The role of Israeli judges in authorising solitary confinement placements: Balancing human rights and risk,or neutralising responsibility?

Abstract

This paper explores the role of judges in authorising the extension of placements in solitary confinement in Israeli prisons for lengthy periods of time. It qualitatively examines, through content analysis of 354 Israeli court decisions, how judges negotiate and rationalise the harmful effects of solitary confinement when balanced against the prison authorities’ reasoning for subjecting prisoners to it. Finding an overall tendency to defer to the expertise of prison authorities, we examine what Sykes & Matza termed ‘techniques of neutralisation’ used by judges to distance themselves from the responsibility for solitary confinement placements and the hardship they inflict. The paper further discusses the socio-legal and organisational structures and contexts which incentivise the prioritisation of prison security/discipline over the protection of prisoners from the ‘pains of solitary confinement’.

Introduction

Solitary confinement, under different names, manifestations and forms, is a common practice in many penal systems (Lobel and Scharff-Smith, 2020; Shalev, 2008). As an extreme penal practice, arguably second in severity only to capital punishment, solitary confinement stands at the heart of the intersection between human rights and penology, explored by a growing body of research (Haney, 2018; Lobel and Scharff-Smith, 2020; Reiter, 2016; Shalev, 2013), and in personal accounts of those who survived solitary confinement, mostly from the United States (Casella et al., 2017; Woodfox, 2019). Serious efforts had been made to describe, theorise and analyse solitary confinement decision-making (Reiter, 2016; Schlanger, 2020; Shalev, 2013) as well as the ‘pains of solitary confinement’ (Coppola, 2019; Haney, 2018; Shalev, 2008) and its meaning for prisoners’ dignity and human rights (Polizzi, 2017; Schlanger, 2020); prison discipline (Logan et al., 2017); and post-release re-entry (Lovell et al., 2007).

Several studies, from different perspectives, methodologies and disciplines, have investigated prison officials’ decision-making generally (Calavita and Jenness, 2015; Feeley and Rubin, 2000; Feeley and Swearingen, 2004), and when imposing solitary confinement specifically (Logan et al., 2017; Schlanger, 2020; Shalev, 2013).These studies point to the broad and ‘powerful’ administrative discretion within prison where ‘no one was watching’ (Reiter, 2016: 3).

But what happens when the decision maker about solitary confinement placements is a judge, who is external to the prison? What are some of the forces, tensions and discourses in their decision making? How do judges balance the competing demands of human rights and management risk? And more specifically, how do they balance the ‘pains of solitary of confinement’ against the prison authorities’ request to extend stays in solitary confinement?

The paper aims to shift the focus from prison to judicial discretion when deciding the ‘law’s violence’ (Calavita and Jenness, 2015) of solitary confinement placements. It aims to better understand how judges – acting as primary decision-makers – interpret and balance the logics of human rights and control/risk management (Garland, 2001; Shalev, 2013), as well as the organisational forces at play when when deciding on solitary confinement placements (Feeley and Rubin, 2000; Feeley and Swearingen, 2004).

Furthermore, research in criminology (Logan et al., 2017; Shalev, 2013), law (Lobel, 2008; Schlanger, 2020), and ‘punishment and society’ (Franco et al., 2020; Reiter, 2016), problematises the application of administrative power to normalise the exceptional through routinised/bureaucratised solitary confinement decision-making (Reiter, 2016; Rhodes, 2009; Shalev, 2013). Prison officials’ discretion had been criticised as largely hidden, generic, arbitrary, racially-biased, and imposed through perfunctory discretionary hearings that lack ‘external oversight’ (Coppola, 2019; Franco et al., 2020; Reiter et al., 2020: 56; Schlanger, 2020; Shalev, 2013). Scholars also highlight the ongoing U.S. constitutional courts’ reluctance to ‘second-guess’ prison officials’ discretion, even where solitary confinement posed a risk of very significant harm to the prisoner (Coppola, 2019; Haney and Lynch, 1997; Schlanger, 2020). In this context, it is worth noting the crucial role that strategic litigation has nonetheless played in curtailing some of the worst excesses of the use of solitary confinement, for example ending the use of indeterminate solitary confinement in California (e.g. Ashker v. Governor of California, No. 4:09-cv-05796-CW (N.D. California, 2012) and in Canada (British Columbia Civil Liberties Association v. Canada, 2018 BCSC 62 (17 January 2018).

In an effort to address concerns about solitary confinement and in recognition of the ‘pains of solitary confinement’ (Haney, 2018; Shalev, 2013; World Health Organisation, 2014), a number of domestic and international policy and legal instruments proposed to control the administrative authorisation of solitary confinement through judicial review. The UN Standard Minimum Rules for the Treatment of Prisoners (2015) (‘Mandela Rules’) require solitary confinement placements to be reviewed by an independent body. The European Court of Human Rights held that ‘it is essential that the prisoner should be able to have an independent judicial authority review the merits of and reasons for a prolonged measure of solitary confinement’ (Ramirez v. France (2006), par. 145). The Inter-American Court of Human Rights similarly demands strict judicial review for imposing solitary confinement and of its duration (Lobel, 2008). Other domestic systems consider or have adopted judicial review for certain types of prisoners (e.g. remand detainees), such as Norway (Shaheen, 2015) or suggested implementing ‘an independent external review process’ for administrative segregation placement beyond certain periods, such as Canada (Casavant and Charron-Tousignant, 2019: 2).

The paper aims to explore how courts, serving as primary decision makers, decide on the extension of solitary confinement placements. It does so through a qualitative analysis of 354 solitary confinement court decisions, authorising the continuation of solitary confinement placements for lengthy periods (longer than 6 months). Using the lens of ‘techniques of neutralisation’ theory (Sykes and Matza, 1957), the paper explores decisions made by Israeli judges on the continuation or caseation of prolonged solitary confinement placements. It asks how they negotiate and rationalise the ‘pains of solitary confinement’. As solitary confinement policies are implemented differently across different systems, the paper also strives to expand the scholarship beyond its U.S.-British contexts (Franco et al., 2020).

The paper is structured as follows. The first part presents key concepts and typologies, including the theoretical framework and legal context of the study. The second part presents the methodology, data and analytic approach used. The third part, the heart of the paper, explores how judges neutralise the ‘pains of solitary confinement’ when authorising prolonged solitary confinement placements. In the final part we discuss the findings and their relevance for theory and policy.

Key concepts and typologies

Techniques of neutralisation

The techniques of neutralisation theory (Sykes and Matza, 1957) seeks to explain how people who commit offences have nonetheless internalised social and legal rules and maintain at least a minimal commitment to the dominant social order. With time, this theory has also been used beyond the limits of criminal behaviour, for example to examine how sentencing judges decide to imprison in borderline cases (Tombs and Jagger, 2006), and how prison officers rationalise the suffering of prisoners in their care (Scott, 2008; Shalev, 2013). The techniques of neutralisation theory will serve as a theoretical framework for exploring the judicial decision-making of solitary solitary confinement placements.

According to Sykes and Matza (1957), to blunt the moral force of the law and rationalise their criminal conduct and associated guilt, delinquents use a variety of ‘techniques of neutralisation’, which Sykes and Matza grouped into five major categories: First, delinquents can rid themselves of a negative self-image through the denial of responsibility by claiming that their behaviours are accidental or due to forces beyond their control. Second, is the denial of injury. Here, the wrongfulness of one’s behaviour is determined by whether anyone was hurt and whether the actor intended to do any harm, which is often a matter of interpretation. Third, sometimes when delinquents admit that their actions cause harm, they can neutralise moral indignation by denial of the victim either by (i) justifying the injuries they cause as a form of rightful punishment imposed for the victims as wrongdoers, or; (ii) claiming that no victim exists, and as such no one was hurt. Fourth, is the condemnation of the condemner. Instead of focusing on their own actions, delinquents focus on the motivations or behaviours of the people who disapprove of them by claiming that their condemners are ‘deviants in disguise’ (Sykes and Matza, 1957: 668). The final technique is the appeal to higher loyalties. Delinquents shield themselves from internal and external controls by claiming that their behaviour is consistent with the moral obligations of their group. Another important neutralisation technique is the claim of necessity: delinquents do not experience guilt about their behaviour if they perceive their acts as necessary.

We also draw on Stanley Cohen’s (2001), ‘states of denial’ framework for understanding the manner in which human suffering can be either denied or acknowledged. Cohen (2001: 7) identifies three main ways in which one can deny the very real human suffering that occurs. Literal denial is ‘the assertion that something did not happen or is not true’. Interpretive denial in where the facts themselves are not denied but are given a different interpretation (e.g., euphemisms), or ‘spin’, thus altering their meaning (Cohen, 2001: 8). In the case of implicatory denial, what is minimised is not information but ‘the psychological, political or moral implications that conventionally follow’. These forms of denial allow the individual to avoid fully acknowledging the reality of human suffering and thus to feel no compulsion to act to mitigate this suffering.

The pains of solitary confinement

The concept of ‘pains of imprisonment’, famously coined by Sykes (1958), captures the key deprivations inherent in ‘total institutions’, including the deprivation of liberty, autonomy, security, goods and services, and lack of heterosexual relationships. Other scholars have added the pains of separation between staff and prisoners, humiliation, loss of identity, regrets about the past, concerns about the future, boredom, and the experience of highly intrusive and emotionally demanding psychological assessments (Crewe, 2011).

People held in solitary confinement have to endure a particularly harsh form of the ‘pains of imprisonment’ experienced by all prisoners, as well as the particularly painful deprivation of human contact and social relationships (Shalev, 2008). Human beings are social creatures who require human contact for their very basic sense of self, hard-wired for social interaction (Lieberman, 2013). Solitary confinement therefore afflicts the individual in two ways: it imposes additional psychological strain, and, by isolating the individual from the main prison society, it takes away coping resources ordinarily available to people in prisons (Haney, 2018; Shalev, 2013).

The harms of solitary confinement are associated with wide-ranging health problems, both psychological and physiological (Coppola, 2019; Haney, 2018; Shalev, 2008; World Health Organisation, 2014). Commonly reported problems include: anxiety, panic attacks, depression, hopelessness, anger, poor impulse control, cognitive disturbances, perceptual distortions, paranoia, psychosis, and a significantly increased risk of self-harm and suicide (Haney, 2018; World Health Organisation, 2014). Physiological problems include gastro-intestinal and genito-urinary problems, insomnia, deterioration of eyesight, weakness, profound fatigue, migraine headaches, joint pains, and an aggravation of pre-existing medical issues (World Health Organisation, 2014). Neuroscientific research confirms that the mental pain caused by lack of social interaction and limited environmental stimulation, which are typical of solitary confinement environments (Shalev, 2008), can alter brain function and structure (Coppola, 2019). Some argue that solitary confinement is no more damaging than ordinary prison conditions (Morgan et al., 2016, but see Haney, 2018 for critical analysis), but the pains of solitary confinement are widely documented and acknowledged.

The extremity of solitary confinement and its potential harms are reflected in the prohibition, under international human rights law, of prolonged solitary confinement, defined as one lasting longer than 15 days. Likewise prohibited is its use with people who are particularly vulnerable to its harms, including the mentally unwell, pregnant women, breast-feeding mothers, children and young people (Mandela Rules, 2015).

The use of solitary confinement in Israel: The legal framework

In 2020, the total prison population in Israel was estimated at 13,700 people (of whom 9,300 were ‘criminal’ and 4,400 were ‘security’ (terror-related) prisoners) (IPS, 2021). official data on the use of solitary confinement is not currently published.

Israeli administrative courts play an unusual role in solitary confinement placements: they need to authorise the continuation of such placements beyond an initial six months period ordered by the prison authorities. Article 19b (‘Separate Confinement’) of the Prisons Ordinance (New Version), 1971, was amended in 2000. This amendment was recommended by a Special Committee set up jointly by the Government and the Prison Service with the aim of limiting the use of solitary confinement. Article 19b allows for the separation of a prisoner – in the absence of alternative measures that could achieve the same goal– for one of the following purposes: (1) State security; (2) Prison security; (3) Protecting the well-being or health of the prisoner or of other prisoners: (4) Preventing substantial damage to the discipline and orderly routine of the prison; (5) Preventing a crime of violence (and several serious offences under anti-crime organisation or drug laws) (6) When a minor’s best interests require that they be held in a cell of their own. Solitary confinement can also be used as short-term punishment for a prison disciplinary offence, which we do not discuss here.

Israeli prison authorities (IPS) may order, or a prisoner may request, to be placed in solitary confinement (on one or more of the grounds listed above) for periods of up to 6 months (or up to a year in where two prisoners are sharing one separation cell, or ‘double bunking’). Each time period in solitary confinement (24 h; 48 h; 14 days; 30 days) requires the approval of an increasingly senior prison official, and placements for longer than 6 months (1 year if double-bunking) require a court order. Extensions of up to six months in single separation, and up to a year if the prisoner shares a cell with another prisoner, may be ordered by the court, with the period extended from time to time, with no maximum duration set. Prisoners have the right to ask for a review of the court’s decision if new facts have come to light or if the circumstances have changed. It is important to note that, in this capacity, the courts exercise primary judicial authority, thus not simply reviewing prison authorities’ administrative discretion (Musli v. State (8.11.11)).

Data on the exact extent of use of solitary confinement in Israeli prisons is difficult to obtain. Official and systematic data regarding solitary confinement placements in Israel is not regularly published (Ministry of Justice, 2012: 7).The latest indication of such placements was obtained by Physicians for Human Rights Israel (PHRI) who reported that, in July 2015, 117 prisoners were held in solitary confinement (PHRI, 2016). Over half (54%, or 63 prisoners) of these prisoners had been isolated for longer than 6 months, meaning that their placement would have had to be authorised by a court (PHRI). The total prison population that year was 17,963 (IPS, 2018).

Methodology

This study examines the degree to which courts acknowledge or deny the hardships (or pains) of solitary confinement, and how they balance these against the prison authorities’ request to extend stays in solitary confinement. It analyses 354 decisions taken by District courts in cases involving requests to authorise solitary confinement placements beyond the initial 6 months (or a year in cases of ‘double bunking’) imposed by prison authorities. The cases were collected from two major Israeli legal databases: Nevo and Takdin. Key search words were: ‘solitary confinement’, ‘isolation’, ‘solitary confinement prisoner’; ‘solitary confinement conditions’ and ‘s. 19 of the Prison law’. Cases cover the period between 13.8.2000 (when this provision was enacted) and 30.12.2019.

Once the data was collected, content analysis of the data included several stages. First, after an initial reading of the cases, we defined the initial coding, based on the legal criteria for placements in law. On finding that the courts approved the continuation of solitary confinement in 93.4% (331) of the cases analysed, we sought to better understand the reasoning provided by the courts to support their decision. We found Sykes and Matza’s (1957) and Cohen’s (2001) analytical frameworks helpful for identifying key themes in the arguments put forward by the courts in their decision to approve or deny requests to extend solitary confinement stays.

We then deconstructed texts from all the data, sorting them into general themes and sub-themes according to our theoretical typology. In the final stage, we coded the data in accordance with the final themes and sub-themes. Each of us performed the analysis separately, and we then discussed our results and reached consensus regarding the coding. Finally, we analysed the data and through reflexive reading of existing literature on solitary confinement.

Neutralising solitary confinement in action

Israeli courts and solitary confinement decisions: General observations

Before moving on to discuss our findings and analysis of decisions made by Israeli courts to grant or deny IPS requests to extend solitary confinement placements, it should first be noted that judges in these cases were clearly aware of the particular ‘pains of solitary confinement’. This is in line with the supreme court’s rhetoric (and the relevant Bill), recognising that solitary confinement is ‘harmful’ and holds ‘inherent difficulties’ and should be limited to use as a ‘last resort’ (Azran v. State) (5.6.17).

To illustrate, in the cases we analysed, one judge proclaimed that: ‘Man is a social animal: take human company away from him, and you take away the essence of life’ (Abutbul v. IPS (20.7.11)). Another judge noted that:

Solitary confinement entails social isolation and reduced environmental stimulation that all human beings are used to in daily life, a deprivation which may lead to real psychological tension, nervousness, impatience, shrinking of the personal experience and despair, as well as other physical symptoms. (IPS v. Mugrabi (16.4.2009))

The harms of solitary confinement, as another judge said, were a given: ‘even without any expert opinion regarding the consequences for the prisoner, I presume that solitary confinement might hold very serious mental consequences even for a person with great mental resilience’ (IPS v. Sa’ad (20.10.10)).

The court’s role is to balance these predictable harms against the reasons put forward for the isolation of any person:

There is no doubt that solitary confinement worsens the conditions in which the prisoner is held, and that can be taken as a given. There must be real reasons to balance against this harm for the court to approve the solitary confinement. (Harush v. IPS (13.07.2010)).

The courts’ analysis corresponds to the requirement in international law that solitary confinement placements satisfy six tests set out in Mandela Rule 45(1). Placements must be: exceptional; used as a last resort; as short as possible; authorised by a competent legal authority; not imposed on people because of their original crime; and, subject to independent review (see PSI, 2003). In contrast to the Mandela Rules’ requirement for solitary confinement stays to last no longer than 15 days, the courts’ involvement in the cases analysed only started after the person had already been isolated for 6 months, and some stays lasted substantially longer, including one stay of 22 years, three stays of 14 years, and 2 stays of 15 years, for example.

In 105 of the 354 decisions analysed (29.6%), the request to extend the person’s solitary confinement was approved by the court without substantive deliberation, on the basis of the prisoner’s consent or the prisoner’s refusal to attend the hearing. In these cases, the courts ‘rubber stamp’ the prison authorities’ request without providing any narrative reasoning for their decision. We grouped these cases together and did not analyse them further. The courts’ reasonings in the remaining 249 decisions are analysed below.

The formal reasons put forward by the state (there may more than one reason) for extending solitary confinement in these cases were: safety of the prisoner or others (194; 77.9%); preventing serious harm to discipline and ordinary life of the prison (87; 34.9%); prison security (63; 25.3%); national security (42; 16.8%); preventing violent, drug and organised crime (35; 14%); and public safety (17; 6.8%). However, notably, when deciding, courts very often conflated all these to a single generic concept of ‘risk’/‘dangerousness’ and rarely referred to any specific distinguishable risk factor.

Finally, we found the supreme court reluctant to intervene in district courts’ decisions. Leave for appeal was denied in 96% of the cases (22 cases were denied fully; 85%) and with some limits in 3 cases; 11%) and only in one case the supreme court ordered release from solitary confinement (juvenile, after serving 3.5 years in solitary confinement) (State v. Doe (22.10.12)).

Denial of alternatives and appeal to necessity

Both Article 19b and international law require that solitary confinement be used only as a last resort when other alternatives have been sought, tried, and failed. Our analysis of decisions suggested that, whilst the courts stated that solitary confinement was ‘inescapable’ in many of the cases we examined, prison authorities were rarely asked about alternatives sought prior to deciding to isolate the prisoner.

The language and claims of necessity, however, were very much part of the courts’ discourse. In 43% of the cases analysed (108), the courts directly stated that there was no alternative to isolating the prisoner, invariably describing solitary confinement as a ‘vital necessity’, ‘last resort’, derived from an ‘inevitable reality’ and ‘inescapable’.

The language of necessity was particularly evident in the framing of decisions involving prisoners sentenced for terror-related offences as part of a wider ‘war on terror’. One judge, for example, noted that ‘[in] war times … the security interest outweighs the rights of the prisoner’ (IPS v. Haj (2.2.2004)).

In these cases, then, the pains of solitary confinement are subordinate to the need to protect the state, as observed by the court in one case involving a prisoner serving a life sentence for serious terror-related offences:

 … In the current state of war that Israel is in against different terror organisations who are conducting an armed campaign against it, the security services and the army are busy protecting Israeli citizens. The Prison Service joins this effort by doing everything it can to neutralise the continuation of terrorist activity of security prisoners within the prison’s walls. I am convinced that the respondent continues to pose a security threat to the state of Israel, and the request to continue his solitary confinement should be approved. (IPS v. Baraguti (6.9.2004))

The protection of state security, in these narratives, calls for exceptional measures and requires all parties– including the IPS and the courts themselves – to be on board. In this context, we should be reminded that approximately 16% of the cases analysed involved a prisoner charged with terror-related offences.

The denial of injury: Minimising the extent of harm to the specific prisoner

In our analysis, this category included all the cases where the court explicitly stated that, in the specific case before it, the prisoner was not harmed by their solitary confinement, or that their isolation did not constitute solitary confinement ‘proper’. This form of neutralisation was found in 8.4% of cases (21).

As noted earlier, the pains of solitary confinement were acknowledged by the courts. Our analysis found, however, a gap between this knowledge and its application to any given case. According to the judicial narrative analysing specific cases, little (or not disproportionate) harm is caused to the prisoner by his solitary confinement, and any harm which may be caused is temporary and/or has been mitigated. In some cases, the courts applied forms of ‘implicatory denial’ (Cohen, 2001), where the conventional psychological or moral implications of solitary confinement are denied.

This form of denial of injury is well illustrated in the court’s reasoning in the case of Yigal Amir, a high-profile prisoner convicted of murdering the late Israeli prime minister Yitzhak Rabin, who had been in solitary confinement for more than 15 years at the time of the hearing. Deliberating IPS’s request to extend his solitary confinement, the court asserted that there is a difference between ‘solitary confinement’ and ‘complete isolation’:

We do not take lightly the conditions of solitary confinement; however, we emphasise that the prisoner isn’t completely separated from the outside world. He has conjugal visits once a month and speaks daily with his relatives … . He can watch TV though he chooses not to and can receive newspapers. It is indeed solitary confinement, but this is not someone who is completely isolated from the outside world. Therefore, the harsh mental consequences of the permanent solitary confinement described in the expert opinion provided on behalf of the prisoner, are irrelevant. (Amir v. IPS (7.12.10)).

In another case, involving a man serving a life sentence for murder, the court approved IPS’s request, noting that the prisoner went for daily walks, received visits, and had been approved for conjugal visits: ‘therefore and without diminishing the hardship of solitary confinement, I believe that the prisoner’s argument regarding his living conditions describes a harsher situation than it is in reality … ’ (Abutbul v. State (20.7.11)).

In other words, these prisoners are not subjected to ‘real’ solitary confinement, which could have potentially caused immense pain had the prisoner been subjected to it. This insistence that the solitary confinement is not really solitary confinement assumes the existence of a ‘total’ solitary confinement, where the individual is well and truly all alone with no contact whatsoever with the outside world. But such forms of complete and total solitary confinement do not exist in most prisons, at least in Western democracies, and by assessing the prisoner’s conditions of confinement against a hypothetical form of total isolation, judges fail to apply the tests prescribed by law, and minimise the injury of the solitary confinement to the prisoner in question.

The denial of responsibility

In this narrative, the judge is not actually responsible for causing the prisoner pain, which is ‘due to forces outside of the individual and beyond his control’ (Sykes and Matza 1957: 667). The decision-making is delegated by the court to the IPS, or to security, social or medical expertise, denying its own agency as primary decision-maker, or only exercising very limited discretion. It should be noted that, in our analysis, we only included in this category cases where the judge specifically stated that they would defer to IPS/security/sociomedical expertise. This form of neutralisation was used by the courts in 67% of the cases analysed (166 cases).

Deference to prison authorities

In the cases we examined, the courts exercised three main forms of denial of responsibility.

First, deference to IPS instead of de novo decision-making. The role of the court in solitary confinement decisions, as note dearlier, is that of a primary judicial authority, not to review prison authorities’ administrative decisions (Musli v. State (8.11.11)). However, in several of the cases analysed (9.6%; 24), courts elected to minimise their own role. In one case, for example, the court noted that it ‘is not authorised to replace the Prison Service’s decision-making with its own, unless a substantial deviation from reasonableness has occurred’ (IPS v. Hoch (17.4.2008)).

A second form of denial of responsibility by the court is to approve solitary confinement, and then direct IPS to modify the practice at the margins: ‘I accept the request and extend the prisoner’s solitary confinement … at the same time, the prison service must take measures to place the prisoner in double bunking or other conditions that are less harmful to him’ (IPS v. Zaguri (20.9.2012)), or to state, having approved the continuation of solitary confinement, that it ‘presumes that the prison service will do their best to find an alternative to solitary confinement as soon as possible, if such alternatives exist’ (Alwachidi v. IPS (14.7.2019)). How and in which way these judicial remarks were followed is unclear.

A third variant involved the courts refusing IPS’s request in its entirety, but allowing the solitary confinement to continue for a significant time. In one case, for example, the court approved a six-months extension of the prisoner’s double-bunked solitary confinement, rather than the one year requested by IPS (IPS v. Ben-Shitrit (30.9.2015)). In another case, the court was highly critical of IPS, demonstrating it ‘failed’ the tests applied to assess solitary confinement placements, but nonetheless approved its continuation for an additional month (IPS v. Simandiev (21.6.2016)).

Deference to the state’s security authorities: Security intelligence reports

Another form of ‘expert opinion’ were security intelligence reports, typically disclosed only to the judge, with the prisoner and their legal representative unable to see or challenge them. At best, the prisoner received a paraphrase of the intelligence information, and no cross examination of the security services was normally allowed; criminal evidence rules did not apply (Abutbul v. IPS (20.7.11)). We found deference to the general security service (GSS) expertise when, for example, the prisoner argued that the information presented against him was inaccurate: ‘it is a professional intelligence work that is part of the professional expertise of the GSS, and the court will not put itself as a professional body instead of the GSS, as the court lacks an appropriate qualification and information’ (IPS v. Mugrabi (3.5.2011)).

In another case, approving IPS’s request to extend solitary confinement, while acknowledging the distress caused to the prisoner, the judge said:

I have been provided with classified information which is so secret and sensitive that I am unable to even paraphrase it. In light of this information, I cannot find that the Prison Service’s decision not to allow double bunking was unreasonable. (IPS v. Ayetal) (22.9.2011).

This almost Kafkaesque reasoning is near impossible to refute. The court and the security services are having a conversation to which neither the prisoner nor their representative are party. In these cases, the judge, rather than being a neutral arbiter, acts as an extension of the state, part of the security apparatus whose role is to protect the state, not the wellbeing or rights of the individual prisoner.

Deference to socio-medical authorities: Socio-medical expert reports

Prisoners themselves rarely appeared before the court to testify about their experience of solitary confinement. Instead, courts appeared to rely to a great extent on written expert reports, including reports by medical doctors and social workers required by Prison Service Regulations (PSO, 2013, s. 11I). These reports are meant to assist the court in its own overall assessment of the reasons put forward by the prison authorities for continuing the person’s solitary confinement, balanced against the predictable harms of solitary confinement. Our analysis found that expert reports were mostly accepted at face value by the courts and taken as an indication that the prisoner’s solitary confinement is not harmful to them, even when there was strong indication of harm. In one case, for example, the Court approved the continued solitary confinement beyond three years of a prisoner described as ‘psychotic’ by the prison service’s own psychiatrist:

The medical report from IPS psychiatrist states that the prisoner is indeed psychotic, that he opposes hospitalisation, and that there is no justification for compulsory treatment in his condition. However, the opinion also states that there had been no deterioration in his condition recently, and that the conditions of solitary confinement are not harmful to his mental state. (IPS v. Dochel (10.4.2010)).

In this context it is important to note that in Israel both social workers and health staff are Prison Service employees, raising a range of potential issues around ‘dual loyalties’ (Shalev, 2008).

The denial of the victim

Denying the victim, in our analysis, refers to a form of neutralisation whereby rather than considering the pains of solitary confinement, the decision-maker focuses on the immoral nature of the prisoner and their own acts, which by implication means that they ‘deserve’ their fate and cannot be considered as ‘victims’. That is often done through framing the prisoner as someone who is not quite ‘like us’, making the decision to place someone in solitary confinement less uncomfortable. The ‘othering’ and ‘estrangement’ of prisoners becomes yet another discursive route for maintaining psychological, social and moral distance from the prisoner as a person, enabling the decision-maker to relinquish responsibility for any damaging consequences of the solitary confinement (Cohen, 2001).

We identified this form of neutralisation in 17% of the cases analysed (43), where the court explicitly denied the prisoner’s victimhood citing either the prisoner’s lacking morality, or emphasising their own riskiness.

Retributive discourses – involved framing the prisoner as morally deserving to be held in solitary confinement. This ‘character retributivism’ logic aims to fit the prisoner’s extreme suffering to his or her ‘inner viciousness’ and entire moral personality. This allows for the assessment to move from the prisoner’s immediate risk to the prisoner’s overall depraved moral character as reflected from their accumulative criminal history and particularly the crime they were convicted of.

The retributive logic was evident in the court’s assessment of a request to extend the solitary confinement of a prisoner sentenced to life for serious terrorist crimes:

Every prisoner is entitled not to be held in solitary confinement, unless the conditions stipulated in sec. 19 are met … And yet, he was indicted of offences that led to a sentence which, as the respondent’s own legal representative put it: ‘when you read his sentence for these offences, it is quite scary: not many prisoners serve 35 life sentences’. Such a person is not entitled to be treated leniently, lest, god forbid, this court acts in a way which fulfils the saying that ‘he who pities the cruel shall end up being cruel to the pitiful’ (State v. Anagdat (12.5.16)).

Another example involved a continued solitary confinement placement for Yigal Amir:

I agree with the claimant’s [IPS] position on the difference between the prisoner and other security prisoners regarding the continuation of his separation, even considering the passage of time, because it correctly expresses the normative public position regarding the prisoner’s deed and the ideological motive behind it. (IPS v. Amir (5.4.2011)).

The focus on the prisoner’s sentence in solitary confinement placements contradicts the principle that solitary confinement is an internal prisoner management tool, which is supposed to be entirely independent of the prisoner’s sentence. This important principle is also reflected in the Mandela Rules (2015) which prohibit solitary confinement placements on the basis of the prisoner’s crime or sentence. For the prisoner in question, the confluence of their original crime with their prison conditions may mean an unbreakable cycle of retributory solitary confinement.

Risk based discourses – denying the isolated prisoner’s suffering involves the framing of the prisoner as part of a highly dangerous risk group. In the court judgements analysed, this form of justification was particularly prevalent with prisoners held on national security grounds:

Security prisoners, that is, prisoners convicted of offences against the state, are a type in their own right. The classification of security prisoners as a type in its own right stems from the special danger ingrained in them. (IPS v. Gabarin (13.7.2004)).

The re-framing of individuals as a group means that risk could be de-individualised and managed with all those belonging to the group treated as high-risk and dangerous (Garland, 2001; Shalev, 2013). This is another form of ‘othering’ and ‘estrangement’ that help to deny the prisoner’s suffering. This time, however, this is done not through punishing the prisoner as a moral subject, but by de-individualising them as a ‘fixed risk object’, whose risk needs to be eliminated through solitary confinement.

The appeal to higher loyalties

As noted earlier, judges in the cases we analysed were aware, at least partially, of the ‘pains of solitary confinement’ and, by implication, of the potential consequences of their decision to continue a person’s solitary confinement. This is a ‘troubling recognition’ (Cohen, 2001: 87) that has to be neutralised. One way of doing this is through ‘moral prioritising’ (Tombs and Jagger, 2006), as exemplified by the court’s assertion in one case that the solitary confinement of a prisoner with severe mental health issues was justified for the prisoner’s own safety: ‘The request to extend the period of solitary confinement and the basis for the request takes precedence over the prisoner’s personal circumstances’ (Harush v. IPS (13.07.2010)).

A perhaps more common form of appeal to higher loyalties which must take precedence over the individual prisoner’s wellbeing involves national security. National security is a particularly sensitive issue in Israel, and this was reflected in the cases we analysed. In one case, involving a prisoner who was a high-ranking member of a designated terrorist organisation, the judge noted that:

The prisoner’s continued separation is necessary to interrupt his involvement in directing terrorist attacks out of the Territories into Israel, and to disrupt this terrorist infrastructure which he seeks to continue to conduct from within the prison wall … The State of Israel is in a state of war, facing various terrorist organisations, waging an armed, violent and bloody fight against the State of Israel. These days, more than ever, the General Security Service, the Israeli police and various military officials are doing their utmost to prevent murderous terrorist attacks, in other words, to safeguard the safety of Israeli citizens. (IPS v Baraguti (26.1.2004)).

We identified an direct appeal to higher loyalties in 10% of the cases analysed (25). The danger with this form of neutralisation is that, where security prisoners are concerned, rather than acting as an independent review authority for the prison service’s decision to isolate a prisoner, the courts are acting as protectors of the state or delivering societal messages aimed at the collective consciousness, thus potentially failing to provide a fair and balanced review of the risk of harm to the individual prisoner.

The condemnation of the condemners

This technique of neutralisation involves the judge shifting the focus of discussion from the extremity and harms of solitary confinement to the prisoner himself. This technique is distinct from that of denying the victim in that rather than denying that there is a victim, the victim is blamed for their own predicament: ‘The state’s repeated claims [about the necessity to isolate] are derived from the prisoner’s repeated behaviour’ (IPS v Ben-Shitrit (30.10.15)). In another case, extending a prisoner’s stay in solitary confinement, the judge noted that:

The request to extend the solitary confinement is mostly based on the prisoner’s own behaviour … therefore, the respondent cannot blame anyone but himself, and the key to leaving solitary confinement is in his own hands. (State v Avital (10.4.2012)).

Once the focus has shifted to the prisoner himself, it follows that anything that happens to them is a result of their own behaviour and they have ‘brought it on themselves’. The implicit assertion is that if the prisoner were to change, their treatment would correspondingly change. We identified this technique of neutralisation in as many as 9% of the cases analysed (23).

Interpretive denial

This important interpretive technique is common when there are factors in favour of the prisoner. Here the court might ‘spin’ the meaning of these favourable factors. For example, where there is no evidence to support the need to isolate a prisoner, instead of framing the lack of negative evidence as an indication of reduced risk and hence denying IPS’s request to extend the solitary confinement, the court may use it in the opposite way: the scant evidence simply means that the prisoner’s solitary confinement thus far ‘works’ and a further period in solitary confinement is justified- indeed required. This was demonstrated in two cases involving lifers convicted of terror related offences:

 … since the prisoner is in solitary confinement, the intelligence information against him is poor and as a result cannot in itself evidence his risk. However, it is clear that this is not all there is with regard to this prisoner and his proven risk … at best, for now, this can demonstrate the relative effectiveness of further solitary confinement and the justification for it. (IPS v. Mugrabi (16.4.2009)).

Indeed, there is no great weight of intelligence information against the prisoner. However, it should be ascribed to the effectiveness of the solitary confinement which, while not completely preventing contact between the prisoner and other prisoners, makes it harder. (IPS v. Baraguti (16.5.10)).

So, in these cases, the prisoner is caught in a ‘catch-22’ situation: (a) if the state presents evidence that proves their risk, this constitutes a strong justification for a further period in solitary confinement; (b) if there is no such evidence (or not enough), that may only demonstrate that the time they have spent in solitary confinement so far ‘works’ in reducing their risk, and a further period is thus justified. This form of neutralisation was present in as many as 12% of the decisions analysed (31).

Discussion

Our analysis explored the judicial authorisation of solitary confinement placements. We demonstrated that the ‘pains of solitary confinement’, whilst acknowledged in the courts’ rhetoric, did not appear to dissuade judges from authorising very long stays in solitary confinement, some lasting over a decade. Judges used a variety of ‘techniques of neutralisation’ to explain away the ‘pains of solitary confinement’ and diminish their own role in authorising the infliction of these pains. This is in a context in which the courts’ involvement only begins after the prisoner had already spent 6 months in solitary confinement, far longer than the 15 days specified in the Mandela Rules (2015) as prolonged, and therefore prohibited, treatment.

In the vast majority (93%) of cases we analysed, IPS requests were approved by the courts (with some conditions in around 8% of these cases). Furthermore, rather than establishing that the prisoner was not isolated because of their original crime, as required by the Mandela Rules, in 17% of the cases, especially those involving very serious crimes, the courts appeared to do quite the opposite. This reflects a ‘diffusion’ of sentencing logic to prison logic through the transformation of solitary confinment into a retributive ‘act of extreme punishment’ for the prisoner’s crime or depraved character (Polizzi, 2017: 39).This is further integrated with findings that solitary confinement may both be experienced as punitive and be often imposed for punitive purposes by prison officials, even when labelled as administrative (Franco et al., 2020; Shalev, 2013).

Judges’ reluctance to see themselves as impartial ‘independent reviewers’, carefully scrutinising the specifics of each request brought before them, may be motivated by socio-legal and organisational contexts of the judicial process.

First, the legal proceedings we analysed are ‘hardly a contest in the usual sense’ (Calavita and Jenness, 2015: 186). One aspect of the weak adversarial nature of the process is the heavy reliance on classified information (secret reports; normally no cross examination; lack of formal evidence rules) submitted by prison authorities mostly for the judge’s eyes only (‘denial of responsibility’). A second aspect is when the process changes from being about the protection of the prisoner’s rights, to a forum for delivering societal messages, like the ‘war on terror’ (‘appeal to higher loyalties’).

Second, the discretionary nature of the judicial process seems to incentivise a risk-averse logic that counterbalances the prisoner’s rights. As we saw, the court may decide to extend solitary confinement on the basis of perceived risk, even if there is no (or not enough) information against the prisoner. Therefore, even ‘false positive’ cases can be explained through ‘interpretive denial’ as a success of the solitary confinement logic, since to release a prisoner from solitary confinement ‘would be to admit a mistake in these categorizations’ (Reiter, 2016: 145). Further, as discussed, the supreme court is very rarely willing to intervene with district courts’ exercise of discretion, which may also reinforce existing practices. That was evident, for example, when the courts directed IPS to modify the a person’s solitary confinement at the margins (‘denial of responsibility’).

Third, the cases we analysed demonstrate an extreme asymmetry of power between the prisoner and the prison authorities. Feeley and Rubin (2000) suggest that judges, by and large, were found to ‘dislike criminals’ and ‘favor established institutions’ when discussing solitary confinement (Feeley and Rubin, 2000: 143). Prisoners are perceived through the lens of punitive risk-based logic, which was the dominant discourse in the cases we analysed, spurred on by IPS’s appeal to protect the state, public, or prison safety/security from the risk posed by the prisoner. In some cases, as we demonstrated, courts perceived the prisoner as deserving of solitary confinement for their past crime or for their extreme risk (‘denial of victim’). In others, prisoners may be seen as bringing this fate on themselves through their own behaviour (‘condemnation of the condemners’). The courts’ rhetoric labels these prisoners as ‘official troublemakers’, ‘worst of the worst’ or ‘the most dangerous’ (Shalev, 2013). These labels serve as a powerful exclusionary stigmatisation of the ‘administrative subject’, even beyond the already stigmatising label of ‘prisoner’ (Reiter, 2016: 216), and frame these prisoners as falling outside the boundaries of the community’s responsibility, undeserving of the human rights of ‘innocent’ society members (Cohen, 2001). On the other hand, prison authorities enjoy not only a ‘repeat player’ status, but also the control of key aspects of the process (e.g. applying for extension) and the symbolic status of experts in security/risk (e.g. security services, prison, or police classified reports).

Finally, courts, when determining prison conditions, seemed to act as part of a quick and (purportedly) efficient organisational prison risk management exercise (see Feeley and Swearingen, 2004). The judicial process in the cases analysed was bureaucratised and routinised in different ways which limited its power to protect human rights.

First, probably the most clear example of routinisation was that in many cases (29.6%), courts served to rubber stamp decisions made by the prison authorities without any real investigation.

Second, routinisation was achieved through the use of ‘generic assertions of safety and security’ to legitimise their decisions, instead of giving specifics as to the exact risk that the prisoner poses (Coppola, 2019: 195).

Third, in some cases judges used risk-management rhetoric addressing the risk of a group (e.g. terror) rather than individual risk. This is similar to the underlying logic of the U.S. supermax prisons, where solitary confinement is imposed for being in a category (e.g., gang, terror) (Reiter, 2016; Shalev, 2013). This logic combines with the Israeli socio-cultural context, where judges –acting in the shadow of the politics of the Israeli-Palestinian conflict – are particularly inclined to defer to governmental agencies in security-related matters (Davidov and Reichman, 2010).

A fourth aspect of routinisation is that courts’ decisions were made as quickly as possible. This, according to one judge, was an unavoidable consequence of the high caseload facing the courts:

Unfortunately, there is a tension between the ideal and reality. The resource of judicial time is limited, and the number of cases and judicial obligations is large. It is customary, because of the ratio between time allocated to prisoner appeals, including requests to extend solitary confinement stays and their large number, to allocate around 15 minutes to each discussion … . (IPS v Agabaria (11.5.2011)).

We must also consider whether hearings lasting 15 minutes from start to finish are sufficient for the careful balancing act required for the court to assess the appropriateness of prolonging stays in conditions acknowledged by the judges themselves as extreme and harmful.

To sum up, the socio-legal and organisational contexts of solitary confinement decision-making seem to incentivise a largely symbolic/expressive process. The various ‘techniques of neutralisation’ we analysed help to bridge the gap this creates. This process enables judges, on the one hand, to express commitment to the ‘dominant normative system’ of human rights and, on the other hand, to normalise lengthy authorisations in solitary confinement as ‘acceptable’ if not ‘right’ practice (Sykes and Matza, 1957: 667).

The findings are also consistent with U.S. prison litigation research which suggests that prison reform may serve as a ‘double-edged sword’– though aimed at enhancing prisoners’ rights, these reform actually strengthen prison officials’ capacity to control (e.g., building more prisons, enhancing authoritarianism through bureaucratisation) (Feeley and Swearingen, 2004: 435). As our analysis suggests, the power dynamics and security/risk logic of prison administrations play a more dominant role in the judges’ decision-making process than the harms of solitary confinement placements which they authorise. This highlights the challenges of reforming solitary confinement through judicial decision-making. It also emphasises the importance of understanding the socio-legal and cultural contexts of the relationships between judicial and prison officials when analysing the use of solitary confinement.

Finally, the findings should be read in light of several limitations, some of which are inherent to the qualitative methods used (e.g., other possible interpretations, the specific context of the study). Further limitations stem from the fact that the data was drawn from open-access legal databases, and does not include judicial decisions which were not published for substantial (e.g. confidentiality) or technical reasons. Finally, this study analyses judicial decision-making for lengthy periods of solitary confinement and does not analyse prison authorities’ decisions of solitary confinement placements of up to 6 months, which are an important part of the overall picture. Future research should explore additional aspects of solitary confinement decision-making (e.g. comparing judicial/prison decision-makers’ narratives) including the provision of a full account of solitary confinement rates (and other aspects of its use) in Israel. Despite these limitations, however, we believe that this study is a useful step for understanding and improving the theory and practice of long-term solitary confinement decision-making and its underlying socio-legal and organisational forces.

Conclusion

This study has shed light on the little explored intersection of judicial decision-making and prison life. It utilised Sykes and Matza’s (1957) ‘techniques of neutralisation’, as developed by Cohen (2001), to analyse 354 decisions made by Israeli courts regarding prison service requests to prolong prisoner placements in solitary confinement beyond an initial period of 6 months. It found that judges recognised the particular pains of solitary confinement, but have nonetheless approved over 90% of requests to extend its duration, explaining away these pains by various ‘techniques of neutralisation’. Our analysis has illustrated the importance of the socio-legal and organisational contexts in decision making regarding solitary confinement placements. The process we analysed was characterised by extreme asymmetries of power between prison authorities and the prisoner, weak adversarial nature (e.g. heavy reliance on classified reports), and routinised-bureaucratised decision-making (e.g. expertise-based, generic, efficient and quick). This appears to incentivise judges’ reluctance to act as impartial ‘independent reviewers’ protecting prisoners’ rights, and thus contributes to a further increase in the use of prison security and disciplinary measures. These findings call for further research (normative, comparative, and empirical), exploring ways to better protect prisoners’ rights in judicial and prison decision-making.