Prison reform and torture prevention under ‘compromised circumstances’

Abstract

In this article, drawing on two decades studying prisons and prison reform practices in (mostly) southern countries undergoing transition, I examine the challenges facing anti-torture professionals and prison reformers working in the global south and critically interrogate the assumptions of dominant models of reform. Rights and health-based entry points to the prevention of torture and inhumane treatment and prison reform are argued to be necessary but insufficient. I propose the concept ‘compromised circumstances’ to counter the structural biases that diminish and erase ordinary everyday experience. The ‘compromised circumstances’ of countries torn by conflict, inequality, poverty and mundane violence call for innovative interventions based on reflexive social scientific description and analysis. The inevitable sense of dizziness and uncertainty such circumstances induce must be embraced not denied. A dynamic, organic and relational entry point to reform is required.

Ambitions, drivers and focus

Classic torture prevention aims to prevent the worst excesses associated with sites of institutional confinement. This article takes torture prevention as a specific kind of prison reform and adds to a growing body of critical scholarship that calls for more thorough contextualisations of the ordinary lives, situations and practices that are the subject of reform efforts. I identify and examine the ‘if only’ and ‘as if’ logics that commonly propel reform efforts with specific focus on the drive to transfer knowledge; the insistence that norms and standards be adhered to; and the promotion of professionalism. Put simply, I critically examine the belief that transformation is best achieved through new knowledge, better rules and more professionalism. To counter the structural biases that diminish and erase ordinary everyday experience, I propose the concept ‘compromised circumstances’ as a novel way of drawing attention to the oft-neglected contexts in which reforms take place. My ambition is that this new term might disrupt the all too easy acknowledgement that ‘of course context matters’ only for the business of reform to continue as usual.

The article is driven by theory, the field-based research of others and my own experience of conducting reflexive, ‘against-the-grain’ research on interventions (rights-based reform) and the sites of interventions (prisons) within a norm-driven activist organisation. It also draws on my experience participating in collaborative reform efforts with partners in the global south. It is inspired by the theorising of Jean Lave (2011) on situated learning and contentious social practice and the work of theorists of everyday life (e.g. Das, 2007, 2020; Povinelli, 2009; Segal, 2015, 2016; Stevenson, 2014) and further exemplifies the ecological approach to torture prevention developed by Danielle Celermajer (2018). ¹

Focus is on the way the compromised and compromising circumstances of prisons in the global south present tough challenges for reformers armed often with models and solutions designed in and for other contexts. Without awareness and recognition of such circumstances, with all their nuances and complexity, reform efforts will inevitably have limited effect and may even fail.

The article is structured as follows. After some further introductory remarks and a brief consideration of common forms of prison reform and torture prevention, I set the scene with an empirical example from my own experience accompanying detention monitors in Sierra Leone. I then elaborate on the concept of ‘compromised circumstances’ before considering the way that prisons and those who encounter them are always already compromised. Then I dismantle the three generic assumptions. In conclusion, I call for more synergy between critical, reflexive social scientific analysis and norm-driven activism. I suggest that exploring the potential of ‘anticipatory’ interventions that protect populations rendered imprisonable and torturable by structures of intersectional oppression might be a progressive move for anti-torture professionals and others made indignant by the pains of penality. ²

Abolitionism, reform and prison failure

The topic of the conference, for which this article was initially prepared, was ‘what kind of prisons for Haiti?’ By the end of the first day, I had an answer: fewer prisons, smaller prisons and prisons where fewer people are incarcerated. This, I felt, would be a step in the right direction and would clearly solve some of the problems of Haitian prisons: overcrowding, degrading conditions, violence and so on (see www.cresej.org). This answer falls within a tradition of critical or abolitionist criminology (e.g. Armstrong and Jefferson, 2017; Canning, 2014; Coyle and Scott, 2021; Drake, 2012; McMahon, 1992; Mathiesen, 1974; Ruggiero, 2010; Scott, 2013a, 2013b; Sim, 1994, 2009). Recognising the harms that prisons do to individuals and society, abolitionists wish to be rid of them. Prisons as compromised and compromising institutions embody and amplify dilemmas and contradictions in a manner that few other societal institutions do. Does the world need them? According to the abolitionists, no it does not. Does the world have them? Most certainly. And there lies the obligation and the quandary for critical prison scholars: to study and engage without sustaining the institution or reproducing the discourses upon which it is founded. This article embraces these quandaries while seeking to take seriously the obligation to engage.

Prisons in the global south fail to live up to the expectations of reformers and anti-torture professionals in at least two ways. First, in a normative sense, they fail to comply with norms and standards ³ agreed by the international community, and more significantly, they fail to match the expectations implicit within dominant models of intervention which assume relative stability (understood as the kind of stability associated with the ideal type of a liberal democratic state). The unstable, often politically volatile, conflict-filled contexts that frame imprisonment in the global south are the ‘compromised circumstances’ that confront reformers with a crucial challenge. This is a challenge they must overcome if reform is to be meaningful rather than imaginary; it is a challenge that calls for illumination if reformers are to cast off structural bias and see and hear in new ways. It is a challenge that calls for collective imagination and new synergies between reformers and researchers.

Torture prevention and prison reform

Two professions have played key roles in the anti-torture movement to date, namely, law and medicine. A reflective anti-torture professional I once interviewed shared with me his version of how the medical and legal professions, each in their own way, have played a role in the spread of the anti-torture movement, a typical pattern being doctors heading clinics for victims and lawyers setting up advocacy organisations and speaking vociferously on their behalf. It is hardly surprising that two such historically powerful disciplines have played a key role. As my colleague put it, ‘they are the two most authoritative languages that exist around these kinds of things: one is the authoritative language of the body, the other the authoritative language of the law’. ⁴ The medical paradigm has in fact informed more than the setting up of clinics, and the legal paradigm has done more than establishing advocacy organisations. Law and medicine powerfully inform common discourses of torture prevention and prison reform, most obviously perhaps through the promotion of legal safeguards, ⁵ on the one hand, and prevention of harm, ⁶ on the other. They have made important contributions. Nevertheless, the analysis in this article suggests that while these approaches are necessary, they are insufficient.

What are the more common types of prison reform and torture prevention? A brief glance at the priorities of Penal Reform International (PRI) gives some clues. ⁷ PRI’s 10 priorities can be meaningfully grouped under three categories: (1) focus on forms of punishment (alternatives to imprisonment; death penalty; life imprisonment; pre-trial detention); (2) focus on specific actors or targets of punishment, often referred to as ‘vulnerable’ populations ⁸ (women in the criminal justice system; justice for children); and (3) promotion of norms and standards, safeguards and best practice (global advocacy; prison conditions; torture prevention; rehabilitation and reintegration). This focus resembles the orientation of dominant models of torture prevention towards institutions, victims and laws. Carver and Handley’s (2016) analysis of whether torture prevention works operates with four main torture preventive mechanisms: (1) detention safeguards, (2) prosecution of perpetrators, (3) monitoring and (4) complaint mechanisms. This offers reasonable coverage of the most common initiatives representing ways of protecting people in custody, ways of holding states and state officials accountable, means through which to open institutions up to surveillance and inspection, and means through which victims can seek redress.

A tale from the field ⁹

In July 2018, I visited a police station in Sierra Leone’s capital Freetown with the advocacy organisation Prison Watch – Sierra Leone (PWSL). ¹⁰ We stood in a dark, miserable cell with 14 detainees. My colleagues and I watched as the detention monitors spoke with the detainees under the watchful eye of a police officer. We observed as they checked the records to see how many detainees were there, how long they had been there, what age they were and so on. The monitors listened patiently as detainees vied with each other for space to speak. The chief detention monitor described their mandate: not to judge guilt or innocence but to ensure they were not subject to torture or inhumane treatment. Our attention was drawn to grimy toilet facilities; detainees complained about the lack of food. There was talk of money having gone missing in the cell and a detainee having been beaten.

We left the cell and went upstairs to lobby the Officer in Charge (OiC). The detainees thanked us as we walked out of the cell. The chief detention monitor shared some concerns with the OiC. Before he had got very far, the officer interjected: ‘There’s no need to lecture on human rights – I know human rights but do you know what these people have done, do you know who they are?’ He then launched into a longer tirade that implied two key assumptions. First, human rights are not for everyone; second, if you have done wrong, you invite punishment. He showed no interest in the idea of innocent until proven guilty and no commitment to the judicial process. Rather, he seemed to see the role of his police station to be detection, capture and punishment, and his own role as sheriff and executioner. He spoke reasonably. He knew human rights. But he justified maltreatment on the grounds that it was deserved, and he cast doubt on the validity of the detainees’ allegations, rendering them a priori suspect individuals. From his perspective, the young men in the cells were members of a torturable, abusable or imprisonable population. He failed to register any contradiction between his ‘knowledge’ of human rights and his attitude and behaviour towards the detainees in his custody.

This is a striking example of what detention monitors face in Sierra Leone. Given his position as the OiC of the police station, it is easy to imagine that torture and/or inhumane treatment could be common in the cells despite the close attention of the monitors. This example is from police cells; it could just as well have been a prison. If common institutional practices associated with policing and prisons are to be transformed, it is necessary to understand such dynamics. We need to explore how come ill-treatment persists despite knowledge of human rights and the presence of monitors. It is noteworthy that the monitors from PWSL are particularly attentive to whether basic safeguards are respected by the police, that is, to whether detainees are charged or released within the 3- and 10-day limits prescribed by law, to whether juveniles are held separately from adults, to whether families are informed about the detention of the detainees and so on. This is the type of activity that Carver and Handley’s (2016) study concludes would be a good idea. However, it is important to notice that even in the presence of legal safeguards and monitoring, torture and ill-treatment persist. I will return to this point later and to this tale from the field.

Acknowledging interference: Towards new questions from new points of departure

My concern is not with whether the interventions of Prison Watch work (though I believe they do, in fact, on a case-by-case basis play an important role in rescuing some people from abusive situations). The question I believe to be important is not does prevention work but how does it work under ‘compromised circumstances’. Under what conditions does it work more or less well? For whom does it work best or worst? Any attempt to understand what works needs complementing by understandings of what probably does not work ¹¹ and, more importantly, understandings of the circumstances under which torture and imprisonment take place and the circumstances under which torture prevention and prison reform take place. Asking whether something works exhibits a concern with effectiveness rather than meaning. These should not be divorced from one another. The aforementioned study by Carver and Handley operates with an overly legalistic and narrow conceptualisation of torture where torture is seen primarily as a rule violation (see Rejali, 2020) and inhumane, cruel and degrading treatment is bracketed off. ¹² An ethnographic orientation approaches torture as an intersubjective, political, relational, institutional practice that demands an organic, dynamic, multi-pronged response. Similarly, an ethnographic approach to imprisonment and prison reform demands that we take as point of departure the everyday dilemma and contradiction-filled worlds of prisons, penal practices and state officials. Carver and Handley’s study implies that it is possible to bracket out the messiness of the world and its disturbing ability to ‘interfere’ with how life ought to be and prevention ought to work. In reality, there is almost always interference. There is almost always a kind of white noise that reformers must deal with, for example, the attitudes of the OiC above. Carver and Handley’s analysis reveals the ‘if only’ logic behind common prevention initiatives: if only we had the right laws in place; if only officials knew about the laws; if only they acted in accordance with the laws; if only circumstances were different. The problem is that circumstances are not different. Circumstances are compromised – by conflict, by poverty, by inequality and so on. As Rejali puts it,

It’s true: when people are good and institutions are sound, torture will cease. All we have to do is get there . . . The problem is what to do when people are bad, institutions broken, understaffed, and corrupt, and habitual serial violence is routine. (Rejali, 2020)

Rejali should not be understood too literally here. His point is it cannot be about the morality of people (goodness/badness) or the soundness of institutions because people and institutions are never perfect but flawed from the outset. Our point of departure must be social and institutional life as it is rather than as ought to be. This is a broader, more organic entry point, a point also made by Kelly et al. (2020) who forcefully call for attention to be paid to ‘other norms, expectations and relationships’ than those associated with the reform of laws and public institutions. Noting that torture does not simply disappear with the rise of democracy and rule of law, they flag the need to explore factors that drive torture that are connected to corruption, extra-custodial violence and contestations around public authority.

Compromised circumstances

Compromise is usually conceived of as a good thing. Its original Latin meaning alludes to the idea of a mutual promise, an agreement reached by two disagreeing parties, something that keeps everyone more or less happy. When siblings squabble, we encourage them to compromise; when factions of political parties must come together for the common good, compromise is necessary. I am using the term with a more negative twist. Compromising is also about accepting a reality that does not quite live up to one’s ideals; a compromise is not necessarily an ideal outcome for either party. It is something arrived at through necessity rather than desire.

I have come to the concept via Lauren Berlant’s (2006) suggestive phrase ‘zone of compromised ordinariness’, and the notion of ordinariness is also important to the concept of ‘compromised circumstances’. Scholars attentive to the ordinary to whom I am indebted include Berlant (2006, 2011), Das (2007, 2020), Povinelli (2009), Segal (2015, 2016) and Stevenson (2014), all of whom are concerned one way or another with social suffering. What is ordinariness or the ordinary? In her poignant analysis of the lives of Palestinian women whose husbands are in Israeli prisons, anthropologist Lotte Buch Segal (2016) deliberately refuses to define the ordinary in advance of her analysis. It is the ordinary that belongs to and imbues the lives of the wives of imprisoned Palestinians that concerns her. It is their ordinary which she privileges. Similarly, I am advocating that reformers refuse to predefine but nevertheless take seriously the ordinary ‘compromised circumstances’ of the institutions and actors who are subject to reform as they encounter them. Acknowledgement of their dilemmas, conditions of possibility and sustaining forces are necessary if structural bias is to be overcome. This means attending to the drivers of both victimisation and perpetration.

The notion of ‘compromised circumstances’ is also connected to the notion of suffering as a quality of life itself. Under conditions of judicial limbo, prolonged injustice and human suffering that characterise criminal justice practices in poor countries such as Sierra Leone, suffering is not an aspect of lives, not an occasional feature, not an abnormality but an endemic, defining and constitutive quality. Suffering is not separate from life but constitutive of life itself (Jefferson, 2014). In the same way, ‘compromised circumstances’ are constitutive of prison life and reformist responses to prison life.

As alluded to above, Kelly et al. (2020) are also occupied with the question of how the common practices of torture prevention and the assumptions underlying them play out under less-than-ideal conditions. What happens, they ask, when the state has limited and inconsistent reach, when state and non-state coercion is routine and where the liberal instruments of authority are under-prioritised by governments? Complementing these important questions, Segal draws attention to more quotidian forms of life that compromise, that is, ‘forms of life that never make the headlines’ (Segal, 2016: 167), and ‘the slow grind of violence that is not spectacularly catastrophic’ (Segal, 2016: 16). Her concern is with ‘what it means to live with violence at your front door as a permanent feature of life rather than as an occasional, discrete occurrence’ (Segal, 2016: 10). Ethnographies like those that feature in the work of Kelly et al. and Segal invite us behind the scenes and remind us of the ordinariness of suffering that endures and fatigues.

Occupying a similar analytical landscape, Danielle Celermajer’s book on ecologies of torture (Celermajer, 2018) offers a holistic, organic approach that is inclusive of the practices that Carver and Handley’s study deliberately brackets off. ‘Poisoned orchards’, as she would have liked to call it, is an insistent book that recognises that the struggle to prevent torture (and by extension transform prisons) is a shared journey that demands imagination, experimentation and a deep grounding in the world of practice. The main argument is that we must examine more thoroughly and systematically the factors that ‘produce and sustain’ torture if we are to be able to develop effective strategies to counter it. It is a book that grapples with complex questions related to old debates about the relation between structure and agency that might feel far removed from the everyday business of preventing torture or transforming prisons but which are absolutely necessary for understanding why some strategies can be expected to be more impactful than others. For example, and most fundamentally, If we know the practice of torture is firmly embedded in structures and systems, why have interventions so often privileged changing the behaviour of individuals? And why do we naively trust that instructing people about what they are believed not to know or changing the rules that supposedly govern their practice will be transformative?

Some circumstances that compromise are general, pertaining to geo-political conditions, global inequalities, structural dynamics, institutional capacity, organisational set up and so on; others are specific, pertaining to available forms of life, misery and deprivation in local context. In many situations, circumstances do not live up to the ideal type conditions under which reform strategies are thought most likely to work. Embedded, embodied, historical practices of mundane, everyday violence under conditions of poverty and socio-political volatility are compromising. But prisons are not only compromised by surrounding circumstances, they are also compromised by virtue of their mandate to deliver punishment and pain (Carlen, 1994; Scott, 2013b). That is, even before we get to concrete conditions and specific practices of imprisonment, we find that prisons are fundamentally compromised by virtue of their paradoxical purpose. Pretty much universally, prisons claim to both punish and reform. The drive to correct and reform through the delivery of pain and punishment is compromising. The early 20th-entury British prison commissioner Alexander Paterson famously identified the dilemma when he asked whether people can be prepared for freedom under conditions of captivity. Prisons are arguably compromised in other ways too. For example, they rarely reduce recidivism; they rarely deter criminals; they rarely protect the public. Furthermore, they dehumanise even as they seek to create law-abiding humans and, while ostensibly serving the interests of justice, act on populations unfairly and disproportionately (e.g. minorities, the poor).

Identifying and dismantling common assumptions

Kelly (2019) has critically examined the way anti-torture professionals are overly preoccupied with sites of interrogation and detention, political prisoners as ‘archetypal victims’, legal reform as a necessary foundation of any kind of change and the monitoring of compliance to norms. This reflects forms of structural blindness or bias (see later). My focus is on the commitment of reformers to introduce new (often moral) knowledge, new rules and new forms of professionalism as a way of promoting change.

Through encounters with prison officer recruits in training schools in Nigeria and prison staff engaged in everyday prison practice, I learned that the kind of knowledge with which they were targeted by external human rights agencies (norms-based, manualised, assumed universal and transferable) was a poor match for what they otherwise learned. What they learned during the 6-month residential training programme was most fundamentally about their position at the bottom of the hierarchy. What they took with them into their future working lives was embodied experience – inculcated via the disciplined, para-militarised practice of learning how to march and conduct themselves in relation to superiors – of how those in such a position should be treated, that is, relatively harshly. The fundamental importance of hierarchy was underappreciated by reformers who, for example, instructed middle-ranking staff to go back to their duty stations and advocate for change based on their new knowledge of human rights norms acquired through lectures and role-plays. This approach failed to recognise their relatively disempowered positions and overestimated the power of new knowledge.

This is not an argument against knowledge. The question is what kind of knowledge is assumed to make a difference and how can it best become embedded in changed practices. Simplistic knowledge transfer models that imagine knowledge as a package transferred from one mind to another are ill-conceived. Reform needs to include the development of embodied knowledge of how-to-do as well as information on what is prescribed and prohibited.

A similar argument can be made about the common desire of reformers to change the rules governing practice, be these hard laws or soft laws. Increasingly, human rights practitioners are recognising the need to get closer to everyday practice. The revision of the standard minimum rules for the treatment of prisoners (the Mandela Rules) is a case in point, yet interventions still tend to privilege awareness-raising about rules rather than on the ground discussions of the implications of rules for everyday prison practice. In some jurisdictions, the push to change the rules can be so fast that nothing meaningful gets changed. For example, the Sierra Leone Correctional Service Act (2014) changed very little except the name of prisons to correctional centres and the name of prison officers to correctional officers. On a more positive note, the Act did empower a correctional council to review the prison rules although the first draft of these rules left much to be desired. But even once rules are in place, few reformers have the necessary trust or basic knowledge of prison practice to advise on how to implement them or what they might mean in practice. A related issue arises when we look at countries transitioning from authoritarian to democratic regimes. Under such circumstances, the whole governance framework alters. State institutions are expected to follow suit. New forms of authority are expected to emerge and be enacted, but these take time to embed in governance structures, society and especially in penal institutions that are arguably more suited to authoritarian logics than democratic ones. The issue of how to democratise a prison has received little attention among reformers. ¹³ This means that while societies and politics might change during transitions, prisons might not quite keep up. Militarised, securitised and punitive discourses and practices are likely to persist and will form part of the conditions of possibility that effect the success of any reform intervention.

A third common assumption is that enhancing professionalism helps. The question is what is meant by professionalism and how do appeals to professionalism travel. Professionalism and appeals to professionalism are means of changing working cultures, but they risk falling into the trap of the ‘if only’ thinking alluded to earlier: if only staff were professional, the prison would be a better place. This position refuses to consider the conditions under which prison staff are obliged to perform their duties. It fails to take point of departure in the local conditions of possibility for professionalism and thereby runs the risk of prescribing rather than facilitating. For example, health professionals in prisons are typically assumed to be potential allies of reformers and health is a common pragmatic and tactical entry point. This is partly because health can be appealed to as a universally agreed-upon good that health professionals are assumed to be universally committed to upholding – also those working in prisons. But health professionals in prisons are often compromised in all sorts of ways: by their quality and motivations; by the tools and facilities at their disposal; by dependencies on prisoners to provide or filter care; by their position in institutional power dynamics; and by the fact that within punitively oriented institutions the sick body can very easily become a more or less deliberate object of punishment. It is easier to let die than to kill, but the effect is the same. Appeals to health professionals to act professionally because it is their duty by virtue of medical ethics ignore the fact that prisons, by virtue of their purpose, easily risk undermining medical ethics, and prison health staff are inevitably caught up within these dynamics, either fighting against them or collusive with them.

By way of further exemplification, let us revert to the OiC of the police station in Sierra Leone. How receptive might we imagine he would have been to new knowledge, new rules and enhanced levels of professionalism? We have already seen that he sees little value in the kind of knowledge he feels is on offer: don’t come to me lecturing on human rights, he says. He explicitly claims to ‘already know’ human rights, and faced daily with members of a troublesome, imprisonable population, he does not find human rights particularly conducive to helping him deal with them. They might empower the local non-governmental organisation (NGO) and potentially empower those with a grievance or simply a desire not to be in the cell, but they do not offer him much, at least not in the short term. In fact, we might speculate that human rights are perceived to be an obstacle to him carrying out his perceived mission as judge, jury and executioner (see Martin, 2009). Of course, one could argue that he lacks knowledge of principles of due process, innocent until proven guilty and so on, and this could feasibly be part of a human rights training package, but in the absence of a pedagogy that took into account his specific circumstances, even attempts to offer him new ways of making sense of his role and position would likely have limited success. The kind of knowledge he needs is about how the habitual practices of the police station might be transformed. This involves a shift in standard pedagogical assumptions focused on attitudes and mentality towards approaches that highlight practice and possibilities.

What about new rules or procedures? Might these not allow for or encourage changed practices at the institutional level. Possibly, but this would ignore the fact that legal safeguards were in place. There are local rules that govern who and how long people may be held in the police cells. It is exactly such situations that Celermajer identifies as potent sources of new understandings about the challenges facing reformers. In her words,

It is here, in closely tracking what actually happens in concrete situations where torture persists even in the face of the existence of formal laws mandating the correct conditions of detention, that we can begin to come close to understanding the types of factors that keep in place problematic detention practices, or that impede their reform. (Celermajer, 2018: 94)

Reasons why rules are not always fully adhered to might on occasion be malicious; they might equally be a product of over-pressured, under-resourced working environments, basic oversights or inattentiveness, or rooted in understandings of imprisonable populations as deserving of maltreatment. The issue is less lack of rules and more the inability of the rules to stick, that is, to have purchase and seem relevant under the conditions facing individual police and prison officers. The structural dynamics and relations that render members of certain communities – the poor, the hopeless, the stuck – as worthy of punishment have a tendency to trump rules anyway. Policing, or the exercise of authority, we might say, like other institutionalised practices takes place despite, not because of the rules.

All this suggests reformers should promote knowledge of how to do rather than what to think and focus on the implementation of rules rather than (or at least as well as) the creation of new rules. This might resemble a new appeal to professionalism. However, what must be remembered is that it only makes sense to call for professionalism if the norms, values, and codes of conduct implied by professionalism make sense in situ. As we have already considered, appealing to the assumed professionalism of health workers in prison might not make sense under conditions where the prisoner’s sick body becomes itself complicit with a punitive apparatus in delivering further pain. Appealing to the professionalism of the police officer in the example I have presented would likely have been to ask him to ignore his frustrations and dilemmas and the meaning he attaches to working in the way he feels obliged to under the circumstances. This is not to excuse conduct that fails to live up to local or international norms. But the norms associated with professionalism cannot be assumed to be the same everywhere for all time.

The above discussion makes clear that the knowledge, rules and professionalism imagined to be required are too often defined in terms of the perspective of the reform-minded outsider looking in. This deficit-oriented perspective defines the state official in terms of what he is perceived to lack in terms of mentality, norms of conduct (morality), values and skills rather than in terms of the specific practices within which she or he is caught up. In fact, she or he already ‘knows’; the rules are often already there, and in their own understanding they are consummate professionals.

Seductive but not fit for travel

The assumptions considered so far are all assumptions one or more steps removed from everyday practice. Implied by each of these assumptions is an idea that modes of intervention are universally applicable and (easily) transferable. But we cannot assume that because a model works under certain circumstances (perhaps those within which the model was designed?) it will travel well to other contexts. As Kelly et al. (2020) put it,

(W)e should not assume that the institutional practices associated with a minority of high-income liberal-democracies can simply be transferred to the world’s poorest countries . . . (W)hilst the prohibition of torture itself might be a near universal aspiration, the institutional arrangements and political relationship that both provide the conditions for torture and its prevention are not the same all over the world.

Interestingly, it seems that the more generic the language, the more authoritative the claim being made for transferability sounds. Inversely, the more specifically tailored the intervention, the less relevant it sounds for anyone outside of the specific context. This is a structural factor impeding the take-up of practice-oriented, context-specific interventions that deserve further attention. Creative ways to resist this structuring, impeding dynamic are necessary.

Often, culture or attitude is blamed for torture or poor prison practice. As Celermajer (2018) puts it, culture or attitude is appealed to ‘as explanatory devices that give the appearance of telling us what is going on, but, as generally used and parsed, provide no operationalizable guidance’ (p. 95). They are also faulty and incomplete explanations. The situational factors that ‘impede the practical implementation of detention-related measures’ (Celermajer, 2018: 94) are part of what I am calling ‘compromised circumstances’. For Celermajer (2018), these include,

(B)ureaucratic inertia, the tendency to continue to tread the well-worn path, the dominance of certain disciplinary or professional perspectives, selective expertise, the hegemony of particular approaches, and even the pleasures associated with adopting certain types of strategies and the ethical positions they entail . . . (p. 96)

When not considered, these factors get in the way. Reform is blocked. And the blockage cannot simply be wished away through fantasies of ‘if only’.

Deficit-oriented assumptions deny the significance of local context and the structural and systemic factors that impinge on reform and give rise to the dynamics that are the target of change. We might call this structural blindness or preferably structural bias. ¹⁴ Lim ¹⁵ defines structural blindness as ‘the inability of people to see something, not because of a defect in their eyes, but because of a long-standing feature of the world’. For zemiologist Victoria Canning, structural blindness is about being ‘blind to structural mechanisms of violence and inequality etc., and embedded blindness of the same within structures and institutions’ (Canning, personal communication). Structural blindness/bias thus refers to the invisibilising effects brought about by dominant, hegemonic and often universalizing discourses and practices. The taken for granted, the obvious, and the unquestioned are products of structural bias, that is, an inability to fully understand that is called forth by the very structures that most demand scrutiny. To accuse reformers of structural bias is not to blame them but to direct attention to the need to decipher the dynamics through which invisibilisation takes place.

Another way to think about structural bias is in terms of dominant languages and an inability to listen effectively. Segal examines how the dominant discourses available to make sense of suffering and political struggle in Palestine are ill-equipped to sense the everyday ordinary experiences facing the wives of imprisoned Palestinians. The ‘compromising circumstances’ in which women find themselves do not lend themselves to the available languages of loss and mourning. In the same way, the ‘compromised circumstances’ of occupants of prisons in politically charged, poor, violence-effected countries do not lend themselves to the standard, easily available languages of reform. Dominant languages limit what can be put into words and what can be responded to. They limit the ability to listen and acknowledge. ‘Compromised circumstances’ evade easy rendition in language. It is perhaps not surprising that dominant languages are unable to render ‘compromised circumstances’ legible since were they to do so it would reveal their own frailty. To paraphrase Segal, to acknowledge the compromised nature of the world is to acknowledge one’s own limits.

Another way prisons in particular compromise is by drawing multiple actors into their carceral web. Occupants of prisons are moulded by the institution and its driving logic as are intervening and reforming agencies. In a study of human rights NGOs’ encounters with prisons in Sierra Leone, the Philippines and Kosovo, Jefferson and Gaborit (2015) discovered that the variation in approaches was more or less determined by the prison systems NGOs were engaging with. Reform agents are not independent but in significant ways in thrall to the institutions they seek to change. Reformers must therefore be aware of the risks associated with structural bias but also of the effects of their ‘capture’ by the institutions they seek to transform. To be effectively ‘held hostage’ makes it risky to engage. When there is no external morally superior position from which to meaningfully speak, the stakes are raised. The potential to be simply window dressing at best and complicit at worst is very real. But these are the conditions of the real world of prison reform and torture prevention.

Embracing uncertainty, discomfort and imbalance

The work of anthropologist Danilyn Rutherford (2012) on ‘kinky empiricism’ is instructive about the dilemmas and almost impossible balancing act required of reformers. ‘Uncertainty and justice’ she writes, ‘go hand in hand in those moments that force us to choose among contending ways of doing the right thing’ (p. 473). Rutherford (2012) advances a kind of social science that is ‘always slightly off kilter, always aware of the slipperiness of its grounds and of the difficulty of adequately responding to the ethical demands spurned by its methods’ (p. 466). The knowledge generated through fieldwork compels ‘those who seek knowledge to put themselves on the line’. Similarly, the obligations created by engaging with prisons and practices of torture are ultimately unfulfillable and the reformer is likely to have to live with a constant sense of gnawing dissatisfaction and discomfort. The reflexivity inherent in good ethnographic practice, what Rutherford (2012) identifies as a second dimension of kinky empiricism, ‘eats away at certainty as well as good conscience’ (p. 468). The same could be said of sound reform practice. This may well leave reformers feeling dizzy. Rutherford argues that one way of embracing and inhabiting ethical quandaries, rather than disowning or denying them, is by building alliances. Alliances will likely also help ameliorate dizziness. Reformers should therefore embrace quandaries and qualms ¹⁶ from positions within shared communities of practice.

Conclusion – a further call for pre-emptive collaborative action

We seem to be fairly good at knowing what needs to be changed, but less successful at knowing how to bring about those changes . . . Critique is all too easy and coming up with constructive strategies for transforming entrenched practices and norms all too difficult. (Celermajer, 2018: 96)

Ideal conditions for the implementation of preventive mechanisms, measures or pathways do not exist, or if they do, they exist in places where the need for them is least. The challenges facing prison reformers and those committed to torture prevention are considerable. This is not helped by the fact that the prison has a remarkably fixed position in the public, political and media imagination. As Scott and Codd (2010) put it,

The great triumph of the prison is to have made itself indispensable in the public imagination, rendering it difficult to conceive of a world without it. It has become entrenched and embedded in our society to such an extent that its continued existence seems beyond challenge. We need to remember, however, that it is possible to challenge the prison persistently, with determination and diligence. (p. 170)

The framing of an issue matters, claim Scott and Codd. This is especially important when issues are framed as implying universality or inherent goodness or when actions are seen to be informed by noble intentions. Under such circumstances, the possibilities to question, challenge or problematise are reduced and the status quo persists. It is imperative to question prisons – their role, their function, their grounds for existing – but also for societies to question their attitude to punishment more generally. ¹⁷ For critical criminologists like Scott and Codd, this is a natural outworking of a prison abolitionist perspective that calls not only for the abolition of prisons but also for a reordering of society along lines more attuned to justice, solidarity and equality.

Research conducted in and around prisons in the global south presents a harsh and troubling image of the circumstances confronting would-be reformers. Climates of misery, political contestation, secrecy, varying degrees of deprivation, stuckness, exhaustion and foreboding are a poor match for the imaginary abstractions that so often inform reform efforts. Reminders about such climates disrupt, complicate and interfere with default ways of working. But faced with embedded practices of mundane violence and punishment under conditions of poverty, volatility and conflict, we should not be surprised if standard modes of intervention do not work as effectively as they are imagined to under conditions of liberal democratic peace. While rights and health-based entry points to the prevention of torture and inhumane treatment (and prison reform more generally) plausibly make sense under certain optimal circumstances, when confronted by the ‘compromised circumstances’ of many prison climates in the global south, something else may be called for. What that something is remains an open question that calls for collective imagination.

Looking ahead, it is incumbent on activist and researcher communities to work more effectively together using multiple entry points. It is necessary to examine more carefully the complex drivers as well as the consequences of abuse in sites of institutional confinement and beyond, taking seriously the inevitably ‘compromised circumstances’. Pre-emptive, anticipatory interventions targeting socio-economic conditions and inequalities and identifying, visibilising and protecting imprisonable and torturable populations may be an innovative direction. And likewise, a focus on social actors and the relationships that keep them torturable or imprisonable. In sum a dynamic, relational entry point to reform is desirable if entrenched practices and norms are to be transformed.