Solitary confinement as state harm: Reimagining sentencing in light of dynamic censure and state blame

Abstract

The continuous perpetration of unjustified harms by the carceral state through its use of solitary confinement justifies the creation of a novel process of automatic sentence review. This process is necessary to account for such state-perpetrated harms and communicate censure more accurately. This article proposes the use of a communicative theory of punishment developed in sentencing to characterise and account for the state's wrongdoing and harms in the context of a sentence that involves solitary confinement. Specifically, it outlines a justification for an automatic review process of the offender's carceral sentence based on an expanded and relational understanding of censure developed in the literature and proposes a two-step process to implement this review.

Introduction

On December 6, 2016, after being held in solitary confinement for 1647 days – more than 4 years – Adam Capay took his life away in his cell. A father of three, member of the Lac Seul First Nation, he was only 26 years old when he passed away (McQuigge, 2019). During his confinement, the young man had been suffering from post-traumatic stress disorder, severe hallucinations, and could barely speak (Prokopchuk, 2019).

The case of Mr Capay is part of a recent pattern of inmates dying in solitary confinement in Canadian penitentiaries (Ramlakhan, 2018). In the wake of these tragedies, civil society organisations challenged the constitutionality of this practice and partially won. Prolonged solitary confinement was deemed unconstitutional under the Canadian Charter of Rights and Freedoms (Charter) (Canadian Civil Liberties Association v Canada (CCLA v Canada), 2019 ONCCA 243 (28 March 2019); British Columbia Civil Liberties Association v Canada (BCCLA v Canada), 2019a; BCCA 228 (24 June 2019)). Parliament was consequently required to reform the Corrections and Conditional Release Act (CCRA) (1992), and the use of segregation in federal penitentiaries (BCCLA v Canada, 2019a). In 2019, Parliament adopted Bill C-83, which the federal government declared would ban solitary confinement and replace this practice with Structured Intervention Units (SIUs) (2019, ss. 10, 39–40; Bryden, 2019). Bill C-83 also amended the CCRA to include oversight mechanisms for reviewing the use of SIUs and avoiding any unnecessary harms they could cause to inmates (2019, s. 10). Nevertheless, reports conducted by Doob and others revealed that solitary confinement remains alive and well under the guise of SIUs and that the new review mechanisms have been highly deficient (Doob et al., 2021).

This article argues that the continuation of unjustifiable harms perpetrated by the state through SIUs justifies the creation of a novel review process for reassessing the sentence to account for state harms and communicate a more accurate censure. This process, referred to as ‘SIU Harms Triggered Sentence Review’, builds on Manikis’ (2022) framework of ‘state blame’ which seeks to qualify the state's wrongdoing and harms by using a novel typology, and review an offender's carceral sentence to convey accurate censure which accounts for the state's blame and related harms. This proposal endeavours to hold the state accountable for unjustified SIU harms and proposes to re-sentence offenders accordingly and provide a response that allows for more accurate censure.

While this article focuses on the Canadian context, the framework and proposition of a novel sentence review process can apply in other common law jurisdictions that value communicative theories of punishment and face similar challenges with solitary confinement. Specifically, this piece uses the unaccountable context of solitary confinement in Canada to illustrate the importance of considering such prison conditions as an extension of sentencing that warrant justification and judicial accountability throughout the sentence. This contrasts with the current approach in most common law jurisdictions which consider this reality as a separate and unrelated dimension of sentencing. The proposed novel review process would be an important communicative aspect of sentencing that would require to communicate, analyze and respond to state-induced harms within the administration of the sentence. Finally, while this paper proposes this novel review process to communicate and address excess state-created harms in the specific context of solitary confinement, this revised censure-based approach can also be expanded to address wider contexts of, or changes to, prison conditions which can also be considered excess state-created harms during the administration of the sentence.

This paper is divided into three parts. The first part provides a background on solitary confinement with the limits that relate to the current review process in Canada. The second part presents the paper's theoretical framework, underpinned by Manikis’ (2022) account and typology for state blame and harms. The third part proposes the SIU Harms Triggered Sentence Review and applies this framework to the analysis.

The challenges and harms of solitary confinement: The Canadian example

According to the United Nations Standard Minimum Rules for the Treatment of Prisoners (Mandela Rules), solitary confinement is defined as ‘the confinement of prisoners for 22 h or more a day without meaningful human contact’ (2015, Rule 44). Prolonged solitary confinement occurs when a prisoner is subject to these conditions for more than 15 consecutive days and constitutes torture (Mandela Rules, 2015, Rule 44).

Research demonstrates that solitary confinement causes severe psychological distress. The cells of such segregation are often small spaces without windows (Dobson, 2020: 102). The American Psychological Association claims solitary confinement is associated with an ‘increased risk of self-mutilation, and suicidal ideation; greater anxiety, depression, sleep disturbance, paranoia, and aggression’ (2016: 1, cited in Haney, 2018: 368). While offenders in solitary confinement are at a heightened risk of exacerbating their pre-existing mental health conditions, even offenders without such pre-existing ailments are susceptible to irreversible psychological damage (Reiter and Blair, 2018; Coppola, 2019: 207).

In response to the harms of solitary confinement, Canadian human rights organisations have constitutionally challenged the CCRA – the legislation which formerly authorised its use. In both CCLA v Canada and BCCLA v Canada, the challenges were heard before the jurisdictions’ respective courts of appeal (2019; 2019a). In 2019, the Ontario Court of Appeal (ONCA) ruled that prolonged solitary confinement violated s.12 of the Charter by amounting to cruel and unusual punishment as it ‘causes foreseeable and expected harm which may be permanent and which cannot be detected through monitoring until it has already occurred’ (2019: 5). Similarly, the British Columbia Court of Appeal (BCCA) ruled that solitary confinement violated s.7 (the right to life, liberty and security) of the Charter, finding that certain provisions of the CCRA authorised indefinite and prolonged solitary confinement and offered primarily internal rather than external reviews of decisions to segregate inmates (2019a).

Following these decisions in 2019, Parliament was required to amend the CCRA and ensure that, at the very least, its practices of segregation in federal penitentiaries were constitutional (British Columbia Civil Liberties Association v Canada, 2019b BCCA 233 (26 June 2019)). That same year, Parliament adopted Bill C-83, declaring that this legislation banned the use of solitary confinement (2019, ss. 10, 39-40; Bryden, 2019) and that a new practice would replace solitary confinement – Structured Intervention Units (SIUs). Under these changes, offenders can only be in SIUs for a maximum of 20 h a day and two of the four hours spent outside their cell must provide for social interaction (CCRA, 1992, s. 36(1)(b)). The permissible use of SIUs falls just below the threshold of solitary confinement as defined by the Mandela Rules.

The changes made to the CCRA also included introducing oversight mechanisms vis-à-vis the segregation of offenders. Currently, the CCRA contains specific provisions for reviewing an offender's placement in an SIU, its conditions, and whether all reasonable steps were taken by correctional staff to provide the affected offender with the requisite hours of human interaction and time outside the SIU (CCRA, 1992, ss. 31–37). Additionally, a Structured Intervention Unit Implementation Advisory Panel (Advisory Panel) was established by the federal government to monitor and assess the implementation of the SIUs (Public Safety Canada, 2016). However, as revealed by Doob, these oversight mechanisms have been deeply flawed.

Deficiencies of the CCRA's new review mechanisms: Burdensome and failing to hold the state accountable for harms

Despite the oversight mechanisms created by the Federal Government following the changes made to the CCRA, Doob and others have discovered that the use of solitary confinement has not ended in federal penitentiaries (Doob and Sprott, 2020; Doob et al., 2021) and that the new review mechanisms within the CCRA are highly deficient (Doob et al., 2021). The two primary review mechanisms under the CCRA are presented below along with their weaknesses.

CCRA review of whether to remove offenders from an SIU or to alter SIU conditions

The first review mechanism under the CCRA serves to determine whether an inmate should be removed from an SIU or whether its conditions should be altered. This mechanism has been characterised as a ‘multi-layered bureaucratic process’ as it occurs in several lengthy steps (Dobson, 2020:101) and fails to provide an external oversight process. These steps are as follows.

1. Mental health assessment by a registered health care professional.

Under the CCRA, a mental health assessment by Correctional Service Canada (CSC) is required within 24 h of an offender's placement in an SIU and subsequently, the offender should, at least once a day, be visited by a ‘registered health care professional’ (1992, s. 37.1(2)(a)(b)).

2. Institutional head decision regarding the altering of conditions or removal from SIU.

Following the mental health assessment, a registered health care professional may recommend to the institutional head that the conditions of the SIU be altered or that the offender be removed from the SIU (s. 37.2, 1992). Following this recommendation, the institutional head shall make this determination ‘as soon as practicable’ (CCRA, 1992, s. 37.3(1)(a) – (2)).

3. Committee or commissioner review of a decision not to alter SIU conditions or remove from an SIU.

If an institutional head determines against the SIU alteration of conditions or the removal of an offender from an SIU, a committee established by the Commissioner of Corrections, consisting of CSC staff members higher in rank, or the Commissioner of Corrections themselves, shall review this decision (CCRA, 1992, ss. 37.31(3), 37.32, 37.4). If an offender is kept in an SIU, there must be reasonable grounds to believe this is necessary for the safety of the offender or the penitentiary, or to ensure a potential investigation is not interfered with (CCRA, 1992, s. 37.41(1)).

4. Independent external decision-maker review of a decision not to alter SIU conditions or remove from an SIU

According to the CCRA, an Independent External Decision-Maker (IEDM) shall review, within 30 days, the Commissioner's decision not to alter an offender's SIU conditions or not to remove them from an SIU (1992, s. 37.8). The IEDM review must occur as soon as practicable when such a decision is made by the committee (CCRA, 1992, s. 37.81). (CCRA, 1992, s. 37.8, s.37.81, 37.82). However, Doob and his colleagues found that when interpreting the CCRA along with the relevant Corrections and Conditional Release Act Regulations (1992), and the Commissioner's Directives (2019), there can be significant delays in the processing of such reviews by IEDMs (Doob et al., 2021: 7). Indeed, it is only after a lengthy period of approximately 60–67 days in an SIU that an offender's file is given to an IEDM for review (Doob et al., 2021: 7). Moreover, there is a lack of transparency regarding the information shared with IEDMs for decision-making, as well as the qualifications of the IEDMs (Doob et al., 2021: 14–15, 20–21).

Thus, the existing administrative review mechanisms under the CCRA are inadequate. Not only are the internal review steps complex and lengthy, but the final external review step is opaque and delayed – failing to hold the state accountable for violations and harms.

CCRA review of minimum hours of meaningful human contact and time outside of SIU

Under the CCRA, IEDMs may also undertake a review to investigate whether reasonable steps were taken by correctional staff to provide the offender with the opportunity to receive hours outside their SIU and for human interaction (1992, s. 37.83). This mechanism also shares substantial limitations, including the failure to hold state correctional facilities to account when failing to offer minimum meaningful contact to individuals detained.

Research conducted by Doob and Sprott (2020: 18) showed that most offenders in SIUs were not leaving their cells and were not receiving at least 2 h of meaningful human contact a day. Moreover, in at least 10% of cases, inmates in SIUs lacked meaningful human contact for more than 15 consecutive days (Doob and Sprott, 2020: 19). These critical findings further support the conclusion that IEDMs are ineffective with respect to overseeing the legislated minimum hours of CCRA s. 36(1).

The literature on prison accountability in Canada has long argued in favour of external oversight by the judiciary (Kerr, 2015; Parkes and Pate, 2006; Arbour, 1996). In a culture where administrative resistance has been prevalent, more robust tools with legally enforceable mechanisms are needed to examine the carceral state's actions in the production of harms and offer responses accordingly.

In light of these findings, and relying on theoretical underpinnings of punishment, this paper proposes a novel framework that recognises an automatic sentence review by a judge in contexts of SIU imprisonment, on the basis that this practice creates presumptive harms in excess of legitimate punishment. Without such review, unjustifiable harms will continue to be perpetrated by the unaccountable state and give rise to unlawful punishment. Interestingly, the American Law Institute (2017) proposed revised appeals in cases where the initial sentence is unlawful. This was meant to enable substantive appeals on the merits of individual sentences. The novel proposition in this article goes further by proposing reviews much later during the administration of the sentence and includes excess harms that were unforeseeable at the time of sentencing. These include contexts where initial sentences were lawful but where conditions during the administration of the sentence changed and may have become unlawful.

The following section proposes a novel ‘sentence review’ process undertaken by a sentencing judge – rather than mechanisms that remain within the administrative purview of CSC – on the basis that conditions relating to imprisonment are a continuation of punishment as censure. This proposal builds on Manikis’ (2022) theoretical framework of ‘state blame/harms’ and ‘dynamic/relational censure’ to support the view that SIU wrongdoing and harms need to be analyzed, communicated, and accounted for within the process of re-sentencing by the judiciary – rather than remain within the administrative purview of correctional services. It then relies on a typology of state wrongs/harms to discuss possible ways to consider SIU harms within this sentence review.

Theoretical underpinnings of sentence review in the context of solitary confinement

This section justifies the ‘sentence review’ process by relying on the framework of ‘state blame’ and dynamic censure developed by Manikis (2022). This framework will then specifically be applied to the development of a sentence review process that accounts for state wrongdoing/harms resulting from SIUs.

Sentence review as part of dynamic censure and state blame

Communicative theories of punishment have long focused on the notion of censure as an important justification of punishment (e.g., von Hirsch 1976; von Hirsch and Ashworth, 2005; Ashworth 2011). Duff's (2001) penance theory considers punishment a communicative enterprise in which the punishment engages with citizens and their shared values. This includes an authoritative, communal condemnation of wrongs that qualify as public. Some have challenged the notion of public wrongs on the basis that defining them as the ‘criminal law's business’ seems circular and needs greater clarification (Madden Dempsey, 2011). For Duff, the public nature of these wrongs suggests that ‘those who commit them should be called to account and censured by the community’ (Duff, 2001, 61). Censure of such wrongs is owed to victims, as members of the community, and as manifesting that concern for them and for their wronged condition. It is also owed to the society whose values the law claims to embody, showing that those values are taken seriously. Finally, it is owed to offenders themselves since an honest response to another responsible moral agent's wrongdoing is criticism or censure of that wrongdoing. Several criticisms have been highlighted towards communicative theories of punishment. Some suggest that they remain focused on a retributive dimension that emphasises moral desert and that instead should consider wider instrumental goals, such as deterrence and rehabilitation. Others have addressed the notion that Duff's theory is rooted in consequentialism and seeks to achieve wider goals such as repentance and apologies, which should not be the aim of punishment (von Hirsch and Ashworth, 2005). Finally, others have highlighted the limits of the dialogue proposed by Duff by suggesting that because punishment is imposed, it does not truly form part of a moral dialogue, even though offenders may receive lighter sentences by repenting and apologising (Brownlee, 2011). Similarly, others have noted the limits of communicative theories by highlighting that Duff's dialogue does not include mutual opportunities for blame, where the offender and other citizens may highlight the state's blameworthiness alongside that of the offender's (Manikis, 2022).

For Duff, communication involves a reciprocal and rational engagement. It requires someone to or with whom we try to communicate. Communication considers the offender as an active participant in the process who will receive and potentially respond to the communication as it appeals to the other's reason and understanding (Duff, 2001).

Building on Duff's responsive censure, certain desert theorists (Roberts and Dagan, 2019) have considered the criticisms towards traditional censure (Ryberg, 2021) and recently re-conceptualised censure and punishment as a more dynamic form of communication that extends beyond the offence. While they suggest that the communicative dimension of punishment is not aimed at provoking a response, they nevertheless highlight that if the offender takes the opportunity to respond, this can be considered at sentencing and during the administration of the sentence (Roberts and Dagan, 2019: 141, 143). Contrary to traditional desert theory, the offender is recognised as a responsive agent who partakes in a dialogue and cannot be ignored (Maslen, 2015). Accordingly, proponents of this approach have developed the notion of dynamic censure – highlighting the importance for authorities to review the original censuring decision over the years and to account for the prisoner's response to the sentence's message in relation to the offence (Roberts and Dagan, 2019). Examples of responsive offence-related factors developed by these authors include: the way the person addresses the harm inflicted by compensating the victim's loss; showing sincere empathy, remorse; apologising; and taking responsibility for the harm.

Relatedly, authors have expanded the traditional concept of censure in punishment theory by suggesting that censure should not be stagnant in time and solely determined during sentencing, but rather dynamic to account for the post-sentence administration of punishment as relevant to state censure. Specifically, Roberts and Dagan (2019) propose a ‘second look’ or ‘late sentence review’ whereby an offender's sentence can be reviewed with the passage of time to account for their engagement and actions in relation to the offence. If an offender exhibits positive behaviour during the administration of their sentence in reaction to the message the sentence intended to convey, the state would provide a ‘second look’ at the offender's sentence to potentially reduce its level of censure and severity (Roberts and Dagan, 2019: 149). This notion of censure was also discussed by Roberts and Maslen (2014), who have highlighted that since the offender's reflective process is important to the rationale underlying the censuring response, an offender who independently engages with the reflective process should consequently be censured. These authors specify that dynamic censure and parole are very different as they rely on different rationales and objectives. Specifically, while parole predominantly relies on a risk-based logic, censure, as described above, is rooted in communicative theories of punishment as part of a predominantly normative logic that focuses on fault and blame. Accordingly, adopting a dialectical model of censure serves a different function than the parole model, making room for input from the offender relating to this wrongdoing. Failing to moderate the degree of censure would be to address the offender as if they lacked any appreciation of the seriousness of the offence. Moreover it ‘would demonstrate deficient communication on the part of the state’ (Roberts and Maslen, 2014: 105).

Relying on this notion of dynamic censure, one of us (Manikis, 2022) has contributed to this communicative framework by expanding Roberts and Dagan's (2019) conception of censure to suggest that this concept should not only focus on individual censure but should also shift the analytical focus towards censuring the state. Manikis’ framework and its expanded understanding of censure specifically provides justifications and space to develop an analysis of the state's responsibilities and various types of harms to which it can be attributed. Her framework does not reject the importance of individual censure. She suggests that censure should not simply focus on the individual's conduct, but also on the state's conduct during the post-sentencing administration of the sentence, since this conduct can affect the level of censure towards the state. Specifically, relying on her notion of ‘state blame/censure’, which expands communicative theories of punishment, she underscores the importance throughout the punishment process of recognising and accounting for the state's relational and contributory role in the creation of harms as part of the process of censure (Manikis, 2022). This framework of ‘state blame’ is separate and more expansive than, yet complementary to the more traditional notions of blame and censure in criminal law, since it recognises that communicative censure should not merely focus on individual wrongdoing but also on the state's relational wrongdoing and related harms. In this sense, contrary to Roberts and Dagan's (2019) framework, the one offered by Manikis (2022) not only values desert-based approaches, but also expands the communicative endeavour towards censuring the state. For some, this might be considered a way to strengthen just deserts by advancing mutual opportunities for blame and therefore strengthening the legitimacy for individual blame. Others might perceive this as a mixed theory that understands communication as both desert-focused when focusing on the individual as well as utilitarian when it aims to achieve expressive and remedial purposes beyond individual desert.

Importantly, this framework is not intended to absolve the offender from responsibility for the crime committed nor necessarily remove the state's standing to recognise and legitimately sanction individual wrongdoing (Duff, 1988; 2010). Rather, the framework expands communicative theory by recognising the state's role and duties in relation to the offender and the offence – suggesting that the state should also be part of the censuring exercise and held accountable for its own contribution to the various harms relating to criminality, criminalisation, and punishment (Manikis, 2022: 317).

This broader understanding of dynamic censure and state blame is important in the context of harms relating to solitary confinement, since unlike Roberts and Dagan's (2019) dynamic censure, the communicative message relating to censure is not just focused on the individual in relation to their wrongdoing, but also on the state's creation of excess harms, which would need to be communicated and responded to as part of the sentence. To emphasise the importance of communicating censure towards the state, Manikis’ (2022) framework of state blame will serve as the foundation for the proposed sentencing review. Its accompanying typology of state wrongdoing/harms, articulated below, will provide relevant analytical tools for recognising and addressing state blame and harms in the context of SIUs.

Typology of state wrongdoing/harms

As part of the framework of ‘state blame’, which recognises the importance of communicating blame as relational, Manikis (2022) has developed a framework to conceptualise the various types of wrongdoings and related harms by the state. This typology can be useful to communicate state blame at sentencing but also during the sentencing review process described in the next section.

Manikis’ (2022) typology of state blame and harms is divided into four categories. The first type recognises the state's ‘systemic criminogenic contribution by the creation and maintenance of societal inequalities’ (Manikis, 2022: 308). This first type of state blame comprises state policies that foster inequalities which themselves create criminogenic conditions, meaning conditions that increase the likelihood of crimes being committed (Manikis, 2022: 308). According to Tadros, the state is responsible through its complicity in creating economic inequalities that promote criminogenic conditions (Manikis, 2022: 308, citing Tadros, 2009: 400). Such inequalities include those like state colonialism, patriarchal structures, racism, and ableism that targets persons with mental health and substance abuse issues (Manikis, 2022: 308). In applying this first typology type, courts should analyze, separately from individual desert, the state's contribution to creating criminogenic inequalities.

Manikis’ (2022: 310) second type of state blame and harms is the state's ‘systemic and criminogenic contribution through criminalisation policies and practices that target or affect marginalised groups’. This typology includes both instances where the state's criminalisation policies and practices directly and intentionally target individuals of marginalised groups, as well as instances where the state knows these are more likely to criminalise and affect individuals of these groups (Manikis, 2022: 310). In this first scenario, this includes, for example, the criminalization of refugees due to migration policies, while the second scenario includes phenomena like over-policing and over-prosecuting (Manikis, 2022: 311, citing National Council of Welfare (Canada), 2000). A sentence could thus consider these criminal state policies and practices that help understand the context surrounding the commission of the offence. This typology type would also communicate the state's relational contribution to the offence (Manikis, 2022: 313).

Manikis’ (2022: 313) third type is the ‘state's contribution in producing harms that result from the sentence’. These harms can be foreseeable or known to the state during the rendering of a sentence but are not intended to be part of the legitimate or justified sentence (Manikis, 2022: 313). Harms within this typology may also be unforeseeable at the time of sentencing and only appear or take place during the administration of the sentence. In current sentencing theory and practice they are generally referred to as ‘collateral consequences’ of punishment and include a wide range of harms that exceed what was decided as justified punishment. Examples of such harms include civil disqualifications suffered by the offender as a result of the sentence, harms relating to the dangerous conditions within prisons, and the harms relating to the type of management within the prison, including the level of restrictions, that might disproportionately affect certain groups of individuals (Office of the Correctional Investigator, 2014). Indeed, according to H.L.A. Hart, among the necessary elements of a legitimate punishment is the intentional administering of unpleasant consequences to an offender by the state in response to a crime committed (Manikis, 2022: 314, citing Hart, 1959–60). Indeed, the individual has a prima facie moral right not to be harmed by the state and therefore any punishment needs to be intended and justified to be considered legitimate (Manikis, 2022: 315, citing Bülow, 2014: 20, and Zedner, 2016: 20). Thus, according to Manikis, (2022) harms excluded from the legitimate sentence are illegitimate additional sentencing harms. An application of this third typology would require the state to acknowledge the illegitimate harms it caused the offender and provide justifications and responses for these harms (Manikis 2022: 315). It may be that these harms risk taking place during the administration of the sentence but that their actual commission remains unforeseeable during the time of sentence. For those reasons, a review of the sentence, as the one proposed below, would be warranted to adequately calibrate and legitimate the actual sentence.

Manikis’ (2022: 316) fourth and final typology type consists of ‘state blame and harms that stem from abusive state actors that disregard fundamental human rights’. This includes violations of rights relating to detention, arrest, search and seizure, abusive conditions of pre-trial detention, and the right to counsel. Manikis (2022: 316) notes that while Canadian courts have partially recognised these types of harms at sentencing, they are usually analysed solely in relation to the offender's punishment rather than through a separate analysis that focuses on the state's responsibility. Arguably, such abuses can also happen during the administration of the sentence and therefore a review process to respond to these state wrongdoings would be warranted.

Manikis’ (2022) typology of state blame and harm will be necessary in analyzing the state's responsibility in causing harms to offenders while in SIUs. Specifically, these harms can be introduced within the implementation of the proposed sentence review process discussed in the next section.

Sentence review for solitary confinement (SIU)-related harms

The following section proposes the development of a novel sentence review mechanism to better examine, acknowledge, communicate censure, and respond to state harms caused in the context of SIUs. This process can replace the current deficient legislated review processes for offenders in SIUs by requiring an automatic sentence review to assess the legality of this detention and degrees of excess harms created by the state. This process will aim to be less bureaucratic and more expedient in contexts of offender placements in an SIU, assessing their conditions, and whether correctional staff took reasonable steps to offer the offender the legislatively-required hours of meaningful human contact and time outside of the SIU. Importantly, the court will qualify the state actions and harms according to Manikis’ (2022) typology and alter the offender's carceral sentence based on this analysis to reflect a more adequate level of censure.

This review process builds on dynamic censure articulated by Roberts and Dagan's (2019) ‘second look’ as adapted by Manikis’ (2022) theory of state blame which suggests shifting the focus onto the state's responsibility in creating harms. Accordingly, unlike Roberts and Dagan (2019), these proposed review processes do not examine the offender's positive behaviour following a sentence but rather focus on the state's contribution to harms experienced by the offender during the administration of the sentence. Moreover, while the ‘second look’ review in theories of dynamic censure articulated by Roberts and Dagan (2019) is intended to occur with regards to lengthy sentences after years have passed (Manikis, 2022: 316), the sentence review proposed in this article is free from such constraints. Rather, it is intended to allow the court to review – as many times as necessary – the harms resulting from the use of SIUs by the state. Realistically, however, there may be limitations in allowing the court to review the sentence as many times as necessary, particularly with regards to the caseloads and court time to which this process might give rise. This position however may nevertheless encourage a more limited use of incarceration with alternative sentences, and more humane conditions within prisons to pre-empt or respond to reviews. Finally, this proposed review process centers the analysis around qualifying state harms and identifying those that exceed the legitimate punishment, regardless of the offender's moral blameworthiness in relation to the offence.

As discussed, since SIU placement causes harms to an offender, the court will engage in an SIU Harms Triggered Sentence Review. This review will occur in two stages. At the first stage, the court will qualify the state's harm according to Manikis’ (2022) typology for state harms and blame. This will allow the court to formally attribute blame to the state and measure the extent of harms it produced towards the offender. At the second stage, the court will determine how to alter the offender's sentence or rely on alternatives responses that take into consideration its evaluation of the state's harms.

Stage one: Characterising SIU harms caused by the state

As will be seen, some of the harms articulated in Manikis’ (2022) typology are more directly applicable to the context of SIUs. The following section will examine these harms from order of direct relevance.

The state's contribution in producing harms that may result from the sentence (third type of harm). Manikis’ (2022) third type of harm will necessarily apply in every SIU Harms Triggered Sentence Review. As previously explained, this typology refers to the state's contribution in creating ‘unintended harms’ that can result from a legitimate sentence but that are not officially part of the punishment and are therefore illegitimate (Manikis, 2022: 313). These additional harms can be foreseeable to the state at sentencing – especially in some institutions – but this is not always the case. In the case of SIUs, it is possible for there to be foreseeable as well as unforeseeable harms. Indeed, the degree of foreseeability at sentencing might vary from one institution to another where there is, for instance, a track record of using this method regularly. For example, in some cases it may be unforeseeable at the time of the original sentence that a person would develop a mental health condition caused by an SIU and that they would suffer from more restrictive conditions of liberty than that intended in the original sentence. Foreseeability might be more feasible in cases where the person already has a mental health condition and would be sent to a prison which uses SIUs frequently for disciplinary as well as protective purposes. In the context of SIU harms that were unforeseeable at sentencing and have occurred during the administration of the sentence, it would be important for a sentence review to respond and address these harms, as they would now be known by the state. Thus, the state has an ongoing responsibility to address SIU-related harms – harms resulting from the sentence – as part of its sentencing role and, as will be discussed below, would need to acknowledge, and respond to such harms through this proposed sentence review process.

Critics of this third category might argue that harms resulting from SIUs are legitimate when used to discipline an offender. While the harms of SIUs are excluded from an offender's initial carceral punishment, they could be seen as legitimate as the offender was potentially placed in an SIU as punishment for their problematic prison behaviour. This view is arguably flawed since to be legitimate, according to Hart (Manikis, 2022: 314, citing Hart, 1959-60: 4), punishment needs to be for an offence against legal rules, intentionally imposed by human beings, and imposed and administered by an authority constituted by a legal system against which the offence is committed. For the punishment to be legitimate, a separate sentencing assessment would need to be made by authorities who are recognised as legitimate bodies that are legally trained to craft appropriate sentences. Moreover, solitary confinement for the purposes of protecting other inmates or the prisoner from other inmates would also be considered illegitimate and would need to be part of a sentence review that considers the specific realities of the institution in which the person is serving the sentence, and the specific needs of the individual. Indeed, the harms resulting from SIUs – as described in most of the contexts discussed above – far outweigh any form of legitimacy in sanctioning misconduct and are ill-adapted to adequately respond to the situation.

The state's blame and harms that stem from abusive state actors that disregard fundamental human rights (fourth type of harm). Manikis’ (2022) fourth type can also apply in the context of an SIU Harms Triggered Sentence Review, notably when CSC staff place an offender in an SIU beyond the legislated limits of the CCRA. As explained above, Doob and Sprott (2020: 19) indicate that prolonged solitary confinement –for 15 consecutive days or more for 22 h a day without meaningful human contact – was occurring in 10% of SIU stays in federal penitentiaries. Thus, prolonged solitary confinement continues to occur even though this practice has been deemed in violation of sections 7 and 12 of the Canadian Charter and constitutes torture according to the Mandela Rules.

The state's criminogenic contribution by the creation and maintenance of societal inequalities (type one) and through policies and practices that target or affect marginalised groups (type two). Although Manikis’ (2022) first and second types of harms do not directly pertain to harms resulting from an offender's sentence, they are nonetheless relevant for the court to consider in its sentence review triggered by SIU harms.

A court application of Manikis’ (2022: 309) type one harms – the state's systemic criminogenic contribution by the creation and maintenance of societal inequalities – would account for the state's creation and maintenance of inequalities that are criminogenic. In the context of SIUs, the court could analyze how the state's use of SIUs might result in creating criminogenic conditions in prisons that are themselves rooted in social inequalities.

Similarly, Manikis’ type two harm – the state's systemic and criminogenic contribution through criminalisation policies and practices that target or affect marginalised groups – would account for the over-policing and over-prosecution of Indigenous and racialised offenders that have led to the over-representation of these groups in prisons. The court could consider documents such as the 2021-2022 Annual Report of the Office of the Correctional Investigator of Canada, which highlight the disproportionate over-incarceration of these groups in federal penitentiaries (OCI, 2022: 45).

These types of harms should also consider the perpetuation of state inequalities and criminalisation policies towards marginalised groups specifically within the context of SIUs. Indeed, data from the Office of the Federal Correctional Investigator illustrates that SIUs are disproportionately used against Indigenous and Black offenders. Specifically, persons of colour experienced more, and longer SIU stays than white persons. Furthermore, Black offenders were almost twice as likely as white offenders to be placed in an SIU and were more likely than other groups to experience SIU stays of 60 days or more. Moreover, Indigenous offenders were transferred to SIUs at a much higher rate and were more likely to experience SIU stays of 15 days or more, compared to non-Indigenous offenders (OCI, 2022: 17).

Thus, an application of Manikis’ (2022) first and second typology types of state harms and blame when performing an SIU Harms Triggered Sentence Review would be pertinent with respect to persons of colour, to account for the state's systemic blame vis-à-vis their incarceration. The qualification of the state's SIU harms through Manikis’ (2022) typology types during this first stage of the SIU Harms Triggered Sentence Review will set the table for the court's adjustment of an offender's sentence at the second stage of the review.

Stage two: Accounting for harm and adjusting the offender's sentence as well as wider responses

While the reduction in the length of a sentence given to an offender could be an adequate remedy to account for certain forms of state blame and harms, methodologies should not be reduced to this single approach (Manikis, 2022: 319). Indeed, Manikis (2022: 319) intends for her separate and complementary analysis of state harm and blame to be dialectical. This view is anchored in Duff's communicative penance theory which considers censure as dialectical between the State, the community, and the offender (Manikis 2022: 319).

After classifying and qualifying the severity and nature of SIU harms within the typology, the sentencing court could review an offender's carceral sentence and through this, respond to its wrongdoing for censure purposes. State harms that solely correspond to one typology type may not necessarily be considered less severe than harms that correspond to two or several types and therefore the qualification for communicative purposes would be more qualitative than quantitative. One response by the court would be to alter an offender's previous carceral sentence in nature or length. The court may apply other forms of responses as well.

Quantitative sentence reductions

Within an SIU Harms Triggered Sentence Review, the court could decide to reduce the number of days an offender serves in prison to account for the unjustifiable state harm they suffered while in an SIU. As previously discussed, such state harms will necessarily be classified under the third typology type, meaning harms resulting from an offender's sentence. If the state's SIU harms are classified under this category, an appropriate sentence reduction could be the deduction of one day from the offender's carceral sentence for each day spent in an SIU where correctional staff did not take all reasonable steps to provide the offender with the opportunity to receive two hours of meaningful human contact, as required by s. 36(1) of the CCRA. For example, if the court determines this occurred for five days, the offender's carceral sentence could be reduced by five days. Nevertheless, additional typology types could be applicable, which could warrant a different response from the court.

The court could apply a greater ratio for state SIU harms that are not only categorised under Manikis’ (2022) third typology type, but also the fourth – human rights abuses by state actors. Indeed, if the court determines the offender did not receive meaningful human contact while in an SIU for over 15 consecutive days, this constitutes prolonged solitary confinement. As previously discussed, prolonged solitary confinement was deemed unconstitutional by both the ONCA and the BCCA and constitutes torture pursuant to the Mandela Rules. Therefore, placing an offender in an SIU during this length is a human rights abuse on behalf of the state, and likely requires the application of a greater ratio in reducing the offender's carceral sentence. For example, if an offender experienced 20 days of prolonged solitary confinement, their carceral sentence could be reduced by 40 days.

State SIU harms that fall within all the typology types – meaning the third, the fourth, the first, and the second – could result in an even greater sentence reduction ratio. In other words, if an offender is Indigenous, and was subject to prolonged solitary confinement, the court could go beyond considering the application of only the third and fourth typology types. The court could also consider that the state's systemic criminogenic contribution through both its maintenance of social inequalities (typology type one) and its criminalization policies of marginalised groups (typology type two) compound state harms against Indigenous offenders. Thus, an Indigenous offender who was subject to prolonged solitary confinement in an SIU for 20 days, could see the reduction of their carceral sentence by 60 days.

Nevertheless, the application of varying ratios based on typology types may not be adequate in all sentence review instances. Certain types of harms or combination of harms may be so substantial that no reduction would be appropriate to convey the level of censure towards the state, and therefore the state would need to craft a different response that may not be considered punishment. Duff's (1988; 2010) concept of loss of standing to punish suggests there are instances where the state has egregiously failed to treat the offender as a citizen which removes standing and ability to blame and punish. Duff's account however is unclear about when such a stage is reached, and this paper is not meant to solve this matter.

Qualitative sentence adjustments. While the reduction of an offender's sentence may be an appropriate remedy to account for state related harms in certain instances, the court should not be limited to such responses. Indeed, Manikis warns that reviewing sentences solely with respect to the quantum of an offender's prison term neglects the full potential of state blame and censure (2022: 320). Therefore, a broader discussion could be undertaken by the court during the sentence review to examine whether a more legitimate sentence, that also addressed individual censure, can be crafted which respects dignity and needs. In so doing, the court could engage with the prison staff that interacted with the offender during their stay in the SIU – such as the prison guards, registered health care professionals, and any other relevant staff – and the offender themselves, to try to understand the conditions that gave rise to the offender's placement in an SIU and the unjustified harms that flowed from it. Perhaps some needs and contexts would be better met outside of prison and give rise to a different type of sentence that would also maintain the individual censuring function of the original sentence, such as conditional sentences. Indeed, the literature on some alternatives to incarceration has made clear these forms of punishment can adequately provide censure (Roberts, 2001; Roberts and von Hirsch, 1998).

Furthermore, during this discussion, if the court or other parties deem it relevant, experts could be called upon to provide further information relating to an offender's background, such as a psychologist, a Gladue report writer, or family members and friends of the offender. This discussion may lead to new findings vis-à-vis the offender that have arisen since the initial rendering of their sentence. For example, the court may find, through the technical expertise of a psychologist, and the testimony of the prison staff and the offender, that the offender has a serious mental illness, such as schizophrenia. Consequently, the court may order that the offender's sentence be altered by being served in a different institution that is better tailored to their needs.

Therefore, the second stage of an SIU Harms Triggered sentence review would utilise the qualification of the state's harms to alter an offender's carceral sentence. This sentence review should not be limited to reducing the quantum of the offender's sentence, but also consider broader adjustments, including some that recognise the state's loss of legitimacy to punish.

Conclusion

Despite the federal government purporting to have banned solitary confinement, reports have shown that this practice continues to be employed in Canadian penitentiaries under the name of Structured Intervention Units (SIUs). Moreover, these reports have highlighted that the review mechanisms added to the Corrections and Conditional Release Act to enable oversight of SIUs have been greatly deficient. Based on these deficiencies and the importance of conveying state blame and responsive censure, this article proposes a novel sentence review process as part of an expanded understanding of sentencing. This process is undertaken by sentencing judges as a review of the initial sentencing decision – enabling judges to order documents and evidence that examines the presence of state harms committed against the offender, their level of unjustifiability, and related responses.

Indeed, this novel sentence review would implement an automatic review process of the sentence to ensure adequate and dynamic censure throughout the administration of the sentence, and to account for state blame. Furthermore, this process makes clear that the review is part of the continuation of sentencing, rather than a separate and distinct process that falls to the hands of separate – and often unaccountable – state authorities. This proposed review is underpinned by Manikis’ (2022) framework of state blame which suggests that the state has continuous obligations to consider the ongoing censure of the individual but also needs to question its own censure. Manikis’ (2022) communicative framework allows for an expanded understanding of censure that provides analytical tools to conceptualise and understand harms that can be perpetrated by the state during the administration of the sentence and that need to be accounted for and responded to within the sentence.

The proposed review process is divided into two stages. The first stage of this review would qualify the state's blameworthiness using Manikis’ (2022) typology for state blame and harms. The second stage would implement Manikis’ (2022) typology by reviewing the offender's carceral sentence to account for the state's blame. Certain harms might be considered more serious and blameworthy than others – depending on intentionality and foreseeability. Accounting for such harms could take various forms of a sentence length reduction but also, and more importantly, a qualitative assessment of the appropriate sentence which can lead to a type of sentence that rejects imprisonment. Moreover, other sentence alterations should be considered by examining the offender's needs through a discussion with relevant actors. Finally, some instances of state blame would be considered so illegitimate and abusive that they result in the state's loss of legitimacy to punish and therefore responses would need to invoke other systems.

Overall, this article analyzes a specific problem in Canada – the use of SIUs and the state's lack of accountability – and provides a novel way to address this current issue. While the article has focused on SIUs as distinct types of harms that currently remain unaccounted for as part of a system that is considered separate to sentencing, we suggest that this original framework and analysis can also apply to other types of harms perpetrated by the state during the administration of the sentence. These harms are numerous and can take different forms, which we hope will be explored and discussed in future research.