NIRAV DEEPAK MODI v GOVERNMENT OF INDIA

This is the judgment of the Court to which we have both contributed.

The appellant, Nirav Deepak Modi, is sought by the Government of India (“GoI”). There are three sets of criminal proceedings. The first, brought by the Central Bureau of Investigation (“the CBI”), relates to a fraud on the Punjab National Bank, which caused losses equivalent to over £700 million. The second, brought by the Enforcement Directorate (“the ED”), relates to the alleged laundering of the proceeds of that fraud.

The GoI submitted requests for Mr Modi’s extradition on 27 July 2018 in relation to the CBI proceedings and 24 August 2018 in relation to the ED proceedings. The requests were certified by the Home Office on 28 February 2019. He was arrested on 19 March 2019. He appeared at Westminster Magistrates’ Court on 20 March 2019 and has been in custody at HMP Wandsworth since then. On 11 February 2020, the GoI made a further extradition request in relation to a third set of criminal proceedings involving alleged interference with evidence and witnesses in the CBI proceedings. This request was certified on 20 February 2020.

The extradition hearing took place over two weeks in May and September 2020 before District Judge Goozée (“the District Judge”), with closing submissions in January 2021. On 25 February 2021, the District Judge handed down his decision. He found that there were no bars to extradition and sent the case to the Secretary of State. On 15 April 2021, she ordered Mr Modi’s extradition to India.

Mr Modi applied for permission to appeal on multiple grounds. On 9 August 2021, Chamberlain J handed down a reserved judgment and gave Mr Modi permission to appeal on two of them. Ground 3 is that the lower court was wrong to decide that his extradition would be compatible with his Convention rights under Article 3, ECHR. Ground 4 is that the lower court was wrong to decide that it would not be unjust or oppressive within the meaning of section 91 of the Extradition Act 2003 (“the 2003 Act”) to extradite him by virtue of his physical or mental condition. All other grounds of appeal were rejected.

In giving permission, Chamberlain J neatly encapsulated the central features of this appeal, stating at [18]:

“I will not restrict the basis on which those grounds can be argued, though it seems to me that there should be a particular focus on whether the [District Judge] was wrong to reach the conclusion he did, given the evidence as to the severity of the appellant’s depression, the high risk of suicide and the adequacy of any measures capable of preventing successful suicide attempts in Arthur Road prison. The application of the Turner test to a case of severe depression also seems to me to warrant consideration by the Divisional Court.”

This appeal is brought under section 103 of the 2003 Act. Given that it is accepted that the appellant would find it harder to establish a violation of Article 3, ECHR than of section 91, Mr Edward Fitzgerald KC rightly concentrated on the latter. In the majority of extradition appeals the principal focus is on whether the district judge erred in law or in fact, but in the instant case the evidential picture has moved on very substantially since February 2021 as we will in due course explain. Although it remains relevant to consider whether the District Judge applied the correct legal test under section 91 to the material that was before him, the parties are agreed that, a mass of fresh evidence having been admitted, we are required to decide the central question of oppression de novo. In that regard, the main issues for us to decide remain those encapsulated by Chamberlain J.

The District Judge gave a lengthy and thorough judgment on a range of issues, most of which have fallen away in the light of Chamberlain J’s permission decision.

On what we have described as the central question, the District Judge received two assurances from the GoI (the Ministry of Home Affairs), dated respectively 8 June 2019 and 11 September 2020. In the first of these assurances, it was stated that in the event of extradition the appellant would be held at Barrack No 12, Arthur Road Jail in Mumbai which is separate from the “general population” at this prison. Aside from matters of personal space and living conditions, the key point was that medical facilities would be available 24/7, four medical officers along with four nursing orderlies and two pharmacists would also be available, there was a prison hospital with 20 beds and outside experts came in when required. There is a public hospital within 3 km of the prison.

The letter of assurance further stated that “the prison authorities are bound by this assurance provided for [the appellant] and there is no discretion whereby any other administrative, local government or judicial authority would override it as per the law of the land”.

In the second assurance, described as a “continuation” of the earlier assurance, it was stated that if extradited the appellant “may receive any relevant and necessary treatment from a private doctor or mental health expert of his choice, including treatment or counselling from psychiatrist, psychologist, as required and paid for by him, including coming into prison / over video-link for consultations”.

The District Judge was also provided with what he described as “a very useful video” of Barrack No 12, both inside and out. We too have studied this video and have borne in mind the circumstances in which it has been provided as well as the fact that there were no inmates at the material time. Barrack No 12 is a dedicated area for high-profile prisoners. The video certainly gives the impression of a spacious and clean cell with plenty of natural light and a well-appointed en-suite bathroom.

From the appellant the District Judge received written and oral evidence from Professor Andrew Forrester, Consultant and Honorary Senior Lecturer in Forensic Psychiatry, Dr Alan Mitchell, currently Chair of the Independent Prison Monitoring and Advisory Group for HM Chief Inspector of Prisons in Scotland, and Professor Coker, Professor of Public Health at the London School of Hygiene and Tropical Medicine. For present purposes, we may dwell only on the evidence of Professor Forrester.

Professor Forrester’s written evidence was contained in an initial psychiatric report dated 15 December 2019 and addendum reports dated 4 March, 27 April, 27 August and 3 October 2020. Before the District Judge there was no evidence from a psychiatric expert instructed by the GoI and Professor Forrester’s evidence effectively went unchallenged. It was only at the very end of 2021 that the appellant agreed to submit to examination by an expert instructed by the GoI and to disclose his prison medical records.

As at August 2020, which was shortly after the first Covid lockdown, Professor Forrester’s opinion was that the appellant met the criteria for a diagnosis of recurrent depressive disorder, current episode severe without psychotic symptoms; that the pandemic was likely to have played a significant role in the worsening of his mood; and that the risk of suicide was high albeit not immediate. That risk was linked to the recurrent depressive order, a family history of suicide, and his reports of suicidal ideas and intentions. Professor Forrester considered that the appellant met the criteria for detention under section 48 of the Mental Health Act 1983. If extradited, the appellant’s mental health would likely deteriorate. It was possible that he might become unfit to plead.

It appears that Ms Helen Malcolm KC’s cross-examination of Professor Forrester was limited. He stated that he “would be prepared to discuss his treatment with a practitioner in India”.

The District Judge addressed Article 3, EHCR before section 91. He described the assurances as “comprehensive” and placed “great weight” on the video of Barrack No 12, concluding that it gave a fair picture of the conditions that the appellant would face as well as being “extremely corroborative” of the assurances. The District Judge further considered that the assurances were reliable and would be honoured, and made the following point at [213] of his ruling:

“The GoI well know that if the assurances are broken, they will be very publicly broken in light of [the appellant’s] high profile. Just as [the then Senior District Judge] observed in her judgment in Mallya, I have no doubt [the appellant’s] lawyers would report any breach of assurance to this court as well as the Courts in India. That in turn would create “a perfect storm of publicity” as the Senior District Judge concluded in Mallya. That conclusion is equally apposite in this case. Extradition arrangements work on the basis of trust and any failure to abide by the assurances given by the GoI in [the appellant’s] case would doubtlessly affect the trust between this court and the GoI. I have no reason at all to think that the GoI will want to breach that trust by not upholding their assurances provided in support of this extradition request.”

That point was repeated before us by Ms Malcolm using slightly different language.

The District Judge did not address the psychiatric evidence under the rubric of Article 3. His conclusions on section 91 were quite brief. In essence, they were that the recent deterioration in the appellant’s mental health was attributable to the restrictive, Covid-related regime at HMP Wandsworth; that the regime awaiting the appellant at Barrack No 12 would represent an “amelioration” of his current conditions of detention; that the therapeutic regime recommended by Professor Forrester was sufficiently guaranteed by the assurances; and then, in the context of the suicide risk (at [225]):

“… when considering the criteria in Turner v Government of the United States of America [2012] EWHC 2426 (Admin) the evidence presented does not in my assessment meet the high threshold to satisfy me that [the appellant’s] mental health condition is such that it would be unjust or oppressive to extradite him. Albeit risk of suicide is assessed as high, Dr Forrester confirms in his report that there are no immediate suicidal intentions. [The appellant’s] mental condition is not such that it removes his capacity to resist the impulse to commit suicide. It is clear from Dr Forrester’s report that while in HMP Wandsworth the ACCT provisions can be deployed to safeguard against risk and having considered the assurances provided by GoI it is clear the Indian authorities have capacity to cope properly with the appellant’s mental health and suicidal risk, bolstered by [the appellant] being able to access private treatments from clinicians. I also weigh up the strong public interest in giving effect to extradition treaty obligations.”

Since the decision of the District Judge, a mass of additional evidence has been placed before the Court. Some may be described as public domain materials evidencing inadequacies in the Indian prison system in general and the prison at Arthur Road in particular. We have read all of those materials and have taken them into account. It is not necessary to provide a separate review of that evidence in this judgment, though we shall refer to it as necessary to explain the background to the seeking of assurances that has been a prominent feature of the case. In addition, there has been updating and development of psychiatric evidence concerning the appellant’s health and the risks that he may face and pose in the event that he is extradited. We deal with this evidence separately at [21-65] below. Going hand in hand with the production of that evidence, further assurances have been sought from and provided by the Indian authorities, which it will be necessary to review in some detail: see [66-82] below.

There has been considerable delay in this case since Chamberlain J’s grant of permission 14 months ago. This Court has stated more than once that, in appeals involving a risk of suicide, the substantive hearing should be listed as soon as possible. In that context we should explain that in December 2021 we adjourned the appeal so that questions could be posed by the Court of the Indian authorities by way of clarification if not amplification of various points that had been made in further assurances, and in June 2022 we were required to rule on the appellant’s submission that the case should be remitted to the District Judge to make evidential findings rather than be retained by us. The parties have provided an agreed chronology of the procedural history for which we are grateful but which it is unnecessary to set out.

We are assisted by having received written and oral evidence from two psychiatrists of very high expertise. Professor Andrew Forrester is Professor of Forensic Psychiatry at Cardiff University and works as an experienced consultant in Forensic Psychiatry with Oxleas NHS Foundation Trust. Professor Seena Fazel is Professor of Forensic Psychiatry at the University of Oxford and also has extensive experience of working in hospitals and prisons. He is acknowledged as having a worldwide reputation for his work on suicide in prisons. Each provided reports and, together, they provided a joint report dated 30 August 2022.

It is necessary to refer to the evidence of the experts in some detail; but it is convenient at the outset to identify matters that are not in dispute between the experts or the parties. First, Mr Modi suffers from a depressive illness which is recurrent in nature and which fluctuates in severity. Second, as the extradition proceedings are a causal factor in his depression, an adverse decision in the extradition proceedings could lead to an increase in the severity of his depressive illness. His suicide risk in the context of extradition can be described as “elevated” when compared with the general population and has been described as “substantial”, though this last term is qualitative rather than quantitative.

Professor Forrester approached his reference to diagnosis of Mr Modi’s condition by reference to the provisions of ICD-11, whereas Professor Fazel’s reports, at least initially, concentrated on the predecessor provisions of ICD-10. We were told that ICD-10 has remained in widespread use even since the formal introduction of ICD-11. This was disputed in a letter from those representing Mr Modi after the hearing. We are not persuaded (if, indeed, it was being suggested) that reliance on and reference to ICD-10 rather than ICD-11 would cause any difference of substance in the diagnosis of Mr Modi’s mental illness by the experts; and we are not persuaded that Professor Fazel’s greater reliance on ICD-10 undermines the value of his overall clinical judgment. Each iteration calls for a clinical assessment based upon multiple features of a patient’s presentation – and Mr Modi’s presentation has been the same whether he was being assessed by reference to ICD-10 or ICD-11. It is possible that the ICD-10 approach of identifying core and other features might lead in a borderline case towards a particular level of assessed severity where ICD-11 might allow for greater flexibility. However, the features and reasoning supporting the clinical judgments of each expert were clear and we do not consider that any further consideration of the differences between ICD-10 and ICD-11 would be profitable. It is, however, convenient to set out the provisions of one set of criteria and, if only because they are the later, we set out below extracts from the provisions of ICD-11 that are most relevant to the present case. These citations highlight the fact, which was emphasised by each expert, that there are multiple features with which a diagnostician has to grapple in exercising their clinical judgment.

Recurrent depressive disorders are described at 6A71 as:

“Recurrent depressive disorder is characterised by a history of at least two depressive episodes separated by at least several months without significant mood disturbance. A depressive episode is characterised by a period of depressed mood or diminished interest in activities occurring most of the day, nearly every day during a period lasting at least two weeks accompanied by other symptoms such as difficulty concentrating, feelings of worthlessness or excessive or inappropriate guilt, hopelessness, recurrent thoughts of death or suicide, changes in appetite or sleep, psychomotor agitation or retardation, and reduced energy or fatigue. There have never been any prior manic, hypomanic, or mixed episodes, which would indicate the presence of a Bipolar disorder.”

Recurrent depressive disorders are described by reference to whether the current episode is mild, moderate or severe and, in the case of moderate or severe current episodes, by reference to whether there are or are not psychotic symptoms. Thus at 6A71.3 “Recurrent depressive disorder, current episode severe without psychotic symptoms” is described as:

“Recurrent depressive disorder, current episode severe, without psychotic symptoms is diagnosed when the definitional requirements for Recurrent depressive disorder are met and the current episode is severe and there are no delusions or hallucinations during the episode. A depressive episode is characterised by a period of depressed mood or diminished interest in activities occurring most of the day, nearly every day during a period lasting at least two weeks accompanied by other symptoms such as difficulty concentrating, feelings of worthlessness or excessive or inappropriate guilt, hopelessness, recurrent thoughts of death or suicide, changes in appetite or sleep, psychomotor agitation or retardation, and reduced energy or fatigue. In a severe depressive episode, many or most symptoms of a Depressive Episode are present to a marked degree, or a smaller number of symptoms are present and manifest to an intense degree. The individual has serious difficulty continuing to function in most domains (personal, family, social, educational, occupational, or other important domains).” (Emphasis added)

The main distinction between the descriptions of current episodes being severe, moderate or mild lie in the replacement of the italicised words above with, respectively:

“In a moderate depressive episode, several symptoms of a depressive episode are present to a marked degree, or a large number of depressive symptoms of lesser severity are present overall. The individual typically has considerable difficulty functioning in multiple domains (personal, family, social, educational, occupational, or other important domains).”

Or

“In a mild depressive episode, the individual is usually distressed by the symptoms and has some difficulty in continuing to function in one or more domains (personal, family, social, educational, occupational, or other important domains). There are no delusions or hallucinations during the episode.”

As with the assessment of the severity of his depression, assessment of the risk of suicide is ultimately a matter of clinical opinion based on multiple contributing factors, both beneficial and adverse. It is broadly common ground between the experts that a risk of suicide increases with the severity of depression (though there is no suggestion that there is anything as simple as a linear relationship between the two); and that the presence of psychotic symptoms (which are wholly absent in the present case) would be a material additional risk factor for suicide. Although there are no bright lines on the spectrum of seriousness of a recurrent depressive order, it is plain from the evidence of the experts that the risk of suicide is materially elevated once the underlying depressive disorder is to be characterised as severe rather than as moderate or mild.

One other point may be mentioned here. Professor Forrester’s early reports were commissioned with a view to supporting applications for bail. In cross-examination by Ms Malcolm he acknowledged that people with moderate depression are often managed in a community setting. This, of course, is not of universal application and it is not possible to extrapolate from the general so as to make meaningful deductions about the level of risk of suicide in Mr Modi’s case; but it provides some “colour” when the Court is assessing the implications of a recurrent depressive disorder being described as mild, moderate or severe.

The main difference between the opinions held by the experts is that Professor Forrester tends to regard both Mr Modi’s depression and the risk of suicide that he presents as rather more serious than does Professor Fazel. While recognising that Mr Modi’s illness fluctuates, both Professors have provided “snapshot” assessments of his condition at the times when they have seen him as well as taking into account his presentation over time. That presentation is evidenced by observations by other physicians and health professionals (as well as those of the Professors) and by information contained in Mr Modi’s medical notes while in custody and his previous medical history. Professor Fazel accepted that, as he had not seen Mr Modi at the times when others had concluded that he was suffering from a severe episode, he could not offer a direct critique or comparative diagnosis as at the time of that snapshot: of course, the other side of the coin is that others were not present or presented with identical material as Professor Fazel when he reached his conclusions. That said, both he and Professor Forrester relied upon the observations of others to a greater or lesser extent, pointing to items that tended to suggest greater or lesser severity.

Mr Modi’s medical records from HMP Wandsworth state (as recorded by Professor Forrester) that he was seen by Dr Blackwood (then the Consultant Psychiatrist at HMP Wandsworth) on 6 August 2019, at which time he denied active suicidal ideation and that he was being treated for a mixed anxiety/depressive disorder although he had “no current active [symptoms] of [the] same”. He was on the minimum therapeutic dose (20 mg) of Fluoxetine. It does not appear that Dr Blackwood carried out a full inquiry into all symptoms. However, the following day a referral was made for counselling services in which it was noted that he “has a treated mixed anxiety/depressive picture (very mild)”.

Professor Forrester first assessed Mr Modi in September 2019, which led to his report dated 15 September 2019. The history taken by Professor Forrester included the death by suicide of Mr Modi’s mother in his presence when he was 8 years old. He had a history of depression, having become depressed in 2008 during a period of business difficulty. His symptoms resolved within about 12 months and he remained well until February 2018 when he was diagnosed as suffering from a recurrent depressive disorder. On examination Mr Modi spoke of reduced sleep, fluctuating appetite, reduced levels of interest and enjoyment, feelings of hopelessness and worthlessness, reduced energy and increased fatigue. He said that he had experienced suicidal thoughts since March 2018 (i.e. before his arrest) and that they involved either jumping or starvation and that he would kill himself if extradition was ordered. Professor Forrester’s opinion was that, based on the previous depressive episode in 2008, Mr Modi met the criteria for a diagnosis under ICD-11 of recurrent depressive disorder, current episode moderate, without psychotic symptoms. The current episode started before he was received into prison custody but Professor Forrester’s opinion was that incarceration had, in all probability, been associated with both a deterioration in, and a prolongation of, his presenting symptoms. He noted that, although Mr Modi did not currently present with suicidal plans or intentions, such plans and intentions could develop as he continued to lose hope. For that reason he would be concerned about a further acute deterioration with more intense suicidal ideas, and with suicidal plans and intentions if extradition were to be ordered. Detention in India would, in all likelihood, be associated with deterioration in his condition and he would be likely to present as a high and persistent risk of suicide in such circumstances. Other features, such as the destruction of his business and reputation were described as “key factors” in the development of his current episode of depression and the subsequent worsening and prolongation of symptoms, as were Mr Modi’s belief that he would not receive a fair trial if returned to India.

This first report establishes the foundations of Professor Forrester’s thinking. Those foundations, though taking into account the fluctuations in Mr Modi’s presentation over time, have not changed and, in our judgment, retain the essence of Professor Forrester’s opinion to this day.

Professor Forrester reassessed Mr Modi face-to-face in February 2020 and produced an addendum report in March 2020. His conclusions about diagnosis and future risks were unchanged. Professor Forrester had access to Mr Modi’s medical records from HMP Wandsworth, which recorded that he was being managed under an ACCT (i.e. Assessment, Care in Custody and Teamwork) system designed for the management of people thought to present a suicide risk. We were told that Mr Modi has been managed under an ACCT during four periods since being in custody. Mr Modi reported to Professor Forrester that he was very down and worse since their last meeting. When asked if he felt suicidal, Mr Modi said that he had concerns about confidentiality and, for that reason felt unable fully to report his experiences. Professor Forrester recorded that he did not describe active suicidal intentions or plans; but he was concerned at Mr Modi’s inability to report his mental state fully. This could, in Professor Forrester’s opinion, present significant harm to his health in the future if he were to develop intense suicidal ideas and intentions but feel unable to report them. Although he had not repeated his statement that he would kill himself if extradited, Professor Forrester considered that his earlier statement should be taken very seriously given the known risk factors of his documented history of depression and his family history of completed suicide.

His medical notes record that Mr Modi was no longer being managed under an ACCT on 18 March 2020.

After assessing Mr Modi by video-link on 20 April 2020, Professor Forrester provided a further report, dated 27 April 2020, in which his position shifted somewhat. His diagnosis of recurrent depressive disorder, current episode moderate, without psychotic symptoms remained unchanged. But he identified a qualitative worsening of Mr Modi’s mood, with the development of suicidal ideas, in the absence of suicidal plans or intentions. He attributed the deterioration in Mr Modi’s presentation to a number of factors including the pandemic and consequent lockdown regime in HMP Wandsworth, the impossibility of social distancing in the prison, the withdrawal of his counselling because of lockdown, a reduction in legal and social visits, and the reduction in opportunities for exercise. Mr Modi had reported to him that, because of Covid restrictions and the absence of all necessary papers he was unable to prepare his legal case properly. He said that he tended to read for between three and four hours a day, reading newspapers (Times, Telegraph, Guardian, Economist) as well as poetry, classics and crime fiction, but said that his concentration was poor (“not as good as it used to be”).

On 28 April 2020 the medical records document a response to a letter from Mr Modi’s solicitor saying that he was feeling suicidal. The nurse recorded recent concerns (including a recent prison suicide involving a young man known to Mr Modi) and that he was low in mood but that he “denied having current suicidal/[deliberate self-harm] thoughts”. The nurse noted that there was “no evidence/report of depressive or psychotic symptoms” and that he was eating well and enjoying reading.

Professor Forrester next assessed Mr Modi in August 2020, the interview being by video-link. Mr Modi described a deterioration in prison conditions and his mood, including reduced ability to concentrate and the development of suicidal intentions. The changes described by Mr Modi led Professor Forrester to the opinion that he now met the criteria for a diagnosis under ICD-11 of recurrent depressive disorder, current episode severe, without psychotic symptoms. He identified a worsening of the depressive symptoms in the four week period before he saw him, with the presence of daily depressed mood, difficulty attending and concentrating, feelings of hopelessness and guilt, thoughts of death and suicide, reduced appetite with weight loss, sleep disorder, mild psychomotor retardation and reduced energy with increased fatigue. Professor Forrester considered that the restricted prison conditions associated with lockdown had played a significant role in causing the worsening in Mr Modi’s mood; and he considered it significant that Mr Modi presented with persisting suicidal ideas, now with some suicidal intentions, albeit that the intentions were not immediate in their nature. In his opinion, Mr Modi should now be considered at substantial (meaning high), albeit not immediate, risk of suicide because he had “assessed his current risk of depression as severe” and because of the deterioration during his time in HMP Wandsworth, the deterioration over the previous four weeks, the close family history of suicide and because he now reported both suicidal ideas and intentions. It was his opinion that Mr Modi was not receiving the care and treatment he needed and that he now met the criteria for detention under Part III of the Mental Health Act 1983. He considered that Mr Modi’s condition would deteriorate further if he were extradited, relying upon his general response to imprisonment in the UK, progressive deterioration over many months in custody, and the complicated nature of the criminal proceedings he would be likely to face, particularly if his concentration and general ability were to be compromised by a severe depressive illness.

The medical records disclose that Dr Blackwood reviewed Mr Modi on 15 September 2020. We have not seen the relevant entry. It is reported in Professor Fazel’s April 2022 report that Dr Blackwood thought that Mr Modi’s affect was flat and that Mr Modi reported suicidal ideas on occasion. Dr Blackwood concluded that Mr Modi was experiencing a “moderate depressive episode currently treated with Fluoxetine” at that point.

Professor Forrester’s next report, dated 3 October 2020, considered the assurances that had been provided by GoI at that point. Professor Forrester did not meet Mr Modi or reassess him for the purposes of the report, which does not add significantly to the diagnostic and prognostic opinions that he had previously expressed.

As reported by Professor Fazel, Dr Blackwood reviewed Mr Modi again on 27 October 2020 and thought he was less flat in mood. Dr Blackwood reported that he was able to laugh, and noted that Mr Modi wanted a role at the prison library. Thereafter Mr Modi engaged in 18 sessions of psychological therapy from November 2020 to April 2021. The psychotherapist conducting the sessions reported that he engaged well and reported that the sessions had been very useful. His mood was reported as stable (though it had fluctuated on occasions from week to week from “low mood” to “stable”). He found conversations with his family “very enjoyable” but they became difficult after an adverse result from his case in February 2021 when he was put on an ACCT.

After permission to appeal had been given by Chamberlain J, Professor Forrester was asked to give his opinion of issues relating to the propositions in Turner, which he did in October 2021. In briefest outline, his opinion was that “it is very likely that the operative cause of Mr Modi’s high and persistent risk of suicide would be depression.” In providing that opinion he referred to studies identifying an association between depression and completed suicide with the risk of suicide increasing as the level of severity of the depression increases. We will return to those studies (and others identified in Professor Fazel’s first report) when we consider the issues arising in relation to Turner, below at [112-130]. For present purposes the most informative aspects of the studies are the findings in Bradvik (2018) that suicides account for 1.4% of all deaths internationally and that most of these deaths “are related to psychiatric disease, with depression, substance use, disorders and psychosis being the most relevant risk factors.” Bradvik estimated the lifetime risk of suicide for mental disorders including depression alcoholism and schizophrenia to be between 5 and 8%. Nordentoft et al (2011) reported similar rates in those with clinical depression diagnoses over 3 decades. According to another paper by Bradvik and others (2008), there is a clear association between the severity of a person’s depression and the long-term suicide risk, which was estimated at 3.1% for those with medium severity and 11.4% for those with severe depression.

In December 2021 Professor Fazel provided his first report. He had not been able to interview Mr Modi and did not have access to his medical records, for reasons that are contentious and do not matter for present purposes. He was therefore constrained to rely upon Professor Forrester’s reports, and upon his summary of some of the medical records relating to the period from 31 January to 14 May 2020 which, in Professor Fazel’s view, did not support the August 2020 diagnosis of severe depression. Given the disadvantages under which Professor Fazel was then operating, he did not contradict Professor Forrester’s August 2020 assessment that the depressive episode was then severe; but he did identify points (and the absence of observations supporting such an assessment) in the reported medical notes and suggestions that Mr Modi retained the ability to experience pleasure as giving grounds to question a diagnosis of moderate or severe depression. Turning to the suicide risk, he expressed the opinion that it was not possible to assess the current suicide risk by reference to an assessment from the previous year when Covid-related restrictions had been at their height.

In March 2022, Professor Forrester provided a further report in which he gave his opinion about the assurances that had then been provided by GoI. He did not carry out a further reassessment of Mr Modi’s condition and prognosis for the purposes of preparing that report.

Also in March 2022 Professor Fazel provided a report (later re-dated as May 2022) having now had the opportunity to interview Mr Modi for two hours in February 2022 and to see Mr Modi’s medical records for the period from 20 March 2019 to 14 May 2020. In addition he spoke to Dr Blackwood. In conversation, Dr Blackwood is reported by Professor Fazel to have said that Mr Modi was not under the care of the Wandsworth mental health inreach team and that he had been only mildly depressed at most in the past but that Dr Blackwood did not think he was clinically depressed. It is not entirely clear what prior involvement Dr Blackwood had had with Mr Modi or what enquiries he may have made. It is material to note that, at this stage, the medical records available to Professor Fazel did not cover or disclose the fact of Dr Blackwood’s reviews in September and October 2020, and Dr Blackwood appears not to have alerted him to them.

Professor Fazel’s opinion was that Mr Modi currently had a clinical diagnosis of depression, which was of mild severity. He rejected a characterisation of Mr Modi’s depression as severe because he had no psychotic symptoms, maintained some reactivity in his mood, and was able to experience some enjoyment when speaking with his children. He referred to Mr Modi reading newspapers and speaking to his family daily on the phone. He also relied upon the assessment of the psychiatric team at HMP Wandsworth and the absence of reported indicators of more serious depression, though he acknowledged the absence of up-to-date medical records. Turning to the question of suicide risk, Professor Fazel accepted that it is elevated and that there are risk factors associated with his being in custody - including his underlying clinical depression and hopelessness about his situation if he is extradited. He recognised that the risk may become further elevated if things go against Mr Modi; but he considered that some features may be modified in ways that help to ameliorate the risk even if he were to be extradited including, for example, how the Indian authorities manage his safety and what his perception of the possibility of release may be.

Professor Fazel provided a follow up report in April 2022 after being provided with the medical records for the period from May 2020 to December 2021 which we have summarised above. The records did not change his opinion that he had a clinical diagnosis of depression of mild severity. Professor Fazel relied upon Mr Modi’s beneficial reaction to counselling and the fact that his mood is reactive and changes according to circumstances, which he considers to be more suggestive of a mild depressive episode than a more severe one. His continued close contact with his family was identified as a further protective feature.

On 6 April 2022, as recorded in the medical notes, Dr Zachariah (consultant psychiatrist at HMP Wandsworth) carried out a mental health review follow up. Mr Modi reported extreme sadness about his current circumstances, suffering caused to others, and loss of contact with family. On mental state examination it was reported that most times he experiences severe despair with no hope of a fair trial. He was having thoughts of ending his life and even had suicidal thoughts when he was in the community before being remanded. His sleep was currently good and his appetite variable. He reported worthlessness and hopelessness with the current situation. He hardly enjoys any activities in prison but is happy to see the family and speak to them. Dr Zachariah’s impression as recorded in her note was that he was suffering from moderate depression. In a communication to Professor Forrester she referred to him coping well on the house block but being “moderately/severely depressed and on 20mg of Fluoxetine” with a risk of suicide that was “not immediate, but very real.”

The medical notes record that Mr Modi was assessed by an assistant psychologist on 13 April 2022 who applied a psychological tool (CORE-10) which indicated “severe psychological distress”. He was described as reporting some suicidal thinking but not having any active plans of self-harm or suicide. The assessor’s opinion was that there were no immediate concerns regarding risk to himself. He was added to the list for a mood management workshop.

On 2 May 2022 Professor Forrester provided a further report based on updated material and an interview with Mr Modi by video link on 16 March 2022 (i.e. before Mr Modi was seen by Dr Zachariah on 6 April). His opinion now was that Mr Modi continued to meet the criteria for a diagnosis of recurrent depressive disorder, current episode moderate, without psychotic symptoms. His illness fluctuates, but there is no evidence of psychotic symptoms at any stage. He remained of the view that Mr Modi’s risk of suicide should be considered high, or substantial, within the context of extradition.

In a passage with which Professor Fazel has agreed, Professor Forrester pointed out that Mr Modi had now been assessed by four Consultant Forensic Psychiatrists (Forrester/Fazel/Blackwood/Zachariah) each of whom had at some stage used the word “moderate” to describe the intensity of his depressive illness. Professor Fazel also agrees that Mr Modi’s is a fluctuating condition which is likely to explain the variations in description and attributions of severity from time to time.

One of the notable features of Professor Forrester’s report of 2 May 2022 is an oblique but clear questioning of the competence of the prison psychiatric services in general and the services at HMP Wandsworth in particular. We are not in a position to enter into an assessment of the competence of UK prison services either in general or at HMP Wandsworth in particular; but we note Professor Forrester’s questioning of the competence of the recording of interventions or observations in Mr Modi’s medical records. While we take that questioning into account, it does not lead us to discount the importance of the medical records as a source of information, even if we have to scrutinise them with care.

The medical notes record that Mr Modi was assessed by a specialist registrar psychiatrist, Dr Cleall, on 25 May 2022. On interview it is recorded that there was no change to his mental state, which was low throughout the day. There was daily communication with his family. His energy levels were very low. He slept a lot during the day and could only focus on a few pages of a book a day. Despite fleeting thoughts of “not wanting to be around any more” he denied suicidal ideation and had no current plans to harm himself. There were no psychotic symptoms. On mental state examination, Dr Cleall noted psychomotor retardation, poor eye-contact, poverty of speech and low mood with a flattened affect. Dr Cleall considered that his presentation was consistent with severe depression and increased his dose of Fluoxetine from 20 mg to 40 mg daily (the therapeutic range generally being between 20 and 60 mg). He considered that there should be a low threshold to commence ACCT if any deterioration in mental state or increased risk to self were to occur.

On 8 June 2022 the notes record that Mr Modi presented as calm and pleasant on interactions and reported he was doing ok. He denied any current thoughts or plans of suicide or self-harm and said that his sleep and appetite were ok. On a further brief review on 24 June 2022 he reported that he was doing well and was observed to be reading a financial newspaper in his cell with his cell mate. He denied any thoughts or plans of self-harm or suicide. No overt issues with depressive features were observed. He was calm and pleasant in mood and on interactions. He was seen regularly by the nurse and other members of the team through June and July and, on 15 July, was noted to be under the care of the psychiatry clinic and was on the psychology waiting list.

Each expert reassessed Mr Modi shortly before the present hearing. Professor Forrester’s report was dated 31 August 2022, after assessment by video link on 28 July 2022. In addition, he was provided with medical records from 22 December 2021 to 2022. His opinion, in summary, was that Mr Modi continues to meet the criteria for a diagnosis of recurrent depressive disorder, current episode moderate, without psychotic symptoms; and that Mr Modi presents a high, or substantial, risk of suicide within the context of extradition. On examination Mr Modi said that he was feeling depressed all the time, with no variation. He sleeps up to 15 hours in a 24 hour period, sometimes eats a lot while at other times he does not eat, and his weight has increased from 65kg to 82kg currently. He reported that he could not read books, for lack of concentration. Asked about the visits he had received from his family, he said that he was just sad, everything was hopeless and pointless and he considered himself to be worthless. He said that he was experiencing suicidal thoughts all of the time. He did not wish to discuss them but said he would kill himself if returned to India. Professor Forrester noted a general slowing of bodily movements (in keeping with a diagnosis of depression) falling short of frank psychomotor retardation, which would be a marker of severity. His mood was subjectively and objectively depressed, his affect flat and lacking in reactivity, but with no evidence of other abnormalities. His thoughts showed signs of slowing. He did not present with features that would indicate the presence of a psychotic illness.

In reaching his opinion on the level of severity of Mr Modi’s depression, Professor Forrester noted the fluctuations over time which, in his view, indicates the clear potential for the further emergence of severe depressive symptoms in the future. In his opinion, Mr Modi’s suicidal ideation has intensified over time and his mental state is likely to deteriorate if extradition proceeds. While acknowledging that the adequate provision of methods to prevent suicide can in most cases help reduce the risk of suicide, at least in the short term, it is his opinion that a suicide risk is likely still to arise in Mr Modi’s case, regardless of whether adequate methods of suicide prevention are used.

Professor Fazel re-interviewed Mr Modi on 12 August 2022 and produced his final report on 29 September 2022. Mr Modi told him that his mood was “up and down” or “the same” with some days being worse than others. The information he provided to Professor Fazel was largely consistent with that recorded by Professor Forrester in his latest report, with some points of interpretation being contentious. For example, Professor Fazel recorded being told that Mr Modi read the Guardian, Times, Telegraph and FT “everything, front to back”, which Professor Fazel understood to mean that he read from cover to cover whereas it was suggested to him it simply meant that Mr Modi did not head straight for the sports pages (or other pages of specific interest) to the exclusion of other sections. Mr Modi repeated to Professor Fazel that he thinks of suicide and that he has plans, though he was not prepared to disclose them. On mental state examination, Professor Fazel found him more animated and brighter in mood than in the last assessment he had conducted. His speech was normal in rate and rhythm and with no signs of psychomotor retardation. Professor Fazel thought his mood was low, although there remained some reactivity “with one or two wry smiles”. He said he was not actively suicidal. His attention and concentration were not obviously impaired and he responded to questions intelligibly. There were no abnormalities of thought content, possession or form and, as before, there was no evidence of psychotic symptoms.

Professor Fazel’s opinion was that Mr Modi continued to have a clinical diagnosis of depression of mild severity. The depression had “clearly improved” since his last assessment based on clinical examination and longitudinal assessment around 6 months apart. Although sustained low mood and fatiguability are present, in Professor Fazel’s opinion marked loss of interest or lack of pleasure is not clearly met. The evidence upon which he relies for this is Mr Modi’s reading of newspapers from front to back, the fact that Mr Modi experiences some pleasure in helping others, maintains an interest in reading and enjoys certain foods (though the extent to which these features were present was challenged in cross-examination). Psychotic symptoms have not been reported and recent medical records referred to Mr Modi “doing well” and being “calm and pleasant.” Mr Modi had told Professor Fazel that his mood is “sometimes good”, which is not typical in clinical depression and is more in keeping with a mild rather than a moderate or severe depression. Professor Fazel did not regard the score on the CORE-10 test (administered on 13 April 2022) to be clinically informative. His views on suicide risk were also unchanged. He considers it to be elevated in comparison with other prisoners of a similar age and sex to Mr Modi. He regards Mr Modi’s unwillingness to discuss his suicide plans as suggesting that he is able to think rationally about his plans, as opposed to being subject to impulsive thoughts. Furthermore, in a case of severe depression, it is Professor Fazel’s view that the subject will typically be unable to delay suicidal plans for a much later date.

The experts were cross-examined before us concisely, courteously and effectively. It emerged for the first time during the cross-examination of Professor Forrester that he had interviewed Mr Modi the previous week and that his conclusion was unchanged. While that comes as no surprise given the consistency of his opinion over time, this latest visit and assessment were not the subject of further evidence.

Professor Forrester’s opinion remained that Mr Modi’s risk of suicide “might lay at some point at high or substantial in the context of extradition.” He could not say exactly when, but it would be sooner or later and he could not guarantee that it would only be in the initial period after extradition. In cross-examining him, Ms Malcolm accepted that there would be a period of high risk but suggested it would settle as he became accustomed to his new circumstances. Professor Fazel’s evidence was rather more nuanced. He did not agree with Professor Forrester’s characterisation of “unalterable” features that would determine Mr Modi’s mental condition because, in his opinion, there may be unpredictable variation both initially and over time; and Mr Modi’s perceptions (e.g. as to the risk of being killed or about his state of isolation from his family) may change with time and altered circumstances. Largely for that reason, as it seems to us, he was unwilling to make a prediction of risk of suicide before Mr Modi has been extradited. But he accepted as a realistic possibility that the risk of suicide may increase after extradition.

We have outlined the evidence about Mr Modi’s fluctuating condition above. In our judgment a fairly clear picture emerges. For much of the time since he was first re-diagnosed as suffering from depression (which was before his arrest and remand in custody), his condition had fluctuated around a level that has reasonably been assessed as being of moderate severity without psychotic symptoms. There was a deterioration in the summer of 2020, to which the restrictions associated with Covid made an important contribution, so that his current episode was realistically assessed as being severe. Fortunately, even when subject to an episode that was assessed as being severe, he has suffered no psychotic symptoms; and (consistently with the fluctuating nature of his condition) the level of severity reduced again so that the predominant view up to 2022 was that his condition was of moderate severity.

The position in 2022 is less clear cut, partly because we have the benefit of Professor Fazel’s evidence to set alongside that of Professor Forrester; and partly because the medical notes (which, as we have noted, are themselves the subject of criticism from Professor Forrester) provide indicators going in different directions. We consider that Professor Fazel is justified in identifying indicators that would tend towards a finding of mild severity, but the evidence is not all one way and fluctuates with time.

We note that Dr Cleall is the only person to have recorded frank psychomotor retardation. It is not clear whether this is attributable to a temporary fluctuation in presentation and seriousness, or to a difference in approach and (possibly) expertise on the part of Dr Cleall. In circumstances where we have not heard from Dr Cleall, the weight we place upon his findings as recorded in the medical notes is somewhat reduced, though we recognise that both Professor Fazel and Professor Forrester paid attention to what he had said. Overall, for the reasons we have just indicated and the criticisms made by Professor Forrester of the apparent level of expertise evidenced by the medical notes, we place less weight upon the recorded findings of Dr Cleall and others from whom we have not heard than we do upon the findings and opinions of those from whom we have.

Viewed overall, the primary evidence of the experts and the secondary evidence contained in the medical notes support the conclusion that this is and remains a fluctuating condition where the predominant level can be described as being of moderate severity, albeit with features that tend to suggest a less severe case within that categorisation. Ultimately, we do not think that there is any very substantial difference of substance between Professor Fazel and Professor Forrester. Where there is a difference of emphasis we consider that it can be accommodated in the way that we have described above and our acceptance that Professor Fazel’s identification of features tending to indicate lesser severity is generally reasonable.

Of more importance is the experts’ assessment of Mr Modi’s likely progress if he is extradited. We accept that the time immediately surrounding and after any extradition would be stressful and difficult for Mr Modi and therefore would carry a risk of detrimental fluctuation in his overall condition. We are not persuaded that it can be predicted that he will experience deterioration or fluctuation so that his “snapshot” condition would be described as being severe, though there is a risk that could happen. Even so, there is no evidence upon which we could conclude that any fluctuation would not be susceptible to treatment. Viewed overall, and on the basis of the evidence we have summarised above, we conclude that the most likely outcome is that Mr Modi’s condition will generally not be worse than of moderate severity. While we agree with Professor Fazel that it would be preferable and more reliable to defer the assessment of the risk of suicide that Mr Modi would present until after any extradition, that cannot absolve us from making an assessment now for the purposes of this judgment. In our judgment it is clear that there may well be a heightened risk of suicide on and after any extradition, though we cannot make a finding as to when that is likely to happen, not least because we accept Professor Fazel’s evidence that Professor Forrester’s concept of “unalterables” is less clear-cut and more fluid than Professor Forrester would suggest; or, putting it slightly differently, it is not predictable how Mr Modi will react to altered or altering circumstances after any extradition.

We cannot quantify the risk of suicide and do not attempt to do so. It may reasonably be described as high or substantial in the context of extradition. Two observations may be made. First, it is a beneficial diagnostic indicator that Mr Modi has never shown any signs or features suggesting a psychotic illness. And, second, the papers to which we have referred at [41] above indicate that, although elevated compared with the general population, the risk he presents will be significantly less than would be the case if the general course of his depressive illness had been severe rather than moderate.

From the outset, the appellant has expressed and evidenced concerns about the quality of the Indian prison system, the conditions in the Arthur Road prison where he would be held if extradited, and the availability of adequate medical services should he be imprisoned in India. The District Judge received two letters of assurance, dated 8 June 2019 and 11 September 2020, to which we have referred above. He summarised them extensively (at paragraphs 160-179) and took them into account (at paragraphs 208-225) in reaching his conclusions that (a) the assurances were reliable, (b) the accommodation provided by Barrack 12 was satisfactory, and (c) the assurances relating to medical care were sufficient even though they had not specifically addressed Professor Forrester’s concerns that there needed to be “medication, high intensity psychological interventions, electroconvulsive therapy, crisis service, combined treatments, multi-professional and inpatient care”.

Both before and since the decision of the District Judge there has been a pattern of assurances being given which have then been taken as the basis for further suggested inadequacies and questions. It was in an attempt to bring finality to the apparently iterative development of the appellant’s stated concerns and the GoI’s attempts to respond that the Court adjourned the appeal on 14 December 2021 and gave directions for the formulation of questions and the provision of responses to those questions. The list of questions was finalised by 6 January 2022. The GoI responded on 3 February 2022 by a letter of assurance and accompanying materials.

It is neither necessary nor convenient to chart the progression of concerns and assurances chronologically. It is common ground that much relevant and helpful material has been provided, not least in response to Professor Forrester’s stated concerns about the availability of suitable medical provision; and, during the hearing, the Appellant abandoned the contention that there was no adequate legal foundation for the assurances. We accept that the assurances are given with appropriate legal foundation and that the GoI and related authorities will treat themselves as duty bound to respect the Sovereign Assurances they have given. The Appellant has pointed to other cases where questions have been raised and criticisms made about compliance with Sovereign Assurances provided in those cases by the GoI. Even if those criticisms are taken at face value, they are case specific and no firm basis for doubting the genuineness or commitment underlying the assurances given in this case has been shown. In our judgment, there is every reason to accept that the GoI will treat its commitment as given in these assurances with appropriate seriousness. If anything, the imperative for the GoI to comply with its assurances will be enhanced by the fact that this is a high profile case so that its conduct and care for Mr Modi is likely to be subject to heightened scrutiny at all times. The GoI will surely appreciate that a failure to honour its assurances would be liable to have a significant adverse effect on the mutual trust that forms the basis of the extradition regime to which India and the United Kingdom are parties.

In these circumstances, we identify the outstanding concerns as identified in the Appellant’s skeleton argument for the current hearing and consider the assurances that have been given in response to those concerns.

First, the Appellant submits that the assurances still fail to provide a sufficiently detailed account of the medical care and treatment that would be made available to the appellant. This criticism is closely allied to the second, which is the Appellant’s submission that “in the absence of any existing protocols in Arthur Road for the management of mentally unwell prisoners, and/or the suicide risks such inmates may pose, there remain serious deficiencies in what has been “assured””. Specifically, the clinical and risk plans that exist (such as they may be) are alleged to be “fairly sketchy and vague, lacking the sort of detail that would be required to ensure a successful and safe clinical transfer”; it is said that Mr Modi’s care will largely be entrusted to prison officers rather than specialist mental health nurses; and, perhaps critically, it is said that there appears to be no comprehensive suicide prevention plan in place.

It is common ground that the Indian authorities do not have a pre-existing protocol for the care and treatment of prisoners who present a suicide risk. The GoI’s response, however is that it will make suitable and sufficient provision for the care and treatment of Mr Modi. It accepts that there is no personalised plan in place as yet; but it rejects the substance of the criticism that the Appellant makes in this regard. Specifically, if extradition is ordered:

The GoI has given specific assurances about the care available to Mr Modi in hospital if required. Specifically:

One point of contention has been the steps that would be required for Mr Modi to access private medical treatment. The GoI’s original assurance was that “Mr Modi may receive any relevant and necessary treatment from a private doctor or mental health expert of his choice, including treatment or counselling from a psychiatrist, psychologist as required and paid for by him.” That care can be in the prison or by video link as and when required. Further, “[i]f on medical advice he requires treatment in hospital, he can be moved to JJ Hospital, or if so advised by JJ Hospital, to a private hospital” with appropriate arrangements for his security.

Evidence was submitted on behalf of the Appellant that, since the position was not covered by the relevant prison rules, private medical care could not be accessed without a Court order. The GoI’s response was that “As already assured Mr Modi may be allowed treatment by a private doctor/mental health expert of his choice. However, the relevance and necessity of treatment by the private doctor/mental health expert will be decided by doctors/expert of JJ Hospital. This assurance is not in contravention within Maharashtra Prison Rules.” Mr Fitzgerald submits that this does not deal with the question whether a court order is required. We do not agree. In our judgment the position is clear (and it was expressly affirmed by Ms Malcolm on behalf of the GoI): the GoI has given an assurance that a court order will not be needed. The GoI has also stated that, “for optimal management of patient, psychiatrist in-charge of Mr Modi’s treatment at JJ Hospital shall liaise with the private consultant, if engaged by Mr Modi.”

For convenience, though it does not technically arise under this general heading, we record the assurances that Mr Modi may be visited by his legal advisers daily (except on Sundays and holidays) and by his family weekly or more often if approved by the doctors.

In Professor Forrester’s oral evidence he accepted that the GoI had “gone a very long way” to answer his questions. Both during cross-examination and in closing submissions, Ms Malcolm demonstrated that this acceptance by Professor Forrester was fully justified. In his submissions, Mr Fitzgerald concentrated on the fact that there is as yet no plan drawn up, even provisionally. He submitted that, with the information that is now available, at least a provisional plan could and should be in being, particularly since there is no pre-existing protocol for the care of suicide risk patients in general.

What was agreed between the experts in their joint statement under the heading “Assurances” was:

“However, we agree that a robust clinical plan should be put in place and agreed before any removal occurs. This plan should include the following details - full psychiatric review at the time when Mr Modi is received into an Indian prison (which can include a nursing and psychiatric assessment – Professor Forrester thinks a psychiatric assessment should be conducted straight away, whereas Professor Fazel thinks it can take place within a few days), the type of mental health care and treatments that will be provided, regular access to clinical care including the sort of care to be provided, and observations that may be applied if necessary.” (Emphasis added)

We consider that the position outlined in the Joint Agreement is reasonable and supportable. Although in his oral evidence Professor Forrester maintained that he would have hoped that at least a provisional plan would have been in place by now and certainly before extradition, a sense of perspective was given by his answer when asked whether, if roles were reversed and he was the receiving clinician in India, he would wish to draw up a provisional plan before speaking to the colleagues who had responsibility for Mr Modi in England and the provision of medical records and other information. His reply was that he would want the records to be available and to have a conversation and to see what exists in the case and then draw up a medical plan. That answer seems to us, with respect, to be entirely sensible; and it should be noted that the authorities in India do not yet have the records and have not yet had those conversations with their English colleagues.

The fact that a robust clinical plan has not yet been drawn up (even provisionally) does not indicate to us that there is a real risk of oppression if extradition is ordered. For a start, drawing up a detailed plan before now (even provisionally) may be said to be premature and potentially wasteful of time and resources since it has not been known whether the Court will order extradition. If we were faced by a blank canvas, there might have been room for concern; but we are not. The assurances provided by the GoI already encompass most aspects of the robust plan as proposed by the joint statement. Specifically, (i) there will be a full psychiatric review at or shortly after Mr Modi’s arrival at an Indian Prison; (ii) assurances have been given that the scope of the services and treatments that may be provided meet the requirements identified by Professor Forrester as important before the District Judge and again before us; (iii) assurances have been given about regular access to clinical care to be provided; and (iv) assurances have been given about observations.

Ms Malcolm pointed out that, if extradition is ordered, Mr Modi will remain the responsibility of the UK until handed over to be taken onto a plane. It is true that no assurances have been given that are specific to his time on a plane, save that his travel will be arranged with his medical condition in mind; but, wisely, that has not been the focus of Mr Fitzgerald’s submissions.

There is no specific assurance that a robust (but provisional) plan will be in place “before any removal occurs”: but does that matter? The “robust plan” that the experts have said should be available before removal is a plan detailing the steps that should be taken on and after his arrival in India. As we have said, the assurances that have been given encompass most if not all of the steps that the experts consider should be included in the robust plan.It is therefore clear that the GoI and others who would be responsible for Mr Modi’s care if he were to be extradited have already given considerable thought to the question how he shall be treated if extradited, even if they have not gathered those steps together into a formal plan as contemplated by the experts. The main substance of such a plan is therefore already in place. Although a concluded plan will have more detail, we reject the criticism that the assurances that have been given and the prospective treatment that they outline are “vague and sketchy”. On the contrary, the GoI has gone to considerable lengths and has provided considerable detail to address the concerns that have been raised in sufficient detail for present purposes.

For obvious reasons, if an order for extradition is made, it would be preferable for steps to be taken to draw the strands of the assurances together and start the process of formulating the plan as contemplated by the experts sooner rather than waiting until the date of extradition or just before it. Equally, as Professor Forrester recognised, any planning that is done before his arrival and assessment by those who will have responsibility for him in India must inevitably be provisional. That said, the extent of the existing assurances gives us confidence that, if an order for extradition is made, a robust (if provisional) plan for Mr Modi’s future health needs as contemplated by the experts can and will be put in place after liaison with Professor Forrester and (we anticipate) those responsible for Mr Modi’s care at HMP Wandsworth but before he leaves; or, if not then, the plan will be formulated either immediately on arrival in India or very soon thereafter at the latest. That being so, we do not regard the current absence of such a plan as a source for significant concern.

The parties have filed detailed skeleton arguments on the substantive issues. We may summarise their submissions respectively as follows.

Mr Fitzgerald criticised [225] of the District Judge’s decision. He submitted that the District Judge applied too narrow and restrictive approach to section 91 in holding that the risk of suicide was not immediate and in interpreting the fourth proposition in Turner to require proof, at the very least, of an inability to control an impulse. Mr Fitzgerald further submitted that it was unrealistic to suggest that there would be an amelioration in the appellant’s conditions overall were he to be extradited. On the contrary, he would find himself in a hostile environment even further away from his family.

Mr Fitzgerald invited us to prefer Professor Forrester’s evidence over Professor Fazel. Although the latter is a world authority on prison suicides, his perspective is more academic than clinical, and it was submitted that Professor Forrester had more “hands-on” experience in that regard. Moreover, Professor Forrester has had the considerable advantage of seeing the appellant on far more occasions than has Professor Fazel.

Mr Fitzgerald pointed out that Professor Forrester’s consistent diagnosis of moderate depression (save for the one occasion in August 2020 when he diagnosed a severe depressive episode) was not merely supported by other clinicians who examined the appellant in prison but was buttressed by a close and accurate application of the criteria set forth in ICD11. There were repeated references to suicide in the prison medical records, and suicide was central to the appellant’s diagnosis. Furthermore, Professor Forrester’s conclusion that the appellant’s mental condition would likely deteriorate if he were extradited was supported by both experience and common sense. If the appellant’s depressive disorder got worse, the suicide risk would necessarily increase. Mr Fitzgerald criticised Professor Fazel for concentrating on the “now” whereas the authorities make clear that a predictive exercise is required. It was said that in his oral evidence Professor Fazel overstated the difficulties that subsisted in conducting that exercise.

Mr Fitzgerald was also critical of Professor Fazel’s evidence in failing to apply ICD11 as opposed to ICD10 and in downplaying the views of clinicians at the prison that the appellant’s depression was at least of moderate severity. He also submitted that Professor Fazel in fact accepted in cross-examination that many symptoms of moderately severe depression were present, that the risk of suicide was elevated, and that if extradited that risk would increase.

Separately, Mr Fitzgerald submitted that the Court should either excise or rewrite the fourth proposition in Turner, which requires that the requested person’s condition be such that “it removes his capacity to resist the impulse to commit suicide”. Both experts were critical of this test, in particular its use of the term “impulse” which in the context of psychiatry is a term of art. Mr Fitzgerald submitted that it was anomalous that cases of severe depression should fall at this hurdle whereas cases of less severe depression but with a concomitant personality order might satisfy it. Mr Fitzgerald argued that Professor Fazel’s suggested solution, which was to import considerations germane to capacity under the Mental Capacity Act 2005 into Turner, was both unwarranted and unduly restrictive. He submitted that “capacity” and “impulse” could and should be read in a lay and common-sense way to mean “ability” and “compulsion”. That would sufficiently differentiate between the rational/voluntary act cases and those where the person has, or perceives that he has, no real choice.

Finally, Mr Fitzgerald was strongly critical of the various assurances given by the GoI. First, he characterised them as vague and unspecific, and submitted that we could not trust the GoI to be loyal to them, referring in particular to the ongoing breach of assurances by the GoI to Portugal in the case of Abu Salem and the recent refusal of the Portuguese Supreme Court of Justice to permit the extradition of another Indian national because of deficiencies in the assurances given.

Second, and relying on the evidence of Justice (Ret’d) Nandrajog dated 4 June 2022, Mr Fitzgerald submitted that the GoI’s assurances about access to private health care should be rejected: a court order would be required before a private doctor might be permitted to enter Barrack No 12. Third, he submitted that the assurances dealing with healthcare provision were in any event inadequate because the Indians had no suicide protocol (c.f. the ACCT regime in England and Wales, first developed in 2005) and there was no robust care plan in place in relation to the appellant. Professor Fazel had accepted on 30 August that such a care plan should be agreed and put in place before removal.

Mr Fitzgerald advanced other submissions which we have considered but it is unnecessary to summarise.

Ms Malcolm’s headline submission was that the appellant’s mental condition was not sufficiently serious to engage section 91, whatever the test as she put it, and that in those circumstances it was unnecessary to consider the adequacy or otherwise of the GoI’s assurances.

It was accepted that in the context of a fluctuating condition the appellant’s depressive illness might meet the threshold of moderate although there was evidence that it has been improving following an increase in the dose of his Fluoxetine. It was also accepted that an adverse decision in these proceedings “could lead” to an increase in the severity of the appellant’s disorder as well as the suicide risk, although Ms Malcolm invited us to consider the various imponderables (e.g. the appellant might be reassured on arrival at Barrack No 12) as well as Professor Fazel’s opinion that the increase in risk would subsist “in the immediate period following arrival in an Indian prison” although an assessment beyond that point was uncertain.

Ms Malcolm invited us to prefer Professor Fazel’s evidence over Professor Forrester’s. She submitted that ICD10 was still current and that in any event, as Professor Fazel has explained, the application of criteria such as these cannot involve a checklist but requires a clinical judgment in the light of all the available evidence. Ms Malcolm submitted that Professor Forrester has overstated the appellant’s current state and future risk. She pointed out that there had been no incidents of self-harm and that the reality here is that the appellant has suffered a “crash in prestige” in the context of such a serious case of alleged fraud.

Next, Ms Malcolm made a series of submissions on the fourth proposition in Turner. She accepted in writing, in line with the view of Professor Fazel, that difficulties would arise if “impulse” were given a specific psychiatric meaning. Ms Malcolm submitted that whatever reformulation this court might adopt, if minded to do so, should not be overly generous or restrictive. She also strongly urged us not “to go down the causation route”. Not merely where there are difficulties in using any adjective such as “substantial”, “operative” and “predominant”, with or without the indefinite article, notions of causation are inapposite in the context of a risk which may arise many years hence.

However, Ms Malcolm did agree that it might be possible to substitute “compulsion” for “impulse”, or to deploy nouns such as “wish” or “desire”; although she stressed that the gravamen of the fourth proposition was that whatever was actuating or overbearing the person’s will, if it were not his voluntary act, had to be urgent.

Ms Malcolm further submitted that, if the issue arose, the GoI’s assurances were both comprehensive and reliable. She contended that the “utmost efforts” have been made to accommodate the concerns of both the appellant and the Court. Although there was no protocol in place, the assurances indicated that a tailor-made care plan would be devised for the appellant shortly after his arrival at Barrack No 12, with the multi-disciplinary team having spoken to Professor Forrester and, if necessary, to HMP Wandsworth before finalising it. There was no need for such a plan to be established in advance of removal, and that might be unnecessarily delimiting. Private care would also be available if necessary or appropriate, although it should be recognised that the responsibility for the care of the appellant resides with the Indian authorities. Contrary to Justice (Ret’d) Nandrajog’s evidence, a court order is not required.

Ms Malcolm also made a number of what she called “sweep-up” submissions which we bear in mind but it is unnecessary to summarise.

This appeal is brought under section 103 of the 2003 Act. Pursuant to sub-section (4)(a), it may be brought on a question of law or fact.

Section 104(2)-(5) provide:

“(2)

The court may allow the appeal only if the conditions in subsection (3) or the conditions in subsection (4) are satisfied.

(3)

The conditions are that—

(a)

the judge ought to have decided a question before him at the extradition hearing differently;

(b)

if he had decided the question in the way he ought to have done, he would have been required to order the person’s discharge.

(4)

The conditions are that—

(a)

an issue is raised that was not raised at the extradition hearing or evidence is available that was not available at the extradition hearing;

(b)

the issue or evidence would have resulted in the judge deciding a question before him at the extradition hearing differently;

(c)

if he had decided the question in that way, he would have been required to order the person’s discharge.”

Both sub-sections (3) and (4) are relied on in this case, and they are in the alternative. An appeal under sub-section (3) based on alleged factual errors by the tribunal below faces the familiar difficulty that an appellate court is slow to overturn findings of primary fact: see, in an extradition context, Polish Judicial Authorities v Celinski and others [2015] EWHC 1274 (Admin). An appeal based on sub-section (4) is governed by the decision of this Court in The Szombathely City Court v Fenyvesi [2009] EWHC 231 (Admin). In short, the fresh evidence is admissible only if unavailable below and decisive.

In the instant case, the evidential landscape has moved on considerably since February 2021. In particular, the Court now has the evidence of Professor Fazel, the further evidence of Professor Forrester (suggesting an improvement in the appellant’s current position since August 2020), and further assurances from the GoI. All of this evidence was initially admitted de bene esse, and it is accepted that it was not available below. It follows that all evidential disputes must be resolved by this Court de novo: there is no sensible alternative, and the most important fresh evidence has been given orally and tested by cross-examination. If we were to accept the appellant’s case, it would follow that section 104(4)(b) would be satisfied and the evidence would, for the purposes of Fenyvesi, be decisive.

The fact that fresh evidence is relied on is potentially two-edged from the appellant’s perspective. Without such evidence, if the appellant failed to demonstrate any error in the District Judge’s decision, that would be the end of his case, and no issue under section 104(4) would arise. Conversely, now that the fresh evidence has been admitted, Mr Fitzgerald accepts that it is for us to decide whether the section 91 test has been met regardless of whether the condition in section 104(3) had been fulfilled. In other words, if the District Judge were right, section 104(4) applies and the appellant has another bite at the cherry; but – and herein lies the sting for the appellant - if the District Judge were wrong, this Court would not simply be allowing the appeal but should reconsider the matter for itself in the light of the fresh evidence. Thus, on this hypothesis the appellant might have won without the fresh evidence but could now lose with it. Mr Fitzgerald did not submit, rightly in our view, that the disjunctive language of section 104(2) means that he can now have it both ways. A reconsideration is inevitable, not least because section 104(2) says “may” and not “must” in the context of allowing appeals.

The Court raised in oral argument the possibility that the District Judge’s decision has been wholly superseded by events. In our judgment, Mr Fitzgerald was correct to submit that it remained important to analyse [225] of the District Judge’s decision because if it contains errors of law this Court, upon a reconsideration, should not be led astray. We have done so.

Section 91 of the 2003 Act provides:

“Physical or mental condition

(1)

This section applies if at any time in the extradition hearing it appears to the judge that the condition in subsection (2) is satisfied.

(2)

The condition is that the physical or mental condition of the person is such that it would be unjust or oppressive to extradite him.

(3)

The judge must—

(a)

order the person’s discharge, or

(b)

adjourn the extradition hearing until it appears to him that the condition in subsection (2) is no longer satisfied.”

Following the refusal of permission by Chamberlain J on other grounds, the sole issue for determination is whether it would be oppressive to extradite the appellant. Given the breadth of the statutory language, that issue requires a wide-ranging factual and evaluative assessment of all the salient features of this appellant’s case. Previous decisions of this Court decided on their own facts are indicative only, inasmuch as no two cases can possibly be the same.

Ms Malcolm relied on the use of the present tense in sub-section (2). Some physical or mental conditions will be stable for the foreseeable future, in which case no issue arises. Some may be serious now but are likely to improve, in which case sub-section (3)(b) may apply. Others may fluctuate or deteriorate. In such cases, although the court must make a finding as to the requested person’s condition now, it is inevitable that an evaluative assessment of the person’s “present” condition will have to recognise the prospect of future fluctuation or deterioration, since that is integral to the condition to which the person is now subject. It will therefore be necessary to consider the likely position (usually expressed in terms of a risk) at the time of extradition and foreseeably thereafter. That was the approach in Dewani (No 2) (see [109] below) and is also precisely what happened in Love v Government of the United States of America [2018] EWHC 172 (Admin); [2018] 1 WLR 2889, at [117-122].

Our attention has been drawn to three general statements of principle made by strongly constituted Divisional Courts.

In Dewani (No 1) v Government of the Republic of South Africa [2012] EWHC 842 (Admin), this Court (Sir John Thomas P and Ouseley J) stated, at [73];

“In our view, the words in s.91 and s.25 set out the relevant test and little help is gained by reference to the facts of other cases. We would add it is not likely to be helpful to refer a court to observations that the threshold is high or that the graver the charge the higher the bar, as this inevitably risks taking the eye of the parties and the court off the statutory test by drawing the court into the consideration of the facts of the other cases. The term "unjust or oppressive" requires regard to be had to all the relevant circumstances, including the fact that extradition is ordinarily likely to cause stress and hardship; neither of those is sufficient. It is not necessary to enumerate these circumstances, as they will inevitably vary from case to case as the decisions listed at paragraph 72 demonstrate. We would observe that the citation of decisions which do no more than restate the test under s.91 or apply the test to facts is strongly to be discouraged. There is a real danger that the courts are falling into a similar error as courts fell into in relation to s.23 of the Criminal Appeal Act 1968 and as described by the Lord Chief Justice in R v Erskine [2009] 2 Cr App R 29, [2009] 2 Cr App Rep 29, [2009] EWCA Crim 1425, [2010] Crim LR 48.”

Then, in Dewani (No 2) v Government of the Republic of South Africa, (Lord Thomas LCJ, Ouseley and Blake JJ) [2014] EWHC 153 (Admin); [2014] 1 WLR 3320 this Court held, at para 61:

“We therefore accept, … that the breadth of the factors to be considered under s.91 include looking at the question of whether it was unjust or oppressive to extradite the person at the time the request was being considered as well as looking forward to what might happen in the proceedings in South Africa if he was extradited. We must take into account all such matters, including the consequences to the requested person’s state of health and age. We accept that this entails a court taking into account the question as to whether ordering extradition would make the person’s condition worse and whether there are sufficient safeguards in place in the requesting state (as the Privy Council held was necessary in Knowles v Government of the USA [2007] 1 WLR 47 at paragraph 31).”

Finally, in Government of the United States of America v Assange [2021] EWHC 3313 (Admin); [2022] 4 WLR 11, this Court (Lord Burnett LCJ and Holroyde LJ) observed, at para 63:

“The law relating to "oppression" and suicide risk for the purposes of sections 25 and 91 of the 2003 Act is well-trodden. It may be collected from the judgments of Aikens LJ in Turner and Sir John Thomas P in Polish Judicial Authority v Wolkowicz [2013] 1 WLR 2402. It will rarely be necessary to look outside those two authorities for the applicable principles. Mr Lewis was concerned that the judge's approach applied a test which amounted to an obligation on a requesting state to guarantee that a requested person could not commit suicide in any circumstances. Mr Fitzgerald did not suggest that such an obligation arises. Section 91 and the decisions of this court do not impose such an unrealistic standard on requesting states. Mr Lewis submitted that the judge went too far in a predictive assessment of what might happen in the long term, depending on a number of contingencies, and failed to focus on Mr Assange’s mental condition at the time of extradition. He further submitted that the judge erred in failing to make the overall determination required by section 91.”

In [28] of his judgment in Turner v Government of the United States of America [2012] EWHC 2426 (Admin), Aikens LJ identified seven propositions established by previous case law:

“(1)

the court has to form an overall judgment on the facts of the particular case: United States v Tollman [2008] 3 All ER 150 at [50] per Moses LJ.

(2)

A high threshold has to be reached in order to satisfy the court that a requested person's physical or mental condition is such that it would be unjust or oppressive to extradite him: Howes v HM's Advocate [2010] SCL 341 and the cases there cited by Lord Reed in a judgment of the Inner House.

(3)

The court must assess the mental condition of the person threatened with extradition and determine if it is linked to a risk of a suicide attempt if the extradition order were to be made. There has to be a "substantial risk that [the appellant] will commit suicide". The question is whether, on the evidence the risk of the appellant succeeding in committing suicide, whatever steps are taken is sufficiently great to result in a finding of oppression: see Jansons v Latvia [2009] EWHC 1845 at [24] and [29].

(4)

The mental condition of the person must be such that it removes his capacity to resist the impulse to commit suicide, otherwise it will not be his mental condition but his own voluntary act which puts him at risk of dying and if that is the case there is no oppression in ordering extradition: Rot v District Court of Lubin, Poland [2010] EWHC 1820 at [13] per Mitting J.

(5)

On the evidence, is the risk that the person will succeed in committing suicide, whatever steps are taken, sufficiently great to result in a finding of oppression: ibid. 

(6)

Are there appropriate arrangements in place in the prison system of the country to which extradition is sought so that those authorities can cope properly with the person's mental condition and the risk of suicide: ibid at [26].

(7)

There is a public interest in giving effect to treaty obligations and this is an important factor to have in mind: Norris v Government of the USA (No 2) [2010] 2 AC 487.”

At [10] of his judgment in Wolkowicz, Sir John Thomas P sitting with Burnett J, as they then were, stated:

“10.

The key issue, as is apparent from propositions (3), (5) and (6), will in almost every case be the measures that are in place to prevent any attempt at suicide by a requested person with a mental illness being successful.”

The Court in Wolkowicz identified three stages in the extradition process: (1) when the requested person is being held in custody in the UK; (2) when he is being transferred to the requesting state; and (3) custody and detention thereafter. Ms Malcolm made submissions about stages (1) and (2), but Mr Fitzgerald’s focus was throughout on stage (3).

In the light of the submissions we received, it is necessary to address the interplay between Turner propositions (3), (5) and (6) as well as Mr Fitzgerald’s attack on proposition (4).

There will be cases where the requested person’s medical condition is not severe enough, and the risk of suicide not high enough, to engage section 91 at all. In our view, the clause “whatever steps are taken” in propositions (3) and (5) are really addressing opposite sides of the coin. If the risk is too low, the meaning and effect of proposition (3) is that it is unnecessary to consider the adequacy of the preventative arrangements referred to under proposition (6) because the requested person’s case has already failed. In this context, therefore, “whatever steps are taken” in proposition (3) may be understood to mean, “ignoring any steps taken”. The primary submission of Ms Malcolm was, on our understanding, that the present case falls into this category. On the other hand, proposition (5) recognises that there may also be cases where the risk of suicide is so great that, whatever steps may be taken, they will not reduce the risk to an acceptable level. Examples of cases falling into this category are Jansons and, subject to the qualification we make below, Fletcher v Government of India [2021] EWHC 610 (Admin).

Between these two poles there will be cases where the risk of suicide may be moderate or even high – too high to be discounted, but not so high that nothing can be done to address it that will render the risk acceptable. By risk of suicide we mean the risk that, in the absence of preventative measures, an attempt at suicide will be made and succeed. In such circumstances, the proposition (5) issue (level of risk) must be considered in conjunction with proposition (6) (steps taken to ameliorate the risk and reduce it to an acceptable level). In practice, most cases will fall into this category, which explains the emphasis in Wolkowicz on proposition (6). This reflects the practical realities including the fact that even in relatively weak cases it will be appropriate, out of an abundance of caution, to have regard to the system in the relevant prison.

In Fletcher, Chamberlain J endorsed the following encapsulation of the Turner propositions proffered by Fordham J in Farookh v Judge of the Saarbrucken Regional Court (Germany) [2020] EWHC 3143 (Admin), at [7]:

“The question is whether, on the evidence, whatever steps are taken – and even if the Court is satisfied that appropriate arrangements are in place in the prison system of the country to which extradition is sought so that those authorities will discharge their responsibilities to prevent the requested person committing suicide – the risk of the requested person succeeding in committing suicide, by reason of a mental condition removing the capacity to resist the impulse to commit suicide, is sufficiently great to result in a finding of oppression.”

The application of this encapsulation to the facts of both Farookh and Fletcher (correctly) led to successful appeals under section 91, although – as we explain below – Chamberlain J appears to have decided the case on a slightly different basis. However, a requested person’s case and/or appeal could fail at an earlier stage of the analysis because the medical condition and level of risk of suicide is not serious enough; or alternatively it could succeed when considering proposition (6) because the medical condition is not so serious that the risk is unacceptable whatever steps are taken, but the steps which would be implemented in the receiving state are inadequate to reduce the risk to an acceptable level.

The concept of oppression entails a value judgment. The focus cannot be solely on the nature and severity of the requested person’s medical condition or, on these facts, the risk of suicide. Turner proposition (7) should not be ignored because there is a cogent public interest in extraditing individuals who may have committed serious offences. Furthermore, as is clear from [38] of Aikens LJ’s judgment in Turner, the reduction of the risk to an acceptable level does not impose an absolute standard:

“I am quite satisfied that Florida has the proper facilities to cope both with Ms Turner’s mental illness and, so far as anyone can, the risk of her attempting to commit suicide if extradited. I think that this conclusion is entirely borne out by the evidence from … to which I have already referred.” (Emphasis added)

It is of some assistance briefly to consider the facts of three previous cases. In Jansons, there had been one suicide attempt and the evidence was that she would commit suicide if extradited. That was an extreme case. In Love, the Appellant had severe depression, Asperger’s syndrome and eczema, and there was a risk of serious deterioration in the event of extradition. The prison regime was inadequate to address the high risks. In Fletcher, the appellant had severe depression as well as a personality disorder, and the risk of completed suicide (in our view the correct test, rather than a risk of attempted suicide) was “very high”. Moreover, the finding in Fletcher was not so much that “whatever steps are taken” would fail to render the risk acceptable but rather that the specific steps the Indian authorities had said they would take were inadequate to do so: see [41, (d)-(f)].

Finally, we must turn to Turner proposition (4) which occupied a considerable part of the oral argument before us.

We have already pointed out that all of the Turner propositions have been endorsed by strong constitutions of this Court on more than one occasion. We do not consider that it matters for this purpose whether, as Mr Fitzgerald submitted, Aikens LJ misread Mitting J’s judgment in Rot (we do not believe that he did). We also recognise that in principle it is open to us not to follow this aspect of Turner if we were satisfied that it was plainly wrong. We are not so satisfied.

It would be preferable, in our judgment, to seek to interpret the fourth proposition of Turner taking into account the evidence we heard in this case and which, as far as we are aware, has not previously been adduced in similar cases.

The difficulty arises because the term “impulse” means one thing to a psychiatrist or clinical psychologist and another to a lawyer or lay person. In its technical meaning, an impulse is a sudden, forceful, urge to do something, which the person who is subject to the impulse may find difficult or impossible to resist. It is a feature of many personality disorders, and impulse control disorder features both in ICD11 and DSM-V. If proposition (4) in Turner were using “impulse” in this technical sense, the consequence would be that someone with very severe depression who was not impulsive would fail at this hurdle whereas someone with less severe depression but a comorbid personality disorder might surmount it.

The term “capacity” is also problematic because it has a technical meaning under statute. We see no attraction in Professor Fazel’s contention that concepts germane to the Mental Capacity Act 2015 might be read across to Turner. That would be to import far too high a bar, and we do not consider that this is what Aikens LJ meant.

Another difficulty, although this was hardly touched on in evidence, is that many psychiatrists would have difficulty with the notion of “voluntary acts”, still less those which may be neatly partitioned from acts which are impulsive and/or generated by an underlying disorder.

In our judgment, to the extent that Turner proposition (4) adds anything to propositions (3) and (5), its function is to indicate that in situations where the decision to commit suicide is voluntary, in the sense of being rational and thought-through, a finding of oppression should not be made. We heed Ms Malcolm’s warning that it would be unwise to gloss Turner proposition (4) with some additional or alternative form of words which imports a specific causation test: the verb “linked” already appears in proposition (3). In particular, we would deprecate any attempt to introduce concepts of causation as are routinely applied in tort or contract: the fact that (in conventional causation terms) a person’s depression would be either a cause or even the dominant cause of a person’s decision to commit suicide does not mean or necessarily suggest that the act was not voluntary within the meaning of Turner proposition (4).

It is always to be remembered that the Turner propositions form part of a judgment that attempted to set out general principles. It is not to be treated in the same way as if it were embodied in a statute. In our judgment, Turner proposition (4) should be read in a common-sense, broad-brush way giving full effect to the question whether the act of suicide would be the person’s voluntary act. This approach does not demand proof of “impulse” as that term is used by clinicians. “Compulsion”, “wish”, “desire” or “intentions”, as terms familiar to lay persons, are suitable synonyms; but none should be given particular precedence after being press-ganged into service. In Assange, the evidence was that the requested person had a “single-minded determination” to commit suicide. Consistently with this approach, “capacity” in proposition (4) is synonymous with “ability” or “capability”. It does not import the provisions or workings of the Mental Capacity Act 2005.

Although it is true that on one reading of the District Judge’s ruling, Turner proposition (4) was dispositive of the appellant’s case under section 91, the appeal before us does not hinge on that feature.

We have set out summaries of:

In the light of these summaries we can now address the issues that fall to be determined on this appeal relatively shortly.

There are two discrete but closely intertwined strands of the medical evidence that must be considered. The first is the prognosis for Mr Modi’s recurrent depressive disorder; the second is the risk of suicide in the event of extradition. The assurances are relevant to both strands and also to the wider question of the suitability of the accommodation in Barrack 12, with the attention of the parties and the Court being overwhelmingly on medical issues.

We therefore return first to the two strands of the medical evidence before addressing the overall question posed by section 91 and the specific questions arising from the application of Turner to the facts of this case.

As explained at [63] above, we accept that Mr Modi’s condition has been and is a fluctuating condition which has generally been characterised as moderate, though there have been times when it has deteriorated temporarily so as to be properly characterised as severe and other times when there have been features indicative of a characterisation as mild. In adopting this general characterisation we accept Professor Forrester’s assessments at the times he provided his “snapshots”, and we accept that Professor Fazel’s assessments on the two occasions that he interviewed Mr Modi are within the bounds of reasonable clinical judgment. The presence of features indicative of a characterisation as mild and Professor Fazel’s findings serve to temper any suggestion that Mr Modi is generally at the severe end of the spectrum of cases that can qualify as moderate. It is common ground that his recurrent depressive disorder is likely to deteriorate in the event of extradition. The extent of any such deterioration is unpredictable and may be relatively short-term, though the timing and duration of any deterioration are themselves uncertain.

The risk of suicide will increase in the event of extradition, reflecting the immediate change in circumstances and likely deterioration in his depressive disorder. We accept that the risk of suicide in the event of extradition may be characterised as “high” or “substantial”. Subject to two qualifications, we accept Professor Fazel’s opinion that the risk is likely to be highest at and shortly after the time of any extradition, as will be appreciated by those who are by then responsible for his care. The qualifications are that (a) it cannot be predicted with any confidence that the risk later will eventuate and (b) the existence of a risk later on cannot be excluded.

However, and leaving aside the Turner questions for a moment, the risk of deterioration of the underlying depression and the risk of suicide cannot be considered in a vacuum. On the basis of the assurances that the GoI has given, we accept that there will be suitable medical provision and an appropriate plan in place for the management and medical care of Mr Modi, which will be provided in the knowledge that he is a suicide risk (i.e. a person who, in the absence of preventative measures, may or will attempt suicide and will or may succeed). The evidence does not support a finding that the assured steps will eliminate the risk that Mr Modi will commit suicide altogether, still less the risk that he will attempt to do so. At the other end of the spectrum, we at one point understood Professor Forrester to give as his opinion that Mr Modi would commit suicide whatever steps were taken. If that were his opinion, we are unable to accept it, for a number of reasons. First, as we have indicated and as the papers to which we have referred at [41] above illustrate, the risk of a person committing suicide increases with the severity of their underlying condition; and Mr Modi neither is nor is very likely to be at the most severe end of the scale of depressive illness. Second, he has so far displayed no features of psychotic illness. Third, although he has exhibited persistent suicidal ideation, he has neither attempted suicide or deliberate self-harm nor disclosed plans to do so, except in the most vague and general way. Fourth, the steps taken to render Barrack 12 safe and to ensure that there is effectively constant monitoring operate to reduce both the risk of attempted suicide and the prospect of suicide being committed.

Turning to the section 91 question of oppression it is convenient first to consider the Turner propositions as we have discussed them at [112]-[130] above. In our judgment the risk of suicide in the present case is high enough to engage section 91. It is therefore necessary to consider the effect of the assurances about the arrangements that will be put in place in the event of extradition. In our judgment, and as recognised by Professor Forrester, the GoI assurances are extensive and, as we have set out above, specific to meet the successive levels of concern that have been advanced on Mr Modi’s behalf. The arrangements that we are assured will be put in place are appropriate to Mr Modi’s present and anticipated mental condition. They are in a number of respects more comprehensive than the regime that has been implemented at HMP Wandsworth. Specifically, the arrangements for weekly attendance (and more often if necessary) by the JJ Hospital psychiatrist and the assurance of attendance by a psychologist as and when required, together with the assurance about the availability of relevant and necessary treatment from a private doctor or mental health expert of his choice go significantly further than the regime at HMP Wandsworth.

While, as we have said, the arrangements cannot entirely eliminate the risk of suicide, that is not the test. The starting point is that a high threshold has to be reached in order to satisfy the court that Mr Modi’s condition is such that it would be oppressive to extradite him. As we have said, the arrangements that will be in place, which have been the subject of assurances in response to the concerns and promptings of those acting for Mr Modi, are appropriate. That is of itself an indication that they will enable the authorities to cope properly with Mr Modi’s condition and the risk of suicide. On the assumption that the arrangements are put into place in accordance with the GoI’s assurances, the residual risk is, in our judgment, greatly reduced.

Furthermore, when applying the principles that we have outlined in relation to Turner proposition (4), we are far from satisfied that any attempt to commit suicide would be other than a voluntary act as there described. In reaching this conclusion we bear in mind that Mr Modi is recorded on multiple occasions in the past as having contemplated the idea of suicide at some point in the future. This does not support the notion that, if he were to attempt suicide, it would be as a result of his having lost the capacity (in the sense that we have discussed) to resist the compulsion/wish/desire/intention; rather, it suggests that his act would be rational and thought-through, which is not to be treated as an involuntary act within the meaning of Turner proposition (4): see [122]-[129] above.

We also remind ourselves that, in the words of Turner proposition (7), “there is a public interest in giving effect to treaty obligations and this is an important factor to have in mind.”

Pulling these various strands together and weighing them in the balance so as to reach an overall evaluative judgment on the question raised by section 91, we are far from satisfied that Mr Modi’s mental condition and the risk of suicide are such that it would be either unjust or oppressive to extradite him.

In the light of our conclusion we can address the judgment of the District Judge shortly. He was conscious that the evidence before him was in some respects limited; and it may be said that his decision to proceed to judgment without requiring further assurances from the GoI at that point was bold to the point of being unwise. Furthermore, his apparent treatment of the absence of any “immediate” suicidal ideations as determinative was, for the reasons we have explained about the significance of future prognosis, incorrect. To that extent, we would accept that there were grounds for challenging his reasoning. His conclusion, however, was sound. It may be that the main benefit of the appeal has been to obtain the extensive further assurances that we have identified in the course of this judgment, which render the position clear to Mr Modi’s advantage and the District Judge’s decision supportable.

For these reasons, the appeal is dismissed.