England and Wales Court of Appeal (Civil Division) Decisions

IN THE HIGH COURT OF JUSTICE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM QUEENS BENCH DIVISION ADMINISTRATIVE COURT
MR JUSTICE CRANSTON
CO202309

		Royal Courts of Justice Strand, London, WC2A 2LL
		19/10/2010

Lady Justice Black :

1. This appeal, from a decision of Cranston J on 13 October 2009, concerns the Secretary of State for the Home Department's powers of detention under paragraph 2(3) of Schedule 3 to the Immigration Act 1971 and the implications of his failure to have regard, when exercising those powers to detain the Appellant, to his own policy as set out in a document entitled "Enforcement Instructions and Guidance".
2. The Appellant is a national of Bangladesh who was born in 1970. He arrived in the United Kingdom with his family when he was 13 years old and was given indefinite leave to remain. He has long standing mental health problems, first referred to in a medical report dating from 1991. The diagnosis at that stage seems to have been paranoid psychosis. Reports from 2008 gave a diagnosis of paranoid schizophrenia, for which anti-psychotic medication was required. A further report, dated July 2009, confirmed the diagnosis of paranoid schizophrenia and also identified that the Appellant was suffering quite prominent depressive symptoms which were attributed to prolonged detention.
3. The Appellant has a long criminal record, commencing in 1991, when he was 21, with offences of assault occasioning actual bodily harm and going equipped for theft. Over the next fifteen years or so, according to the Respondent's counsel's tally, the Appellant's criminal record came to contain, in total, 27 offences committed in 20 different names. Many of the offences were for dishonesty and there were two further assaults. Other offences of which he was convicted included perverting the course of justice, obstructing the police, and failing (on multiple occasions) to surrender to bail or to custody. Short terms of imprisonment were imposed on a number of occasions. The last offence in the sequence of the Appellant's criminal offending was robbery, of which he was convicted in January 2007 together with two offences of theft. On 14 February 2007, he was sentenced to four years imprisonment for these offences, the sentencing judge remarking that he had been found guilty of a persistent and premeditated course of conduct in stealing money from Ladbrokes and noting that the robbery had caused great distress to the young woman whom he had attacked and hit in the stomach.
4. In December 2007, the Respondent decided to deport the Appellant. The Appellant did not appeal against that and the deportation order was signed and served on him in February 2008. When he finished his prison sentence in mid May 2008, he was transferred to immigration detention where he remains.
5. The Respondent's policy at all material times was not to detain mentally ill people unless there were "very exceptional circumstances" which justified that. Cranston J found that the Appellant's mental health issues were such that he came within this policy but that the Respondent had failed to engage with the policy in detaining the Appellant. There had been reviews during his time in detention but the judge found that they had focussed on other issues and hardly touched on the Appellant's mental health until the hearing before him was imminent (the hearing date was 19 August 2009). He granted a declaration that the Respondent "unlawfully failed to consider the implications of his policy for the detention of [the Appellant]". Otherwise, he dismissed the Appellant's claim, giving permission to appeal to this court.
6. The remedies sought by the Appellant at trial were, in fact, far reaching. As the judge records in paragraph 3 of the judgment, he sought an award of damages for breach of his Article 5 rights under the ECHR, a grant of temporary admission, damages for false imprisonment, and aggravated and exemplary damages. On none of these claims did he succeed.
7. The Appellant had also sought a declaration that his detention was, and continued to be, unlawful and a mandatory order for his release. Refusing this, the judge found that he had not demonstrated that his detention was in breach of the policy. His conclusion about this is set out in paragraphs 68 and 69 of the judgment, which I will quote in full because they are central to this appeal:

"68. In my judgment, however, given the way the "very exceptional circumstances" test must be interpreted, the claimant has not demonstrated that his detention is in breach of the policy. The policy means that there is a strong presumption in favour of his release because of his mental illness, and that the balance of other factors has to be substantial indeed for detention to be justified. There needs to be a qualitative judgment of these various factors. In the balance will be the factors which typically operate in detention cases, absconding and re-offending risk, because the object of immigration detention is removal, detention not being an end in itself. Of course there needs to be an elevated risk of absconding and re-offending to counter the strong presumption in favour of release.

69. Here the claimant has a history from 1991 of prolific offending, some forty offences involving twenty-six convictions. The offending culminated in the robbery conviction, serious enough to warrant a four year sentence. I have quoted what the trial judge said about the impact of the robbery on the young woman victim. Apart from the escalating seriousness of the claimant's pattern of offending and the risk posed as a result of that, there is also the very high risk of absconding. That is evident in the considerable number of offences relating to bail and failing to surrender to custody. Closely associated with that is the disruptive behaviour to frustrate removal, and the claimant's failure to co-operate in an attempt to document him on the expiry of his Bangladesh passport. There are also the unmeritorious applications he has made for asylum and judicial review and the frequent and unsuccessful applications he has made for bail. Given all these factors, the balance is against release."

8. The judge refers to the various applications made by the Appellant. The Appellant had, in fact, been making applications of one kind or another from the end of April/beginning of May 2008. We are told that his response to the deportation order was by way of fax on 7^th November 2007, which can be found in the appeal bundle. Towards the end of April 2008, the Respondent made an attempt to remove him from this country. According to the Immigration History and Chronology compiled by the Respondent, this was preceded by last minute representations by the Appellant which were refused but, in the event, he was not removed because he became disruptive and was taken off the flight.
9. Removal directions were reset for 8 May 2008 and the Appellant was notified of them on 1 May 2008. On 2 May 2008, his representatives submitted an out of time appeal in relation to the deportation order and, on 7 May 2008, lodged an application for asylum. The removal directions were cancelled.
10. In July 2008, further representations were made raising Article 8 issues in connection with the Appellant's medical condition.
11. In August 2008, the Appellant's appeal against the deportation order was dismissed by the Asylum and Immigration Tribunal and his application for reconsideration of that dismissal was refused. In September 2008, his application for asylum was refused; that decision was subsidiary to the deportation decision and did not carry a right of appeal.
12. In December 2008, directions for removal were set again but cancelled when the Appellant submitted a fresh claim for asylum and lodged the first of the judicial review claims which were ultimately determined by Cranston J.
13. Early in February 2009, the Appellant's further representations were refused. He made more than one unsuccessful claim that month to the Asylum and Immigration Tribunal for bail. He also issued his main judicial review application concerning the legality of his detention. He applied again for bail in March 2009, withdrawing his application on the day of the hearing and then re-applying the following day. Further applications for bail in succeeding months were all refused.
14. In May 2009, permission was granted for the judicial review challenge in relation to the legality of the Appellant's detention. The Appellant's ancillary judicial review challenges were not permitted to proceed to trial. Cranston J himself refused permission for a proposed challenge to the validity of the deportation documentation, commenting that the various arguments advanced were "deeply unimpressive", "simply untenable", and had nothing in them.
15. In June 2009, the Respondent, without prejudice to his position, agreed to proceed on the basis that the Appellant had made an application to revoke the deportation order and to agree to him having a right of appeal from the decision not to do so. We have been told that the appeal was unsuccessful but that reconsideration of the decision was ordered and, in consequence of the changes to the procedure in relation to such matters, that has been treated as permission to appeal to the upper tribunal. That appeal is now pending.

The focus of the appeal

16. The focus of this appeal has been Cranston J's refusal to grant a declaration that the Appellant's detention had been and remained unlawful and to grant a mandatory order for his release. Very little has been said on the subject of damages, whether for false imprisonment or otherwise, no doubt because the reality is that those matters are dependent on the principal issue of the legality of the detention and also because the Appellant's main concern must inevitably be to obtain his release.
17. The Appellant argues that, having rightly held that the Respondent had failed to take into account his own policy in relation to those who are mentally ill when deciding whether initially to detain the Appellant and then to keep him detained, Cranston J was obliged, without more ado, to grant a declaration that the Appellant's detention was unlawful. At its most radical, the submission advanced by Mr Macdonald QC and Mr Sayeed for the Appellant is that a failure by the Secretary of State to apply his published policy in making a decision to detain necessarily renders that decision a nullity and the resulting detention unlawful. Accordingly, they argue that the judge fell into error by requiring the Appellant to surmount a further hurdle before such a declaration would be appropriate, wrongly proceeding on the basis that a failure to apply the policy would not render the detention unlawful unless that failure "caused the detention"/"in fact led to the detention" (as the judge put it at paragraph 42 of the judgment). This supplementary hurdle has been referred to in shorthand by counsel as a "causation analysis" and I will adopt that terminology, even though it has its shortcomings.
18. In the event that existing authority in this court is found to preclude this radical submission, and some sort of causation analysis is necessary, Mr Macdonald QC and Mr Sayeed argue that the question that the judge should have asked himself was whether the detention would inevitably have resulted even if the Respondent had applied his policy. They submit that Cranston J did not ask himself this question, and if he had done, he could not properly have concluded that detention would inevitably have resulted. They argue that the judge actually put himself in the role of the primary decision maker, balancing the various factors and determining for himself whether the Appellant should have been detained, which they submit was not his role. Their case is that the judge's error in this respect was compounded by the judge wrongly putting the burden on the Appellant to establish that the detention was unlawful. They further submit that even if it is not correct to say that the judge approached matters in this way, the decision that he took amounted, at best, to a decision that the Respondent could have decided to detain the Appellant and not to a decision that his detention was inevitable.
19. For the Respondent, Mr Waite does not accept that a failure by the Secretary of State to apply one of his policies when deciding whether to detain someone necessarily renders that person's detention unlawful. He submits that the breach of policy in this case was relevant to the issue of whether the Appellant's detention was unlawful but not determinative of it. The lawfulness of the detention is, he submits, dependent on the well known principles set out in Re Hardial Singh [1984] 1 WLR 704. He does not accept that the court's role is confined to a consideration of whether detention would have been inevitable even if the Secretary of State had applied his policy, although he submits that this Appellant would, in fact, inevitably have been detained. He submits that it is for the judge to reach his own view as to whether the detention was lawful under the Hardial Singh principles, taking into account all the relevant factors, including the Secretary of State's policy, not only with regard to people with mental health problems but in all other relevant respects as well, and that Cranston J was entitled to conclude, as he did, that the Appellant's detention was lawful.

The relevant statutory provision and policy considerations

20. The Appellant was detained pursuant to the power contained in paragraph 2(3) of Schedule 3 of the Immigration Act 1971. That paragraph reads:

"Where a deportation order is in force against any person, he may be detained under the authority of the Secretary of State pending his removal or departure from the United Kingdom (and if already detained by virtue of sub-paragraph (1) or (2) above when the order is made, shall continue to be detained unless he is released on bail or the Secretary of State directs otherwise)."

21. Immigration detention has been the subject of consideration in a manual for officers dealing with immigration enforcement in the United Kingdom entitled Enforcement Instructions and Guidance. This manual contains the Government's policy in relation to immigration detention. We have been provided with the Guidance in a form which we are told incorporates amendments made after the Appellant's detention began in May 2008, but it is not suggested that such changes as there were make any material difference to this case. The part of the Guidance which is relevant to detention is Chapter 55.
22. Chapter 55.10 sets out the policy not to detain mentally ill people except in very exceptional circumstances. It is headed "Persons considered unsuitable for detention" and, in so far as is material, reads:

"Certain persons are normally considered suitable for detention in only very exceptional circumstances, whether in dedicated Immigration accommodation or elsewhere. Others are unsuitable for Immigration detention accommodation because their detention requires particular security, care and control. In CCD cases, the risk of further offending or harm to the public must be carefully weighed against the reason why the individual may be unsuitable for detention.

The following are normally considered suitable for detention in only very exceptional circumstances, whether in dedicated Immigration detention accommodation or elsewhere:

……

those suffering from serious medical conditions or the mentally ill - in CCD cases, please contact the specialist Mentally Disordered Offender Team; …."

"CCD" is an abbreviation for Criminal Casework Directorate Cases which include cases, such as this one, concerning foreign national prisoners.

23. Elsewhere in Chapter 55, other aspects of the Government's policy on immigration detention are set out.
24. Chapter 55.1.1 sets out the general presumption in favour of temporary admission or release rather than detention.
25. Chapter 55.1.2 says that cases concerning foreign national prisoners are subject to the general policy in 55.1.1 and that the starting point in such cases "remains that the person should be released on temporary admission or release unless the circumstances of the case require the use of detention". However, 55.1.2 goes on to say that the nature of these cases means that special attention must be paid to their individual circumstances and provides that in any case in which the criteria for considering deportation action are met (as they are here)

"the risk of re-offending and the particular risk of absconding should be weighed against the presumption in favour of temporary admission or temporary release. Due to the clear imperative to protect the public from harm from a person whose criminal record is sufficiently serious as to satisfy the deportation criteria, and/or because of the likely consequence of such a criminal record for the assessment of the risk that such a person will abscond, in many cases this is likely to result in the conclusion that the person should be detained, provided detention is, and continues to be, lawful. However, any such conclusion can be reached only if the presumption of temporary admission or release is displaced after an assessment of the need to detain in the light of the risk of re-offending and/or the risk of absconding."

26. The Guidance returns elsewhere in Chapter 55 to the issue of CCD cases, for example in 55.1.3 it is said that:

"[s]ubstantial weight must be given to the risk of further offending or harm to the public indicated by the subject's criminality. Both the likelihood of the person re-offending and the seriousness of the harm if the person does re-offend must be considered. Where the offence which has triggered deportation is included in the list at 55.3.2.1, the weight which should be given to the risk of further offending or harm to the public is particularly substantial when balanced against other factors in favour of release. In cases involving these serious offences, therefore, a decision to release is likely to be the proper conclusion only when the factors in favour of release are particularly compelling. In practice, release is likely to be appropriate only in exceptional cases because of the seriousness of violent, sexual, drug-related and similar offences. Where a serious offender has dependent children in the UK, careful consideration must be given not only to the needs such children may have for contact with the deportee but also to the risk that release might represent to the family and the public."

No list to speak of can be found at 55.3.2.1 but there is a list elsewhere in the Guidance entitled "Crimes where release from immigration detention or at the end of custody would be unlikely" and there can be little doubt that it is to this list that 55.1.3 refers. The list includes robbery.

27. Chapter 55.3A concerns the decision to detain in CCD cases. It includes the following passage related to "more serious offences" which appears to concern those offences on the list to which I have just referred:

"More serious offences

A conviction for one of the more serious offences is strongly indicative of the greatest risk of harm to the public and a high risk of absconding. As a result, the high risk of public harm carries particularly substantial weight when assessing if continuing detention is reasonably necessary and proportionate. So, in practice, it is likely that a conclusion that such a person should be released would only be reached where there are exceptional circumstances which clearly outweigh the risk of public harm and which mean detention is not appropriate. Caseworkers must balance against the increased risk, including the particular risk to the public from re-offending and the risk of absconding in the individual case, the types of factors normally considered in non-FNP detention cases, for example, if the detainee is mentally ill or if there is a possibly disproportionate impact on any dependent child under the age of 18 from continued detention. Caseworkers are reminded that what constitutes a "reasonable period" for these purposes may last longer than in non-criminal cases, or in less serious criminal cases, particularly given the need to protect the public from serious criminals due for deportation."

28. Similar themes are re-worked elsewhere in Chapter 55, with 55.3.1 setting out a list of the factors influencing a decision to detain, and 55.3.2 providing further guidance on deciding whether to detain someone in a CCD case, including the following passage:

"In cases involving these serious offences, therefore, a decision to release is likely to be the proper conclusion only when the factors in favour of release are particularly compelling because of the significant risk of harm to the public posed by those convicted of violent, sexual, drug-related and other serious offences. In practice, release is likely to be appropriate only in exceptional cases."

The judge's approach

29. Before I consider the authorities which bear on the issues that we have to decide, it may be helpful to examine in a little more detail the approach that the judge took.
30. He reviewed a considerable number of authorities under two main headings. In the first category of authorities, he considered Hardial Singh itself and the restatement of the Hardial Singh principles in R(I) v Secretary of State for the Home Department [2001]EWCA Civ 888 as well as the factors identified by Dyson LJ (as he then was) as relevant to the issue of how long it was reasonable for the Secretary of State to detain someone pending deportation. Dyson LJ had said in R(I) that the risk of someone absconding should not be overstated and was not a trump card whereas, in R(A) v Secretary of State for the Home Department [2007] EWCA Civ 804, Toulson LJ and Keene LJ attached considerable significance to the risk of absconding.
31. The second category of authorities was concerned with non-compliance with policy. Cranston J identified the duty of a public authority to adhere to its published policy unless there was a compelling reason to depart from it and he cited a number of decisions in which a failure to apply published policy had resulted in detention being unlawful. Then, at paragraph 37, he stated that non-compliance with policy does not ipso facto lead to detention being unlawful and that it must be shown that but for the breach the person would not have been detained. In this connection, he cited R(D); R(K) v Secretary of State for the Home Department [2006] EWHC 980 (Admin) affirmed by the Court of Appeal under the name of HK (Turkey) v SSHD [2007] EWCA Civ 1357 (failure to carry out a medical examination within 24 hours of detention) and Abdi v Secretary of State for the Home Department [2008] EWHC 3166 (Admin), the decision which was later to be considered by the Court of Appeal as R(WL) and others v SSHD [2010] EWCA Civ 111 to which I shall need to come in some detail in due course. He also considered R (SK)(Zimbabwe) v SSHD [2008] EWCA Civ 1204 to which, again, I shall need to come in some detail.
32. Having reviewed the authorities, at paragraph 42 of his judgment, Cranston J distilled the principles which he drew from them as follows:

"In my judgment the following principles apply when judicial review is sought of a decision to detain, on the basis of the non-application or a breach of the Secretary of State's policy:

i) at the outset there must be a non-application or a breach of the policy. To determine whether there has been a breach of policy, the policy is to be construed in the ordinary way;

ii) any non-application or breach of the policy must have caused the detention. Of itself the non-application or breach of policy cannot lead to a conclusion that detention is unlawful without an additional inquiry into whether this in fact led to the detention. That turns partly on the nature of the policy in issue: for example, there is a difference between a policy requiring the medical examination of detainees and the one at issue in this case which limits the detention of those with mental health issues to very exceptional circumstances.

iii) the non-application or a breach of policy causing the detention may give rise to ordinary public law remedies such as a declaration. Ordinarily damages are not available in judicial review, but may be awarded if the court is satisfied that they would be awarded on private law principles (in this case, the tort of false imprisonment) or as a result of the Human Rights Act 1998 (in this case, just satisfaction for breach of Article 5)."

33. Cranston J considered the provisions of the policy set out in Chapter 55.10 of the Enforcement Instructions and Guidance. He took the view that the provision that the mentally ill should be detained in only very exceptional circumstances does not stand in isolation. As he pointed out, 55.10 commences with the requirement that in CCD cases the risk of further offending or harm to the public must be carefully weighed against the reason why the individual may be unsuitable for detention and, elsewhere in the Guidance, there is a requirement that particularly substantial weight should be given to the risk of further offending or harm to the public where someone has been involved in a serious offence. He also pointed out the express reference to mental illness in the passage in 55.3A dealing with more serious offences, requiring case workers to balance against the risk to the public from re-offending and absconding other factors such as the mental illness of the detainee. The judge concluded, in paragraph 55 of his judgment, that the import of the Guidance was:

"that although a person's mental illness means a strong presumption in favour of release will operate, there are other factors which go into the balance in a decision to detain under the policy. The phrase needs to be construed in the context of the policy providing guidance for the detention of all those liable to removal, not just foreign national prisoners. It seems to me that there is a general spectrum which near one end has those with mental illness who should be detained only in "very exceptional circumstances" along it – the average asylum seeker with a presumption of release – and near the other end has high risk terrorists who are detained on national security grounds. To be factored in, in individual cases, are matters such as the risk of further offending or public harm and the risk of absconding. When the person has been convicted of a serious offence substantial weight must be given to these factors. In effect paragraph 55.10 demands that, with mental illness, the balance of those factors has to be substantial indeed for detention to be justified."

34. Applying the law and considerations of policy to the facts before him, Cranston J concluded that the Secretary of State had failed to engage with his policy about the detention of mentally ill people. The Appellant had argued that even the review of 14 August 2009, which made mention of his mental illness and of exceptional circumstances, did not in fact represent a proper consideration of the policy. The judge did not deal with this expressly but merely said, in paragraph 65:

"The detention reviews clearly focus on other issues and hardly touch on his mental health up until the point when the current hearing was imminent."

35. As we can see from paragraph 56 of the judgment, the Appellant had argued that the failure to engage with the policy meant that he was entitled to damages for false imprisonment. The judge did not accept that that followed. He embarked upon his conclusions about the case by stating, in paragraph 64, what, in his view, was the issue he had to determine which he formulated as "whether, because of the claimant's mental health issues, his detention has been lawful" and "whether the power of detention has been properly exercised in the particular circumstances of this case". This required, in his view, a "fact specific assessment as to whether the claimant's past and continuing detention has been consistent with chapter 55 of the Enforcement Instructions and Guidance and Hardial Singh principles". At paragraph 66, he said:

"However, the failure of the Secretary of State to engage with the claimant's mental health, and its relevance to detention under the policy, is not the end of the matter. The authorities make clear that a court must be satisfied that a failure to apply or breach of the policy has in fact caused the claimant prejudice. In other words the claimant needs to satisfy me that even if the Secretary of State had engaged with the policy, his continued detention would not have been justified under it."

36. Dealing with the issue as he had identified it, the judge accepted that the Appellant's mental health problems were such that the policy applied but he did not accept that the Appellant had demonstrated that his detention was in breach of the policy. This conclusion can be found in paragraph 68 of the judgment which I have set out earlier together with paragraph 69. Those two paragraphs contain an identification of the factors that the policy requires to be put into the balance in deciding whether to detain a mentally ill person and of the facts relating to this Appellant which are pertinent in that regard, in particular what the judge classed as "the escalating seriousness of the Appellant's pattern of offending" and the risk posed as a result, the very high risk of absconding, the Appellant's disruptive behaviour to frustrate removal and his failure to co-operate with regard to travel documentation, his unmeritorious applications for asylum and judicial review and his frequent unsuccessful bail applications. This led the judge to the view that "[g]iven all these factors, the balance is against release".
37. The Appellant had advanced an alternative submission to the judge to the effect that even if he did not succeed in his argument that the failure to apply the Guidance rendered the detention unlawful, it was unreasonable in Hardial Singh terms because his mental illness and other vulnerabilities meant that the period of detention which would be classed as reasonable would be significantly shorter than otherwise and should not have exceeded a matter of months at most. The judge concluded that the detention was, and continued to be, justified not only under the policy but also, for similar reasons, under the Hardial Singh principles. He referred in this regard to the Appellant's own disruption of legitimate attempts to remove him, by refusing to co-operate with attempts to document him and making belated challenges aimed at frustrating his deportation, whilst taking into account the possibility that some of the Appellant's behaviour might have been attributable to his fluctuating mental health and also taking into account the medical evidence suggesting that detention and the prospect of removal were having an adverse impact on the Appellant in terms of depression.

Analysis

38. It is important to recognise that when Cranston J considered this matter, the case of Abdi had not yet been considered by the Court of Appeal. Judgment in that appeal (R(WL) and others v SSHD, see above) was given on 19 February 2010. It is of central importance to the present appeal and I will have to deal with it in some detail just as I must also deal with another important recent authority, R (SK (Zimbabwe) v SSHD [2008] EWCA Div 1204 to which I now turn.

R (SK (Zimbabwe) v SSHD [2008] EWCA Div 1204

39. SK was a foreign national who was sentenced to imprisonment for common and sexual assaults. He was detained pursuant to paragraph 2(2) of Schedule 3 to the Immigration Act 1971 pending the making of a deportation order in relation to him. He alleged that his detention was unlawful because regular reviews of it had not been carried out as required by Rule 9 of the Detention Centre Rules 2001 and the Secretary of State's policy as set out in the Home Office's Operations Enforcement Manual and he was not provided with written reasons for his detention on a monthly basis as the Rules required. He sought an order for his release, a declaration and damages. At first instance, he was granted a declaration that he had been unlawfully detained for specified periods for which he was entitled to damages but his release was not ordered because his detention had since become lawful. The Secretary of State appealed successfully to the Court of Appeal.
40. Laws LJ (with whom Longmore LJ agreed without adding anything and Keene LJ agreed in a short judgment), treated the issue as a question of the true construction of paragraph 2(2) of Schedule 3. It is well established that the power contained in paragraph 2(2) is subject to the limitations set out in Hardial Singh but the judge had found that none of those limitations had been breached and that was not challenged so the focus of the appeal was, instead, on the judge's conclusion that it was a necessary prerequisite to the continuing legality of the detention that the reviews dictated by the Detention Centre Rules and the manual had been carried out. Laws LJ pointed out that if that approach was right, it would represent another limitation on the power to detain, namely that it is subject to compliance with the Rules and the manual. However, he did not consider that that approach was right. In his judgment, compliance with the letter of the Rules or manual could only be a sine qua non of a lawful exercise of the power to detain if paragraph 2(2) (or "other main legislation") made it so and that was not the case. Paragraph 2(2) does not contain anything analogous to section 34(1) of the Police and Criminal Evidence Act 1984 (PACE) which had featured in the case of Roberts v Chief Constable of the Cheshire Constabulary [1999] 1 WLR 662 which had influenced Munby J (as he then was) at first instance. The review that had not been carried out in the Roberts case was a review of the detention of a man being held in police custody under section 37 of PACE and it was required by section 40 of that Act. Section 34 of PACE expressly provided that a person should not be kept in police custody except in accordance with the provisions of "this part of this Act" and section 40 was one of those provisions. As Laws LJ explained it, the Roberts case depended on the proper construction of PACE which made section 40 reviews a condition precedent to the legality of the suspect's detention. Accordingly, whilst breach of the Rules or the manual might attract other remedies in public law, for example a declaration that the Secretary of State had failed to comply with both, that would not turn the detention into false imprisonment.
41. Keene LJ had been troubled initially by the breaches of the Detention Centre Rules, a piece of subordinate legislation laid before Parliament and therefore of greater significance than the manual, but with some hesitation was persuaded that compliance with the Rules was not a condition precedent to the exercise of the powers under Schedule 3 of the Immigration Act and that a breach of Rule 9 did not of itself render detention unlawful.
42. Laws LJ also put forward an alternative analysis of the situation, giving consideration to whether the imperative that the courts would not allow a discretionary power to detain to be exercised arbitrarily required the Secretary of State to have in place some effective mechanism, not necessarily the same as those in the Rules or the manual, by which detention is regularly monitored to ensure that the purpose of the confinement remains capable of fulfilment within a reasonable time and there are no countervailing factors which should mandate the detainee's release. His answer to this starts with the observation that if no steps were taken to see that the Hardial Singh principles were being complied with, any detention would be vulnerable to challenge on the footing that the detainee was as liable to remain incarcerated in breach of the principles as in fulfilment so his detention would necessarily be arbitrary and the Secretary of State would be in no position to establish by evidence that the principles were systematically met. But Laws LJ considered that the means by which that state of affairs was to be avoided was not prescribed by either article 5 of the Human Rights Convention or the common law. Deprivation of liberty could only occur in accordance with a procedure prescribed by law which he interpreted to be a requirement that any interference with liberty must not be random and arbitrary but governed by clear pre-existing rules. He went on to say:

"Here the "rules" are the Hardial Singh principles. Their fulfilment in any given case saves a detention from the vice of arbitrariness. A system of regular monitoring is, no doubt, a highly desirable means of seeing that the principles are indeed fulfilled. But it is not itself one of those principles".

43. He pointed out:

"34. It is important to notice that if this approach is wrong, it means that a detention will be unlawful in the absence of (or failure to fulfil) a system of internal monitoring even though it can be shown on the particular facts that the detention, far from being arbitrary, is wholly justified. Such a position, however, is at odds with authority in this jurisdiction which tends to show that a failure of a published procedure which a detainee is entitled to have applied to him will not of itself invalidate his detention".

He then quoted R(D); R(K) v Secretary of State for the Home Department (see above, no medical examination within 24 hours of detention does not of itself mean the detention was wrongful) and R (Saadi) v SSHD [2002] 1 WLR 3131 (forms given to detainees setting out the reasons for their detention gave wrong or inappropriate reasons but this did not affect the legality of the detention).

44. In May 2009, leave was given to appeal to the House of Lords against the decision of the Court of Appeal in SK, and we are told that the appeal is to be finally determined later this year in the light of the appeal which is now also pending in WL. At least for the moment, however, the law in this court stands as set out in the Court of Appeal's decision. The thrust of that is that it is the Hardial Singh principles which determine the legality of detention; it is when those principles are not satisfied that a detention is unlawful. The Secretary of State has an obligation to comply with the Rules and the manual (now replaced by the Guidance) but failure to comply does not inevitably make a detention unlawful although compliance with the Rules and the manual would be an effective and practical means of demonstrating that the principles have been and are being fulfilled.
45. The appellant in SK was detained under paragraph 2(2) of Schedule 3 and the Appellant here was detained under paragraph 2(3). However, it has not been argued that this makes any material difference to the applicability of the principles enunciated in SK to this case.

R(WL) and others v SSHD [2010] EWCA Civ 111 (WL)

46. WL concerned the detention of two foreign national prisoners, following the completion of their prison sentences, with a view to deportation. The Home Secretary's published policy provided that such prisoners would continue to be detained only if that was justified, in other words the presumption was that they would be released. There was also in existence an unpublished "policy" (or practice) which the trial judge, Davis J, found set up a presumption of detention unless release was justified. The Court of Appeal, having had access to additional material, found that there was, in fact, a blanket policy excluding virtually all foreign national prisoners from being considered for release and that such a policy was unlawful, even though a policy involving a presumption of detention would not necessarily have been unlawful. They concluded, furthermore that, in any event, it was unlawful to operate a secret policy which conflicted with the government's published policy. Importantly for this case, they considered that the detention of an individual was not rendered unlawful "unless the unlawful practice or policy was a material cause of the detention" (paragraph 48(4)). The summary of their conclusion on this point continues as follows:

"….It is necessary, therefore, in every case in which it is relevant to do so, to ascertain whether detention was authorised by reference to the blanket practice or policy or by consideration of a presumption or, indeed, without reference to any administrative practice or presumption."

They concluded that on the facts of two of the cases before them, "materiality" was not established.

47. The reasoning behind these conclusions is set out in some detail but I will only rehearse it where it bears on the decisions we have to take. At paragraph 58, Stanley Burnton LJ, giving the judgment of the court, sets out that although policy is not to be equated with law, it may give rise to obligations or restrictions in public law. The existence of these obligations or restrictions may be explained in different ways depending on the context, for instance a failure by the Secretary of State to apply his own published policy without good reason may amount to a breach of legitimate expectation. In other contexts the failure may generate a Wednesbury challenge or a challenge on the basis of inconsistency or unfairness amounting to an abuse of power. Stanley Burnton LJ says:

"Indeed, we may have arrived at the point where it is possible to extract from the cases a substantive legal rule that a public body must adhere to its published policy unless there is some good reason not to do so."

48. The court begins, at paragraph 80, to set out its reasoning in relation to the question "whether the fact that the decision to detain was made against the background of an unlawful policy was in itself sufficient to provide the foundation for the cause of action for false imprisonment, and an award of at least nominal damages". Davis J had held that this was not, in itself, sufficient, accepting the submission of the Secretary of State that (as Davis J summarised it in his judgment) "where the unlawful policy was of no causative effect (because detention would still have been lawfully directed irrespective of the new policy) then the detention is not to be styled as unlawful detention at all". Davis J accordingly directed himself that inquiry had to be made as to "whether the introduction of the unlawful and unpublished policy in fact caused each claimant unjustifiably and unlawfully to be detained".
49. As the Court of Appeal sets out at paragraph 82, Davis J regarded the causation analysis as dictated by established authority, although he also added his own reasons why he considered it to be an appropriate approach. He cited Saadi (see above, in which Lord Slynn in the House of Lords endorsed Collins J's view that giving no reasons or wrong reasons for detention did not in the end affect the legality of the detention) and Nadarajah [2003] EWCA Civ 1768 (applicant unlawfully detained in reliance on an unpublished policy not to have regard to the notification that judicial review proceedings were imminent, as opposed to such proceedings actually having been issued, when such a notification would normally have been expected to be sufficient to secure his release and the applicant's solicitors would have commenced proceedings had they known of the unpublished policy). He then expressed the view that what disposed of the matter was a pair of Court of Appeal decisions, namely D v Home Office [2005] EWCA Civ 38 and SK. In both of those cases, damages were being claimed for false imprisonment and both, in the judge's view, required causation to be established before a policy breach would give rise to damages for unlawful detention.
50. The Court of Appeal had some difficulty in following the passage in the judgment of Brooke LJ in D v Home Office upon which Davis J had relied. They also record that it was attacked by counsel for WL and KM as being inconsistent with Christie v Leachinsky [1947] AC 573 and may also have been thought inconsistent with Roberts (see above). However, at paragraph 87, they conclude that Christie v Leachinsky and Roberts were both cases in which there was no lawful authority for the detention of the detained person whereas there was no doubt, in WL, that the statutory powers relied on by the Secretary of State were apt for the purpose, and the case was not based on the breach of any specific regulation on which the legality of the detention was dependent but concerned the manner in which the power was exercised. The Court of Appeal distinguished between detention under paragraph 2(1) of Schedule 3 of the 1971 Act and detention under either paragraph 2(2) or 2(3). Sub-paragraph (1) is itself legislative authority for the detention of a foreign national prisoner who has been sentenced to imprisonment and recommended for deportation. That is not so with sub-paragraphs (2) and (3) as the Court of Appeal set out at paragraph 89, which I will quote in full because it also contains the nub of the decision concerning the causation issue:

"89.     The position is different when the decision to detain is made under sub-paragraph (2) or (3). In these cases, there is no lawful authority to detain unless a lawful decision is made by the Secretary of State. The mere existence of an internal, unpublished policy or practice at variance with, and more disadvantageous to the FNP than, the published policy will not render a decision to detain unlawful. It must be shown that the unpublished policy was applied to him. Even then, it must be shown that the application of the policy was material to the decision. If the decision to detain him was inevitable, the application of the policy is immaterial, and the decision is not liable to be set aside as unlawful. Once again, however, once a decision to detain has lawfully been made, a review of detention that is unlawful on Wednesbury principles will not necessarily lead to his continued detention being unlawful."

51. If further insight is required into the principle that is to be derived from this paragraph, it is perhaps available from paragraph 96. That paragraph falls in a section of the judgment dealing with what was, in WL, the hypothetical issue of the proper approach to the assessment of damages in a case where the actual detention was unlawful but where a lawful decision might have produced the same result. The concluding remarks in the discussion on that subject include this:

"96. If we had to decide the matter, we would not necessarily regard ourselves as precluded by Roberts from considering, for the purposes of assessing damages, whether an unlawful detention had in practice caused any real loss. If, on the evidence it was clear that, even assuming a lawful consideration, there was no realistic possibility of a different decision having been reached, and no realistic possibility of earlier release, then we do not see why that should not be reflected in an award of nominal damages only. However, on the view we have taken of the issue of causation, such a set of facts in the present context would mean that there was no liability at all, so that the issue of damages would not arise. It follows that the determination of the appeals depends on the facts of the individual cases…." [my emphasis]

52. To my mind, paragraph 89 determines, in this court, the issue of whether a causation analysis is or is not required in determining whether someone's detention is or is not unlawful. It is. This is entirely in line with the decision of the Court of Appeal in SK which is itself an example of a breach of rules and policy not being sufficient alone to render detention unlawful.
53. Mr Macdonald indicated in argument that he had no issue with the principle set out in paragraph 89 of WL but that he sought to distinguish both WL and SK from the present case. His argument is that the failure of the Secretary of State in WL and in SK did not relate to a "centrally relevant" consideration in the determination of whether to detain a person and that the reach of those cases is confined to cases of that type. He argues that, in the instant case, the failure of the Secretary of State is "qualitatively different in that it relates to a centrally relevant consideration about the appropriateness of detaining mentally ill persons".
54. The main difficulty that I have with this argument is that I am not convinced that it is possible to say that WL concerned a failure which was not centrally relevant to the decision to detain. The corollary of the application of the secret policy (if indeed it had been applied to the individuals concerned) would necessarily have been that the published policy, which was diametrically opposed to it, was not being applied. Accordingly, instead of the decision maker approaching the case on the basis of a presumption of release, he would approach it on the basis that detention was required. It is difficult to see what could be more fundamental to a decision about detention than that and it is certainly a state of affairs at least as fundamental as omitting to take into account a provision requiring that mentally ill people should be detained only in very exceptional circumstances. I acknowledge that it is arguable that the failure in SK was less fundamental but, even in SK, the consequence of the failure to review the detention was that a person who was entitled to have the issue of his detention revisited at regular intervals had no decision taken at all about it for a significant period of time. On one view of that state of affairs, it is at least as fundamental as carrying out a review but omitting to pay attention to an extremely important factor.
55. Furthermore, there is no suggestion in the language of paragraph 89, or elsewhere in the WL judgment, that the principle there enunciated was intended to be confined in the way in which Mr McDonald submits. Indeed, the causation analysis set out in paragraph 89 appears designed to identify (amongst other cases) the very cases which he would say were completely outside its application. This is achieved by a two stage process of first eliminating cases where the failure or breach has nothing at all to do with the case ("It must be shown that the unpublished policy was applied to him"), and then requiring that the application of the policy (or in our case, read "failure to apply the policy") was material to the decision. A case in which the policy which was not applied related to a centrally relevant consideration in the determination is surely a prime example of a case which falls squarely within the parameters of paragraph 89, unqualified as they are.
56. It would be wrong to consider paragraph 89 without also considering paragraph 90. This reads:

"90.     For completeness, we would add that the test of materiality may not be precisely the same as in the context of an application for a quashing order in judicial review. In that context, a court, faced with a judicial review claim made promptly following the original decision, would be likely to quash a decision, and require it to be retaken, even if the evidence showed only a risk that it might have been affected by the illegality. However, in the context of a common law claim in tort, which is concerned not with prospective risk, but actual consequences, we think it would be entitled, if necessary, to look at the question of causation more broadly, and ask whether the illegality was the effective cause of the detention (see e.g. Galoo Ltd v Bright Grahame Murray [1994] 1WLR1360, 1374; and the the discussion of "Causation in Law" in Clerk & Lindsell Torts 19^th Ed, paras 2-69-71)."

I confess to finding this paragraph somewhat difficult. However, Mr Macdonald did not advance any argument as to why its contents should alter my conclusion about the meaning and reach of paragraph 89 and I cannot find any myself.

57. Accordingly, on my reading of WL and SK, they establish, between them, that the legality of detention is to be determined according to whether the Hardial Singh principles have been observed. This was plainly the approach of the Court of Appeal in SK and the court in WL did not seek to question that in any way. A failure to apply a relevant policy or breach of a policy or even a rule such as rule 9 of the Detention Centre Rules will not on its own establish that the associated detention is unlawful. SK decided that and the court in WL applied that approach.
58. I think the emphasis of the two cases may be perceived to be slightly different. SK puts the Hardial Singh principles very much at the forefront of any consideration of the legality of a particular detention, certainly requiring observance of rules and policy/guidance, but treating the fact of observance or non-observance as something which may assist in determining whether the Hardial Singh principles are satisfied, rather than as a pivotal consideration in its own right. WL, whilst acknowledging that SK is binding authority that a failure in breach of procedural rules to review a man's detention does not necessarily render the detention unlawful (see paragraph 88 of the judgment in WL), does not deal with SK at any length. Hardial Singh is much less prominent in the reasoning of the court in WL than in SK. The concentration in WL is upon the need to show that the application of the unpublished policy was material to the decision to detain, without which the detention will not be unlawful. The outcome of WL's appeal resulted from the conclusion of the court that there was nothing to show that he was detained or kept in detention by the application of the unpublished policy, and that he was, in fact, properly and lawfully detained in the light of the high risk of him absconding and reoffending and the fact that at all stages there was a prospect of removal within a reasonable period; in these circumstances, the fact that there was an unpublished practice with regard to the detention of foreign national prisoners was irrelevant to his claim. The outcome of KM's appeal was determined by the conclusion of the court that even if the unpublished practice was applied to him, it made no difference to the decision to detain him and that he would inevitably have been detained anyway under the published policy, given the nature and seriousness of his offences; the reasons given for his detention were said to be "cogent in the extreme". The Hardial Singh principles were applied by the Court of Appeal to the case of WL but this was not so much as a part of the analysis of the impact of the application of the secret policy as by way of a discrete exercise in considering WL's argument, advanced separately, that the period of his detention was too long to accord with Hardial Singh (see, in particular, paragraph 98). As for KM, the court does not, itself, measure his detention against the Hardial Singh principles, but it endorses a conclusion reached by Davis J in a passage in his judgment which included an evaluation of whether KM's detention was justified in Hardial Singh terms.
59. R(MC (Algeria)) v SSHD [2010] EWCA Civ 347, which was decided after both SK and WL and in the light of them, is of importance in that, in my view, the approach of the court includes an analysis both in terms of the implications of the non-application of a policy and on the basis of the Hardial Singh principles. I will come to it in more detail shortly. First, I must consider what form the causation analysis that is dictated by the existing authorities takes.
60. It will be recalled that the argument between the parties on this issue is (if I may paraphrase it) as to whether, as the Appellant argues, the court will only find that a detention made in the face of some error in relation to the application of a policy (be it applying the wrong policy or omitting to apply a policy or to take account of a feature of it) is not unlawful if the person would inevitably have been detained notwithstanding the error or whether, as the Respondent argues, the court should examine for itself whether the person should have been detained, in which case the detention will have been unlawful only if the court, exercising its own discretion, would not have required it.
61. Approaching matters by means of a Hardial Singh analysis, as the court did in SK, obviates the need for any express consideration of the nature of the causation analysis. The error in relation to policy (and the impact it has actually had on the decision) becomes one of the circumstances in the evaluation of whether the detention complies with Hardial Singh. In contrast, the court's reasoning in WL did include passages which are relevant to the issue although, in reality, nothing there turned on the nature of the causation analysis. In WL's case, the issue did not need to be determined because there was no error in that the unpublished policy was not applied to him; in KM's case, it was inevitable that he would have been detained. In neither case, therefore, was it essential for the decision for the court to consider whether anything less than inevitability would have saved the detention from being unlawful. Nevertheless it is important to see what was said about it.
62. Paragraph 89 is the obvious starting point. It could be argued, however, that that paragraph embraces two degrees of probability. It says that it must be shown that the application of the unpublished policy was material to the decision and then goes on to say that if the decision to detain was inevitable then the application of the policy is immaterial. Depending on precisely how the concept of materiality is interpreted, one can, I think, contemplate cases in which the application of the policy was undoubtedly material to the decision but it is not possible to go so far as to say that it was inevitable that, even if the error had not occurred, the person would still have been detained. Depending on the subject matter, one may, for example, be inclined to view it as possible that a decision would still have been taken to detain him or probable that that would have occurred, but not inevitable. A slightly different formulation (but one which seems to me to be related to the formulation in paragraph 89) is used in paragraph 96 of the WL judgment which contains the court's observation that if, on the evidence, it was clear that even assuming a lawful consideration, there was "no realistic possibility" of a different decision having been reached, then that would rule out any liability. However, there being no realistic possibility of a different outcome must surely, to all intents and purposes, amount to the same thing as it being inevitable that the outcome would be the same so this probably takes the matter no further. The final paragraph which potentially bears on the issue is paragraph 90, which contrasts the test of materiality in the context of a prompt application for a quashing order in judicial review where, it is said, the court may quash the decision even if the evidence shows only a risk that it might have been affected by the illegality and the test in the context of a common law claim in tort which is concerned retrospectively with actual consequences and where the court would be entitled to look at the question of causation more broadly, asking whether the illegality was the effective cause of the detention.
63. The issue was raised for consideration in R(MC (Algeria)) v SSHD (see above) but not determined by the Court of Appeal because the claimant's past record of failing to comply with bail conditions, absconding and reoffending was so bad that despite his mental condition, he would inevitably have been detained.

R(MC (Algeria)) v SSHD [2010] EWCA Civ 347

64. Unusually, MC was an instance of the Court of Appeal itself undertaking the judicial review of the lawfulness of the claimant's detention in pursuance of the Schedule 3 powers. The claimant had a long string of offences, many committed on bail. Ultimately he was sentenced to a period of 33 months imprisonment. Thereafter, he was served with notice of intention to make a deportation order and kept in immigration detention. Whilst detained, he harmed himself on numerous occasions, his conduct becoming so extreme that he had to be restrained with a body belt. His history was convoluted but included a period released on bail by the AIT during which he failed to report and spent a further period in prison for more offences. He was again placed in immigration detention where his behaviour was disturbed. A personality disorder was diagnosed. Chapter 55 of the Guidance applied to him, just as to the Appellant in the present case, and should have been considered by those making decisions about his detention, once they were put on notice that there were issues about his mental health, but was not.
65. Several passages of the judgment of Sullivan LJ (with whom the other members of the court agreed) are relevant to the question of the nature of the causation analysis. At paragraph 44 the issue is identified as follows:

"44……In the Detailed Grounds of Defence it was submitted that even if the Defendant erred in failing to consider the policy of detaining the mentally ill only in exceptional circumstances, the Court could be satisfied on the balance of probabilities that this Claimant would have been detained even if the policy had been taken into account. In view of the Claimant's high risk of absconding and re-offending, and the ability to adequately treat his mental condition while in detention, there were very exceptional circumstances which justified his detention."

66. Sullivan LJ said that this submission was based upon Davis J's first instance decision in R (on the application of Abdi and Others) v SSHD [2008 EWHC 3166 (Admin) and recorded that after the Detailed Grounds of Defence had been drafted, the Court of Appeal, in WL, had dealt with the appeal from Davis J. He went on to say:

"50. [Counsel for the claimant] relying on paragraph 89 of the Court's judgment in WL submitted that it was necessary for the Defendant to demonstrate not merely that the Claimant would have been detained, but that his detention was "inevitable" if the policy relating to the detention of the mentally ill had been applied in his case. [Counsel for the Secretary of State] submitted that detention would have been inevitable in any event.

51. In view of the unusual circumstances of this case, it is unnecessary to resolve the question raised in paragraphs 90 and 96 of the Court's judgment in WL….. "

67. There then follows an account of why the claimant's past record of failing to comply with bail conditions, absconding and re-offending led to the conclusion that he "would inevitably have been detained".
68. MC is of particular interest in this appeal because the facts are similar to those in this case and it is an example of the application of the principles derived from WL and SK to a mental health case. It is to be noted that counsel for the claimant in MC accepted that a mere failure to consider the policy contained in Chapter 55.10 of the Guidance did not render the detention necessarily unlawful and the court endorsed this as being a proper concession (see paragraph 44). The court in MC was not, therefore, troubled by any disquiet as to whether or not that principle extended to a failure to take into account the policy considerations in relation to mental illness, albeit that, in the light of the claimant's concession, they did not receive submissions on the subject as we have done. This reinforces my view of the scope of the principle set out in paragraph 89 of WL.
69. In some ways, the approach of the court in MC is, it seems to me, an amalgam of the approaches in WL and SK. Sullivan LJ first considered the impact of the failure to have regard to the Guidance on the claimant's detention and concluded that he would inevitably have been detained even if it had been applied. He then turned to consider the lawfulness of the claimant's continuing detention in relation to which it was apparently common ground that the Hardial Singh principles should be applied. Three matters initially concerned the court in this regard – the length of detention which amounted to over 29 months, the fact that there was still no end to the detention in sight, and the adverse effect of the detention on the claimant's mental health. The matter was therefore adjourned for further evidence dealing with some of these issues. This established that there was a real prospect that the defendant would be in a position to return the claimant to Algeria within three months. The court concluded that the continued detention was not unlawful saying:

"In these circumstances, and bearing in mind the fact that the length of the Claimant's detention has been due, at least in part, to his non-co-operation, I am not persuaded that his continued detention is unlawful. However, the Defendant should not be complacent. Despite the Claimant's poor criminal record and propensity to abscond, this is a finely balanced case. The Claimant has been in immigration detention for 30 months. The Defendant could, and should, have acted with greater diligence during that period."

OM (Algeria) v SSHD [2010] EWHCA 65 (Admin)

70. We were referred to a number of first instance decisions in this area of the law, amongst them the decision of Mr CMG Ockleton sitting as a Deputy High Court Judge in OM (Algeria) v SSHD [2010] EWCA 65 (Admin) (a decision made before WL). Like the Appellant, the claimant in OM had a mental illness, had been convicted of a considerable number of criminal offences, and was the subject of a deportation order. He challenged his detention under the Immigration Acts in judicial review proceedings. The deputy judge found that the Secretary of State had not applied the policy contained in the Guidance in maintaining the claimant's detention, not having made any real assessment of the claimant's circumstances and having failed to make the judgment required, in the light of his mental illness, that his circumstances were very exceptional. Mr Ockleton referred to Cranston J's decision in the present case and noted that Cranston J had, as Mr Ockleton perceived it, struck the balance himself, making his own decision as to whether there were very exceptional circumstances that would have meant that the claimant would have been detained even if the Guidance had been followed. Mr Ockleton said that he took it that Cranston J thought that the result in his case was "obvious". In contrast, he considered the decision he himself had to take was not obvious. His view was that it was by no means clear that, if the balancing exercise required by the Guidance had not been done, the court ought to be the first to make the assessment. He had sought to address the issue in relation to OM by requesting a comprehensive review from the Secretary of State but the resulting product fell short of what was required and did not contain a proper justification for the claimant's detention so the judge found that the detention was unlawful.
71. Counsel for the Appellant submitted that an "obvious" result is very similar to an inevitable result and that Mr Ockleton, in apparently considering that that was sufficient to justify Cranston J taking his own view of the circumstances but doubting that it was correct to do so in a less obvious case, was rightly interpreting WL as setting a standard of inevitability.
72. The Respondent submits that the Court of Appeal has not laid down a new test of inevitability. He submits that the use of that word in WL merely reflects the fact that on the evidence before the court, detention was indeed inevitable. As can be seen from what I have already said, I agree that this is certainly a possible interpretation of WL although I do not think that, alone, it reliably disposes of the possibility that the court in WL did contemplate that unless the same outcome would have been inevitable without the error, the resulting detention would (or might) have been unlawful. However, the Respondent goes on to submit that this cannot have been the intention of the court because a test of inevitability would be inconsistent with the role of the court in unlawful detention cases which is to reach its own view as to whether the detention was arbitrary. A consideration of inevitability would be part and parcel of a conventional public law approach which, Mr Waite submits, is not the proper approach dictated by the authorities to the issue of detention.
73. The role of the court in relation to immigration detention has been considered recently in R (A (Somalia)) v SSHD [2007] EWCA Civ 804. That case also concerned paragraph 2(3) of Schedule 3. At first instance, it was held that there had been a period of unlawful detention but the Court of Appeal substituted a decision that the appellant had been lawfully detained throughout. Toulson LJ considered the role of the court in a passage commencing at paragraph 60. The question was whether it was "for the court to decide if A's detention for the period in question was reasonably necessary or whether the court's role is limited to reviewing on a narrower basis the reasonableness of the Home Secretary's decision to exercise his power of detention during that period." Toulson LJ said that a "subtle argument" had been advanced in support of a limited role and said:

"I intend no disrespect by not going into the refinements of Mr Giffin's argument but dealing with the matter on a broader basis. Where the court is concerned with the legality of administrative detention, I do not consider that the scope of its responsibility should be determined by or involve subtle distinctions. It must be for the court to determine the legal boundaries of administrative detention. There may be incidental questions of fact which the court may recognise that the Home Secretary is better placed to decide than itself, and the court will no doubt take such account of the Home Secretary's views as may seem proper. Ultimately, however, it must be for the court to decide what is the scope of the power of detention and whether it was lawfully exercised, those two questions being often inextricably interlinked. In my judgment, that is the responsibility of the court at common law and does not depend on the Human Rights Act (although Human Rights Act jurisprudence would tend in the same direction)." [my emphasis]

74. Longmore LJ agreed with Toulson LJ, as did Keene LJ who gave his own judgment. Keene LJ identified the principal issue there arising as concerning the exercise of the Secretary of State's power to detain, and in particular, whether the detention was for a reasonable period of time. He said, "There is also another issue, logically arising at an earlier stage, as to whether it is for the court or for the Secretary of State to determine whether the detention was for a reasonable period and therefore lawful". He recorded the argument as being that the court should not itself determine whether the period of detention exceeded a period that was reasonable but rather should ask whether it was open to the Secretary of State to regard detention as appropriate in the circumstances. Keene LJ's answer to this is contained in paragraph 71 of the judgment:

"71. It is to my mind a remarkable proposition that the courts should have only a limited role where the liberty of the individual is being curtailed by administrative detention. Classically the courts of this country have intervened by means of habeas corpus and other remedies to ensure that the detention of a person is lawful, and where such detention is only lawful when it endures for a reasonable period, it must be for the court itself to determine whether such a reasonable period has been exceeded. That has been the approach adopted in practice in the domestic cases to which we have been referred: Hardial Singh, R (I) v Secretary of State for the Home Department and, to my mind, Khadir. In addition, this issue fell to be considered explicitly in the case of Youssef v The Home Office [2004] EWHC 1884, where Field J held that the court was the primary decision-maker as to the reasonableness of the length of detention: see paragraph 62.

"72. The Privy Council seems to have adopted a similar approach in Tan Te Lam, finding that the facts which had to be found for the power to detain to exist were jurisdictional facts and hence for the court to determine. Mr Giffin has pointed out that the decision went to the existence of the power rather than to its exercise, which is true, but the reasoning in that decision seems to be of broader significance……"

75. Later Keene LJ continued:

"74…..Nonetheless, this court is still required by s 6(1) [of the Human Rights Act 1998] to decide whether or not the detention of this individual is compatible or not with his rights under Article 5, because only by so doing can the court ensure that it is acting lawfully. It cannot do that merely by asking whether it was open to the Home Secretary to decide that the length of detention was reasonable, as opposed to whether it was actually reasonable in the eyes of the court….." [my emphasis]

76. At paragraph 75, he concluded his consideration of the issue by saying

"…the ultimate decision is, in my judgment, for the court. I therefore would reject the Secretary of State's submission as to the limited role of the court in cases such as this."

77. I take A (Somalia) as binding authority that the court must assume the role of primary decision maker when considering the lawfulness of detention rather than simply reviewing the decision of the Secretary of State along traditional public law lines. This disposes of the Appellant's argument that Cranston J was wrong to conduct his own assessment of the justification for the Appellant's detention as if he were the primary decision maker. In my view, it does also shed some light, as the Respondent submits, on how one should interpret the causation analysis that is required when considering the implications of an error in the application of policy. I am inclined to agree that if the court confines itself to a consideration of whether it was inevitable that the decision would have been the same if the error had not been made, it is abdicating the responsibility that it has as primary decision maker in cases concerning the liberty of the subject. It might have been argued by the Appellant that the abdication produces a position which in fact favours the person who has been detained and is therefore, at the very least, entirely acceptable because it removes the usual acceptance of a band of reasonable decisions and requires the Secretary of State to surmount the ultimate hurdle of inevitability if the detention is not to be declared unlawful. In fact this was not argued and, in any event, it seems to me that the "inevitability hurdle" introduces an undesirable degree of sophistication and technicality into what could be, and I think is, essentially a simple position i.e. that the court itself decides on the legality of the detention, attaching appropriate weight in its deliberation to matters such as government policies, risk assessments, and the evidence as to likely time-scales for the deportation of the individual. It also gives rise to the challenging question of how such an approach fits alongside the rather different approach required of the court in applying the Hardial Singh principles. I am not persuaded that the interests of justice would be served by there being two different regimes (or distinct stages) – consideration of the implications of a breach of policy and consideration of the Hardial Singh principles - when issues relating to the application of policy can, in fact, be subsumed into the court's consideration on Hardial Singh lines in the way which I think was contemplated by SK.
78. Against this analysis, I return to Cranston J's decision in order to look again at the way in which he determined the implications of the Respondent's failure to apply his own policy in relation to those who are suffering from mental illness. The relevant passage is essentially that in paragraphs 67 to 69, and particularly paragraphs 68 and 69 which can be found rehearsed earlier in this judgment.
79. The Appellant argues that the judge erred in that he placed the burden of proof on the Appellant to show that even if the Respondent had applied his own policy, his detention would not be justified under it. It is certainly correct that Cranston J said, at the beginning of paragraph 68, that "the claimant has not demonstrated that his detention is in breach of the policy". The Respondent accepts that the burden of proving that the detention was authorised by statute and not in breach of the Hardial Singh principles was not on the Appellant but on the Respondent. However, he submits that when the judgment is read as a whole, it is plain that the judge reached his own decision about the lawfulness of the detention and that any misstatement of the burden of proof did not infect his approach to the evidence on this issue. I accept that argument and do not consider that, when read in the round, the judgment reveals any material error in relation to the burden of proof.
80. Was the judge's evaluation of the evidence in relation to the Appellant's circumstances and his determination about the lawfulness of his detention flawed? The Appellant argues that it was and that Cranston J was not entitled to determine, as he did, that the "balance is against release". This was an exercise of judgment by the judge, weighing up all of the factors in the case. He not only carried out the exercise against the backdrop of Chapter 55 of the Guidance but also, in paragraphs 70 and 71 of his judgment, in the light of the Hardial Singh principles. He obviously had in mind the provisions of Chapter 55 as they affect those with a mental illness and, in particular, that the Secretary of State's policy was that people with mental illness should only be detained in very exceptional circumstances. It can be seen from paragraph 71 that he did not ignore the fact that there was medical evidence that detention was having a detrimental effect on the Appellant's mental state. He commented, entirely accurately, that "the balance of other factors has to be very substantial indeed for detention to be justified". The factors that he identified as weighing against release were factors that are well recognised by the authorities (see, for example, A (Somalia), above) as relevant in this regard, namely the risk of absconding and the risk of further offending, as well as the Appellant's own disruption of legitimate attempts to effect his removal, his refusal to co-operate with attempts to obtain documentation for him and his belated challenges, aimed at frustrating his otherwise lawful deportation.
81. As can be seen in the passage from paragraph 55 of the judgment which I set out earlier, Cranston J's interpretation of Chapter 55 was that when dealing with mentally ill people and requiring that they should be detained only in very exceptional circumstances, the Guidance contemplated that what was very exceptional should be judged taking into account the whole spectrum of those who are liable to be removed in the immigration context. Mr McDonald argues that this is not an accurate interpretation of the Guidance. He says that the phrase "very exceptional circumstances" is designed to invite comparison only with other CCD cases, indeed I think he would go so far as to confine the group for comparison purposes not just to criminals but to criminals with mental health issues. This, to my mind, is far too narrow an approach to the Guidance. There is nothing in it to suggest that it is intended to be confined in this way and, in my view, the judge was right to look at the issue of very exceptional circumstances, as he did, in the context of a general spectrum with the average asylum seeker at one end and high risk terrorists near the other end. He was also right to give substantial weight to serious criminal offences in considering the balance of factors, this being precisely what the Guidance requires, as can be seen from the passages I quoted at the beginning of this judgment. Indeed, I do not consider that the Appellant has been able identify any material respect in which Cranston J erred in his analysis and determination of the question of the lawfulness of his detention or his conclusion that the detention was justified.
82. I would add, for the sake of completeness, that had I concluded that the correct question for the judge was, in fact, whether it was inevitable that the Appellant would have been detained even if Chapter 55 had been properly applied, although the issue would have caused me a degree of anxiety, it is likely that I would have been persuaded that on the facts here, it was inevitable.
83. In all the circumstances, I would dismiss this appeal.

Lord Justice Longmore:

84. I agree with Black LJ that this appeal should be dismissed but would like to add something on what has been termed the "causation question", namely Mr Macdonald's argument that detention ordered as a result of an error in relation to the application of a published policy of the Secretary of State will only be lawful if the applicant would have been "inevitably" detained notwithstanding the error. The word "inevitablity" has its source in para 89 of WL (Congo) and I agree with Black LJ that this court was not intending to lay down a legal test of inevitability as the test of causation. But I would respectfully question whether this court needed, in the light of SK (Zimbabwe), to lay down a causation test of any kind at all.
85. As my Lady has explained, SK challenged his detention, ordered pursuant to section 2(2) of Schedule 3 of the Immigration Act 1971, on the basis that there had been no regular review of his detention as required by both rule 9 of the Detention Centre Rules 2001 and the policy of the Secretary of State as set out in the Home Office's Operations Enforcement Manual. This court held that compliance with the requirements of the Rules or the Manual could only be a condition of the power to detain if legislation so provided and it did not. Laws LJ summarised his views (with which the other members of the court agreed) in para 35 of his judgment in the following way:-

"In seeking to formulate the issue before us I posed the question, what is the reach of the power conferred by paragraph 2(2) of Schedule 3 to the Immigration Act 1971, and characterised it as a question of statutory construction. In light of all the matters I have canvassed I would summarise my conclusions on this issue as follows. (i) Compliance with the Rules and manual as such is not a condition precedent to a lawful detention pursuant to paragraph 2(2). Statute does not make it so (contrast section 34(1) of PACE, and the Roberts case [1999] 1 WLR 662). Nor does the common law, or the law of the Convention. (ii) Avoidance of the vice of arbitrary detention by use of the power conferred by paragraph 2(2) requires that in every case the Hardial Singh principles should be complied with. (iii) It is elementary that the power's exercise, being an act of the executive, is subject to the control of the courts, principally by way of judicial review. So much is also required by Convention article 5(4). The focus of judicial supervision in the particular context is upon the vindication of the Hardial Singh principles. (iv) In the event of a legal challenge in any particular case the Secretary of State must be in a position to demonstrate by evidence that those principles have been and are being fulfilled. However the law does not prescribe the form of such evidence. Compliance with the Rules and the manual would be effective and practical means of doing so. It is anyway the Secretary of State's duty so to comply. It is firmly to be expected that hereafter that will be conscientiously done."

In other words, on the question of the lawfulness of the detention, the court should apply the Hardial Singh principles and that should be that.

86. There is no mention here of any causation test and it seems that Laws LJ thought causation was irrelevant since in paragraph 34 he had said this:-

"It is important to notice that if this approach is wrong (namely if the Hardial Singh principles are not enough to determine the lawfulness of the detention and it is relevant to consider whether there has been an error in application of Home Office policy in relation to monitoring) it means that a detention will be unlawful in the absence of (or failure to fulfil) a system of internal monitoring even though it can be shown on the particular facts that the detention, far from being arbitrary, is wholly justified."

In other words, even if it was inevitable that on proper application of the relevant policy detention would have been ordered, detention made on a misapplication of the policy would be unlawful.

87. When therefore the cases of WL (Congo) and KM (Jamaica) came before Davis J and this court it would, as it seems to me, have been open to the Secretary of State to submit that, although the appellants had been detained pursuant to an unlawful (unpublished) policy and not pursuant to the lawful (published) policy, the only question that needed to be determined was whether detention was lawful according to Hardial Singh principles. As it is the submission made and recorded in para 80 of the judgment was

"where the unlawful policy was of no causative effect (because detention would still have been lawfully directed irrespective of the new policy) then the detention is not to be styled as unlawful detention at all."

It was this submission that this court accepted in para 89 of its judgment despite the opposition of Mr Raza Husain for the appellants.

88. No doubt this court in WL (Congo) did not wish to go further than the Secretary of State's submissions but the logic of SK must be that causation is irrelevant to the question of the lawfulness of the detention which is to be assessed only according to Hardial Singh principles. It is a curiosity, moreover, that the court drew a distinction between detention pursuant to sub-para (1) of para 2 Schedule 3 of the 1971 Act and detention pursuant to sub-paras (2) and (3) of para 2. As far as sub-para (1) was concerned, it accepted that SK was authority for the proposition that causation was irrelevant; yet the court did not advert to the fact that SK was actually an authority on sub-para (2) of para 2 (see paragraph 3 of the judgment of Laws LJ) of Schedule 3, when it accepted that causation was relevant in cases coming under that sub-para and sub-para (3).
89. Be all that as it may, I would myself give, as an additional reason for upholding the judgment of Cranston J, the fact that the detention of the appellant was not unlawful because there was no infringement of the Hardial Singh principles.

Lord Justice Maurice Kay:

90. I too would dismiss this appeal.
91. I am most grateful to Black LJ for her detailed review of the authorities. For my part, I do not consider them to be entirely consistent in approach. SK gives pride of place to the Hardial Singh principles and marginalises the Secretary of State's published policies. Compliance with the Hardial Singh principles renders the detention lawful. Later authorities, in particular WL, seem to proceed on the basis that compliance with the Hardial Singh principles is a necessary condition of lawful detention but not always the whole story. If, notwithstanding compliance with Hardial Singh, there has been a failure to have regard to the published policy of the Secretary of State, that failure may render the detention unlawful unless consideration of the policy would have inevitably resulted in the same decision to detain. This is variously described as the materiality or causation approach. I am bound to say that, if the question were free from authority, I would favour the materiality approach. I can see no reason why a Secretary of State who promulgates a policy which augments existing judicially-developed requirements should not be under a duty to have regard to his policy unless there is a good and articulated reason not to do so.
92. In the present case, I am entirely satisfied that the outcome would be the same, whichever approach is adopted. The Secretary of State acted consistently with the Hardial Singh principles and the failure to apply the policy was of no materiality because, as Cranston J found, the appellant's history meant that a decision to detain was inevitable.
93. In these circumstances, and as I understand that the Supreme Court will be hearing the appeal in WL next month, any further analysis from me would be otiose.