U3 v Secretary of State for the Home Department

The question of law at the heart of this appeal is what functions the Special Immigration Appeals Commission (‘SIAC’) has when it hears an appeal against a decision of the Secretary of State to deprive a British citizen of her nationality on the ground that she is a risk to national security. The answer depends on the meaning of section 2B of the Special Immigration Appeals Commission Act (‘the 1997 Act’) read in its context, and on a correct understanding of two decisions, in particular: Secretary of State for the Home Department v Rehman [2001] UKHL 29; [2003] 1 AC 153 (‘Rehman’) and Begum v Secretary of State for the Home Department [2021] UKSC 7; [2022] AC 765 (‘Begum’). Rehman did not concern section 2B. As I will explain, Begum did. The appeal also concerns SIAC’s powers under section 2 of the 1997 Act when it considers a human rights appeal against a later refusal of entry clearance.

After a hearing lasting a week between 29 November and 3 December 2021, SIAC dismissed the appeals of the Appellant (‘U3’) against decisions of the Secretary of State to deprive her of her British citizenship (‘decision 1’) and to refuse her entry clearance (‘decision 2’). I will refer to those appeals as ‘appeal 1’ and ‘appeal 2’ respectively. On this appeal, U3 was represented by Ms Harrison KC, Mr Grieves KC, Mr Toal, and Mr Clark. The Secretary of State was represented by Mr Sheldon KC and Ms Thelen. The intervener, JUSTICE, was represented by Mr Hickman KC, Mr Molyneaux and Mr Fakhoury. I thank all counsel for their written and oral submissions. There was also a CLOSED hearing, which was attended by the Special Advocates and by counsel for the Secretary of State. None of the issues which we were asked to consider in that hearing has any bearing on the reasoning in this judgment.

In outline, U3’s counsel argued that SIAC had taken too narrow a view of its powers, by limiting itself to asking, in short, whether, in making decisions 1 and 2 on the grounds of national security, the Secretary of State had made a public law error. U3 also argued that SIAC should have made various findings of fact and that had it done so, it would, or might, have allowed U3’s appeal. That approach was broadly supported by JUSTICE. The Secretary of State, by contrast, defended SIAC’S approach.

Paragraph references are to SIAC’s OPEN judgment, unless I am referring to an authority.

The issues on this appeal are issues of law. It is convenient therefore, first, to summarise the legal position, so as to provide a context for the arguments that SIAC misdirected itself. I will then summarise SIAC’s reasoning. I will say more about the submissions in the course of the last part of this judgment, in which I will analyse and decide the issues. In summary, I have reached three broad conclusions.

The 1997 Act distinguishes between the apparently unqualified right of appeal conferred by section 2B (‘a person may appeal to SIAC…’) and rights to apply for statutory review conferred by sections 2C-2E. There are no express limits on that right of appeal. By contrast, when, under sections 2C-2E, SIAC considers whether an impugned decision should be ‘set aside’, it is required to ‘apply the principles which would be applied on an application for judicial review’.

General Medical Council v Michalak [2017] UKSC 71; [2017] 1 WLR explains some of the differences between an appeal and an application for judicial review. The claimant was a doctor. She complained to an Employment Tribunal (‘ET’) that the appellant and its officers had discriminated against her on the grounds of her race in the course of its fitness to practise procedure. There was a preliminary issue about whether the ET had jurisdiction over her claim under section 120 of the Equality Act 2010. The answer depended on whether the acts she complained of were subject by virtue of an enactment to ‘an appeal or proceedings in the nature of an appeal’. The ET held that it had jurisdiction, because while steps taken in the course of the GMC’s complaints procedure were amenable to judicial review, the appellant did not have a right of appeal against them, and judicial review did not amount to ‘proceedings in the nature of an appeal’. The Employment Appeal Tribunal allowed the GMC’s appeal, but this court reversed that decision. The Supreme Court upheld the decision of this court.

Lord Kerr gave a judgment with which the other members of the Supreme Court agreed. In paragraph 20, he said that ‘an “appeal” (if it is not qualified by any words of restriction) is a procedure which entails a review of an original decision in all its aspects. Thus, an appeal body or court may examine the basis on which the original decision was made, assess the merits of the conclusions of the body or court from which the appeal was taken and, if it disagrees with those conclusions, substitute its own. Judicial review, by contrast, is, par excellence, a proceeding in which the legality of or the procedure by which a decision was reached is challenged…in the human rights field, the proportionality of a decision may call for examination in a judicial review proceeding…But an inquiry into the proportionality of a decision should not be confused with a full merits review’. He cited paragraph 272 of R (Keyu) v Secretary of State for Foreign and Commonwealth Affairs [2015] UKSC 69; [2016] AC 1355.

In his view, judicial review ‘even on the basis of proportionality, cannot partake of the nature of an appeal’ (paragraph 21). A complaint of discrimination illustrated the point. A tribunal considering such a complaint had to ‘conduct an open-ended inquiry into that issue’. Whether there was discrimination in fact depended on the judgment of the tribunal. It could not be answered by looking at the reasons of the alleged discriminator, and deciding whether or not what he did was within the range of reasonable responses. ‘The latter approach is the classic judicial review investigation’. On a judicial review, the High Court ‘merely either declares the decision to be unlawful or quashes it. It does not substitute its own decision for that of the decision-maker. In that sense, a claim for judicial review does not allow the decision of the GMC to be reversed. It would be anomalous for an appeal or proceedings in the nature of an appeal to operate under those constraints. An appeal in a discrimination case must confront directly the question whether discrimination has taken place, not whether the GMC had taken a decision which was legally open to it’ (paragraph 22, original emphasis). Lord Kerr’s conclusion was that judicial review in the context of that case was not in the nature of an appeal.

The appeal to this court from SIAC is an appeal on point of law. It is convenient now to summarise the approach which this court takes to criticisms of a judge’s treatment of the evidence in appeals when the test is whether or not the first instance decision is ‘wrong’. That approach is well established. I emphasise that that approach does not apply to this appeal. The approach to such criticisms in this appeal must be even more stringent. In other words, if a challenge to a judge’s treatment of the evidence could not succeed on the first type of appeal, it certainly could not succeed on an appeal on a point of law. Two decisions are relevant: Fage UK Limited v Chobani [2014] EWCA Civ 5; [2014] FSR 29, paragraph 114, and Volpi v Volpi [2022] EWCA Civ 464; [2022] 4 WLR 48, paragraph 2. The last four points in paragraph 2 of the second decision are especially pertinent here.

Part II of the Immigration Act 1971 (‘the 1971 Act’) continued the Immigration Appeal Tribunal and adjudicators, which were provided for by the Immigration Appeals Act 1969 (section 12(1)). Subject to stated exceptions, Part II conferred rights of appeal against certain immigration decisions. The provisions of Part II required an adjudicator to dismiss an appeal if he was satisfied of some matters (see, for example, sections 13(4) and 16(4)). Section 19 provided that, subject to sections 13(4) and 16(4), and subject to any restriction on the grounds of appeal, an adjudicator was required to allow an appeal if he considered that ‘the decision or action against which the appeal is brought is not in accordance with the law or with any immigration rules applicable to the case’, or ‘where the decision or action involved the exercise of a discretion by the Secretary of State or an officer, that discretion should have been exercised differently’. In any other case, the adjudicator was required to dismiss the appeal. For the purposes of section 13(1)(a) the adjudicator was given an express power to review any determination of a question of fact on which the decision or action was based (section 13(2)). Section 13 gave the adjudicator power, if he allowed an appeal, to give such directions to the Secretary of State as he considered requisite for giving effect to his determination, and power to make recommendations. With a stated exception, the Secretary of State had to comply with any direction (section 19(3)).

Part II of the 1971 Act was repealed by the Immigration and Asylum Act 1999 (‘the 1999 Act’). Part IV of the 1999 Act dealt with appeals. Part III of Schedule 4 was enacted by section 58(4). Paragraph 21 of Schedule 4, headed ‘Determination of appeals’ was in similar terms to section 19 of the 1971 Act. Paragraph 22 of Schedule 4 governed the powers of the Immigration Appeal Tribunal (‘the IAT’) on an appeal from an adjudicator. Any party to such an appeal might, ‘if dissatisfied with his determination, appeal to [the IAT]’ (paragraph 22(1)). On such an appeal, the IAT might ‘affirm the determination or make any other determination which the Adjudicator could have made’ (paragraph 22(2)).

In Subesh v Secretary of State for the Home Department [2004] EWCA Civ 56; [2004] INLR 417 it was common ground that paragraph 22 enabled the IAT to set aside a factual decision of an adjudicator (paragraphs 30 and 40). In paragraph 25, Laws LJ said that paragraph 22 conferred ‘an unqualified right of appeal to the IAT, that is a right of appeal not limited by reference to issues of any particular kind such as matters of law only (in contrast to the right of appeal to this court created by paragraph 23(1))’. The appellant in Subesh argued that the IAT could only overturn a factual decision of an adjudicator if his conclusions were ‘plainly wrong or unsustainable’ or outside the ‘generous ambit within which a reasonable disagreement is possible’ (paragraph 27).

In paragraph 37, Laws LJ cited paragraph 13 of the judgment of Hale LJ (as she then was) in Indrakumar v Secretary of State for the Home Department [2003] EWCA Civ 1677. She said that the IAT was no different from this court, or any other court with jurisdiction to hear appeals on fact and law. It could only interfere if there was an error, that is, if, on analysis, the decision of the adjudicator was wrong. It was not enough that the IAT might have reached a different conclusion. The application of the test would vary, depending on the nature of the evidence on which the finding of fact was based. Findings of fact based on oral evidence and an assessment of credibility could only rarely be overturned on appeal. Findings based on documentary evidence could be overturned more readily, unless they were linked to an assessment of credibility. The IAT was at least as well placed as an adjudicator to make findings based on conditions in-country. The IAT would be entitled to draw its own inferences from such evidence if it detected an error by the adjudicator.

Laws LJ said, in paragraph 40, that the IAT was not limited to a Wednesbury approach. He accepted, in paragraph 41, that the IAT must be slow to impose its view in relation to a finding which depended on an assessment of oral evidence. That had ‘nothing to [do] with the reach of the appellate court’s jurisdiction. It merely recognises the pragmatic limitations to which the appeal court, not having heard the evidence, is subject’.

Finality is an important factor. The appeal was not a ‘re-run second time around of the first instance trial’. The appellant did not ‘approach the appeal court as if there had been no first decision, so to speak, he and his opponent were to meet on virgin territory. The first instance decision is taken to be correct until the contrary is shown…The true distinction is between the case where the appeal court might prefer a different view (perhaps on marginal grounds) and one where it concludes that the process of reasoning, and the application of the relevant law, require it to adopt a different view. The burden which the appellant assumes is to show that the case falls within this latter category’ (paragraph 44) (original emphasis).

To detect an error and to substitute inferences were not part of a two-stage process; the error might be that the adjudicator had drawn the wrong inferences (paragraph 45). This court’s reasoning was not ‘merely …an exercise in the construction of paragraph 22’. The reasoning was based not on the statute, but on the principle of finality. ‘It is what might nowadays be called a default position, defeasible in any particular case by a statutory provision inconsistent with it’. Laws LJ gave the example of an appeal from the magistrates’ court to the Crown Court as ‘in effect… a new first instance hearing’. Evidence was called again. It might differ from the evidence in the magistrates’ court. That process was different from the process which he had described, but did not undermine it. The regime in the Crown Court ‘merely shows the working of a particular statutory regime as it has been interpreted. Cases where statute prescribes a specially restricted right of appeal will equally involve a departure from the default position’ (paragraph 48).

Part IV of the 1999 Act was repealed by the Nationality, Immigration and Asylum Act 2002 (‘the 2002 Act’). From its commencement, Part V of the 2002 Act governed immigration appeals. Its initial scope was substantially curtailed by the Immigration Act 2014. The current provisions limit immigration appeals to appeals against decisions which refuse a protection or a human rights claim.

Section 40(2) of the British Nationality Act 1981 (‘the BNA’) gives the Secretary of State power, by order, to deprive a person of her citizenship status (as defined in section 40(1) of the BNA) if the Secretary of State is ‘satisfied that deprivation is conducive to the public good’. By section 40(3), the Secretary of State may also deprive a person of citizenship status which results from her registration or naturalisation if ‘the Secretary of State is satisfied that’ the registration or naturalisation was obtained by means of fraud, false representation or concealment of a material fact. Subject to the exception described in section 40(4A), section 40(4) prevents the Secretary of State from making a deprivation order if the Secretary of State ‘is satisfied that the order would make a person stateless’. Before making an order under section 40, the Secretary of State must give the person concerned written notice complying with the three requirements in section 40(5). The third requirement is that the person must be told of her right of appeal under section 40A(1) of the BNA, or under section 2B of the Special Immigration Appeals Commission Act (‘the 1997 Act’). Section 40(6) makes provision similar to that made by section 40(3) in cases where a person obtained her citizenship by operation of law, as further described in section 40(6).

When it was first enacted, section 40 did not provide for a right of appeal. Instead, a person against whom the Secretary of State proposed to make an order could ‘apply in the prescribed manner for an inquiry’, and if she did, the Secretary of State was obliged to refer the case to a committee of inquiry. The Secretary of State also had a general power to refer any case to such a committee (section 40(7)).

The 2002 Act substituted for section 40 of the BNA a revised section 40 and a new section 40A. Section 40A(1) created a right of appeal to an adjudicator appointed under section 81 of the 2002 Act. If, however, the Secretary of State certified that the deprivation decision was taken wholly or partly in reliance on information which, in the opinion of the Secretary of State, should not be made public, section 40A(1) did not apply (section 40A(2)).

When section 40A(1) was first enacted, it provided that a person who was given a notice under section 40(5) could appeal to an adjudicator. There were no express limits on that right. Between 4 April 2005 and 15 October 2014, section 40A(3)(a) applied section 87 of the 2002 Act to appeals against deprivation. Section 87 permitted the appellate body to make a direction to the Secretary of State if the appeal succeeded. Such a direction could provide that the impugned decision be treated as if it had never had effect. Section 40A(3) also applied other provisions of the 2002 Act to such appeals.

The current version of section 40A(3) ((c), (d) and (e) only) provides that only three provisions of the 2002 Act are to apply to appeals under section 40. They are section 106 (‘rules’), section 107 (‘practice directions’) and section 108 (‘forged documents’). Section 40A(3)(a), the only provision which applied to section 2B appeals, was repealed with effect from 20 October 2014.

In immigration cases in which the Secretary of State made a certificate similar to a certificate under section 40A(2), and in deprivation cases, the 1997 Act created a right of appeal to SIAC. As originally enacted, section 2(1) of the 1997 Act gave a person a right of appeal to SIAC against any matter in relation to which he would have been able to appeal to an adjudicator under various provisions of the 1971 Act, of the Immigration (European Economic Area) Order 1994, and of various provisions of the Asylum and Immigration Appeals Act 1993. Section 2(2) created a right of appeal against some refusals of entry clearance. As originally enacted, section 2 did not confer a right of appeal in deprivation cases, nor does it do so now.

Section 4 of the 1997 Act was repealed from 1 April 2003. It was headed ‘Determination of appeals’. Section 4(1)(a) required SIAC to allow an appeal if it considered that ‘the decision or action against which the appeal is brought was not in accordance with the law or with any immigration rules applicable to the case’ or ‘where the decision involved the exercise of a discretion by the Secretary of State or an officer, that the discretion should have been exercised differently’. Section 4(1)(b) required SIAC to dismiss the appeal in any other case. If SIAC allowed the appeal, it was required to ‘give such directions for giving effect to the determination as it thinks requisite’. It could also make ‘recommendations with respect to any other action which it considers should be taken in the case under [the 1971 Act]’. The Secretary of State, and any officer to whom directions were given, were required to comply with them (section 4(2)). Section 4 therefore echoed the parallel provisions about immigration appeals in the 1971 Act (and later, in the 1999 Act).

Section 2B was inserted in the 1997 Act with effect from 1 April 2003. It still governs appeals to SIAC against deprivation decisions. It gave and gives a person a right of appeal to SIAC against a decision to make an order under section 40A of the BNA if she was not entitled to appeal under section 40A(1) because of a certificate under section 40A(2). It provided and provides ‘A person may appeal if…’. Section 2B was amended with effect from 4 April 2005 so as to incorporate section 40A(3)(a) of the BNA by reference. That reference was repealed in 2014 by paragraph 26(3) of Schedule 9 to the Immigration Act 2014, but article 2(e) of the Immigration Act 2014 (Commencement No 3 Transitional and Savings Provisions) Order 2014/2771 has postponed that repeal. Nevertheless, the direct repeal of section 40A(3)(a) of the BNA by paragraph 25 of Schedule 9 to the 2014 Act (which has, as article 2(e) of the 2014 Order shows, been brought into force), empties the reference in section 2B to section 40A(3)(a) of any effect.

Section 5(1) deals with procedure. It gives the Lord Chancellor power to make rules ‘(a) for regulating the right of appeal …conferred by section …2B above’ and ‘(b) for prescribing the practice and procedure to be followed on or in connection with appeals under section…2B above including the mode and burden of proof and admissibility of evidence on such appeals’. Section 5(6) requires the Lord Chancellor, when making such rules, to ‘have regard, in particular, to – (a) the need to ensure that decisions which are the subject of appeals are properly reviewed…’.

The upshot, therefore, is that appeals to SIAC against deprivation decisions have, ever since that right was first conferred by section 2B on 1 April 2003, been governed by section 2B of the 1997 Act. Section 4 of the 1997 Act was repealed at the same time as section 2B was brought into force. Section 4 has never, therefore, applied to appeals to SIAC against deprivation decisions, although, until its repeal in 2003, it applied to appeals to SIAC under section 2 of the 1997 Act.

The parties submitted an agreed note, after the hearing, about the effect of a decision by SIAC to allow an appeal under section 2B. In E3, N3 and ZA v Secretary of State for the Home Department [2023] EWCA Civ 26, the Secretary of State had withdrawn three orders depriving appellants of their nationality because, after litigation in SIAC, she had accepted that the effect of those orders in those cases had been to make the appellants stateless. The issue (paragraph 2) was the effect of the withdrawal of those orders. Did the withdrawal of the orders mean that they should be treated as never having been made, with the result that the appellants retained their nationality throughout, or did the withdrawal only take effect from the date when the order was withdrawn?

Lewis LJ noted that no statutory provision dealt with the powers of SIAC on an appeal against a deprivation decision. It was accepted that where SIAC held on an appeal that the effect of the order was to make a person stateless, the Secretary of State could withdraw the decision and order in order to give effect to SIAC’s decision to allow the appeal. It was also common ground that the Secretary of State could anticipate such a decision by SIAC and withdraw a decision and order before an appeal was heard (paragraph 20). He also noted until its repeal in 2004, section 40A(6) of the BNA gave an appeal under section 40A a suspensive effect, and the (now repealed) power to give a direction to the Secretary of State (originally conferred by the cross-reference in section 40A(3) to section 87 of the 2002 Act) (see paragraph 22 and 23, above). Section 40(4) has been described as a condition precedent to the making of an order, but it ‘is more accurate to describe it as limitation on the circumstances in which the power may be exercised’ (paragraph 30). The limitation was expressed not by reference to a state of affairs, but to a state of mind (paragraphs 31 and 40).

He added that ‘On an appeal relating to section 40(4)… SIAC will have to determine whether the making of the deprivation order did render the person stateless. Statelessness will be a question of fact which will depend on the evidence available including expert evidence… SIAC will need to determine whether or not the order does on the facts have the consequence of rendering the person stateless. (The position is different in relation to an appeal against a decision that deprivation is conducive to the public good under section 40(2) where SIAC is concerned with reviewing the exercise of a discretion on public law principles [see Begum, paragraphs 66-71])’ (paragraph 32).

As there were no express provisions governing SIAC’s powers on an appeal it would simply allow the appeal if the effect of the order was to make the appellants stateless. SIAC did not have power to quash the decision or the order. The Secretary of State would need to take further steps to give effect to SIAC’s finding. The Secretary of State would have to withdraw the decision and the order, thus ‘removing the legal measure that deprives the individual of his status as a British citizen’ (paragraph 33). SIAC’s judgment would not amount to a finding that the decision was unlawful. It was not a finding that the Secretary of State had no power to make the decision so that the order was a nullity. Nor was it a decision that the Secretary of State could not have been satisfied, when she made the decision that the effect of an order would not be to make the appellants stateless. The appellants had not made such a challenge. Any finding would have been a finding that, as a matter of fact, the order did make the appellants stateless. The lawfulness and validity of the decision and of the order were separate from SIAC’s findings on statelessness. The withdrawal had effect from the date when the order was withdrawn by the Secretary of State (paragraph 34).

The power to give a direction which SIAC had previously had supported that view, although it was not decisive. The existence of that power was inconsistent with the suggestion that a successful appeal under section 40A automatically meant that the deprivation order was a nullity. It suggested that Parliament thought in 2004 that other provisions in section 40 and section 40A did not have that effect (paragraph 37).

Rehman concerned SIAC’s powers on an appeal under section 2(1)(c) of the 1997 Act against a decision by the Secretary of State to make a deportation order against the respondent on the ground that to do so would be conducive to the public good in the interests of national security because of his association with Islamic terrorist groups. SIAC allowed his appeal.

The judgment of this court was given by Lord Woolf MR. He held that SIAC was right to see its task as determining questions of fact and law. The fact that it had power to review the exercise of the Secretary of State’s discretion ‘inevitably leads to this conclusion. Without statutory intervention, this is not a role which a court readily adopts’ (paragraph 42). The fact that SIAC could decide issues of fact for itself did not determine what standard of proof it should apply. SIAC rightly said that the assessment of the executive was entitled to great weight. SIAC had then found that five specific allegations against the respondent had not been proved, applying a ‘high civil balance of probabilities’. ‘On one approach’ the standard applied by SIAC was appropriate; SIAC was entitled to say that in so far as the Secretary of State was relying on ‘specific allegations of serious misconduct’ those ‘had not been proved’ (paragraph 43).

The Secretary of State, however, was entitled to make a deportation order on the basis not only that the respondent had in fact endangered national security, but on the basis that he was a danger to national security. It was therefore necessary not only to look at the question whether particular allegations had been proved, but also to ‘examine the case as a whole against the individual and then to ask whether on a global approach that individual is a danger to national security, taking into account the executive’s policy with regard to national security’. The cumulative effect might show that the person was a danger to national security even if no particular allegation could be proved against him to that high standard. He had not been charged with a specific criminal offence. The danger he posed to national security had to be balanced against his personal interests (paragraph 44). The Court of Appeal allowed the Secretary of State’s appeal and remitted the case to SIAC.

The respondent appealed. All the members of the Appellate Committee agreed with the reasoning of the Court of Appeal on this issue. In paragraph 22, Lord Slynn said that if specific acts which had already occurred were relied on, fairness required that they be proved to the civil standard of proof. He added that that was not the whole exercise, because there was a range of other information to which the Secretary of State was entitled to have regard. He was also entitled to have regard to ‘precautionary and preventative principles rather than to wait until directly harmful activities have taken place’. In doing so, ‘he is not merely finding facts, but forming an executive judgment or assessment. There must be material on which proportionately and reasonably he can conclude that there is a real possibility of activities harmful to national security but he does not have to be satisfied, nor on appeal to show, that all the material before him is proved, and his conclusion is justified, to a “high civil degree of probability”. Establishing a degree of probability does not seem relevant to the reaching of a conclusion on whether there should be a deportation for the public good’.

This was not to confuse proof of facts with the exercise of a discretion ‘…specific acts must be proved, and an assessment made of the whole picture and then the discretion exercised as to whether there should be a decision to deport…’ (paragraph 23). SIAC had ‘powers of review both of fact and of the exercise of discretion, [but it] must give due weight to the assessment and conclusions of the Secretary of State…’ (paragraph 26).

Lord Steyn agreed with Lord Slynn. In paragraph 29, he specifically rejected a submission that the civil standard of proof applied to the Secretary of State and to SIAC. On the contrary, the task of the Secretary of State was to ‘evaluate risks in respect of the interests of national security’. He endorsed paragraph 44 of the judgment of this court (see paragraph 36, above). He also agreed that section 4 of the 1997 Act gave SIAC the power to decide questions of fact and of law (paragraph 30). He dismissed the appeal for the reasons given by this court, for those given by Lord Slynn, and for his own brief reasons (paragraph 32).

Lord Hoffmann explained in paragraph 34 of his speech that until 1998, the respondent would have had no right of appeal against the decision to deport him, as his right of appeal to an adjudicator was excluded by section 15(3) of the 1971 Act. The respondent could only have made representations to an extra-statutory panel.

Lord Hoffmann noted in paragraph 36 that the European Court of Human Rights (‘the ECtHR’) had held, in Chahal v United Kingdom (1996) 23 EHRR 413 that that procedure was not adequate to safeguard two of Mr Chahal’s Convention rights. The first was his article 13 right to an effective remedy from an independent tribunal to protect his right not to be deported to a country in which there was a serious risk that he would be subjected to a breach of his article 3 rights. The second was his right under article 5.4 to a determination by an independent tribunal of the lawfulness of his detention. The ECtHR rejected the Government’s argument that considerations of national security made such a right of appeal impossible. The ECtHR referred to a procedure under the Canadian Immigration Act 1976, which enabled the national security case to be tested, in the absence of the appellant.

That was the background to the passing of the 1997 Act. The 1997 Act was ‘intended to enable the United Kingdom to comply with the European Convention as interpreted by the court in Chahal’s case’. The procedure followed the Canadian model in allowing part of the appeal to be held in the absence of an appellant, with representation by a Special Advocate (paragraph 38). In paragraphs 39-41, Lord Hoffmann referred to the Secretary of State’s summary of the relevant facts and of his reasons for the decision. SIAC had held the appeal raised two issues: whether the respondent was engaged in the activities alleged by the Secretary of State, and in so far as SIAC found those activities proved, whether they were against the security interests of the United Kingdom. SIAC considered the principal allegations and found that the Secretary of State had not proved them to the necessary standard of proof (paragraph 42). It held that whether or not any proven activities were a threat to national security was a question of law for it to decide (paragraph 43).

Lord Hoffmann then summarised the reasoning of this court. This court identified three errors of law. Two are not directly relevant. The third error was to treat the Secretary of State’s ‘reasons as counts in an indictment and to ask whether each had been established to an appropriate standard of proof. The question was not simply what the appellant had done but whether the Home Secretary was entitled to consider, on the basis of the case against him as a whole, that his presence in the United Kingdom was a danger to national security. When one is concerned simply with a fact-finding exercise concerning past conduct such as might be undertaken by a jury, the notion of a standard of proof is appropriate. But the Home Secretary and [SIAC] not only do have to form a view about what the appellant has been doing. The final decision is evaluative, looking at the evidence as a whole, and predictive, looking to future danger’. He quoted, with approval, Lord Woolf’s statement that ‘the cumulative effect may establish that the individual is to be treated as a danger, although it cannot be proved to a high degree of probability that he has performed any individual act which would justify this conclusion’ (paragraph 48).

Lord Hoffmann endorsed the approach of this court on all three points. He said that while SIAC had been right to decide that section 4(1) gave it ‘full jurisdiction to decide questions of fact and law, they did not make sufficient allowance for certain inherent limitations, first, in the powers of the judicial branch of government, and secondly, within the judicial function, in the appellate process’. The exercise of the judicial function must recognise its constitutional boundaries. The limitations of the appellate process arise from ‘the need, in matters of judgment and evaluation of evidence, to show proper deference to the primary decision-maker’ (paragraph 49).

He described the effect of the separation of powers in paragraphs 50-54. The question whether something is in the interests of national security is a ‘matter of judgment and policy’. This is not a question for the courts. Lord Hoffmann noted the submission that section 4(1) gave SIAC the same powers as adjudicators had under the 1971 Act. The question was not the extent of SIAC’s appellate jurisdiction, but whether ‘the particular issue can properly be decided by a judicial tribunal at all’ (paragraph 52).

In paragraph 54 he explained that, despite these limitations, the whole decision on whether deportation would be in the interests of national security was not ‘surrendered to’ the Secretary of State. He referred to Lord Scarman’s analysis in Chandler v Director of Public Prosecutions [1985] AC 374, 406. That analysis showed that SIAC served ‘at least three important functions’ (my emphasis) which were necessitated by Chahal. First, ‘the factual basis for the executive’s opinion that deportation would be in the interests of national security must be established by evidence’. It was therefore open to SIAC to say that ‘there was no factual basis’ for the Secretary of State’s view that the appellant in that case was actively supporting terrorism in Kashmir. He added that SIAC’s ability to differ from the Secretary of State’s ‘evaluation may be limited…by considerations inherent in the appellate process’. The effect of that (subject to the next point) was that SIAC could not say that although that opinion ‘had a proper factual basis’ it did not accept that that was contrary to the interests of national security. Second, SIAC could reject the Secretary of State’s opinion ‘on the ground that it was “one which no reasonable minister advising the Crown could in the circumstances have reasonably held”’. Third, an appeal to SIAC might lie on issues which ‘at no point lie within the exclusive province of the executive’. Lord Hoffmann gave, as an example, the point which arose in Chahal, which was whether his deportation would breach article 3. The Strasbourg cases made it clear that whether deportation was in the interests of national security was irrelevant to rights under article 3.

In paragraphs 55-56 Lord Hoffmann considered the standard of proof. “‘A high civil balance of probabilities” is an unfortunate mixed metaphor’. Some things are more likely than others, and cogent evidence is generally needed to persuade ‘a civil tribunal’ that someone is guilty of fraud. Nevertheless, the question is always whether the tribunal thinks it more probable than not. In any event, the concept of standard of proof was not helpful. It may be sensible in a criminal or in a civil trial when the question is whether an event happened or not. The question in Rehman was not ‘whether a given event happened, but the extent of future risk. This depends upon an evaluation of the evidence of the appellant’s conduct against a broad range of factors with which they may interact. The question of whether the risk to national security is sufficient to justify the appellant’s deportation cannot be answered by taking each allegation seriatim and deciding whether it has been established to some standard of proof. It is a question of evaluation and judgment, in which it is necessary to take into account not only the degree of probability of prejudice to national security but also the importance of the security interest at stake and the serious consequences of deportation for the deportee.’

In paragraphs 57-58 Lord Hoffmann added some remarks about the limitations of the appellate process. SIAC was not the primary decision-maker. The decision was the Secretary of State’s. He had a ‘wide range of advice from people with day-to-day involvement in security matters which [SIAC], despite its specialist membership, cannot match’. This case did not involve a ‘Yes’ or ‘No’ answer about whether it was more likely than not that a person had done something, but an evaluation of risk. An appellate court would usually give a ‘considerable margin’ to the primary decision-maker. It should not normally interfere with the view of the Secretary of State unless it considers that the view of the Secretary of State is not one which can reasonably be entertained. Such restraint would not be necessary in relation to every issue, for example an article 3 issue. But it was required in relation to the question whether deportation was in the interests of national security. This limitation was not based on any limit to SIAC’s appellate jurisdiction. ‘The amplitude of that jurisdiction is emphasised by the express power to reverse the exercise of a discretion. The need for restraint flows from a common-sense recognition of the nature of the issue and the differences in the decision-making processes and responsibilities of the Home Secretary and [SIAC]’.

In paragraph 62 he added a postscript, prompted by the events of 9/11. In the field of national security, ‘the cost of failure can be high’. This underlined the need for judicial restraint. Not only did the executive have access to special information and expertise. Such decisions required ‘a legitimacy’ which could only be conferred by entrusting such decisions to those who have been democratically elected, so that the voters who elected them could also remove them.

Lord Clyde agreed with Lord Hoffmann. Lord Hutton agreed with Lords Slynn, Steyn and Hoffmann. He agreed that the Court of Appeal was right to hold that the Secretary of State had to assess the extent of future risk and that he was entitled to decide to deport the respondent on the ground that he was a danger to national security ‘viewing the case against him as a whole, although it cannot be proved to a high degree of probability that he has carried out any individual act which would justify the conclusion that he is a danger’(paragraph 65).

The respondent in Begum was born and brought up in the United Kingdom. She was a dual British-Bangladeshi citizen. She travelled to Syria when she was 15 years old, married an ISIL fighter, and lived in Raqqah. The Secretary of State decided to make an order depriving her of British nationality and certified that the decision was based on material which could not be made public. The respondent was detained in a camp in Syria. She appealed against that decision (‘decision 1’) to SIAC under section 2B of the 1997 Act. She later applied for entry clearance to return to the United Kingdom to take part in her appeal to SIAC. The Secretary of State refused that application. She also appealed to SIAC against that decision (‘decision 2’), under section 2 of the 1997 Act. She accepted that she could only appeal against decision 2 on human rights grounds, and that she had no such grounds. She therefore also attacked decision 2 on common law grounds in an application for judicial review of decision 2.

SIAC decided three preliminary issues against the respondent. It held that the effect of decision 1 was not to make her stateless (‘issue 1’). It held that decision 1 was not a breach of the Secretary of State’s extra-territorial human rights policy (‘issue 2’). SIAC also decided that the fact that the respondent could not have a fair and effective appeal from Syria did not mean that it was required without more to allow the section 2B appeal (‘issue 3’). The Administrative Court, also held, on the application for judicial review of decision 2, that the Secretary of State was not required to give the respondent entry clearance so that she could take part in her appeal in the United Kingdom (‘issue 4’). SIAC also dismissed the appeal against decision 2.

The respondent appealed against SIAC’s decision to dismiss the appeal against decision 2, and against the Administrative Court’s decision on issue 4. She also applied for judicial review of SIAC’s decision on issues 2, 3 and 4. This court also sat as a Divisional Court. For simplicity, I will refer to the court, wearing each of those hats, as ‘this court’. This court overturned the decisions below on issues 2 and 4. It held, on the application for judicial review, that SIAC had been wrong to apply public law principles to decide issue 2, and that it should have decided for itself whether or not decision 1 led to a breach of the respondent’s article 3 rights. It remitted issue 2 to SIAC. It agreed with SIAC that the fact that the respondent could not have a fair and effective appeal did not mean that her section 2B appeal should automatically succeed. It held, however, that the respondent’s right to a fair and effective appeal outweighed any risk which she might pose to national security, allowing the appeal against decision 2 and ordering the Secretary of State to give her entry clearance to take part in her appeal.

The Secretary of State appealed against this court’s decisions on issues 2 and 4. The respondent appealed against this court’s decision on issue 3. The Supreme Court overturned this court’s decisions on issues 2 and 4.

Lord Reed gave a judgment with which the other members of the Supreme Court agreed. He explained that the Supreme Court was dealing with appeals in three sets of proceedings: an appeal by the Secretary of State against this court’s decision on issue 2 (in the appeal against decision 1); an appeal by the Secretary of State against this court’s decision to allow the appeal against decision 2; and a related appeal by the Secretary of State in the application for judicial review of decision 2 (paragraph 13).

In paragraphs 28-81, Lord Reed considered, in detail, SIAC’s jurisdiction and powers in appeals under sections 2 and 2B of the 1997 Act. SIAC and this court had disagreed about that. SIAC had considered that the section 2 appeal was limited to human rights grounds. Its view, in the section 2B appeal, ‘in relation to the issue of the Secretary of State’s compliance with his policy’ was that it should not decide the issue on its merits, but apply the principles of judicial review (paragraph 28). This court, by contrast, considered that, on both appeals, SIAC should have decided for itself whether ‘the decision of the Secretary of State in question was justified on the basis of all the evidence before it, not simply determine whether the decision of the Secretary of State was a reasonable and rational one on the material before him as in a claim for judicial review’ (paragraph 29, citing paragraph 123 of the judgment of this court).

Lord Reed described the legislative history in paragraphs 30 and 31. In paragraph 33, he said that, under section 2 of the 1997 Act, the respondent had an appeal to SIAC against decision 2, ‘in so far as it refused her human rights claim’. He noted the terms of section 4 of the 1997 Act and its repeal with effect from 2003 (paragraph 34). He described the later history of amendments, concluding (paragraph 36) that the repeal of various provisions of the 2002 Act had ‘the effect of restricting the scope of appeals and narrowing the powers of the Tribunal and SIAC’. An appeal against the refusal of a human rights claim must now be brought on the grounds that the decision is unlawful under section 6 of the Human Rights Act 1998 (‘the HRA’). In an appeal under that ground, ‘SIAC must decide for itself whether the impugned decision is lawful’ (paragraph 37).

He noted, in paragraph 38, that deprivation appeals had an ‘entirely separate history’, which he then described. In paragraph 40, he said that there does not seem ever to have been a provision governing the nature of an appeal under section 2B. Section 4 of the 1997 Act was repealed on the day section 2B came into force, and sections 84-86 of the 2002 Act were not applied to section 2B appeals, although ‘The same appears not to be true of an appeal to the Tribunal under section 40A’ of the BNA.

In paragraphs 41-45, Lord Reed considered decisions of the UT concerning the scope of an appeal to the Ft-T under section 40A. In paragraph 46, he said that it was clear from the authorities that the principles to be applied to an appellate body, and the powers available to it, are by no means uniform. They depend on the relevant legislation and the procedures followed by the relevant courts. Some appeals, such as licensing appeals to the magistrates’ court, are a complete re-hearing. In appeals to this court against the exercise of a discretion by a lower court, the scope of the appeal is much more limited. A similar approach had been taken to appeals against the exercise of a discretion by a statutory decision-maker (paragraphs 47-49).

In paragraphs 51-62, he summarised the decision in Rehman (paragraphs 34-50, above). He pointed out that section 4 of the 1997 Act applied to section 2 appeals when Rehman was decided (paragraph 52). In paragraph 59, he said that a contrast might be drawn between ‘the hybrid approach’ of Lord Slynn and what he described as ‘Lord Hoffmann’s more orthodox (in public law terms) identification of the relevant questions’ as whether there was a ‘proper factual basis’, or (conversely) whether there was ‘no factual basis’ for the Secretary of State’s opinion, and whether the Secretary of State’s opinion was one which no reasonable minister could have held. Lord Reed concluded that however the law stood at the time, the repeal of section 4 of the 1997 Act, and the absence of a similar provision in the current scheme, meant that ‘Lord Hoffmann’s approach is now the more relevant’. He referred to Lord Woolf’s statement in this court, approved by Lord Hoffmann (see paragraphs 35 and 43 above). He said that the points made by Lord Hoffmann had been repeated in later cases (paragraph 62).

Against that background when considering SIAC’s functions and powers in an appeal under section 2B it was necessary to examine the nature of the decision and any statutory provisions which shed light on the question (paragraph 63).

It was also necessary to bear in mind that ‘the appellate process must enable the procedural requirements of the [European Convention on Human Rights (‘the ECHR’)] to be satisfied, since many appeals will raise issues under the [HRA]’. In an immigration case (before the repeal of section 4 of the 1997 Act) those would generally include the potential to challenge the legality of a measure, its compatibility with absolute rights and the proportionality of any interference with qualified rights. SIAC must also be able to allow an appeal when the Secretary of State’s assessment of the requirements of national security has ‘no reasonable basis in the facts or reveals an interpretation of “national security” that is unlawful or arbitrary’. In deprivation cases, a more limited approach had been taken. An arbitrary denial or deprivation of citizenship might raise an issue under article 8 in some circumstances. The question was not whether the revocation was proportionate, but whether it was arbitrary. The ECtHR had considered the impact on the applicant. In deciding whether the deprivation is arbitrary, the ECtHR has considered whether the deprivation was in accordance with the law, whether the authorities acted diligently and swiftly, and whether the person concerned had the procedural safeguards required by article 8 (paragraph 64).

Section 2B confers a right of appeal, in contrast with sections 2C to 2E, which provide for a right to apply for a ‘review’ (in which judicial review principles are to be applied). There are no such limitations on an appeal under section 2B. The power to make procedure rules conferred by section 5(1)(b) contemplates appeals involving questions of fact as well as points of law (paragraph 65).

In paragraphs 66-68 Lord Reed considered the nature of the decision under appeal. He explained that section 40(2) gives the Secretary of State, and not SIAC, a power to decide to make a deprivation order. There is no statutory indication that Parliament intended that discretion to be exercised by SIAC. SIAC could, however, ‘review the Secretary of State’s exercise of his discretion and set it aside in cases where an appeal is allowed, as explained below’ (paragraph 66). He added that a statutory condition must also be met. The Secretary of State must be satisfied that deprivation is conducive to the public good. The condition does not require SIAC to be satisfied of that. The right of appeal enabled that decision to be challenged, but did not make SIAC the decision-maker on that issue. The authorities show that appellate courts and tribunals cannot generally decide how a statutory discretion conferred on a primary decision maker should be exercised, or exercise the discretion themselves, unless they are so authorised by statute (as they were by section 4, before its repeal, and by sections 84-86 of the 2002 Act, before they were repealed). They could decide whether the decision-maker had acted in a way in which no reasonable decision-maker could have acted, whether he had taken irrelevant considerations into account, or failed to take into account something which he should have taken into account, or had erred in law. They must also decide questions of compatibility with Convention rights for themselves (paragraph 68).

That was an apt description of SIAC’s role in a deprivation appeal. That did not mean that SIAC’s jurisdiction was supervisory rather than appellate. References to a supervisory jurisdiction were apt to confuse. The fact that a jurisdiction is appellate does not decide what principles of law the appellate body is to apply. Those depend on the decision appealed against and the relevant statutory provisions. ‘Different principles may even apply to the same decision, where it has a number of aspects giving rise to different considerations, or where different statutory provisions are applicable’. When SIAC reviews the exercise of the Secretary of State’s discretion conferred by section 40(2), it must apply principles which are ‘largely the same as those applicable in administrative law…’ But it has to decide whether the Secretary of State has acted compatibly with an appellant’s Convention rights ‘objectively on the basis of its own assessment’ (paragraph 69).

In applying the relevant principles to discretionary decisions, SIAC must take into account the nature of the discretionary power in question, and the Secretary of State’s ‘statutory responsibility for deciding whether the deprivation…is conducive to the public good’. The exercise of the power ‘must depend heavily upon a consideration of the relevant aspects of the public interest…Some aspects of the Secretary of State’s assessment may not be justiciable…Others will depend, in many if not most cases, on an evaluative judgment of matters, such as the level and nature of the risk posed by the appellant, the effectiveness of the means available to address it, and the acceptability or otherwise of the consequent danger, which are incapable of objectively verifiable assessment’. SIAC had to give the Secretary of State’s assessment ‘appropriate respect’, for reasons of ‘institutional capacity (notwithstanding the experience of members of SIAC) and democratic accountability…’ (paragraph 70).

In paragraph 71, he further explained that SIAC has important functions, nevertheless. He gave four examples.

He added that SIAC might have to consider relevant evidence. Some decisions might ‘involve considerations which are not justiciable’. SIAC also had to bear in mind that ‘due weight’ had to be given to the ‘findings, evaluations and policies of the Secretary of State’. In reviewing compliance with the HRA, SIAC ‘has to make its own independent assessment’.

The premise of the approach of this court to the appeal had been a different understanding of SIAC’s jurisdiction and powers, as explained in paragraph 72 of Lord Reed’s judgment. This court had not been referred to Rehman, but to two decisions which were either wrong, or irrelevant (paragraphs 73-76 and 77-79). In Pham v Secretary of State for the Home Department [2015] UKSC 19; [2015] 1 WLR 1591, the Supreme Court appeared mistakenly to have thought that section 4 applied to section 2B appeals, when it had been repealed, and had never applied to them, but nevertheless held that SIAC was required to review the reasonableness or rationality of the Secretary of State’s decision under section 40(2) (paragraph 80). A passage from the judgment of Lord Sumption in Pham on which the appellant relied was in fact a clear reference to the common law test of rationality or reasonableness. His observations were not consistent with ‘an approach which would place SIAC “in the shoes” of the decision-maker and treat it as competent to re-consider the matter de novo or re-take the decision itself’ (paragraph 81).

In paragraphs 84-96, 98-111 and 112-131, Lord Reed gave reasons for his conclusions on the four strands of the appeal and cross-appeal.

Lord Reed summarised his conclusions about the judgment of this court in paragraphs 133-136.

The issue in R (Pearce) v Parole Board of England and Wales [2023] UKSC 13; [2023] 2 WLR 839 was the lawfulness of guidance issued by the Parole Board (‘the Board’). The guidance advised the Board, if it was not able on the balance of probabilities to make a finding of fact about an allegation against a prisoner, that it should make a judgment about the ‘level of concern’ which arose from an unproven allegation, and then consider what impact that level of concern had on its decision whether or not to direct the release of a prisoner. The Supreme Court held that the guidance was lawful. Lord Hodge, giving a judgment with which the other members of the court agreed, considered the prisoner’s submission that, in law, there is a binary concept of ‘fact and non-fact’ (paragraph 30). He referred to the judgments in Rehman and Begum. He said (paragraph 43) that the judgments of Lord Slynn and of Lord Hoffmann in Rehman were consistent in saying that the original decision-making process was not confined to deciding whether or not past facts had been proved. Lord Slynn was, instead, saying that where such facts are relied on, fairness required that they should be proved. Lord Hodge rejected the prisoner’s submission. He held (paragraph 44) that the judgments of Lord Slynn and Lord Hoffmann ‘support the view that a decision-maker, whether a member of the executive branch of government or a judicial body, when assessing future risk, is not as a matter of law compelled to have regard only to those facts which individually have been established on the balance of probabilities; the decision-maker, from the assessment of the evidence as a whole, can take into account, alongside the facts which have been so established, the possibility that allegations, which have not been so established, may be true’.

In P3 v Secretary of State for the Home Department [2012] EWCA Civ 1642; [2022] 1 WLR 2869 this court allowed the Secretary of State’s appeal against a decision of SIAC to allow P3’s appeal against a refusal of entry clearance, on the ground, among others, that SIAC had erred in law in substituting its assessment of the risk posed to national security by P3 for that of the Secretary of State. I gave the main judgment. In paragraphs 114-115, I added a ‘Postscript’ which was not necessary to the decision, in which I expressed concern about submissions from the Secretary of State which indicated that, in the light of Begum, SIAC was taking an unduly narrow approach to section 2B appeals. I said that Begum ‘is authority for the proposition that, broadly, SIAC should take a public law approach to challenges to the Secretary of State’s assessment of national security. It is not authority for any wider proposition’ (paragraph 114). The 1997 Act clearly distinguishes between appeals and reviews. SIAC was not confined to applying public law principles on all the issues which might arise in such an appeal, nor was it confined to considering only the materials which were before the Secretary of State when he made the impugned decision. SIAC could take into account material which emerged on an exculpatory review, which might not have been before the Secretary of State. SIAC might also exclude material which was before the Secretary of State, for example if there was a risk that it was obtained by torture. ‘In any event, SIAC hears evidence on an appeal, which was not before the Secretary of State, and is entitled to make of that evidence what it may’(paragraph 115).

Sir Stephen Irwin, and Bean LJ, who agreed with him, said that ‘a proper degree of respect for the national security assessment of the Secretary of State …must not be mis-translated into an erosion of the right of appeal under section 2B so that it becomes indistinguishable from a review under sections 2C to 2E’ (paragraph 118).

Sir Stephen Irwin added that it is SIAC’s function to ‘scrutinise all the evidence, OPEN and CLOSED, with a critical and expert intelligence, to test the approach and the evidence bearing on the assessment, both for and against the assessment, both for and against the conclusions of the Secretary of State, and then applying due deference, to decide whether the conclusions of the Secretary of State were reasonable and, adopting the phrase of the Strasbourg court, conformed with common sense. In doing so, SIAC is bound to show deference at all stages, and at all levels…Proper deference there must be, but it does not amount to a simply supine acceptance of the conclusions advanced by the Secretary of State. I do not understand that to be in any way implied by the decisions in Rehman or Begum’ (paragraph 126; and 129).

SIAC described the focus of the appeals in paragraphs 7-12. U3’s case was that she did not pose a risk to national security. She invited SIAC to find that the Secretary of State’s contrary view ‘was and is flawed’. In appeal 2, she submitted that it was for SIAC to decide for itself whether any interference with article 8 rights was outweighed by the risk to national security. She conceded that if she could not impugn the assessment that she posed such a risk, her appeal was unlikely to succeed (paragraph 7).

In paragraph 8 SIAC said that that concession was realistic, if it could not interfere with the Secretary of State’s assessment that she posed such a risk. SIAC had nevertheless formed a view about the weight which should be given to the article 8 rights of the family and to the children’s best interests, bearing in mind the duty imposed by section 55 of the Borders, Citizenship and Immigration Act 2009 (‘section 55’). It added that the focus of the appeal, nevertheless, and rightly in its view, had been the national security assessment. U3’s counsel had chosen not to cross-examine the Secretary of State’s national security witness, because the OPEN allegations were only a bare outline. He maintained that stance when told that the Special Advocates had also decided not to cross-examine that witness (paragraph 10). SIAC would return to that issue later in the judgment, in paragraph 213 (see paragraph 157, below).

Counsel explained that U3’s positive case, in both appeals, was based on U3’s written and oral evidence and the evidence of her expert witnesses. As a whole, that evidence showed that when she travelled to Turkey, and then to Syria, U3 was in ‘an abusive, coercive and controlling relationship with her then husband (”O”)’. She travelled in the context of that relationship. She did not then know about ISIL’s ideology or atrocities. She was not, then, or later, ideologically aligned with ISIL. She did not become radicalised in Syria and does not now support ISIL or extremism more generally (paragraph 11).

With that introduction, SIAC divided the judgment into five significant parts. As SIAC observed in paragraph 12, this positive case brought ‘into sharp focus the approach that SIAC should adopt when deciding appeals’ such as appeals 1 and 2, ‘where both decisions are based on an assessment that the appellant poses a risk to national security’. This was the first substantive SIAC appeal since Begum and Secretary of State for the Home Department v P3. SIAC therefore considered, in paragraphs 13-43, the parties’ submissions, and the authorities, and decided what approach it should apply to each appeal.

The second significant part of the judgment is SIAC’s analysis of the OPEN evidence (paragraphs 49-179). In the light of the criticisms of the judgment on this appeal, I consider that it is necessary to summarise that part of the judgment at some length. In the third significant part of the judgment (paragraphs 181-204), SIAC ‘addressed’ nine factual issues which, in written submissions after the hearing, U3 had submitted that SIAC had ‘institutional competence’ to decide. The fifth part of the judgment was SIAC’s conclusions on the two appeals (paragraphs 205-215, and 216-219).

The background to SIAC’s analysis of the authorities is that much of the law was common ground between the OPEN representatives, although the Special Advocates did not agree with that common ground. SIAC recorded that common ground in paragraphs 14 and 15. One area of dispute was whether SIAC’s role in reviewing the national security assessment was limited to the four functions mentioned by Lord Reed in paragraph 71 of his judgment in Begum or was wider than that (paragraph 16). The Special Advocates submitted that appeal 1 was governed solely by public law principles (paragraph 20). The effect of those submissions, if correct, would be that all of U3’s evidence would be inadmissible on appeal 1 (paragraph 21).

SIAC noted that it was important to distinguish between appeal 1 and appeal 2. There were two distinct but related questions in relation to each: first, the grounds on which SIAC can interfere with the relevant decision, and second, the evidence which is relevant and admissible (paragraph 22). Begum was a decision about cases in which the Secretary of State has exercised the discretion conferred by section 40(2) of the BNA to deprive an appellant of her nationality on grounds of national security. It was based on the principle that the judgment whether something is in the interests of national security is ‘constitutionally reserved to the executive’ (Rehman, paragraph 50, cited in paragraph 56 of Begum). Moreover, despite its expert membership, SIAC is not ‘institutionally competent’ to substitute its judgment for that of the executive, especially where this involves an assessment of future risk (Rehman, paragraph 57, cited in paragraphs 60-61of Begum) (paragraph 23).

SIAC understood Begum to be ‘driven by constitutional and institutional considerations specific to national security assessments, and not only by the terms of the statutory regime’. SIAC noted that Lord Reed had said, in paragraph 69, that the principles of law which apply to an appellate body depend on ‘the nature of the decision under appeal and the relevant statutory provisions (emphasis added)’. SIAC referred to paragraph 70 of Begum: the correct approach depends on the issue. Some assessments are not justiciable at all, such as the question whether the promotion, abroad, of terrorism in a foreign country is contrary to the interests of national security. Other assessments, such as the level and nature of the risk posed, the effectiveness of measures to combat the risk, and the acceptability or otherwise of the consequent risk, are justiciable, but only on limited grounds, which reflect the constitutional and institutional considerations which SIAC had mentioned (paragraph 24).

SIAC rejected the Special Advocates’ submission. It was clear, both from paragraph 54 of Rehman and from paragraph 71 of Begum, that SIAC had to make its own assessment of compliance with the ECHR. SIAC considered, by reference to paragraph 114 of P3, that it was ‘more accurate’ to say that “Begum is authority for the proposition that, broadly, SIAC should take a public law approach to challenges to the Secretary of State’s assessment of national security” (emphasis added)’. SIAC considered that that formula, even if obiter, was correct, although SIAC did not think that it was obiter. It was a summary of the reasoning in paragraphs 95-102 of the judgment, which explained why, even in a case in which ECHR rights are in issue, the reasoning in Rehman and Begum applies to assessments of national security (paragraph 25).

In paragraph 26, SIAC summarised the grounds on which it could interfere with an assessment of national security (‘at least three’ according to paragraph 54 of Lord Hoffmann’s judgment in Rehman). Lord Reed had added, in paragraph 71 of Begum, a further function in a deprivation appeal. That function concerns the issue of statelessness.

SIAC added that neither Lord Hoffmann nor Lord Reed claimed to have given an exhaustive list of the grounds on which a deprivation decision could be impugned. Nor did either suggest that the available public law grounds did not include all the grounds which would be available in other cases. It listed four such examples, rejecting a narrower submission of the Secretary of State (paragraph 27). SIAC therefore accepted the Special Advocates’ submissions that the grounds on which an assessment of national security could be impugned included all the grounds which would be available on an application for judicial review, and that if a deprivation decision was flawed on public law grounds, SIAC would set it aside unless the Secretary of State could show that, the decision would inevitably have been the same ‘irrespective of the error’ (paragraph 28).

SIAC then asked whether the permissible grounds of challenge went beyond those available in public law. It considered the judgment of Sir Stephen Irwin in P3. SIAC accepted the submission of the Secretary of State that it could not interfere with an assessment of national security ‘on the ground that it “does not conform to common sense”’ (paragraph 30). His statement was obiter, and not supported by the relevant decisions of the ECtHR. On this appeal, Ms Harrison did not criticise SIAC’s approach to this point.

SIAC was not the alter ego of the Administrative Court. It has special expertise in immigration law and in the assessment of intelligence. ‘Its procedures allow for the detailed consideration of evidence, OPEN and CLOSED, including exculpatory evidence’. It often hears evidence from a national security witness about the national security assessment. In this respect, the tools available to SIAC go beyond those which would be available to the Administrative Court, even in a case where CLOSED material procedures apply (paragraph 31).

That meant that there might be cases in which SIAC could see a flaw which could not be detected in ordinary judicial review procedures, because ‘it has a more powerful microscope, not because it is looking for a wider range of flaws’. SIAC gave as an example the way in which the Special Advocates can test the details of an assessment in the course of CLOSED hearings. The question whether such a flaw vitiates a decision is answered by applying the same standards as on an application for judicial review (paragraph 32).

Conversely, there might also be cases in which such close scrutiny could show that what seemed at first sight to be a flaw was not, on closer inspection, a material flaw (paragraph 33).

In paragraph 34, SIAC returned to the logical distinction between two related questions: the grounds on which an assessment of national security can be challenged and the evidence which is relevant and admissible on an appeal. The ‘extreme’ consequence of the Special Advocates’ submission was that SIAC could only look at the evidence which was before the Secretary of State when she made the decision, and all U3’s evidence on appeal 1 would be inadmissible.

That was the position in most ‘judicial review contexts’ and was not unfair because the claimant would have had the opportunity to make representations before the decision was made. But deprivation decisions are often taken without giving the person affected any such opportunity. If she could not give evidence on her appeal, no-one would consider her answer to the case against her. Parliament had given her a right of appeal. SIAC found it ‘impossible to interpret the regime it enacted as precluding SIAC from taking into account [U3’s] own evidence that she is not a risk to national security’ (paragraph 35).

SIAC then asked how it could take U3’s evidence into account ‘as relevant to the national security assessment’ while still applying Begum (paragraph 36). SIAC concluded that it could in at least three ways.

First, in a deprivation appeal, if SIAC could consider evidence which was not before the Secretary of State, but that was limited to evidence about things which happened before the deprivation decision. SIAC might identify something which the Secretary of State did not consider, but which (given what she did know, or ought reasonably to have known) she should have taken into account. That might be relevant to a Wednesbury review or to a Tameside review (Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014) (paragraph 37).

Second, once an appellant appeals, the national security assessment will be updated to take account of her evidence and of any material uncovered by the exculpatory review. In that way, those advising the Secretary of State keep the deprivation decision under review during the appeal. Officials would be bound to draw to the attention of the Secretary of State anything which undermines or materially changes the original assessment. The updated assessment will in practice replace the original assessment for the purposes of the appeal, even if it has not been shown to the Secretary of State. If that updated assessment ‘is shown to be flawed in the public law sense’ the appeal would be allowed. This approach differs from the approach in an application for judicial review, in which ‘rolling’ judicial reviews have been deprecated. The appeal is designed to be a ‘one-stop shop’ in which the appellant either wins ‘(with the result that her citizenship is restored)’ or loses, in which case it is not. In the light of this court’s later decision in E3 v Secretary of State for the Home Department (see paragraphs 29-33, above) this statement is not right. I must make it clear that I do not criticise SIAC on that account. SIAC said that this approach could be reconciled with ‘the public law approach in Begum if SSHD keeps the challenged decision under review during the appeal and SIAC treats SSHD’s updated national security assessments as superseding the original ones’ (paragraph 38).

Third, the appellant might ask the Secretary of State to take a further step, such as giving her entry clearance. When taking that second decision, the Secretary of State will be able to consider the up-to-date position, including any representations from the appellant. The national security assessment for that decision will also have to take those representations into account. The evidence adduced by the appellant on the entry clearance appeal will be relevant in so far as it shows that the national security assessment is flawed ‘in public law terms’ and the Secretary of State cannot show that the decision would inevitably have been the same. In such a case, the appeal to SIAC will succeed (paragraph 39).

So SIAC is not limited to evidence which was before the Secretary of State at the time of the decision. What it could do with that evidence, however, was strictly limited, ‘when its purpose is to undermine a national security assessment’. Those limits should be clear, as should their implications for a case like this case. SIAC then made three points.

First, in this case the deprivation decision turned on an assessment of national security, as, ‘in practical terms’, did the entry clearance decision. If the Secretary of State was, and was still, entitled to conclude that U3 is a risk to national security, and SIAC could not interfere with that assessment, the national security risk was likely to outweigh the article 8 interests of U3’s family.

Second, the appeal regime does not allow SIAC, on either appeal, to reach its own view about whether U3 is a risk to national security.

Third, that was so even though SIAC had heard U3’s oral evidence, and the evidence of other factual and expert witnesses (paragraph 40).

SIAC summarised the position in paragraph 41. The challenged decisions had an effect which is ‘in some respects more severe than a prison sentence… U3 cannot return to the country where she was born and grew up, and must remain separated from her children for the foreseeable future.’ Nevertheless, the law allows the Secretary of State to deprive U3 of her nationality on the grounds of national security, without giving her an opportunity to show, before the decision was made, that she was not a risk to national security, without requiring the Secretary of State to prove, even on the balance of probabilities, that she was such a risk, and without giving her the chance to prove to an independent tribunal that she is not such a risk. SIAC could not decide for itself on the entry clearance appeal whether U3 is a risk to national security. The extent to which she could challenge the assessment of national security was limited to public law grounds on both appeals.

In paragraphs 42 and 43, SIAC considered how the extent of any risk should be assessed. A court or tribunal considering proportionality for the purposes of article 8 ‘needs a notional set of scales’. It was for the court or tribunal to decide on which side the scales came down. In reaching that view, it must show the respect which is due to the legislative and executive branches. There was a statutory framework in deportation cases, but no such statutory framework in national security cases. The approach in Begum required ‘great respect to be shown to the judgment of the executive about whether a risk to national security is made out’ (paragraph 42).

The same applied to the judgment about the extent of any risk. In P3, this court had allowed the Secretary of State’s appeal because SIAC had not given enough respect to the judgment of the Secretary of State about the extent of the risk to national security which was posed by P3. Assessing the extent of any such risk ‘is likely to be an imprecise evaluative exercise. It may be very difficult to predict with any certainty what a particular individual might do if permitted to return to the UK. There are many possibilities…This is the kind of predictive assessment which the Supreme Court in Carlile said was, in the first instance, for the executive. The later authorities indicate that when assessing the level of risk, and judging whether it is acceptable, SIAC is required to accord great weight to the view of SSHD’ (paragraph 43).

In paragraphs 44-48, SIAC considered article 8 and section 55. It decided that as U3 was not within the territory of the United Kingdom, and was not in any category of non-territorial jurisdiction, the United Kingdom was not obliged in international law to secure her article 8 rights, but was obliged to secure those of her children. In this case it was artificial to distinguish her article 8 rights from theirs. The family’s interests in reunification were engaged in appeal 2. It was common ground that section 55 was engaged when decision 2 was made and that the duty to have regard to the needs specified in section 55 did not mean that they should be given precedence.

In paragraphs 49-61, SIAC summarised the OPEN case against U3. In paragraphs 49-57, it described the OPEN submission to the Secretary of State. The introduction said that the assessment of the Security Service was that people who had travelled to Syria or to Iraq to align with ISIL were a ‘serious and credible threat to UK national security as a result of being in ISIL-controlled territory’, whether or not they were combatants. Such a person ‘is likely to have been radicalised, has contributed to the continuation of ISIL as an entity, and may have received military training, fought with ISIL or taken part in terrorist attacks. Individuals who remain in theatre are likely to pose a continuing threat to the UK, but we assess the threat would be significantly higher if they returned to the UK’ (paragraph 49).

Paragraph 9 of Section C said that those who had travelled to Syria since the formation of the ‘so-called caliphate’ and those who had travelled before June 2014, and had stayed there with ISIL, ‘will have been radicalised and exposed to ISIL’s extremism and violence’. If they returned to the United Kingdom, it was assessed that ‘they will present a national security threat to the UK’.

Paragraph 13 summarised sentencing guidance for terrorism offences given by this court, in three broad types of case. The first was those who joined or otherwise supported a ‘terrorist organisation usually engaged in conflict overseas’ and who ‘participat[ed] on the periphery of actual combat’ or trained for combat. The second was those who tried but failed to join a terrorist organisation. The third was those who had such an intention but did not do much to act on it. The sentencing range was 21 months’ imprisonment to 20 years’ imprisonment. Paragraph 14 said that the Secretary of State considered that the Court’s view of the seriousness of travelling to join ISIL was ‘an accurate reflection of the seriousness of the threat posed by such individuals’.

There was an assessment in paragraph 22 of the ministerial submission that anyone who had travelled voluntarily to ISIL-controlled territory to align with ISIL since the declaration of the caliphate, or who had gone before that and had stayed voluntarily ‘is aware of the ideology and aims of ISIL and the attacks and atrocities it has carried out’. The declaration of the caliphate was ‘a key cornerstone of the group’s rhetoric and propaganda as well as its ideological appeal’ (paragraph 52).

In paragraph 53, SIAC quoted paragraph 33 of the ministerial submission, which was headed ‘Radicalisation in theatre’. The assessment was that everyone is ‘exposed to routine acts of extreme violence in ISIL-controlled territory’. That included, but was not limited to, ‘armed conflict’ between ISIL and other factions, ‘crimes against civilian populations carried out by invading fighters (genocide, ethnic cleansing, mass rape, looting) air strikes against civilian targets, and public executions and corporal punishment administered by ISIL’. The assessment was that this was likely to de-sensitise people to acts of brutality, and to encourage them to see ‘terrorist activity in the UK or against UK interests as an acceptable and legitimate course of action’. The assessment therefore was that even those who had travelled to ISIL-controlled territory ‘involuntarily’ (original emphasis) were likely to have been radicalised during their time there ‘due to their daily exposure to ISIL indoctrination and extreme violence’.

In paragraph 54, SIAC identified the five ways in which it was assessed that people who have spent time in ISIL-controlled territory may pose a threat to national security.

Paragraph 46 of the ministerial submission referred to the al-Khansaa brigade. This is an all-woman group which was said to police women civilians in Raqqa. In paragraph 50, the ministerial submission described the assessment that any man or woman who returned to the United Kingdom having spent a long time in ISIL-controlled territory was likely to have the capacity to carry out an attack independently ‘regardless of whether or not they have had contact with ISIL’s external operations section’. The judgment was that ‘the risk of an ISIL-enabled attack occurring in the UK will increase’ if people ‘who have been in ISIL-controlled territory return to the UK’ (paragraph 55).

In paragraph 56, SIAC quoted paragraphs 53 and 54 of Section C of the ministerial submission. It was possible that some of these people who returned to the United Kingdom in the future might not get involved in any of the five activities described in the assessment ‘at least not immediately on their return’. The assessment, nevertheless, was that they were a ‘potential threat’ to national security. Those people had already shown a ‘demonstrable commitment to living under extremist rule’. There was ‘every possibility that, if the same circumstances were to arise again, they would look to travel to another area of conflict in answer to a call for violent jihad’. They would therefore continue to be a threat to national security ‘for all of the reasons set out in this statement’.

Annex D of the ministerial submission considered U3’s case. The ‘key assessment’ was that she travelled to Syria and aligned with ISIL. Police reporting indicated that she travelled from London to Istanbul via Zurich on 24 June with her two children. The same reporting indicated that in October 2014 she contacted her younger sister, who was based in the United Kingdom, via Skype. According to her sister, she said that she was ‘Ok’, was in Turkey, and was not returning to the United Kingdom. The police report also said that in July 2015, she contacted members of her family on Facebook. She was said to have reported that she was living in Istanbul with her two children. She said she was still married, but that O spent most of his time working abroad and she rarely saw him. The assessment was that U3 was not in Istanbul, but in Syria at the time of that contact, and that she had travelled to Syria and aligned with ISIL.

SIAC recorded, in paragraphs 58-61, four submissions by counsel for the Secretary of State. The first was that there was no information at the time of the deprivation decision that U3 was coerced into travelling to Syria. She did not say so in her first statement, which she drafted herself, or in her second or subsequent statements. She went ‘willingly and independently’ (paragraph 58). He also submitted that there was no evidence that U3 had tried to contact the authorities about O’s coercive behaviour in the three years between her arrival in Syria and the deprivation decision. She lied to her family about where she was. There was nothing to indicate that, despite moving to Raqqa, she was not aligned with ISIL (paragraph 59). The Secretary of State, he submitted, was entitled to take a ‘preventive and precautionary approach’ to the potential future risk posed by U3. ‘It was not necessary to reach findings on the balance of probabilities or otherwise as to precisely what U3 had been doing in Syria, or her state of mind as far as support for ISIL was concerned. Her task was, rather, to take account of all the material, OPEN and CLOSED, generic and specific, and reach “an overview” as to the national security risk posed’ (paragraph 60).

Finally, the national security assessment was reconsidered and maintained in the light of U3’s second witness statement. It was accepted that she had ‘a difficult relationship’ with O, but the assessment that she travelled to Syria and aligned with ISIL was maintained, and that she was ‘an active and willing’ traveller to Syria. It was also highly likely, for the reasons given in that assessment, that U3 would have been aware of some of the atrocities committed by ISIL and nevertheless chose to go to territory controlled by ISIS (paragraph 61).

SIAC said that U3’s OPEN case was based on two points. First, she did not go to Syria with the intention of aligning with ISIL, and was not radicalised after that. Second, it was in the best interests of her children that she be allowed to join them in the United Kingdom (paragraph 62). Her factual evidence included a statement she wrote in 2018, and four witness statements made in the course of the appeal (in August 2020, November 2020, March 2021 and on 11 October 2021). Those four statements gave ‘an extremely detailed account of U3’s life in the UK with O, his violence towards her and his control of her life, her travel to Turkey and Syria and her life there’ (paragraph 63). There were also factual statements from U3’s brother and sister in law, from her current partner, M, from another person she met in Syria, from her friend SM, from her mother, and from two of her solicitors. SIAC had read those, but did not summarise them in the judgment. In key respects, they corroborated U3’s account. But ‘it was obviously not possible’ to put in cross-examination any of the CLOSED evidence on which the Secretary of State relied (paragraph 64).

SIAC summarised what it referred to as U3’s ‘self-drafted statement’ in paragraphs 65-83. She said she was a ‘normal teenager’, interested in fashion and music. Her father was not part of her life, and she did not get on with her mother. She got to know O when she was studying for her ‘A’ levels. He was respectful towards her at first. Her mother disapproved of the relationship. She moved out to live with O instead of finishing her education. They married in a religious ceremony after living together for about three weeks. She and O were raised in Islamic traditions, but were not practising Muslims. They ‘turned to God’ because of financial difficulties. O met YC at the Mosque. YC was from Algeria. O’s attitude to U3 then changed. She was never allowed to socialise, to have friends, or to leave the house. She was given a basic phone. O manipulated her into believing that there was something wrong with her.

She ‘came to understand’ that O had been caught twice trying to travel to Syria and had been banned from entering Turkey for ten years. He left her and the children without warning, taking her bank card. He was away for three weeks. She learnt that he had married another woman. She did not feel she could go to her mother. She was supported by a woman at the Mosque. She decided to go to Turkey, believing that O could not reach her there. She stayed in Istanbul with a family known to O. She had no intention of travelling to Syria. O joined her in Turkey. He ‘manipulated’ her into going to Syria, by saying that he loved her so much that he had gone all the way to Turkey illegally, and that anything could have happened to her. He made her believe him. She did not know why she could not say ‘No’. She wanted him to leave her and she wanted him to treat her well. ‘That was a stupid thing to have done’. She described the journey to Syria, how they joined a group and how the men travelled first, and the families followed, under the direction of people smugglers. She gave a detailed account of the perils of that journey with her two children. At first she was only allowed a brief meeting with O, and only after her protests. She had no friends. Living conditions were poor and ‘prison-like’. Eventually the men and women were reunited and removed to Raqqa. Living conditions were no better and she was still separated from O.

He joined them after three weeks. She understood that he had done the theoretical part of his ISIL training, but not the practical part. They lived in an old house with no sanitation, and then in a car which O bought. He then got a job as a court clerk, and they were able to rent a house. She got to know two unconnected British citizens, A and R. R wanted to leave. ‘U3 was disillusioned with ISIL to point [sic] of hating it. She had heard too many horror stories and knew that she had made a mistake travelling to Syria’. She could not travel without O’s permission. He continued to abuse her. O hated his job and used her as a punch bag. He lost his job and was briefly in prison. They had to find somewhere else to live. O then got a job which involved frequent travel. That suited U3. She became very friendly with a neighbour who owned an internet cafe with her husband. The friend supported her during her pregnancy and when her third child was born.

O lost his job again and decided that he hated ISIL. But there was no safe way to leave with the children. They had to leave their house in Raqqa. O was stressed and ‘as usual’ U3 was ‘his punching bag’. They had to abandon an escape plan because of artillery attacks. Eventually O found a way for them all to escape. Her account of the escape was ‘detailed’. She was arrested by the Syrian authorities, tried with belonging to ISIL and ‘acquitted’. I note that Ms Harrison accepted in the course of her oral submissions in this court that U3 was not in fact acquitted, but that the charges were dismissed for lack of evidence. She was taken to a camp. She was told that O had been detained in Turkey and that she would be sent there. Conditions in the camp were ‘very challenging’ but they improved after a ‘chance meeting’. She was interviewed by Turkish intelligence. Her mother was allowed to send her money. She was given permission to leave the camp, but returned. Conditions there ‘had improved greatly’. She had never had any bad intentions. She did not take part in any criminal acts. She did not radicalise anyone. She was very normal, and ‘would love to have opportunity [sic] to come home’.

In paragraphs 84-144, SIAC summarised U3’s statements in the appeal. These are more detailed than her first statement. The main difference is that U3 said that O began to be jealous of her and to try to control her as soon as they began to talk about getting married. At first she found this romantic. Her mother established that O ‘had no papers’. She would approve of the marriage when O got a good job. She had a heated argument with her mother and left home to live with O. They did not sleep together until they got married. They married after about two weeks. She was 18. As they were leaving the Mosque he told her that she would have to stop wearing trousers. He thought that a woman became her husband’s property when they got married. She thought that this was romantic.

O began to be violent to her on the first day of their marriage. She described his frequent violence, his increasing attachment to conservative Islam, and his control over her dress and conduct. He did not allow her to listen to music on her laptop, or to watch television or films, strictures which he enforced with physical violence. She went along with these, in order to avoid the violence, but without understanding their basis. She became pregnant with their first child. He controlled her interactions with the health service. At his insistence, they contracted a civil marriage. She inferred that he had been advised that this would help him to get leave to remain. She restored her relationship with her mother, but O resented that. On one occasion he beat her, even though she was heavily pregnant. O became violent again soon after the birth. He would apologise, with apparent sincerity. He was given leave to remain but did not try to get a job. He resumed sexual relations with her against her will and she became pregnant again after about four months. She feared his violence if she resisted.

U3 knew that O was in contact with his friend, YC, who was in Syria. Her evidence was that she did not know what was going on there, apart from a civil war. O hinted that he wanted to fight there. He told her that it was her duty as a Muslim wife to travel with him. She did not want to. O said that if she did not, he might take the children and marry another wife. He became more distant and when she was six months pregnant, he beat her until she fainted because she asked whether he had found another woman. O had a profile on a Muslim dating website. He explained that as a Muslim he was allowed another wife. Their son was born in May 2013. They were relatively happy after that. When the baby was about six weeks old, O went to Algeria. When he returned, he insisted that she wear gloves and cover her eyes in public. O went to Saudi Arabia in January 2014. On his return, he was fixated with going to Syria. He insisted that U3 go with him. There was a violent argument at her mother’s house and she never saw her mother again. O’s demands about Syria increased. U3 agreed to follow him there. ‘She was, in part, frightened that he would just take the children’ (paragraph 115).

O left the United Kingdom. He returned, saying that he had been deported to Algeria from Turkey, banned from Turkey for ten years and interviewed by MI5. He tried to leave the United Kingdom again, but was stopped. O threw her out of their flat in April 2014. O warned her not to be in touch with a friend via the internet. U3 felt conflicted about O. ‘It was like I had an obsession with him, like a sickness’ (paragraph 118). O did not come back. YC persuaded U3 to win O round by pretending that she would go to Syria. YC suggested that U3 join him in Turkey as a stepping stone to Syria. She booked an indirect flight to Turkey, intending to stay with YC. O was thrilled. They were reconciled. She travelled to Turkey with her children and was met by YC. She was pleased to see O. Her willingness to go to Syria meant a lot to him. U3 told O, however, on 20 August that she did not want to go to Syria. He told her that he had been stopped by someone from the Security Service. If she returned to the United Kingdom, she might have problems keeping the children. He could not stay in Turkey because of the ban. He then destroyed the children’s passports. He attacked her. YC’s wife intervened. O was emotional. If she went to Syria he would have no other women. She agreed to go, but only for the sake of their marriage. She did not agree with extremist values. As a young person she had promoted human rights. She knew ‘next to nothing’ about ISIL and had no wish to align with them.

She described increasing restrictions on personal freedom, especially for women, ‘her astonishment that there were slaves in Syria and the discomfort she experienced living in a society that allowed stoning and beheading’. In February 2015, O married a second wife. When U3 complained, he beat her with a metal rod designed for cleaning his gun, so badly that she was bruised for many weeks. He divorced the woman after a few weeks. This was a turning point. She realised that she no longer loved O. She contacted her mother and brother in July 2015. She was worried that if she died, her children would be brought up in an ISIL orphanage. She nevertheless told her mother that she was in Turkey (not Syria). She told her brother the truth and hoped he would tell her mother.

She gave evidence about other violence by O. Once he cracked a bone in her leg. She wanted to leave but it was difficult. She realised she was pregnant in March 2016. She was ‘devastated’, mainly because of concern about the child. She was appalled at news of ISIL attacks in Europe. She thought about leaving and resumed contact with her mother, who said that she could only help if U3 was in Turkey. In October 2016, O told her he had married a third woman. He said hurtful things to her. U3 continued to look for ways to leave. O was disillusioned with ISIL. He took two more wives. They all lived together. U3 was totally humiliated. Eventually he divorced the two new wives, and told U3 that they were going to leave. He found a people smuggler in October 2017. They were to travel separately, and have not seen one another since they parted in December 2017. She was tried in Az’az as an ISIL supporter and ‘acquitted’ (paragraph 135). As I have already said, Ms Harrison accepted in the course of her oral submissions in this court that U3 was not in fact acquitted, but that the charges were dismissed for lack of evidence.

She contacted a charity with a view to returning her children to the United Kingdom. She was then approached by ‘Lauren’ and ‘Matt’ from the Security Service. She co-operated by answering questions and wrote her first statement which she sent to them. It was written in difficult circumstances and parts were not accurate. Matt told her in April or May that she had been deprived of her citizenship. She contacted her current solicitors and eventually, in March 2019, the children left for the United Kingdom, with help from the Foreign and Commonwealth Office. She became friendly with M, and started to live with him. She left him briefly to flee to Turkey. When she failed, she returned to him. She applied for entry clearance, prompting decision 2.

SIAC also referred to U3’s third and fourth statements. She gave further details of O’s violence towards her. She had never looked for ISIL videos. She had seen recordings of executions on O’s phone when she was looking for other things. She denied knowing about ISIL’s atrocities when she left the United Kingdom for Turkey. She repeated her denial that she was aligned with ISIL in her fourth statement. She had assumed that Syria would be a stricter society, under Sharia law. She knew nothing about ISIL’s atrocities. She disagreed with the contentions that it was ‘likely’ that she was aware of events in Syria and ‘highly unlikely’ that she would not be interested in O’s motives for travelling. O made it difficult for her to look at electronic media, and she did not read serious newspapers. She did not accept that as a housewife and mother, she was supporting ISIL. She did not support ISIL.

U3 gave her oral evidence remotely, we were told. SIAC summarised her cross-examination in paragraphs 145-159. She accepted that she had said that O wanted to go and fight, but had not asked what he meant. She accepted that if O had succeeded in getting to Syria at his first attempt, she would have taken the children to join him. She agreed that ‘part of me wanted to follow him’. She had understood that it was against the law to go to Syria yet she had twice agreed to join O there. She had then refused. She accepted that there was a conflict between her first statement and the later professionally drafted statements; she had said that she went to Turkey with no intention of going to Syria but later, that she went to Turkey hoping to join O there and expecting to travel to Syria. ‘…she had always endeavoured to tell the truth but that she did not know how to tell her story’. She denied that ‘she went to Syria in the hope and expectation of travelling with O to Turkey’ (paragraph151). I think that this is likely to be a slip and that SIAC inadvertently reversed ‘Syria’ and ‘Turkey’ in this sentence.

U3 accepted that YC worked for ISIL. He worked as a taxi driver. She had ‘no idea’ who his customers were. She denied that going to stay with him was a step in a planned journey to Syria to support ISIL. She did not accept that she was an extremist who had chosen to expose her children to extremist ideas. She confirmed that she had started to study ‘A’ level law. She accepted that she had asked YC’s wife about life in Syria and the journey there. Despite being reminded that she had said that YC’s wife had told her about the declaration of the Caliphate, and showed a clip of Baghdadi’s declaration, that there would be an Islamic state which applied Sharia and would protect Muslims, she continued to deny that she had done any significant research before agreeing to travel to Syria with O and their children.

She was asked about her Twitter account. She assumed she had set it up, but O had access to it. Many of the tweets were about the importance of modest clothes. She accepted that she had tweeted or re-tweeted ‘Whoever dies without having fought or having resolved to fight has died following one of the branches of hypocrisy’. She could not remember how this had been sent, she claimed, but she had sent it or let O send it, to please him. She had ‘liked’ tweets in order to impress him. They included ‘And whoever his sins are plenty then his greatest remedy is Jihad’. This was from a scholar whom O respected. It was in Arabic script, and she could not read that at the time. She sent it to please O. She had also liked ‘When liberating your land, store ten bullets into your gun, one for the enemy and nine for the Traitors’. This was also to please O. She was asked about her relationship with two people. She described them as ‘Dawlah fanatics’. She denied that she shared their ideas. ‘Nevertheless, she remained in contact with them’ (paragraph 159).

In paragraphs 160-166, SIAC summarised the evidence of Dr Agnew-Davies. She adopted her two reports. SIAC described her experience and qualifications. She had ‘an impressive CV’. SIAC found her ‘a balanced and thoughtful witness’ who was familiar with the evidence in the case. Her answers were ‘considered and nuanced’. She fully understood her duties as an expert. In her first report, she concluded that O’s behaviour ‘amounted to a very severe, extensive coercive control through a number of dimensions’ (which SIAC then listed) (paragraph 162). She considered that U3 was ‘not ‘thinking freely’ when she went to Turkey. Her conduct was produced by irresistible instinctive drives and subconscious forces.

The ‘thrust’ of her evidence was that by ‘his extreme and sustained coercive and controlling behaviour’ O had so affected U3’s thinking that the desire to please him and win him back was the ‘dominant reason for many of her actions, particularly travelling to Turkey and, eventually, Syria. Any views she might hold about ISIL were peripheral as explanations for her behaviour’. Occasional contact with others did not lessen his power over her. SIAC observed that she was not able to take into account the CLOSED evidence in reaching her conclusions (paragraph 166).

SIAC then summarised the evidence of Nic Newman and Siobhan Coward (paragraphs 167-170). Mr Newman had concluded that stories about ISIL atrocities were rare in the British media before 24 June 2014. Ms Coward and her team responded with evidence about Google and YouTube searches using terms such as ‘ISIL’. They found 94 relevant reports before that date. There were 12 articles about a massacre of 1700 Iraqi soldiers near Tikrit on about 12 June 2014. Many reports were on mainstream sites. Some included ‘graphic images depicting beheadings, crucifixions and other gruesome killings’, or detailed descriptions of such images. 76 were published in the three weeks before 24 June 2014. Mr Newman’s second report responded to that material. He maintained in cross-examination that reporting of ISIL’s atrocities was relatively sparse.

SIAC also heard evidence from two ‘radicalisation experts’ who produced a joint report. They interviewed U3 remotely. Their conclusion was that ‘we do not assess [U3] as currently radicalised’. They did not think that it was ‘clear’ that she was radicalised before she left the United Kingdom. There were indications that she did not support ‘the move to Syria to the extent that [O] did’. Their view was that the evidence was more consistent with a view that U3 went to Syria because she was in an abusive relationship rather than because she was radicalised. There was no evidence that she had tried to radicalise her children when in Syria. In cross-examination they said that the tweets ‘could be properly explained without concluding that U3 was herself radicalised or ideologically aligned with ISIL’.

In paragraphs 174-175 SIAC summarised the evidence about U3’s children.

SIAC then made some findings about U3’s evidence. SIAC accepted her evidence, not challenged and corroborated ‘in significant respects’ that O subjected her to ‘serious and sustained violence’; and that it was ‘the dominant feature of the relationship’. SIAC also accepted her evidence about ten separate incidents of violence in the United Kingdom (paragraph 176). Her account of his violence after she left the United Kingdom was consistent with that account, and ‘powerfully corroborated’ by things the children had later said to U3’s mother and brother. SIAC described those incidents in paragraph 177. In paragraph 178, SIAC said that U3 was not merely subjected to violence, but was also subjected to other conduct by O, described in the evidence of Dr Agnew-Davies, such as ‘threats, surveillance, degradation, shaming, isolation from family and friends, false imprisonment’, control of many important aspects of her existence, ‘gaslighting’, and his mood swings. This led to ‘pathological bonding’. SIAC held that U3 was not exaggerating this behaviour.

SIAC made findings about the best interests of the children in paragraph 179. ‘On balance’ SIAC accepted that U3’s return to the United Kingdom would be in their best interests.

In paragraph 180, SIAC referred to written submissions after the hearing in which U3’s counsel had identified ‘nine factual issues’, which, he submitted, SIAC had ‘institutional competence’ to decide.

The first was the impact which O’s behaviour had had on U3. SIAC said there was only one answer. ‘The impact of this behaviour on U3 was severe. SSHD did not suggest the contrary.’ His treatment ‘had a major impact on the way she made decisions for herself and her children’ (paragraph 181).

As SIAC pointed out, the second such issue posited ‘a binary choice’ between two mutually exclusive explanations, by asking whether U3’s motive for leaving the United Kingdom for Turkey on 20/21 June 2014 was as a result of her abusive relationship with O (that is, to stop him leaving for another woman) or to align with ISIL as a radicalised person. SIAC had no doubt that her experience of O’s violence, coercion and control ‘contributed materially’ to her decision to leave the United Kingdom for Turkey (paragraph 182). How that experience affected U3 was less clear. Her first statement was inconsistent with her later statements. SIAC found that her first statement was false, and was made to support a narrative that she had never wanted to go to Syria and had done so because she was forced by O (paragraph 183).

SIAC held that she had mixed motives for travelling to Turkey. In part, she wanted to convince O that she would go to Syria to dissuade him from going there with another woman. Her knowledge that O was banned from Turkey played a part. There was a chance that U3 could prove her commitment to him and that he would not be able to join her. SIAC found Dr Agnew-Davies’ evidence that such conflicting motives were common in women who were subject to behaviour like O’s ‘compelling’ (paragraph 184).

But that reasoning left out of account one important element ‘in the process of reasoning which led U3 to decide to travel, with her two children, from the UK to Turkey. We use the word “decide” deliberately’. Affected though she was by O’s conduct, she ‘was still able to make reasoned decisions for herself’. Travelling to Turkey would only show her commitment to O ‘only because, and insofar as, Turkey was a stepping stone to Syria’. She must have recognised, when she left the United Kingdom, ‘and in our view, on the evidence, did recognise - that she and her children might end up going to Syria and was prepared to take them there if necessary’ (original emphasis). When she left the United Kingdom, she had a ‘contingent intention to travel to Syria’ with them. ‘There is no doubt that she knew that O’s intention in travelling to Syria was to align with ISIL. She therefore intended, contingently, to offer support to him in that end, at least as his wife and as the mother of his children’ (paragraph 185).

The role played in U3’s motives by her desire to stop O going to Syria with another woman did not determine what, SIAC considered, was a key question for the Secretary of State: ‘the extent, if any, of U3’s alignment or support for ISIL.’ Even with the evidence about the nature of the relationship and its effect on U3, it was possible that an ideological commitment to, alignment with or sympathy for ISIL played some part in her decision to leave for Turkey. In order to make a decision about that, the Secretary of State needed ‘to make an assessment taking into account all the evidence, OPEN and CLOSED.’ SIAC added that the authorities ‘indicate unequivocally that the relevant judgment is, in the first instance, for SSHD. Only if that judgment can be shown to be vitiated by a public law error can we interfere with it’ (paragraph 186). SIAC said it would consider whether there was any such error in due course.

Issue (iii) was what U3 knew about ISIL on 20/21 June 2014, and to what extent any such knowledge was significant. SIAC was not persuaded by the evidence of Mr Newman. A person of average intelligence with access to the internet who wanted to find out about ISIL and its activities ‘would have discovered [the relevant reports] with ease’ (paragraph 188). U3 had been studying ‘A’ Levels in Media Studies and Law. She is an intelligent woman. The evidence did not support a suggestion that O had so isolated U3 that she could not have done the necessary research. U3 had done internet searches ‘at earlier stages’. It was not likely that she would have done no relevant research before leaving the United Kingdom, particularly given her contingent intention to travel to Syria with her two children (paragraph 189).

The fourth question was whether U3’s motive for leaving Turkey to travel to Syria was O’s abusive behaviour or to align with ISIL. SIAC observed that this question had the same flaw as the second question. Its premise, again, was two alternatives which were assumed to be mutually exclusive. There was strong evidence that O’s violent outburst, threats, and destruction of the children’s passports played a part in her decision. The question for the Secretary of State, however, was whether, and to what extent, she was ideologically aligned with ISIL when she did (paragraph 190). SIAC considered that there were elements of the OPEN and CLOSED evidence which could suggest such alignment. U3 had herself said that she went to Syria without being forced. That was an important starting point for an analysis of the degree of risk U3 might pose. U3 accepted posting material on her social media accounts which could suggest such alignment. SIAC reminded itself that it was not its function to decide the significance of those posts, but did not accept that U3 knew as little as she said she did about the ideology and activities of ISIL. SIAC considered that the Secretary of State could properly have taken that material into account, with ‘other evidence in the case, as suggesting an ideological alignment to ISIL.’ The weight to be given to this material was a matter for the Secretary of State. There was also CLOSED evidence which, in SIAC’s view, the Secretary of State could properly regard as containing elements suggestive of alignment with ISIL when U3 left Turkey for Syria (paragraph 191). The question was not the extent to which SIAC considered that U3 was ideologically aligned with ISIL when she left Turkey for Syria, but whether it was open to the Secretary of State to make such an assessment (paragraph 193).

The fifth question was whether U3’s stay in Syria was voluntary throughout. Her evidence strongly suggested that she was very soon disillusioned with ISIL, that there was no practical way for her to leave safely and that if she could have done so, she would have. Nevertheless there were elements in the CLOSED evidence which, the Secretary of State could properly consider, cast doubt on that evidence. Again, the question was not what SIAC concluded, but whether it was open to the Secretary of State to conclude that U3 stayed voluntarily (paragraph 193).

Question (vi) was whether U3 posed a risk to national security at the date of decision 1. SIAC recorded that U3’s counsel accepted that, for this question ‘the appropriate respect slider is moved in favour of SSHD/security services and such conclusions afforded more weight’. He submitted, nevertheless, that ‘in this case, it will depend, potentially decisively in U3’s favour, on SIAC’s view’ about questions (i)-(v) and (vii)-(viii).

SIAC said that counsel was right to accept that ‘this question (which is fundamental to the deprivation appeal)’ was one on which, according to the authorities, SIAC was obliged to give great weight to the Secretary of State’s assessment. SIAC did not agree, however, that the assessment depended on SIAC’s view about questions (i)-(v), let alone questions (vii) and (viii), which related to things which happened after decision 1. Questions (i)-(v) were ‘the building blocks on which the national security assessment is based’. In each case, the answer depended on an assessment of the OPEN and CLOSED evidence. ‘If the lawfulness of the national security assessment depended on SIAC’s assessment of the key factual elements underlying it, SIAC would in effect be substituting its own judgment for that of SSHD’ (original emphasis). The appeal could only succeed if after SIAC had considered all the evidence for itself, including the oral evidence, SIAC concluded that the Secretary of State’s assessment left out something material or was ‘vitiated by some other public law flaw’.

Question (vii) was whether U3 was ‘radicalised’ when she left ISIL territory in the autumn of 2017. U3 submitted that this was an issue which SIAC had institutional competence to decide and that no particular respect was due from SIAC to the approach or conclusions of the Security Service or of the Secretary of State ‘other than to carefully weigh them’. SIAC repeated its view of limits of its role (paragraph 196).

In paragraph 197, SIAC described the five strands of evidence on which the Secretary of State’s view that when she emerged from its territory, U3 was still ideologically aligned to ISIL, was based. SIAC doubted whether four of those, alone, were a sufficient rational basis for that conclusion. The CLOSED evidence was important. The Secretary of State had to assess the evidence as a whole. The question was whether against the background of the Secretary of State’s view that U3 was ideologically aligned with ISIL when she entered its territory the strands were ‘a rational basis for concluding that U3 remained ideologically aligned when she left ISIL-controlled territory’ (paragraph 198). SIAC concluded that they were.

The eighth question was whether U3 was ‘radicalised’ at the date when the appeal was heard. Counsel submitted that this was a question which SIAC could decide for itself, without any particular respect for the view of the Secretary of State (paragraph 199). SIAC rejected that submission. The question, rather, was whether the Secretary of State was entitled to reach the view that, at that date, U3 was still aligned with Islamist extremism (paragraph 200). SIAC noted that U3 was in very frequent remote contact with her children. There was no evidence that she tried to radicalise them. SIAC doubted whether that was entitled to much weight, as U3 must know that, even in relation to conversations which were not overheard by adults, there would be a significant risk that the children might, intentionally, or not, reveal what she had said (paragraph 201). The Secretary of State had taken ‘an avowedly precautionary approach’ to this assessment. ‘The authorities indicate that she is entitled to do so’. This assessment was based on the preceding assessments ((ii)-(vii)) and on a generic assessment about the likely alignment of individuals who have spent time in ISIL-controlled territory. SIAC could not say that the Secretary of State’s assessment was ‘irrational or …otherwise flawed in the public law sense’ (paragraph 202).

The final question was whether, as at the date of the hearing, U3 posed a risk to national security. U3’s counsel again accepted that this question ‘might be one where the appropriate respect slider is moved in favour of the security services/SSHD and such conclusions afforded more weight’. He nevertheless submitted that in this case, the answer would depend ‘potentially decisively in U3’s favour’ on SIAC’s view on questions (i)-(viii) (paragraph 202).

SIAC’s response was that, applying Begum and P3, it could only interfere with the Secretary of State’s conclusion on this question on public law grounds. That approach would be ‘entirely undermined’ if the answer could depend ‘whether decisively or not’ on SIAC’s view on ‘the key factual inputs to the assessment’. Each of those inputs was a matter for the Secretary of State. SIAC could only interfere if the overall assessment was flawed on public law grounds (paragraph 204).

SIAC had reviewed the Secretary of State’s assessment that on 17 April 2017, U3 posed a risk to national security. The aspect which most concerned SIAC was whether the Secretary of State had ‘properly taken account of’ the strong and unchallenged evidence about O’s violent controlling behaviour. Paragraph 6 of the assessment was put into OPEN as a result of the rule 38 process (ie disclosure). It read ‘We do not dispute that [U3] had a difficult relationship with [O]’. U3’s counsel submitted that this was ‘a wholly inadequate description’ of O’s behaviour towards U3. If the assessment was based on a view that the relationship was no more than difficult, the assessment had a false basis and the Secretary of State had left something relevant out of account. SIAC agreed that the word ‘difficult’ was not a fair or adequate description. Many couples have ‘difficult’ relationships, but very few ‘are dominated, as was the relationship between U3 and O, by repeated, serious violence, humiliation, degradation and control’ (paragraph 208).

SIAC did not, however, accept that the description was a public law error, for three reasons (paragraphs 209-212). First, it was clear that the author of the assessment was responding to U3’s second witness statement (which contained the substance of her case). The contents of that statement had not been left out of account. The structure of the assessment made clear that the Secretary of State maintained her assessment, despite U3’s evidence. Second, even if fully accepted, U3’s evidence about the nature of the relationship did not determine the separate question whether or not she was aligned to ISIL at the relevant times and during the relevant period. The Secretary of State considered that U3 had minimised her own alignment with ISIL. That was not inconsistent with an acceptance of her evidence about the relationship, given her evidence that she had chosen to leave Turkey. Third, in those circumstances, the description given to the relationship did not matter. Read as a whole, the Secretary of State’s assessment was that, despite U3’s evidence, she was assessed to be ideologically aligned with ISIL at the relevant times. That was based, in part, on the Secretary of State’s rejection of U3’s evidence that she did not at that stage know about ISIL’s ideology or atrocities. Significantly, SIAC added, ‘We also reject U3’s evidence in that regard’.

‘These points would have been enough on their own to convince us that there was no public law error in [the Secretary of State’s] description of the relationship as “difficult”.’ SIAC added that it was also significant that the Secretary of State’s national security witness had attended the hearing, and had not been cross-examined by U3’s counsel or by the Special Advocates. That would have been an opportunity for this point to be put and answered (paragraph 213).

In the light of the conclusions which SIAC had reached on the factual questions (ii)-(v) and ‘on the basis of OPEN and CLOSED evidence as a whole,’ SIAC concluded that the Secretary of State could rationally assess ‘in April 2017’ that U3 had been ideologically aligned with ISIL when she left for Turkey and after that. The Secretary of State could rationally assess that she posed a risk to national security. It did not matter whether SIAC agreed with that view. ‘The deprivation decision was not vitiated by any public law flaw’ (paragraph 214). SIAC dismissed appeal 1 (paragraph 215).

SIAC then considered appeal 2. For the reasons it had given in relation to questions (vii) and (viii), it concluded that it was rationally open to the Secretary of State, having considered all of the OPEN and CLOSED material, to assess that U3 posed a danger to national security at the date of decision 2 and at the date of the appeal hearing. SIAC had to balance that assessment (‘rather than any that we might undertake ourselves’) against the article 8 rights of U3’s children, giving appropriate weight to the Secretary of State’s view about where the balance lay (paragraph 216). It would be in the best interests of the children for U3 to return to the United Kingdom. Those interests were a ‘primary consideration’. But those interests were ‘firmly outweighed’ by the Secretary of State’s assessment that U3 posed a risk to national security. SIAC’s conclusion would have been the same, if, contrary to its view, U3’s article 8 rights had also been engaged. That conclusion was supported by SIAC’s assessment that the current arrangements for the care of the children, though not ideal, provided ‘a substantial degree of protection for their interests’ (paragraph 218).

SIAC dismissed appeal 2 (paragraph 219).

U3 applied to SIAC for permission to appeal. She argued, in appeal 1, that SIAC had erred by applying a public law approach to ‘the factual aspect’ of the national security case. Her further grounds in appeal 1 were that

She had four grounds of appeal from the decision on appeal 2.

On 8 April 2022, SIAC gave permission to appeal on ground i. in each appeal but refused permission on the other grounds of appeal. SIAC rightly recognised that while it had reached a clear view on the issue raised by ground i., it raised a significant point of law of wider importance. U3 renewed her application to this court on the remaining grounds of appeal. I refused that application initially. U3 asked me to reconsider that refusal. I did so, and gave limited further permission to appeal. I accepted, in effect, that the question of SIAC’s powers on a section 2B appeal might not be a binary choice between a public law and a merits-based approach, and that there might be scope for a hybrid approach, but it was not clear how it would apply in practice. I gave U3 permission to rely on her other grounds; not for the purpose of re-arguing her case before SIAC, but in order to help this court to decide how, in practice, if she were to lose on ground i., those parts of the appeal which were challenges to the assessment of national security were to be distinguished from other issues, and in what respects and to what extent, SIAC might adopt a different approach. I also gave permission to JUSTICE to intervene by written submissions, supplemented by oral argument.

I have indicated the parties’ broad positions in paragraph 3, above. Rather than lengthen this long judgment further by reciting them in detail in a further section, I will consider Ms Harrison’s main submissions in the next section of this judgment, in the course of my discussion of the issues.

There are three broad issues.

The parties to this appeal disagreed about whether Lord Reed’s account of SIAC’s functions on a section 2B appeal was necessary to the decision or not. In Begum SIAC decided three preliminary issues in the section 2B appeal. Its decision on issue 2 was the subject of a successful application for judicial review to the Divisional Court, on the basis that its legal approach to issue 2 was wrong. The Secretary of State then appealed to the Supreme Court. The Supreme Court was required to decide what the correct legal approach on a section 2B appeal is. It could have done so either by giving a general account of SIAC’s role on such an appeal, or by considering, only, the narrower question of the correct approach to preliminary issue 2, that is, whether or not the Secretary of State had complied with his extra-territorial human rights policy. The Divisional Court’s reasons for overturning SIAC’s public law approach to that discrete issue were that the section 2B appeal was a full merits appeal, and that SIAC had wrongly treated it as raising an issue of public law. The Supreme Court could have confined itself to deciding that, whatever the nature of a section 2B appeal, the question whether the Secretary of State had complied with his policy was a question which SIAC was right to treat as being reviewable only on public law grounds. However, the Supreme Court could also have gone further, as it did, and explained why this court’s approach to preliminary issue 2 was wrong, by setting out the general nature and limits of a section 2B appeal; an explanation which reinforced its reasons for upholding SIAC’s approach and rejecting the approach of this court. I consider that the Supreme Court’s account was necessary to its decision. In any event, whether or not it is obiter, this account is highly persuasive. I would follow it, whether or not this court is strictly bound by it. I observe, nevertheless, that because the Supreme Court was not considering an appeal against a decision of SIAC about the merits of a section 2B appeal after a full hearing, its description of SIAC’s functions on such an appeal is necessarily at a high level, and not prompted by any detailed analysis of an actual decision on the merits after such a hearing. The description of SIAC’s functions, therefore, cannot, on any view, be considered to be a binding and exhaustive account of those functions.

The parties’ submissions concentrated on two linked issues in particular: the relationship between a section 2B appeal and an application for judicial review, and whether and if so, to what extent, SIAC has a power or a duty to make findings of fact.

References to applying judicial review principles on an appeal are apt to cause and have caused confusion in SIAC, for two reasons. First, such language may be an imprecise proxy for the bases on which, on the authorities, an appellant can challenge the Secretary of State’s assessment of national security on an appeal (I will refer to this as ‘the Rehman approach’). Rehman does not tie those bases expressly to judicial review, but to the need to comply with Chahal (see paragraph 46, above), whereas Lord Reed’s focus was the discretionary nature of the decisions which are challenged on a section 2B appeal. This confusion may partly stem from Lord Reed’s endorsement of SIAC’s public law approach to the second preliminary issue concerning the Secretary of State’s human rights policy (see paragraphs 56 and 71(iv), above). But that is intrinsically a public law question, and sheds no light on SIAC’s overall functions on a section 2B appeal. Second, and more importantly, the authorities do not require SIAC to apply the Rehman approach to other aspects of a section 2B appeal, apart from a challenge to the assessment of national security. It is not necessary for us to decide whether (as SIAC held) there are other bases, available in judicial review, on which a national security assessment can be challenged, as this question does not arise here.

U3 argued that SIAC’s approach in this case, and in other judgments to which we were referred, is incoherent because SIAC, particularly in the other cases, has stated that it must apply judicial review principles to section 2B appeals, while at the same time taking into account material which was not before the Secretary of State when she made the impugned decision, such as later national security statements which respond to evidence from an appellant, and exculpatory material which comes to light after the decision and which the Secretary of State therefore cannot have taken into account when she made the decision. I agree with U3 that if a section 2B appeal is the same as an application for judicial review, such an approach would be incoherent, and wrong. But a section 2B appeal is not an application for judicial review, as I will now explain.

A section 2B appeal is not an appeal from a court. It is an appeal from an administrative decision-maker who has particular expertise and institutional competence in questions of national security. Section 2B confers an apparently unqualified right of appeal. This is relatively unusual in this field, as my survey of the rights to challenge immigration decisions and other decisions which have been certified shows. My starting point is that, subject to the constraints exerted by the authorities, that right of appeal gives SIAC power to decide questions of fact and of law. If it were for me to decide this point, I consider that Lord Woolf’s observation (see paragraph 35, above) that such a power on an appeal is unusual is inconsistent with the later reasoning of this court in Subesh (see paragraphs 13-17, above). That is supported by section 5(1)(b) and 5(6) of the 1997 Act. Parliament has tended, in this field, to define or limit rights of appeal. Even section 4, before its repeal, was a limited right of appeal. If, after considering the appeal, SIAC considered that one of the two conditions in section 4(1)(a) was met, it was required to allow the appeal. ‘[I]n any other case’ it was required to dismiss the appeal. The 1997 Act now clearly distinguishes between appeals and statutory reviews. The provisions about the latter require SIAC to apply ‘the principles which would be applied in judicial review proceedings’. The contrast between those review provisions, section 4 (now repealed) and section 2B is eloquent. It is, in any event, also clear from Michalak that an appeal and an application for judicial review are fundamentally different, in that an appellant can challenge the merits of a decision, whereas a claimant in an application for judicial review cannot do so. For those reasons, SIAC did not err in law in refusing to treat the appeal as an application for judicial review.

The parties agreed that, on an appeal, the appellate body can take into account evidence which post-dates the impugned decision, as long as that evidence is capable of casting light on events before and at the time of the decision. The procedure for deprivation is inherently unfair, on two accounts: the appellant has no input into the decision and the decision is based in part on material which she never sees. SIAC was very aware of those factors (see paragraph 41, which I summarise in paragraph 101, above). Those factors suggest that the appeal which Parliament has given an appellant is a forum in which such decisions should be examined as meticulously as is possible, again within the constraints established by the authorities (a view supported by the terms of section 5(6)(a) of the 1997 Act: see paragraph 27, above). SIAC has an important role in scrutinising all that evidence independently, with the invaluable help of the Special Advocates, who press for as much disclosure to the appellant as possible, and who rigorously test the CLOSED evidence. In the course of a hearing, SIAC sees more intelligence materials than the Secretary of State will have done: she will normally only see a ministerial submission perhaps with some annexes, as, SIAC recorded, she did in this case. SIAC’s reference in this case to its ‘powerful microscope’ in paragraph 32 (see paragraph 89, above) was apposite. Moreover, SIAC will also have a potentially wide range of evidence from the appellant (as in this case), which will not have been considered by the Secretary of State, either. The appeal may well be the appellant’s first and only opportunity to influence a decision-maker. SIAC is entitled to expect that by the time of the appeal it will have been given an updated national security statement which takes on board the evidence which the appellant has served for the purposes of the appeal. It is also entitled to expect that the Secretary of State will make available a national security witness who is immersed in the detail of the case, and who is ready to be cross-examined by the OPEN representatives and by the Special Advocates. These considerations mean that the arguments about incoherence, which rest on the confusion I describe in paragraph 169, above, fall away.

The authorities do not in any way restrict the evidence which SIAC is required to consider on an appeal; nor do the relevant procedure rules. Two factors are highly significant in a case like this. First, the appellant had no opportunity to give her own account, or to influence the Secretary of State’s thinking, before the deprivation decision was made. Second, because this was an appeal, and because of its special expertise, SIAC was in a uniquely good position, within the limits expressed in Rehman and Begum, to review and rigorously to test the assessment of an appellant’s risk to national security. It was bound to do so on the basis of all the evidence which it had received on the appeal, OPEN and CLOSED. It is also significant that, while some appeals to, and reviews in, SIAC do not concern any provisions of the ECHR, the statutory procedure was itself created in order to comply with the procedural component of relevant provisions of the ECHR, as Lord Hoffmann observed in Rehman.

None of the authorities prevents SIAC, on a section 2B appeal, from making findings of fact on the balance of probabilities. I accept Ms Harrison’s submission that it can. SIAC, which is not bound by the rules of civil evidence, is in a good position to make such findings, as, in most cases, and in this case, it will have heard oral evidence and cross-examination, and will have considered a range of documentary evidence, OPEN and CLOSED, most of which will not have been before the Secretary of State when the Secretary of State made the deprivation decision. There are many examples of cases in which SIAC may or must find facts. Statelessness is often an issue on a section 2B appeal. In cases where statelessness is an issue, SIAC will usually hear expert evidence about foreign law. SIAC has a duty to make findings on that evidence, and to decide whether or not the effect of the deprivation order was to make the appellant stateless. Nothing in the authorities prevents SIAC from making relevant findings of fact about that issue. Indeed, E3 obiter (see paragraph 31), above, supports that view. Nor does anything in those authorities prevent SIAC from making findings of fact in an appeal which raises issues under section 40(3) (if such an appeal were to be certified). Again, having heard the witnesses, it has a duty to make such findings. SIAC can, and must, also make relevant findings of fact on any issues about the ECHR which may arise on an appeal, and, as in this appeal, in relation to section 55.

In other words, if it considers that it can, and that they are appropriate, it may make findings of fact which may be relevant to the assessment of national security, as long as it does not use those findings of fact as a platform for substituting its view of the risk to national security for that of the Secretary of State. Those findings of fact are the material to which SIAC must apply the tests set out in Rehman and Begum when it considers a challenge to the Secretary of State’s assessment of national security. For example, one of SIAC’s tasks is to allow the appeal if there is no factual basis for the assessment. That would mean, in my judgment, that if there were evidence, which SIAC accepted, which showed, for example, that, on the balance of probabilities, U3 had never been to Syria, and that the Secretary of State had mistaken someone else for her, SIAC’s duty would be to make that finding and to allow the appeal. As long as it respects the limits expressed in Rehman and Begum when it considers a challenge to the Secretary of State’s assessment of national security, SIAC can also make whatever findings of fact it considers it is able to make on the evidence, on the balance of probabilities, and which, in its expert judgment, it considers that it is appropriate to make.

There is a wide area in which it is for SIAC, as the specialist court, to judge whether it can, and whether it is appropriate for it to, make a particular finding of fact. SIAC is very well placed to judge whether it has enough material on which to make a finding of fact, and, as importantly, whether, given the limits imposed by Rehman and Begum, it is also appropriate for it to make such a finding. SIAC’s task is then to see whether the Secretary of State’s assessment can withstand its view of the evidence, provided that it remembers that the Secretary of State’s assessment is frequently not solely or even primarily based on specific findings of fact made on the balance of probabilities, but is an evaluation, based on a range of different types of material, many of which are not evidence for the purposes of litigation. Sometimes, as in the example I have just given, there will be a pivotal finding, such as that the appellant travelled or stayed somewhere, which SIAC is in position to contradict. Very often, as in this case, however, those core facts will not be in dispute. The contentious aspects of the assessment, in a case like this, will often turn on questions of motivation, which will depend on inferences, or on similar issues. SIAC is not a jury deciding whether or not a defendant intended to commit the crime charged on an indictment. Nevertheless, in an appropriate case, SIAC may judge that the evidence enables it to make a finding about a person’s motivation. If it can, it may do so. SIAC has to bear in mind, however, that if it considers that such a finding is possible and appropriate, the use which SIAC can make of such a finding is limited. The finding is part of the factual picture to which it must apply the tests in Rehman and Begum. Since a finding about motivation necessarily involves an assessment based on inferences from primary facts, SIAC must bear in mind that its finding about motivation cannot displace a contrary assessment by the Secretary of State, as long as there is material which would rationally support such a contrary assessment.

The conceptual distinction between an application for judicial review and an appeal is only a starting point. The reasoning in Rehman, Subesh and Begum illustrates a similar point. The point is that the language conferring a statutory right of appeal is not the sole guide to the functions of the appellate body, which will depend on other factors. Those include which body Parliament has entrusted with the power to make the decision which is challenged, whether that body is a court, the nature and subject matter of that decision, the relative expertise of the decision maker and of the appellate body, and the particular issue or issues which the appellate body has to, or may, decide on an appeal. Authority which binds this court is clear that, on a section 2B appeal, Parliament has conferred the power to make two discretionary decisions on the Secretary of State, and that the decision whether deprivation is conducive to the public good depends on knowledge and expertise which even SIAC lacks, and political accountability, which SIAC obviously also lacks.

In so far as the appeal was a challenge to the Secretary of State’s assessment that, at all relevant times, U3 was a danger to national security, SIAC understood those authorities as requiring it to review the assessment by the Secretary of State of the interests of national security within the limits described in those authorities, rather than substituting, for that of the Secretary of State, its view about the danger posed by U3. The Secretary of State’s assessment was the starting point for SIAC’s consideration of the appeal. U3 (and the Special Advocates in the CLOSED hearing) took on the burden of displacing that assessment, and of showing that it was mistaken, within the limits described in the authorities. In closely examining all the evidence to see whether the assessment was mistaken, SIAC had to, and did, bear in mind, among other things, that an assessment is not based on findings, on the balance of probabilities, that particular things have happened, or that particular things have been thought, but on an overall consideration of all the material, so as evaluate future risk.

In summary, SIAC can, and in some cases must, make findings of fact based on its own assessment of the evidence on the appeal. As long as it respects the limits of the Rehman approach, it may make whatever findings of fact it considers, in its expert judgment, it is able to make and which are appropriate in the appeal it is considering. A judgment SIAC makes about whether to make a finding, or not, is unlikely to be susceptible to challenge on an appeal on a point of law.

There are three issues.

The first issue is whether SIAC erred in law in making findings of fact at all. I have already explained why it did not do so. SIAC had four statements from U3, and she was cross-examined. SIAC was right to consider that it could make relevant findings, based on its assessment of all the evidence on the appeal. It is not the ‘alter ego’ of the Administrative Court. In this case it made, and was entitled to make, at least 13 important findings, which include findings of fact, based on the evidence as a whole, and which it explained fully.

The second issue is whether, in the respects relied on by U3, SIAC nevertheless erred in law in its approach to some aspects of the evidence, either by making wrong findings of fact or by declining to make findings of fact. The apparently unlimited statutory right of appeal is qualified by the reasoning in Rehman, a deportation case, which has more recently been applied to section 2B appeals by Begum. The potential tension between the approaches of Lords Slynn and Hoffman was resolved by Lord Reed in Begum (see paragraph 60 above). I do not consider that the observations of Lord Hodge in Pearce (see paragraph 72, above) undermine that resolution in this context. This court is bound by that reasoning. This was an appeal in which U3 directly challenged the Secretary of State’s assessment of the risk to national security which she posed (see paragraphs 7 and 11 of SIAC’s judgment, summarised in paragraphs 76 and 78, above).

I will now consider U3’s main criticisms of SIAC’s approach to the nine questions posed by her counsel. Ms Harrison accepted in her oral submissions that the ‘assessment of overall risk’ was for the Secretary of State. She argued, however, that ‘the building blocks, the facts’ were matters which SIAC ‘can and should decide’. She added that ‘on our case, that includes alignment and the underlying factual conclusions’. Whether or not U3 was aligned with ISIL was exactly the sort of factual question which is routinely considered in the criminal courts. It was a precedent fact. SIAC’s overall approach had led it not to consider the evidence as a whole. SIAC’s approach was inconsistent because it had made some findings which were adverse to U3, while declining to make other findings. There had to be a principled basis for drawing the line. The distinction which SIAC had made between motivation and alignment was artificial. The Secretary of State’s assessment had to be established by evidence.

SIAC’s reasoning had been made without any reference to the expert evidence or explanation of how it had taken that evidence into account. Nor did SIAC deal with the evidence about U3’s later conduct after she left ISIL-controlled territory, which was capable of casting a different light on the issues. SIAC had not considered the report which expertly analysed the risk U3 posed. If SIAC had considered the evidence properly, it should have concluded that U3’s motive was based on the dynamics of her relationship with O. This point could be considered on an appeal on a point of law. If not, U3 was making a reasons challenge, because SIAC had not explained why it rejected the expert evidence. She was also making a rationality challenge, as the only reasonable conclusion was that U3 was under the coercive control of O.

If U3’s account and her expert evidence had been accepted by SIAC, that would have removed the underpinning of the Secretary of State’s assessment, and the Secretary of State would be required to reconsider it. Ms Harrison nevertheless accepted that if SIAC was entitled to conclude that U3 had mixed motives, the Secretary of State’s assessment was reasonable.

Ms Harrison did not criticise SIAC’s findings in paragraphs 181 or 189, one of which was favourable to U3’s case (see paragraphs 140 and 143, above). Ms Harrison was particularly critical of paragraphs 186, 192, 193, 195 and 204 (see paragraphs 144, 146, 147, 148-149 and 154, above). The issue which is common to the criticisms of these paragraphs is that SIAC abdicated its function too soon, and should have made relevant findings of fact (or that it made the wrong finding and/or did not explain its finding). I will now consider those paragraphs of SIAC’s judgment. I bear in mind that a premise of this submission was that on certain key issues, SIAC could make findings of fact which would then enable it to undermine the Secretary of State’s assessment of the risk to national security which U3 posed.

In paragraph 193, SIAC accepted that U3’s evidence was a strong basis for her case that she was disillusioned with ISIL and would have left its territory if she had been able to. SIAC did not make a finding about that, instead relying on elements of the CLOSED evidence which the Secretary of State could properly consider cast doubt on that. The question was not what SIAC concluded, but whether it was open to the Secretary of State to conclude that U3 stayed of her own free will. I consider that SIAC could have made a finding of fact on this question, if it had thought it appropriate to do so. If and to the extent that it considered that it could not, in principle, make such a finding, it erred in law. But I also consider that any such error of law was immaterial, for the reason which SIAC gave in the last sentence of paragraph 193. This question was part of the Secretary of State’s overall assessment of risk, and even if SIAC had found that U3 had not freely stayed in Syria, it could not lawfully have overset the Secretary of State’s assessment on the basis of such a finding, either on its own, or in combination with others. On the authorities, the question it had to ask, and did ask, was whether there was material which rationally supported the Secretary of State’s assessment to the contrary.

In paragraphs 186 and 192, SIAC expressed its conclusions about questions (ii) and (iv), which concerned U3’s motives for leaving the United Kingdom for Turkey, and for leaving Turkey for Syria, respectively. SIAC observed that both of those questions were flawed because they posited a binary choice between, in short, coercion and control by O and alignment with ISIL as mutually exclusive explanations for U3’s decisions. SIAC found, in relation to question (ii), that ‘there was no doubt’ that the former factor ‘contributed materially to’ her decision to leave the United Kingdom. Her motives, however, were mixed. She had a contingent intention to travel to Syria and to offer support to O, who intended to align with ISIL there. It was ‘possible’ that an ideological commitment to ISIL played a part, even if not the major part, in her decision to leave for Turkey. In relation to question (iv), SIAC said that there was ‘strong evidence’ that O’s violent behaviour and threats ‘all played a part’ in U3’s decision to leave for Syria. SIAC did not make a finding about her motives for leaving Turkey, instead saying the elements of the OPEN and CLOSED evidence could suggest alignment.

SIAC’s answer to question (ii) is a finding of fact about U3’s motives which is damaging to her case. SIAC was, in principle, entitled to make that finding. This was one of nine factual issues which SIAC was invited to decide. They were framed as questions which SIAC had ‘institutional competence’ to decide. If SIAC had institutional competence to decide any of those issues, and decided them, an attack on SIAC’s findings of fact is both paradoxical, and very difficult to make good.

Ms Harrison submitted that this finding was either irrational, or insufficiently explained. First, I do not consider that, as a matter of logic, the evidence of Dr Agnew-Davies compelled a different conclusion. That would only have been the case if, as Ms Harrison argued, SIAC was obliged to accept that evidence. I do not consider that SIAC was obliged to accept that evidence. First, a premise of that evidence was that U3’s account was truthful. SIAC, having heard her give evidence, found, and was entitled to find, that her evidence was untruthful in some respects. Second, as SIAC observed (paragraph 166), Dr Agnew-Davies had not seen the CLOSED evidence. Third, SIAC said (paragraphs 184 and 185) that it found her expert evidence about conflicting motives in such cases ‘compelling’; but that that evidence left out of account one element in U3’s reasoning (see paragraphs 142 and 143, above). Fourth, where a tribunal has expertise of its own, it is not obliged, as a matter of law, to accept the uncontradicted evidence of an expert: see Kentucky Fried Chicken v Secretary of State for the Environment [1978] EGLR 139.

Nor do I consider that SIAC was obliged to explain its reasoning in any more detail than it did (see Volpi v Volpi, paragraph 10, above). For this purpose, SIAC, which heard the relevant evidence and made some findings of fact, is the first instance judge. The observations in Volpi v Volpi have even greater force when an appellant, as here, tries to suggest that her criticisms of the approach of a fact-finder show, not that the fact-finder was ‘wrong’ about the evidence, but that the fact-finder erred in law. At first sight, there is an inconsistency in U3’s criticisms of the findings of fact which SIAC did make. In any event, if it had been necessary, as a matter of law, for SIAC to have explained its approach to that evidence, it did sufficiently explain it, by referring to the fact that Dr Agnew-Davies had not seen the CLOSED evidence, and that her approach left one important matter out of account.

The premise of this criticism is that SIAC was bound to accept that evidence as the only explanation for U3’s decisions, and, more fundamentally, that SIAC could, on these two questions, having accepted that evidence, substitute its view of U3’s motives for the assessment of the Secretary of State that U3 was, at the relevant times, aligned with ISIL. A similar premise underlies Ms Harrison’s criticisms of SIAC’s approach to the evidence of Drs Silke and Brown. Their evidence also relied, to varying degrees, on their own assessment of matters which were essential elements of the Secretary of State’s assessment. Even if SIAC had accepted all that expert evidence, that would not have entitled it to find that the Secretary of State’s assessment was flawed on any relevant basis. First, the Secretary of State’s assessment did not depend and could not depend on an informed analysis of U3’s psychological state at any point, or on making findings on the balance of probabilities about what U3 did, or did not, do, or about her motives. It was based, rather, on all the evidence, OPEN and CLOSED, which was relevant to the question whether or not she was aligned to ISIL. Second, it was not, as a matter of law, SIAC’s function to make such an assessment for itself, whether or not that assessment was informed to any extent (or none) by the evidence of the experts. These factors all mean that the weight which SIAC could give to this evidence was, necessarily, limited. SIAC expressly found that there was material to support both the relevant assessments of the Secretary of State. I do not consider that SIAC was required to say more about these issues than it did, but even if it was, any such error was immaterial, because whatever view SIAC might have formed about whether or not U3 was aligned with ISIL at the relevant times, that view could not have displaced the assessment of the Secretary of State.

Questions (vi) and (ix) concerned the risk to national security posed by U3 at the date of decision 1, and at the date of the hearing. SIAC’s conclusions about questions (vi) and (ix) were in paragraphs 186 and 204. U3’s counsel expressly accepted, in relation to those questions, that ‘the appropriate respect slider’ is or might be moved in favour of the Secretary of State, and that her relevant conclusions ‘afforded more weight’. He nevertheless submitted that the answers to questions (vi) and (ix) depended ‘potentially decisively’ in U3’s favour, on SIAC’s answers to the other questions which he had posed. SIAC said that if the lawfulness of the national security assessment depended on SIAC’s assessment of the key factual elements underlying that assessment, SIAC ‘would in effect be substituting its own judgment for that of’ the Secretary of State. SIAC could only interfere if there was a public law flaw in the assessment (paragraph 195). SIAC made a similar point in paragraph 204.

SIAC was right to say that questions (vi) and (ix) were questions for the Secretary of State, with which SIAC could only interfere on limited grounds. I consider that SIAC was right to reject the submission that the answers to question (vi) and (ix) could depend, to any extent, on SIAC’s answers to the other questions. Nor do I consider that SIAC was bound to make its own findings on those questions. As it happened, SIAC decided that it could, and it did, make findings of fact in relation to questions (ii) and (iii). It did not err in law in doing so. I do not consider that SIAC erred in law in declining to make findings of fact on questions (vi) and (ix). Even if it could have made such findings of fact, they could not have displaced the Secretary of State’s assessment, absent one of the errors identified in Rehman. So if I am wrong, and SIAC should have made any such findings of fact, any such error was immaterial. I should consider, here, Ms Harrison’s criticism based on SIAC’s approach to evidence adduced by U3, other than her own, which post-dated decision 1. I reject the submission that SIAC wrongly ignored that evidence. It referred to it in paragraph 64 (see paragraph 116, above), explaining that it had not summarised it, and why, though it corroborated U3’s evidence, it did not take matters further. Finally, I reject the submission that SIAC’s conclusions on any of these questions were irrational.

The third issue is Ms Harrison’s challenge to the second national security assessment (see paragraphs 206-213 of the judgment, summarised in paragraphs 155-157, above). The premise of this criticism is that SIAC should have treated this part of the appeal as an application for judicial review of the second OPEN national security statement. That premise is wrong, as I have already explained. The question for SIAC, rather, was whether the Secretary of State’s assessment that U3 was and remained a danger to national security could be challenged on the grounds described in the authorities. The issue, therefore, was not whether the Secretary of State had misdirected herself by understating or misunderstanding U3’s evidence about her relationship with O. SIAC carefully considered this point, and gave three good reasons for rejecting this criticism. First, it could not be said that the Secretary of State had left U3’s evidence out of account. Second, even if her evidence was accepted, it could not be decisive of the relevant question. Third, in all the circumstances the description which was given to the relationship did not matter. I do not consider that SIAC erred in law in rejecting this criticism.

In addressing the nine questions which, U3’s counsel suggested, it had ‘institutional competence’ to decide for itself, SIAC said that it was observing the limits established in the authorities. I consider that it did not err in law in seeing those limits as the touchstone. That does not necessarily mean that it could not have made more findings of fact than it did, but, first, given those limits, the primary judgment about whether any such findings were appropriate was for SIAC. I would not second-guess, in an appeal on a point of law, the exercise of that judgment. In any event, even if SIAC erred in law in not making further findings of fact, any such errors were immaterial, given the limits which it was bound to, and did, observe. In every respect in which it was asked to substitute its own assessment for the relevant assessment of the Secretary of State, it refused to do so, but, instead, reviewed the Secretary of State’s assessment to see whether it had a rational basis. It was right to do so.

For those reasons, I would dismiss this appeal.

I agree.

I also agree.