R (on the application of O) v Secretary of State for the Home Department

The appellant, O, is a woman of Nigerian nationality, aged 38. In November 2003, with her son, then aged three, she illegally entered the UK. In July 2008 she pleaded guilty to offences of cruelty towards her son, who had returned to live in Nigeria, and the court sentenced her to 12 months’ imprisonment and recommended that she be deported. On 8 August 2008 her sentence came to an end, whereupon the respondent, the Home Secretary, detained her - at first pending the making of a deportation order and then, following the making of such an order, pending her deportation pursuant to it. O’s detention, which was at the Immigration Removal Centre (“IRC”) at Yarl’s Wood in Bedfordshire, continued until 6 July 2011 when, pursuant to a grant of bail on 1 July 2011, she was released. It follows that O was detained at Yarl’s Wood for almost three years. The court knows nothing about her circumstances after 6 July 2011 but infers that she has not, or not yet, been deported.

In her first claim for judicial review, O, acting (as now) by the Official Solicitor, her litigation friend, challenged the lawfulness of the earlier period of her detention, namely from 8 August 2008 to 22 July 2010. In the Administrative Court her claim failed entirely but on appeal it succeeded to a limited extent. By its decision, entitled R (OM) v Secretary of State for the Home Department [2011] EWCA Civ 909 and dated 28 July 2011, the Court of Appeal held that for most of that earlier period, namely until 28 April 2010, O had been the subject of unlawful detention but was entitled only to nominal damages in respect of it and that for the remainder of that earlier period she had not been the subject of unlawful detention at all.

In the present proceedings, which - chronologically - encompass her fourth claim for judicial review, O challenges the lawfulness of the later period of her detention, namely from 22 July 2010, and in particular from say 4 March 2011, until 6 July 2011. The object of the present proceedings has never been to secure her release, which had already occurred at the time of their issue. The object has been to secure a declaration that the detention was unlawful and, perhaps in particular, an award of substantial damages for false imprisonment. But on 3 April 2012 Lang J refused to grant permission for this claim to proceed and on 17 July 2014 the Court of Appeal (by a panel which comprised Arden LJ, who gave the substantive judgment, and Underhill and Floyd LJJ, who agreed with it) dismissed her appeal: [2014] EWCA Civ 990, [2015] 1 WLR 641. O now asks this court to grant permission for the claim to proceed and therefore to remit it to the Administrative Court, so that, following the filing by the Home Secretary of detailed grounds for contesting it and of any written evidence on which she wished to rely, it might proceed to substantive determination.

O has the misfortune to have suffered for many years from serious mental ill-health. So the appeal requires this court to consider the Home Secretary’s policy relating to the detention of the mentally ill pending deportation; and perhaps also to identify the criterion by which the court should determine a complaint that she has failed to implement some aspect of her policy relating to it. Furthermore the Home Secretary is obliged to conduct monthly reviews of whether a person’s detention pending deportation should continue. There were, as the Court of Appeal held, defects in the Home Secretary’s conduct of the monthly reviews of O’s detention between March and July 2011. The appeal requires this court to identify the effect of the deficiencies on the lawfulness of her detention during those four months, particularly in the light of the Court of Appeal’s decision in R (Francis) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2014] EWCA Civ 718, [2015] 1 WLR 567. Supported by Bail for Immigration Detainees which, jointly with Medical Justice, also intervenes in the present appeal, O contends that the Francis case was wrongly decided.

(a) It was within days of their arrival in the UK in November 2003 that O

During the period of her imprisonment and of her detention at Yarl’s Wood O displayed signs of serious mental ill-health, including by a number of attempts at suicide and other acts of self-harm; by suffering hallucinations; and by unpredictable mood-swings and impulsive outbursts. There she was mainly treated with high doses of anti-psychotic and anti-depressant medication.

In May 2008, for the purposes of the court in sentencing her for the offence of child cruelty, Dr Olajubu, a specialist registrar in forensic psychiatry, diagnosed O as suffering a recurrent depressive disorder and an emotionally unstable personality disorder. He considered that in prison she would have access to all appropriate psychological interventions.

On 30 April 2009 Professor Katona, a consultant psychiatrist, made the first of a series of reports on O at her request. At that time he agreed with the diagnosis of Dr Olajubu but on 21 September 2009 he reported a considerable deterioration in O’s condition and recommended that the Home Secretary should direct her transfer from Yarl’s Wood to hospital under section 48 of the Mental Health Act 1983.

On 12 March 2010, following an attempt to suffocate herself, O was admitted to the psychiatric wing of Bedford Hospital for assessment. On 15 March 2010 Dr Ratnayake, a consultant psychiatrist there, led the assessment and, by letter of discharge to Yarl’s Wood of that date, he expressed agreement with Dr Olajubu’s diagnosis of O as having an emotionally unstable personality disorder, which Dr Ratnayake said was of a borderline type. He added that his team found no true psychosis in her and that her needs, in particular for constant observation and continued medication, would be adequately met at Yarl’s Wood, to where accordingly she was returned.

On 16 July 2010 Professor Katona reported that, although O’s acts of self-harm had become somewhat less frequent and her depression less profound, he maintained his recommendation for her transfer to hospital.

On 10 February 2011 Dr Agnew-Davies, a clinical psychologist with special expertise on the impact of trauma on the mental health of women, reported on O. Her report forms the foundation of this appeal. It runs to 69 pages. Instructions to her to make the report came from O’s solicitors, who asked her to comment in particular upon whether O’s detention was detrimental to her mental health and upon the effect on it of her forcible return to Nigeria. Dr Agnew-Davies reported:

that O gave a plausible history of having suffered frequent physical and sexual abuse at the hands of an uncle when she was aged between 11 and 14 and in his care in Nigeria;

that staff at Yarl’s Wood had told her that since the summer 2010 O’s behaviour had been more stable and that she had undertaken a short course of cognitive behavioural therapy;

that her study of O’s records, her lengthy interview with O and the results of application to her of mainstream psychological tests led her to diagnose in O not only a major depressive disorder but, in particular, a severe, complex and chronic form of post-traumatic stress disorder (“PTSD”) arising out of her uncle’s protracted abuse of her;

that, unlike Dr Ratnayake, she considered that O could not access the necessary mental health services at Yarl’s Wood;

that release from detention would greatly benefit her mental health;

that she needed a long-term structured package of mental health services;

that she was not fit to live independently without professional support;

that recovery of her mental health could occur only over the long term;

that neither medication nor general counselling services would alone be enough to secure her recovery;

that she needed to be referred to a specialist trauma-focussed psychiatric clinic, such as those in London provided by St Bartholomew’s Hospital, by the Maudsley Hospital and by three others, for treatment which would take place in phases over years;

that such a referral was in accordance with guidance issued by the National Institute for Health and Care Excellence (“NICE”) to the effect that neither general counselling services nor treatment with medication could alone provide sufficient interventions in a severe, complex case of this sort; and

that her deportation to Nigeria would have grave effects upon her mental health and be likely to precipitate unsuccessful attempts at suicide followed perhaps by a successful one.

On 30 June 2011, for the purposes of the application for bail which proved successful on the following day, Dr Agnew-Davies wrote an addendum report. She noted the apparent absence of any acts of self-harm on O’s part during the previous six months and, from her psychological perspective, she urged O’s immediate release. She recommended that O should receive medical support in the community from a home treatment team and later from a community mental health team and, as before, that in the long term O should engage in treatment at a specialist clinic.

Also on 30 June 2011 Professor Katona wrote a further report in which, without having again interviewed O, he reviewed the reports of Dr Agnew-Davies and the up to date medical records from Yarl’s Wood. In the light of the marked improvement in O’s self-harming behaviour, he withdrew his recommendation for her transfer to hospital. He now agreed with Dr Agnew-Davies that O was suffering PTSD in addition to her depressive disorder. He also agreed with her recommendation for O to access medical care in the community and in the long term for her to undertake treatment at a specialist clinic, which, he added, would probably continue for two or three years.

Paragraph 2 of Schedule 3 to the Immigration Act 1971 (“the 1971 Act”), entitled “Detention or control pending deportation”, provides:

In that she had been the subject of a recommendation for deportation, it follows that from 8 August 2008 to 25 November 2010 O was detained under para 2(1) of Schedule 3 to the 1971 Act and that, from 25 November 2010, when the deportation order was made in respect of her, until 6 July 2011 she was detained under the words in parenthesis in para 2(3) of the schedule.

In that the Home Secretary made the deportation order in accordance with section 32(5) of the 2007 Act, it is worthwhile to note section 36(2) of that Act, which provides:

“[I]mmigration detention powers need to be transparently identified through formulated policy statements”, observed Lord Dyson in R (Lumba) v Secretary of State for the Home Department (JUSTICE and another intervening)[2011] UKSC 12, [2012] 1 AC 245 at para 34.

The Home Secretary’s published policy in this regard is set out in Chapter 55, entitled “Detention and Temporary Release”, of a manual addressed to caseworkers and entitled “Enforcement Instructions and Guidance” (“the manual”). It states:

at para 55.1.1 that the power to detain had to be retained in the interests of maintaining effective immigration control but that there was a presumption in favour of release;

at para 55.1.2 that the presumption applied even to foreign national offenders (such as O) but that, in relation to detention pending their intended deportation, the risks of their re-offending and absconding might well outweigh it; and

at para 55.8 that, following the start of any detention, reviews of it were necessary in order to ensure that it remained lawful and in line with policy; that in a criminal case (such as that of O) they should take place at least every 28 days; and that the law required detainees to be provided every 28 days with written reasons for their continued detention, based on the outcome of the reviews.

Paragraph 55.10 of the manual is entitled “Persons considered unsuitable for detention”. It states:

Those suffering from serious mental illness which cannot be satisfactorily managed within detention …”

On 14 January 2016 Mr Stephen Shaw CBE made a report to the Home Secretary entitled “Review into the Welfare in Detention of Vulnerable Persons”, Cm 9186. His eleventh recommendation was that the phrase “satisfactorily managed” should be removed from para 55.10 of the manual. Mr Shaw noted suggestions that the meaning of the phrase was inexact and obscure and he stated that, irrespective of whether it was “satisfactorily managed”, serious mental illness among detainees was clearly not being treated in accordance with good psychiatric practice.

A central inquiry mandated by this appeal is into the treatment of the report of Dr Agnew-Davies in the Home Secretary’s reviews of O’s detention.

The report was submitted to the Home Secretary under cover of a letter from O’s solicitors dated 16 February 2011. It is of some relevance that it was expressly submitted in support of O’s application, then recently issued, for judicial review of the Home Secretary’s certificate that the application to revoke the deportation order had been clearly unfounded. In the letter O’s solicitors quoted at some length from the report and stressed passages relevant to the claim for judicial review, including doubts about O’s ability to conduct an out-of-country appeal and the risk of her suicide in the event of deportation. Although in the letter they did refer to the diagnosis of PTSD, the solicitors did not refer to the recommendation of treatment at a specialist clinic in London; did not allege that O’s illness could not be “satisfactorily managed” at Yarl’s Wood; and, generally, did not question the legality of O’s continued detention in the short term.

By letter dated 8 April 2011 the Home Secretary, by her caseworker, replied to the letter dated 16 February 2011. Again the context of the letter was O’s claim for judicial review rather than the legality of her continued detention in the short term; and the gist of it was that the Home Secretary found nothing in the report of Dr Agnew-Davies to lead her to abandon her defence of the claim. Presumably in an attempt to show that she had carefully read it, the writer quoted at length from the report, including that the doctor had diagnosed PTSD. Oddly, however, she then twice asserted that the report contained no new diagnosis. She said that, in response to Dr Agnew-Davies’ report, the medical officers at Yarl’s Wood had explained that O’s condition had become more stable; that her last attempt at self-harm had occurred more than a year previously; and that she was compliant with her medication.

In the six reviews of O’s detention which were written between 4 March 2011 and 4 July 2011, each prepared by the caseworker who wrote the letter dated 8 April 2011 and each duly countersigned by senior officers, only the briefest reference was made to the report of Dr Agnew-Davies. Inserted into the lengthy recital in each review of O’s protracted immigration history was reference to “yet another psychiatric report”, which had been “treated as a further request to revoke” the deportation order. Again oddly, the reviews identified O’s most recent diagnosis as being that of Dr Ratnayake on 15 March 2010. In each case the senior officers in effect indorsed the caseworker’s conclusion that the risk of O’s reoffending and absconding outweighed the presumption in favour of release.

One has some sympathy for the caseworker because the report of Dr Agnew-Davies had been submitted to the Home Secretary as relevant to an issue different from that of the legality of O’s continued detention in the short term. Nevertheless on any view the report bore some relevance to the Home Secretary’s policy relating to the detention of the mentally ill and should have been properly addressed in the reviews. The reviews

failed to refer to Dr Agnew-Davies’ diagnosis of O as suffering PTSD;

indeed wrongly stated that the most recent diagnosis of O’s mental condition was that of Dr Ratnayake;

failed to refer to Dr Agnew-Davies’ assessment of O’s need for treatment at a specialist trauma-focussed psychiatric clinic; and

failed therefore to consider whether O could be “satisfactorily managed” at Yarl’s Wood and, even if not, whether there were very exceptional circumstances which nevertheless justified her continued detention.

In the above circumstances the Court of Appeal concluded that the Home Secretary had unlawfully failed to apply the policy set out in para 55.10 of the manual when deciding to continue to detain O between March and July 2011. This conclusion the Home Secretary now accepts. She does not suggest that the evidence which she would be entitled to file in the event that the claim was permitted to proceed would be likely to throw a different light on it. The defects in the reviews already filed speak for themselves.

Appendix 4 to Mr Shaw’s recent report, referred to in para 20 above, is an assessment by Mr Jeremy Johnson QC of six High Court cases in which since 2010 the Home Secretary’s treatment of immigration detainees has been held to be inhuman or degrading and therefore in violation of their rights under article 3 of the ECHR. He also assessed at least six other cases in which, without identifying a violation of article 3, the High Court or the Court of Appeal held the detention to have been unlawful. For the purposes of these assessments Mr Johnson made a detailed study of the Home Secretary’s detention reviews and concluded:

The next question is: were she to have applied her policy correctly, how would the Home Secretary have reacted to the report of Dr Agnew-Davies? The first part of the answer is to consider the meaning of the phrase “satisfactory management”. There is lively dispute between the parties as to the nature of the court’s review of the legality of the Home Secretary’s application of policy (which presupposes that she has purported to apply it: see para 37 below). But in this appeal there is no dispute that the court’s approach to the meaning of the policy is to determine it for itself and not to ask whether the meaning which the Home Secretary has attributed to it is reasonable: R (Raissi) v Secretary of State for the Home Department[2008] EWCA Civ 72, [2008] QB 836, paras 107 to 123.

NHS England is responsible for commissioning the provision of all health services in IRCs as well as in prisons in England pursuant to regulations made under section 3B(1)(c) of the National Health Service Act 2006. What level of health services should NHS England arrange to be provided there? The answer is to be found in the following two principles identified in the Partnership Agreement, first published in 2013 and republished in April 2015, between Home Office Immigration Enforcement, NHS England and Public Health England, at p 12:

Health and wellbeing services in IRCs should seek to improve health and wellbeing (including parity of esteem between services which address mental and physical health) …”

In formulating policy that, save very exceptionally, management of serious mental illness in an IRC, if not “satisfactory”, should precipitate release, the Home Secretary has adopted a word of extreme and appropriate elasticity. It catches a host of different factors to which the circumstances of the individual case may require her to have regard. In R (Das) v Secretary of State for the Home Department (Mind and another intervening) [2014] EWCA Civ 45, [2014] 1 WLR 3538, in a judgment with which Moses and Underhill LJJ agreed, Beatson LJ, at paras 45 to 47 and 65 to 70, offered a valuable discussion of the phrase “satisfactory management”. I respectfully disagree with him only in relation to an aside in para 71 of his judgment. Beatson LJ there expressed an inclination to accept the Home Secretary’s contention that, if the management of the illness in an IRC was likely to prevent its deterioration, it would be satisfactory even if treatment was available in the community which was likely to secure its improvement. I would not exclude the relevance of treatment, available to the detainee only if released, which would be likely to effect a positive improvement in her (or his) condition. If it was likely that such treatment would actually be made available to the detainee (rather than be no more than on offer in principle to all members of the community in NHS publications), its availability should go into the melting-pot; and the burden would be upon the Home Secretary to inquire into its availability. If, contrary to the Partnership Agreement quoted in para 29 above, the standard of care (expressly aimed at improving health as well, of course, as preventing it from deteriorating) provided to a detainee in an IRC were for some reason not equal to that which would be made available to her if released, it would in my view be questionable, subject to the strength of other relevant factors, whether the management of her illness in the IRC was satisfactory. While satisfactory management does not mean optimal management, a narrow construction of the word “management” as meaning no more than “control” of the illness would lack principled foundation, particularly when in very exceptional circumstances the detainee may continue to be detained in the IRC pursuant to the policy notwithstanding the unsatisfactory management of her illness there.

Above all the policy in para 55.10 of the manual mandates a practical inquiry. As Beatson LJ stressed in the Das case, the phrase “satisfactory management” should be interpreted with regard to its context and purpose (para 45); should not be subjected to the fine analysis appropriate to a statute (para 47); nor invested with a spurious degree of precision (para 65). An important part of its context is that the management of the illness takes place in detention pending likely deportation. Treatment of a patient who finds herself in the doubly stressful circumstances both of detention and of likely deportation has its own considerable, extra challenges; treatment in those circumstances might be satisfactory even if it would not otherwise be satisfactory.

The reliance by Dr Agnew-Davies on highly generalised words of guidance issued by NICE, set out in para 11(k) above, leads O to refer the court to the paragraph in The NHS Constitution for England, updated to 14 October 2015, which tells the patient:

Had she sought to ask herself whether, in the light of the report of Dr Agnew-Davies, O’s illness would satisfactorily be managed at Yarl’s Wood, the Home Secretary would have sought to obtain answers to questions along the following broad lines:

Was Dr Agnew-Davies likely to be correct in diagnosing PTSD in O? No one had previously diagnosed it. In particular it had been diagnosed neither by Dr Ratnayake nor (until 30 June 2011) by Professor Katona.

In particular did the clinicians treating O at Yarl’s Wood agree with the diagnosis and, if not, what diagnosis did they favour? Their intimate and protracted exposure to O might, subject to the quality of their response, invest their views with considerable authority.

What was the nature of the treatment currently provided to O at Yarl’s Wood?

How satisfactory did the clinicians regard the current treatment and would they confirm the improved stability of O’s behaviour reported by Dr Agnew-Davies?

Was it necessary to instruct an independent psychologist to comment on the diagnosis of Dr Agnew-Davies?

What was the likely length of time before the Home Secretary could achieve O’s deportation? In February 2011 she had come close to achieving it but O’s second claim for judicial review in relation to revocation of the deportation order was pending.

Insofar as Dr Agnew-Davies was recommending that O needed at once to embark on lengthy treatment at a specialist trauma-focussed clinic, would the doctor responsible for O approve it and was the recommendation in any way practical? Was there evidence that any such clinic could and would accept O, as a foreign citizen awaiting deportation, even for immediate assessment let alone for early treatment?

If O were released into the community, what accommodation should be provided for her and would its location be compatible with her need to undergo the treatments appropriate for her?

What medical services (in particular, what mental health services) and what local authority community care services would be available to O in her locality immediately following any release?

Realistically O accepts that the proper application of the Home Secretary’s policy to her case in the light of the report of Dr Agnew-Davies would not have led to her immediate release in March 2011. She correctly contends that the report should have led the Home Secretary to make inquiries. We cannot predict the result of the inquiries, most of which, judged by the contents of the reviews, seem never to have been made. Indeed, even if, which is doubtful and which indeed the Court of Appeal expressly rejected, the appropriate conclusion would or might have been that O’s illness could not be satisfactorily managed in detention, the Home Secretary, in considering whether there were very exceptional circumstances which nevertheless justified her continued detention, would have had to consider the risks of her absconsion and (possibly also) re-offending. On 1 July 2011 the tribunal judge rated them as acceptably low. But, in his judgment given later that month on O’s appeal in her first claim for judicial review, and therefore by reference to the circumstances which existed only up to 22 July 2010, Richards LJ at para 36 assessed the risk of her absconsion as very high. At least, however, the limited period between March and her release on bail on 6 July 2011 makes one thing clear: even on the dubious assumption that proper application of her policy should in due course have led the Home Secretary to direct O’s release, it is unrealistic to consider that the conditions necessary for her release would have been in place prior to 6 July 2011.

For the above reasons, in agreement with the Court of Appeal, I regard it as already clear that, although the Home Secretary unlawfully failed to apply her policy under para 55.10 of the manual to O’s continued detention between March and July 2011, a lawful application of her policy would not have secured O’s release from detention any earlier than the date of her actual release on bail.

I have referred at para 28 above to the dispute as to the nature of the court’s review of the legality of the Home Secretary’s application of policy. It is now settled at the level of the Court of Appeal - at first sight unsurprisingly - that the nature of the review is the traditional public law inquiry into whether the application of it was rational: R (ZS) (Afghanistan) v Secretary of State for the Home Department [2015] EWCA Civ 1137. In para 30 above I have explained the open texture of the concept of “satisfactory management”, which reflects the wide range of factors relevant to it and explains the broad nature of the Home Secretary’s decision-making process. If indeed the inquiry is into the decision’s rationality, a process of that breadth may very well yield more than one rational, and thus more than one lawful, decision. But, supported by the interveners, O vigorously commends a more muscular approach. She insists that the subject is liberty; that indeed it is liberty denied by executive diktat; and that nothing less than an intense judicial inquiry into whether the application of policy was correct can be warranted in circumstances so controversial and of such fundamental importance.

I do not descend more fully into the rival contentions noted above because I consider that this appeal does not afford to the court the opportunity to choose between them. For the Home Secretary failed to address the satisfactory management or otherwise of O’s illness at Yarl’s Wood in the light of Dr Agnew-Davies’ report and so there is no decision for a court to be able on either basis to appraise. Instead the overall refusal to release O betrays a different type of public law error: it was procedurally flawed. What however is clear is that, even in the absence of any flaw, no decision to release O would in any event have been made prior to 6 July 2011.

In the Lumba case, cited at para 17 above, two foreign nationals, Mr Lumba and Mr Mighty, were sentenced to terms of imprisonment, apparently without being recommended for deportation by the sentencing judge. Once their sentences came to an end, and following notice of her decision to make deportation orders against them, the Home Secretary detained them; and, following the making of those orders, her detention of them continued. So initially, unlike O, they were detained pursuant to para 2(2) of Schedule 3 to the 1971 Act, set out in para 14 above; and subsequently, like O, they were detained pursuant to the words in parenthesis in para 2(3) of the schedule, also there set out. In proceedings for judicial review they challenged the lawfulness of their detention and claimed damages for false imprisonment.

At the time of the detention of the two men the Home Secretary’s published policy was that, even in relation to foreign national prisoners such as them, there was a presumption that they should be at liberty pending their intended deportation. In fact, however, the Home Secretary detained them pursuant to an unpublished policy which, inconsistently with her published policy, amounted almost to a blanket resolution on her part to detain foreign national prisoners pending intended deportation. It had been patently unlawful for the Home Secretary to apply to them an unpublished policy which was inconsistent with the published one. It was also clear, however, that, had the Home Secretary applied her published policy to them, her decision would, similarly, have been to detain them. This had led the Court of Appeal to hold that her unlawful application of policy had not made their detention unlawful. By a majority, this court disagreed. Giving the leading judgment, Lord Dyson said:

I provisionally conclude that, were O’s claim for judicial review permitted to proceed, it should therefore lead to the same result.

The Home Secretary, however, contends that the Lumba case is distinguishable from the present case; and that, where the detainee is initially detained, as here, under para 2(1) of Schedule 3 to the 1971 Act, rather than under para 2(2) of the schedule, an unlawful application of policy does not make the detention itself unlawful; and that therefore it does not generate a right even to nominal damages for false imprisonment. In this respect the Home Secretary relies on the Francis case, cited at para 4 above.

Although the initial detention of the two men in the Lumba case had been effected pursuant to paragraph 2(2) of Schedule 3 to the 1971 Act, the Court of Appeal in that case, [2010] EWCA Civ 111, [2010] 1 WLR 2168, in the course of explaining its decision (later reversed), had in passing addressed the effect of para 2(1) of the schedule. It had clearly had in mind the difference between the words “shall” in para 2(1) and “may” in para 2(2), both of which I have set in italics in my quotation of the sub-paragraphs in para 14 above; and at paras 88 to 89 it had proceeded to observe that, unlike detention under para 2(2), a person’s detention under para 2(1) was authorised by that sub-paragraph itself and that, even were the Home Secretary to have made an unlawful decision not to direct that person’s release, the lawfulness of the detention would therefore remain unaffected. In para 55 of his judgment in this court in the Lumba case Lord Dyson had specifically put those observations to one side.

In the Francis case the Court of Appeal, by a majority (Moore-Bick and Christopher Clarke LJJ), reached its decision by reference to the observations which that court had made in passing in the Lumba case.

Mr Francis, who for the purposes of the proceedings was assumed to have Jamaican rather than British nationality, had been sentenced to a term of imprisonment and recommended for deportation. On 4 December 2007 his sentence came to an end and he was detained pending the making of a deportation order. Following the making of that order on 21 May 2008, he continued to be detained until 29 September 2011. So, like that of O, his initial detention was effected pursuant to para 2(1) of Schedule 3 (being a “shall” provision) and his subsequent detention was effected pursuant to the words in parenthesis in para 2(3) (also being a “shall” provision).

The Court of Appeal divided the detention of Mr Francis into three periods:

The first period was from 4 December 2007 to 9 September 2008. In respect of this period the court adopted the conclusion of the trial judge that the Home Secretary had, as in the Lumba case, unlawfully applied to Mr Francis an unpublished policy in favour of detention which was inconsistent with her published policy.

The second period was from 9 September 2008 to 1 June 2010. In respect of this period the court adopted the conclusion of the trial judge that the Home Secretary had unlawfully failed to apply her policy by failing to cause the reviews of the continued detention of Mr Francis to be conducted by persons with authority to direct his release.

The third period was from 1 June 2010 to 29 September 2011. The court adopted the finding of the trial judge that during this period there was no longer any prospect that the deportation of Mr Francis would take place within a reasonable time.

Although the focus required by the present case is upon the court’s treatment in the Francis case of the first and second periods, its treatment of the third period remains important. This requires reference to principles which are no longer in play in the present case, namely the Hardial Singh principles, named after the decision of Woolf J in R v Governor of Durham Prison, Ex p Hardial Singh [1984] 1 WLR 704. In due course Lord Dyson distilled the decision of Woolf J into four principles of public law and he repeated them in para 22 of his judgment in the Lumba case. The second Hardial Singh principle is that the Home Secretary should detain a person pending intended deportation only for a reasonable period and the third is that if, before the expiry of the reasonable period, it becomes apparent that the Home Secretary will not be able to effect deportation within a reasonable period, she should direct release. In the Francis case the Court of Appeal concluded, in the light of the judge’s finding, that in respect of the third period the Home Secretary had been in breach of the third Hardial Singh principle. What, however, is of importance is that it then proceeded to hold that the detention of Mr Francis during the third period had been rendered unlawful by the Home Secretary’s breach of the third principle and that accordingly he was entitled to damages for false imprisonment during that period.

By contrast, however, the Court of Appeal proceeded to hold that neither of the different deficiencies in the Home Secretary’s application of policy during the first and second periods rendered the detention of Mr Francis during those periods unlawful. The court felt obliged to give a different value to the word “shall” in para 2(1) of Schedule 3 from that to be given to the word “may” in para 2(2) of it. Moore-Bick LJ said:

Notwithstanding the proper predisposition of any court to discern a difference of effect in any difference of language between statutory provisions, I have, with great respect to Moore-Bick and Christopher Clarke LJJ, come to the conclusion that their decision in relation to the first and second periods was wrong. It was wrong for the following reasons, taken cumulatively:

Any claim by the Home Secretary to be entitled to detain a person pending deportation must be clearly justified by the statutory language: Khawaja v Secretary of State for the Home Department [1984] AC 74, 122 (Lord Bridge of Harwich).

The Home Secretary’s duty to review the continuation of detention applies as much to those detained under para 2(1) as to those detained under para 2(2). Why would Parliament intend that the same unlawful deficiencies in her conduct of those reviews should have such different legal consequences?

Why should the effect of a recommendation for deportation, with the result that detention falls into para 2(1) rather than para 2(2), be that it remains lawful notwithstanding the Home Secretary’s unlawful application of policy? A Crown Court judge’s recommendation, perhaps made several years previously, has no other legal consequence, let alone one of such significance, and it is not even a recommendation for detention pending deportation.

Both men in the Lumba case were detained initially under para 2(2) (being a “may” provision) and subsequently under the words in parenthesis in para 2(3) (being a “shall” provision). This court decided that they had been unlawfully detained throughout both periods as a result of the Home Secretary’s unlawful application of policy; and it clearly considered that their later detention under the “shall” provision was no impediment to its decision. It was no doubt respect for this court’s decision which led Moore-Bick LJ in the Francis case to suggest at para 17 that

Section 36(2) of the 2007 Act, set out in para 16 above, refers to the Home Secretary’s exercise of the “power” of detention under para 2(3). But, according to the decision in the Francis case, Parliament’s reference to a “power” under para 2(3) was incorrect in circumstances in which detention under para 2(3) has been preceded by detention under para 2(1).

The court’s treatment in the Francis case of the Home Secretary’s breach of the third Hardial Singh principle seems to me to have been at odds with its treatment of her unlawful application of policy. The former was held to have rendered detention during the third period unlawful. The latter was held not to have rendered detention during the first and second periods unlawful. But why the difference? Moore-Bick LJ suggested at para 47 that “the Hardial Singh principles can be understood as implied limitations on the scope of an otherwise unqualified direction”. But why should the requirement in public law for the Home Secretary properly to apply her policy have any lesser effect than the requirement in public law for her to comply with the Hardial Singh principles?

In my view, therefore, the preferable analysis is along the lines sketched by Sir Stephen Sedley in his concurring judgment in the Francis case at paras 56 and 57, namely that the mandate to detain conferred by para 2(1) and by the words in parenthesis in para 2(3) is subject to two conditions. At the risk of oversimplifying the Hardial Singh principles, I would summarise the first condition as being that there is a prospect of deportation within a reasonable time. I would summarise the second as being that the Home Secretary will consider in accordance with her policy whether to exercise the power expressly given to her to direct release. Were either condition not to be satisfied, the mandate would cease and the detention would become unlawful.

The second condition was not satisfied in respect of the first and second periods of Mr Francis’ detention, with the result that, as in respect of the third period, the mandate to detain him ceased and therefore his detention during those periods should also have been held to have been unlawful.

Accordingly there is no difference in effect between, on the one hand, the conditional mandate to detain conferred on the Home Secretary by para 2(1) and by the words in parenthesis in para 2(3) and, on the other, the power to detain conferred on her by para 2(2) and by the words not in parenthesis in para 2(3).

The conclusion postulated in para 40 above need no longer be provisional: were O’s claim for judicial review permitted to proceed, the result in all likelihood would be a declaration that her detention from 4 March 2011 to 6 July 2011 was unlawful and an award to her of damages in the sum of £1. The Court of Appeal decided that, since such was - “at most”, so it added - the likely result of the claim, it was appropriate to uphold the refusal of Lang J to grant permission for it to proceed. I agree. By the time of its issue O had been released and it could bring her no practical benefit. To the extent that her contentions in these proceedings have deserved to be vindicated, she has secured their vindication in this judgment. I would dismiss the appeal.