AC-2023-LON-002337 - [2025] EWHC 1985 (Admin)

The Applicant's primary case under Article 3 ECHR is that extradition would not be compatible with his right not to be subjected to inhuman or degrading treatment due to the high risk of assault, particularly sexual assault, and the alleged inefficacy and inhumanity of any protective measures within the United States prison system. This contention formed a significant part of his original grounds of appeal and was elaborated upon in the various additional written and oral submissions before me.

Mr Fitzgerald submitted that the SDJ erred in concluding that the Applicant was not at risk of sexual assault, a risk known to be higher in the case of non-heterosexual prisoners. Thus, he argued the Applicant is at a heightened risk of prison violence and sexual assault in US federal custody due to his particular vulnerabilities, specifically his homosexuality, demonstrable effeminacy, and prior victimisation.

A central aspect of the Applicant's Article 3 argument is the challenge to the reliability of PREA audits, upon which the SDJ had relied. The Applicant contends that the SDJ's reliance on a PREA audit of MCC San Diego was “wholly unreasonable” given compelling general evidence about the inadequacy of such audits in other cases. Specifically, evidence from Ms Abbate and letters from Senators highlighting concerns about the accuracy and reliability of PREA audits and the BOPs failure to protect prisoners from sexual abuse. The Applicant further pointed to incidents at FCI Dublin, where favourable PREA audits were conducted despite notorious sexual abuse occurring there. He contended that the SDJ “wrongly discounted” Ms Abbate's criticisms of the PREA audit.

The Applicant's case acknowledges that, for his own safety, he might be detained in an SHU. However, he challenges the SDJ's findings regarding the compatibility of such detention with Article 3.

He argues that the SHU regime is “extremely restrictive” and its likely effects on him, given his particular personality traits and psychological difficulties, would be deleterious to his mental health, effectively amounting to solitary confinement. The Applicant submitted that the SDJ was wrong in his conclusions that solitary confinement in a single cell for protective reasons was highly unlikely and would not be inhuman in any event.

If he was not housed in a single cell, then he further argued that the SDJ was wrong to conclude that housing in a double cell in an SHU would be compatible with Article 3. For this proposition he relied on the evidence of Dr Grassian, a psychiatric expert, that double-celling could be “even worse” due to forced sharing of a small space for prolonged periods with no external stimulation, leading to potential interpersonal frustration and anger, which would “all the more clearly breach Article 3 ECHR”.

The Applicant's mental condition, specifically his personality disorder and psychological difficulties, was also directly linked to the argument advanced by reference to SHU placement and its compatibility with Article 3 (Dr Grassian, Professor Forrester, and Professor Greenberg concurred that solitary confinement would have a deleterious effect on the Applicant and could lead to a risk of serious psychiatric injury).

The Applicant further raised concerns about foreign prisoners detained in CAR prisons, suggesting harsher conditions for US citizen prisoners. While linked to the proposed new grounds under section 81(b) and Article 14, this evidence was also initially relevant to Article 3 and was considered by the SDJ in this context.

It was common ground before the SDJ that sexual assaults occur within the US prison estate. The Applicant relied on statistical material, notably from the National Inmate Survey (NIS) conducted by the Bureau of Justice Statistics (“BJS”). The purpose of the NIS is to gather data on the prevalence of sexual victimisation in prisons. It draws upon large, randomised samples of adult prison populations, and its methodology includes statistical adjustments for non-response to enhance its rigour. The BJS utilises the data collected from these surveys to compile reports on various topics, including the experiences of Lesbian, Gay, Bisexual, Transgender, and Queer (“LGBT+”) individuals within the prison system. The survey relies on self-reported information without subsequent verification.

The NIS results suggest a significant underreporting of sexual abuse incidents by prison administrators. For example, the NIS-3, covering 2011-2012, estimated 80,600 victims of sexual abuse in 2012, while facility administrators reported only 10,047 incidents in the same period, implying administrators reported approximately 12.5% of inmate-reported allegations. The BJS has estimated that 200,000 prison inmates are sexually abused annually in the US. This underreporting is attributed to factors such as fear of retaliation, stigma, a “code of silence,” and staff scepticism or premature dismissal of claims. Only a small percentage of allegations are substantiated (approximately 2.6% in Federal prisons in 2018), with the low substantiation rates arising not from false allegations but as a result, it is said, of systemic issues such as a lack of clear policies, insufficient investigators, poor evidence collection, and inadequate coordination. A key statistical point advanced by the Applicant is the disproportionate risk to LGBT+ prisoners. Evidence from the BJS indicates that a gay or bisexual man or woman is at least 10 times more likely to be sexually victimised by another incarcerated person, and 2.6 times more likely to be victimised by staff, compared to heterosexual individuals. This is acknowledged to be a “high-risk group for being the victims of violence and harassment”. The SDJ accepted that a higher risk to such a defendant was “obvious”.

The SDJ considered the prevalence of sexual assaults, finding that “the level of sexual violence reported within the US and UK prison estates is broadly speaking equivalent, or the UK prison estate appears (statistically) to be worse”. He concluded that the level of sexual violence in the US prison estate is not “endemically unusual” and has a lower level of reported abuse than the UK. While the Applicant criticises this comparison as flawed due to US underreporting, the SDJ was entitled to weigh the evidence as presented to him and draw contextual comparisons based on the data available. He specifically noted that Professor Hamilton conceded in cross-examination that sexual violence was not a “peculiarly American problem” and existed in other Western jurisdictions. This concession, though subject to the argument as to underreporting, informed the SDJ's view on the comparative prevalence of abuse within prison systems and was based upon the evidence.

In assessing the risk that the Applicant would in fact face against this background, the SDJ evaluated the evidence from the Applicant's expert witnesses, including Professor Hamilton, Ms Abbate, Ms Hill, and Mr Zoukis. The SDJ found their evidence to contain “non-specific, general assertions”.

He concluded that Mr Gregory Houska's evidence was inadmissible as he was not an expert on prisons, lacking the necessary qualifications or experience. He had not been inside a US prison for six years (save for a family visit) and his knowledge of MCC San Diego was based on tours from the late 1990s and 2005.

Mr Christopher Zoukis's evidence was given “very little weight” by the SDJ, who found him not to be an expert and to have failed the impartiality requirements that attach to expert evidence. Mr Zoukis accepted in cross-examination that examples he provided to undermine the effectiveness of the PREA system were “wholly irrelevant” and that he “regretted” including them.

The SDJ approached Ms Maureen Baird with “real concerns over the veracity of her evidence”. She had not been inside MCC San Diego since 2009 and relied on contact with only two former colleagues, neither of whom still worked at the facility. She was unaware of a multi-million-dollar overhaul of the prison in 2017/18 and the existence of the I-Unit.

He decided that Ms Hill's statistical analysis was inadmissible as being beyond her expertise.

While the Applicant presented evidence of alleged failures of PREA audits at FCI Dublin and CFCC Coleman (where the concerns were of abuse of female prisoners by staff), and noted Congressional concerns about their accuracy, the SDJ was required to make an evaluative judgment as to how this evidence was to be weighed against the specific audit information available for MCC San Diego.

The SDJ placed reliance on the recent (April 2021) and comprehensive PREA audit of MCC San Diego, from which he said he “drew significant comfort”. This audit confirmed that “proper and robust systems are in place for the prevention and detection of sexual assault” and that the prison was “100% compliant with PREA regulations”. He considered this “up to date and cogent evidence of the steps taken, in accordance with the law, to minimise sexual assaults in Prisons in the US”. The Respondent referred to a number of further recent PREA audits for various facilities which consistently showed “Meet Standard” for PREA compliance. They detail procedures for inmate education, risk screening, reporting mechanisms, and investigation which the SDJ accepted, represent “practice” rather than merely “policy”.

The SDJ rejected Ms Abbate's opinion on the adequacy of the MCC San Diego audit, questioning her objectivity and suitability for providing expert evidence given her willingness to express opinions without acknowledging whether she had raised her concerns with Mr Coudriet, the auditor, prior to finalising her conclusions. The SDJ considered Ms Abbate's criticism that the auditor only spent three days on site when conducting the audit to be “bordering on partisan” and insufficient to undermine the audit. Her assertion that the PREA audit report was “dishonest or incompetent,” led the SDJ to question her objectivity. The SDJ concluded that he could not agree “...with her ultimate opinion that Mr Coudriet’s report was ‘valueless’”.

His finding that Ms Abbate's criticisms did not sufficiently undermine the reliability of the MCC San Diego audit, particularly given his assessment of her objectivity, was a determination of fact based on evidence. In relation to protective measures for vulnerable prisoners, including LGBT+ individuals, the SDJ found that the US has “a range of procedures in place which are appropriate to protect vulnerable prisoners,” notably housing in SHUs or Integration Units. He concluded that these measures would provide a “reasonable level of protection from sexual violence”.

The Applicant contends that reliance on SHUs as a protective measure does not meet Article 3 concerns because they are themselves “extremely restrictive” and could lead to deleterious mental health effects, potentially amounting to solitary confinement incompatible with Article 3. However, the SDJ’s findings and conclusions were to the opposite effect. He concluded that detention in an SHU does not amount to “solitary confinement” and that inmates are held in the “least restrictive setting necessary” with access to prison staff and facilities.

He summarised the Applicant’s case in relation to SHU’s observing that it appeared to involve a Catch 22 argument [282-283]:

“a.

Firstly, he should not be housed in general population for his own safety.

b.

Secondly, he should not be housed in protective custody as this would amount to “solitary confinement” and breach his article 3 rights.

c.

Thirdly, he should not be housed in protective custody with a cell mate (i.e., not in solitary confinement) because this would also breach his article 3 rights.”

Statistical evidence presented indicated that only 0.289% of federal inmates were held in SHUs for protective custody reasons as of December 2021. Furthermore, Professor Hamilton herself accepted that the SHU regime described by the Respondent was “similar to that adopted in many Western prisons throughout the world and was in principle a sensible approach”. The conflicting evidence regarding the 9th floor I-Unit at MCC San Diego, with some indicating it housed general population rather than vulnerable prisoners, was a matter for the SDJ to weigh in his assessment. His conclusion, that these units would also serve the purpose of providing reasonable protection, was a finding he was entitled to make on the evidence before him.

Whilst Mr Fitzgerald highlighted the pervasive nature of sexual violence in US prisons, significant underreporting, and the particular vulnerability of LGBT+ individuals, the SDJ was plainly aware of and took it into account this background. He made findings on witness credibility, such as that concerning Ms Abbate and Mr Zoukis, which fall squarely within his remit as the trial judge. His reliance on the positive PREA audit for MCC San Diego, despite the Applicant's broader criticisms of the system, was a permissible exercise of evaluative judgment on the evidence, given that he had assessed the specific audit in question.

Article 3 ECHR provides an absolute prohibition against torture, inhuman or degrading treatment or punishment, making no provision for exceptions or derogations. This absolute character underscores its fundamental importance. In the context of extradition, it is unlawful to extradite an individual to a requesting state where there is a real risk of them being treated in a manner which is prohibited by Article 3 (see R (Ullah) v Special Immigration Adjudicator [2004] 2 AC 323).

The threshold for the minimum level of severity required to engage Article 3 is relative. The ill-treatment must go “beyond that inevitable element of suffering or humiliation connected with a given form of legitimate treatment or punishment” (see Elashmawy v Court of Bresica, Italy [2025] EWHC 28 (Admin)). Generalised evidence of human rights violations is insufficient; it must be demonstrated that the individual requested person is specifically at real risk.

The requesting state may adduce evidence to rebut a claim that Article 3 standards would be breached and show that no real risk exists. While there is a presumption that ECHR signatory states will comply with human rights standards, including Article 3, a similar presumption applies to trusted extradition partners such as the USA. At [240-243] of his judgment the SDJ observed:

“240.

The starting point when looking at the prison conditions in the US is framed by some fundamentals about the USA and how the law is to be applied in those circumstances.

241.

The United States is a mature democracy, with an internationally recognised commitment to the rule of law, it is a signatory of the Universal Declaration on Human Rights, the US is a trusted UK extradition partner to which the UK courts regularly surrender requested persons and an individual surrendered to the United States enjoys the protection of the US constitution and the ability to litigate any alleged infringements of that constitution within the US legal system.

242.

A consequence of these fundamental matters is that, despite the frequency with which UK courts are asked to consider conditions within the United States prison system, it is very rare for the UK to refuse surrender to the United States based on concerns relating to prison conditions.

243.

As much is evident from even a cursory look at some of the most recent occasions upon which UK courts have found that surrender to the US would be compatible with an individual’s Convention rights.”

Where the alleged ill-treatment stems from non-state agents, such as fellow prisoners, the governing test is “whether the state can provide reasonable protection against such violence” (R (Bagdanavicius) v Secretary of State for the Home Department [2005] 2 AC 668 and Lord Advocate v Dean [2017] UKSC 44). This necessitates consideration of both the objective risk of violence to the requested person and the extent to which the authorities in the receiving state will implement protective measures.

It is settled law that Article 3 does not impose a requirement for a guarantee of absolute safety. The state's responsibility under Article 3 is engaged if it fails to provide reasonable protection.

The SDJ concluded that extradition would not be incompatible with Article 3 concerning prison conditions and the risk of the Applicant being subjected to inhuman and degrading treatment. This conclusion was based on a thorough assessment of the evidence, which he considered in detail.

In relation to the risk of sexual violence, the SDJ found, on the basis of the evidence referred to above, that sexual assaults do occur in US prisons, much as they do in the UK prison estate. While acknowledging a degree of under-reporting, the SDJ considered the reported level of sexual violence in US and UK prisons to be broadly equivalent, or even lower in the US estate. The SDJ also noted that the problem of sexual violence in prison is not peculiar to the United States [245]:

“245.

The RP submits that because there is evidence of sexual assault within the US prison estate, or because incarceration in certain conditions is considered detrimental to the health of those with pre-existing mental health conditions, this amounts to a bar to extradition, that is flawed and illogical thinking, it is obvious that all prisons in all parts of the world suffer from degrees of sexual assault and / violence and health conditions may well impact negatively on health but that is not the test or the jurisprudence laid down by the authorities. The approach of the ECtHR is illustrative of this.”

The SDJ was satisfied that the US has a range of procedures in place that are appropriate to protect vulnerable prisoners. These measures include the possibility of housing vulnerable inmates in the I-Unit at MCC San Diego, or if necessary, in an SHU. The SDJ also noted that in other BOP facilities, housing is available in Integration Units. He found that inmates have proper avenues to express concerns about vulnerability, and these concerns are taken seriously within the BOP estate.

In relation to the adequacy of protective measures, the SDJ expressly found that it could not sensibly be argued that SHUs or the I-Unit would fail to provide the Applicant with a reasonable level of protection from sexual violence, as this would be the very purpose of housing him in an SHU. The SDJ thus concluded that, “whatever the level of risk,” there were undoubtedly measures in place to provide a reasonable level of protection, thereby rejecting the Applicant's case on non-state agent inter-prisoner violence. This conclusion was squarely within the governing test in cases of mistreatment by non-state agents and open to the SDJ on the evidence.

In relation to SHUs, the SDJ found that detention within an SHU does not amount to “solitary confinement”. Inmates in SHUs are housed “in the least restrictive setting necessary to ensure [their] safety,” either alone or with another inmate, with those housed alone having “periodic contact with other inmates”. They participate in recreation enclosures that accommodate multiple inmates at once. The SDJ concluded that SHU detention is for a legitimate purpose (protection), is in the least restrictive regime necessary, would be subject to weekly reviews, and provides adequate living space and lighting, regular contact with staff and other prisoners in the SHU. Such conditions were similar to Article 3 compliant regimes adopted in many Western prisons. Detention in an SHU was not arbitrary, subject to review, and did not involve complete sensory deprivation. This accords with established case law that removal from association for security, disciplinary, or protective reasons does not in itself amount to inhuman or degrading treatment, provided specific conditions are met regarding the stringency, duration, objective, and effects on the person concerned.

In R (AB) v Justice Secretary [2022] AC 487, the Supreme Court held that:

“43 The application of article 3 in relation to what can broadly be described as removal from association or solitary confinement has been considered by the European court in a substantial number of cases. The court has repeatedly held that removal from association is not in itself inhuman or degrading. In Van der Ven v The Netherlands (2003) 38 EHRR 46, para 51, it stated, under reference to earlier decisions: “the removal from association with other prisoners for security, disciplinary or protective reasons does not in itself amount to inhuman treatment or degrading punishment.”

To the extent that the applicant sought to establish that he had specific mental health problems which militated against the use of protective measures such as the SHU this was also considered and taken into account by the SDJ.

In assessing the Applicant’s mental health, the SDJ considered the evidence provided by various psychiatric witnesses. Only one of the four doctors, Professor Forrester, diagnosed the Applicant with a depressive disorder, and even Professor Forrester considered this to be of a moderate level.

The SDJ was sceptical about the extent to which the Applicant's self-reporting to psychiatric experts could be relied upon. This scepticism was shared by some of the psychiatric experts; Dr Picchioni noted the Applicant's “tendency to exaggerate and lie,” as did Professor Greenberg. Professor Forrester “would not be drawn on the issue” of the Applicant's self-reporting. The SDJ concluded that “any conclusions based on self-reporting should – for obvious reasons – be carefully drawn”. Dr Picchioni's assessment, which the SDJ favoured, also noted the Applicant's unreliability. The Applicant himself admitted to lying to achieve his objectives. When asked at the outset of cross-examination, “Do you lie to get what you want?”, the Applicant replied, “Yes”. The SDJ noted, “it is hard not to frame my assessment of my assessment of his evidence through that prism”. He found, not surprisingly perhaps, that the Applicant was “dishonest and manipulative”.

The SDJ noted that the Applicant had managed in an Indonesian prison and subsequently in a Victorian Category B prison in the UK without any significant negative impact on his mental health which did not suggest that the Applicant was ‘unsuitable for prison life’. Consequently, the SDJ found no basis to conclude that extradition would be unjust or oppressive due to the Applicant’s mental state.

He referred to Dr Picchioni’s evidence [171]:

“When asked about the possibility Mr Tahilramani might need to be housed away from other inmate for his own protection, Dr Picchioni said “we are discussing a man who has spent two years at HMP Wandsworth, sometimes sharing a cell, sometimes not. A man who has a job off the wing, and on the wing, meets inmates, has experienced prejudice and bigotry. He has remained on normal location; he has not ended up on the vulnerable prisoner wing. He has demonstrated more resilience than might be expected just based on the diagnosis, he has strengths as well as weaknesses and an ability to navigate a course.”

Furthermore, the Applicant had no significant history of self-harm or suicide attempts and was assessed as ‘not significantly unwell’ (see further below in relation to section 91).

Notwithstanding the evidence presented by the Applicant regarding systemic issues of sexual abuse and the criticisms levelled against the PREA auditing process, the SDJ’s judgment reflects a careful assessment of the statistical data and the expert evidence. He was entitled, as the primary arbiter of fact, to make findings on the credibility and weight of the expert witnesses, such as Ms Abbate, Mr Zoukis, and Ms Baird, whose evidence he treated with caution. The SDJ's reliance on the specific, recent PREA audit for MCC San Diego, as “cogent evidence” of robust protective systems, was a permissible exercise of his evaluative judgment, particularly given his critical assessment of attempts to undermine its reliability. His conclusions that the US authorities would be able to offer a “reasonable level of protection” to the Applicant, “whatever the level of risk,” and that protective measures such as SHUs would not, in themselves, constitute inhuman or degrading treatment, were findings he was entitled to make on the evidence before him. The test remains whether the state can provide “reasonable protection” and does not require “a guarantee of safety”.

The Applicant argued that the SDJ erred in law by refusing to admit further evidence, particularly material said to undermine the reliability of PREA audits. The SDJ's judgment deals with the application expressly [204-207]:

“204.

During the Extradition Hearing the Court invited Ms Abbate to provide an example of a PREA audit report which she considered to be adequate. Ms Abbate has provided such an example and I have admitted and considered that because it was at the courts request.

205.

However, Ms Abbate has also provided a letter (which she referred to in evidence) from Senators Brian Schatz and John Cornyn to the Director of the BOP requesting a review of recent PREA audit findings considering ongoing reports of sexual abuse of inmates, dated 3 June 2022 (‘the June 2022 Congress Letter’.

206.

In addition a further 1300 pages of “Further Evidence Bundle” were also served, they seem to be irrelevant and served late. It is not in the interests of justice for the defendant to serve, after the closing of evidence, substantial bodies of evidence which do not realistically go to any issue. This includes, for example, open- source material dating back to 2016 concerning an individual prison in Illinois, which has not been the subject of any evidence or argument, nor was such evidence anticipated by the Court of the US Government.

207.

The service of this volume of material fails to identify with precision relevant evidence that is germane to the issues at this stage of the proceedings. It risks court being dragged into satellite litigation. The evidence fails to comply with either the CPR overall or Rule 50’s special objective in extradition and fails to identify with particularity what issue it relates and is NOT admitted.”

The SDJ's decision to refuse the admission of the further evidence was, in my view, entirely proper and consistent with the principles governing the management of evidence in extradition proceedings for the reasons he gave. There is nothing to suggest that the SDJ's exercise of discretion in this regard was erroneous or outside the wide ambit of deference to be accorded to a trial judge in relation to a decision of this type.