The Government of South Africa v Shrien Dewani judgment (original) (raw)

City of Westminster Magistrates' Court, sitting at Belmarsh Magistrates' Court

The Government of South Africa v Shrien Dewani

10 August 2011

Findings of fact and reasons

This is an application by the Government of South Africa (represented by Mr Hugo Keith QC and Mr Ben Watson) under Part 2 of the Extradition Act 2003. I am asked to send this case to the Secretary of State for a decision whether the defendant is to be extradited. The defendant, Mr Shrien Dewani, is represented by Miss Clare Montgomery QC and Mr Julian Knowles QC. The application is opposed. South Africa is a Category 2 territory.

South Africa requests extradition so that Mr Dewani can be prosecuted for murder, kidnapping, robbery with aggravating circumstances, conspiracy to commit murder, and obstructing the administration of justice. The allegations relate to the murder his new bride, who was killed during their honeymoon in South Africa. The requesting authority is not required to demonstrate that there is a case to answer. It follows that it is no part of my function to consider the strength of any evidence there may be on the substantive allegations against him.

The Secretary of State has issued a certificate and sent the request and the certificate. My first task as the appropriate judge at the extradition hearing is to determine that the documents sent to me by the Secretary of State include the documents and information required by section 78(2), namely the documents referred to in section 70(9) (that is the request and the section 70 certificate; particulars of the person whose extradition is requested; particulars of the offences specified in the request; and as this is a case of a person accused of an offence, a warrant for his arrest issued in the Category 2 territory).

If the documents are deficient then the defendant must be discharged. As I am satisfied that the paperwork is in order, and this is not in dispute, I must go on to decide the three questions specified in section 78(4).

The first question is whether, on a balance of probabilities, the person appearing or brought before me is the person whose extradition is requested. This is not in dispute. The second question is whether the offences specified in the request are extradition offences. This is not in dispute.

The third question is whether copies of the documents sent to me by the Secretary of State have been served on the person, namely the defendant. This is not in dispute.

Having decided these questions affirmatively, I must go on to decide whether any of the bars to extradition in section 79 are applicable. If one or more of them is applicable then the defendant must be discharged. None of the bars is argued and none is found. However the defence argues that these proceedings are an abuse of the process of the court. It is also argued that extradition would be a breach of the defendant's human rights, particularly Articles 2 and 3 of the European Convention on Human Rights. It is further argued that extradition is barred by s91 Extradition Act 2003.

The hearing

The hearing took place at Belmarsh Magistrates' Court on three days in early May 2011, and resumed over four days in July. Mr Dewani attended on Tuesday 3 May, promptly, and an application was made on his behalf to excuse his attendance for the remainder of those three days. This application was based on a letter to Mr Dewani's solicitor from Dr Paul Cantrell, a consultant forensic psychiatrist with the West of England Forensic Mental Health Service. I agreed to excuse Mr Dewani's further attendance until a date to be agreed. There were a number of reasons for this decision. The application was not opposed by the government of South Africa. The defendant was represented by a solicitor and two counsel. At the time he was detained under section 2 of the 1983 Mental Health Act in Fromeside, a hospital ward near Bristol. Attending court involves a considerable disruption to his treatment regime and indeed to the staff at the hospital. His presence in these proceedings is not essential. The progress of the case did not depend on further instructions from him, and it was not intended that he give evidence.

The case was opened by Mr Keith QC on 3 May and then evidence was received by way of video link from South Africa, and by live evidence and written evidence. We concluded submissions on 21 July 2011 and adjourned for judgment until today.

The evidence about prison conditions

On 3 May I heard evidence by video link from Miss Amanda Dissel and Miss Sasha Gear. They separately affirmed and separately outlined their areas of expertise. Thereafter they gave evidence in chief in effect together, answering those questions in their specific area of expertise or sometimes, where their expertise overlapped, both providing answers to a question. In the circumstances I hope it will be acceptable that I refer to the evidence of Miss Dissel and Miss Gear collectively, as "their" evidence. To a large extent they confirmed the main points in the draft report dated 22 March 2011, consisting of 103 pages. They added some comments about the custodial facilities at the prisons named in the undertakings given by the South African government (Goodwood, Malmesbury A, and Brandvlei). Necessarily only a brief summary of the live and written evidence can be provided here. To some extent that evidence is paraphrased.

The witnesses have considerable experience working in the field of criminal justice reform in South Africa, with a specialization on penal reform (in the case of Miss Dissel) and on issues of sexual violence and sexual health in South African men's prisons since 2001 (in the case of Miss Gear).

There are 239 prisons in the country, with a designated capacity of 118,158. On 31 March 2010 there were 164,793 prisoners, of whom just under a third were on remand and the balance were sentenced offenders. The prison population has fallen slightly over recent years, but the system remains seriously overcrowded. The extent of overcrowding fluctuates at different times of the year. It also fluctuates depending on the correctional centre. Some unnamed prisons have been operating at well over 200% of capacity. However 49 centres have been operating at less than 100% capacity. There are particularly acute problems associated with 19 centres that recorded occupancy levels of 200% or over.

Overcrowding means that prisoners are held in cells intended to accommodate far fewer prisoners. There are insufficient toilets, inadequate sleeping facilities, lack of free flowing air, and lack of isolation facilities for infected inmates, which contributes to the spread of disease. Although prison rules require that prisoners be provided with eating utensils, it is not always possible to meet that basic requirement, because of overcrowding and possibly insufficient budget. The problems of overcrowding are well recognised by the authorities, with the Judicial Inspectorate noting particular concern for the conditions of awaiting-trial prisoners. Nevertheless significant vacancies remain in critical occupations within the prison service. In particular health care in many prisons is in crisis because of a lack of medical staff, prison overcrowding, poorly resourced prison hospitals and operational inefficiencies. As a consequence the compulsory health status examination, which must take place as soon as possible after admission to a Correctional Centre, is in fact not happening in all instances. As a result there is an increased risk of the spread of disease. These witnesses are also concerned about the recording of deaths in prison, and in particular as to how those deaths are labelled. It appears to them likely that some avoidable deaths are not being avoided and are not being completely recorded in the statistics. Although prisoners are allowed to obtain private medical treatment, at their own expense, this can be problematic. It is not always possible to secure treatment, and medication is not necessarily retained by the patient. There is very little understanding and research on the HIV/AIDS situation in South African prisons. It is very likely that, for a variety of reasons that were given, the rate of infection amongst prisoners is considerably under-recorded and far higher than in the broader population. Similarly there is considerable concern for offenders who are mentally ill. The witnesses say that the evidence suggests that the Department of Correctional Services is ill equipped to handle the health and care needs of the inmate population it houses. Many of those who need mental health attention remain in prison because of lack of resources or overburdened facilities.

There is also a serious, and well recognised, problem with gang culture in prisons. Such problems have existed for over 100 years, and exist across the country. Western Cape has a significant gang problem. Non-gang members bear the brunt of gang violence and intimidation. The violence includes sexual violence such as rape. Examples of specific violence were provided by the witnesses. Official figures were provided but the witnesses thought it very likely that many assaults are unreported, for example for fear of victimization. Over the years great concern has been shown by the authorities about the problem. The experts have both worked with government departments who are attempting to improve the position. However progress has been impeded by various factors including lack of continuity and different levels of commitment by key members of staff. This means that, in the view of the witnesses, staff are not given enough guidance and not enough is done to prevent gang violence and sexual violence in particular.

A number of specific heightened risks were highlighted for Mr Dewani. The risk to him of contracting HIV in prison springs from his vulnerability to sexual violence and coercion in prison. He fits the profile of someone who is particularly vulnerable to such abuse. He appears to lack "street wisdom". He does not involve himself in the direct perpetration of violence. He appears to be good-looking, youthful and physically well preserved. There has been a suggestion that he may be gay. If he is mentally disabled this is another vulnerability factor. Although wealth can be a protective factor, there is evidence that wealth needs to be linked to aggression, and an ability to fight back. The witnesses said that even if Mr Dewani were held as the sole occupant of a single cell, he would nevertheless be vulnerable to assault. He would not be in his cell all the time, but would exercise, take meals, shower, and have other contact with prisoners. Moreover it is not uncommon for prisoners to stray from their section to another section of the prison. Much of the serious violence occurs at night. Where there is only one person on duty at night after lock-up, as appears to be the case at Goodwood, a variety of situations could arise that would leave a prisoner exposed. It is usual for a member of staff to be locked in with the inmates, although the witnesses had not visited Goodwood, and could not comment upon the particular set-up there. The member of staff has no master key, therefore cannot unlock, and so needs to call for assistance from elsewhere. In addition, there are documented cases, and in particular anecdotal evidence, about the prevalence of attacks, including sexual attacks, on the way to court or in the cells at the court. In South Africa police convey prisoners to court. Although arguably the supervising judge has a mandate to oversee conditions in court cells, in practice the judge does not visit holding cells.

The witnesses considered the undertakings given by the South African authorities (attached at Appendix A). A single cell is necessary, but not sufficient, to provide protection. It is not enough to know that Mr Dewani will have a single cell. There should be additional information about protection while he is at exercise, collecting his cooked food, or being taken to court in a van. What are the mechanisms to protect him? The witnesses were unable to give an estimate of the level of risk to Mr Dewani. However he has special vulnerabilities as outlined above.

There was then a break in the evidence of these witnesses so that the lawyers representing the government could read and consider the several volumes of material that had been supplied to them shortly before the hearing began. On Wednesday 4 May I heard evidence from Judge Deon van Zyl, who gave evidence in person.

Judge van Zyl is Inspecting Judge in the Judicial Inspectorate of Correctional Services of South Africa. He gave evidence at the request of the Director of Public Prosecutions of the Western Cape. He was a judge of the High Court of South Africa from 1985 until 2008. He has held the position of Inspecting Judge since 1 May 2008, initially in an acting capacity but since 31 October 2008 by appointment by the President for a three year period. Unusually for proceedings in a magistrates' court, his evidence was recorded and is transcribed. I can therefore deal briefly with his evidence and with my findings of fact arising from that evidence. Again I summarize from his written evidence and from the answers he gave to questions in court.

Judge van Zyl had considered the various documents, including the report on "Treatment and Conditions in Prison and Police Custody in South Africa" by Dissel and Gear, the witnesses from whom I had heard the previous day. He had also considered the report on undertakings given by the Department of Correctional Services in regard to the correctional detention envisaged for Mr Dewani, should he be extradited.

The witness impressed on me the total independence of the Judicial Inspectorate. It is based on the British system. I have no doubt having seen him and heard him that he is, as he says, motivated by a strong sense of justice; a concern for what is fair and reasonable; and for fundamental values. It is also clear, from his reports and from his evidence, that he does not hesitate to criticize prison conditions when he thinks criticism is appropriate. Some of his comments to me could be considered to be outspoken criticism of conditions in South African prisons.

He gave his opinion of the three establishments where the South African authorities have undertaken that Mr Dewani would be held if he were extradited and then remanded in custody, or sentenced. His observations on those facilities, namely the Goodwood Correctional Centre for remand prisoners and the Malmesbury Medium A Correctional Centre for medium-term detainees and the envisaged new Brandvlei Maximum Correctional Centre, are covered in detail in the evidence. In summary they are all suitable and appropriate for accommodating Mr Dewani in compliance with his human rights. He said his own office can monitor Mr Dewani if he is placed in any correctional centre in the country. He undertook, on behalf of the Inspectorate, that Mr Dewani would have full access to independent visitors and other officials of the Judicial Inspectorate should he be detained. Judge van Zyl said that Mr Dewani would have no difficulty in making special arrangements for private care and access to psychiatrists and psychologists. He also referred to the publicly available health care facilities at the three centres mentioned. He expressed "perfect confidence" that suitable treatment would be available in these specific institutions. If extradited Mr Dewani would make an early appearance before the courts. Judge van Zyl was hopeful an early trial would be arranged, as it had been in other high-profile cases.

Concluding his evidence in chief the judge said he had no reason to doubt the undertakings provided by the Department of Correctional Services. If extradited Mr Dewani will receive whatever treatment he needs. He will be well cared for and it is unjustifiable to suggest that there is a real risk of a violation of his human rights.

He accepted in cross-examination that the undertakings given by the Department of Correctional Services, and by himself, are not legally binding on the authorities in the future. However he would expect the undertakings to be honoured in the absence of any unforeseen change. He accepted that a crisis could change the position, but said that was highly speculative. In this case there is a high profile detainee and there has been a specific undertaking by the National Commissioner that this person will be given special treatment. It would be most unusual for any successor to the present National Commissioner to deviate from that unless there are very good reasons. A change in position is not impossible, but in practice any successor will probably follow the undertaking of his predecessor unless there are very good reasons to deviate and those reasons can be justified.

He was cross-examined about his decision not to cooperate with Miss Dissel in the provision of facts for her report in these proceedings. He said he had originally been favourable to her request, and had asked her to put it in writing. However, when he learned that the report was in connection with this case, the Dewani case, which is highly controversial in South Africa, he decided it would not be appropriate apparently to side with one of the parties against another. Miss Montgomery QC asked him whether it would not be more consistent with an independent approach for him to supply to Miss Dissel such factual information as he was in a position to supply. I thought it was a pertinent question, and did not think it received a direct reply, possibly because the witness did not share my understanding of the meaning of the question. The explanation he gave was that initially he did not give full consideration to the request from Miss Dissel. On reflection he could not provide some of the information requested, particularly information about Independent Visitors. More importantly, when he learned that the information was in connection with a controversial case his instinct was to avoid involvement. He changed his mind after seeing the Dissel and Gear report prepared for this case, and at the request of the DPP. Most of the facts in the Dissel and Gear report are entirely accurate, and many of those facts are based on official reports including his own official reports. He has great respect for Dissel and Gear. Nevertheless their report runs the risk of presenting a misleading picture. He emphasised that his decision to attend court was to present an accurate picture and does not compromise the independence of the Inspectorate. Insofar as there was possibly a suggestion in the questioning that his judgement had been influenced by pressure, then I am satisfied that is not the case.

Judge van Zyl was cross-examined at some length about the Jali Report prepared in 2005 by a High Court judge. The Commission of Enquiry was directed at incidents of corruption and maladministration and the impact of those activities on the delivery of services within the prison system. The report referred to officials of the Department of Correctional Services being unwilling to respond to recommendations or advice. The witness confirmed that in a report of his own he observed that serious cases involving assaults, deaths, and suicides were not being adequately investigated or otherwise addressed. He was getting no response from the Department. The Jali Report also said that gangs present a complex challenge to the Department both in terms of corruption and violence. It said that South African prisons have become places where people are at risk of contracting HIV/AIDS due to prison rapes and sexual abuse. At the outset Judge van Zyl made it clear that he was not an expert in conditions in prisons for the period covered by that report, 2001 to 2005, but was familiar with the general contents of the report as a concerned citizen. He knows the author of the report. The thrust of the questions was that conditions in South African prisons were so dire in respect to corruption and violence in the period leading up to 2005 that this must cast doubt on the accuracy of his assessment of current conditions. Although the judge accepted that his colleague would have based his report on the evidence available to him at the time, he consistently made it clear that his experience since his appointment in 2008 does not reflect the picture painted in 2005. He persistently, and in my view properly, refused to speculate or give evidence outside his own area of expertise and understanding about current prison conditions. He repeated that current conditions in many prisons are in many ways unacceptable. He believes strongly that many more improvements can be made and should be made. He acknowledges that he does not necessarily have a full picture, and some incidents are not reported to him or to his inspectors. He does not suggest that corruption has been rooted out, but does not believe that this kind of corruption has had any effect on the implementation of his mandate. He had heard the allegation that in the past some people had been infected with HIV/AIDS deliberately, but hadn't experienced such an allegation in his term of office. He hadn't heard of any single case of deliberately transferring HIV/AIDS. He is not suggesting that it never happens, but it certainly hasn't come to his attention. He accepts that virtually all prisons have gang members to some extent. However the question of whether Mr Dewani would be exposed to that kind of activity is a totally different question. Although Goodwood, Malmesbury A and Brandvlei have had gang problems in the past, this is not something that has been brought to the attention of the Inspecting Judge as requiring his attention. He concluded by saying that the whole situation is changing and has changed since the time that Judge Jali wrote his report. He was also asked about a "guesstimate" of another colleague, Judge Fagan, that there is a prevalence rate of more than 50% of HIV/AIDS infection in South African prisons. The witness was cautious about that figure and points out that it is an estimate based on a guess. He was also asked about a newspaper article reporting comments from another judge, Judge Erasmus, that prisoners continue to be forced into dehumanizing conditions. In particular the article suggests that a prisoner entering a prison is going to be raped, mugged and forced into a life of crime. The witness said that he would never generalize to that effect, and cast doubt on the reliability of the article.

Judge van Zyl gave evidence all day. He had a plane to catch that evening. By agreement the remainder of the cross-examination was reserved until his return on 18 July. It was considered that having given his evidence thus far in person (and in view of the considerable paperwork that had to be considered) it would be best for him to return in person, if that could be arranged.

On Thursday 5 May Miss Dissel and Miss Gear were cross-examined by video link from South Africa. I need do no more than summarize briefly. Their campaigning has been to establish the serious problems with overcrowding, violence and sexual violence in prisons. There have been statistical difficulties because rape, for example, is not separately recorded from general assault. Changes in policy have been dramatic and encouraging but have not so far been fully implemented in practice. The position with Independent Visitors was explored as was the right to complain. The available statistics on violence and unwarranted sexual attention were also considered. The experts do not dispute the findings of the Judicial Inspectorate in 2009 and 2010. The relationship between prison overcrowding and other present problems was discussed. There are grounds for cautious optimism that new policies will bear fruit in the future. The statistical basis for their research in 2000 and 2001 was also explained. Segregation of vulnerable inmates from those who appear to be less vulnerable is necessary. So is the practice of separating gang members from non-gang members. However some institutions do this well, whereas others are very poor. The situation varies markedly. The position where there is only one person in a cell bears no relation to what happens in communal cells. There was some discussion as to the personal vulnerability of Mr Dewani.

In short, there was little disagreement between these two witnesses and Judge van Zyl on significant matters. They did not have any information to contradict the judge's assessment of the prisons named in the undertaking, but had not had the opportunity to undertake their own research. The researchers were not able to comment on the Valkenberg Hospital in Cape Town. Neither witness has any personal knowledge concerning the provision of psychological or psychiatric services in the relevant penal institutions, or the provision of psychiatric services in centres outside them. Had they been told earlier the prisons where Mr Dewani would be detained, if extradited, they would have been able to give a better picture of what the issues are in those specific prisons and how they function. This would have been gained by talking to the Independent Visitors.

During the rest of the day the court was shown a short DVD. This was a television advert advising against drink driving. It implied that a motorist jailed for drink driving would be sexually molested in prison. The court was then taken through the main evidence bundles. The case was adjourned until 18 July, for psychiatric evidence. It was not realistically practical to fix an earlier date in view of the further preparation required.

The hearing did resume on 18 July. Again Mr Dewani appeared promptly, and with some hesitation I excused his further attendance for the three days of evidence. The defence application was not opposed by the Republic. I gave my reasons at the time.

The July evidence has also been transcribed, and it is unnecessary for me to do more than briefly summarize that evidence.

Judge van Zyl was further cross-examined by Miss Montgomery QC. To a large extent this cross-examination involved putting to him comments by others on prison conditions. It was put to him that the Jali Commission referred to gangs in prisons as not only a pervasive problem, but one that had led to gangs taking control of the prisons. The judge said that was a total exaggeration. He did not believe the evidence was there to reach that conclusion. Although there is gang activity in many of the prisons, it does not create the problems described by the Commission and is not the kind of issue that has been brought to his attention as Inspecting Judge. He also dismissed claims by the Portfolio Committee in May 2010 that gangs had control of prisons. If that was a reference to the Jali Commission then that is a totally outdated document. "I'm not aware of a gang being in control of any single prison in the country." He suggested the Portfolio Committee makes generalizations that are totally unacceptable, based on hearsay rather than facts, or based on information that is now out of date. He doesn't agree that in 2008/9 sexual abuse and prison gangs in the Western Cape Prisons were rife. "Rife" is an exaggeration. It's not true. He did accept that when he was a judge the waiting list at Valkenberg was long and that mentally ill patients were frequently mixed with other inmates. However you cannot generalize and say this applies in all 239 centres. He also accepted that incidents relating to gangs are underreported. Reports alleging a high incidence of sexual violence in the Western Cape are unreliable. He accepts that individual incidents can and do happen and that violence in prison is not isolated and is linked to problems of infrastructure and overcrowding. He is very concerned that many of the natural deaths in prison could have been prevented, as could many of the unnatural deaths, and in particular the suicides. Among the problems is a lack of training and a lack of facilities. South Africa is a mix of first and third world, and problems will remain until the country is able to overcome financial difficulties. He was asked about a report from Families Against Abuse, referred to by the Portfolio Committee, that violation and abuse are guaranteed en route to prison. He was also asked about a practice whereby drugs are forced into the rectums of prisoners so that they can be transported into prison. The judge accepted that this practice occurs. However much of the information is unreliable based on exaggerated hearsay. He cannot agree that many rapes occur in prison vans. He was asked about a survey of some 750 prisoners published in 2008. He thought the sample was tiny. There was no way of linking the statistics to the proportion of vulnerable inmates. He agreed that some old prisons should be demolished and replaced. He gave evidence about the steps being taken to determine the cause of prison deaths. At the moment it is impossible to say whether a death was caused by a prison-acquired infection. They are not receiving reports of prison rape. Nevertheless many prisoners arrive with HIV/AIDS and the virus can then easily be transmitted to someone else. It was put to him that Gideon Morris had said that a prison sentence is tantamount to a death sentence by HIV and that 70% to 80% of suspects are raped by fellow prisoners before they are charged. He agrees that Gideon Morris started the Judicial Inspectorate but says there is no factual basis for the statements attributed to him there. As for 45,000 prisoners dying of AIDS in South African prisons that is absolute nonsense. Similarly the report that gang members carrying HIV were ordered to rape disobedient inmates in a ritual known as slow puncture, by which the victim would die over a period of time, is absolute nonsense. He rejects it unconditionally. He does not believe Mr Morris made that kind of allegation. He does not know why the Head of Centre at Goodwood regarded gang activity as a problem. There is a shortage of psychiatrists who do forensic work in South Africa. Prisoners are not always treated in hospital because there are insufficient hospital places for them. There has always been a problem with delay at Valkenberg. The judge does not know who is currently responsible for investigating Valkenberg. He has not investigated, although he is aware that there have been renovations and reparations at the hospital. However he cannot guess what the situation is there now.

Counsel also returned to the reasons why Judge van Zyl had not given the defence researchers access to his inspectors' reports, and what he had meant about embarrassment. He had said in examination in chief that this case is "a very high profile matter, and certain statements have been made by people in high places, which makes this a very embarrassing kind of situation for me personally." He did not remember making the comment at the earlier hearing, and thought the embarrassment was because he had not realized earlier that the case he was being asked about was the high profile case of Dewani. When taken to the passage in the transcript, he said he thought the embarrassment is more to do with the fact that he didn't originally enquire sufficiently about the case being referred to. Also, because of the sensational press and publicity, he did not want his office to become involved. When he thought about it he realized it was totally unacceptable for his Independent Visitors to be contacted. Counsel suggested to him that this demonstrated an inappropriate partiality. This was denied by the witness who repeated that his embarrassment was probably as a result of his own failure to apply his mind to what he was asked.

Judge van Zyl was asked about a quote from the Police and Prisons Civil Right Union General Secretary that the proposed treatment of Mr Dewani is unfair. He must be treated the same as other prisoners. The judge says this trade union is nothing to do with the Inspectorate. It represents employees in Correctional Services. It doesn't deal with prisoners. "We look after the prisoners."

In re-examination he was asked again about the figure from Friends Against Abuse that there was a 99.9% guarantee of abuse on the way to court. That is a total exaggeration and based on misinformation. No evidence to support the allegation has ever been brought to his attention. This morning [that is, 19 July] he received notice of the 466th death in prison this year. That figure includes natural deaths. He was asked about the suggestion that Mr Dewani was very likely to be sexually assaulted in prison, and totally rejected it. "That this kind of thing does take place cannot be denied, but at that level there is no basis for accepting or even suggesting [that it would happen to Mr Dewani]." The reason that the Dissel and Gear report was misleading was because it made no reference to conditions at Goodwood, Malmesbury A and Brandvlei. The last two Judicial Inspectorate reports deal with complaints, but the complaints raised have not included gang activities. He was taken to the report of the Department of Correctional Services that says that at Goodwood there were in the past year only 35 complaints of alleged assaults, and not one of those took place in the area where Mr Dewani would be detained. Similarly at Malmesbury the recent record of assault and violence is "very low". There is no reason to doubt the good faith of the undertaking given by the DCS.

Moreover the judge gave an undertaking on behalf of the Inspectorate that: "We personally will ensure that this undertaking is complied with." He has not the slightest doubt as to the practical efficacy of the undertaking. Since the last time he testified at Belmarsh he has required his officials to visit all three of the centres and all the reports have been extremely positive. He is confident that there will be no gang violence involving sexual abuse of Mr Dewani, who will be held in a single cell, if he is held at all. Most sexual violence occurs in communal cells. He is "perfectly sure" Mr Dewani will never be exposed to the risk of gang activities. He has no reason to doubt anything that Dr Kaliski, from Valkenberg Hospital, says. Dr Kaliski is a man of incredible experience with a very strong staff under him. This includes a number of psychiatrists who have varying degrees of expertise. Many of them are Fellows of the Royal College of Surgeons. The impression he had of Valkenberg was that even though it may not be the best in the world, it is certainly far better than it had been previously.

The medical evidence

Dr Paul Cantrell is a Consultant Forensic Psychiatrist with the West of England Forensic Mental Health Services. He is treating Mr Dewani at Fromeside. He provided reports and gave evidence as treating psychiatrist. Professor Nigel Eastman is Professor of Law and Ethics in Psychiatry and an Honorary Consultant Forensic Psychiatrist at St George's, University of London. He prepared reports at the request of Mr Dewani's solicitors, Hickman and Rose, and also gave evidence. Professor Michael Kopelman is Professor of Neuropsychiatry, a Consultant Neuropsychiatrist and a Chartered Psychologist working in the South London and Maudsley NHS Trust and based at St Thomas' Hospital. He prepared a report at the request of the Extradition Unit at the Crown Prosecution Service and also gave evidence. Professor Sean Kaliski is a specialist psychiatrist based at the Valkenberg Hospital in the Western Cape. He sent a letter, authenticated, dated 13 July 2011, which I accepted as admissible as to conditions at that hospital. Dr Paul Dedman treated Mr Dewani while at the Priory Hospital in Bristol. He is a Consultant Psychiatrist and provided written information at the request of the defence solicitors. Dr Larissa Panieri-Peter is a specialist psychiatrist in South Africa. She also provided evidence in written form at the request of Mr Dewani's solicitors. Dr Martin Eales is a Consultant Adult Psychiatrist at the Cygnet Hospital who wrote to Mr Dewani's solicitor in April.

The evidence of the expert psychiatrists was to a large extent agreed. Very helpfully, Professors Eastman and Kopelman had met before the hearing and agreed a number of key facts, which are attached (as Appendix B). An even briefer summary is that it is agreed that Shrien Dewani is suffering from a depressive illness and from post-traumatic stress disorder (PTSD), and that each disorder is severe. The current risk of self-harm or suicide is real and significant. The doctors are agreed that Shrien Dewani is currently unfit to plead. The prognosis is very hard to predict, but Professor Eastman estimates it as "poor to bad" while Professor Kopelman is somewhat more optimistic. If the defendant is extradited it is highly likely that his mental disorders will worsen and there is a significant risk of further relapse into psychosis. His fitness to plead would be further reduced and the suicide risk would further increase.

In evidence Dr Cantrell fully agreed that the illnesses were severe, in the categories "mild", "moderate" and "severe". He also agreed that there is a real and significant risk of suicide. This is currently managed by Mr Dewani being in an appropriate therapeutic environment. He also agreed that his patient is unfit to plead. He referred to the defendant as hyper-sensitized and suffering from hyper-arousal. He asked, as he had previously, for Mr Dewani's bail conditions to be varied to allow him to return to the family home unescorted by a member of staff. By this he apparently meant that hospital staff need not stay within arms' length of him. He would like to move to taking him, leaving him and then fetching him back, but at a later stage. There is a risk but "all psychiatrists in dealing with mental disorders juggle risks".

All the medical experts were asked about compliance with treatment and whether he is faking his symptoms. They were particularly asked about the fact that he had been exercising, which may well have had an adverse affect on his CK levels. CK is a specific muscle enzyme, creatine kinase, and Mr Dewani suffers from an unusual variant of it. The CK levels are highly significant. They are far too high in his case, which means that pharmacological treatment is at the moment too risky to undertake. Both pharmacological treatment and skilled psychotherapy are important in the treatment of the illnesses from which Mr Dewani suffers. The accepted medical belief is that to start pharmacological treatment on a patient with this CK level produces a risk of death of maybe 20%. When asked about the expected timescale on therapeutic conditions, Dr Cantrell said:

"I don't know. There are many psychological factors which are maintaining his current disturbed mental state, not least court proceedings. The usual response times for depressive illness alone is of the order of 4 to 8 weeks, perhaps 12 weeks. The difficulty with Shrien Dewani is that there are two disorders multiplying the effect of each other and hampering treatment of each other. The improvement to fitness may take a long time. It may take months, if not many months. I cannot be more specific."

Later he said that he does not think his patient is fit to be extradited. "He may become so, but it is my opinion that he is not at present." The risk of suicide would be raised if extradited. He broadly believed Mr Dewani's statement that he would kill himself if sent to South Africa.

In cross-examination he was asked about his comment that the defendant's health will improve. There are two reasons for that opinion. First, it is highly unusual for cases of depression and/or PTSD not to improve. It is his opinion and experience that Mr Dewani will make a full recovery, but that it will be a long, slow process. The second reason is that he hopes that the current problem with CK levels can be remedied so that his patient can start to take the relevant medication. He did not rule out the possibility, at a later stage, of prescribing those medications even if the CK levels do not drop. He confirmed that antidepressants are likely to be available in South Africa in the same way as they are here. Although he cannot predict a time for recovery, if antidepressants were used he would hope for some improvement, although it will take considerably longer than the usual 4 to 12 weeks.

It is Dr Cantrell's view that there is a significant risk of suicide, but that the risk is not immediate. He did not consider the February overdose to mean that Mr Dewani wanted to kill himself, but that he did this to avoid being part of the proceedings any more. There is a difference.

Mr Dewani has continued to exercise by skipping, doing sit-ups and push-ups and also using his multi-gym at home.

The defendant suffers from profound psycho-motor retardation and he gave details about that. The doctor initially had underestimated the extent of that retardation. The exercise he takes is not inconsistent with it. He was asked about the account his patient had given of events in South Africa. Mr Dewani is extremely bright but is severely disabled.

"He can hold a conversation. You can with difficulty get lots out of him. The difficulty is that he drifts away into his inner world. That is characteristic but deeply problematic for treatment and for court proceedings."

He described the risk of suicide as "high", even in the absence of extradition. Those with severe depression commonly harm themselves. The risk is mitigated here because Mr Dewani is checked every 15 to 30 minutes by staff. Also there are protective factors such as his religious beliefs and a desire to clear his name. It is not necessary for a patient to make repeated suicide attempts or to say he would commit suicide before finding that there is a high risk of suicide. His illnesses themselves carry the risk. Indeed Mr Dewani has said that if he did plan to commit suicide, then he would not tell the doctors.

Mr Dewani had no history of mental health before December. There is no evidence of psychosis at the moment.

The phases "high", "very high", "probably very high", "very serious", "serious", and "significant", are all much the same. You cannot quantify risks in numbers. "It is not 72%, it's not 0.5%, it is "high". There is no other way of doing it. He was pressed on this but maintained that none of the terms used in other cases helped the court. He can only say the risk is "high".

The defendant would be more at risk in South Africa for a number of reasons. First, he would be in prison. Secondly, he needs an understanding therapeutic team. Thirdly, he and the medical team need an appropriate therapeutic environment. Fourthly, the journey to South Africa would be highly psychologically distressing.

Professor Eastman's evidence was that in this case the prognosis is very difficult to make because the pharmacological treatment is imponderable and because of the chronic nature of the maintaining factors, both in relation to depression and to PTSD. PTSD can become chronic. The professor agreed with Dr Cantrell about the risks of extradition. Mr Dewani has hyper-perception of threat. He would undoubtedly feel at great risk in a South African prison. Fear could lead to him becoming pathologically frightened which would feed into his inability to cope with legal matters. The proposal to place Mr Dewani in a single cell runs counter to UK practice with those who are a suicide risk. He was cautious about facilities in South Africa because he hasn't been to them. He was asked about Dr Kaliski's letter. Dr Kaliski describes the conditions at Valkenberg Hospital. This witness did not find reassuring the reference to the wards being "habitable". He had seen no evidence that the defendant would be adequately treated in South Africa.

Although Mr Dewani can give a history of events in South Africa, this does not mean he is fit for trial. He has very substantially impaired ability to listen to evidence and then instruct counsel accordingly.

There is a hierarchy of descriptions of the risk of suicide. He agrees with Dr Cantrell that adjectives such as "high" and "very high" try to describe something quantitively without using adjectives. The professor believes the risk to the defendant will increase with the length of the illness, even if extradition does not take place. However extradition would substantially increase the risk. This risk would be "unacceptable" in the absence of proper and adequate management of the illnesses and of the risk.

The prognosis for recovery depends on whether Mr Dewani can receive medication. If he cannot then the prognosis is poor. "However, one has to wait until the next chapter in order to come to a firmer view". If CK levels fall and he receives anti-depressant treatment the professor would be more optimistic. He cannot say whether the defendant would improve with medication alone. "All of this is very hypothetical because we do not know what the future holds in terms of what can be done and what the effect of what can be done will be".

There is an unacceptable risk of suicide now, which would increase if he is returned to South Africa. A therapeutic environment is important and there is a huge difference between hospital and prison. As someone starts to improve from depression, the risk of suicide can increase. Travel to South Africa would greatly enhance Mr Dewani's distress and cause him substantial further psychological harm. It may result in him killing himself. It is very unusual to have both depression and PTSD present to such a severe degree as exists here.

Even if Mr Dewani is acquitted, recovery is likely to take a long time. However the professor is concerned that, as things stand, Mr Dewani will not be sufficiently well to be able to complete the process.

Professor Kopelman referred to the accounts that Mr Dewani had given him. He agreed that over a number of hours he was able to obtain a fairly detailed account of his past, his personal history, his family life and the circumstances of the events in South Africa last year. He also gave medical details that the witness supplemented with information from the medical records. The written summary makes the account more fluent than in fact it had been. There had been long hesitations and repeated questions. Progress was very slow. He also discussed the risk of suicide, saying that the predominant factors in relation to a risk of suicide were severe clinical depression, compounded by PTSD and a predominant feeling of hopelessness. Mr Dewani has had protracted suicidal ideas for an extensive period of time and there has been a previous attempt. About 25% of those who attempt suicide try again within a year, and 5% succeed, although those figures may not be very relevant to this case. The professor gave reasons as to why the incident in February might be considered an intention to commit suicide. He discounted, as did the other medical witnesses, the possibility that the defendant might be actively trying to delay recovery.

He said the defendant is still at "significant" risk of suicide and self-harm. The terms "suicide" and "self-harm" are used interchangeably. However he declined to be drawn further. He said that "once suicide is classified as a high risk, it is impossible to grade further". He said the risk of suicide would increase if extradition took place. The fact of return to South Africa could bring back memories of whatever happened there. Returning to a previous environment can lead to re-traumatization and marked deterioration. Family support would be less available. Although he could not comment on the psychiatric care that would be offered in South Africa, it would inevitably involve a substantial change. One reason for slow progress to date may be that he has had disruption in clinical care because there have been three treating teams. It would be harder for him to recover to a point that he could be fit to be tried. Fears about what might happen in South Africa are very real perceptions for Shrien Dewani. Even acquittal is not likely to terminate all his symptoms, although it is likely to be helpful. There is no physical reason why he cannot go on a plane but that would be immensely distressing for him. In short there is a high risk, and it will be made worse if extradition takes place.

"I would hope for a significant improvement over the course of six months. Actually Dr Cantrell said something similar, if anti-depressants are resumed� so the optimistic view would be that in six months he would be sufficiently improved for the suicide risk to have diminished, and for his clinical depression to have diminished� and therefore much more fit to plead, despite probably still having some PTSD symptoms. But we cannot say for sure. That is where psychiatry and the law clash, because you need a definite timescale and we psychiatrists cannot give one."

The most optimistic view is that Shrien Dewani would improve in six months.

"The problem with extraditing now is that, even if they have the best possible psychiatric facilities in South Africa, once Shrien Dewani is extradited in his present clinical state, it will be very difficult to get him to a state where he is fit to plead. The goal should be to get him clinically better in terms of depression/PTSD, so that here, in the UK, we now think he is fit to plead. It is unusual for depression/PTSD to be grounds for unfitness, so we would expect him to come back to a state of fitness to plead at some stage."

He agreed with Professor Eastman that extradition would produce an "unacceptably high risk" of suicide. It would make his condition even worse and would be harder to get him into a position where he would be fit to plead.

On the basis of what he was told about prison conditions in South Africa, he thought they would provide totally inadequate care for this particular person. This was on the basis that there would be twice monthly visits by psychiatrists (which may in fact not be accurate). He was also asked about a report entitled _The Insanity of the Criminal Justice Syste_m. He said that if the article was accurate it sounds ghastly but that Judge van Zyl is the person to say whether or not the article is accurate. He was not able to comment about Valkenberg Hospital from personal experience. However he expressed concern about the statement of Professor Kaliski that the ward is "habitable".

I had the great advantage of hearing evidence from a number of witnesses of real integrity who are leading experts in their field. I will make a separate comment about Judge van Zyl later. In addition this case was prepared and presented to the highest standard. Hard work in preparation had produced for me nearly 20 lever arch files of papers to read. If my decision today is appealed, it may be that the court above would benefit from rather less material to read.

Abuse of Process

An allegation of abuse of process is made by the defendant. The conduct alleged to constitute the abuse is identified as follows. First, says the defence, a senior South African police officer (General Cele) has described this defendant as a monkey who is guilty. Secondly, says the defence, the South African authorities have threatened to reveal the motive for the killing if Shrien Dewani contests extradition. The material relied on in support of the motive would inevitably expose the defendant to humiliation beyond that warranted by a mere exposition of the case against him. Thirdly, a South African Prosecutor, Advocate Simelane, has prejudged the issue of guilt in this case, as evidenced by an interview he gave.

I must consider whether this conduct, if established, is capable of amounting to an abuse of process. If it is, I must next consider whether there are reasonable grounds for believing that such conduct may have occurred. If there are, then I should not accede to the request for extradition unless I am satisfied that such conduct has not occurred.

I will deal separately with the complaints made by the defence. The first complaint can be dealt with shortly. If a senior police officer, with full understanding of the evidence in the case, publicly describes Mr Dewani as a monkey in circumstances that assume the proposed defendant is guilty, that is not capable of amounting to an abuse of the process of the court. Even if it comes to the attention of the trial court, it would not make the court more likely to convict. It is no surprise to a court to learn that a police officer believes a defendant to be guilty. Such a belief carries no weight for the prosecution. In some circumstances one can imagine it carrying weight for the defence, particularly if the officer is a key witness in the case or collected the evidence. Similarly, offensive comments by a police officer are more likely to harm than to help a prosecution. In South Africa any trial of these allegations will be heard by a judge, almost certainly sitting with two lay assessors. The assessors themselves may well be experienced in the law. It is common ground that South Africa provides a fair trial process. There is nothing in this complaint.

The second complaint is more complicated. There are circumstances when the authorities in the requesting state place undue pressure on the defendant so as to make it an abuse of process that he be extradited. The facts in Cobb provide such an example. Cobb is a Canadian case where, among the other things, a judge had stated that non-cooperation with extradition to the USA would result in "the absolute maximum jail sentence that the law permits me to give." It is therefore necessary to examine in more detail what happened.

On 18 January 2011 a South African newspaper, The Telegraph, reported that the Police Commissioner, Bheki Cele, had said that detectives knew why Mrs Dewani had been killed and planned to disclose the details at an extradition hearing in London on Thursday. The disclosure by Mr Cele, said The Telegraph, appeared to be aimed at persuading Mr Dewani to return to South Africa voluntarily to face his accusers. Mr Cele refused to give details of the alleged motive but said: "It might come out on Thursday as we'll be having this hearing and if it needs to come out there it will be fine". The paper continues that a police representative had explained these comments by saying: "If Mr Dewani comes forward and agrees to come to South Africa, this will no longer be necessary". There was then a reference to further "good leads" being collected by police in London.

The defence says this is a reference to a statement taken the previous day. They are probably right about this. The statement in question is from a man referred to in these proceedings as LL. I have seen his full statement. The requesting authority wanted to adduce a small part of that statement in their case. However I ruled it inadmissible as irrelevant and likely to engender further satellite litigation that would not help me in these proceedings. Nevertheless, it has been necessary for me to reconsider the statement in the light of defence submissions. It would be artificial and arguably unfair to the defence not to do so. Nevertheless, I see no need to repeat the contents of the statement here. The facts do not establish the "chilling" motive ascribed to them by Miss Montgomery.

The events described in the statement of LL undoubtedly have the potential to embarrass the defendant. They could be described as humiliating, particularly if true, if secret, and if they breach taboos in the defendant's culture. However, they are less significant, less humiliating, and less damaging than the main allegation, which is that he arranged for the murder of his new bride. It is hard to imagine how a man who does not wish to face the allegations of murder through the trial process in South Africa (which is the assumption being made, whether true or not) would change his mind in the face of a statement that the police now know the motive. That would be the case even if Mr Dewani realized from the statements what the alleged motive might be. There is no evidence that the comments were directed at this defendant, or that he knew of them, or that he was likely to act on them (still less that he did act on them). Had he thought about the matter at all, he would have concluded that returning to South Africa would at best have delayed the alleged motive from coming into the public domain. It would not have kept it secret from his family or those members of the public who would follow the case. The statement of Mr Cele is no more than a statement of fact. They believe they know the motive. It was not necessary to reveal it in uncontested extradition proceedings. In any event the decision whether to introduce the motive was not a matter for Mr Cele. He has no influence on these proceedings. The speculation of The Telegraph was just that. The defence has tried, unsuccessfully although with commendable endeavour, to trace the source of the speculation. Even if I am wrong, and the motive was to put some pressure on Mr Dewani to return voluntarily, that would fall short of an abuse of the process of the court. Abuse of process is a rare and exceptional remedy. Here the statement was not made by a judicial officer (as in Cobb), or by a prosecution lawyer, or by anybody in a position to influence the proceedings. Mr de Kock confirms that all decisions relating to the prosecution are his, based solely on the evidence at his disposal. What was published does not show bad faith on the behalf of the prosecutors. It was not acted on. Disclosing the alleged motive in these proceedings would not be unlawful. The newspaper article has not caused this defendant to take any step to his detriment. While the facts in Cobb suggest that this is not an essential feature of this type of abuse of process, at the very least it must be an important factor. Cobb was considered by the House of Lords in McKinnon v Government of the United States of America [2008] UK HL 59 where it was said that it would only be in a wholly extreme case that the court should properly regard any encouragement to the accused person to surrender for trial and plead guilty ... as to constitute an abuse of process justifying the requested state's refusal to extradite the accused. "It is difficult, indeed, to think of anything other than the threat of unlawful action which could fairly be said so to imperil the integrity of the extradition process as to require the accused, notwithstanding his having resisted the undue pressure, to be discharged ..." [Lord Brown para 41]. The facts here do not amount to an abuse of process.

I can deal very briefly with the complaint against Advocate Simelane. The complaint is made in a statement by Advocate Slabbert. He refers to having been given copies of comments made by Advocate Simelane in an interview. This was investigated further by Rodney de Kock, the DPP, who has provided further details in his statement. The interview appears to be a live television broadcast and is set out in Annex 22, Appendices to Volume 2. Advocate Simelane was mistaken in saying that Mr Dewani is a fugitive from justice, and has acknowledged that. Taken in isolation, some parts of the interview can be considered to assume guilt. However, taken as a whole the comments make it clear that the court will make a decision based on the evidence presented before it. When read, and presumably when viewed, the interview does not offend against a sense of natural justice, still less come close to establishing an abuse of process of the court.

Finally I should add that Mr de Kock has provided details of the abuse of process doctrine as applied in South Africa. This appears to be similar to our own.

Nothing presented to me in these proceedings could lead to the conclusion that extradition is an abuse of the process of our courts. It would be a matter for the trial court whether there has been an abuse of process of the South African system.

Findings of Fact

The remaining defence submissions are based on the medical evidence and on the conditions Mr Dewani will face if extradited. Most of the facts are not in dispute but they do need to be recorded and to some extent discussed.

a. Mr Dewani's health

I substantially adopt Miss Montgomery's summary of the medical evidence in paragraph 7 of her closing submissions.

(i) Shrien Dewani suffers from two severe and incapacitating mental illnesses.

(ii) There is currently a real and significant risk of suicide.

(iii) The suicide risk will increase if he is extradited to South Africa.

(iv) The prognosis is uncertain. The treatment he can receive is not yet known because of his very high CK levels.

(v) He will suffer real harm and distress if he is returned.

In so far as I have not adopted the remaining points of the summary, that is not because Miss Montgomery incorrectly summarizes the agreement between the experts (she does not), but for the following reasons.

First, the question of whether any risk is unacceptable is ultimately a question for the court to consider. The court's conclusion may legitimately be different to that of the medical experts, particularly those whose primary concern is for the care of Mr Dewani.

Secondly, the opinions of the professors and the doctor on the therapeutic facilities in South Africa are not conclusive. They are not familiar with them and are not best placed to comment.

Thirdly, the fact that the experts agree that the defendant is unfit to stand trial is not determinative of the issue.

b. Is Mr Dewani fit to plead?

The parties agree that it is not my function to determine whether this defendant is currently unfit to plead and stand trial. There is no dispute between the doctors on this. The witnesses from whom I have heard are leading experts in the field. They correctly state the legal test for fitness to plead. Professor Eastman, I notice, is a Professor of Law and Ethics in Psychiatry, a barrister, and has been involved as an adviser to the recent Law Commission report on the law of murder, which included consideration of capacity. It is unlikely that any court would disagree with the combined weight of opinion that has put before me.

Nevertheless it remains a question for the court, and not the experts. The experts and the parties explicitly recognize this. If the question of fitness to plead comes to be considered by a court then in my view any judge would almost certainly explore the matter further.

In his report dated 19 June 2011 Professor Eastman says this (page 21):

"In my opinion, Mr Dewani's mental disabilities call into serious question whether he would be "fit to plead" in terms of the Prichard or Davies tests. Albeit these legal tests of fitness are set at a high level, his mental state is so acutely disturbed much of the time, and disturbed in such a way that he is unable to concentrate effectively, that it is open to serious doubt whether he satisfies all of the required criteria. Specifically, although he is fit to enter a plea, and in very broad terms is fit to instruct his lawyers in very broad terms "outside of the court room", in my opinion he is likely to be profoundly disabled in terms of following the detail of evidence in any trial, such that he would not adequately be able to instruct his counsel as evidence emerged during the trial, which evidence he would presumably be required to address. Such disabilities would, in my opinion, arise from disabilities of concentration arising from both slowing and restriction of his thinking, originating from his depressive illness, and hyper-arousal and hyper-vigilance, with resultant distraction from matters at hand, arising from his PTSD."

In his report dated 11 July 2011 Professor Kopelman states, at paragraph 7:

"Mr Dewani understands the nature of the charges against him. He would be able to understand that (in British law) he would be allowed to challenge a juror. However, in his present state of mind (with very impaired concentration, distractibility, forgetfulness, mental slowness, and problems in comprehension), I am not persuaded that Mr Dewani would be able to follow or to remember the details of legal proceedings accurately, nor that he would be able to instruct his lawyers appropriately. Hence, in my view, he is not currently fit to plead, but this is likely to change when Mr Dewani's clinical state has substantially improved."

Dr Cantrell said, in his report of 14 July, paragraph 6:

"Although fitness to plead as understood in the Prichard criteria may not come directly within the legal framework with which the court is dealing in this set of circumstances, it is my opinion that Mr Dewani is currently not fit to plead according to those criteria. This is principally on grounds of his lack of current capacity to instruct his defence, and to continue to do so by necessarily following Court proceedings in sufficient detail. This is the result of both his disorders, acting synergistically, and each complicating the treatment of the other."

Mr Dewani is highly intelligent, an advantage not shared by many defendants before the criminal courts. It appears that he has already set down in detail to the South African authorities his account of his brief stay in that country. He was able to repeat part of that account, albeit in very difficult circumstances and with considerable expert care, to Professor Kopelman. He has also shown awareness of some of the developments in South Africa, such as being called "a murdering monkey" and the fears of being raped (interview dated June 2011, Professor Eastman).

In this country fitness to plead only arises in the crown court. The concept does not apply in the summary courts, including the youth court. Children can be and are tried in the youth court for serious crimes such as rape and section 18 assault (but not homicide). Unsurprisingly, capacity is often an issue. The courts have learned to explore ways to enable children to give instructions and to give evidence in difficult circumstances (see DPP v P [2007] EWHC 946 at para 52). For example, long breaks can be allowed so that the defendant can, painstakingly if necessary, give instructions on new evidence or evidence that deviates from the anticipated case. Intermediaries can be used. In some circumstances a psychiatrist can give evidence in the trial of a youth to illuminate issues of capacity. In the current case I assume (having no direct experience of current crown court practice) that the court would want to explore with the experts whether there are any measures that the court of trial can take to enable the defendant to participate at the required level (de Kock, para 9.18). This has not been done so far, and indeed cannot be decided without further full consideration by the court and the psychiatrist as to what adjustments to court process can be made. This court cannot embark on those enquiries: it is a matter for the trial court that has conduct of its own procedures. Those procedures may vary depending on the nature of the trial: jury trial or trial by judge and two lay assessors.

It is right to mention that Mr Keith concedes that Mr Dewani is currently unfit to plead. The evidence on that is agreed. His approach is that it does not damage his case that Mr Dewani is currently unfit. The question, he says, is whether if returned it is inevitable that a finding of unfitness to stand trial will be reached. Nevertheless I have considered the current position. I do not consider it unfair to the defendant that I have taken account of the possibility that on this evidence a trial court might consider the defendant currently fit to plead. The advocates understood me to raise it in submissions (as I did) and confirmed that a court deciding this issue would not be bound by the conclusions of the psychiatrists. There will be many cases where the incapacity is so severe that it is fanciful to consider the possibility that the court will take a different view to that taken by the psychiatrists. This is not such a case. On the evidence that I have heard, a court is likely to conclude that he is unfit to plead. However it is not inevitable, either today or in the future.

This is a convenient place to consider the defence argument that extradition is not for the purpose of prosecution, but for civil proceedings, namely proceedings under the fitness to plead provisions. I do not accept that argument. Even if Miss Montgomery is right that fitness to plead proceedings are not criminal proceedings, the purpose behind this extradition request is for the purposes of a criminal prosecution. It is the hope of the prosecuting authorities that a full criminal trial will take place. It is not unrealistic to expect that such a trial could take place. If however a defendant is unfit to plead, a hearing would determine whether he was responsible for the act complained of. If not, this would lead to his discharge without a full trial. Such a hearing and conclusion would form a very valuable process. An alternative determination would be that the defendant was responsible for the act complained of. As I understand it, such a determination would result in detention in a hospital until such time, if ever, as the defendant was fit to face a full trial.

c. Degree of risk of suicide.

Another topic that needs to be considered further is the degree of risk of suicide. The experts agree. The risk is high. It was only towards the end of the evidence that I began to wonder whether I understand what is meant by this.

Unfortunately it is not uncommon for this court to have to consider the risk of suicide in extradition cases. Several such cases have been considered by the Divisional Court on appeal. The degree of risk has been considered relevant. In Jansons there was uncontested psychiatric evidence that if the defendant were extradited to Latvia "he will commit suicide. The report says that in terms or, at the least, one supposes, make every effort to do so". In other cases there is a reference to "a substantial risk that he will commit suicide"; a "significant risk of self harm and potentially suicide"; In Wrobel Professor Hirsch described the risk as "substantially high "and in other circumstances as "extremely high" and later "I think suicidal acts a very high risk". In Prosser it was said that "a very high risk would doubtless be capable of achieving the article 3 threshold" and there is later a reference to " an extremely high risk", to "a severe risk" of suicide, and later still it is said that in a case based on the risk of suicide there must be independent and convincing evidence of "a very high risk of suicide if the fugitive is returned". In Griffin the defendant was assessed as having "a chronic ongoing risk of suicide". Running through these various decisions, and others, is the need to make an assessment of the degree of risk.

However, the evidence before me stated clearly that it is not appropriate to distinguish between high levels of risk, for example by categorizing as very high, virtually certain, or real and substantial. The levels are "mild", "moderate", and "high". Professor Eastman describes the other gradations as all much the same. Professor Kopelman said that "once one is classified as a high risk, it is very difficult to grade further ... within high risk you cannot classify further ..." Perhaps this explains why the agreed report refers to the risk of suicide on extradition being "probably very high" in paragraph 7 and "high" in paragraph 8c. Professor Eastman refers to a "high risk of suicide were he to be extradited to South Africa" (p22, 19 June). Professor Kopelman does not expressly state his own view in his report of 11 July, but refers to the views of Professor Eastman (p20) and Mr Dewani (p22). In evidence he said he agreed with Professor Eastman that extradition would produce an unacceptably high risk of suicide. In his report dated 14 July Dr Cantrell refers to "high risk of completed suicide" (p5), and the increased risk after partial treatment, but does not specifically refer to the risk if extradited.

I had previously assumed, and perhaps I have been alone in this, that measuring the degree of risk involved looking at likelihood or probability. However the evidence in this case, and in particular from Professor Eastman and Dr Cantrell, suggests that this is not the case. For example Professor Eastman said he cannot answer the question in terms of statistics of likelihood. So there is a high risk of suicide, but high risk cannot be more subtly graded. High risk does not necessarily mean statistically likely.

d. 20 February 2011: was it a suicide attempt?

On 20 February 2011, Mr Dewani was admitted to the Bristol Royal Infirmary having taken an overdose. It is believed that he had taken 12/14 tablets of mirtazapine, 28 tablets of zopiclone, and 14 tablets of diazepam. (He told Professor Eastman it was 4-6 diazepam, 4 zopiclone, plus his normal tablets for the day, but the weight of evidence suggests otherwise.) He was very drowsy, and remained so for longer than 12 hours. On examination he said he didn't want to live.

By any standards this is a significant incident, and I have had to consider the facts of it during bail hearings and in these proceedings. He is reported to have said in the A and E department: "I don't want to live." Later he told Dr Dedman that he just wanted to sleep. He told an Approved Mental Health Act worker that this was not a suicide attempt. In her report she pointed out that Mr Dewani had some understanding of different medications, because of his work background and that of his father, and claims to know how much would kill him". (His brother Prien said something similar to Professor Eastman, although in evidence Shrien's knowledge of pharmaceuticals was described as rudimentary.) The same Senior Social Worker records that "he continues to deny that the overdose he took on 20 February 2011 was a suicide attempt but has expressed that he now wishes that he had taken enough to kill himself." In bail applications before me it was asserted on his behalf (apparently disputed by the Republic, at least at one stage) that this was not a deliberate suicide attempt. On balance I accept this. It appears to be agreed that he did not take a potentially fatal overdose. There has been the slightest of suggestions � more a hint really - that Mr Dewani took medication knowing that it would not kill him, but intending it to look like a suicide attempt. In view of all the evidence I have heard, I think that is very unlikely. On balance I accept the conclusion of Dr Cantrell that this was a deliberate overdose to avoid engaging with these proceedings, which is different from an intention to kill himself.

e. Is he faking the symptoms of mental illness?

Although Mr Dewani has not undertaken the so called malingering test, the experts have all considered this question. Their conclusion is that this is highly improbable. Professor Eastman, for example, considers the question in some detail on pages 19 and 20 of his report dated 19 June 2011. He gives six reasons for his conclusion that faking is highly improbable. The reasoning is convincing and on the evidence before me I am satisfied that the symptoms and the illnesses are genuine.

f. Is Judge van Zyl a reliable independent witness?

In her closing submissions Miss Montgomery suggested that Judge van Zyl is not a reliable independent witness. I understand her forensic need to make such an assertion. The judge's evidence is very damaging to a central plank of her argument. Moreover she had laid the groundwork without which such an assertion would have been unacceptable.

Mr Katzen, the defence solicitor, has explained in his witness statement of 22 March the difficulties he found in locating experts on prison conditions who were prepared to assist the defence. He says "We have consistently encountered problems in finding expert witnesses who would be prepared to help us." He believed one witness who withdrew after being formally instructed may have been influenced by concerns that his future funding and access to prisons might be in doubt. The defence witnesses who were called, Miss Dissel and Miss Gear, had asked for co-operation from the Inspecting Judge's office, but had not received it. They wanted access to statistics and Independent Prison Visitors. Among other things, they have not been able to comment on the conditions of the three prisons referred to in the assurances given by the DCS. Judge van Zyl was asked about this. In the notes I made after the first hearing, which are set out above, I said this:

"Miss Montgomery asked him whether it would not be more consistent with an independent approach for him to supply to Miss Dissel such factual information as he was in a position to supply. I thought it was a pertinent question, and did not think it received a direct reply, possibly because the witness did not share my understanding of the meaning of the question. The explanation he gave was that initially he did not give full consideration to the request from Miss Dissel. On reflection he could not provide some of the information requested, particularly information about independent visitors. More importantly, when he learned that the information was in connection with a controversial case his instinct was to avoid involvement. He changed his mind after seeing the Dissel and Gear report prepared for this case, and at the request of the DPP. Most of the facts in the Dissel and Gear report are entirely accurate, and many of those facts are based on official reports including his own official reports. He has great respect for Dissel and Gear. Nevertheless their report runs the risk of presenting a misleading picture. He emphasised that his decision to attend court was to present an accurate picture and does not compromise the independence of the Inspectorate. Insofar as there was possibly a suggestion in the questioning that his judgement had been influenced by pressure, then I am satisfied that is not the case."

Miss Montgomery came back to this question on 18 July. In particular she asked the witness why he was embarrassed. His explanation was that the embarrassment was more to do with the fact that he didn't originally enquire sufficiently about the case being referred to. Also, because of the sensational press and publicity, he did not want his office to become involved. When he thought about it he realized it was totally unacceptable for his Independent Visitors to be contacted. He repeated that his embarrassment was probably as a result of his own failure to apply his mind to what he was asked. I did not consider this an entirely convincing explanation and it caused me to re-examine the impression I had of the evidence after the February hearing.

I would have preferred the defence experts to have had as much co-operation as possible in the preparation of their report. I would have liked them to have had access to information about the three prisons named in the undertaking given on behalf of the Republic of South Africa. I can understand why Miss Montgomery says that it is inconsistent with complete impartiality to deny access to the defence for the reasons given, and then be prepared to give evidence for the government. However, after careful consideration I am fully satisfied with the integrity and independence of the judge. In reaching that conclusion the following matters have weighed with me.

• His qualifications and background.

• The impression I gained of the judge during the long period during which he gave evidence before me.

• The respect with which he is undoubtedly held in South Africa. For example a press cutting criticizing him for referring to a prison as a five-star hotel, and casting doubt on his motives, nevertheless said: "Van Zyl is generally accepted as a judge of the highest calibre and highest probity" (Saturday Star 7 May 2011, Vol 3 tabs 14 and 16).

• The defence witnesses themselves, Miss Dissel and Miss Gear, raised no suggestion about the lack of independence of the judge.

• The court and the defence have been provided with the April inspection reports for the three named centres.

• He is a fierce critic of prisons in South Africa. His reports have consistently condemned conditions in those prisons and urged improvements.

• On reflection, I accept the reasons he gives. It would have been a normal reaction for a cautious man undertaking a difficult and independent job with the Prison Inspectorate not to want to be associated with a controversial case. I understand that it could be seen as embarrassing. It would also have been a normal reaction to reverse that opinion when he saw a report prepared for this hearing that, in his opinion, did not reflect the current situation accurately.

There was a second string to Miss Montgomery's argument. She pointed out that the witness had consistently and persistently failed to endorse critical comments of the conditions in South African prisons made in apparently reputable sources, such as the Jali Report, parliamentary committees, and newspapers. That is true. The judge made it clear that he is able to answer questions about the current state of the prisons in South Africa. He was not prepared, for the most part, to comment on conditions before his appointment as Inspecting Judge. He made it clear that he could not endorse comments when he did not know their provenance, or was not familiar with the data supporting them. He pointed out that much of the Dissel and Gear data was collected many years ago. He was clearly not prepared to endorse hearsay or comments that may have had a political motive rather than a statistical basis. Sometimes facts and figures were put to him that were demonstrably false, and he said so. For example there is a repeated claim that it is 99.9% guaranteed that an inmate would be violated or abused en route to prison. This claim can be found on a website headed Parliamentary Monitoring Group where an organisation known as Friends Against Sexual Abuse made the allegation. In the same document there was an allegation about "pokers" which is where gangsters insert drugs forcibly [into the rectum] of a rape victim. The allegations have been repeated in a variety of contexts and have been presented as an almost inevitable consequence of being imprisoned. The judge was prepared to describe them as without foundation, nonsense, and at variance with the other statistics provided. It was pointed out that the founder of what is elsewhere described as Friends Against Abuse is Ms Lizelle Albertse, whose inspection report on Brandvlei in April 2011 is referred to later in this judgment. Similarly, a report put before an Australian court in 2004, De Bruyn, quoted Gideon Morris of the South African Judicial Inspectorate Office as saying that "a prison sentence is tantamount to a death sentence by HIV/AIDS. Morris estimated that between 70% and 80% of suspects held in South African jails are raped by fellow prisoners before they are even officially charged." It was put to Judge van Zyl that the same Australian case referred to a UNAIDS report as saying that the situation in South African jails "has reached catastrophic proportions with 45,000 prisoners dying of AIDS- related diseases every year". This same report referred to prison gangs using HIV infection as punishment. It was reported that gang members carrying the HIV virus were ordered to rape disobedient inmates in a ritual known as "slow puncture", by which the victim would die over a period of time. He said that is obviously nonsense. It is wrong to criticize him for his approach. He should not endorse a comment unless he is satisfied it is true, and he could not be satisfied with many of the quotations that were put to him. For example the quotes referred to in this paragraph look authoritative at first sight. However they can only be described as nonsense. If 45,000 prisoners are dying every year of AIDS-related conditions that would be well over a quarter of the prison population at any given time, and maybe 50 times higher than the true figure. The Judge told me the total when he gave evidence in mid-July was 466 for the year so far. In contentious areas the witness, and indeed this court, should be very cautious about statistics whose provenance is unknown.

Overwhelmingly, when the whole picture is considered, I have no doubts whatsoever about the integrity and independence of Judge van Zyl. On the contrary, the record of his work in his current position shows a man of integrity, and independence, who has demonstrated a willingness to attack prison conditions where appropriate.

g. If Mr Dewani is extradited to South Africa, will he be detained and if so where?

Professor Sean Kaliski, of the Forensic Mental Health Service at Valkenberg Hospital, wrote on 13 July that there is no official injunction that prevents awaiting trial prisoners from being admitted to the forensic mental health service in the Western Cape. Most admissions are by court order. However a patient can be admitted either as an assisted, voluntary or involuntary patient under the Mental Health Care Act. He says there would be no obstacles to admitting Mr Dewani directly into the maximum secure unit at Valkenberg Hospital. He suggests that in this case a place would be found for Mr Dewani urgently, even though there is normally a waiting list. It is very likely that Mr Dewani will be referred under section 79 of the Criminal Procedure Act for a formal assessment. If he is as seriously ill as indicated in some of the reports, he will remain in the ward until he is triable. Alternatively, he may be certified as a state patient and remain indefinitely in the facility. "Given the prominence this case has in the media we would assure the court that we would admit Mr Dewani to our facility on his arrival in Cape Town."

It was explained to me in submissions that one possibility is that on arrival in South Africa Mr Dewani would be assessed and if found to be seriously ill he could be admitted to the hospital without appearing in court. I cannot tell whether this is likely.

Alternatively, and more probably in my view, Mr Dewani would be produced in court where he would have a right to apply for bail. Rodney de Kock, the DPP for the Western Cape, has provided details as to the legal position with bail. In summary, the defendant in this case would have to satisfy the court, on a balance of probabilities, that exceptional circumstances exist which permit his release on bail in the interests of justice. He confirms that in all the circumstances the State would seriously consider opposing bail, but the issue is determined by the court, bearing in mind all relevant circumstances, including health.

In her oral submissions Miss Montgomery said that it is most unlikely that this defendant would be transferred to a hospital. She pointed out that the Republic had consistently opposed bail in these proceedings. They would continue to do so. Therefore, she says, it is highly likely that her client would be detained in prison. I do not accept the logic of that submission. It is true, and may well remain true, that the government opposes bail. However South Africa has an independent judiciary. I have every confidence in my colleagues in South Africa to consider an application for bail on its merits. Or course the courts there will require their own assessments and their own evidence. Of course they might not take the same decision on bail as the courts in this jurisdiction have so far.

Their assessment of the appropriateness of bail may well be different. Here I can say with some authority that the initial decision to grant bail, and subsequent decisions to continue bail, have been marginal. I would not presume to suggest to my colleagues in South Africa how, if the question arises, they should determine an application for bail. I have complete confidence that the courts there would take that decision on its merits and in the light of all the evidence put before them. I accept as a real possibility that if extradited in his current state of health Mr Dewani could be held in Valkenberg Hospital, either by direct transfer or by court decision.

If on the other hand he were refused bail, or later were to serve a sentence as a convicted prisoner, we have assurances as to the prisons in which he would be detained. Miss Montgomery says that those assurances would not necessarily be adhered to. There may be a change in the official position, or alternatively a change in circumstances.

Nobody can rule out unforeseen circumstances. To a certain extent all assurances potentially could be overtaken by events. An assurance can never be seen as a cast iron guarantee. However I accept that the assurances of the South African authorities are made in good faith. Moreover I am satisfied that there are a number of reasons to believe that they would be adhered to in all but the most extreme change of circumstance. The first is that we have an assurance, given to this court, by Judge van Zyl that he will ensure that the conditions are monitored and adhered to. He accepts that the undertaking has been made in all good faith and that the Department should have little difficulty in carrying out the undertaking it has given. Secondly it is clear that the Republic has a lively and healthy parliamentary democracy that scrutinizes the activities of the government and the administration with considerable care. Thirdly there is a strong independent press. Finally I have reason to believe that the South African people consider this case to be of considerable importance, not least for their standing in the world community, and expect their representatives to honour an undertaking given to the courts of this country.

h. Likely impact of extradition on mental health and risk of self-harm or suicide.

The experts' joint statement says that they believe that Mr Dewani's mental disorders would be highly likely to worsen further, and his risk of suicide would become even higher, probably very high, in the event of an order being made for extradition. Professor Eastman agrees with Dr Cantrell's observation that Mr Dewani is less likely to kill himself in England. However, were he to be extradited to South Africa the risk would rise greatly. Also, were he to be detained in prison rather than in hospital the risk would rise even further. The possibility of managing that risk, and reducing the chances of completed suicide in prison by comparison to being in hospital, would also be far lower. The professor had only limited information about what therapeutic environment and what specific treatment might be available in South Africa. He therefore found it impossible to offer a realistic opinion about the prospects of treatment of Mr Dewani's mental illnesses in prison there. On the limited information made available to him he was not satisfied that his illnesses could be adequately managed in that context. Professor Kopelman commented that Mr Dewani made it very clear in April, and strongly intimated in July, that suicide would be a very serious risk if he were extradited to South Africa. "My impression was that this accurately reflected his views and was not something just said to help his extradition case."

The experts did not explain in detail in their report why the risk of suicide would increase if Mr Dewani is extradited. Perhaps they thought it obvious. In evidence Professor Kopelman said there is extensive literature on how reminders, such as going back to a previous environment, lead to re-traumatization and marked deterioration, which in his view was "a big factor". There is no suggestion that Mr Dewani would go back to the scene of these intensely traumatic events, and I assume the professor means that simply being in the same country is likely to have a harmful effect.

Professor Kopelman said that Mr Dewani benefited from family support that would be less available in South Africa. He did not explain the basis of that statement. Again perhaps he thought it obvious. The defendant would not be able to visit his family home and presumably would not see his extended family or friends. However I would be surprised if his close family did not remain with him, at least in the early months. Miss Montgomery points out that there are frequent complaints in the South African prison system that prisoners do not have access to their family. In view of the other assurances and factors in this case I believe it is likely that this defendant would have access to his family in South Africa, even if detained in prison. Nevertheless the professor's assumption that family support would be less available is justified.

The professor made the further point that disrupting current clinical care would itself be a disadvantage. He thought that the care in South Africa would not be as good as Fromeside. Mr Keith on behalf of the Republic conceded that care in a South African prison would not be as good as the current treatment. This concession was based, I think, on the lack of access to expert psychiatric care in prison. It had been my understanding from earlier evidence that private care can be made fully available in prison. Judge van Zyl and Dr Panieri-Peter said as much. Even so this is thought likely to fall short of the care provided by the current treatment, and on reflection Mr Keith was right to make the concession. There appeared to me to be an occasional underlying assumption that the treatment offered in hospital in South Africa would not be equal to the treatment in Fromeside. Professor Kaliski confirms that his hospital is also a teaching hospital for the Department of Psychiatry at the University of Cape Town. "As is commonly known in the UK, many eminent and excellent South African psychiatrists now practising in the UK trained in our department and at this hospital." The judge told me that the facilities in the Valkenberg Hospital are in order and that he had heard nothing negative about them in the past three years, although he has not visited and inspected them himself (nor, as I understand it, have members of his staff or the inspectors). He referred favourably to the experience of Professor Kaliski and his staff. Dr Panieri-Peter (instructed by the defence) has many years of working at Valkenberg. It is her view that Mr Dewani would be referred there for observation. She refers to "considerable negative press" over the years of Valkenberg, and plans for building a new unit. However she does not offer her own opinion on the quality of care offered.

Professor Kopelman also said that the defendant's fears about what might happen in South Africa, for example assault or rape, are very real perceptions for him. I accept that even if those fears are not based on events that are likely to happen, they are real.

The flight itself would be extremely distressing and "may make the disease worse" (Professor Eastman in evidence). "He would probably need two escorts. It would be extremely distressing for him. But yes he can get on a plane" (Professor Kopelman in evidence).

The conclusion was that on arrival, and probably for a considerable period thereafter, there would be massive anxieties. Even acquittal, although helpful, is not likely to terminate all his symptoms. "I do know that if he were sent to South Africa there will be an exacerbation of his conditions to a very serious degree. This may be partly mitigated by anti-depressants, but we can't rely on that alone."

When Dr Cantrell, and later Professor Kopelman, gave evidence about the reasons why the risks to Mr Dewani would increase on extradition, I did not find those reasons compelling. It occurred to me that some at least of the reasons were based on assumptions that may or may not be accurate. For example at times I thought there was an apparent assumption that treatment here is generally better than in South Africa. It is not obvious on the evidence that this is true for those with means. Dr Panieri-Peter, a specialist psychiatrist in South Africa, is critical of facilities in prison. She says it is very difficult to arrange private psychiatric treatment in prison. While that may well be the case normally, I am satisfied by the assurances I have had that Mr Dewani would receive such treatment. Nevertheless the fact remains that the experts agree that there is a high risk to Mr Dewani even if he remains in this country. Court proceedings are a maintaining factor, and paradoxically the risk of suicide increases as he begins to recover. The risk will be made worse if extradited. That is a conclusion that I accept.

i. General prison conditions in South Africa

It is obvious, not least from the reports of Judge van Zyl and from the discussions in Parliamentary committees and in newspapers, that there is considerable informed concern about conditions in prisons in South Africa. It is equally clear that some of the allegations that are made are unjustified. For example I was taken on a number of occasions to a suggestion that someone arriving at a prison is almost certain to be raped. This is obviously untrue.

Nobody in this case, and certainly not Judge van Zyl, has said that prison rapes do not occur. This may be a convenient place to mention some recent statistics whose provenance is known. It comes from the annual report of the Judicial Inspectorate of Prisons for 1 April 2007- 31 March 2008, that is just before Judge van Zyl was appointed. It is described as a "customer survey" amongst some 750 prisoners with a view to testing directly from them experiences and perceptions about conditions in prison. The sample size was small (at about 0.5% of the total prison population) and selected from 46 prisons (20% of the total). To the question "while in prison, were you subjected to unwanted sexual attention?", the answer was "yes" 7%; "no" 93%. To the question "are you currently suffering or have you suffered from any medical condition?", the answer was "yes" 32%; "no" 68%. To the question "have you ever been a victim of violence whilst being incarcerated?" the answer was "yes" 15%; "no" 85%. The perceptions about how often sexual abuse happens in prison are different. Only 34% say it never happens while 12% say it happens very often. This difference between experience and perception is unexplained, although it is not uncommon in other situations and perhaps not surprising when there is publicity suggesting that rape is all but inevitable in prison. Miss Montgomery said that 7% being subjected to unwanted sexual attention is still too high, which indeed it is. However I do not accept that unwanted sexual attention is invariably physical abuse, or "unwanted sexual violence" as she describes it. It can be comments. Also it appears very likely that those in communal cells receive unwanted sexual attention far more frequently than those in single cells. The defence submission that 7% means that 100% of all vulnerable prisoners are reporting sexual predation is ingenious but unconvincing. As for those suffering from medical conditions, the figures are obviously at variance, if any evidence were needed, with the death figures of 45,000 a year reported in the 2004 Australian case of De Bruyn. That figure is manifestly and obviously wrong. In that context it is worth mentioning that the death rate in custody is falling. The judge gave me figures of 466 for the year so far. He says that is far too high a total. He has consistently suggested that arriving prisoners should be given a proper health screen and treated appropriately. That, it seems, is not happening so that (he believes) many people are dying in prisons of conditions they had contracted before admission. In 2009-10 there were 30 suicides and 19 homicides in South African prisons. In the institutions where this defendant would be held, according to the undertakings, the level of violence and death is reported below.

The 2009 operational guidance notes issued by the UK Border Agency in relation to South African prison conditions say: "whilst prison conditions in South Africa are poor, conditions are unlikely to reach the Article 3 threshold." In immigration cases (and the same applies in extradition) the individual factors of each case should be considered to determine whether detention will cause a particular individual in his particular circumstances to suffer treatment contrary to Article 3. Relevant factors include the likely length of detention, the likely type of detention facility and the individual's age and state of health. Miss Montgomery in her opening submissions argues that the evidence in this case shows that the guidance notes understate to a significant degree the severity of conditions in the South African prison system. She points out that even within those guidance notes it is recognized that there are circumstances where the risk to a potential prisoner will give rise to a violation of Article 3. Evidence was called in this case to show that Mr Dewani has personal characteristics that make him a particularly vulnerable prisoner. In addition it is argued that his health makes him particularly vulnerable.

I have accepted the assurances as to the prisons in which Mr Dewani would be detained, if he is detained at all. In those circumstances I need not consider the conditions in the general prison estate. I hope this is not seen as a discourtesy to the witnesses who took so much trouble to provide expert opinion. In fact there was a considerable measure of agreement between the witnesses as to current conditions.

j. Conditions in the prisons where Mr Dewani would be held if detained in South Africa.

The National Commissioner of the Department of Correctional Services (DCS) on behalf of the Republic of South Africa has given a series of undertakings. Those undertakings are attached to this judgment (Appendix A). In summary they confirm that if extradited Shrien Dewani would only ever (potentially) be detained at one of four facilities, namely:

a. If remanded in custody pending trial, at Goodwood Correctional Centre in a single cell in the sick bay area.

b. If convicted, sentenced to a term of imprisonment and classified as a "medium" security risk, at Malmesbury Medium A Correctional Centre in a separate cell with a flush toilet and hot and cold water.

c. and d. If convicted, sentenced to a term of imprisonment and classified as a "high" security risk, at Brandvlei New Correctional Centre in a separate cell with a flush toilet and hot and cold water. Following the scheduled upgrade of Brandvlei Maximum Correctional Centre he would be moved in there and detained in similar circumstances.

There is no reason to doubt the sincerity of those undertakings or the ability of the South African authorities to fulfil them. The undertakings were described as "encouraging" by the defence experts. They were endorsed by the Inspecting Judge who in turn undertook to monitor them. Earlier in this judgment I have given my reasons for concluding that the undertakings would be fulfilled.

A very considerable body of evidence about conditions in these prisons was put before me. That evidence comes from the DCS report dated 15 April 2011, and from the Inspecting Judge, Judge van Zyl. It is most unfortunate that the defence experts did not have the opportunity recently to visit those institutions themselves or provide their own account of them. However they did not, as might have been open to them, cast doubt in any significant way on the overall conditions there. Indeed Miss Dissel appeared to agree that some correctional centres "don't appear other than in relatively favourable terms". Moreover, it is obvious from the written evidence provided in the July hearing that the judge's conclusions were widely available in South Africa for several months before he was further cross-examined by Miss Montgomery. They were commented on. I have no doubt that in an active parliamentary democracy, with a free press, significant evidence would have been forthcoming to contradict the judge's conclusions, had that evidence existed. There was some new material, but more limited in scope than might have been expected, and overall it casts no doubt on the judge's evidence.

Judge van Zyl gave his overall opinion that each of the correctional centres mentioned above is "suitable and appropriate for the purpose". I am satisfied that if Mr Dewani is extradited and detained in custody he will be held in a single cell in one of the four named institutions. None of those institutions suffers from the level of overcrowding or staff shortages that have been criticized as a significant cause of poor conditions in other parts of the prison estate in South Africa. There is no reason to doubt that he will receive appropriate health care and treatment. The judge told me that Mr Dewani would have "no difficulty at all" making special arrangements for private care and for access to psychiatrists and psychologists perhaps of quite renowned levels. Dr Panieri-Peter said it would be very unlikely that delays would occur in remanding Mr Dewani for medical observation in hospital or that he would "get lost" in the prison system.

Mr Keith, on behalf of the Republic, concedes that the conditions in these prisons would not match those in Belmarsh, as far as medical treatment is concerned. Concessions by counsel are of great value in court hearings. They remove the need for the court to consider uncontested issues. They prevent time being spent on evidence that does not help resolve the case. I therefore do not look behind the concession made by the Republic. However the concession does not go to the heart of any issue I must consider. The question is not whether facilities are better in London than in the Western Cape. It is clear from the detailed evidence put before me that conditions in the four named prisons are appropriate. Judge van Zyl gave evidence that Shrien Dewani would be able at all times to be seen by a private doctor of his own choosing. It has not been suggested that the defendant and his family do not have the means to pay for that treatment.

Conditions in the named correctional centres have, as I have said, been set out in considerable detail in the papers and I have been taken to the relevant passages. There is a report on the undertakings given by the DCS dated 15 April 2011 and running to 57 pages. The conclusion of that report prepared by the DCS is that the undertakings given as to the specific circumstances under which Mr Dewani would be detained mitigates the risks identified. It says that extradition within the parameters of these particular undertakings would be compatible with his rights under the European Convention on Human Rights. Among other things, the report describes conditions in Goodwood, Malmesbury Medium A and Brandvlei New Correctional Centre. I have also been provided with comments from Judge van Zyl dated 18 April 2011 on that report. Further I have been provided with the inspection report for Goodwood dated 5 April 2011; the inspection report on Malmesbury Medium A dated 7 April 2011; and the inspection report on Brandvlei Maximum Correctional Centre dated 6 April 2011. That evidence is summarized in the Preliminary Skeleton Argument dated 26 April 2011 prepared by counsel for the Republic. I can summarize that still further.

Goodwood it is a modern build that meets UN Standard Minimum Rules for the Treatment of Prisoners. It is one of the DCS's "centres of excellence". The sick bay has never exceeded its maximum occupancy level. In April 2011 the Centre as a whole was operating at an occupancy rate of 113%. Goodwood has its own health care unit which is separate from the general prison population and comprises a clinic and the sick bay where Shrien Dewani would be detained. On a recent visit three of the patients displayed psychiatric conditions (and according to Michael Prusent, who inspected on 5 April 2011, a further four with schizophrenia). There are two full-time psychologists and a medical doctor who visit three times per week. The conditions there were described as "certainly in the upper echelon" by the judge. In the event of an emergency, there is a named referral hospital. Any inmate can at any time (with permission) be seen by his own doctor and/or psychiatrist, at his own expense. In the event of an emergency he can request that he be taken to a doctor of his own choosing, again at his own expense. Mr Dewani would be entitled to visits from his family, legal representatives, and consular officials. In the last financial year (2010/2011) there were 35 assaults on inmates, two of which were sexual in nature. None was in the sick bay area. Two officials were warned for assault on inmates. There was one suicide. There were no suicides in the sick bay area. The injury register reflected, according to the inspection report, the following injuries: laceration, head injury, face and neck bones and stab wounds. The Head of the Centre reported to the inspection that "gang activities in the community at times negatively influence the situation in the centre." According to Judge van Zyl, Goodwood is in a class of its own compared to other prisons such as Pollsmoor (p59 transcript tab 2). The defence experts appeared to acknowledge that Goodwood is a centre of excellence.

Malmesbury Medium A is under-populated. A variety of recreational activities are available and inmates can take at least one hour of exercise a day. There are programmes addressing spiritual care and other rehabilitative programmes. Malmesbury A incorporates a hospital section or primary health clinic that has 4/5 nurses and a doctor visiting twice a week. In the opinion of the Investigating Judge the hospital section is well maintained. An inspection on 7 April by Mr Professor Mohlaba described the hospital as excellent, clean and hygienic. Inmates with mental health problems are referred for assessment by an outside psychiatrist who visits once per month. In some circumstances an inmate can be transferred to a public hospital or, at his own expense, to a private hospital. Arrangements are in place for those exhibiting suicidal ideation. In the past financial year a total of 25 assaults were recorded on offenders by other offenders or by officials. In the calendar year 2010 there were four other assaults of a sexual nature. There have been no unnatural deaths (including suicide) since 2009. The case officer interviewed a few inmates who raised no concerns.

Judge van Zyl said Malmesbury Medium A impressed him, when he first went there, as likely to be considered to be too much like a five-star hotel (transcript p60).

Brandvlei Maximum/Brandvlei New Correctional Centre. The New Correctional Centre is currently being upgraded, so if Mr Dewani is convicted and sentenced and classified as a high security inmate it is likely he would be detained in Brandvlei Maximum. This is a new generation prison. When upgraded it is expected to comply with all the required International Standard Minimum Rules. It is said that the new build is likely to be no less good than those at the operational new generation prisons mentioned above. The April 2011 inspection report prepared by Ms Lizelle Albertse refers to the current centre having 43 inmates on psychiatric treatment. Two of those inmates have been referred to Valkenberg by the court, but are still at the clinic as the hospital cannot accommodate them at the moment. In 2010 there were 67 inmate-on-inmate assaults reported, seven member on-inmate assaults, 23 self-inflicted injuries, 35 injuries caused by accidents and seven attempted suicides of which one was fatal. There was then an update of the injury register so far in 2011. The inspector conducted interviews with inmates that revealed that they are satisfied with access to the internal complaints procedure. Judge van Zyl said that when the new prison is complete it will compare with the best security rehabilitation facilities in the world (transcript p106).

Valkenberg Hospital was not specifically mentioned in the undertaking given by the DCS and I have far less detailed evidence about it. There was a report in April 2005 that described the hospital as "barely fit for human habitation" and there have been unfavourable references to conditions since. The defence witnesses from whom I have heard had not visited the hospital and were relying on other accounts. The premises have to some extent been refurbished recently. Judge van Zyl has been to the hospital, but not to conduct an inspection. He did say that the facilities themselves are in order and that he had never heard anything negative about them in the past three years. Dr Panieri-Peter worked at the hospital for many years and referred to considerable negative press about the conditions, and especially those within the Forensic Unit. She referred to plans for building a new unit and says that Professor Kaliski has been quoted as describing the poor conditions in the unit. However she did not directly criticize conditions there. Professor Kaliski says that "in the past the conditions in the maximum secure ward were abject, and were often decried in the media. The ward has been renovated substantially in the last four years. The conditions are not luxurious, or perhaps comparable to the best that perhaps exist in the UK, but the ward is habitable." He refers to the fact that the hospital is also a teaching hospital for the University and that many eminent psychiatrists trained there. The witnesses who gave evidence commented on accounts given to them and were not in a position to comment directly. They were not reassured by the description of the ward as "habitable ". I will return later to the possibility of Mr Dewani being detained at Valkenberg. However there is no reliable current information to contradict the evidence of Judge van Zyl and Professor Kaliski. I have no reason to believe that conditions there are unacceptable.

In addition to the specific assurances about prisons in this case, the Republic of South Africa has explicitly adopted all relevant international and regional treaties. It is bound by its own Constitution which incorporates a full Bill of Rights. The specific assertion made in these proceedings - that since the end of apartheid South Africa has consistently upheld both the rule of law and individual rights - has not been challenged. The Republic is now a well established democracy with a vibrant and free press. There is an established Judicial Inspectorate, with Independent Prison Visitors. "My office is able to monitor every single inmate at every prison throughout the country, which would include Mr Dewani should he at any stage be placed in any of those or any other correctional centre in the country" (Judge van Zyl in oral evidence). There are internal complaints procedures. It is undoubtedly the case that there are severe problems in some of the prison estate, caused at least in part by overcrowding which leads to a number of serious difficulties. However even those problems are under scrutiny. Nobody is saying that there is not room for significant improvement in the general prison estate. Campaigners have understandably concentrated on the worst institutions and those that fall below an acceptable standard. However, I accept the assurances given and accept the findings about the conditions in which Mr Dewani would be detained, if detained at all. Problems with gangs, sexual assaults, and possible victimization of Mr Dewani if in custody, have all been comprehensively reviewed in the evidence. Mr Dewani will be held in a single cell, and the defence experts agreed that "what happens in a single cell if there is only one person in that cell bears no relation to what happens in communal cells". I am satisfied the authorities will take all reasonable steps to protect him, and that the risk of assault (including sexual assault) will be reduced very considerably by the conditions in which he will be held.

I am also satisfied that concerns about attacks at court have been considered, and dismissed in the evidence. The suggestion that the risk of an attack on the way to court was 99.9% has been completely discredited by the evidence and is not accepted by the defence experts. There was a body of evidence about gang-related activities at prisons across the country but Judge van Zyl told me that he didn't think this defendant will ever be exposed to any risk as far as gang activities are concerned.

I accept the conclusion of Judge van Zyl that there is no reason to doubt that Mr Dewani would be well cared for in those institutions where he may be detained.

Articles 2 and 3

The defence argues that it would breach Articles 2 and 3 of the European Convention on Human Rights to order extradition of this defendant to South Africa. It would breach Article 2 if there is known to be a real risk to Mr Dewani of loss of life in the receiving country. To establish Article 3 the defence must show that there are substantial grounds for believing that there is a real risk that the requested person will be subjected to torture or to inhuman or degrading treatment or punishment in the requesting state. In each case it is necessary to show there where the risk is from non-state agents, that there is in addition a lack of reasonable protection in the receiving country. It is common ground that an assessment must be made according to the specific circumstances as they would apply to Mr Dewani rather than generalized concerns.

A high threshold is required to establish Article 3. The ill-treatment must necessarily be serious such that it is an affront to fundamental humanitarian principles to remove an individual to a country where he is at risk of serious ill-treatment.

The defence argues that its case is strengthened by the fact that the South African government thought it necessary to give special undertakings as to the specific prisons in which this defendant would be held, and the fact that he would be placed in isolation in a single cell. Counsel adds that Judge van Zyl thought it appropriate to strengthen these. A further argument is that this defendant is especially vulnerable as a young, light-skinned and good-looking man who is accused of a "sissy" crime. Further he has been described as a monkey, which has marked him out as a trophy suitable for murder, rape "or worse". He is further vulnerable because it has been suggested that he is homosexual. A number of deeply offensive and threatening electronic or website messages (of the type that unfortunately often accompany high profile cases) were brought to my attention. He is at high risk of suicide. He is at risk from gang violence, sexual violence, threats to his physical health (such as HIV/AIDS) and his mental health and vulnerability. No undertakings have been given about Valkenberg Hospital, and the prison officers association has said publicly that it is not appropriate for this defendant to receive special protection. Further if this defendant is detained indefinitely in solitary confinement (which would be the effect of the assurances if he remains in prison after a finding that he committed the act or was guilty of the offence) then this would be inhuman.

I have considered the risk of suicide earlier, and will return to it later. That risk does not mean that extradition is in breach of Article 2. Bearing in mind the assurances given, there is no evidence from which I could properly conclude that there is a real and immediate risk to life if extradited.

I have already said that I accept the assurances given. I do not accept that following those assurances would amount to cruel and inhuman punishment if the defendant remains in custody for a very long time, or even indefinitely. The evidence as to the risk of violence is best assessed in the light of reported allegations of violence in the penal institutions where this defendant may be detained, namely Goodwood, Malmesbury A and Brandvlei. The most recent statistics are set out above. It is likely that there is under-reporting of violence. In almost any circumstances there is a risk of violence, including sexual violence. It is generally accepted that the risk of violence inside prison is greater than in the community at large.

It follows that there is a generalized risk of violence to a prisoner entering the South African prison system. There may or may not be increased vulnerability for Mr Dewani, as the risk factors and the protective factors point in different directions. Dissel and Gear believe he is at increased risk and I accept their assessment as likely rather than definite. There is a high incidence of HIV/AIDS in prisons as in society generally. There are repeated reports of sexual violence, and this carries with it the threat of contracting disease. I am satisfied that the general level of risk has been greatly exaggerated in some cases. The incidence of serious sexual assault is not 99.9%: it is hugely lower than that. Forty-five thousand prisoners a year are not dying in South African prisons, as was reported to the Australian courts. Nevertheless it is undoubtedly the case that there is some serious violence and some serious sexual violence.

Mr Dewani's defence team has worked hard to consider the safety of their client in a number of different situations. For example it is said that a prisoner is particularly vulnerable on the way to court and in the cells at court. Not all of the evidence for that is anecdotal, of the "99.9% certain to be raped" variety. Individual allegations have been brought to my attention and some time has been spent on them in evidence. Judge van Zyl said that arrangements could be made for this defendant to be brought to court singly in a police car. He dismissed the possibility of him being attacked in court. Similarly the defence has queried the safety of the system in single cells after lock-up, when a single member of staff may be on duty. Concern has also been expressed about Mr Dewani's vulnerability on his way to shower or to exercise. If held in hospital, he may not be in a single room and may be exposed to danger from other patients. Although in each case it is impossible to quantify the risk, there is no doubt that some risk exists in each of these situations.

However the harm identified in this case does not come from the South African authorities themselves. Any harm inflicted by non-state agents will not constitute Article 3 ill-treatment unless in addition the state has failed to provide reasonable protection. Here I have been impressed by the willingness of the authorities to provide assurances. Taken as a whole, those assurances are sufficient to persuade this court that the South African state will comply with its positive duty to provide reasonable protection against criminal acts directed at Mr Dewani. Although some prison officers and others may think the special treatment offered here is unfair to other prisoners, I see no reason to believe that they will disobey instructions or act in an unprofessional way. No assurance can provide a complete guarantee. No prison authority anywhere can say with certainty that a prisoner in their care will not suffer any harm. However in this case the authorities have gone as far as they need, and perhaps further, to satisfy me that neither Article 2 nor Article 3 will be breached by extradition.

In summary, I accept there is a risk to Mr Dewani, and that the risk comes primarily from the risk of suicide and from other prisoners, if he is held in prison. However those risks have been drawn to the attention of the authorities, and I am impressed by the level and the nature of the assurances given. The Articles 2 and 3 points fail because I am satisfied that the South African authorities can provide proper care and protection to this person, and will take all reasonable steps to do so.

Section 91 Extradition Act

Section 91 reads:

"(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."

I accept that section 91 provides additional protection; in other words even if a defendant fails to show a human rights barrier to extradition there can be circumstances in which extradition would be barred by section 91. However insofar as the defence in this case sometimes appeared to say that the physical and mental condition of the defendant should be looked at in isolation, without considering other competing public interests, then I do not accept that argument.

It is common ground that to satisfy a finding of "unjust or oppressive" a high threshold must be attained. That threshold is not reached by showing hardship. The bar is very high.

In Tajik v USA [2008] EWHC 666 (Admin) it was said:

"Whilst a judgement has to be made in every case by reference to the particular facts, it is clear from those authorities that in practice 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."

In Spanovic v Croatia [2009] EWHC 723 (Admin) it was said at paragraph 39:

"It is plain to us that the bar is set very high, and the graver the charge the higher the bar, in that there is a heightened public interest in the alleged offender being tried: provided, of course, that the trial and the conditions in which he will be held will be fair."

So it is clear that the bar is very high. There is good reason for this. There is a strong public interest in honouring our extradition treaty obligations. There is a strong public interest in alleged offenders being tried.

Whether or not it is unjust or oppressive to extradite a defendant is a matter for the court to determine on the facts of a particular case. Generally it is unhelpful to consider the facts of other cases in reaching that determination. The court must apply a sense of justice to all the facts and reach a decision in accordance with conscience.

There is no doubt that Shrien Dewani suffers from two severe and incapacitating mental illnesses. He is not faking them. There is currently a real and significant risk of suicide. That suicide risk will increase if he is extradited to South Africa. No father would find that an acceptable risk for his son. No doctor would find that an acceptable risk for his patient.

I accept the assurances given by the South African authorities as to the circumstances in which Mr Dewani would be detained if extradited.

I have complete confidence in the South African system of justice. I have been provided with a statement from Rodney de Kock, the DPP for the Western Cape Provincial Division of the High Court of South Africa. He has summarised the relevant constitutional rights underpinning the independence of the judiciary, the separation of powers and the framework of the constitutional law, particularly as it applies to criminal proceedings. His statement confirms the independence of the courts; the constitutional guarantee of a fair trial; that fair trial rights are enforceable; that the burden of proof is "beyond a reasonable doubt"; and that the court gives reasons for its decisions. The defence has not disputed the fairness of the South African judicial system. I have no reason to doubt that unless the evidence satisfies the court that Mr Dewani did the act complained of, then he will be acquitted or discharged, as the case may be.

Other than that, the future for Shrien Dewani's health is speculative and uncertain. Whatever happens he may remain seriously unwell. The longer he remains seriously unwell the greater the risk of suicide. These proceedings, or the expectation of future proceedings, will be a maintaining factor for his conditions. On the other hand there are reasons for optimism. Most people with his condition improve (although the risk of suicide increases at the early stages of that process). There is every reason to believe that he will receive good medical care in South Africa, even though it has been conceded that the care would be better in this country. There is a high risk that he will attempt suicide, particularly if extradited. However I am satisfied that the authorities are alert to that possibility and will take appropriate steps to protect him from that risk.

Professor Kopelman said (see above):

"The problem with extraditing now is that even if they have the best possible psychiatric facilities in South Africa, once Shrien Dewani is extradited in his present clinical state, it will be very difficult to get him to a state where he is fit. The goal should be to get him clinically better in terms of depression/PTSD, so that here, in the UK, we know he is fit to plead. It is unusual for depression/PTSD to be grounds for unfitness, so we would expect him to come back to a state of fitness at some stage."

It is an attractive argument that the best outcome in all the circumstances would be for Mr Dewani to remain in his current treating environment, at least for the next six months or so. That, at its most optimistic, is the minimum time that would be required for him to be clinically better in terms of depression/PTSD, to be fit to plead, and for extradition then to take place. However that is not the question I am asked. The question is whether extradition now is oppressive.

The defence position can be put simply, I hope not too simply, in the following way.

If our other arguments are unsuccessful, then we accept that Mr Dewani must be extradited once he is fit to stand trial. There is a reasonable prospect that if left in his current therapeutic setting he could improve sufficiently to be fit for trial at some stage. If he is extradited now to South Africa that process is likely to be set back indefinitely. He will get worse rather than better. The risk of suicide will increase to an unacceptable level. Mr Dewani is not at the moment in a position to give instructions. The decision to contest these proceedings is taken on his behalf by his lawyers on the advice of the medical experts. Because of his medical condition it is oppressive to extradite now, but another application can be made as and when he is fit.

The defence argument has been more nuanced and complicated than that, but I hope I do not do the argument an injustice by summarizing it in this simple way. The argument is persuasive and carries weight. It is a respectable argument, and in the circumstances may be the only approach the defence legal team can properly take. Criticism of that approach (which I have seen in some of the background material in the defence bundles) is unjustified. When the defendant was in a position to make an informed decision about these proceedings, he indicated through counsel that he wished to return to South Africa to contest them. He has consistently asserted his innocence and has, as I understand it, provided detailed statements to that effect to the South African police. The brief accounts he has given to his medical team here confirm that position. On the other hand, he is now ill, has expressed a fear of returning to South Africa and has said that there is a high risk of suicide if that happens.

Against the argument put forward on his behalf the court must consider the strong public interest in honouring our extradition treaties. There is a very strong public interest in this case coming to trial and the facts being determined. I have heard nothing about the victim's family during this hearing, rightly so, although they have sat in quiet dignity throughout the proceedings. The court should not overlook their right to have the factual position determined as soon as is reasonable and fair to do so. The alternatives are in stark opposition to each other. Either Mr Dewani arranged for his new bride to be brutally murdered, or he has been the victim himself of the most terrible tragedy.

I start, as always, with an assumption of innocence. I have complete confidence in the South African judicial system to provide a fair trial. That means he will not be convicted unless the evidence satisfies the court beyond reasonable doubt that he is guilty. Alternatively, if he is unfit to plead I am satisfied that the court will not determine that he committed the acts complained of unless the evidence satisfies the court that he did. In any such hearing the court will take into account the difficulties for the defence in presenting its case. It is sometimes overlooked that it is almost always easier for a court to determine the truth when it hears evidence sooner rather than later. If the prosecution witnesses are lying, that is normally more easily exposed at an earlier than at a later hearing (when inconsistencies can more readily be accepted as the result of fading memories). That is a small, but not negligible, factor.

Inevitably, the court has had to consider speculative arguments. Much is uncertain.

There is undoubted hardship for Mr Dewani if extradited. That hardship is more than ordinary hardship. The chances of an early recovery, or even any recovery, reduce, and the risk of suicide increases. However, when all relevant factors are considered, the hardship falls short of oppression. The public interest in extradition and trial outweighs the competing hardship.

As the issues arising above have been decided adversely to the defendant, I must send this case to the Secretary of State for a decision whether the defendant is to be extradited.

Howard Riddle Senior District Judge (Chief Magistrate)

10 August 2011