WAS (Pakistan) v Secretary of State for the Home Department

The Appellant (‘A’) appeals from a determination of the Upper Tribunal (Immigration and Asylum) Chamber (‘the UT’) promulgated on 22 March 2022 (‘determination 4’). Asplin LJ gave permission to appeal on two out of four grounds of appeal. A did not renew his application for permission to appeal, so I will refer to the grounds for which he was given permission as ‘ground 1’ and ‘ground 2’.

On this appeal, A has been represented by Mr Fripp and Mr Lester, and the Secretary of State by Mr Holborn. I thank counsel for their written and oral submissions. Paragraph references are to the determinations of the UT or of the First-tier Tribunal (Immigration and Asylum Chamber) (‘the F-tT’) as the case may be, or to an authority if I am referring to an authority.

For the reasons given in this judgment, I would allow the appeal on grounds 1 and 2.

I have taken A’s immigration history from the determinations in his case. A entered the United Kingdom on 19 June 2012 as Tier 1 (General Student) Migrant. In December 2015 he made a timely application for further leave to remain as the spouse of a British citizen. The Secretary of State refused that application on 16 February 2016. A then made a protection claim on 19 October 2017 on the basis of his actual or perceived political opinion. He said that he was a member of the Muttahida Qaumi Movement (‘MQM’), an opposition movement in Pakistan. The further representations to which I refer in the next paragraph were based on an assertion that he had also been disinherited by his father.

I have also taken the procedural history from the determinations. The Secretary of State refused A’s protection claim on 18 April 2018. A appealed to the F-tT. The F-tT dismissed A’s appeal on 27 June 2018, in determination 1. The F-tT found (paragraph 11) that A was ‘an active member of MQM in Pakistan for approximately 3 years, from 2009 to 2012, prior to his arrival in the UK. [A’s] father is an active supporter of MQM…. In December 2015, [A] had been an active member of MQM for 6 years overall and MQM-London for approximately 3 ½ years’. The F-tT found that his credibility was damaged by the fact that he had not claimed asylum until October 2017, ‘in circumstances where his political affiliations have not changed…’. In paragraph 16, the F-tT accepted that A was a current member of MQM ‘and is likely to continue to be a supporter of MQM should he return to Pakistan’. The F-tT appears to have accepted his evidence that he was ‘an active member of MQM from 2009, going from house to house in support of their causes and MQM London from 2012 until 2017’ (ibid). The UT dismissed a further appeal on 14 May 2019. A made further representations in September 2019. In a decision dated 18 November 2019, the Secretary of State accepted that A’s further representations were a ‘fresh claim’ for the purposes of paragraph 353 of the Immigration Rules (HC 395 as amended), but refused that claim.

A again appealed to the F-tT. The F-tT dismissed his appeal in a determination promulgated on 5 February 2020 (‘determination 2’). The F-tT summarised the Secretary of State’s decision in paragraphs 12-21. That summary recorded (paragraph 13) that the Secretary of State relied on determination 1 and on the later decision of the UT (referring to Devaseelan [2002] UKIAT 00702). ‘Both tribunals had found that [A] was a low-level MQM member in London and would not be at risk on return to Pakistan’. In paragraph 24, the F-tT recorded that it had heard evidence from Abdul Hafeez (‘Mr Hafeez’), a member of the ‘MMQ UK Organising Committee’. The UT gave A permission to appeal against determination 2. In a determination promulgated on 18 September 2020 (‘determination 3’), the UT found a material error of law in the F-tT’s determination and set it aside. In short, the UT held that while the F-tT had taken into account the relevant passage in the expert report of Dr Bennett-Jones, it had ‘failed adequately to consider the risk…that a person who aligned himself with MQM-London may face from the Pakistani authorities (and MQM-Pakistan)’ (paragraph 20). The UT noted (paragraph 21) that there had been no challenge to the F-tT’s adverse credibility findings.

In paragraph 21 of determination 3, the UT directed that several findings of fact made by the F-tT should be preserved. Those were the F-tT’s findings of fact in paragraphs 29 to 32 and 53 about ‘threats to his family in Pakistan’, a finding in paragraph 39 about the receipt of financial support, and findings in paragraphs 33-37 and 51 to 52 about A’s involvement in MQM and his Facebook account ‘so far as they related to his sur place activities up to the date of’ the F-tT hearing. I describe those findings in more detail in paragraphs 9-18, below. The UT did not refer in determination 3 to the findings in determination 1.

The UT also decided in determination 3 that it would retain the appeal. The decision would be re-made ‘to determine whether [A] would face a real risk of persecution if removed to Pakistan on account of his actual or perceived involvement with the MQM-London’ applying the principles in HJ (Iran) v Secretary of State for the Home Department [2010] UKSC 31; [2011] AC 596. The UT would also consider, if necessary, the issue of internal re-location.

Paragraphs 29-32 are headed ‘Threats to [A’s] family in Pakistan’. In paragraphs 29-32, the F-tT described A’s evidence that he had been publicly disowned by his father in July 2019, that his father had, in January 2018, been approached by the authorities in Pakistan, and questioned about A’s activities, and that he had not been in contact with his family in Pakistan for over two years, except for his sister. The F-tT commented adversely on the defects in this evidence. For example, A had not produced any external evidence that his family had been harassed in Pakistan. In paragraph 32, the F-tT recorded that A’s evidence was that he had forgotten to say in his witness statements that his father had been questioned, even though he accepted that it was a very important part of his asylum claim. The F-tT said that it did not find A ‘to be a credible witness in this matter’.

In paragraph 53, the F-tT returned to this topic. The F-tT did not find A’s evidence about lack of contact with, and financial support from, his family in Pakistan credible. The F-tT accepted that the deed of disinheritance was issued and publicised by A’s father, but the F-tT did not accept that A had shown ‘on the lower standard of proof’ that this was because of adverse attention from the authorities.

Paragraphs 33-39 are headed ‘[A’s] political activities in the UK’. In paragraph 33, the F-tT described photographs of A’s activities in the United Kingdom. They were said to show him sitting round tables at various meetings, and meeting ‘Mr Hussain, the MQM founder’ (‘AH’) ‘at what appeared to be a lecture event’, which A said was on 29 March 2019. Two photographs showed A ‘as one of many in a large crowd’. A said that those photographs showed him taking part in a protest at 10 Downing Street in May 2018. The F-tT said that there no evidence to show that the photographs had come to the attention of the authorities in Pakistan. The F-tT noted that the photographs all post-dated January 2018, when his father had been questioned by the authorities.

The F-tT found that the photographs did not show that A had ‘any significant political or activist role’ in MQM in London. The F-tT summarised the evidence of Mr Hafeez as support for that conclusion. Mr Hafeez had ‘no knowledge of [A] making any political speeches or statements on behalf of MQM London’. The F-tT recorded an assertion in a letter from Mr Ehsan that A would be at risk in Pakistan, but also noted that ‘there was no further evidence from Mr Ehsan to support this opinion/statement’ (paragraph 34).

A had said that he joined MQM in Pakistan in 2009. He had come to the United Kingdom in 2012, but had not joined MQM-L until August 2016. A explained that he had been too busy with his studies, and had not been able to play an active part until 2016. The F-tT noted that this was after A had been refused leave to remain on the basis of his marriage to a British citizen (paragraph 35).

In paragraph 36 the F-tT referred to new evidence about A’s Facebook profile. This evidence was not in A’s bundle. It consisted of laminated cards which A had handed out at the hearing. His oral evidence was that the Facebook profile for ‘Vick Shapman’ was his; it was a nickname used by his friends and family. The F-tT accepted that there were pictures of A on this account and that it was his account.

There was no political content on this account. There was a birthday message to Mr Altaf Hussain (‘AH’) on 16 September 2019, but it was not political. There were other messages in Urdu, but as they had not been translated, the F-tT could not take them into account. A did not claim that the cards he produced at the hearing had any political content, or that they would put him at risk in Pakistan (paragraph 37).

A and Mr Hafeez both gave evidence that MQM had split into two factions, one in Pakistan and the other in London. AH, who had founded MQM, was in exile in London and could not return to Pakistan (paragraph 38). In paragraph 39, the F-tT recorded A’s evidence that after his student visa expired, he had not been able to work part-time. Since 2016, he had been supported by a friend with whom he lived rent-free. He had a very old laptop which he used for his social media activities. A did not name the friend. There was no evidence to confirm that a friend supported A financially. The F-tT did not find A’s evidence that he received no financial support from his family credible. As I have already said, the F-tT also made a finding about financial support in paragraph 53 (see paragraph 10, above).

In paragraph 51, the F-tT said that it had found that A had shown that he was a member of MQM London and a committee member of the North London unit. He was an active member ‘in the sense of administration and organizational activities’. He went to some meetings and some demonstrations. But, on the lower standard of proof, he was not ‘active politically in MQM-London’. That was confirmed by his own witness, Mr Hafeez, who was not aware of A’s having made any political speeches or public statements on behalf of MQM London.

The last preserved finding was in paragraph 52. The F-tT repeated that it had accepted A’s evidence that the ‘Vick Shapman’ Facebook account was associated with A, but that there was no evidence of any political content which would put A at risk in Pakistan.

The UT hearing was on 12 July 2021. Determination 4 was promulgated on 30 March 2022. A and the Secretary of State were represented by counsel. A relied on three witness statements. He gave evidence. The UT considered a range of in-country information about Pakistan, including reports and news articles. Appendix A to determination 4 is a list of those materials: nearly a page of reports, two pages of articles, and a book by Owen Bennett-Jones, ‘The Eye of the Storm’ (third edition). Dr Bennett-Jones also gave evidence at the hearing.

The UT gave a ‘brief overview’ of MQM (paragraphs 6-9). It was founded in 1984 as a party for the Urdu-speaking Muslims, known as Muhajirs, who had left India for Pakistan in 1947 at the time of Partition. It is a secular party based in Karachi, and it promotes the rights of Muhajirs. It also has bases in two districts of Sindh Province. Before the 2018 election, it had 50 out of 167 seats in the Sindh provincial assembly. In 2019 it had 21 seats. It was a divided political force. In 2016 it split into two factions: MQM-L, led by AH, who has lived in London ever since he fled there in 1992, and MQM-P, led by Farooq Sattar, and then by Khalid Maqbool Siddiqui. The split followed an anti-Pakistan speech by AH in 2016, which led to political violence in Karachi.

AH faces charges in Pakistan, though he benefitted from a partial amnesty in 2009. In 2018 an anti-terrorism court in Pakistan decided that AH had ordered another MQM leader, Dr Imran Farooq, to be killed in London in 2010. Three members of MQM were sentenced to imprisonment for life for that murder. It was reported in 2020 that the Federal Investigation Agency had put AH on its ‘most wanted terrorists’ list. MQM-L is not a proscribed organisation in Pakistan or in the United Kingdom.

MQM-L boycotted the 2018 general election. MQM-P won 7 seats and became a member of the governing coalition. It is the main opposition party in Sindh Province.

AH was arrested in London on charges of encouraging terrorism in Pakistan. He was recently acquitted. He had been the subject of two previous criminal investigations. Neither resulted in charges.

The UT summarised the evidence in paragraphs 12-43, starting with A’s (paragraphs 12-24). A had mentioned, for the first time, that his aunt and her son had been killed in a massacre of Muhajirs by Sindh nationalists. He had not been born at the time and had not had time to raise this before (paragraph 14). The UT returned to this point in paragraph 120 (see paragraph 53, below). He had been a member of MQM since 2009 and got ‘more heavily involved’ in 2016 after seeing the news and watching AH give a speech on 22 August. He was part of the North London Committee Unit. He organised events and social media. He received direct orders from headquarters (paragraph 15).

The party organises the dates and places of events. Sympathisers are asked for donations. A distinguished between sympathisers and members. He did not use social media for fundraising (paragraph 16). The UT described A’s evidence about screenshots on his social media accounts in paragraphs 18 and 19. He translated his 2019 birthday message to AH as ‘happy birthday brother – my leader’. He had had a Twitter account for two months (paragraph 20). MQM members mark particular dates, for example, Martyrs’ Day on 9 December. Events cannot be organised in Pakistan because people will be arrested by the Rangers (a paramilitary law enforcement organisation) (paragraph 21). He would never join MQM-P. He would continue to fight for the rights of the Muhajirs. His life would be at risk if he returned to Pakistan because of his political activities. He did not wish to hide his political identity (paragraph 22). He would never have said that he supported MQM-P (paragraph 24). He had a current passport which would expire in 2023. The Home Office had it.

The UT summarised the evidence of Dr Bennett-Jones in paragraphs 25-43. He is a freelance journalist and writer, and a former BBC correspondent and presenter. He has made a particular study of MQM. He gave oral evidence at the hearing. MQM had started by representing refugees to Pakistan and their descendants, but had become violent, and had, in turn suffered violence. It survived in part by fear. It is strong in the Sindh, where most of the refugees settled, particularly in Karachi and Hyderabad. Its support as a coalition partner is generally needed by the government. AH became estranged from the state after his speech in 2016 in which he denounced the army and the idea of Pakistan as: ‘a cancer’. MQM supporters then rampaged through Karachi. The state decided to break MQM and relied on the Rangers to do that.

The speech prompted, and the repression which followed it caused, two factions to break away from the MQM: MQM-L and MQM-P. MQM-P said that AH should be tried for treason. Since the speech, MQM-L members have faced risks in Pakistan. They have been killed without trial. They tend to be tried in military, not civilian, courts. Many political activists in Pakistan are held without being charged. His sources in Karachi say that the state targets MQM-L members but this is difficult to corroborate.

After Imran Khan became Prime Minister in 2018 there had been greater restrictions on the press. The 15-20 United Kingdom journalists writing for the media in Pakistan realise that their editors will not like it if they give prominence to MQM-L. ‘His instinct is that there are unreported cases’. In 2016 the United Nations Human Rights Committee (‘the UNHRC’) had expressed concern about 97 cases of enforced disappearances, nearly all of whom had been arrested in Karachi and many of whom were reported to be affiliated with MQM. The state did not have a dispute with MQM-P, so it was reasonable to infer that these were MQM-L supporters.

In 2019, the UNHRC reported that there had been 300 enforced disappearances since 2016. The true figures were probably significantly higher. Some of those who had disappeared were violent jihadis, but not recently. The numbers were down in recent years, probably reflecting the success of the campaign. Few MQM members in urban Sindh declare allegiance to AH. It is not possible for MQM-L to have an open presence in Pakistan. Those who express their support ‘face prison or worse’.

MQM-L and MQM-P have similar policies. MQM-P is not proscribed in Pakistan. The authorities can control it. He did not know how many members each faction had. MQM-P had broken with AH ‘and effectively accepted a degree of state control in exchange for their lives’.

The authorities are now better at monitoring social media. A state organisation known as Inter-Services Public Relations (‘ISPR’) can identify foreign posts on social media, and hand over an article by a journalist to Pakistan Inter-Services Intelligence (‘ISI’). The organisation can identify Facebook posts and social media posts form a foreign source. ISPR is primarily aimed at journalists in Pakistan and abroad. If a pro-AH tweet is posted, and its author cannot be identified, the state would ask an informer in MQM-L. ISPR is primarily aimed at journalists in or outside Pakistan. One of his own Tweets, Dr Bennett-Jones was aware, was picked up by ISPR.

There was no evidence that the authorities had lists of the members of MQM-L or MQM-P. They have people working in the organisation. A person who regularly attends demonstrations outside the High Commission is probably known by the authorities. The authorities would definitely know if MQM-L organised a seminar. They probably would not know everyone who attends, but would know those who attended regularly.

The risk to a member of MQM-L would depend on whether he had come to the attention of the ISI. The risk would depend on the member’s level of activity rather than on his seniority. A was from Hyderabad. It was plausible that he made contact with MQM-L in London. Even though he was not very senior, as a committee member with a profile on social media, he was likely to have come to the attention of the ISI. If they have not done so already, someone in MQM-L was likely to pass on this information to the ISI. It was likely that the authorities would be waiting for A at the airport. That there was no arrest warrant would make no difference.

Dr Bennett-Jones knew that members of MQM-L passed on information to the authorities. He had spent 20 years talking to people in intelligence, in MQM and to political leaders. He did not know how many people suspected of involvement with MQM had been returned to Pakistan. If A was not detained at the airport, he would have to move to another part of Pakistan. He would have to go to Lahore and start again. He would have to be careful. It would be difficult to get a job. He would be picked up by the ISI. He could save himself by giving information about MQM-L, denouncing AH and switching to MQM-P. If not he would be tortured or killed. If he continued to post messages supporting MQM-L, it would be ‘game over’.

All the MQM-L supporters whom Dr Bennett-Jones knows have switched to MQM-P. He gave one example, but not much information about it. No-one in Pakistan who was in his ‘right mind’ would say they were in favour of AH. Only very hard-core members would admit their support for MQM-L.

The Pakistan High Commission review passport applications. Applications from MQM-L members would definitely be of interest to the authorities. Although this could not be proved, Dr Bennett-Jones thought that if A needed to apply for a passport, he would come to the attention of the authorities. Those who were arrested or prosecuted by the authorities were known as ‘target killers’. A could face a claim that he was a target killer even if that was not true. Information about the number of MQM-L target killers arrested would not be public. Most are processed by military courts. Some are tortured and killed. Cases would only be publicised if they were high profile. The position was chaotic and those arrested and charged as target killers might be associated with the killings with which they were charged. It was hard to say whether ‘target killer’ was an accurate description.

The UT summarised the Secretary of State’s submissions in paragraphs 44-58 and A’s in paragraphs 59-62.

The Secretary of State relied on a specific response to information about MQM (‘the RTIR’). The RTIR showed very little evidence of adverse action against members of MQM-L. The Secretary of State accepted that A’s motives for sur place activities could not be decisive, but argued that they were relevant to the risk he might face on return, since, the less sincere those activities were, the less likely it was that A would persist in them in Pakistan. Much of the evidence of Dr Bennett-Jones was speculative. A was a low-level member of MQM-L and it was not likely that the authorities would know about him. Those at risk were people who were suspected of violent terrorism.

A submitted that the question concerned the risk to A given the unchallenged expert evidence that anyone who declared support for AH or aligned themselves with MQM-L might be targeted for their actual or imputed political views. A asked the UT to find that he had been a member of MQM since 2009. He was committed to AH. He was loyal to a person whom Pakistani authorities see as a terrorist.

The UT’s assessment is in paragraphs 63-148 of determination 4. The UT accepted, against the background evidence, that there were good reasons why Dr Bennett-Jones’s sources had not wanted to be identified, and why he did not want to identify them (paragraph 72). The overall picture presented by the background evidence supported significant aspects of his evidence (paragraph 74). The UT decided to give weight to his evidence, which, in any event, had not been significantly challenged in cross-examination (paragraph 75).

The background evidence showed that in 2019, 2020 and 2021, the authorities arrested and killed MQM-L members whom the authorities described as ‘hitmen’ or ‘target killers’, and those with significant positions in MQM-L. Most of those arrested, if not shot and killed, confessed to ‘very serious offences’ (paragraph 81). In 2021, an acquittal of an MQM worker was reported. There were also reports of convictions and significant prison sentences, and of further acquittals, one concerning several prosecutions for ‘21 hate speech cases’. The overall picture suggested that ‘those perceived as criminals and described as “workers” within MQM-L were being targeted by the authorities and prosecuted’. ‘Worker’ seemed to mean ‘member’. The evidence suggested a fall in disappearances and violent conflict. The situation was ‘far more nuanced’ than the RTIR suggested.

The UT was satisfied that the authorities were responsible for human rights abuses, including detention without charge, extrajudicial killings and enforced disappearances. Corruption was also a problem in the criminal justice system. There were many reports of killings of MQM-L workers, and of missing persons, who were also likely to be people linked with MQM-L in Sindh province. The numbers seemed to be going down. The authorities wanted to eliminate AH and his power base. MQM-L workers ‘may’ be perceived as ‘target killers’ because they support AH, who is seen as a terrorist. They might be genuinely suspected of committing serious crimes. The UT was satisfied that the RTIR did not reflect ‘a true picture of arrests and convictions’.

The decrease in the numbers of those who openly identified with MQM-L and those whom the authorities therefore identified as a threat was explained by the threat of violence which caused MQM supporters to distance themselves from AH. The UT accepted that a person could save himself by switching allegiance, because the authorities did not want to crush MQM-P, but rather to crush AH and his supporters. There had been a decrease in state violence but there was clear evidence that the authorities were still interested in MQM-L.

There was support for the expert’s view that support for AH was a risk factor. That risk could be obviated in some cases if the person concerned renounced AH or gave information to the authorities. It was not an option open to genuine supporters of AH, however (see HJ (Iran)). The phase ‘target killers’ could be used both to describe genuine criminals and those who refused to denounce AH. Some ‘target killers’ were arrested on false charges. The UT accepted that those who were identified by the authorities as associated with MQM-L as members supporters and ‘workers’ were at risk. The UT summarised its conclusions about this part of the case in paragraph 103.

The UT then considered the evidence of Dr Bennett-Jones about the authorities’ knowledge of activity by MQM-L in London. There was some indirect support for his evidence. AH’s presence in London made monitoring of activities in London more likely. It was plausible that the authorities would ‘invest resources in identifying genuine supporter/members overseas’. They ‘would go to lengths’ to ensure that AH’s power base ‘did not gain traction in Pakistan’. It was reasonably likely that if the authorities were aware that a person was or was perceived to be involved in pro-MQM-L activities in London that person would be targeted and at risk of persecution. Such a person would be expected to renounce MQM-L and to give information to the authorities. Someone who was not a genuine supporter could reasonably be expected to renounce MQM-L. Providing information in exchange for safety was fraught with risk and uncertainty. To expect someone to do that was contrary to the spirit of the Refugee Convention and of the European Convention on Human Rights (paragraph 107). I understand the UT to have meant that this was so, whether or not the person was a genuine supporter.

The key to interest from the authorities was whether or not the person concerned was, or was perceived to be, a supporter of AH or of MQM-L because they had done things which were reasonably likely to have come to the attention of the authorities, either as a result of monitoring in the United Kingdom by those authorities, and/or from monitoring of media activity. What determined risk was not whether the person’s activity was described as ‘low’ or ‘high’, but whether activity in support of MQM-L or of AH had come to the attention of the authorities. The more activity there was, the more likely it was to be detected by the authorities. A person could be at risk even if he had no genuine political motives: but if there was an issue about the level of activity, his motives were a relevant consideration.

There was no evidence that the authorities had a list of MQM-L members or supporters. There was no evidence of such a list in the background material or in the evidence of other MQM-L supporters. The UT did not accept the expert’s evidence about MQM-L members giving evidence to the authorities in order to curry favour. The UT found that his evidence that informers identified those responsible for media posts was ‘similarly speculative and insufficient to discharge even the lower “real risk” standard’, although the UT did not rule out that it could happen. There was no evidence about or from returnees about information being provided to the authorities by members of MQM-L. No such members had given evidence. The UT took into account the evidence about this in A’s 12 September 2019 witness statement (paragraph 110).

It was reasonably likely that the authorities intended to monitor MQM-L activity in London. They were likely to attend meetings of which they were aware if they were able to attend covertly. There was no evidence which ‘establishes’ that ‘all’ meetings/events would be known to the security services, or ‘establishing’ the size or place of meetings or whether covert monitoring was possible, nor whether meetings were open to the public. There was no evidence about various aspects of MQM-L meetings (paragraph 111). There was ‘insufficient evidence’ from which the UT ‘could reasonably draw conclusions about the level of and mechanics of monitoring in the United Kingdom’. The UT would not have expected the expert to know about this, but would have expected evidence from senior members of MQM-L about this. The UT ‘reasonably’ inferred that MQM-L would be ‘extremely keen’ to protect its members and that senior members would be willing to give evidence, in private if necessary, about this in order to give ‘a coherent and detailed account of a person’s role within the organisation clearly explaining how and why that person’s activities would likely come to the attention of the authorities’ (paragraph 112).

The UT had no doubt that the authorities were able to detect social media posts. The evidence did not ‘however establish that it was reasonably likely that the authorities have the resources or the capability to detect all posts which mention AH or MQM-L, let alone identify the individuals responsible’. The expert’s evidence was that the main target of monitoring was journalists. Detection was more likely if the person was posting material as a journalist ‘and/or on behalf of MQM-L’ (paragraph 113).

If a person had come to the adverse attention of the authorities as a result of MQM-L activity, then, whatever his motivation, he would be at risk from the authorities. Even if a genuine supporter’s activities had not come to the attention of the authorities, he might be at risk, following HJ (Iran). Relocation would not be possible as the risk was from state actors. The Secretary of State did not argue otherwise (paragraph 114).

The UT summarised its conclusions on this aspect of the appeal in paragraph 115. In paragraph 115.b it said that a person might come to the adverse attention of the authorities by attending an MQM-L meeting or event which had been monitored by the security services or High Commission. The authorities were reasonably likely to monitor meetings/events if they were aware of them and monitoring was practicable. It was reasonably likely that public demonstrations were monitored. Not every post in favour of AH/MQM-L was reasonably likely to be detected. Whether detection was reasonably likely would depend on all the circumstances (paragraph 15.c). A genuine supporter/member of MQM-L might be at risk on return even if his activity had not come to the attention of the authorities (paragraph 15.d).

A tribunal assessing an argument that an appellant has, or is reasonably likely to, come to the attention of the authorities would expect evidence covering the five topics which the UT listed in paragraph 116: details of any activities and how they were said to have come to the attention of the authorities; full details of all meetings or events a person has attended; whether details were public; whether events were public or not, and whether MQM-L regulated attendance; and whether a person played a role at a meeting such as being a speaker, which was reasonably likely to attract attention. Support for MQM-L and for AH could properly be characterised as a political opinion (paragraph 117).

In Part E of determination 4, the UT explained why it had decided to dismiss A’s appeal. The UT started this part of determination 4 by listing, in paragraph 119, the preserved findings of the F-tT. That list did not include the contents of paragraph 39 of determination 2, which while recording A’s evidence, was described in determination 3 as a ‘finding’ (see paragraph 13, above). In paragraph 120, the UT said that A had been found to be ‘a witness lacking in credibility’. The UT found expressly that A had ‘exaggerated his involvement with MQM-L’. The UT was not satisfied that he was politically involved with MQM before he came to the United Kingdom in 2012. The UT expressly did not accept that he became a member before 2016,although it was prepared to consider that he might have supported MQM before the split in 2016. That was supported by Facebook posts before 2016. The evidence did not, however, support ‘the kind of political commitment to the party suggested by’ A. The UT expressly found that A had joined MQM-L in 2016 before he claimed asylum in October 2017. The UT found that ‘much of [A’s] evidence is exaggerated or unsupported and that he has used his membership of MQM-L to bolster his claim.’ He claimed at the hearing, for the first time, that his aunt and nephew were killed in 1988. The UT found that ‘this is a recent fabrication designed to embellish his account’.

In paragraph 121, the UT said that its starting point was that A was a member of MQM-L and a committee member of the North London unit. The F-tT had found that he was an ‘active member in the sense of administration and organisational activities’. He went to meetings and to some demonstrations. The UT had analysed the evidence about A’s role by reference to the F-tT’s findings and to the evidence it had heard (paragraph 122). In paragraph 123, the UT listed the features of A’s role about which he had given evidence. In his oral evidence he had said that he did not organise meetings, but that he made sure that people went to them. He had said that he asked for donations from ‘sympathisers’, but there was no evidence to support that.

In paragraph 125, the UT described the evidence which ‘AF’ (a member of the MQM-L Organising Committee) had given to the F-tT. AF did not attend the UT hearing. That evidence was that AF had no knowledge of any political speeches or statements made by A on behalf of MQM-L. The UT quoted a passage in determination 2 which summarised AF’s evidence about A’s role. AF’s evidence was not consistent with A’s evidence. A denied in his oral evidence that he was responsible for organising events.

The UT referred to a witness statement from ‘SM’ dated 7 January 2020. SM said he could not attend the hearing because he was travelling abroad. SM described himself as ‘the unit [sic] in charge of MQM North London Unit’. He described the unit committee of eight, one of whom was A. The UT quoted his evidence about one of A’s activities in paragraph 127 and summarised more of his evidence in paragraph 128. SM’s description of A’s activities differed from AF’s and from A’s (paragraph 129).

In paragraph 130, the UT referred to several letters from MQM-L, the most recent of which was dated 13 June 2021. The UT quoted that letter and said that ‘All the letters are written in similar terms’. The letters and the evidence did not help the UT to assess the extent of A’s activities. It explained why in paragraphs 131 and 132. A’s evidence was that he had been actively participating in the affairs of MQM since 2012 and that he had attended events since 2014. The UT did not accept that. There was ‘no cogent evidence’ of activities in 2014. The UT found some of his evidence difficult to understand.

A did not explain in his witness statement how he helped to raise funds. He gave evidence about it at the hearing. A’s account was ‘wholly unsupported by cogent evidence’. MQM-L did not mention fund-raising, although one witness did. It would have been helpful to hear evidence from a ‘high-level member of the MQM-L’ (paragraph 133). A’s list of his roles and responsibilities described ‘a vast array’. There was some overlap in the evidence of witnesses and the MQM-L letter, but there were also internal and external inconsistencies. Much of A’s evidence and that of witnesses was unsupported. The evidence of his role was ‘nebulous’. The lack of ‘consistency, detail and coherence’ led the UT to conclude that the evidence was not reliable. The UT was not able to make proper findings about A’s role (paragraph 134).

The UT knew that A sat on a committee with seven others. There was no evidence about place or frequency of meetings. There was no evidence about whether people who were not on the committee went to the meetings. It was not clear whether the meetings were public or not. The UT did not find that A had shown that his role as a committee member was reasonably likely to come to the attention of the authorities. There was no evidence of social media posts in the public domain identifying A as a member of the committee (paragraph 135).

In paragraph 136 the UT described the general activities A was likely to have taken part in as a committee member. The UT said that the evidence did not ‘establish’ that any activity in that role was likely to have come to the attention of the authorities. The UT referred to YouTube and other links about protests on 17 December 2017, and on 13 May 2018, A’s attendance at AH’s 65^th birthday on 19 September 2018, meetings on 4 March 2019, 23 March 2019 (35^th Foundation Day), a dinner in May 2019 and the anniversary of the death of Imran Farook’s death in September 2018. A did not play the links at the hearing. It was claimed that he appeared on the platforms. He did not, however, draw the attention of the UT to material in the public domain which identified him at a MQM-L or at a pro-AH meeting.

Paragraph 138 is headed ‘Demonstrations attended by [A]’. The UT summarised A’s evidence on this topic. He had attended five demonstrations between August 2017 and July 2019. Four were at Downing Street and one, on 17 September 2017, in front of the Pakistan High Commission. He had said in his asylum interview that between 50 and 70 people went to these. He prepared banners, distributed leaflets and chanted slogans. I note that A is recognisable in two photographs of demonstrations in the bundle of documents. The UT had no evidence that the fact of the demonstrations was in the public domain. A had said in his asylum interview that he had received invitations to demonstrations from the MQM-L WhatsApp group. The UT nevertheless accepted that the authorities were able to, and did monitor these public activities. A had not given any evidence that he had done anything at a demonstration which would ‘single him out’ (paragraph 139).

Paragraph 140 is headed ‘Photographs and social media’. The UT found that the photographic evidence could support A’s ‘attendance at meetings and some involvement with the organisation’. There was a photograph of A at what looked like a meeting with AH. There was no ‘cogent evidence’ that all these were ‘in the public domain’. A relied on evidence of social media. He said that the authorities would know about it. The Secretary of State accepted that the Facebook account belonged to A. A had given limited evidence about specific posts. There was no evidence about whether the Facebook account was in the public domain, or whether A had used privacy settings. The UT was not satisfied that A would be identified from the Facebook account, for the reasons it gave in paragraph 141, although the account had his picture on it, because nothing linked the account name to A. The posts were limited. Those in 2013 ‘establish[ed] nothing more than that [A] was at the time a supporter of MQM and AH (before the split)’. The UT could see no activity between 2013 and 2019. It was not clear when the photograph of A had been added to the account. The media activity would support affiliation with MQM-L/AH, but did not reflect A’s role as described by the witnesses. A was not posting from a MQM-L account. The posts were from someone who clearly supported MQM-L, but to pick them up and identify A as responsible for them ‘would require a level of investigative work which we do not believe is reasonably likely’.

The UT considered that A realised that the Facebook material had limited evidential value. That explained why, shortly before the hearing, he opened a Twitter account, using his own name ‘and that of the organisation’. The UT found that his explanation for this switch was not credible. He accepted in his oral evidence that his name is a common name. Significantly in the UT’s view, the activity was limited. The account was relatively new. The evidence did not show that A was communicating on behalf of MQM-L in any capacity. ‘Even applying the low standard of proof’ the UT was not satisfied that A would be identified from the limited activity on Twitter. He was not already known to the authorities, which would make detection more likely (paragraph 142).

A’s description of his social media role was ‘wholly unsupported’. He relied solely on the Facebook account ‘until a recent conversion to Twitter’. Some of the letters from MQM-L suggested that members of the security services were operating under cover inside MQM-L. The evidence was not ‘wholly implausible’, but was ‘too speculative’ to be given weight. The UT referred to an answer A had given in his asylum interview to the effect that agents in MQM-L had seen his photograph on the MQM social media site. A did not give any evidence about this. He did not adduce any evidence to support that answer (paragraph 144).

The UT took into account that the F-tT had not accepted that A’s father was of interest to the authorities because of A’s activities (paragraph 145). A had exaggerated his evidence. He was not a credible witness. His evidence that he was involved with MQM before 2016 was not supported. The letter from MQM-L said he had been a member since 2009, but the UT gave little weight to that for the reasons it had given already. That may be a reference to paragraph 120 of determination 4 (see paragraph 53, above). The UT found expressly that A had not ‘established a genuine commitment to MQM-L’. He had established activity since 2016. Anything he had done for MQM-L was to improve the prospects of his appeal.He had not established that he had done anything which was reasonably likely to have come to the attention of the authorities (paragraph 146).

A was not genuinely committed to MQM-L or to AH so as to engage HJ (Iran). The UT was confident that if A was genuinely committed to MQM-L, so that his sur place activity would put him at risk, MQM-L would have provided better evidence to support his claim (paragraph 147). The UT concluded that A had not established that he would be at risk on return to Pakistan. The UT dismissed the appeal (paragraph 148).

There are two grounds of appeal.

Mr Fripp relied on the UT’s findings in paragraphs 91, 95, 104, 106, and 107-8, 115(a) and (d) of determination 4 that the Pakistani authorities wanted to eliminate AH and his power base, that there was a well-resourced campaign against AH and MQM-L, that this extended to monitoring MQM-L in London, that it was likely that the authorities would devote resources to identifying genuine supporters and members overseas, and that they would go to any lengths to ensure that AH did not get any traction in Pakistan. He relied on the finding that a person identified as a supporter of MQM-L would face a risk of persecution on return, and that a ‘genuine’ supporter could not be expected to hide that to avoid persecution on return.

Mr Fripp referred in paragraph 38 of his skeleton argument to the findings about A’s past affiliation to MQM in determination 1 and in paragraph 39 of determination 2. I have described both (see paragraphs 5 and 13, above). He also relies on the preserved findings from determination 2 that A was an active member of MQM-L ‘in the sense of administration and organisational activities’, who attended ‘some meetings and demonstrations’, and a member of the committee of the North London unit of MQM-L.

Mr Fripp then submitted that, on the UT’s findings about the risk to AH and MQM-L supporters in Pakistan, A’s case is made out. He argued that there was a reasonable likelihood that, on return to Pakistan, he would be identified as a supporter of AH/MQM-L in London and face serious risks. The UT raised the standard of proof and required an undue level of corroboration as regards knowledge by the authorities in Pakistan of A’s activities.

He also relied on a statement by Sedley LJ in paragraph 18 of YB (Eritrea) v Secretary of State for the Home Department [2008] EWCA Civ 360. Sedley LJ said that where there was objective evidence ‘which paints a bleak picture of suppression of political opponents’ by a government, ‘it requires little or no evidence or speculation to arrive at a strong possibility – and perhaps more – that its foreign legations not only film or photograph their nationals who demonstrate in public against the regime but have informers among expatriate oppositionist organisations who can name people who are filmed or photographed. Similarly it does not require affirmative evidence to establish a probability that the intelligence services of such states monitor the internet for information about oppositionist groups’.

The UT were not cautious enough about the knowledge of the Pakistani authorities and the resultant risk. Given A’s characteristics, and the findings about the high level of risk to MQM-L supporters, the UT should have allowed the appeal.

The fact that A had tried to ‘bolster his claim’ was not the end of the story, as the relevant authorities show.

The UT’s reasoning about whether or not A was a genuine supporter of AH/MQM-L, did not define what that meant. The UT’s reasoning was unclear. It is not necessary to be a ‘genuine’ supporter to attract the protection of the Refugee Convention. The UT made no adequate findings about A’s state of mind. It missed the significance of the pro-MQM Facebook post in 2013. None of the factors described by the UT was inconsistent with some attachment to AH/MQM-L, so that a requirement to hide that would be inconsistent with HJ (Iran). The UT did not explain how the Pakistani authorities would be able to tell whether or not his commitment was genuine. The UT gave too much weight to its own sense that A’s activity was self-serving.

Mr Holborn submitted that none of the very clear findings on which Mr Fripp relied was relevant because the UT had clearly found that it was not reasonably likely that A would be detected by the authorities in Pakistan. That meant that he would not be identified as a member of MQM-L on his return. The fact that the UT had found that A was not a genuine supporter meant that HJ (Iran) was not engaged. As his support was not genuine, A would not be forced, on his return to Pakistan, to hide his true affiliation in order to escape persecution. Mr Holborn did not accept that YB (Eritrea) was authority for any general principle of law, but accepted, nevertheless, that a lack of specific evidence of surveillance was not fatal to a protection claim.

He did not accept that if ground 1 succeeded, the appeal should be allowed. The case should be remitted to the UT.

I have not found this an easy case. Determination 4 is evidently the product of careful and thoughtful reading and analysis of the extensive evidence of the relevant political history and current conditions in Pakistan, and of a meticulous and thorough evaluation of A’s evidence. The UT is an expert tribunal. It should be assumed to have understood, and correctly to have applied, the relevant law, unless it is clear that it has not done so. Nevertheless, after taking into account those considerations, and giving the appropriate respect to the UT’s industry and expertise, I have reluctantly concluded that its reasoning is unsafe.

The correct legal approach was not in dispute. In brief, the parties agreed that A had to show not that he would be persecuted by reason of his political opinion, but that there was a serious, or real risk that he would be. They also agreed that, if A was a genuine supporter of MQM-L and of AH, he could not be required to hide that on his return to Pakistan in order to escape persecution.

I start by recalling the approach of the UT to risks arising from membership of MQM-L and to the authorities’ coverage of MQM-L’s activities in London. The UT decided to give weight to the expert’s evidence, which had not been significantly challenged and which was supported in significant respects by other evidence. The UT specifically found that the seniority of an MQM-L member was not relevant to the level of risk he might face. There were many reports of killings of MQM-L ‘workers’. The situation was ‘far more nuanced’ than the RTIR suggested. The authorities wanted to ‘eliminate’ AH and his power base. Support for AH was a risk factor. Those identified by the authorities as associated with MQM-L as ‘members supporters and workers’ were at risk.

There was some indirect support for the expert’s evidence about the Pakistani authorities’ knowledge about MQM-L in London. They were aware of MQM-L’s activities in London. The authorities were committed to destroying AH and his power base. The fact that AH was in London made it more likely that the authorities were monitoring activity by MQM-L there. It was plausible that they would ‘invest resources in identifying genuine supporter/members overseas’. They ‘would go to lengths’ to ensure that AH’s power base ‘did not gain traction in Pakistan’.

The UT took into account that the military ‘closely and effectively monitors’ social media activity. That was supported by the background evidence. It was also reasonably likely that the security forces ‘monitor activities of MQM-L on the ground’ and ISPR monitors media activity. If the authorities were aware that a person was or was perceived to be involved in pro-MQM-L activities in London, he would be targeted and at risk of persecution. Such a person would be expected to renounce MQM-L and give information to the authorities. It was contrary to the spirit of the Refugee Convention to expect a person, whether or not he was a genuine supporter, to give information to the authorities.

The interest of authorities would not depend on the seniority of a member of MQM-L. It would depend on whether the person was, or was perceived to be, pro-AH or pro-MQM-L ‘which was reasonably likely to have come to the attention of the authorities ‘(as a result of monitoring in the United Kingdom by the authorities and/or monitoring of media activity)’. The greater the level of activity, the greater the risk of detection. A person who had no genuine political motives could be at risk.

The significant obstacles to the success of A’s appeal on ground 1 are three findings of the UT. The first, in paragraph 110, was its rejection of the expert’s evidence that there was a real risk that members of MQM-L would give information to the Pakistani authorities to curry favour, and his evidence that those responsible for social media posts would be identified by informers. The second was that there was not enough evidence to enable the UT reasonably to draw conclusions about ‘the level of and the mechanics of monitoring’ in the United Kingdom. The third was that although the security services could detect social media posts, they did not have the resources or capacity to detect all relevant posts, let alone to identify those responsible.

I paraphrase a question which Phillips LJ asked Mr Holborn in argument, ‘What evidence did the UT expect?’ It is very improbable that there would be any direct evidence of covert activity by the Pakistani authorities, whether it consisted of monitoring demonstrations, meetings and other activities, monitoring social media, or the use of spies or informers. I do not consider that Sedley LJ was suggesting, in paragraph 18 of YB (Eritrea), that a tribunal must infer successful covert activity by a foreign state in the circumstances which he described. He was, nevertheless, making a common-sense point, which is that a tribunal cannot be criticised if it is prepared to infer successful covert activity on the basis of limited direct evidence. Those observations have even more force in the light of the great changes since 2008 in the sophistication of such methods, in the availability of electronic evidence of all sorts, and in the ease of their transmission. To give one obvious example, which requires no insight into the covert methods which might be available to states, it is very easy for an apparently casual observer of any scene to collect a mass of photographs and/or recordings on his phone, without drawing any adverse attention to himself, and then to send them anywhere in the world.

I consider that, on this aspect of the case, the UT erred in law by losing sight of the fact that direct evidence about ‘the level of and the mechanics of monitoring’ in the United Kingdom is unlikely to be available to an asylum claimant or to a dissident organisation, and by imposing too demanding a standard of proof on A. The UT repeatedly said that A had not ‘established’ things, that ‘cogent evidence’ of something was absent, and that parts of A’s evidence were not supported (see further, the next paragraph).

A related point is that the UT’s approach was to posit two mutually exclusive alternatives: a tiny level of support for MQM-L which was not capable of drawing the attention of the Pakistani authorities, and, therefore, of putting A at risk on return, and the level of support which A described in his exaggerated but nevertheless nebulous evidence. If that was the UT’s approach, its danger is to obscure a third possibility, which is that, on the UT’s other findings, A did support, or could be perceived to support, MQM-L to an extent which might, to the lower standard, attract the attention of the authorities and therefore put him at risk. I consider that the UT’s findings that A had exaggerated his role (which were open to it on the evidence) dominated the UT’s analysis of potential risk; and that the UT erred in law in this respect. There were photographs of A at demonstrations, and the UT accepted that he had been to four outside Downing Street and one outside the Pakistan High Commission. The UT accepted that the authorities would keep an eye on the High Commission. There was also a photograph of AH on A’s Facebook account.

A recurrent theme of determination 4 is that A’s evidence about aspects of his claim was not supported by other evidence, and, by implication, for that reason alone, to be rejected, without the need to consider, to the lower standard, its intrinsic probability. As I have indicated, on at least three occasions, the UT observed that there was ‘no cogent evidence’ that something was the case (paragraph 132, line 4, paragraph 133, line 3, paragraph 140, line 3). That theme indicates a linked error. That error is that the UT treated the specific preserved findings that A was not credible about particular aspects of his claim, coupled with their own findings that he was not credible about other aspects of his claim, as a proxy for analysing the relationship between their own general findings about risk, A’s evidence generally, and the uncontested evidence about A’s role. It is a trite proposition that credibility is not ‘a seamless robe’, even if, on analysis, some, or most of the evidence proves to be incredible. Findings that some aspects of a witness’s evidence are not credible should not, in a protection claim, be generalised to all his evidence. The fact-finder must also consider the intrinsic likelihood, to the lower standard, of the significant aspects of his claim.

Mr Fripp reminded us of the need for anxious scrutiny. I would allow the appeal on ground 1.

My concern about ground 2 is that the impression given by determination 4 is the UT’s view that A had fabricated or embellished his role in order to bolster his asylum claim distracted it from asking, as it should have done, whether his affiliation with MQM-L was genuine, or genuine enough, to mean that he would or might nevertheless be at risk on return to Pakistan, to the lower standard.

Mr Fripp had reminded the UT, in paragraph 5 of his skeleton argument for the UT hearing, that, per Devaseelan, the starting point was paragraphs 16-18 of determination 1. In determination 1, the F-tT had concluded that A had been an active member of MQM in Pakistan from 2009, and in England from 2012. In determination 1, his claim was rejected, in part, because of this long-standing allegiance to MQM which long pre-dated his asylum claim in 2017. There was no suggestion in determination 1 that this affiliation was not genuine. In paragraph 39 of determination 2, the F-tT apparently also found that A had been a member of MQM in Pakistan in 2009.

In determination 4, the UT departed from that starting point without, in my judgment, explaining why. It did not accept that A was a member of MQM before he came to the United Kingdom. It did not accept that he was a member of MQM in the United Kingdom before 2016, although it acknowledged in paragraph 141 (see paragraph 62, above) that there were Facebook posts in 2013 showing ‘nothing more than that [A] was at that time a supporter of MQM and AH (before the split)’. It did not accept his evidence that he had been taking an active part in the affairs of MQM since 2012 and attending some events since 2014. More fundamentally, without explaining why, it departed from the basic thrust of determination 1, which was that A had been a member of MQM in Pakistan since 2009 and in England since 2012, and that his exposure as a result was no different in 2017, which cast doubt on the credibility of his asylum claim in 2017. The UT turned that approach on its head, by downplaying, if not ignoring, A’s earlier allegiance with MQM in order to find, in paragraph 146, that A ‘had not established a genuine commitment to MQM-L’, and that the activity since 2016 was for the purpose of improving his prospects on appeal. The unexplained contradiction between the approach in determination 1 and in determination 4, and the acknowledgement by the UT of evidence of support by A for MQM and for AH in 2013, before the split, and before his unsuccessful application in 2015, and long before his asylum claim, seriously undermine the safety of the UT’s conclusions that A’s support for MQM was not genuine.

I would also allow the appeal on ground 2.

We heard limited submissions about remedy at the hearing. I would be inclined to remit the case on both points. I do not consider, in the light of the complications in this case, and of the apparent contradictions and deficiencies in A’s evidence (for example, about when he joined MQM-L in the United Kingdom), that it is obvious that there is only one right answer. In my judgment, it is for a specialist fact-finder, and not for this court, to assess these questions again. I would not prevent the parties, if so advised, from making further written submissions about remedy, and, in particular, about whether the appeal should be remitted to the UT or to the F-tT.

For these reasons, I would allow the appeal on grounds 1 and 2. I would be inclined to remit the appeal, but would be willing to consider further written submissions on that question.

I agree.

I also agree.