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

The main issue in this appeal is whether the Secretary of State has the power to place a person on bail under para. 1(2) of Sch. 10 to the Immigration Act 2016 (“the 2016 Act”) in circumstances where it would be unlawful to detain them. According to the evidence before the Court there may be more than 90,000 people who are currently on “immigration bail”, as it is described in the 2016 Act, some of whom (like this Appellant) cannot lawfully be detained.

The High Court (Elisabeth Laing J) decided that there is a power to bail a person in those circumstances. The Judge granted permission to appeal to this Court, recognising that this case raises an important point of statutory interpretation, which could affect a large number of people.

We have been assisted by detailed written and oral submissions by Ms Stephanie Harrison QC, who appeared with Mr Alex Goodman and Mr Matthew Fraser for the Appellant; Ms Laura Dubinsky, who appeared with Mr Anthony Vaughan and Ms Eleanor Mitchell for the Intervener, Bail for Immigration Detainees (“BID”); and Mr Robin Tam QC, who appeared with Ms Emily Wilsdon for the Respondent. I express the Court’s gratitude to them all.

According to the Appellant, he first entered the United Kingdom (“UK”) clandestinely on 17 December 2006 and sought asylum. He claims to be a Guinean national.

On 16 January 2007 his first application for asylum was refused and on 5 March 2007 his appeal from that decision was dismissed by the First-tier Tribunal (“FTT”). On 28 September 2007 a reconsideration application was refused and his appeal rights became exhausted.

On 9 June 2009 the Appellant was convicted of possession of false documents at Glasgow Sheriff Court and was sentenced to 14 months’ imprisonment. On 2 July 2009 the Respondent served him with a notice of liability to deportation and in response he made submissions under Articles 2 and 3 of the European Convention on Human Rights (“ECHR”).

On 10 August 2009 the Respondent submitted a request to the Guinean Embassy for an Emergency Travel Document (“ETD”) for the Appellant and on 5 November 2009 served him with a Reasons for Refusal Letter rejecting his ECHR claim.

On 25 November 2009 the Respondent served a Deportation Order on the Appellant.

On 4 February 2010 the FTT dismissed the Appellant’s appeal against the refusal of his ECHR claims and the deportation order.

The Appellant had been given temporary admission in 2006, when he applied for asylum, but he was in custody from June 2009, initially serving his prison sentence and then in immigration detention pending deportation. On 21 January 2011, he was granted bail by the FTT. He was released on 25 January 2011. Since that date he has not been in detention.

The Respondent’s records show that, on 25 January 2011, the Appellant was granted temporary admission but, in a Note filed with this Court after the hearing before us, she considers that, in fact, he was still on the bail granted by the FTT but that Form IS.96 (Tag & Track) was used to set out the conditions.

In accordance with the bail granted by the FTT, the Appellant was required to report to a Chief Immigration Officer on 3 March 2011 and did so. His bail conditions were varied. There were conditions as to residence, reporting and not working. There was also an electronic monitoring condition but, in July 2011, this was removed.

On 3 March 2011 the Respondent was informed by the Guinean Embassy that there was still no response from the authorities in Guinea to the ETD application and on 7 June 2013 the Embassy requested that the application be re-submitted.

Between 12 January 2012 and 29 November 2013 the Appellant’s solicitors filed three further sets of submissions supporting his asylum claim and each of these was refused.

On 20 January 2014 the Guinean Embassy advised the Respondent that the ETD application had been referred to Conakry for “verification checks”. The Respondent’s records show that almost monthly notes were made stating that she was awaiting the outcome of those checks until March 2016.

On 7 July 2014 the Appellant made further submissions under the legacy scheme and on 20 August 2015 made an application for leave to remain in the UK on the basis that he is a stateless person. On 29 February 2016 the Respondent refused this application.

On 10 May 2016 the Appellant completed a further bio-data form to assist the ETD application. On 27 May 2016 the Respondent made a fresh ETD application to Guinea and the Appellant attended a further interview at the Embassy. On 19 August 2016 the Guinean Embassy rejected this application and stated that he was not a Guinean national.

On 12 September 2016 a further application for leave to remain was made on the ground that the Appellant is stateless but was again refused.

On 16 September 2016 the Appellant was issued with a Notice of Restriction letter (Form DO4), which imposed restrictions under para. 2(5) of Sch. 3 to the Immigration Act 1971 (“the 1971 Act”) relating to reporting, residence and prohibiting employment. These were similar to bail conditions to which he had been subject previously. For convenience this is often called a “restriction order” although that phrase is not used in the 1971 Act.

On 22 November 2016 the Appellant completed a third bio-data form in the presence of an immigration officer and, on this occasion, questions were put to him regarding the Guinean Embassy’s assertion that he was in fact a Ghanaian national.

On 15 February 2017 the Appellant made further submissions in support of his application that the deportation order should be revoked. On 26 May 2017 he responded to a “Section 120 Notice”; and on 23 June made further submissions on medical grounds.

The relevant provisions of Sch. 10 to the 2016 Act came into force on 15 January 2018. As will become apparent later, transitional provisions had the effect that the Appellant was moved from having been subject to a restriction order imposed by the Secretary of State under the 1971 Act to “immigration bail” under the 2016 Act, and subject to the same conditions as previously. These conditions were then varied on 5 June 2018: the Appellant was permitted to work but only in one of the occupations on the Shortage Occupation List.

On 6 December 2018 the Appellant made an application for leave to remain in the UK on protection grounds.

On 4 May 2020 the Appellant responded to a request from the Respondent to clarify aspects of the various applications he had made. On 27 May 2020 (one week before the High Court hearing), the Respondent refused the four sets of submissions made between 15 February 2017 and 6 December 2018 and this decision gave the Appellant a right of appeal to the FTT. That appeal was recently allowed in part, in a decision dated 25 June 2021, and, by a decision dated 16 July 2021, the Respondent was given permission to appeal to the Upper Tribunal: that appeal is still pending.

At the hearing before the High Court the Respondent took the position that the Appellant is a Ghanaian national, based on two pieces of information: notes in the Respondent’s case files, recording that an official in the Guinean embassy had stated that Mr Kaitey’s name and that of his school were Ghanaian; and evidence of the Appellant’s connections through Facebook to people in Ghana, including close family members. The Appellant contends that this is unsurprising given that he has family who fled to Ghana at the same time that he fled Guinea.

The Appellant has not committed any offences since his release from detention in January 2011 and has complied with his reporting conditions.

On 28 October 2019 the Appellant’s solicitors sent a pre-action protocol letter to the Respondent to challenge the lawfulness of the imposition of conditional bail on him.

On 3 December 2019 the Appellant filed his claim for judicial review in the High Court. On 23 January 2020, May J granted permission to bring that claim.

Following a remote hearing on 4 June 2020, Elisabeth Laing J handed down judgment on 13 July 2020, dismissing the claim for judicial review.

On 15 July 2020 the Appellant applied to the High Court for permission to appeal. On 17 July 2020 the Judge granted permission to appeal, not because she thought that it had a real prospect of success but on the ground that there was some other compelling reason, given that the case concerned a new statutory concept (“immigration bail”) and it was said that the outcome could affect around 90,000 people.

On 20 November 2020, BID were granted permission by Lewis LJ to make written and oral submissions in this Court, as they had been in the High Court.

The main provision which we must construe on this appeal is para. 1(2) of Sch. 10 to the 2016 Act:

Para. 1(1) provides:

Para. 1(5) of Sch. 10 provides:

Para. 10 of Sch. 10 concerns arrest for breach of immigration bail. Para. 10(1) provides:

Under para. 10(9) a person arrested under that paragraph must, as soon as is practicable, be brought before the relevant authority and may be detained under the authority of the Secretary of State “in the meantime”. The relevant authority is the Secretary of State if it was the Secretary of State who had granted immigration bail: para. 10(10). The relevant authority must then decide whether the arrested person has broken or is likely to break any of the bail conditions: para. 10(11). Para. 10(12) provides:

Section 61 of the 2016 Act, as it was from 12 May 2016 to 14 January 2018, provided:

Section 61 of the 2016 Act, as it has been since 15 January 2018, now provides as follows. Subsection (1) gives effect to Sch. 10 (Immigration Bail). Subsections (3)-(5) of the former section 61 were repealed on the coming into force of the repeal of paras. 22 and 29 of Sch. 2 to the Immigration Act 1971 by para. 20 of Sch. 10 to the 2016 Act.

As I have mentioned, on the date when Sch. 10 to the 2016 Act was brought into force, the Appellant was subject to a restriction order imposed by the Secretary of State. We have been shown the version of the Immigration Act 1971 which was in force from 12 July 2016 to 14 January 2018, the day before the main provisions of Sch. 10 to the 2016 Act came into force.

During that period the relevant legislative regime which applied to this Appellant was as follows. Para. 2 of Sch. 3 to the 1971 Act provided:

Para. 13(1) of Sch. 10 to the 2016 Act provides that regulations under section 92(1) may, in particular, provide for a person to whom that sub-paragraph applies to be treated, for such purposes as may be specified, as having been granted immigration bail in such circumstances and subject to such conditions as may be specified. Sub-para. (2) then lists the persons to whom that provision applies. This Appellant fell within sub-para. (2)(d), as he was a person liable to be detained under para. 2(3) of Sch. 3 to the 1971 Act.

The relevant regulations were the Immigration Act 2016 (Commencement No. 7 and Transitional Provisions) Regulations (SI 2017 No. 1241).

Regulation 2 brought into force the relevant provisions of Sch. 10 on 15 January 2018. Regulation 3 gave effect to the Schedule to the Regulations, which contained transitional provisions.

Para. 1 of the Schedule provides:

The Judge held that the imposition of immigration bail under para. 1(2) of Sch. 10 to the 2016 Act is not dependent on a power lawfully to detain a person under immigration powers.

In summarising the facts of this case the Judge said, at para. 8, that the Appellant was released from immigration detention on bail on 27 January 2011 (in fact, we were informed, he was released on 25 January) and has been on bail since then. That was the impression that the Judge had been left with but, as became clear during and after the hearing before this Court, that is not entirely accurate because it does not give the full picture. At the Court’s request, the parties filed an Agreed Note setting out in detail exactly what the position was between 2011 and 2018.

Although there is some confusion in the documents, because they sometimes refer to temporary admission, it is clear that the restrictions which were imposed by the Secretary of State, certainly from 16 September 2016, were not in the form of conditions attached to bail but were rather restrictions imposed pursuant to powers in para. 2(5) of Sch. 3 to the 1971 Act. As I have mentioned earlier, the effect of transitional provisions was to continue those restrictions more recently, from 15 January 2018, when Sch. 10 to the 2016 Act came into force, so that they became conditions of immigration bail.

At paras. 14-50, the Judge set out in detail a chronological history of the legislation as it was at various times, how it was amended, and how it related to the decisions of the courts in R (Khadir) v Secretary of State for the Home Department [2005] UKHL 39; [2006] 1 AC 207; and B (Algeria) v Secretary of State for the Home Department [2018] UKSC 5; [2018] AC 418.

At paras. 64-69, the Judge considered the potential relevance of the legislative history to the issue of interpretation which she had to decide. She referred, in particular, to Bennion on Statutory Interpretation (7^th ed., 2017); the decision of the House of Lords in Barras v Aberdeen Steam Trawling and Fishing Co Ltd [1933] AC 402; and the decision of the Supreme Court in R (N) v Lewisham London Borough Council [2015] AC 1259, at para. 53, where Lord Hodge JSC said that:

At para. 69, the Judge reminded herself that this is a presumption and not a rule of law. She also observed that there is a further question, which is whether, in provisions which potentially impinge on the right to liberty, Parliament has used “the clearest possible words” to achieve the result for which the Secretary of State contended, that is to make the power to grant bail available in a case in which the underlying power of detention cannot lawfully be exercised.

Having considered the respective arguments in detail, the Judge concluded, at para. 81, that the term “immigration bail” is not “ordinary” bail, precisely because it is available when a person is liable to detention (rather than being detained); and because it is available when a person can no longer be detained (whether as a matter of law or in practice). She said that:

The Judge considered that Parliament intended that immigration bail should replace the previous concepts of temporary admission, temporary release and bail. Immigration bail in Sch. 10 covers the field of those three former powers and should be available even when the underlying power of detention cannot lawfully be exercised.

The Judge said that para. 10(12)(a) of Sch. 10 does not undermine this construction, as it does not provide an independent authority for detention: it does not make detention lawful if it would otherwise be unlawful.

At para. 83, the Judge said that she did not consider that this case raises any discrete Article 5 issue. It was not suggested that the Appellant’s bail conditions amount to a deprivation of liberty. She said that no purpose would be served by abstract theorising about this issue divorced from any actual facts.

At para. 84, the Judge was not persuaded by the alternative argument advanced by BID. She regarded Ms Dubinsky’s argument based on R v Governor of Durham Prison, ex p. Hardial Singh [1984] 1 WLR 704 as essentially a circular one. She said that either the principles in Hardial Singh apply or they do not.

At para. 85, the Judge said that it was unnecessary for her to decide whether or not, if the Secretary of State did not have the power to impose bail conditions, she would have been obliged to give the Appellant leave. Nevertheless, she said that it would be surprising if it were appropriate for a court to step, in effect, into the Secretary of State’s shoes and exercise the power which Parliament has conferred on the Secretary of State, that is whether or not to grant any form of leave to enter or remain in the UK.

The Appellant advances four grounds of appeal:

Ground 1 (misinterpretation of Sch. 10 to the 2016 Act): the judge erred in not interpreting the power to grant immigration bail as being predicated on its being lawful to exercise the power to detain in accordance with the principles in Hardial Singh, with the consequence that the phrase “liable to detention” means “liable to lawful detention”.

Ground 2 (failure to consider whether “some prospect” of removal): in the alternative, if bail can be imposed even where detention would breach the Hardial Singh principles, the Judge failed to determine whether there was, on the facts of this case, “some prospect” of removal since that is necessary to give rise to the existence of the detention power in the first place.

Ground 3 (implied limits on bail power): the Judge erred in rejecting the alternative argument that there are implied limits, similar to the Hardial Singh principles, on the exercise of the power to grant bail.

Ground 4 (obligation to grant leave): the Judge erred in not holding that the absence of a lawful bail power would compel the grant of some form of leave to be in the UK.

Ground 3 is in the alternative to Ground 1 and was argued more fully by Ms Dubinsky, who appeared for BID: her arguments were adopted on behalf of the Appellant by Ms Harrison.

Ground 4 only arises if either Ground 1 or Ground 2 succeeds. For reasons that will become apparent, it is unnecessary, in my view, to address Ground 4 because I would dismiss the other grounds of appeal.

Under Ground 1, Ms Harrison contends that the striking effect of the Judge’s decision is that the Appellant can be held subject to conditional bail indefinitely, even though there is no realistic prospect that he will be deported within a reasonable time and even though, for that reason, he may not lawfully be detained under immigration powers.

She submits that the concept of immigration bail in the 2016 Act should be interpreted consistently with the “ordinary” concept of bail, which is akin to custody: this is why the writ of habeas corpus may issue even in relation to a person who is on bail, for example awaiting extradition.

Ms Harrison accepts that the decision of the House of Lords in Khadir is binding on this Court but reserves the Appellant’s position should this case go further. In any event, she submits that the phrase “liable to detention” was interpreted by the House of Lords in Khadir in the context of temporary admission but not in the present context of bail provisions. She submits that this difference in context is critical because (1) the former temporary admission power was more closely aligned to the power to grant conditional leave than to bail; and (2) the concept of bail has a specific legal meaning predicated on a right lawfully to detain. In support of that submission she places particular reliance on the decision of the Supreme Court in B (Algeria).

Ms Harrison points out that, in the scheme of the 2016 Act, a person who breaches bail conditions may be detained in consequence of the breach. She submits that the Judge’s reasoning erred in not treating para. 10(12) of Sch. 10 as casting light on the interpretation of the Schedule as a whole.

Finally, Ms Harrison submits that the result of the Judge’s construction of the 2016 Act is an incompatibility with Article 5 of the ECHR. Detention is permitted under Article 5(1)(f) in the case of a person against whom action is being taken “with a view to deportation”. It is well established in the case law of the European Court of Human Rights that, where there is no realistic prospect of deportation within a reasonable period of time, detention in the meantime cannot lawfully be with a view to deportation.

I do not accept those submissions for the following reasons. I will first set out an overview of what I consider to be the correct interpretation of Sch. 10 to the 2016 Act. I will then consider the main authorities cited to us. Finally, I will address specific aspects of the submissions made to us on Ground 1 in more detail.

First, the phrase “liable to detention”, on its natural meaning, refers to a person who can in principle be detained: in other words that there exists a legal power to detain them. It does not say “liable to lawful detention”. Nor does it say that the power to detain must not only exist but must be capable of being exercised lawfully.

Secondly, this is how the phrase “liable to detention” was interpreted by the House of Lords in Khadir. In my view, that was in a similar context to the present context, since it involved restrictions on what somebody who did not have leave to enter or remain could do while in this country. Although that case concerned the concept of temporary admission, it was still a similar context. Indeed, I note that, in her separate opinion, at para. 4, Baroness Hale of Richmond referred to the mental health context, where a similar phrase is used: that context is far removed from immigration law and yet she clearly thought that the analytical difference between the existence of a power to detain and the question whether it can lawfully be exercised was relevant there too. It follows therefore that the presumption set out in decisions such as Barras is applicable. It is nevertheless only a presumption.

Thirdly, the legislative history points strongly towards this interpretation. As is made clear in Bennion, the legislative history of a statutory provision is relevant and can be taken into account in its interpretation. Here the chronology of events makes it clear that the intention of Parliament in enacting the original version of section 61 of the 2016 Act, and then bringing into force Sch. 10 to that Act, was to reverse the effect of the decision of the Court of Appeal in B (Algeria). This is also supported by the Explanatory Notes to the 2016 Act.

Fourthly, it is important to bear in mind an important distinction between the terms of the original version of section 61(3) of the 2016 Act and para. 1(5) of Sch. 10 to that Act. The original version of section 61(3) was confined to persons who were currently in detention, which is why it referred to their being “released”. In contrast, para. 1(5) does not require the person to be currently detained. It simply speaks of the grant of bail.

Fifthly, the words of para. 1(5) of Sch. 10 to the 2016 Act are entirely apt to describe (or at least to include) the category of persons who cannot be deported because of the Hardial Singh principles: “even if the person can no longer be detained”.

Sixthly, I am not persuaded by the argument for the Appellant that the historical concept of bail is significant in this context. It is important to bear in mind that the legal concept which needs to be interpreted on this appeal is not “bail” generally but “immigration bail”, which is the concept used in the 2016 Act. It is clear from the power to enact regulations setting out transitional provisions (in para. 13(1) of Sch. 10) that the concept of immigration bail was intended to include cases where previously there had been temporary admission (as in Khadir) or a restriction order such as the one that was imposed in this case.

In this context it is important to bear in mind that using all of these aids to statutory interpretation, including the legislative history and the purpose of the legislation, is a means of arriving at the true meaning of the enactment of Parliament. This exercise is not divorced from the question, to which the submissions for the Appellant have drawn attention, of whether sufficiently clear language has been used to override the presumption that Parliament does not intend to interfere with personal liberty (the principle of legality). As the exercise of answering that question is not a purely linguistic one, all of the various aids to interpretation which I have mentioned have a role to play in assisting the Court to arrive at the true meaning of the legislation enacted by Parliament.

In any event, I am not persuaded that the interpretation reached by the Judge would have the result that there would be a violation of fundamental rights. The grant of bail is not inherently a violation of fundamental rights. The conditions attached to bail may, depending on the facts of an individual case, amount to such a violation but, if they do, it will be open to a person subject to them to complain that they are unlawful.

Furthermore, the Convention rights as set out in the Human Rights Act 1998 (“HRA”) will still be there. No one, including the Respondent, has suggested that the 2016 Act in some way overrides the provisions of the HRA. If it were necessary to do so, it would have to be read in a way which is compatible with the Convention rights: see section 3 of the HRA. In any event, if, on the facts of a particular case, a person is a “victim” within the meaning of section 7 of the HRA, they will be able to rely on their Convention rights to argue that the Secretary of State has acted unlawfully under section 6 of that Act. If that argument succeeds, they will in principle have a remedy under section 8. None of that, in my view, leads to the conclusion that the primary legislation itself must be read in any different way from its ordinary meaning.

In Tan Te Lam v Superintendent of Tai A Chau Detention Centre [1997] AC 97 the Judicial Committee of the Privy Council approved the Hardial Singh principles, at pp. 108 and 111 (Lord Browne-Wilkinson). At p. 111, Lord Browne-Wilkinson went on to state that, although it was open to the legislature to vary or even exclude those implied restrictions by express provision:

At p. 113, in similar vein, Lord Browne-Wilkinson said that:

All of that I would readily accept. But, in my view, it does not lead to the conclusion that the interpretation of the 2016 Act reached by the Judge in the present case is wrong. I would observe that those passages, important as they are, relate to actual detention by the executive. The decision in Tan Te Lam itself was not concerned with any issue about bail.

Khadir concerned an Iraqi national from the Kurdish autonomous area of Northern Iraq. He entered the UK clandestinely in November 2000 and, having claimed asylum, was granted temporary admission under para. 21 of Sch. 2 to the 1971 Act. By section 11 of that Act someone temporarily admitted under Sch. 2 is deemed not to have entered the UK. In other words, although they are physically on the territory of the UK, they are treated in law as if they were not here.

At first instance, Crane J had held that the Secretary of State did not have power to grant temporary admission in the circumstances of that case because it was not possible to deport Mr Khadir within a reasonable time and therefore the power to detain him could not lawfully be exercised. Crane J ordered further consideration forthwith of an application for exceptional leave to remain. He then stayed his order pending determination of the Secretary of State’s appeal to the Court of Appeal.

Following the decision of Crane J, Parliament enacted section 67 of the Nationality, Immigration and Asylum Act 2002 (“the 2002 Act”). I will set out its terms later. This provision had retrospective effect. The Secretary of State’s appeal was allowed by the Court of Appeal on the basis of section 67 of the 2002 Act.

The appellant then appealed to the House of Lords. In giving the main opinion, Lord Brown of Eaton-under-Heywood dealt with the law as it stood before the amendment made by section 67.

Para. 21 of Sch. 2 to the 1971 Act was critical. It provided:

The crucial part of the reasoning of Lord Brown appears at paras. 31-33:

At para. 36, Lord Brown said that it followed that section 67 had been an unnecessary enactment:

B (Algeria) did not concern temporary admission but did concern the grant of conditional bail. In that case the claimant was served with notice of the Secretary of State’s intention to deport him on the ground that he was suspected of being an Algerian terrorist. He was detained under para. 2(2) of Sch. 3 to the 1971 Act pending the making of the deportation order. He appealed to the Special Immigration Appeals Commission, which granted him conditional bail, pursuant to its power under para. 29 of Sch. 2 to the 1971 Act, as modified by para. 4 of Sch. 3 to the Special Immigration Appeals Commission Act 1997, pending determination of his appeal. The claimant then served four months in prison for failing to comply with directions made by the Commission. The Commission then again granted him conditional bail. Subsequently the Commission found that there was no reasonable prospect of his removal to Algeria and that therefore he could not lawfully be detained pursuant to para. 2(2); but it continued his bail on slightly relaxed conditions.

The Court of Appeal allowed the claimant’s appeal, holding that the Commission had no jurisdiction to grant or impose conditions on bail since any statutory provision which purported to permit the deprivation of individual liberty by administrative detention should be construed strictly and, since the power to grant bail presupposed the existence of, and the ability to exercise, the power to detain, the word “detained” in paras. 22 and 29 of Sch. 2 to the 1971 Act meant “lawfully detained”: [2015] EWCA Civ 445; [2016] QB 789.

The Supreme Court dismissed the Secretary of State’s appeal. The lead judgment was given by Lord Lloyd-Jones JSC.

At para. 29, he said that it is a fundamental principle of the common law that in enacting legislation Parliament is presumed not to intend to interfere with the liberty of the subject without making such an intention clear. For that proposition he cited the decision of the House of Lords in Khawaja and the decision of the Privy Council in Tan Te Lam.

Lord Lloyd-Jones acknowledged that in B (Algeria) the particular focus was not on a power of executive detention but on a power to grant bail. He continued:

In my view, what is of crucial importance to the reasoning of Lord Lloyd-Jones is what he said at para. 30:

Lord Lloyd-Jones concluded, at para. 31, that the word “detained” in those paragraphs refers to “lawfully authorised detention.” One of the reasons for that conclusion was, as he explained at para. 32, that a person who is in breach of the conditions of bail may be re-detained but this would not be possible in the absence of a subsisting power to detain.

I note that, at paras. 35-39, Lord Lloyd-Jones considered the decision of the House of Lords in Khadir but concluded that it provided no assistance to the Secretary of State, for the reasons given by Lord Dyson MR, at paras. 29-31 of his judgment in the Court of Appeal, which could be summarised as follows:

It is also important to note what was said by Lord Lloyd-Jones at para. 53:

The words which I have emphasised in that passage indicate that Lord Lloyd-Jones considered that section 61 of the 2016 Act might have the effect of making it clear that the power to grant bail was not dependent on the ability to detain a person lawfully. He contemplated that possibility, otherwise it is difficult to see why he would draw specific attention to section 61.

It is a little curious that the Supreme Court did not decide the appeal before it simply on the basis of the original version of section 61, since that provision had retrospective effect and was intended to reverse the decision of the Court of Appeal in that very case. At the hearing before us, we were informed by Mr Tam, who appeared for the Secretary of State in B (Algeria) too, that the parties wished to have the underlying issue of principle resolved and so he did not invite the Supreme Court to decide the case on the basis of section 61.

In the light of his conclusion in B (Algeria), Lord Lloyd-Jones did not consider it necessary to address the arguments based on Article 5 of the ECHR which, in his view, added nothing to the resolution of the issues before the Supreme Court in that case: see para. 56 of his judgment.

Section 67 of the 2002 Act provides:

The first thing to note about section 67 is that it has never been repealed and remains in force. Secondly, it applies to the construction of any legislation which falls within its terms. It is not confined to legislation which preceded the 2002 Act. In principle therefore it applies to the construction of the phrase “a person who is liable to detention” in the 2016 Act. Thirdly, it is clear from the language of subsection (2) that the factors which are mentioned in paras. (a), (b) and (c) are precisely the sort of factors which will often lead to the conclusion, in accordance with the Hardial Singh principles, that a person cannot be detained pending removal from the UK.

True it is that, in Khadir, the House of Lords considered that it had been unnecessary for Parliament to enact section 67 to achieve the result which it sought to achieve at that time: see para. 36 (Lord Brown). Nevertheless, the legislative history makes it clear that the purpose of section 67 was to reverse the effect of the first instance decision in Khadir by Crane J. It provided the basis for the decision in that case by the Court of Appeal, which was upheld by the House of Lords but for different and broader reasons.

In any event, in the present case, it seems to me that section 67 does have some relevance. Ms Harrison submitted that it is not applicable because the present context is one where there is conferred a power to detain a person and therefore section 67 by its own terms does not apply to the construction of such a provision. I disagree. The power to grant bail in the provision with which we are concerned, namely para. 1(2) of Sch. 10 to the 2016 Act, is not a provision which confers power to detain a person. To the contrary, the grant of bail is not the same as detaining a person. It is to grant that person their liberty.

At the hearing we were shown the version of para. 16 of Sch. 2 to the 1971 Act as it was in force from 28 July 2014 to 11 July 2016. It provided that (1) a person who may be required to submit to examination under para. 2 may be detained under the authority of an immigration officer pending his examination and pending a decision to give or refuse him leave to enter.

We were also shown the provisions of para. 21 of Sch. 2 as it was in force from 31 August 2006 to 14 January 2018. It provided that (1) a person liable to detention or detained under para. 16(1) (and other relevant provisions) may, under the written authority of an immigration officer, be temporarily admitted to the UK without being detained or be released from detention. Sub-para. (2) provided that, so long as a person was at large in the UK by virtue of this provision, he shall be subject to such restrictions as to residence, employment or occupation and reporting to the police or an immigration officer as may from time to time be notified to him in writing by an immigration officer. This was the concept of temporary admission which was considered by the House of Lords in Khadir.

It should be noted that that power did not depend on a person actually being in detention but also applied to a person “liable to detention”.

In contrast, para. 22, of which we were shown the version in force from 28 July 2014 to 14 January 2018, only applied to a person who was in fact detained, for example under para. 16(1). It provided that such a person may be released on bail in accordance with that paragraph. This was the provision which was considered by the Court of Appeal and Supreme Court in B (Algeria).

I have set out earlier the transitional provisions, in para. 13 of Sch. 10 to the 2016 Act.

It is clear from those transitional provisions that the concept of “immigration bail” was created at least in part to include the category of persons who had previously been subject to temporary admission or, like this Appellant, had been subject to a restriction order imposed by the Secretary of State. Those persons had not previously been granted bail in the traditional sense.

It is also clear, in my view, that Parliament did not intend to effect any radical change. In particular, it did not intend to prevent the Secretary of State from being able to impose conditions on a person such as this Appellant, who had no leave to be in this country and indeed was the subject of a deportation order, but was not in fact being detained. If there had been any intention to make such a radical change in the law which was applicable to persons such as this Appellant, one would have expected this to be foreshadowed in a document like a White Paper or some other policy proposal. There is no such policy background to the radical change in the law which Ms Harrison submits was achieved in 2016. I do not accept that submission.

To the contrary, in my view, it is clear that the legislation, when read as a whole, envisaged a seamless transition from the conditions which had previously been imposed on a person who had temporary admission or was the subject of a restriction order and the concept of “immigration bail”. I reach that interpretation on the basis of the 2016 Act alone, without reference to the Regulations which were enacted in late 2017. It is clear from para. 13(1) of Sch. 10 to the 2016 Act that Parliament envisaged that there would be transitional provisions of the kind that were in due course enacted in those Regulations.

As Bennion explains, at Principle 24.14(1), Explanatory Notes to an Act of Parliament may be used to understand the background to, and context of, the Act and the mischief at which it is aimed. It was held that Explanatory Notes are admissible for this purpose in R (Westminster City Council) v National Asylum Support Service [2002] UKHL 38; [2002] 1 WLR 2956. However, as Bennion goes on to explain, although Explanatory Notes may therefore be useful as an aid to construction, the courts will resist attempts to elevate those Notes to a status where they supplant the language of the legislation itself. This is not least because, as the Explanatory Notes themselves state, they are prepared by the Government in order to assist the reader in understanding the Act but do not form part of the Act and have not been endorsed by Parliament.

The Explanatory Notes to the 2016 Act, at para. 286, said that the original version of section 61(3) and (4):

In Bennion, Principle 24.5 is stated as follows:

The comment on that principle states that:

On this appeal Ms Harrison submitted that the Judge was wrong to give the weight that she did to the legislative history. In my view, that history is clearly of relevance to the question of interpretation we have to decide. It is not dispositive but I do not think that the Judge treated it as such.

Before us Ms Harrison complained that the Judge fell into the error of having regard to the purpose of Parliament in enacting the 2016 Act. She submitted that this was to have regard to the subjective intention of Parliament, whereas legislation must be construed objectively.

It is important to emphasise that having regard to the purpose of legislation is not to fall into the trap of having regard to the subjective intention of Parliament, even if that could be discerned. When the courts speak of the purposive approach to the interpretation of legislation, that is not a reference to the subjective intentions of Parliament or any individual member of it. What is meant is that the meaning of legislation is not to be arrived at simply on a linguistic basis.

In Uber BV v Aslam [2021] UKSC 5; [2021] ICR 657, at para. 70, Lord Leggatt JSC said that:

This is but one example of similar statements which have been made by judges at the highest level in recent times. Courts do not any longer (if indeed they ever did) adopt a purely linguistic approach to statutory interpretation. We adopt a purposive approach. But it is important to appreciate that a purposive approach is still an objective approach; it is not an attempt to divine the subjective intentions of Parliament or individual members of it. Even if that were possible, it would be irrelevant.

This was explained in R v Secretary of State for the Environment, Transport and Regions, ex p. Spath Holme Ltd [2001] 2 AC 349, at p. 396, by Lord Nicholls of Birkenhead:

The modern approach to statutory interpretation is to give the words used by Parliament their true meaning in the light of their context and their purpose. In my view, therefore, it is preferable to speak of the purpose of the legislation rather than the intention of Parliament, a phrase which is sometimes apt to mislead.

We heard interesting arguments based on old case law and treatises concerning the question whether the writ of habeas corpus may issue in circumstances where a person is no longer in detention but is on bail: for example, in Re Amand [1941] 2 KB 239, at 249, Viscount Caldecote CJ said that, although the applicant was now on bail, this made no difference and the court had to deal with the application as if he were still detained in custody. This line of authority was cited with approval by the Court of Appeal in B (Algeria), at para. 33, where Lord Dyson MR cited the decision in Re Amand and also the following dictum in Mitchell v Mitchinham (1823) 2 D & R 722, at p. 723:

In B (Algeria) in the Supreme Court, Lord Lloyd-Jones did not find those decisions of any great assistance: see para. 48 of his judgment.

In the present case, we were referred again to a large number of such authorities, including modern authorities on the availability of the writ of habeas corpus in extradition cases where a person is not in fact in detention but is on bail. In my view, these authorities are of no material assistance in resolving the question of statutory interpretation which we have to on the present appeal. I have no doubt that the writ of habeas corpus will still be available in the present context if it is appropriate on the facts of a particular case.

In R v Secretary of State for the Home Department, ex p. Simms [2000] 2 AC 115, at p. 131, Lord Hoffmann set out the principle of legality in a well-known passage as follows:

Lord Hoffmann went on, at p. 132, to observe that the principle of legality would be expressly enacted as a rule of construction in section 3 of the HRA, which at the time of the decision in Simms had been enacted but had not been brought into full force, as it subsequently was on 2 October 2000. I shall have more to say about this later.

It is important to keep in mind that the situation which Lord Hoffmann envisaged was one where Parliament can “if it chooses” legislate contrary to fundamental principles of human rights. In my view, that is not the present situation. There is no indication that, in enacting the 2016 Act, Parliament sought to legislate contrary to fundamental principles of human rights. Nor, in my view, does the interpretation given to the phrase “liable to detention” in the present context by the Judge give rise to any such violation.

In Pham v Secretary of State for the Home Department [2015] UKSC 19; [2015] 1 WLR 1591, at para. 118, Lord Reed JSC said that, in a number of cases concerned with important rights, such as the right of access to justice and legal professional privilege, the courts have interpreted statutory powers to interfere with those rights as being subject to implied limitations and have adopted an approach amounting in substance to a requirement of proportionality, although less formally structured than under the HRA. He gave as examples R v Secretary of State for the Home Department, ex p. Leech [1994] QB 198 and R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532. He continued, at para. 119, that one can infer from these cases that, where Parliament authorises significant interferences with important legal rights, the courts may interpret the legislation as requiring that any such interference should be no greater than is objectively established to be necessary to achieve the legitimate aim of the interference: in substance, a requirement of proportionality. Similar observations were made by Lord Reed in R (UNISON) v Lord Chancellor [2017] UKSC 51; [2020] AC 869, at paras. 80-89.

I readily acknowledge that the principle identified in those authorities means that, for example, the power to impose conditions on bail in the present context would have to be exercised in a way which does not unjustifiably interfere with fundamental rights, including the right to personal liberty. By way of example, if on the facts of a particular case the individual were able to show that a condition had been imposed which was so onerous and restrictive that it unjustifiably interfered with their private life, or in substance amounted to imprisonment by way of a curfew requirement, it might not be lawful. But that would depend on the particular facts of an individual case. It does not lead to the conclusion that the primary legislation which authorises the grant of bail is itself incompatible with fundamental rights.

On behalf of the Appellant, and in particular in the submissions made on behalf of BID by Ms Dubinsky, emphasis was placed on the approach which the courts have taken to ouster clauses. In R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22; [2020] AC 491, in the lead judgment for the majority by Lord Carnwath JSC, it was emphasised, at para. 107, that the interpretation advocated on behalf of the Secretary of State in that case (who was an Interested Party) treated the exercise as one of “ordinary statutory interpretation”, designed simply to discern “the policy intention” of Parliament but that this downgraded the critical importance of the common law presumption against ouster. Lord Carnwath continued, at para. 111, that judicial review can only be excluded by “the most clear and explicit words”, citing the decision of the Divisional Court in R (Cart) v Upper Tribunal [2009] EWHC 3052 (Admin); [2011] QB 120, at para. 31. He said:

In my view, no material assistance is to be derived in the present context from the very special line of authority concerning ouster clauses such as Privacy International.

I have noted that, in Simms, Lord Hoffmann predicted that the principle of legality would be expressly enacted as a rule of construction in section 3 of the HRA. That prediction did not turn out to be entirely accurate. In Ahmed v HM Treasury [2010] UKSC 2; [2010] 2 AC 534, at paras. 112-117, Lord Phillips PSC observed that subsequent decisions of the House of Lords on the HRA had made it clear that section 3 goes further than the principle of legality. At para. 117, he said:

Since the early days of the HRA it has been clear that the strong obligation of interpretation in section 3 only applies if otherwise the legislation would be incompatible with the Convention rights. In Poplar Housing and Regeneration Community Association Ltd v Donoghue [2001] EWCA Civ 595; [2002] QB 48, Lord Woolf CJ said, at para. 75, that courts should always first ascertain whether, in the absence of section 3, there would be any breach of the Convention. He said that, unless the legislation would otherwise be in breach of the Convention, section 3 can be ignored.

The correct approach was explained by Lord Reed JSC in S v L [2012] UKSC 30; 2013 SC 20, at paras. 15-17:

I would summarise the approach which needs to be adopted in the following order:

First, ascertain the ordinary meaning of legislation, having regard to all the usual aids to interpretation. This is not a purely linguistic exercise but seeks to give effect to the purpose of the legislation. The aids to interpretation include the presumption that Parliament does not intend to put the UK in breach of its international obligations, including those under the ECHR.

If – but only if – that ordinary interpretation would give rise to an incompatibility with the Convention rights, section 3 requires a different interpretation so far as possible. This is a strong form of interpretation, which is not the same as ordinary interpretation.

If, even then, it is not possible to give the legislation a meaning which is compatible with Convention rights, the court has a discretion to make a declaration of incompatibility under section 4 of the HRA (if it is one of the courts specified in that section).

In the present case I have reached the conclusion that, at stage (1) of that exercise, the ordinary interpretation of the 2016 Act is not incompatible with the Convention rights and therefore there is no warrant for going on to apply section 3 of the HRA.

Article 5 of the ECHR provides:

Ms Harrison relied on the decision of the European Commission of Human Rights in Zamir v United Kingdom (1983) 5 EHRR 242. At para. 109, the Commission noted that Article 5(4) of the ECHR requires a remedy that entitles a detained person to a judicial ruling on the lawfulness of his detention. However, the Commission said, this right must be seen as independent of the possibility of applying to a court for release on bail.

Ms Harrison also relied on the decision of the European Court of Human Rights in Ismail v United Kingdom (2014) 58 EHRR SE6. Although in that case the application was held to be inadmissible, she draws attention to the fact that, at para. 16, the Government submitted that Article 5(4) was not applicable, since the bail application was not a procedure under domestic law to challenge the lawfulness of immigration detention. They emphasised that, under domestic law as interpreted by the courts, a decision to release a person on bail, subject to conditions designed to ensure his future attendance, presupposed the legality to detain; and they pointed out that the applicant had accepted that his application for bail could not and did not determine the legality of his detention. The Government went on to submit that, under domestic law, the applicant could have challenged the lawfulness of his detention through either habeas corpus or judicial review proceedings, and the existence of those remedies was sufficient to meet the obligation under Article 5(4). At para. 28, the Court decided that Article 5(4) was not engaged on the facts of the case because the applicant had chosen to apply for bail. He did not seek to challenge the lawfulness of his detention and, had he chosen to do so, he could have issued a writ of habeas corpus or made an application for judicial review. At para. 30, the Court observed that the applicant’s immigration bail hearing had not involved any examination of the lawfulness of his detention, either under national law or in terms of the Convention.

It does not seem to me that those decisions are germane to the issue of law which we have to decide in this appeal. They concerned different issues and I did not understand Mr Tam to disagree with what is said in those cases.

On behalf of the Appellant Ms Harrison emphasised that there could be circumstances in which a condition imposed on bail is so onerous, for example as to a curfew requirement, that it will constitute an imprisonment at common law and/or a deprivation of liberty for the purpose of Article 5 of the ECHR: see R (Jalloh) v Secretary of State for the Home Department [2020] UKSC 4; [2021] AC 262. That may be so but it has not been suggested that the bail conditions imposed on this Appellant amount to either an imprisonment or a deprivation of liberty.

Further, I am not persuaded by the arguments for the Appellant based on the fact that there are powers of detention contained for breach of bail conditions in para. 10(9) and (12) of Sch. 10 to the 2016 Act. I note again that this Appellant has not been detained under those powers. Indeed it was emphasised on his behalf that he has complied with his conditions at all times since 2011.

I see force in Mr Tam’s argument that there is no reason in principle why a separate power of short-term detention following arrest for failure to comply with bail conditions should not apply even in circumstances where the powers of immigration detention mentioned in para. 1(1) of Sch. 10 cannot lawfully be exercised. As for the power to detain in para. 10(12)(a) of Sch. 10, I agree with the Judge that this is dependent on the underlying powers of immigration detention. In circumstances where those powers cannot lawfully be exercised, the consequence would be that the person must be bailed again but of course this could be on more stringent conditions under sub-para. (b).

It does not appear to me that Article 5, still less Article 5(4), of the ECHR prevents the interpretation of Sch. 10 which is otherwise the ordinary and correct one. As I have already said, if the facts of a particular case justify it, an application can still be made under the HRA arguing that a person’s Convention rights have been breached, including an argument that his detention is unlawful. An appropriate remedy can therefore be granted in the domestic courts, thus ensuring compliance with Article 5(4).

I turn to Ground 2 in this appeal, which is advanced as an alternative to Ground 1. Under Ground 2, Ms Harrison submits that there must be “some prospect” of removal for the power to detain (and therefore the power to grant bail) to exist at all: see paras. 32-33 of Lord Brown’s opinion in Khadir. In the present case, she submits, there was at the time of the hearing in the High Court, and certainly is now, no prospect of the Appellant’s removal at all.

For the Respondent Mr Tam submits that the Appellant does not have permission from the Judge to advance Ground 2 and that permission should not be granted by this Court. If permission is granted by this Court, he invites us to reject Ground 2 because the Appellant failed to afford the Judge an opportunity to consider setting out fuller reasons in relation to this argument, in accordance with English v Emery Reimbold & Strick Ltd (Practice Note) [2002] EWCA Civ 605; [2002] 1 WLR 2409. In any event, he submits, the Judge did in substance address the point: she repeatedly cited Khadir, including, at para. 27, the precise proposition now relied on, that, for the detention power to exist, there must be “some prospect” of removal. Given this reference, Mr Tam submits, the Judge must have concluded that there was “some prospect” of the Appellant’s removal.

I do not accept that the Appellant has permission to advance Ground 2 in this appeal. First, the Judge was not presented with this ground anywhere in the written application for permission to appeal dated 15 July 2020. Although she did not have formulated grounds of appeal as such before her, she did have that written application, which made detailed submissions over three pages but they focussed on what has become Ground 1 and, in the alternative, at para. 8, raised what is now Ground 3. Before us reference was made to where this ground of challenge had been advanced in the claim for judicial review: we were shown the Statement of Facts and Grounds and even the transcript of the oral hearing before the Judge. This is neither here nor there. The scope of the permission to appeal to this Court is not governed by such matters. It is governed by the grant of permission, either by this Court or by the trial judge. In this case permission to appeal was granted by the trial judge. She clearly did so on the basis of the written application before her, nothing more.

Secondly, in granting permission in writing on 17 July 2020, the Judge said that she was doing so on the “other compelling reason” limb of the test for an appeal to this Court: see CPR 52.6(1)(b). She noted that the application raised an issue of statutory construction and that it potentially affected around 90,000 people. She clearly had in mind Ground 1. On any view Ground 2 concerns only the facts of this particular case and does not raise any issue of principle, let alone provide a compelling reason why this Court should hear an appeal. It is clear therefore that it was not within the scope of the permission granted by the Judge.

I turn to consider the application made to this Court for permission to include Ground 2 in the grounds of appeal. I would refuse that application. This is because it has no real prospect of success and there is no other compelling reason for this Court to hear it.

The reason why Ground 2 has no real prospect of success is that the law requires there to be only “some prospect” of removal of the Appellant at some point. Mr Tam accepts on behalf of the Respondent that, if there were truly no prospect of removal, then, even in accordance with the House of Lords decision in Khadir, there is no legal power to detain at all. Once that point is reached, if it is, then he accepts that there is also no power to grant bail. But, on the material which was before the trial Judge, that was not the Respondent’s position and the Judge must have agreed. It is important to recall that an appeal to this Court is from the decision of the trial court and the principal question for us is whether that decision was “wrong”: see CPR 52.21(3)(a). An appeal to this Court is usually by way of “review” and not a “re-hearing”: CPR 52.21(1). Further, this Court will usually determine that question on the basis of the material which was before the first instance court: CPR 52.21(2).

True it is that there is further material which the Appellant seeks to place before this Court, in particular the decision of the FTT, dated 25 June 2021, which postdates the decision of the High Court in this case. The FTT decided that, as a matter of fact, the Appellant is not a Ghanaian national. It allowed the appeal on grounds which are unrelated to the present appeal. The Secretary of State has been granted permission to appeal to the Upper Tribunal but we were informed that she does not seek to challenge the factual finding as to the Appellant’s nationality. Mr Tam did not object to our being informed of these developments but he does object to any suggestion that developments which have taken place since the decision of the High Court are relevant to the issues we have to decide.

Even after the FTT’s determination, the Respondent disputes whether there is some prospect of removal in due course. In those circumstances it seems to me to be inappropriate for this Court to determine what is a disputed question of fact. Rather this Court should do what it would normally do, which is to decide whether the decision of the High Court was wrong on the basis of the material which was before it. I have reached the conclusion that it was not wrong on that material, nor is it arguable that it was. Accordingly, I would refuse permission to appeal on Ground 2. In those circumstances I would also refuse the application to adduce fresh evidence.

Ms Harrison contends that the Judge erred in rejecting BID’s alternative argument that there are implied limits on the exercise of the power to maintain conditional bail. She adopted the written and oral submissions made by Ms Dubinsky, who took the lead on Ground 3.

Ms Dubinsky makes five key submissions:

First, the Hardial Singh limits are a safeguard imposed by the common law upon executive detention or imprisonment relating to expulsion or deportation.

Secondly, those limits flow from the strict application of the Padfield principle (Padfield v Minister of Agriculture, Fisheries and Food [1968] AC 997) and the need to act reasonably in the Wednesbury sense: Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223.

Thirdly, the Hardial Singh limits can be excluded only by express words or necessary implication.

Fourthly, the immigration bail powers are analogous to the powers of administrative detention to which the Hardial Singh limits apply.

Fifthly and in consequence, the lawful exercise of the power is subject to the implied limitations analogous to those applicable to powers of administrative detention (the Hardial Singh limits).

The principles which were first set out by Woolf J in Hardial Singh, at p. 706, have been approved in many cases subsequently, including by the Supreme Court in R (Lumba) v Secretary of State for the Home Department [2011] UKSC 12; [2012] 1 AC 245, at para. 30, where Lord Dyson JSC said that those principles reflect the basic public law duties to act consistently with the statutory purpose (Padfield) and reasonably in the Wednesbury sense.

I do not accept the submission that the Hardial Singh principles must be imported into the context of the grant of bail. The whole point about the Hardial Singh principles is that they were developed in the context of administrative detention. A person on bail is not in detention. He is at liberty, although there may be conditions attached to his bail (and, in the present context, there must be at least one condition): see the decision of the Privy Council in Syed Mahamad Yusuf-ud-Din v Secretary of State for India in Council (1903) LR 30 Ind App 154. In that case the plaintiff had been released from imprisonment on bail. He sued for false imprisonment but his action was barred by a limitation period if time ran from the date when he had been released from prison on bail. The Privy Council held that it was time-barred. At p. 158, Lord Macnaghten said that:

Similarly, in Stellato v Ministry of Justice [2010] EWCA Civ 1435; [2011] QB 856, at para. 23, Stanley Burnton LJ said:

In any event, the Respondent accepts that the normal principles which govern the exercise of a discretionary power still apply in this context. For example, the power to impose conditions on bail must be exercised for a proper statutory purpose (Padfield); and the power must be exercised rationally. Depending on the facts of a particular case, a person may be able to argue that the conditions imposed are so onerous and unreasonable that they should be held to be unlawful on the facts. None of that, however, means that the general principles in Hardial Singh should be imported into this context.

What the Appellant (and BID) seek in advancing Ground 3 is to impose the temporal limitation in the Hardial Singh principles. Although this was not how it was put by BID, the argument must be that, because a person cannot be deported within a reasonable time, they cannot be lawfully detained and therefore no bail can be granted either. In my view, the conclusion simply does not follow from the premise. There is all the world of difference between concluding that a person can no longer be lawfully detained because they cannot be deported within a reasonable time; and the conclusion that they cannot be the subject of bail while they are not detained. The two things are clearly different. There is no logical connection between them.

For the reasons I have given I would dismiss this appeal.

I would like to return to the procedural difficulties which this Court has faced in dealing with Ground 2. I would remind practitioners and lower courts and tribunals of the guidance which was given by this Court in McDonald v Rose [2019] EWCA Civ 4; [2019] 1 WLR 2828, at para. 21 (Underhill LJ); and in Municipio de Mariana v BHP Group plc [2021] EWCA Civ 1156, at paras. 113-114 (Sir Geoffrey Vos MR). I will not set out that guidance again here but it is essential reading and needs to be read in full.

That guidance reinforces what is said in Practice Direction 52C (Appeals to the Court of Appeal), at para. 5.

As both the Practice Direction and this Court have made clear, the grounds of appeal are not the same thing as submissions, which should be set out in a skeleton argument in support of those grounds. Furthermore, it is important that the grounds are clearly and concisely formulated so that everyone concerned, including the court which is asked to grant permission and this Court when it has to consider an appeal, knows exactly what is within the scope of the appeal. In those cases where this Court is asked to grant permission to appeal, there is usually no difficulty, because by then the applicant has formulated the grounds of appeal. Where, however, it is the lower court or tribunal (as in the present case) which grants permission to appeal, it is essential that there should be properly formulated grounds of appeal, and not simply submissions, before it. It is only in that way that the judge who grants permission to appeal can know precisely what it is for which permission is being granted.

I agree.

I also agree.