A post to draw attention to an interesting administrative-law case decided by the Supreme Court today. The central question in R (Bourgass) v Secretary of State for Justice [2015] UKSC 54 was straightforward: Had the prison authorities acted lawfully by subjecting the claimant prisoners to solitary confinement for unbroken periods of several months? Both prisoners were alleged to have been involved — although, in one of the two cases, only indirectly — in assaults upon another prisoner, and to have proselytised. The Court concluded that the claimants’ solitary confinement (or “segregation”) had been lawful for the first 72 hours but unlawful thereafter. This post is concerned with two issues that the Court considered in Bourgass; it does not address a further issue concerning the applicability of Article 6 ECHR to segregation decisions.

The Carltona Doctrine

The Supreme Court’s conclusion was reached principally on the basis of statutory construction and the application of the ultra vires principle, the question simply being whether the relevant legislation authorised solitary confinement beyond 72 hours without the involvement of the Secretary of State. Rule 45 of the Prison Rules provides as follows:

(1)  Where it appears desirable, for the maintenance of good order or discipline or in his own interests, that a prisoner should not associate with other prisoners, either generally or for particular purposes, the governor may arrange for the prisoner’s removal from association accordingly.

(2)  A prisoner shall not be removed under this rule for a period of more than 72 hours without the authority of the Secretary of State and authority given under this paragraph shall be for a period not exceeding 14 days but it may be renewed from time to time for a like period.

(3)  The governor may arrange at his discretion for a prisoner removed under this rule to resume association with other prisoners at any time.

Notwithstanding rule 45(2), in the claimants’ cases — and, indeed, as was standard practice — the continuation of solitary confinement beyond 72 hours was not authorised by the Secretary of State, but by prison a manager. However, it was argued before the Court that rule 45(2) did not actually require the involvement of the Secretary of State on account of the applicability of the Carltona doctrine. That doctrine stipulates that, in general, government officials for whose decisions the Secretary of State is constitutionally accountable may exercise powers conferred by statute upon the Secretary of State.

There is no suggestion in Bourgass that Carltona does not apply to the Secretary of State’s power under rule 45(2). It would therefore be perfectly lawful for one of the Secretary of State’s departmental officials to grant a rule 45(2) authorisation. The question, however, is whether the Carltona doctrine permits the exercise of the power to be delegated further, such that it can be exercised by prison, as distinct from departmental, officials. The Supreme Court held not.

It rejected parallels that counsel attempted to draw with cases — like R v Secretary of State for Social Security, ex parte Sherwin (1996) 32 BMLR 1 — holding that Carltona can legitimise the exercise of ministerial powers by executive agencies. In Sherwin, noted Lord Reed (writing for a unanimous Court), detailed evidence as to the relationship between the agency and the department had been supplied; in contrast, in Bourgass, the Court had been provided “with only minimal information” about the relationship between prisoners, the Prison Service and the Ministry of Justice, and with “no information about the governance arrangements or the arrangements in relation to accountability to Parliament”. However, the Court inferred that the relationship between prison officials and the Secretary of State was not such as to warrant application of Carltona. For example, Lord Reed noted that whereas departmental officials are inherently required to act on a Minister’s instructions, the Prison Rules (in certain instances) explicitly oblige prison governors and officers to do so. The need to impose such an explicit legal obligation, observed Lord Reed, demonstrates that the relationship between prison governors and the Secretary of State “bear[s] no resemblance to those governing the relationship between a minister and his departmental officials”.

Lord Reed thus concluded that the constitutional rationale underpinning the Carltona doctrine — namely, the existence of a relationship between the Secretary of State and the relevant official implying an adequate degree of constitutional accountability of the latter to the former — was absent. And so the doctrine did not apply. Indeed, he went further, at least implying that even if the rationale had not been absent, the operation of Carltona would have been legislatively excluded by operation of what he took to be the rationale underpinning rule 45(2):

The apparent rationale of rule 45(2) is clear. The governor can order segregation at his own hand for a maximum of 72 hours, but any longer period requires the authorisation of the Secretary of State – in practice, senior officials from outside the prison – in order to protect the prisoner against the risk of segregation for an unduly protracted period.

The implication is that any possibility under Carltona of the rule 45(2) power’s being exercised by officials within a prison would be displaced by the statutorily implied necessity of supplying, for the protection of prisoners, the form of external scrutiny provided by the involvement the Secretary of State or his departmental officials.

This aspect of the judgment is to be welcomed. It serves as a fitting reminder that the Carltona doctrine is not — as its application in the agency context may have been taken to imply — infinitely elastic, and demonstrates a healthy willingness of the part of the Supreme Court to scrutinise with some care whether the constitutional concerns raised by delegations of ministerial authority are adequately assuaged by the reality of the relevant institutional relationships.

Procedural Fairness

Lord Reed noted at the beginning of his judgment that the parties’ printed cases had focussed upon the procedural fairness of the process adopted when making the rule 45(2) continuation decisions. Although the conclusion that the prison authorities had acted straightforwardly ultra vires was sufficient to dispose of the case, Lord Reed went on the address the procedural-fairness issues on account of their general importance. The essence of the argument here was that the common-law standard of procedural fairness had not been met: so little information had been given to the prisoners about the reasons for their confinement that any opportunities to make representations as to why segregation should not be continued were rendered nugatory. This argument had been rejected in the Court of Appeal in Bourgass ([2012] EWCA Civ 376, [2012] 1 WLR 3602) because it had considered itself bound by its own decision in R v Deputy Governor of Parkhurst Prison, ex parte Hague [1992] 1 AC 58. In Hague, Taylor LJ said:

No doubt in many cases the governor will be able, as here, to give reasons at the time of the decision or shortly after. But the same considerations of public policy as persuaded me … to hold that reasons are not in law required as a matter of course before a decision to segregate may apply with equal force after the decision. Again, the guiding factors must be the subject matter and the circumstances … I would not be prepared to hold that in all cases a prisoner has a legal right to be given the reasons for his segregation.

The use of the term “reasons” here is somewhat unhelpful, the issue being not whether there is a free-standing entitlement to ex post reasons for a segregation decision, but whether, in order to ensure the fairness of continuation decisions, what is more commonly called “notice” must be provided. The point, however, is that in such circumstances, “reasons” and “notice” are functionally indistinguishable, the reasons for a decision to subject a prisoner to solitary confinement amounting to the same thing as notice of the case in favour of continuing segregation. The crucial issue, then, is whether the quality of procedural justice due to a prisoner who is subject to ongoing solitary confinement is such as to entitle him or her to an amount information about the perceived case in favour of confinement sufficient to facilitate the making of meaningful representations.

In Hague, the Court of Appeal had little time for this argument: it noted that, unlike certain other provisions in the Prison Rules, what is now rule 45 sets out no procedural requirements. From this, the Court inferred that no such requirements obtained in respect of rule 45. In Bourgass, however, Lord Reed observed that since Hague was decided, the “law relating to procedural fairness has not stood still”, pointing out that in R (Osborn) v Parole Board [2013] UKSC 61, [2014] AC 1115 the Supreme Court had emphasised

both the instrumental value of enabling persons to participate in decision-making when they may be able to contribute relevant information or to test other information before the decision-maker, and the ethical value of allowing persons to participate in decision-making which concerns them and is liable to have a significant effect on their rights or interests, where they may have something to say which is relevant to the decision to be taken.

Against this background, Lord Reed concluded that “a prisoner should normally have a reasonable opportunity to make representations before a decision is taken by the Secretary of State under rule 45(2)”, and that such an opportunity would be “largely valueless unless he knows the substance of the case being advanced in sufficient detail to enable him to respond”. Lord Reed held that this standard had not been met in respect of the claimants: telling someone that they are in solitary confinement “pending an investigation into a serious assault” or because “we are trying to transfer you [to another prison” was wholly inadequate.

It did not, however, follow that in every segregation case prisoners would be entitled to full chapter and verse: the objective of facilitating meaningful participation might sometimes have to be pursued in a way that also takes account of the safety of other prisoners or the general maintenance of prison order. In such circumstances, said Lord Reed, “It will be sufficient to inform the prisoner in more or less general terms of the gist of the reasons for seeking the authority of the Secretary of State.” Of course, this does not answer the really hard question that arises when (for example) the preservation of others’ safety tells against any disclosure to the prisoner of crucial information, without which the right to make representations would be emptied of meaningful content. That is a question with which the courts have wrestled in other contexts, and need not be dwelled upon here.

Instead, it suffices to say that the rejection of the Court of Appeal’s decision in Hague represents a progressive step, and underlines the willingness of the courts to read an obligation to act fairly into a statutory scheme. What is particularly to be welcomed is the Supreme Court’s rejection of the approach in Hague whereby the duty to act fairly was taken to have been effectively excluded by implication (on the ground that other rules imposed, but that rule 45 did not explicitly impose, procedural obligations upon decision-makers).

In this sense, Bourgass is of a piece with classic modern authorities on procedural fairness at common law, in which the court’s willingness to read in procedural duties — as well as the extent of such duties as are read in — turns principally upon the seriousness of what is at stake for the individual. Against this background, it is worth emphasising that while placing someone in solitary confinement for months on end may be necessary, it is no small matter. As Lord Reed pointed out, successive UN Special Rapporteurs on Torture and Other Cruel, Inhuman or Degrading Treatment have highlighted the sometimes-irreversible psychological effects of prolonged solitary confinement, as has the European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment. In the light of this, irrespective of whether the instrumental case for natural justice is engaged, the ethical value of permitting meaningful participation is clear. As Lord Reed put it:

The imposition of prolonged periods of solitary confinement on the basis of what are, in substance, secret and unchallengeable allegations is, or should be, unacceptable.

It is hard to argue with that.