenge imports a very high hurdle, particularly where, as here, it is concerned with a specialist risk assessment decision (axiomatically falling squarely within the Board’s wide discretion) taken by a very experienced judicial member of the Parole Board.

As for the process irrationality ground, Mr Zenshen contends: (i) the set aside mechanism was clearly intended to be a summary, on-paper, process, which did not call for a lengthy decision or for re-running the original process: the decision maker was well aware of the content of the release decision and the documents that underpinned it (expressly stating these were considered) and it was not arguable there had been any failure to take into account relevant considerations or any “critical gap” in reasoning; (ii) the SoS’s second point was really an outcome rationality challenge – an attack on the substantive conclusion (at para 31 of the set aside decision) – but the decision-maker had considered the evidence regarding motivation and was entitled to take the view that further enquiries were unnecessary (and it was notable that, in the set aside application, the SoS had not suggested the additional steps that were now relied on); (iii) as for how the decision-maker had read the set aside application, the SoS’s argument was hopeless: the application had not made the positive assertion now relied on, but, in any event, the decision maker had plainly understood and addressed the nature of the determination he was required to make; (iv) equally, the decision maker had obviously been aware of Mr Zenshen’s previous convictions (apparent from the release decision and the dossier that underpinned it) and could be taken to have had the full background in mind in reaching his decision; and (v) the same was true of the licence conditions: the decision maker plainly had these in mind but was entitled to give weight to the COM’s opinion that Mr Zenshen’s risk could be managed in the community.

Finally, it is Mr Zenshen’s case that the outcome irrationality argument is hopeless, it being unarguable that the decision not to set aside the release decision on the basis of the new information (attempted theft of goods valued at around £30) was so outrageous in its defiance of logic that no reasonable Parole Board, applying its mind to the question posed under rule 28A(4), could have made it. The set aside decision involved the exercise of a judicial function, concerning the assessment of risk, and the court should be very slow to interfere.

In addressing the issues raised on this rolled-up hearing, it is convenient to first consider Mr Zenshen’s contention that the claim was brought out of time.

The three month limit under CPR 54.5(1) is the longstop, not the primary requirement (Mauritius Shipping), and judicial review proceedings that have not been brought “promptly” may be out of time even when brought within three months. Determining whether or not a claim has been brought “promptly” will, however, always require a case-specific assessment, having regard to the nature and context of the decision under challenge. In the present case, the immediate consequence of the decision in issue was that the SoS’s obligation to give effect to the Parole Board’s release decision was no longer suspended (which had been the consequence of the set aside application: rule 28A(10) Parole Board Rules). Thus, at the point the set aside decision was communicated (28 January 2026), the SoS was aware that, by virtue of section 28(5) of the 1997 Act, he was then required to release Mr Zenshen pursuant to the Parole Board’s direction (provisionally issued on 4 November 2025; rendered final by the reconsideration decision of 8 January 2026). As the minutes of the MAPPA meeting of 20 February 2026 make clear, the SoS was kept informed of the arrangements being put in place to ensure compliance with that direction, with Mr Zenshen’s release being set for 11 March 2026.

Although I am unsure as to whether the SoS’s decision regarding the bringing of judicial review proceedings took into account allthe potential implications of delaying until after Mr Zenshen’s release (indeed, it was because there was a lack of clarity about the SoS’s position as to the effect on Mr Zenshen of the relief sought (an order quashing the set aside decision) that it was necessary for further written submissions to be made after the oral hearing), it is apparent the SoS was alive to the significance of this event and as to the prejudice that would be suffered by Mr Zenshen if he then had to be returned to custody (as Mr Davison has said, this informed the SoS’s decision not to seek interim relief). Although the commencement of proceedings prior to his release might not have favoured Mr Zenshen (the indication is that the SoS would then have applied for interim relief to delay his release), seen from his perspective, the failure to issue this claim for some seven weeks after the communication of the set aside decision has led to greater uncertainty at a time when he is making the transition to non-custodial life (and I have taken note of what Mr Creighton has said about the effect on Mr Zenshen in this regard). In such circumstances, the need for promptness and speed (per Carr LJ, para 38 GLP) would seem paramount.

Recognising the potential prejudice for Mr Zenshen (particularly if the SoS is correct that the immediate consequence of the claim being upheld would be his return to custody – all the more difficult given he has now spent over two months released on licence), I must, however, also have regard to other perspectives when assessing the issue of promptness in this case. Mr Davison has explained that it was an unusual and significant step to seek a judicial review of a decision of the Parole Board, and it required the personal consideration of the SoS; I accept that evidence, and that this was not a decision that was taken lightly. Recognising, as I do, that this was a complex matter, requiring reflection on the part of the SoS (and those advising him), I am unable to say that the time taken (effectively the five weeks between the communication of the set aside decision and the SoS’s instruction to commence proceedings) was not reasonably prompt. Thereafter, the lawyers involved moved with some speed, sending out the letter before claim within five days, and issuing proceedings nine days later. As Mr Davison has acknowledged, considering Mr Zenshen’s individual position, the time taken was not “ideal”, and I do not underestimate the impact this will have had, but I also accept that there were a number of other considerations which the SoS had to take into account and, having regard to the full context, I am prepared to accept that this claim was brought “promptly” for the purposes of CPR 54.5(1).

Given my conclusion that the claim was brought in time, it is unnecessary for me to rule on the SoS’s (precautionary) application for an extension of time.

It is Mr Zenshen’s submission that the court should decline to entertain this application for judicial review given that it was open to the SoS to use the alternative remedy provided by section 32ZAA of the 1997 Act, whereby this matter could have been referred to the High Court. In argument, this was characterised as akin to a statutory appeal, providing suitable alternative relief, which the SoS ought to have pursued in preference to seeking judicial review (Glencore Energy, paras 55-58; Watch Tower Bible & Tract Society, para 19).

As I have referenced at para 60 of this judgment, there is a preliminary issue between Mr Zenshen and the SoS as to whether it would have been open to the SoS to seek to use the referral power afforded by section 32ZAA, given that the initial release decision in this case pre-dated its commencement (see section 32ZAA(6)). On this point I consider there is merit in the submissions made on Mr Zenshen’s behalf: the focus of section 32ZAA is on the SoS’s duty to release, which only arises once the release direction of the Parole Board becomes final, which, in this case, was after section 32ZAA had been brought into force. That focus, in my judgement, is made clear by subsection 32ZAA(6), which expressly refers to the duty of the SoS to release a prisoner whose release has been directed by the Parole Board, which was, in Mr Zenshen’s case, only after the determination of the reconsideration application on 8 January 2026.

Proceeding then on the basis that it had been open to the SoS to refer Mr Zenshen’s case to the High Court under section 32ZAA, I do not, however, agree that this provided a suitable alternative means of relief such as to mean that the SoS ought to have utilised the referral power in preference to pursuing the present claim for judicial review.

The first point to make in this regard is that section 32ZAA is not akin to a statutory appeal process. As section 32ZAC makes clear, upon a referral under section 32ZAA, the High Court is not reviewing the decision of the Parole Board; it is, rather, required to reach its own view as to whether it is no longer necessary for the protection of the public that the prisoner should be confined.

Secondly, as the language of section 32ZAC makes clear, the referral power does not provide a means of challenging a set aside decision made by the Parole Board under rule 28A of the Parole Board Rules. Allowing that it is possible that, if he had considered this was a power open to him at the time, the SoS might have wished to use the section 32ZAA referral process in Mr Zenshen’s case, that would, nevertheless, have afforded a very different form of relief to that sought in these proceedings. Concerned, as I am, with this claim, I cannot see that section 32ZAA would have provided a suitable alternative means of challenging the set aside decision that is its particular focus.

Accepting, therefore, that there might have been different forms of legal proceeding available to the SoS (the power to refer Mr Zenshen’s case to the High Court; the ability to seek judicial review of the set aside decision), it is apparent that each would give rise to different questions for the court, and, if successful, different forms of relief. Focusing on what is sought by the SoS in these proceedings – a challenge solely to the set aside decision - I am satisfied this is not a claim for judicial review that the court should decline to entertain because of the existence of an alternative – very different – form of relief.

Although raised as a further preliminary objection, Mr Zenshen’s “futility” argument really relates to potential questions of relief. Had it been necessary for me to consider the issue of relief, for reasons I will later explain, this is a point on which I would wish to receive further oral submissions.

In considering the merits of the SoS’s challenge to the set aside decision, it is common ground that I should adopt the approach to rule 28A of the Parole Board Rules laid down in Jones. Thus, in determining whether the conditions specified by rule 28A(4) are satisfied, the decision maker must have concluded that the release decision “would nothave been given”, a phrase not qualified by probability or likelihood (Jones, para 41). In order to be able to determine whether the release decision “would not have been given”, the decision maker will have to give “sufficient consideration” to the “nature, relevance, reliability and cogency” of the evidence relied upon in support of the set aside application (Jones, para 42). What will be “sufficient” will depend upon the facts of the case and upon the evidence relied upon (in particular, its “nature, relevance, reliability and cogency”) in support of the application (Jones,para 42).

In Jones, the allegations relied on in support of the application to set aside were just that: allegations. Specifically, seven “gists” of security intelligence were identified which contained allegations relating to Mr Jones, who vigorously denied them, pointing out that these matters had neither been raised with him, nor the Parole Board panel charged with determining the question of his release. Notwithstanding Mr Jones’ denial of the primary facts on which the new information was based, in allowing the application to set aside the release decision, the decision maker did not make any finding as to the truth or otherwise of the allegations. That, the court held, was an error: it meant the decision maker could not have determined whether, given the new information, the release decision would not have been given. Moreover, although not necessary for its decision to uphold Mr Jones’ application for judicial review, the court considered there had been a breach of the Tameside duty in the failure to ask the SoS to provide the reliability ratings attached to the allegations in issue (by the time of the court hearing, it was known that a “low” reliability rating had been attributed to most of the “gists”).

There are a number of points of distinction between Mr Jones’ case and that of Mr Zenshen. Most obviously, there was no dispute of fact in respect of the new information relied on in Mr Zenshen’s case: Mr Zenshen accepted he had been caught attempting to steal (low-value) items from his employer; he had no explanation for his actions but did not deny what he had done. The nature of the evidence relied on in support of the application to set aside the release decision was not in dispute; it was plainly both reliable and cogent. The question for the decision maker was not one of fact (as to the truth or otherwise of what was alleged); he was, rather, required to give sufficient consideration to the relevance of the new information so as to be able to determine whether – in the light of that information - the release decision would not have been given. He thus had to evaluate whether this required further investigation/additional information, and, more generally, form a view as to how the information impacted on the assessment of risk underpinning the release decision; these were all matters that clearly fell within the particular expertise of the decision maker – an experienced judicial Parole Board member – and a court should be slow to interfere with an exercise of judgement (substantive or procedural) in such circumstances (DSD paras 117-119; James para 134).

More than that, as those acting for the SoS have accepted, for this claim to succeed it must meet the high threshold of showing the set aside decision was irrational, whether as a matter of process or outcome; it would not be sufficient to merely point to some inadequacy of reasoning. The SoS’s claim is expressly not put as a reasons challenge and, although I would not necessarily wish to accept the characterisation of the set aside procedure as being “summary” in nature (this, it seems to me, will rather depend on the new information in issue and any further investigation this might require), it would be wrong to assume that every failure to mention a specific point demonstrates a gap in the reasoning, still less that the final conclusion was perverse.

Ground 1 is put squarely as a claim of process irrationality: a failure to take into account that which was obviously relevant, or a critical gap in the process of reasoning (KP, para 56; Balchin, para 13; Wells, para 33). It is part of the SoS’s case in this regard that the decision maker failed to understand the way in which the application had been advanced, erroneously stating that “crucially” the SoS had not argued “there has been a change of circumstances such that the new information affects the risk management assessment”.

I have started with this objection because I consider it reflects a failure to engage with the substance of the reasoning provided for the set aside decision, an error of approach that can be seen to infect other aspects of the SoS’s case. On its face, the set aside application did not expressly state there had been a change of circumstances such that the new information affected the risk management assessment; at most, in the final paragraph, the Parole Board was asked to consider the “implications” of the new information “for the assessment already made”. That said, it is in any event clear that the decision maker did consider whether the new information had implications for the risk assessment that had informed the release decision, finding: “Crucially, the professional judgement is that this behaviour does not indicate an increase in [Mr Zenshen’s] causing harm”. The professional judgement thus referenced was that of Mr Zenshen’s COM, who had made clear (as recorded within the set aside application) that the new information “does not indicate an increase in risk of harm”; it would seem to be this that the decision maker considered to be crucial. The decision demonstrates a clear understanding of the question raised by the application (see para 9), and a careful consideration of the new information in the context of the evidence and assessments that had informed the release decision. There can be no proper objection to the approach thus adopted; it was not tainted by any error of understanding as to the nature of the application.

Separately, the SoS has objected to what is said to be the decision maker’s failure to consider the key factors relied on in the release decision and to then consider the impact of the new information on those factors. In particular, the SoS points to the references in the release decision to “sustained evidence of a pro-social attitude”, “absence of any ongoing problematic traits”, “reasonable prospects for employment”, and “more than adequate evidence of Mr Zenshen’s insight, coping skills and motivation/wherewithal to lead a prosocial life”; the point is made that the new information, relating to the unexplained attempted theft, would seem to stand in contradiction to these factors, and the apparent failure to engage with that amounts to an error or a gap in the reasoning process.

The concerns identified by the SoS are not invalid: Mr Zenshen’s attempted theft from his employer – a clear breach of trust in a number of respects – raised “unanswered queries” (as it was put in the email discussion between Mr Zenshen’s COM and Mr Singh Here on 5 and 6 January 2026, see para 29 above). It is, however, apparent that the decision maker had this very much in mind, identifying that (as the COM had opined) “it will be necessary “to ponder” [Mr Zenshen’s] motivation for his behaviour”.

The question raised by the set aside application was, however, whether the decision maker was satisfied that the decision for release would not have been made if the new information had been available. The list of features identified at para 4.14 of the release decision formed part of the panel’s assessment of risk, whereby - taking account of all the factors identified, together with the proposed risk management plan, and considering what was necessary for the protection of the public - it had concluded that “the risk could be effectively managed in the community” (release decision para 4.15). I can see it might arguably have been more helpful for the set aside decision to expressly refer to each of the features listed at para 4.14 of the release decision, explaining in what way they were considered to be impacted (or not) by the new information; to that extent, I would give permission for the SoS’s claim. Upon proper consideration, however, it is apparent that this is really a complaint as to the detail of the reasons provided, it does not reveal an error or gap in the process of reasoning such as to rob the decision of logic (Balchin para 13; Wells para 33; KP para 56).

Critically, the set aside decision demonstrates a clear focus on the assessment of risk, and, throughout the decision, it is apparent that the decision maker was mindful of the primary requirement that the Parole Board had to be satisfied that it was no longer necessary for the protection of the public that Mr Zenshen should be confined (section 28(6)(b) of the 1997 Act). Concerned with the need to safeguard the public from the danger that Mr Zenshen might pose, the decision demonstrates a careful consideration of the evidence available (both in the dossier underpinning the release decision, and in the new information that had become available) to assess whether there was a change in circumstances such as would affect the risk management assessment. Thus having regard to the substance of the decision, there is no error or irrationality in the process of reasoning.

The SoS, however, seeks to make an additional point, relating to what is said to amount to a breach of the Parole Board’s Tameside duty, in failing to make further inquiries into Mr Zenshen’s motivation for the theft. In this regard, it is suggested that, whether or not this had been suggested in the set aside application, consistent with the inquisitorial role of the Parole Board, the decision maker might have sought a further psychological assessment, exercising his power to adjourn to obtain that information, pursuant to rule 6(11) of the Parole Board Rules. Additionally, it is contended that the decision maker was wrong to delegate the assessment of risk to the probation service.

In considering the application to set aside, the decision maker evidently took into account the reasoning that had led to the release decision, and, in particular, to the regard had to the psychologist’s report and recommendation; it is also apparent that he gave careful consideration himself to that psychological assessment. In evaluating the impact of the new information, it was for the decision maker to determine the manner and intensity of the enquiry required (Balajigari; Campaign Against Arms Trade). Determining that Mr Zenshen’s motivation for the attempted theft could be further considered under supervision in the community was not a delegation of the assessment of risk; it was, rather, the consequence of the decision maker’s evaluation of the impact of the new information on that assessment. The SoS’s objection in this regard seeks to attack that evaluation, but, as those acting for Mr Zenshen have put it, this is really an outcome challenge, dressed up as process rationality. Given the prison psychologist’s assessment of there being a low risk of violence, with low imminence, taken together with the recognition in the release decision of there being effective means of managing risk in the community, I am unable to say that the decision maker’s evaluation of the enquiry necessitated by the new information was irrational.

Although the new information did not reveal any act, or threat, of violence by Mr Zenshen, the SoS further complains that, in rejecting the set aside application, there was a failure to take into account that many of Mr Zenshen’s previous convictions were for theft, robbery and burglary (the record suggests that between 1975 and 1985 (when Mr Zenshen was a child or young adult) he had some 23 previous convictions for “theft and kindred offences”), and that it was wrongly assumed that the previous assessment of risk would have remained unchanged. The SoS says the decision maker thus failed to have regard to relevant matters, in particular noting that Mr Zenshen’s rehabilitative work revealed how difficult it had been for him to address his anti-social attitudes and offence supporting beliefs.

As the SoS has acknowledged, the dossier relating to Mr Zenshen - included within the material considered on the set aside application - contained all the available information regarding his previous convictions, together with the detail of the therapeutic and rehabilitative work he had undertaken over a number of years to address his core beliefs, thinking, and behaviour relevant to his earlier offending. The decision maker was a very experienced judge and member of the Parole Board, who expressly stated that he had considered the dossier, alongside the release decision, the set aside application, and other relevant material. I do not infer that a failure to expressly refer to those earlier convictions in the set aside decision suggests that this part of the factual background was not taken into account by the decision maker.

Moreover, the substance of the set aside decision makes clear that the decision maker applied the correct test: considering whether the new information was such that, had it been available to the panel, the decision for release would not have been made. That required the decision maker to evaluate the impact of the new information on the risk assessment at the heart of the release decision, namely that: Mr Zenshen’s “risk of reoffending in a manner likely to cause serious harm was not imminent ... and that his risks could be effectively managed in the community on the basis of a robust risk management plan”. Accepting, as I do, that the decision maker had regard to the information contained within the release decision and the dossier that underpinned it, it cannot be said there was a failure to take into account relevant material. Equally, having regard to the test that the decision maker was required to apply, to the focus on the risk of harm in the release decision, and then to the substance of the decision in respect of the set aside application, no error or gap in the process of the reasoning can be identified.

Finally, under the heading of process irrationality, the SoS contends that the set aside decision reveals a failure to take into account the relevance of the new information to the likelihood that Mr Zenshen would comply with his licence conditions. In this regard, it is submitted that it is no answer to suggest that this should be assumed to be implicit: given the materiality of the issue, the SoS says that it was incumbent on the Parole Board to deal with this matter expressly.

The objection raised seeks to infer that the decision maker’s failure to expressly reference the question of compliance with licence conditions must mean he did not have regard to this plainly material factor. Given the specialist experience of the decision maker (both as judge and Parole Board member), this is not an inference that I would readily draw, not least as the set aside decision: (i) took into account the detailed release decision, which set out licence conditions proposed in Mr Zenshen’s case; (ii) expressly referred to the risk management plan, which plainly included the licence conditions that would apply; and (iii) demonstrated careful consideration of the opinion of Mr Zenshen’s COM, that his risk could be managed in the community, which (again) included the licence conditions to which he would be subject. I do not find that the decision maker failed to have regard to the relevance of the new information to the likelihood that Mr Zenshen would comply with his licence conditions; on the contrary, I am satisfied that this was a factor that was taken into account when considering the impact of the new information on the earlier conclusion that Mr Zenshen’s risks could be “effectively managed in the community on the basis of a robust risk management plan” (set aside decision para 29).

Having considered the SoS’s “process rationality” objections individually, I have then stood back to consider the decision maker’s reasoning taken as a whole. Doing so, I am satisfied that he applied the correct test, and, exercising the broad case management discretion afforded him under rule 6(2) of the Parole Board Rules, formed a rational view of the scope and nature of the enquiry required. This was not a case such as that of Mr Jones, where a fact finding assessment had to be undertaken; here, there was no issue as to the reliability and cogency of the new information, and the task for the decision maker was one of evaluating the relevance and impact of that information on the assessment of risk that informed the original release decision. Acknowledging that the determination of the set aside application took full account of the release decision and the dossier that underpinned it, it cannot be said that there was a failure to have regard to matters that were relevant to the task required under rule 28A. Furthermore, considering the substance of the decision, no error of approach, or gap in the process of reasoning is revealed. Although I would be prepared to allow that permission should be given on ground 1, I am satisfied that it is then to be dismissed on its merits.

In advancing the SoS’s case under ground 2, it is accepted that this faces a particularly high threshold: it must be demonstrated that, notwithstanding the specialist nature of the determination in issue, the outcome – the refusal to set aside the release decision – fell outside the range of reasonable responses open to the decision maker, applying the correct test to the circumstances of this case (Boddington, p 175; KP, para 57). For the SoS, it is nevertheless contended that: (i) the language of rule 28A(4) must allow for a situation where the Parole Board would have felt unable to reach a final view on whether to direct release and would then have adjourned its decision pending the obtaining of further information; and (ii) given the critical importance of understanding Mr Zenshen’s (unexplained) motivation for the attempted theft, had the new information been made available to the panel shortly before/at the oral hearing, the only rational course would have been for the panel to then adjourn to obtain additional evidence and/or assessments (in particular from the prison psychologist).

Although the language of rule 28A(4) of the Parole Board Rules requires the decision maker to be satisfied that the release direction “would not have been given”, I accept that this could encompass a decision to adjourn to obtain further information under rule 6 (11), or to take some other step, such as would have the effect that the release direction would not then have been given. Such a possibility would, however, have been something firmly within the understanding and experience of the decision maker, who was well placed to determine whether or not such a step would have been taken. The set aside decision evinces a clear focus on the reasoning that had led to the release decision in Mr Zenshen’s case, and on what matters had weighed with the panel (which, I note, had referred to the availability of psychological support in the community; release decision, para 4.11). Considering the same evidence together with the new information, it is apparent that the decision maker did not judge that it was necessary to adjourn to obtain additional information, or to request a further report.

The SoS contends, however, that, had the new information become available at/shortly before the hearing on 21 October 2025, the panel could not lawfully have moved to direct Mr Zenshen’s release, but would have been bound to have adjourned to make further enquiries, in particular as to how the new information might impact upon the psychological assessment, and as to Mr Zenshen’s motivation more generally. That, it seems to me, is a straightforward, and impermissible, substitution of the SoS’s view for that of the expert decision making entity in this case. The unarguable nature of the challenge in this regard is apparent when one considers the complex and specialist assessment that had to be undertaken.

Cases such as Mr Zenshen’s plainly involve a very difficult assessment of risk. I started this judgment by setting out the terrible facts of the murder that led to Mr Zenshen’s incarceration; the brutal nature of the crime he committed, and the serious and immediate danger he posed, led to a recommendation that “life” in his case should indeed mean “life”. I do not lose sight of the gravity of Mr Zenshen’s crime, or of the view formed by the judge who presided over his trial. More generally, I recognise that Mr Zenshen remained an extremely violent man throughout his 20s, and did not demonstrate a willingness to address his behaviour until he was in his 40s.

These are, however, all matters that each of the experienced professionals and Parole Board members plainly had well in mind when making their recommendations and decisions in this case. It was for them to evaluate the risk that Mr Zenshen might continue to pose, having regard to the work he had undertaken over some 12 years, through a variety of specialist programmes and in different settings. The importance and complexity of that evaluation will have been obvious to all involved. The reasoning provided for the release decision makes it apparent that the assessment of risk at all times focused on the question of public protection that lies at the heart of the statutory release test, with a clear understanding of the mandatory considerations that must inform the application of that test. Undertaking the task required of it, the original panel concluded that it was no longer necessary for the protection of the public that Mr Zenshen remains confined in prison.

Returning to the evaluation of risk in the light of the new information, the decision maker at the set aside stage was best placed to assess how that new information would have impacted (applying the past conditional tense, per Jones) upon the original release decision. Having undertaken a detailed assessment of the reasoning leading to the release direction, along with the evidence that underpinned it, the decision maker was “very far from satisfied” that, had the panel been aware of this incident, it would have done other than give a direction for Mr Zenshen’s release. Allowing for the complexity of the evaluation required, and the expertise of those involved, I am unable to say that conclusion fell outside the range of reasonable responses open to this decision maker. On this point, I do not consider an arguable ground has been shown.

Given my conclusion on each of the grounds of claim, it is unnecessary for me to consider the question of relief. I make clear, however, that, had I concluded that the claim ought to be upheld on ground (1), I would have wished to receive further oral submissions on the question of relief. That, it seems to me, would be necessary given the further written clarification of the SoS’s position in this regard, and the impact that would have on Mr Zenshen’s position, notwithstanding his release on licence for over two months.

For the reasons provided, I: (1) grant permission on ground 1, but dismiss the claim on this ground on the merits; and (2) refuse permission on ground 2.