JCK (Botswana) v Secretary of State for the Home Department

The Appellant is a national of Botswana born in 1989. He appeals with permission against the decision of the First-tier Tribunal (Judge Brooks) dated the 30^th August 2023 to dismiss his appeal on protection and human rights grounds.

The Appellant claimed asylum on the 3^rd August 2022. His case was that he is at risk of persecution in Botswana by fellow members of the Herero tribe. He claims that this persecution will be for reasons of his religious belief (the Appellant is a ‘born-again’ Christian) and for his imputed political opinion/race (his rejection of tribal custom in refusing to take the leadership position he was expected to inherit from his deceased father).

Protection was refused by way of a letter dated the 10^th February 2023. This being a claim made after the 28^th June 2022, it was one to which sections 31-39 of the Nationality and Borders Act 2022 (NABA 2022) applied. The letter therefore addressed the Appellant’s claim in stages, corresponding to sections 33, then 31, then 32 of the Act. The Respondent accepted that the Appellant is a ‘born-again’ Christian, and that he is a member of the Herero tribe, but rejected the contention that he would be at risk for any of the reasons claimed; the Respondent concluded that if he were, the Botswanan government could provide sufficient protection from hostile members of the Herero tribe, and/or the Appellant could relocate within Botswana.

The Appellant appealed to the First-tier Tribunal which dismissed his appeal, finding that he had not shown, on balance, that he had any characteristics which could cause him to fear persecution, that he did in fact fear such persecution, or that on the lower standard of proof he objectively faced a real risk of harm in Botswana.

The Appellant now appeals inter alia on the grounds that the Tribunal erred in its approach to varying standards of proof in a NABA 2022 appeal. I am grateful to the parties for their submissions on this point, and it is here that I begin.

The standard of proof to be applied in protection claims has, for some thirty-five years, been whether there is a “real and substantial risk” of persecution should the claimant be removed to his or her country of nationality/former habitual residence (Footnote: ¹). Otherwise expressed as a “reasonable degree of likelihood”, or a “reasonable chance” this is a lesser standard than the balance of probabilities otherwise applicable in civil law. The rationale for the application of this lower standard was twofold: the difficulty that the putative refugee faces in proving his claim, and the potentially grave consequences of getting it wrong. Judges and decision-makers in this jurisdiction have become accustomed to applying this standard, but now we must get used to a new paradigm.

For any protection claim made on or after the 28^th June 2022, judges and other decision makers must now comply with the part of NABA 2022 dealing with the ‘Interpretation of the Refugee Convention’. Section 30 provides that sections 31-35 of the Act shall apply for the purpose of determining whether a person is a refugee. Section 31 defines persecution, covering actors and acts. Section 33 is concerned with defining the reasons for persecution; section 34 with whether the home state can be considered to provide sufficient protection, and section 35 addresses internal flight. Here we are concerned with section 32, which sets out the approach to be taken when analysing whether the claimant’s fear is ‘well-founded’. It reads:

The Explanatory Notes to NABA 2022 assert [at page 348] that this “provision establishes a clear two-limb test for assessing whether an asylum seeker has a well-founded fear of persecution and will raise the standard of proof which an asylum seeker must satisfy for certain elements of the test”. The two limbs are of course delineated by the two standards of proof that must be applied at different stages in the process.

The exercise begins at section 32(2) which mandates the decision maker to determine, on the balance of probabilities, whether the asylum seeker “has a characteristic which could cause them to fear persecution” for one of the five reasons set out in Article 1(A)(2) of the Refugee Convention. Next, applying the same standard, consideration must be given to whether the claimant “does in fact fear” such persecution. On a plain reading of the text, the decision maker is therefore to apply the civil standard of proof to whether there is a ‘Convention ground’, and to whether the claimant has a genuinely held subjective fear.

Moving on to the second limb, sub-sections (3) and (4) together provide that where those questions are answered in the affirmative, the decision maker must then go on to conduct the future-focused risk assessment by applying the familiar ‘refugee standard’ of reasonable likelihood, looking first at persecution and then at the availability of protection. Finally, consideration must be given to internal relocation.

There are therefore in fact five questions hanging over the two limbs. Taking each question in turn, the following considerations should be applied.

Beginning at s32(2)(a), the decision-maker is required to determine, on the balance of probabilities:

The decision-maker is not required here to consider whether the characteristic, or imputed characteristic, has in fact attracted persecution, or whether it will do so in the future. The simple question is whether the claimant has a protected characteristic which could cause them to fear. In many cases this will be straightforward. Applicants fearing persecution because they have an outwardly obvious characteristic such as their gender or race will have little difficulty in discharging the burden of proving this matter on a balance of probabilities. Other, more opaque, characteristics could be more challenging to discern. Whether someone is gay, or holds a particular religious or political belief is not something that can be seen with the naked eye, or by making windows into souls. It is something that must be evaluated on the evidence in the round, but care should be taken not to automatically reject, at this first stage, a claimed characteristic by reference to the overall credibility of the claim. The focus must be on the characteristics. Thus in this case, the Respondent accepted that the Appellant is a Christian, and a member of the Herero tribe, even though he rejected the claim that he had suffered harm as a result. Put simply the question raised by s32(2)(a) is whether, taking the claim at its highest, there is a Convention reason.

The answer is not, however, always going to be that simple. There are certain classes of applicant for whom it will be necessary to consider the country context in order to answer the question at s32(2)(a). Sub-sections 33(2)-(4) NABA 2022 require a member of a particular social group to demonstrate not only the innate characteristic possessed by, for instance, an ethnic group, but they must also demonstrate that they have “an identity in the relevant country because it is perceived as being different by the surrounding society” (Footnote: ²)^. That ‘social visibility’ test can only be applied by looking carefully at the country background material (both expert and general), which is, at this stage, to be assessed on the balance of probabilities. Decision-makers must however be mindful that they are not here evaluating risk.

Section 32(2)(b) requires the decision-maker to also determine, on the balance of probabilities:

Given the reference here to section 8 AITCA 2004, decision makers could be forgiven for thinking that this question is all about the credibility of the claim. Certainly, this sub-section was, during the Act’s passage through parliament, widely trailed as applying the civil standard to claims of past persecution, and in its published guidance Assessing credibility and refugee status in asylum claims lodged on or after 28 June 2022 (Version 13.0) the Home Office instructs caseworkers that “this is an assessment of the material facts which the claimant is presenting to you”. That is not however what sub-section 32(2)(b) says.

The provision simply requires the decision maker to consider whether the asylum seeker “does in fact fear”: it is what is otherwise referred to as the ‘subjective fear’ test. Assessing that fear is a discrete exercise from assessing whether past events occurred. Consider a claimant who has been horribly persecuted in the past but whose persecutors have now fallen from power: he could prove, on balance, that the material facts he has presented are true, but he may no longer in fact be afraid. Conversely it is well understood in this jurisdiction that claimants who are “in fact afraid” may seek to exaggerate, or even falsify, past events in order to prove their case. There will be cases in which the acceptance or rejection of historical facts presented by a claimant will inform the decision on whether or not he is “in fact afraid”. As it happens, this is one of them. That is not however always the case. Section s32(2)(b) asks decision-makers to consider a different question, and in doing so relegates the matter of ‘credibility’ to where it belongs in the refugee risk assessment: it can be relevant, but will not on its own be determinative.

The country background material is always going to be relevant here, because it provides context to assessing someone’s claimed fear. If a claimant is from, for instance, a particular ethnic group widely persecuted in his country of origin, that is going to be a relevant consideration when assessing whether or not he is in fact afraid. Again, it is important to remember that this is, at this stage, not an assessment of risk: this is an assessment of whether someone is afraid.

Now we are back in the familiar territory of evaluating objective risk, applying the refugee standard of proof in a rounded, holistic assessment of all of the evidence. Section 32(4)(a) requires the decision maker to determine whether there is a reasonable likelihood that if the asylum seeker were returned to their country of nationality/ country of former habitual residence:

It has been established that there is a subjective fear of persecution for a Convention reason, and now it is a matter of assessing the claim as a whole, in the context of the country background material, to evaluate whether there is a real risk of persecution. Persecution is defined, for the purpose of the Act, at section 31, and here the decision maker is asked to consider whether the actor of persecution is one of those listed at s31(1)(a)-(c), and then to evaluate whether the feared harm is sufficiently serious, or sufficiently severe.

Section 32(4)(b) requires the decision maker to determine whether there is a reasonable likelihood that if the asylum seeker were returned to their country of nationality/ country of former habitual residence:

This obviously requires regard to be had to the terms of section 34, which replicates in material terms the now revoked 4 (1)-(2) of The Refugee or Person in Need of International Protection (Qualification) Regulations 2006. In cases where protection is not in issue – for instance where the state is the actor of persecution – this should simply be recorded in the decision.

Finally, section 32(5) mandates that:

As above, this is not always an issue between the parties, and where it is, decision makers are familiar with the matters to be taken into account in determining whether there is a reasonable alternative to seeking international protection.

The proper approach to s32 is then to address each question expressly and sequentially. If a matter is agreed, that simply needs to be recorded by a single sentence. Addressing each question under a separate heading will aid decision-makers in identifying matters in issue between the parties, and setting out competing arguments and conclusions. Moving between the varying standards is an intellectual exercise which will require discipline, but it does not, cannot, change what decision-makers have always done in taking an ultimate, holistic view of the evidence. It is not possible to evaluate subjective fear – and in many cases Convention ground - without having some regard to the context in which that fear is said to arise. Decision makers will therefore need to consider the country background material twice over. In evaluating the matters raised in s32(2) that material will provide vital context to deciding whether, on a balance of probabilities, the tests are met. The decision-maker must then revisit that material afresh when considering s32(4), and apply the lower refugee standard of proof to the question of risk. This may prove laborious, but it is necessary in order to avoid conflating the matters of subjective fear and actual risk, or conversely, to avoid overlooking important context.

A separate question arises in respect of whether, if any condition of the first limb is not made out, decision makers should proceed to address the second limb at all. Sub-sections (3) and (4) of section 32 provide that where Convention ground and subjective fear are established, the decision maker must proceed to considering those matters arising under the second limb. As we shall see, in this case both Respondent and First-tier Tribunal proceeded to address risk albeit that they were not satisfied that both conditions in s32(2) were made out. That is, in my view, a sensible ‘belt and braces’ approach in circumstances where there are onward appeal rights.

Those principles in mind, I now turn to the decision of the First-tier Tribunal, and the Appellant’s grounds of appeal.

The First-tier Tribunal begins its decision by directing itself to s32 NABA 2022. It records [at §8], those matters placed in issue by the refusal letter. The Appellant is accepted to be Herero, and Christian. The Tribunal then directs itself, entirely properly, to the task ahead. With express reference to s32, it sets out those matters to be determined on the balance of probabilities, and then those matters to be determined on the lower standard of reasonable likelihood.

Then it makes its findings. They begin like this:

This is followed by a further 10 paragraphs in which the Tribunal identifies significant inconsistencies and deficiencies in the account, before saying this at paragraphs 28 and 29:

The conclusion is then expressed:

The appeal was thereby dismissed.

It is argued on behalf of the Appellant that the First-tier Tribunal erred in its approach to the five distinct questions in s32, overlooking concessions of fact made by the Respondent, and conflating matters so that ultimate question of risk on return was decided applying the civil, rather than the refugee, standard of proof.

There is certainly some merit in the first submission. It was clear from the refusal letter that the Respondent had accepted that there was a Convention ground in this case. Indeed there were two: the Appellant is Herero, and he is Christian. The Tribunal records that concession of fact at its §8, but by the time that it reaches its conclusions at §28, it expressly rejects the Appellant’s claim to have those characteristics:

It seems to me however, that this error has arisen not because the Tribunal rejects the evidence that the Appellant is a Herero Christian: if it does, it nowhere says so. Rather it has arisen because the Tribunal has given the word “could” in s32(2)(a) a rather wider reading than it should have done. The sub-section asks whether the Appellant “has a characteristic which could cause them to fear persecution”. As I note above, it is clear from the refusal letter that the Respondent has rightly read this to simply mean “is there a possible Convention ground in this case?”. The Tribunal on the other hand has read it as asking whether the Appellant has a characteristic which could credibly cause him to fear persecution. This leads the Tribunal to undertake a comprehensive critique of the Appellant’s narrative, before reaching the conclusion it expresses at its §28. I accept that the Tribunal has erred in importing questions of credibility into its assessment under s32(2)(a). The Respondent had accepted that there was a Convention ground here, and on a plain reading of the Act, I agree that this is all that s32(2)(a) is asking decision makers to consider.

In her oral submissions Ms Patel went further, suggesting that the Respondent had also accepted that the Appellant had a subjective fear: this, she said, was another concession that the Tribunal had gone behind. I am unable to accept that submission. The refusal letter was set out in tabular form, referring to various sections of the Act. Unfortunately the decision maker omitted to address s32(2)(b) directly. Had the letter been set out in the way I suggest above, in sequence, this important question would not have been overlooked. I am not however able to read into that omission an acceptance that the Appellant was “in fact afraid”. The rest of the letter, in which the entirety of the claim – his narrative and the objective risk – is rejected, makes perfectly clear that the Respondent did not believe that he was.

This brings me to the real nub of the appeal. In its analysis the Tribunal makes reference not only to internal inconsistencies in the account, but to the fact that the Appellant’s claims are unsupported by the country background material: for instance at its §20 where it highlights the evidence that conversion to Christianity is widespread amongst the Herero.Ms Patel submits that in doing so, the decision conflates the questions of Convention ground and subjective fear with the objective assessment of risk, made with reference to the country background material [s32(4(a)]. That is problematic, it is submitted, because in doing so the Tribunal has allowed the civil standard of proof to bleed into a question that can only be answered by applying the ‘refugee standard’ of reasonable likelihood: risk should only ever be determined with reference to the lower standard.

Having reviewed the decision with care, I reject the central premise of this ground, that the Tribunal’s approach impacted on the application of the appropriate standard of proof.That is because it is perfectly clear that whatever else it does when considering matters under s32(2), the Tribunal properly directs itself and applies the lower refugee standard when it comes to its ultimate conclusions on risk under s32(4): see its §29, set out above.There is nothing in the decision which suggests that this final risk assessment was treated as an inevitability, the Tribunal having already decided the s32(2) questions against the Appellant on a balance of probabilities. On the contrary the Tribunal is, in this respect at least, careful to delineate the two exercises. So although the Tribunal has erred in respect of whether there is a Convention ground in this case, it is not an error that is material to the outcome overall.

Turning to the remaining grounds, it is the Appellant’s case that the First-tier Tribunal decision is vitiated for a lack of procedural fairness. It is submitted that in four different passages the Tribunal drew adverse inference from inconsistencies it identified in the Appellant’s account, and that this was unfair because they were not points made in the refusal letter. The Respondent had not attended the hearing and although the Tribunal had directed itself to the Surendran guidelines (Footnote: ³) it had not sought to clarify certain matters with the Appellant. The result was that he had been denied the opportunity to respond to those forensic challenges.

One of the matters cited in the grounds arises at §20 of the decision where the Tribunal notes that many Herero have converted to Christianity. This was the context in which the Appellant’s claim to be facing persecution by his fellow Herero because he was a Born Again Christian had to be evaluated. It is simply not correct to say that this was a new point. It is a central plank of the refusal letter.

As to the remaining matters, these are all internal inconsistencies in the Appellant’s own evidence. It is true that they are not the same ones which formed the basis of the Respondent’s decision. I am not however satisfied that this means that the decision should be set aside.

Credibility was squarely placed in issue in the refusal letter. The Respondent had identified a number of issues with the account which led to it being rejected, and it is clear from the decision of the First-tier Tribunal that it preferred the Respondent’s analysis of the evidence to the one advanced by the Appellant. In those circumstances it is not incumbent upon the Tribunal to raise with the Appellant each and every matter which concerns it: Abdi v Secretary of State for the Home Department [2023] EWCA Civ 1455. The matters cited in the grounds are all quite obviously matters which should have been addressed either by way of examination in chief and/or submissions: they are glaring inconsistencies. The Appellant was represented by solicitors and experienced counsel who were clearly aware of these difficulties with the narrative. Furthermore I am satisfied that even if the Tribunal had not made the findings now criticised by Ms Patel, its decision would uncontrovertibly have been the same.

The grounds next take issue with the Tribunal’s treatment of 9 affidavits produced by the Appellant, and said to have been sworn by him at various times in Botswana after attacks on him by members of his family. The grounds submit that this documentary evidence was not taken into account, and that these affidavits were important because they established that the Appellant had not, and would not, receive adequate protection from the Botswanan government.

In fact the Tribunal does address the affidavits in its decision. It notes that at least two of these documents were couched in identical terms, despite the fact that they were said to relate to two different incidents. It was not minded to attach any significant weight to them. I do not see any error in that. The affidavits were, even if genuinely produced in Botswana at the stated times, only statements of the Appellant’s own evidence. They did no more than repeat the same evidence that he gives in this appeal. It is evidence full of internal inconsistency, but moreover it is evidence which derives no support whatsoever from the country background material. It is notable that in the skeleton argument produced for the First-tier Tribunal hearing the only background evidence that the Appellant is able to point to are reports of police brutality, something which had no bearing whatsoever on this case. Neither that skeleton, nor the grounds to this Tribunal, have identified any background material to support the central contention advanced by the Appellant: that 11 years after his father’s death his Herero relatives would try to kill him because as a Christian he does not want to be their leader.

The final criticism made in the grounds is that the First-tier Tribunal does not identify any particular place in Botswana that the Appellant could internally relocate to if he wished to avoid his family. Given the clear conclusion that there is no risk of harm in this case that is not material, and in any event the Respondent had himself identified Gabarone and Francistown as possible destinations. Again, neither the skeleton argument below nor the grounds to this Tribunal advance any coherent reason why either of these options would be unreasonable in the event that the claimed risk, contrary to all of the evidence and findings, was in fact real.

The appeal is dismissed.

There is an order for anonymity in this ongoing protection appeal.