Evaldas Urbonas v The Prosecutor General's Office of the Republic of Lithuania

These three appeals concern decisions of the Westminster Magistrates’ Court to order the extradition of the three appellants to Lithuania. The sole ground of appeal now pursued in each of the three cases is that the return of the appellants to Lithuania would involve a breach of Article 3 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) because there is a real risk that the Lithuanian authorities will fail to provide reasonable protection for the appellants from violence or other ill-treatment by other prisoners in the Lithuanian prison system. In particular, the appellants rely on the published findings of the Council of Europe’s Committee for the Prevention of Torture and Inhuman or Degrading Treatment (“the CPT”) published on 23 February 2023 (“the 2023 CPT Report”). That report was published after the decisions of the district judge in the cases of Mr Urbonas and Mr Zemaitis and, although published one day before the judgment in Mr Surpickij’s case was handed down, was not considered by the district judge in that case. Consequently, the courts below did not deal with the 2023 CPT Report in their judgments. The appellants appeal on the basis that, if that report had been before the district judge in each of these cases, the decision of the district judge would have been different and extradition would not have been ordered because there were substantial grounds for believing that the appellants would be at risk of treatment in the Lithuanian prison system which would contravene Article 3 of the Convention. McGowan J. granted permission in all three cases for that ground of appeal.

In brief, the Lithuania authorities seek the extradition of Mr Urbonas for prosecution for an offence of causing bodily harm by intentionally stabbing another person. The details of the alleged offence are set out in a European Arrest Warrant (“EAW”) dated 10 May 2022. The Lithuanian authorities seek the extradition of Mr Zemaitis for two offences. One involved stealing, while acting with a group of accomplices, a property of high value (namely a BMW car with an estimated value of 14,000 euros); the other involved an alleged assault of his partner. The details of the alleged offences are set out in an EAW dated 3 May 2022. The third appellant, Mr Surpickij, has been convicted of offences of robbery, manslaughter and theft. The outstanding total period of imprisonment for those offences that remains to be served is 18 months. The Lithuanian authorities seek his extradition by an EAW issued on 13 September 2022 to serve that term. District judges in each of the three cases ordered the extradition of the appellant.

The appellants are subject to EAWs which are governed by Part III of Title VII of the Trade and Co-operation Agreement between the European Union and the United Kingdom. The European Union (Future Relationships) Act 2020 (“the 2020 Act”) implements that agreement into domestic law. Member States of the European Union are designated as Part 1 territories for the purposes of the Extradition Act 2003 (“the 2003 Act”) by virtue of section 11 of the 2020 Act which inserts those territories into the Extradition Act 2003 (Designation of Part 1 Territories) Order 2003. By virtue of those provisions, Lithuania is a Part 1 territory for the purposes of the 2003 Act and extradition to Lithuania is governed by the provisions of that Part of the 2003 Act. Extradition may not be ordered if that would be incompatible with the Convention rights of the individual concerned (see section 21 and 21A of the 2003 Act).

The material provision of the Convention in the present case is Article 3 which provides that “no one shall be subjected to torture or inhuman or degrading treatment”. The relevant legal principles governing Article 3 are not in dispute and can be stated shortly. Article 3 imposes an obligation on a state not to remove a person to a country where there are substantial grounds for believing that the person would face a real risk of being subjected to ill-treatment contrary to Article 3 in that country. In order to come within Article 3, the ill-treatment must attain a minimum level of severity, which depends upon all the circumstances of the case including the duration of the treatment, its physical and mental effects and, in appropriate cases, the sex, age and health of the victim. In cases such as the present, where the risk of ill-treatment is said to emanate from non-state actors (here other prisoners), such ill-treatment will not constitute a breach of Article 3 unless, in addition, the state has failed to provide reasonable protection against such ill-treatment. Where the requesting state is a signatory to the Convention and a member of the Council of Europe (as is Lithuania, the requesting state in the present case), there is a presumption that that state will comply with its obligations under Article 3 of the Convention. That presumption may be rebutted by clear, cogent and compelling evidence, amounting to something approaching an international consensus, identifying structural or systemic failings. If the benefit of the presumption is lost as a result of such authoritative evidence, the requesting state must show by cogent evidence that there is no real risk of a contravention of Article 3 in relation to the particular requested person in the prisons in which he is likely to serve his sentence. Assurances as to the treatment of individuals may be given by a non-judicial authority and those assurances will then need to be evaluated. See, generally, the decisions of the Supreme Court in R (Bagdanavicius) v Secretary of State for the Home Department [2005] UKHL 38, [2005] 2 AC 668, especially at paragraph 24; Lord Advocate v Dean [2017] UKSC 44, [2017] 1 WLR 2721, especially at paragraphs 25 to 27 and the decision of the Divisional Court in Bazys and Besan v The Vilnius County Court, Republic of Lithuania and another [2022] EWHC 1094 (Admin) at paragraph 13 per Holroyde LJ (with whose judgment Swift J. agreed).

The CPT has made periodic visits to inspect conditions in Lithuanian prisons. It has published periodic reports on those conditions including the issue of inter-prisoner violence and ill-treatment. As this Court has observed, the CPT is a committee operating under the terms of the European Convention for the Prevention of Torture and Inhuman or Degrading Treatment (“the Torture Convention”). It is a highly respected body which exists to prevent the ill-treatment of prisoners through dialogue with states (see Bartulis v Panevezys Regional Court (Lithuania) [2019] EWHC 3504 (Admin) at paragraph 26).

Reports of the CPT were published in 2014, 2018 and 2019. Those reports, and particularly that published in 2019, were considered in detail by this Court in Bartulis. The Divisional Court noted that the CPT reported in 2019 that there had been extraordinary levels of inter-prisoner violence, intimidation and exploitation at three prisons, namely Alytus, Marijampole and Praviiskes. The report noted that the CPT had the impression that the only parts of the prisons under the full control of the administration were the punishment blocks: these were frequently used to accommodate prisoners seeking protection from other prisoners and prisoners were punished for refusing to stay in their ordinary units. As the Divisional Court noted, there were a number of factors identified as contributing to that situation. These included the sub-culture among prisoners, staffing levels and the omnipresence of illegal drugs in prisons. The nature of the prison accommodation also contributed to the problem. Prisoners were generally accommodated in open dormitory style accommodation, with low levels of staff. The more such accommodation was used, the more unofficial hierarchies within prisons were able to operate and maintain authority and influence. See generally, paragraphs 39 to 46 and 115 to 117 of the judgment. The Divisional Court noted that the Lithuanian authorities had adopted an action plan in 2018 and described the contents of that plan at paragraphs 46 to 48 of its judgment. Its conclusion was expressed in the following terms:

“118.

Therefore, we begin by acknowledging that the problem of the "caste system" and of inter-prisoner violence is real, not fanciful. If the authorities had not made a positive response to the 2019 CPT report, then that would have been a strong indicator that there was a proper basis for setting aside the presumption. In our view, those circumstances might be regarded as "exceptional".

119.

For these purposes we regard the successive CPT reports and the 2019 report as being "objective, reliable, specific" and up-to-date as at the summer of 2018. It cannot, of course, be up-to-date today, nor can that body of evidence accommodate the Action Plan and the steps taken in response to CPT 2019. It is for that reason that we have carefully analysed the evidence from Liutkevicius, and the factual evidence from Sakalauskas.

120.

The letters from the Respondent authorities, in fact comprise two elements: an indication of how the Lithuanian authorities have already responded to the problem of inter-prisoner violence and to the CPT report, and assurances for the future.

121.

Piecing together the evidence of response to the problem, while it cannot be said that the steps taken, or in hand, abolish the problem completely, we consider they constitute an adequate response. The allocation of specific funding, the increase in front-line staffing, the existing and planned refurbishments taken together seem to us to demonstrate that a significant effort is being made. On its own, the displacement of ring-leaders and their assistants might not be an adequate response, but it does represent a marked step. It is obviously not possible at this remove to gauge closely the effect of the changes made, wing by wing, prison by prison, amongst the three prisons concerned. We are able to see that prison population density has declined which, combined with the progressive refurbishments and even in the absence of a new prison, must give the prison authorities more flexibility than they have previously had to move prisoners with the intent of displacing those who attempt to exert malevolent authority.

122.

Some of the statistical evidence produced from official figures by Dr Sakalauskas puts bounds on the level of Article 3 risk. We begin by noting that there is ready access to lawyers and the domestic courts. This is not a Member State without a functioning apparatus for the investigation and legal vindication of complaints. We are not, of course, so naïve as to imagine that such access abolishes the fear attendant on complaint. But accepting the imperfections of reporting in any prison system, the collected figures of reported criminal offences tabulated in paragraph 69 above provides some bounds to the risk of Article 3 breaches.

123.

Offences of murder, and crimes which in this jurisdiction would be classified as assaults occasioning grievous or actual bodily harm, do not generally depend on reporting. These are serious crimes but the incidence is low.

124.

It is also relevant to bear in mind the heightened focus on this problem, both domestically and in the context of extradition. The Lithuanian authorities are beyond doubt aware of the eyes of their own press, the domestic courts, the relevant European bodies, and other Member States on this aspect of their prison system. The impact of the assurances offered here, before we consider them as an answer to a system which, it is argued, would be deficient without them, do inevitably mean that the Respondents will be fully aware of the impact if any extradited prisoner were to suffer serious harm.

125.

There is no consensus amongst Member States that the presumption is lost. There is no evidence that another Member State has declined to extradite to these three correction houses. There is no "pilot judgment" from the ECtHR concerning Lithuanian correction houses.

126.

Taking all these factors together, we conclude, after a careful balancing exercise, that the presumption of compliance has not been displaced. Without the Action Plan and the evidence of implementation, real if incomplete, our decision might have been otherwise.

The issue was considered again by this court in Bazys and Besan in the light of further evidence adduced by the appellants in that case. At paragraph 72, Holroyde LJ noted that progress in implementing the action plan “had not been as rapid or as extensive as had been intended; but nonetheless, the picture is one of further real progress, not of decline”. Holyrode LJ (with whose judgment Swift J. agreed) concluded at paragraph 73 that:

“73.

Drawing these threads together, the proposed fresh evidence as a whole certainly raises continuing concerns as to prison conditions in Lithuania. In my view, however, it does not support the submission that this court should reappraise the existing case law.”

The CPT carried out visits between 10 and 20 December 2021 to three prisons in Lithuania, namely Alytus, Marijampole and Pravieniskes to review the measures taken to implement recommendations made by the CPT after previous visits. It published its report on those visits on 23 February 2023.

In the Executive Summary, it said this:

“The Committee must note with grave concern that many of the CPT long -standing recommendations, some of them dating back to the very first periodic visit to Lithuania in 2000, remain unimplemented (or only partially implemented). This concerns recommendations on widespread inter-prisoner violence and the informal prisoner hierarchy, as well as the abundance of illegal drugs combined with a lack of targeted strategies to help the large numbers of drug users within prisons. For these reasons, the CPT has decided, during its 108th plenary meeting in July 2022, to set in motion the procedure provided for in Article 10, paragraph 2, of the Convention.”

The summary dealt with prison establishments and, in relation to inter-prisoner violence and abuse, said this:

“The Committee is equally highly concerned to note that no significant progress has been made in reducing the scale of inter-prisoner violence, which has been repeatedly criticised by the CPT during its previous visits. During the 2021 visit, the CPT delegation once again received many credible allegations of inter-prisoner violence, including beatings and violence of a sexual nature, as well as extortion, psychological pressure, and coercion to commit new crimes.

The roots of inter-prisoner violence appear to have remained the same, namely – the informal prisoner hierarchy (the “caste” system), the omnipresence of illegal drugs, and a lack of adequate custodial staff presence in prisoner accommodation areas (coupled with an infrastructure composed mainly of large-capacity dormitories).

Whilst acknowledging that the Lithuanian authorities have taken some steps and made some efforts to address these deeply rooted problems, based on the findings of the 2021 visit, the Committee concludes that these efforts have, so far, been ineffective and the situation remains unacceptable.

The CPT calls upon the Lithuanian authorities to take resolute action, without further delay, to address the systemic and persistent shortcomings throughout the penitentiary system outlined in this and previous reports of the Committee.”

The main body of the 2023 CPT report addresses this issue. At paragraph 38, the CPT said it was highly concerned to note that no significant progress had been made in reducing the scale of inter-prisoner violence. It noted that it had received many credible allegations of inter-prisoner violence including beatings, violence of a sexual nature, extortion, psychological pressure and coercion to commit new crimes. Such allegations had been received from different units within the three prisons visited but the situation was said to be “particularly dire” in Alytus Prison. The report noted that the root causes of the problem appeared to be the same as those identified on previous visits, namely the informal prisoner hierarchy, the omnipresence of illegal drugs and the lack of adequate custodial presence (and reference was made to the influence of dormitory-type prison accommodation). At paragraph 39, the report says this (footnotes omitted):

“The informal prisoner hierarchy, or the caste system still seems to be the key foundation of prisoners’ life in all three prisons visited, with its traditions dictating internal order and being given priority over official rules. Clearly, those worse affected by this are the relatively numerous lowest cast prisoners, the “untouchables” (nuskiraustieji); the majority of the accounts received of inter-prisoner violence came from them. It was obvious that becoming an “untouchable” significantly increased the risk of victimisation” (emphasis in the original).

Paragraphs 39 and 40 then detail the allegations received by those described as “untouchables”. The 2023 CPT report, and these paragraphs, should be read in full. They include, however, allegations that those at the bottom of the prisoner hierarchy could only be safe (that is, free from risk of physical violence, extortion, or psychological pressure) if they shared a dormitory room with other prisoners at the same position in the informal prisoner hierarchy, that they were not welcome elsewhere, had to obey orders given by prisoners higher up the hierarchy, were given the worst places to sleep within a dormitory and had to do all the cleaning of common areas and the rooms of other prisoners and that they faced the risk of being beaten if they refused. They were only allowed to use certain showers and toilets, were required to eat at separate tables using marked crockery and cutlery and using only certain serving hatches, were only allowed to shop after other prisoners and were sometimes required to use their own money to buy things for prisoners higher in the hierarchy.

At paragraph 41, the 2023 CPT report says this:

“41.

The CPT’s findings during this and previous visits to Lithuanian prisons, clearly illustrate that year after year the caste system continues to exist before the very eyes of prison management and custodial staff, with their clear knowledge and a degree of desperate resignation, even though, when asked, their standard response is that all prisoners are equal.

Although the Committee notes that during the last few years many informal prison leaders have been transferred to other prisons mainly to Vilnius, where they are accommodated in a separate unit, the findings of the 2021 visit show that this measure has been only marginally effective. It has not broken the back of the informal prisoner hierarchy.

In the CPT’s view, the situation of “untouchable” prisoners in Lithuania could be considered to constitute a continuing violation of Article 3 of the European Convention on Human Rights, which prohibits, inter alia all forms of degrading treatment and obliges State authorities to take appropriate measures to prevent such treatment, including that carried out by fellow prisoners” (emphasis in the original).

The 2023 CPT report does not expressly define or identify which groups of prisoners are treated as being at the bottom of the prisoner hierarchy or, in the words of the report, the “untouchables”. Paragraphs 42 and 43 refer to two groups. First there are those who owe other prisoners debts for drugs. Secondly, it implies that those who are seen to co-operate with the authorities (including those who seek treatment for drug addiction) become the lowest group or “untouchables”. There may be other situations in which a prisoner is treated as being within the lowest group but these are not identified in the 2023 CPT Report.

In addition to the presence of drugs and the informal prisoner hierarchy, the report deals with the lack of adequate staff presence inside prisoner accommodation areas. It refers to the fact that supervision during the day was essentially carried out by the heads of unit (one per unit) and a group of staff on a 24 hour shift and observed that this was not enough to ensure order inside the prisons and to ensure safety of more vulnerable prisoners. At paragraph 46, the report noted that dozens of prisoners were refusing to live in their assigned units and were requesting to be isolated. At Alytus and Marijampole prisons, where cell-type accommodation was limited, such prisoners ended up being subject to disciplinary punishment and being placed in disciplinary cellular confinement (referred to as KPT premises) for 30 days. The situation was better in Pravieniskes prison because the vast majority of prisoners who refused to live among the general population were transferred to sector 3 of the prison where the majority of the units were cell-type accommodation. Accordingly, the 2023 CPT report notes that such prisoners in Pravieniskes were not being subjected to disciplinary punishment. At paragraph 48, the 2023 CPT report said this:

“48.

Concerning KTP detention in Alytus and Marijampole Prisons, most of the prisoners therein who were interviewed by the delegation had actually been accommodated in the KTP premises for months, or even years on end and under a very impoverished regime. Throughout these very long periods, they were not allowed to receive any visits, were able to make only one call per month (after finishing a 30-day disciplinary confinement and starting a new one) and were not allowed to buy anything except for clothes, footwear, stationery supplies and hygienic necessities (many complaining of really missing access to fruits and vegetables). As regards their daily routine, this was also highly deprived throughout; they were not offered any activities, were not allowed to work, were not even allowed to have a TV in the cell, and had only one hour of outdoor exercise daily, in a cell-like exercise yard.

Such a situation, almost identical to that found during previous visits leads to the conclusion that, despite repeated recommendations from the CPT, this punitive regime is still how the Lithuanian authorities continue to treat prisoners who ask for protection. The grim picture is this – disciplinary confinement premises are almost exclusively filled by inmates who do not feel safe being accommodated in more regular custodial environments, where they have experienced a deplorably low staff presence and are expected to reside in large dormitories full of drugs, violence, and intimidation. Attempting to escape the violence and temptation to use drugs, they are then forced to languish in an impoverished disciplinary confinement regime for months, not benefitting from any preparation for reintegration, while their fellow inmates, from whom they are trying to escape (higher caste prisoners, drug dealers, etc.) continue to live unpunished in better conditions. The persistence of such an intolerable situation suggests a lack of effective leadership and assertive action on the part of the Lithuanian authorities and could amount to inhuman or degrading treatment” (emphasis in the original).

At paragraphs 49 to 55, the CPT sets out its views on addressing the problem. Again, those paragraphs should be read in full. They include first, custodial staff needed to be in a position to exercise authority. That implied adequate staffing to supervise prisoners, attentiveness to signs of trouble and proper training to intervene effectively. In that regard, it considered that positive relations between staff and prisoners (based on a concept described as dynamic security) was important. Further, initial and continuing training programmes for staff should address the issue of managing inter-prisoner violence. The CPT commented adversely on the use of a 24 hour shift pattern for staff, noted that allegedly due to staff resistance this was still in existence, and asked the Lithuanian authorities to reconsider that. Paragraphs 51 and 52 of the 2023 Report state (footnotes excluded) that:

“51.

Moreover, a holistic approach is urgently needed to tackle the phenomenon of informal prisoner hierarchy, preferably in the form of a targeted strategy including the introduction of a comprehensive risk and needs assessment upon admission, the creation of separate units/sectors for prisoners who do not (or no longer) wish to be involved in the informal prisoner hierarchy, offering adequate (or even superior) conditions and regimes which reward pro-social and cooperative behaviour, allowing separation of informal prison leaders, promoting activities which reject prisoners’ classification into casts, developing opportunities for work and education and further strengthening individual sentence planning; reference is also given to paragraph 32 on the conversion of large-capacity dormitories into cell-type accommodation.

There must also be an improvement in the proper recording, reporting and thorough investigation of suspected cases of inter-prisoner violence and intimidation and, where appropriate, the adoption of suitable sanctions or other measures, as well as the development of effective inter-prisoner violence reduction interventions.

52.

Finally, as described above, the high inflow of drugs and their widespread use in the prisons visited contributes to violence and vulnerabilities within prisons, hinders processes of preparation for reintegration, and must be addressed as a matter of priority. It can only be achieved through a multi-faceted approach, which must include a range of measures already mentioned above namely significant ratios of properly trained staff, improved physical security measures, the use of dynamic security, and the application of a daily regime offering prisoners meaningful activities for at least eight hours a day. Moreover, the functioning of rehabilitation centres should be considerably improved (and new ones opened in prisons where they do not yet exist) with serious consideration given to the creation of drug-free units or sectors in every prison.”

The appellants also seek permission to rely on a report prepared by Karolis Liutkevicius dated 12 September 2023. Although the document is described as an expert report, the appellants do not seek permission to adduce the report on the basis that it is expert evidence. Rather they seek to rely on Mr Liutkevicius as a witness of fact. As his report makes clear, he has little or no current relevant direct experience of conditions in penal institutions and that his first-hand knowledge is limited to meeting with the CPT on its visits in 2012 and 2016. Rather, he has gathered factual information from, principally, publicly available sources. The appellants indicate that they wish to rely on certain particular factual matters. First, Mr Surpickij and (if convicted) Mr Urbonas and Mr Zemaitis could be held in one of the three prisons visited by the CPT. Secondly, in relation to the number of cells, rather than dormitory-type accommodation, the report says that Alytus Prison had 10 cells available, capable of accommodation up to 150 prisoners which was 16% of its 965 maximum capacity or 18% of its average population in 2022; Marijampole Prison had 78 cells available that could hold up to 208 prisoners which was 29% of its maximum capacity of 1050 prisoners or 18% of its average population in 2022; and Pravieniskes Prison had 211 cells that could hold up to 842 prisoners which was 32% of its 2660 maximum capacity or 48% of its average population in 2022. Those figures were based on information provided by the Lithuanian authorities as at March 2022. Thirdly, the appellants rely on figures showing a decline in the number of prison staff. However, as Mr Liutkevicius fairly acknowledges the decrease did not mean a dramatic deterioration as the number of prisoners had also decreased at a rate higher than the decrease in the number of staff. There was in fact a slightly increased staff to prisoner ratio, with the ratio moving from 1 to 3 in 2019 to 1 to 1.27 in early 2023. Fourthly, Mr Liutkevicius refers to court records about convictions involving violence by prisoners from 2021. These show 3 cases involving Alytus, 6 involving Marijampole and 10 involving Pravieniskes. There were also 5 cases involving Vilnius Prison and 1 involving Kaunas.

The Lithuanian Government provided a response dated 23 February 2023 to the 2023 CPT report together with an additional response dated April 2023. The Lithuanian authorities also provided information on the report by Mr Liutkevicius on 13 October 2023 and information dated 28 September 2023 on the risk of violence from prisoners and information dated 30 November 2022 in relation to Mr Surpickij. Other information was also provided but is not relevant to the issues arising in this appeal. The responses and information from the Lithuanian authorities should be read fairly and as a whole. It is important to bear in mind that the CPT visits took place in December 2021. This court needs to consider the position as at the present time.

The following themes emerge from those responses. Firstly, in relation to staff levels, the prison population has declined by 47% over the period 2013 to 2023 while staffing levels had decreased only slightly. The ratio of officers to prisoners had therefore improved from 1 officer per 4 prisoners in 2013 to 1 to 2.7 in 2023. It is right to note that there was a smaller improvement in relation to two of the prisons visited by the CPT: the ratio in Marijampole is 1 to 3.7, in Alytus 1 to 3.4. In Pravieniskes, the ratio in 2023 was 1 to 2.8. Legislative changes due to come into force in April 2023 were expected to reduce the prison population further. Other changes which came into force on 1 January 2023 had led to an increase in the number of prisoners sent to open prisons (the 2023 CPT Report noted that the informal prisoner hierarchy did not appear to exist in open prisons).

The responses indicate that one officer is to have contact with no more than 30 prisoners (although there is some suggestion in the evidence that this process began before 2021 and, if so, had not itself produced a significant improvement by the time of the CPT visit in December 2021). To increase the number of staff, and to raise the prestige of the role, salaries had been increased by 12.8% in 2023. The responses note the criticism of a 24-hour shift but noted that changes had been resisted by trade unions. With that in mind, efforts had been made gradually to phase out the 24-hour shift pattern by introducing a 12-hour shift pattern for new recruits. Two prisons had moved to a 12 or 8 hour shift. In relation to the three prisons visited, the information is that 60% of staff in Marijampole worked a 24-hour shift, 66% in Alytus, and 73% in Pravieniskes, the remainder working a 12 or 8 hour shift.

Secondly, training programmes have been developed (in conjunction with the Norwegian prison system) to improve interaction with prisoners. The responses refer to the training programmes available for staff in 2021-2022 which deal with matters such as recognising and managing conflicts, and establishing and maintaining inter-personal contacts and building relationships with prisoners. The numbers of officers involved in violence prevention training in 2021 and 2022 are given. At Marijampole the figures were 47 in 2021 and 85 in 2022; in Alytus the figures were 96 in 2021 and 125 in 2022; and in Pravieniskes it was 13 in each year. The evidence is that this amounts to 7% of officers in 2021 and 12% in 2022. A new training centre is planned to be opened in 2023.

Thirdly, as of 2023, a director of each prison became responsible for the organisation and implementation of the process of safety and resocialisation of prisons. Research carried out in 2022 on the informal sub-culture within prisons has been provided to all prison governors. Prison governors were obliged to prepare and implement a plan of measures for the gradual elimination of the criminal subculture by choosing different targets (such as improvement in the arrangements for shopping by prisoners, participation of prisoners in common events, and the organisation of canteen services). Specific tasks for governors have also been formulated to enable them to demonstrate in their work and decisions that criminal subculture is not to be tolerated. The effectiveness of those measures will, it is said, be evaluated.

Fourthly, steps are said to be taken in relation to identification and isolation of certain prisoners. There is reference to an initial psychological assessment of new arrivals to identify a tendency to violence, hostility to others and proneness to conflict and to arrangements for meetings to discuss risks posed by prisoners’ behaviour and taking steps to reduce the risk including, for example, transferring these prisoners to other accommodation. It is right to note, as recorded in paragraph 41, that during the years prior to the visit in December 2021, many informal prison leaders had been transferred to other prisons but that measure had been only marginally effective. The responses refer to amendments made in December 2022 to the Code of Enforcement of Sentences which authorise the use of additional preventative measures against the leaders of the informal prison hierarchy (including temporary bans on communication with other prisoners). That legislative change is said to contribute to the identification and more effective isolation of prisoners who have had a negative effect on other prisoners.

Fifthly, in relation to drugs, the April 2023 response indicates the steps that the Lithuanian government is taking to avoid the influx of letters, magazines etc. soaked or sprayed with psychoactive substances. The 2023 CPT had noted that it was reported that more than 50% of drugs entered prison by this route. A pilot system was to be established in one prison to scan all prisoners’ correspondence (subject to specified exceptions) and, after assessment of the results, the implementation of that system could be started in all prisons. Other measures were contemplated to address the importation of illegal drugs by such drugs being thrown or being carried by drone over the perimeter fence although more details are not given. The responses also record the steps being taken to address drug addiction amongst prisoners. Prisoners are starting to be assessed, a number of addiction counsellors have been employed, and, in 2022, day centres for those suffering from addiction were established in all prisons. Long-term rehabilitation centres were operating in five prisons and post-rehabilitation centres were operating in three, with a centre to be established shortly after April 2023 in Alytus and one being planned in Pravieniskes. Work, and funding, with non-governmental agencies is being undertaken. Increased efforts have been made to disrupt organised crime networks supplying psychoactive substances to prisons to prevent supply and prosecuting prison offers who contribute to the supply. The number of criminal cases opened in relation to psychoactive substances had increased by 49% between 2020 and 2023 and the amount of psychoactive substances supplied in prison was believed to have decreased.

Sixthly, the practice of disciplinary confinement has been significantly changed. There have been legislative changes. Further, all prison governors were instructed to revoke the practice of applying disciplinary punishment for those who refuse to live among the general prison population or who did not feel safe. From, it seems about January 2023, in most cases such inmates are being confined in cells without application of disciplinary sanctions (that is, a punitive regime and additional sanctions are not applied to them). The figures provided show a marked decline in persons subject to such discipline. As at April 2023, there were in total 22 inmates in the prison population subject to that regime (as compared with 64 in Alytus Prison alone at the time of the CPT visit in December 2021). It is also relevant to note the further information provided in relation to Mr Surpickij. That refers to Article 70 of the Lithuanian Code of Execution of Sentences which provides that a person serving a sentence may be temporarily isolated from other prisoners when his or her life may be endangered or where it not possible to prevent by other means a serious breach of the law. The further information states that if there was a real threat to the health or life of Mr Surpickij during his sentence, the provisions of that legislation would apply.

Seventhly, the responses give details of the modernisation of the prison estate. The responses recognise that this is a long-term project and full elimination of the criminal subculture from prisons may not be possible without modernisation and full conversion of all prison infrastructure by conversion to cell-accommodation. Hence, the responses say, the need to adopt the measures discussed above to contribute to a reduction of the negative impact of the criminal subculture and the encouragement of a zero tolerance of that phenomenon by prison staff. In terms of the prison infrastructure, the most relevant appear to be the reconstruction of a prison dormitory at Alytus is planned to be completed by June 2024, which will provide cell-type accommodation with 199 places. On the figures before this Court, that would more than double the number of such places available in that prison but that would still only amount to approximately a third of total capacity at the prison (and about 36% of the average population in 2022). There is also reference to the completion of the reconstruction of the disciplinary unit premises at Marijampole into 11 new cell-type accommodation with 44 places. Assuming that occurred, as is the implication, relatively recently, that would still only make a marginal improvement in the position at that prison.

Finally, the responses refer to the fact that the 2023 CPT report had said that the CPT intended to set in motion the procedure set out in Article 10(2) of the Torture Convention at its plenary session. Article 10(1) provides for the CPT to draw up a report on the facts found during a visit. Article 10(2) provides that if a state “fails to co-operate or refuses to improve the situation in the light of the [ CPT’s] recommendations”, the CPT may decide to make a public statement on the matter. The responses emphasise that, certainly as at October 2023, the CPT had not made a public statement (and we were told during the hearing in November 2023 that no such statement had been made). The responses emphasise that Lithuania is seeking to maintain a close co-operation with the CPT and it is engaging in discussion with the CPT. As that correspondence is confidential, the responses say that the Lithuanian authorities cannot release the correspondence, but sets out in its views and provides further information and updates on improvements made.

The respondent seeks permission to adduce the evidence of a senior Crown Prosecutor, Mr Barrowcliffe. Part of that evidence shows the number of persons extradited to Lithuania during the period from 1 June 2019 to October 2023. The purpose is to show that there is as yet no international consensus among states, including states who are a party to the Convention and the Torture Convention, that there are systematic failings in the Lithuania such as would render returns a breach of Article 3 of the Convention. In brief, there were 146 returns to Lithuania in 2019, 130 in 2020, 124 in 2021, and 106 in the first 10 months of 2023. States returning persons to Lithuania over the period 1 June 2019 to the end of October 2023 (just over four years) include Denmark (10), Germany (105), the Netherlands (32), Spain (34), and France (15). In the first 10 months of 2023, the figures for these states were respectively 4, 22, 5, 13 and 4.

Mr Hall KC for the appellants submitted that the 2023 CPT report was evidence that inter-prisoner violence remained out of control and was systemic. He submitted that the report showed that staff were aware of and resigned to the informal prison hierarchy system that was the root cause of the problem but sought to deny that the problem existed. He submitted that that demonstrated that the Lithuanian authorities’ minimisation of the problem demonstrated an unwillingness, and inability, to provide reasonable levels of protection.

Ms Hinton for the respondent accepted that the 2023 CPT report, following the visits in December 2021, highlighted inter-prisoner violence as a continuing concern. Ms Hinton highlighted a series of measures (most of which are described in the Lithuanian authorities’ responses summarised above). She submitted that the detailed evidence shows that there is reasonable protection against violence and there is a concerted drive underway to deal with the violence and reduce it, thereby meeting the concerns identified by the CPT. Ms Hinton relied upon the absence of any pilot judgment from the European Court of Human Rights establishing that there was an international consensus that the presumption that Lithuania would not comply with its obligations under Article 3 of the Convention. She further relied upon the fact that other states who were members of the Council of Europe and parties to the Convention, and the Torture Convention, continued to extradite persons to Lithuania as evidence of the absence of any such international consensus.

In the light of the earlier case law, and the detailed review of the evidence above, we can state our conclusions relatively shortly. We consider first the issue of whether the authorities are able and willing to ensure reasonable protection against the risk of ill-treatment from other prisoners. We start first by considering if the presumption that Lithuania will comply with its obligations under Article 3 of the Convention has been rebutted in respect of those prisons where convicted prisoners may be held during their sentence (in particular, Alytus, Marijampole and Pravieniskes Prisons). We accept, as this Court has in other cases, that the successive reports of the CPT, including the 2023 CPT report, are “objective, reliable, specific” evidence which is up to date as at the time of the visits in December 2021 (see the similar approach taken in Bartulis at paragraph 118 of the judgment). We give those reports very considerable significance and weight and regard them as of particular importance given the status and role of the CPT and its expertise and experience (factors essentially similar to those regarded by the Supreme Court in assessing the evidence of the United Nations Commissioner for Human Rights in R (AAA) v Secretary of State for the Home Department [2023] UKSC 42, [2023] 1 WLR 4433, especially at paragraphs 64 to 70).

We accept that there is a real, long-standing and continuing problem of inter-prisoner violence within some of the prisons in Lithuania. This problem affects all units within all three prisons visited by the CPT in 2021 but is particularly acute in the case of Alytus prison. The problem manifests itself in ill-treatment of prisoners particularly those who are at the bottom of the informal prisoner hierarchy (described by the CPT as “untouchables”). The ill-treatment has manifested itself in inter-prisoner violence, and also extortion, pressure and coercion of prisoners and other forms of ill-treatment as described in paragraphs 38 to 41 of the 2023 CPT report summarised above. The three principal factors contributing to this are the existence of the informal prisoner hierarchy, lack of adequate staff supervision and control, and the presence of illegal drugs. Further, we accept that there the measures taken (and largely foreshadowed in the 2018 action plan) had not been effective in eliminating the problem at the time of the latest CPT visit in December 2021.

First, however, we do not see evidence that the problem has deteriorated since the decisions of this Court in Bartulis and Bazys and Besan, although it is right to acknowledge that the problem has persisted for a further period of time and that is a factor to consider. Secondly, the evidence is that the Lithuanian authorities recognise the existence of the problem and recognise the need to adopt measures to address the risk of inter-prisoner violence and abuse and, importantly and significantly, to ensure protection for vulnerable prisoners.

In that regard, the measures adopted prior to December 2021 and those adopted since are intended to bear down on the problem of inter-prisoner violence. We have summarised them above. All those should be considered together. We note the following. First, there have been additional management changes since the CPT visit in December 2021. Since 2023 each prison is required to have a director with responsibility for the safety and resocialisation of prisoners. Further, prison governors are now tasked to plan and implement measures to counter-act ill-treatment by prisoners. Secondly, there have been further reductions in the prison population and an increase in the number of prisoners transferred to open prisons. Legislative changes indicate that that should continue. That should assist in maintaining the improved staff-prisoner ratios (and possibly improving ratios further). In addition, measures have been taken in relation to recruiting staff (by increasing salaries), improving professionalism (by increased training) and attempting to move away from a 24 hour shift system although it is right to note that those changes have not yet affected all staff. Thirdly, there are the attempts that have been begun to reduce the introduction of psychoactive substances into prisons, and to promote the rehabilitation of those suffering from drug addiction. Fourthly, along with earlier attempts to transfer prisoners who are violent to other prisons, legislative changes adopted in December 2022 enable additional measures to be taken against the leaders of the informal hierarchy. Those measures, along with the range of other measures described, do appear to us to represent a real recognition of the problem and a continued, and increasing, willingness to address it.

Furthermore, and very significantly, the Lithuanian authorities have specifically addressed the problem of protection of those prisoners who do not wish to remain in the general prison population because they are vulnerable or otherwise fear for their safety. This was one of the specific issues identified in paragraphs 46 to 48 of the 2023 CPT Report which criticised the placing of prisoners who ask for protection in disciplinary confinement (the KPT premises) and subjecting them to disciplinary measures. The evidence is that legislative changes enable prisoners whose life or health is at risk to be separated from other prisoners and that instructions have been given to all prison governors to revoke the practice of application of disciplinary penalties for those who do not feel safe living in the general prison population. As a result, there had been a very significant and marked decline by April 2023 in the number of persons subject to the application of such sanctions. That, we consider, is a strong indication of the practical and concrete steps taken by the Lithuanian authorities to offer reasonable protection to prisoners whilst steps are taken to address the underlying factors giving rise to the problem of inter-prisoner violence and abuse.

We do not accept Mr Hall’s submission that the Lithuanian authorities are minimising the problems and that this demonstrates an unwillingness, and an inability, to provide reasonable levels of protection. We regard the responses as evidencing a clear acceptance of the problem, a continued attempt to address the problem by a variety of measures and the offering of practical protection to those prisoners who fear violence or abuse if they remain within the general prison population. Principally for those reasons, we do not consider that the evidence yet demonstrates that there is clear, compelling evidence amounting to an international consensus that there are structural or systemic failings such that the presumption that Lithuania will not afford reasonable protection to prisoners against the risk of inter-prisoner violence and ill-treatment as would constitute a violation of Article 3 of the Convention.

We have reached our conclusion by considering the reports of the CPT, particularly the 2023 CPT report, and the responses made by the Lithuanian authorities to that report. Other matters were relied upon by the respondent. We regard these matters as, at most, confirming the conclusion that we have reached, having regard to the CPT reports and the Lithuanian responses, rather than as separate reasons for concluding that the presumption has not been rebutted. In particular, we note there is no pilot judgment issued by the European Court of Human Rights in relation to prisons in Lithuania where convicted prisoners may serve their sentences. The existence of such a judgment might well have indicated the existence of an international consensus that there were general failings in the Lithuanian system. We do not regard the absence of such a judgment as conclusive that there are no such failings. At most, the absence of such a judgment would, on the facts of this case, have reinforced the conclusion that we have reached based on an analysis of the successive CPT reports, including in particular the most recent 2023 CPT report, and the evidence of the response of the Lithuanian authorities to that report. We also note, de bene esse, the evidence that parties to the Convention and the Torture Convention continue to extradite persons to Lithuania. If that evidence is admitted, that may give some indication that parties to those conventions do not regard Lithuania as failing to meet its obligations but we would not regard that matter as being of any great significance. We accept that, on the evidence before us, the CPT has not made a public statement under Article 10(2) of the Torture Convention, which it could have done if it considered that Lithuania had failed to co-operate or was refusing to improve the situation. That may indicate that the CPT is satisfied that Lithuania is co-operating and is seeking to improve the situation in the light of the CPT reports. We would, however, have reached that conclusion in any event in the light of the responses of the Lithuanian authorities and we do not regard the absence of a public statement as carrying much, if any, significance.

In the circumstances, therefore, we do not accept that the presumption that Lithuania will comply with its obligations under Article 3 of the Convention to provide prisoners serving sentences of imprisonment with reasonable protection against the risk of ill-treatment from other prisoners has been rebutted.

In those circumstances, it is not necessary to consider whether, in the case of each of the individual appellants, the Lithuanian authorities could demonstrate that there were no substantial grounds for believing that there was a real risk of a breach of Article 3 in each of their cases. In that regard, we note that there is before us no evidence as to whether Mr Urbonas or Mr Surpickij do fall within the group of prisoners who are at the bottom of the informal prisoner hierarchy (or “untouchables” to use the language of the 2023 CPT report). There is nothing to suggest that they are drug users likely to fall into debt with other prisoners or would be perceived as having co-operated with the authorities (the two groups specifically referred to by the CPT as being vulnerable to the risk of ill-treatment). These matters were not considered by the district judges below in these two cases as the 2023 CPT report was either not available or not provided to those courts and the issue of inter-prisoner violence was not raised before them. If we had accepted that the presumption of compliance with the obligations in Article 3 had been rebutted (which we did not), then we would have been minded to remit the matter to the district judge in each case so that this issue could have been considered.

There is evidence from the second appellant, Mr Zemaitis, that he was bullied and physically abused in 2014 when he was serving a sentence for an earlier offence because he did not belong to what he described as the higher caste. The district judge did not accept that there was a risk of a breach of Article 3 in part because of the 2018 action plan and because Mr Zemaitis had left prison in 2018 and there was no obvious reason why he should be singled out for violence at the hands of people he does not know. Those findings were made, of course, in the absence of knowledge of the 2023 CPT report. If we had found the presumption rebutted (which we did not), we would have remitted the matter to the district judge to consider this issue in the light of the evidence of the 2023 CPT report and other relevant evidence. In the event, this issue does not arise for decision in the case of any of the three appellants as we do not consider that the presumption is rebutted for the reasons given above. Nor in these circumstances is it necessary to consider the question of seeking assurances in relation to the individual appellants if they were returned.

There are certain ancillary matters which arise. First, there is an application to admit the evidence of Mr Liutkevicius contained in a document dated 12 September 2023 and entitled expert report. Initially, it seems, an application was made by Mr Urbonas to extend the representation order to cover the costs of obtaining expert evidence on a number of issues concerning prison conditions in Lithuania. That application was granted and the representation order extended to cover the cost of the preparation of a report by Mr Liutkevicius in the amount of £4,500. We understand that Mr Liutkevicius has prepared reports in a number of cases involving prison conditions in Lithuania. In the event, the appellants no longer seek to rely upon the evidence of Mr Liutkevicius as expert evidence. It will be for other courts to consider whether or not similar applications should be granted in future cases although we doubt that preparation of a report of the type produced by Mr Liutkevicius in the present case would ordinarily justify the extension of a representation order.

We turn next to the application to admit the evidence of Mr Liutkevicius on the basis that it contains factual material as identified at paragraphs 35 and 36 of the appellants’ skeleton argument. First, the appellants seek to rely on paragraph 23 of the report as evidencing that Mr Surpickij and, if convicted, Mr Urbonas and Mr Zemaitis could serve their prison sentence at prisons which include Alytus, Marijampole and Pravieniskes. We do not understand that to be in dispute and it is not necessary to admit the report for the purpose of establishing those facts. Secondly, the appellants seek to rely on the figures in paragraph 62 to 63 of the report detailing the amount of cell-type accommodation established by March 2022. The number of additional cell-type units created by that date and subsequently, appear elsewhere in the evidence before the court. The respondents do not submit that any increase in the amount of cell-type accommodation means that the concerns of the CPT have been addressed. Indeed, the response from the Lithuanian government recognises that full elimination of the criminal subculture will not be possible until the accommodation of all prisoners in cells is possible and that is a longstanding task which has not yet been completed. The responses deal with other measures taken to address the concerns of the CPT. Our decision is not based on the increases in the amount of cell-type accommodation available. Whilst we have read paragraphs 62 and 63 of the report and considered them, we do not consider that the admission of that evidence is justified having regard to the principles set out by this court in its decision in The Szombathely City Court and others v Fenyvesi [2009] EWHC 231 (Admin). Thirdly, the appellants seek to rely on the facts relating to staffing levels. That material is in fact available in other parts of the evidence before the court and it is not necessary to admit the report to establish those facts. Finally, the appellants seek to rely on the evidence at paragraph 73 to 75 of the report summarising the reported cases involving inter-prisoner violence. Again, it is clear that there have been incidents of inter-prisoner violence and abuse from the 2023 CPT report. While we have considered the limited material referred to by Mr Liutkevicius, we do not consider that it adds materially to the information before the court. Accordingly, we refuse permission to adduce the evidence contained in the report dated 12 September 2023.

Similarly, we refuse permission to adduce the evidence of Mr Barrowcliffe on the number of extraditions from other states to Lithuania. Whilst we recognise that information may well have been admissible if it had been adduced during the extradition proceedings, we do not consider that it materially alters the position for the reasons given above and we would not admit it in these appeals.

The appellants in their skeleton argument submitted that the respondent failed to comply with its duty of candour by not referring to the 2023 CPT report at the extradition hearings before the district judge. That report had been transmitted to the Lithuanian authorities in July 2022. The hearings in these cases took place after that date. The Lithuanian prison service has responded to those allegations by letter dated 13 November 2023 indicating that it had always responded to any requests for information but it did not hold data as to the dates of extradition hearings.

The position appears to be as follows. The visits by the CPT were undertaken in December 2021. It appears that the 2023 CPT report was provided to the Lithuanian authorities in about July 2022. Article 11(1) of the Torture Convention provides that the information gathered by the CPT in relation to a visit, its report and consultations with the state concerned are confidential. Article 11(2) provides that the CPT shall publish its report, together with any comments of the state concerned, whenever requested to do so by that state. The 2023 CPT report was published on 23 February 2023.

The extradition hearing in Mr Urbonas’ case was held on 11 November 2022 and judgment given on 28 November 2022. The final extradition hearing in Mr Zemaitis’ case was held on 10 November 2022 and judgment given on 15 December 2022. The extradition hearing in Mr Surplickij’s case took place on 30 January 2023 (i.e. before the 2023 CPT was published) and judgment given on 24 February 2023 (i.e. the day after it was published but without the report having been considered by the district judge).

The position in relation to the duty of candour is set out at paragraphs 132 to 135 of the decision of this court in Bartulis in the following terms.

“132.

That approach was endorsed and re-stated by the Divisional Court in R (Gambrah) v CPS [2013] EWHC 4126 (Admin) as follows:

"8.

The duties of the requesting state and of the CPS can, in my view, be summarised as follows: the duty of the requesting state includes, pursuant to its duty of candour and good faith, the obligation to disclose evidence which destroys or very seriously undermines the evidence on which it relies. The CPS has independently a similar duty. It also has a duty to ensure that the requesting state fulfills its duty. Finally, it has a duty to withdraw from the proceedings if it finds itself put in the position where its duty to the court conflicts with its duties to the requesting state. That is, I believe, a full and accurate statement of the law as expounded in Raissi."

133.

We accept the broad points made by the Respondents as to the nature of the CPT system of inspection and response. We do not conclude that a Member State has an obligation to disclose a CPT report, or the state's response, in advance of the point when it would otherwise become available. To impose such an obligation would be likely to frustrate the CPT process. However, the duty of candour must also mean that evidence or assertions should not be advanced which are inconsistent with the factual position known to the requesting state. That basic component of the duty of candour must arise in relation, for example, to concerns raised by a CPT inspection, not yet published as a report, which are either accepted or cannot be contradicted by the requesting state. As often in such matters, there will frequently be room for argument as to what can and cannot properly be said. But in our view the principle is clear: a requesting state cannot in candour advance a position which the representatives of the state know to be false or misleading, on the basis of a CPT inspection or as yet unpublished report, or otherwise.

134.

We do not intend to engage in a detailed examination of what was said to be misleading by the Appellants. We have reached no conclusion that the Lithuanian authorities set out to mislead, and we are not convinced there is the basis for such a conclusion. There is no basis for saying there was any deliberate or undue delay in the publication of the CPT 2019 report. It is of significance that the parties and the District Judge were fully aware of the 2018 inspection, of the earlier reports, and of the issues which arose from them.

135.

We intend to make no more general observations on disclosure in cases such as this, save to emphasise that it is the obligation of the Crown Prosecution Service when assisting a requesting state to ensure that state is alerted to their duty of candour, including the matters we have spelled out above.”

We have not had full submissions on the question of the duty of candour. We do not consider it necessary to make any rulings on the issue. What follows is our provisional view on some of the issues that might arise. First, the obligation is on the Crown Prosecution Service assisting a state requesting the extradition of an individual to ensure that the duty of candour is complied with. Secondly, that will involve the CPS alerting the requesting state to their duty of candour, and of matters relevant to that duty, and may require the CPS to seek specific information from the requesting state. Whether it needs to do so will depend on all the circumstances including the issues that are raised or which might reasonably arise in relation to extradition proceedings. In the present case, we note that the issue of inter-prisoner violence had been the subject of successive reports by the CPT between 2012 and 2018, and had been the subject of analysis by the Divisional Court on two occasions, the last raising the issue of whether the position had altered since the previous case. Furthermore, Mr Zemaitis, and it seems Mr Surplickij, did raise the issue of possible ill-treatment by other prisoners if returned. In those circumstances, our provisional view is that there was sufficient reason for the CPS to inquire of the Lithuanian authorities whether there had been further visits or reports by the CPT. In the light of the information, whilst the reports could not have been disclosed at the extradition hearings in these three cases, our provisional view is that consideration should have been given to informing the district judge in each case of the possibility of further information becoming available so that the district judge could determine how to proceed including, by way of example, whether to adjourn pending publication. We are told that the problem may now be resolved in that Lithuania has agreed to a procedure whereby reports of the CPT are published without the need for a request from Lithuania and this should enable the publication of future CPT reports without delay. Whilst we have been provided with a letter from the Lithuanian Prison Service confirming that Lithuania has opted into the automatic publication procedure, we do not have sufficient information to know how the procedure will operate in practice. In particular, we do not know the timetables involved and whether there is a period between the CPT providing the report to the state concerned and publication of that report. If that were the case, issues similar to those which arose in that case may still arise in the period between finalisation of the report and publication. There may be other issues of which we are unaware.

For those reasons, these three appeals are dismissed.