Authoritarian exclusion and laissez-faire inclusion: Comparing the punishment of men convicted of sex offenses in England & Wales and Norway*

Comparative penologists have described neoliberal and social democratic jurisdictions as though they exist at opposite ends of a continuum of inclusion and exclusion, and as though neoliberal states are inactive and social democratic states are invasive. This article, which is based on more than 129 interviews with men convicted of sex offenses in England & Wales and


K E Y W O R D S
comparative penology, exclusion, inclusion, Nordic exceptionalism, offender rehabilitation, sex offender imprisonment The history of social control can be told in many ways, and one way would be to rewrite it as a choice between exclusion and inclusion. (Cohen, 1985, p. 266) There is a familiar history of penal change in the Global North, and it is a story of increasing exclusion. The early-to-mid-twentieth century was shaped by social democracy and penalwelfarism (Garland, 1990), or so the story goes, and penal practitioners sought to rehabilitate wrongdoers and thereby bring them (back) into the law-abiding mainstream. In the last halfcentury, as neoliberalism has grown in significance and as discourses of individual responsibility have become ubiquitous, practitioners, politicians, and the public have lost faith in the effectiveness and appropriateness of rehabilitation (Allen, 1981), and prison populations have boomed as lawmakers punish and incapacitate increasingly large numbers of people. This story of degradation and descent is sometimes told alongside another, which is presented as a story of hope (Pratt, in press): that penal moderation and the rehabilitative ideal persist and can be found in Scandinavia and other social democratic countries (Cavadino & Dignan, 2006;Pratt, 2008aPratt, , 2008b. Both stories are built on a series of moralized binaries-social democracy versus neoliberalism, rehabilitation versus punishment-but the least definitionally specific and most evaluatively weighted is inclusion versus exclusion. The purpose of this article is to explore and complicate this binary to tell both stories more precisely. To do this, we will engage with writing on political economy, Nordic Exceptionalism (Pratt, 2008a(Pratt, , 2008b, and rehabilitation, and we will describe the experiences of men convicted of sex offenses in England & Wales and in Norway. By foregrounding the perspective of the punished and connecting these ethnographic analyses to contemporary debates on punishment regimes, the article offers an innovative contribution to the field of comparative penology. Inclusion and exclusion are often described as both cause and consequence of penal difference: Societies that are inclusionary-which believe that even deviants are people like us-deal with crime as a social problem and ensure that people who have been punished "find themselves in the same or better individual and social conditions to voluntarily conduct their lives in accordance with the law" (Díez-Ripollés, 2013, p. 66), whereas exclusionary societies are built on a "criminology of the other" (Garland, 2001, p. 184), insisting that wrongdoers are fundamentally different and treating them in ways that reinforce this difference. A significant body of comparative research has supported the distinction between punitive exclusionary societies and mild inclusionary ones. Researchers have empirically demonstrated correlations between rates of imprisonment and levels of welfare spending and/or social democratic systems of government, and they have interpreted this by arguing that some (social democratic) countries enact social control through inclusionary mechanisms like welfare spending, whereas other (neoliberal) countries rely on exclusionary mechanisms like the criminal legal system (Beckett & Western, 2001;Cavadino & Dignan, 2006;Díez-Ripollés, 2013;Downes & Hansen, 2006;Greenberg, 2001).
What this literature makes less clear, however, is what inclusion and exclusion look like and how they are experienced. There is a yawning gulf between abstract nouns and specific policies and another between policies and lived experience. As Stanley Cohen (1985) convincingly argued, inclusion and exclusion are often described as more distinct than they are in practice, for three key reasons. First, what is intended by a policy is not always what it achieves, and "reforms motivated by the inclusionary impulse often end up being exclusionary" (p. 219) in their effects-by subjecting people to forms of intervention and classification, for instance, or by bringing in reforms that expand the "correctional continuum" (p. 57). It is even possible to see imprisonment itselfthe exclusionary technology par excellence-as motivated by an inclusionary impulse as it was initially intended as a tool of rehabilitation that would normalize deviants (Rothman, 1971). Second, and relatedly, Cohen highlighted that inclusion's "assimilative" (Young, 1998, p. 66) drive can be deeply invasive, seeking to change individuals so that they can be drawn back into society, whereas exclusion can be benign and allow people to live as they choose: "It is also by no means obvious that exclusion must be the more extensive and less tolerant mode. We might separate a group only to ignore it completely, while inclusion might entail massive efforts to achieve normative or psychic change" (Cohen, 1985, p. 219). It is therefore useful, we argue, to distinguish between two forms of inclusion (and correspondingly of exclusion), one of which is formal, liberal, and involves treating people as citizens, and the other of which is substantive, interventionist, and involves doing something to people and their lives that helps to bring them back into society. Third, it is not necessarily in people's interests for them to be formally or substantively included, nor do they necessarily want it. "People are different not just in the labels attached to them" (p. 268), and some groups-Cohen suggested queer people or people with serious learning difficulties-might need or want to be treated differently and to live to some degree distinctly from other people, particularly if assimilation requires them to change in ways that are not acceptable to or possible for them.
Inclusion and exclusion are more complex and ambivalent concepts than much of the comparative literature indicates, then, and we should therefore pause before assuming that they are easily distinguishable, that inclusion is always benevolent, or that the intentions of policy makers are easily converted into reality. Cohen's warnings also apply to the analysis of rehabilitation, which is often described as an unadulterated social good, but is in reality much more complex: Policies that are intended to be rehabilitative do not always achieve it (Martinson, 1974;Mews et al., 2017); it is possible to rehabilitate in ways that are deeply controlling and invasive (Crewe & Ievins, 2020), and although some people want to be "rehabilitated," not everyone does (McEvoy, 2001). Taken together, this ambiguity invites us to conduct research on how inclusion, exclusion, and rehabilitation are experienced from the ground up, focusing on fine-grained ethnographic analysis. In so doing, we can move away from comparative penology's tendency to make use of reductive binaries based on macro-level metrics such as imprisonment rates or generosity in welfare expenditures and pay closer attention to the relationship between state policies and lived experience. Our approach thus responds to Garland's call for "small-n" studies, providing "detailed, in-depth comparisons [. . . ] focused on a few comparable jurisdictions selected for their relevance to issues of theoretical importance" (2013, p. 490; see also Brangan, 2020). This is the goal of this article, which describes the experiences of men imprisoned for sex offenses in England & Wales and Norway. These jurisdictions were chosen as examples of neoliberal and social democratic modes of government, and the rationale for studying the experiences of men convicted of sex offenses is both empirical and theoretical. The two jurisdictions have drastically different rates of imprisonment: In 2018, 142.4 people were in prison per 100,000 citizens in England & Wales (total prison population of 84,373) and 65.4 in Norway (total prison population of 3,461; Aebi & Tiago, 2018). Despite the significant difference in imprisonment rates, prisoners convicted of sex offenses make up a similar proportion of the prison population in both jurisdictions. In 2000, they accounted for ten percent of the prison population in England & Wales, and in 2002, they accounted for eight percent in Norway. By 2017, they represented 18 percent-a significant proportion-of the prison population in both jurisdictions. Generally, people convicted of sex offenses experience multiple exclusions-from wider society by virtue of their incarceration, and from other prisoners by virtue of their conviction-and are described by Wacquant as one of several "dangerous categories" (2009, p. 209, emphases in original) whose experiences of punishment represent a magnified version of the dominant penal style. As a result, we posit that studying how these men experience the interaction between intrusive controls and treatment interventions will give us important insights into the exclusionary and inclusionary properties of state punishment more widely. Overall, we find that although their experiences of punishment broadly mirrored the expected patterns-England & Wales was more exclusionary, and Norway was more inclusionary-they did so in ways that trouble the simplistic binaries between punishment in neoliberal and social democratic jurisdictions, and that speak to Cohen's fine-grained distinctions.

POLITICAL ECONOMY AND OFFENDER REHABILITATION
The contemporary penological literature consistently describes the increasingly exclusionary aims and practices of state punishment reflected in the extreme growth of the size of the prison population and of the population subjected to community punishment, as well as in the density of regulations applied to people subjected to the latter (McNeill, 2019). Common across these accounts is a description of the growth of the Preventive State (Steiker, 1998;Zedner & Ashworth, 2019)-that is, a state preoccupied with the management of risk and the provision of security for the majority, and that does so by weakening individual rights and extending its own power at the bottom end (Wacquant called this "a centaur state, liberal at the top and paternalistic at the bottom"; 2009, p. 312, emphases in original; see also Feeley & Simon, 1992;Garland, 2001). People convicted of sex offenses are often described as the archetypical target population of these preventive and exclusionary strategies, and many scholars have described how their citizenship rights are systematically stripped after their convictions. Spencer (2009), for example, used the work of Agamben to argue that the recent expansion of punitive policies directed toward men convicted of sex offenses in the United States, as well as in Canada and Britainregistration requirements, civil commitment, and housing limitations among them-banishes them into "a lawless space" (p. 220), the camp, which is used by the sovereign state to expel people from the citizenry. Wacquant (2009), however, highlighted that the removal of certain citizenship rights does not mean people convicted of sex offenses in the United States are no longer the target of state power. In fact, they are subjected to tight laws regarding, for example, public notification and limits on where they can live. Rather than treating people as banished noncitizens, this "punitive panopticism" (p. 209) turns them into highly controlled "carceral citizens" (Miller & Stuart, 2017) who are deeply familiar with the bottom end of the centaur state.

Political economy and Nordic exceptionalism
As influential as these "grand narrative" analyses have been, they risk, as Lacey noted, "elevating an explanatory framework largely informed by the specificities of the US situation to the status of a general theory" (2008, p. 26). John Pratt's work on the Nordic exceptionalism thesis has challenged these analyses by highlighting that imprisonment levels in the Nordic countries are much lower than in other Western countries and by arguing that their punishment practices are milder, more humane, and more consistent with penal-welfarist principles (2008a, 2008b, in press; see also Pratt & Eriksson, 2013). Part of this benevolence is demonstrated the Nordic commitment to mitigate (some of) the exclusionary effects of imprisonment through practices such as the extensive use of open prisons, decent material conditions, well-trained and motivated prison officers, respectful staff-prisoner relationships, and the provision of work and educational programs (Pratt & Eriksson, 2013, pp. 8-21). Furthermore, Pratt and Eriksson asserted that Nordic prison systems have retained a "faith in rehabilitation and inclusion" (p. 191, emphases in original) even as the rehabilitative ideal has declined in the neoliberal countries (Allen, 1981;Feeley & Simon, 1992;Garland, 1990). Although Pratt warned that the distinctly Nordic approach is at risk, as penal policies and practices move in more punitive directions (2008b; see also Shammas, 2017), he and Eriksson maintained that the comparative study of Anglophone "penal excess" and Scandinavian "penal exceptionalism" reveal "[t]wo very different ways [. . . ] of thinking about punishment" (2013, p. 3). Pratt and Eriksson's work has contributed to significant academic and policy interest in the Nordic countries. Some scholars have highlighted practices that do not fit into Pratt's model, such as the widespread use of restrictions and isolation in pretrial detention (Smith, 2012), the overrepresentation of foreign nationals (Ugelvik & Damsa, 2018) and drug users in Nordic prisons (Mjåland, 2014), and the deportation of failed asylum seekers and criminal aliens (Barker, 2012). They have also developed our conceptual understanding of Nordic penality by arguing that its apparent benevolence has a dark side and that Nordic inclusion takes an invasive form. Barker (2012), for example, described Nordic penal regimes as "Janus-faced": one face mild and caring, and the other intrusive and controlling. Smith and Ugelvik (2017) built on these insights, supplementing the two ideal types they see in Pratt's work-the punitive Anglophone and the mild Nordic-with a third, the "Big Mother penal welfare state model" (Smith & Ugelvik, 2017, p. 513), which represents a culture of benign but intrusive intervention. Like the Preventive State (Zedner & Ashworth, 2019), the Big Mother State is highly active, and it overlooks individual rights to serve its own purposes; unlike the Preventive State, its desire is to help rather than to manage. In the words of Smith and Ugelvik, the embrace of this State "is simultaneously loving and forceful, and she both wants and knows what is best for you" (2017, p. 513; see also Pratt & Eriksson, 2013, p. 190).
According to Smith and Ugelvik, the Big Mother State pursues a distinct kind of offender rehabilitation, characterized by a maternalistic and interventionist penal strategy. Whether Big Mother is an apt metaphor to describe Nordic penality, however, is a question that demands closer empirical scrutiny, as Smith and Ugelvik themselves acknowledged. It is certainly the case that some of the worst excesses of the interventionist Nordic states led to an increased focus on prisoners' and due process rights from the 1970s onward, limiting what the state could legitimately do to rehabilitate offenders (Fransen, 2017;Pratt, 2008a, p. 132). It is also the case that the renewed interest in therapeutic interventions in Nordic prisons in the last two decades (Kolind et al., 2013) coexists with forms of rehabilitation that are more voluntary and, in Rotman's terms, anthropocentric or humanistic, "conceived of as a right of the citizen rather than a privilege of the state" (1994, p. 292). These forms of rehabilitation include the provision of substantial work and education opportunities in Nordic prisons (Pratt & Eriksson, 2013). Furthermore, rehabilitation's intrusive tendencies are certainly not only found in the Nordic region (Cohen, 1985). Indeed, some scholars argue that rehabilitation in neoliberal countries has not so much died as changed form, increasingly moving in the direction of what Rotman called "authoritarian rehabilitation"-that is, "a technical device to mould the offender and ensure conformity to a predesigned model of thought and behaviour" (1994, p. 292). Hannah-Moffat (2005) and Robinson (2008), for example, argued that rehabilitative principles have been transformed rather than eradicated through the neoliberal move toward risk management and managerial practices, and Crewe (2011;Crewe & Ievins, 2020) used the concept of "tightness" to describe how men in English prisons feel pressured to transform themselves to live up to the expectations of late-modern rehabilitative practices. The presence of these authoritarian strands in non-Nordic countries suggests that the relationship between rehabilitation and political economy is more complex than the Big Mother metaphor implies.

Four forms of rehabilitation
A route to a more nuanced understanding of the rehabilitative ideologies that exist in England & Wales and in Norway and of their relationship to inclusion and exclusion is offered by McNeill's typology of four forms of rehabilitation (2012, 2014; see also Burke et al., 2019). McNeill developed this typology in part as a result of his frustration with the persistent conflation of rehabilitation with treatment-psychologically-informed programs that seek to promote "individual level change in the offender" (2012, p. 14). He acknowledged that what he called psychological rehabilitation (or sometimes personal or correctional rehabilitation) has a substantial evidence base, and that people in conflict with the law often want to be encouraged and helped to change (see also Crewe & Ievins, 2019). He warned, however, that giving this rehabilitative form too dominant a position leads us to think about rehabilitation too narrowly. By highlighting three additional (as opposted to other) rehabilitative forms, McNeill contributed to a less reductionist theory. The first of these, legal or judicial rehabilitation, involves "when, how and to what extent a criminal record and the stigma that it represents can ever be set aside, sealed, or surpassed" (McNeill, 2012, p. 15). This "deontological conception of rehabilitation" (p. 5, emphasis in original), which has its roots in the work of Beccaria, is not based on utilitarian justifications of reducing crime or increasing public protection, and instead involves the full restoration of people's rights and duties once their punishment has concluded. The second form in McNeill's typology, moral rehabilitation, operates on the conflict and the harm (as opposed to the individual) and refers to the extent to which "moral debts are settled" (p. 15). Here, McNeill built on the work of penal theorist Duff (2001), who suggested that the wrongness of the offense should be communicated to both victim and offender, creating the conditions for recognition, reparation, and social healing, and Maruna (2011, p. 5, emphasis in original), who suggested that "moral inclusion" could be facilitated by rituals of reintegration and reconciliation. The last form is social rehabilitation, which involves "informal social recognition and acceptance of the reformed ex-offender" (McNeill, 2012, p. 15). It is less concerned with formal and legal barriers to the integration of ex-offenders and more with informal processes in communities, but the two are intertwined: "Social rehabilitation is not something that the state itself can do or order or provide; rather, it is a duty and function of civil society, but one that the state retains the duty to underwrite and support" (Burke et al., 2019, p. 14). It is central to McNeill's argument that none of these forms of rehabilitation can operate independently and, in particular, that "[n]o amount of correctional or psychological or personal rehabilitation and no amount of supporting people to change themselves can be sufficient to the tasks and challenges of reintegration, resettlement and re-entry, if legal and practical barriers are left in place" (Burke et al., 2019, p. 13).
At the heart of this typology lies a deep, if implicit, interest in exclusion and inclusion and, in particular, the relationship between formal and substantive inclusion. It also raises questions about the role the state can and should play in people's private lives. Psychological rehabilitation can help people to address their own real sense that there is "something wrong with them," but it can also impose intrusive and controlling demands as a requirement for reintegration (see also Crewe & Ievins, 2020). Legal regulations can facilitate formal inclusion if released people regain their full rights on release from prison and if legal measures oblige employers not to discriminate, or they can shore up formal exclusion, as when they are, for example, banned from having contact with the people they care about. The idea of moral rehabilitation is consistent with Duff's (2005) idea that punishment can be a rehabilitative and inclusionary practice as it involves settling a debt and undergoing due penance. Enabling moral rehabilitation, however, might require the state to position itself more intimately in relation to the conflict than it is willing or able to (Christie, 1977). Finally, social attitudes and interactions can either promote or impede substantive inclusion, depending on the communities involved.

METHODOLOGY
The data on which this article is based were collected between 2017 and 2019 as part of a largescale research project comparing penal policymaking and the prisoner experience in England & Wales and Norway. The goal of this project was to provide a detailed empirical analysis of the Nordic Exceptionalism thesis and to understand how large-scale penal difference is experienced. The project had four main substudies, two of which were relevant to this article. 1 Throughout the project, we sought to describe the experiences of all prisoners but with a particular focus on men convicted of sex offenses and women, two groups that have been neglected in prison sociology. This focus rendered the selection of sites and participants challenging. The first substudy was a longitudinal study of prisoners' entry into, progression through, and release from prison, and it involved interviewing people before, during, and after their prison sentences (entry-exit study). In total, 455 interviews were conducted for this study. We started by choosing prisons serving entry functions. In England & Wales, these were local (category B) prisons, usually located in inner-city areas, and designed for remand prisoners and those newly sentenced. Three prisons were initially selected, one for women and two for men. In Norway, the "local prison" category does not exist. Instead, all high-security prisons hold remanded people, and they are typically the first point of entry for people with sentences of two years of imprisonment or more. Low-security prisons do not normally accept people on remand, but they are the main point of entry for those sentenced to less than two years of imprisonment. We therefore selected two high-security establishments (of which one was a women's prison and the other a 1 The substudies that will not be discussed in this article are an analysis of penal policymaking in the two countries and a comparison of "deep end" confinement. For this latter substudy, we interviewed seven men convicted of sex offenses sentenced to forvaring (preventive detention) in Norway. We do not draw on these interviews for the current article as the forvaring sentence is rarely used in Norway (<3 percent of the prison population are serving a forvaring sentence). Their experiences are therefore unique; for more details, see Crewe et al. (submitted for publication). men's prison with 12 places reserved for women), and one open prison for men. Whenever possible, we interviewed research participants shortly after entry, shortly before their release (they had often been transferred to different institutions and we tried to interview them there), and two or three months after their release. Because we could not follow all participants through (because their sentences were too long, because we lost contact, or because they did not want to participate further), and because we wanted to include people with longer sentences than a straightforwardly longitudinal design would have allowed, we added extra prisons to our original sample in which we could interview people shortly before their release, with the aim of interviewing them again after release. In England & Wales, the additional research sites were a closed training prison (category C) and an open prison (category D), and in Norway they were two high-security prisons initially designed to hold men on longer sentences and one smaller prison located in a rural area, which had both high-and low-security units.
The second substudy involved ethnographic studies of penal power and prisoner social relations in prisons for women and for men convicted of sex offenses (ethnography study). The total number of interviews for this substudy, including the women, was 154, and we engaged in extended periods of participant observation. 2 For this study, the selection of research sites was easier in England & Wales, where men convicted of sex offenses are typically held apart from "mainstream" prisoners, either in separate prisons or on separate wings (Vulnerable Prisoners' Units or VPUs). Our main research site was a large medium-security (category C) prison for men convicted of sex offenses (operational capacity 1,206), and we conducted a smaller study in a VPU in a prison serving both a "local" and a training purpose (categories B and C, operational capacity 768 in the prison and 108 on the VPU). The Norwegian Prison Service, on the other hand, does not separate prisoners based on the type of index offense. As the ethnographic component of this substudy would work better in sites holding a certain number of men convicted of sex offenses, we selected one of the few wings in the Norwegian high-security prison estate that offers a treatment program for men convicted of sex offenses as our main research site (operational capacity 30 on the wing and approximately 250 in the prison), and we conducted research in a low-security prison in which half of the population were convicted of sex offenses (operational capacity approximately 50). Most men in the open prison were convicted of serious sex offenses and had progressed to this prison toward the end of long sentences, and therefore, they had considerable prior experience from a wide variety of high-security prisons. The ethnographic fieldwork lasted for four-six months at each research site, except for the smaller study at the VPU in England & Wales (two months of fieldwork).
For all substudies relevant to this article, the selection of research participants was based on a combination of opportunistic and purposive sampling criteria (Gobo, 2007). In most prisons, a member of staff served as our contact person. In the prisons selected for the entry-exit study, the contact person often provided us with information about newly arrived or soon-to-be-released prisoners. In most prisons, we carried keys and moved around unescorted, and we therefore introduced ourselves and asked if they wanted to be interviewed. In the few prisons in which we did not carry keys, the contact person provided prisoners with an information letter about our research project and asked them to participate. In the prisons in which we did not have a contact person, we visited the wings and asked officers on duty if any of the prisoners had recently arrived or were approaching release. Generally, when we were provided information about the prisoners' index offense, we purposively tried to recruit men with sex offense convictions. Although most prisoners who were asked agreed to take part in our study, many of the Norwegian men with sex offense convictions did not. We believe this had to do with the fact that, in Norway, men convicted of sex offenses are held alongside "mainstream" prisoners, and therefore often try to hide the nature of their convictions to avoid abuse and victimization. Keeping a low profile was a way to stay safe, and saying no to being interviewed was part of this adaptive strategy.
We had slightly more control over the recruitment process of interview participants for the ethnography study. At these research sites, we spent considerable time observing and talking informally with prisoners as part of the ethnographic fieldwork. In all four research sites, we tried to take part in as much of daily life as possible: We played sports, hung out in the yard or communal areas of the prison, visited prisoners in their cells (although this was only allowed in Norwegian prisons), watched television with them in the evenings, played cards, and visited people at their workplaces. As a result of our familiarity with the prisons and the prisoners, we could therefore rely on a slightly more purposive sampling strategy, with the aim of increasing the variance of experiences in our sample of participants (Gobo, 2007, p. 17). Although we did not operate with a specific set of criteria, we tried to recruit prisoners with different convictions, who were held in different wings or housing units, who were differently placed in the informal prisoner hierarchy, and who were of different ages. We made a deliberate attempt to approach not only the most vocal and sociable prisoners but also those who kept to themselves and who did not take part in social activities in the prisons.
In England & Wales, we are confident that our sample of research sites and participants gave us access to "typical" experiences of imprisonment for men convicted of sex offenses. We are slightly less confident that we achieved the same in Norway as our sample of participants includes an overrepresentation of men involved in rehabilitation programs and an underrepresentation of those living alongside "mainstream" prisoners, who may have been hiding their convictions. We try to accommodate for this in the analysis that follows by drawing heavily on the interviews with the men we interviewed for the entry-exit study, who often represent the latter category. In our analysis, we have also drawn from retrospective accounts, as when prisoners on the rehabilitation unit told us about their experiences of imprisonment before they progressed to the unit. Overall, then, we believe that the sample includes a satisfactory variance in terms of experiences of imprisonment for men with sex offense convictions in the two jurisdictions.
In total, our sample for this article consists of 85 interviews with men convicted of sex offenses in England & Wales, and 44 in Norway, and it includes interviews conducted in seven prisons in England & Wales (we visited the same institutions for multiple substudies), and six in Norway (see table 1). 3 The interviews were typically conducted in private rooms in the prisons. Interviews normally lasted between one and three hours, although some were longer. Five researchers worked on the project, all of them with substantial experience of qualitative prisons research. All members of the research team interviewed participants for the two substudies we draw on for this article, but the first author conducted the majority of the interviews with the men convicted of sex offenses in England & Wales, and the second author interviewed approximately half of the sample in Norway. Almost all interviews were audio-recorded, and all interviews were transcribed verbatim; those conducted in Norwegian were then translated into English. The content of the interviews varied with the substudies, but all of the fieldwork was driven by an interest in four main sensitizing concepts (see Crewe, 2015): depth (the distance between the prison and the outside world), weight (the oppressiveness of imprisonment), tightness (how psychologically invasive imprisonment is), and breadth (the extent to which the sentence is carried with people after release). We also maintained an interest in shame and penal consciousness in both countries-that is, how the state shapes how people think about their offenses, and what prisoners think the state is trying to do to them by punishing them (see Sexton, 2015). All interviews were semistructured (Kvale, 1996, p. 124). We used an interview schedule to organize the conversations, while allowing both interviewer and interviewee to bring up topics and questions flexibly. The interview guide for the entry-exit study was structured around the sensitizing concepts referred to above, but we also asked different questions depending on the phase. In the postrelease interviews, for example, we asked detailed questions about how people dealt with the restrictions imposed on them. The interview schedule for the ethnography substudy was organized somewhat differently, and it included questions related to social life and relationships (between prisoners and between prisoners and staff), power, rehabilitation and change, shame and guilt, and views on the aims and consequences of punishment.
The data we draw on for this article were coded using NVivo software, mostly by the two authors. Based on careful readings of a sample of interview transcripts, and informed by discussions and reflections throughout the ethnographic fieldwork, we developed a coding scheme reflecting key themes from the interviews and observations (e.g., "shame and guilt," "social relationships"), as well as our sensitizing concepts ("tightness," "penal consciousness"). Under each of these broader nodes, we developed more specific "child" and "grandchild" nodes, of both descriptive ("purpose of sentence," "bullying from 'mains'") and more conceptual ("shame management strategies," "stain") character. When encountering data with a poor fit to our existing nodes, we added new ones (e.g., "cover stories"). To ensure consistency in coding practices, we wrote and exchanged notes on what kind of data we included in the various nodes, and why.

LEGAL LANDSCAPES AND PUNISHMENT STRUCTURES
Before we turn to the analysis of the interviews with men convicted of sex offenses in England & Wales and Norway, we offer a brief sketch of the relevant legal contexts in both jurisdictions. The aims and justifications for punishment and sentencing are based on different and sometimes conflicting principles in both England & Wales and Norway. In the former, the Criminal Justice Act 2003 states that punishment, crime reduction, rehabilitation, public protection, and reparation are the aims of sentencing, although it offers no guidance on which should take priority. In the latter, crime prevention through general and individual deterrence, and upholding the "social peace" are the main principles justifying punishment (White Paper no. 37, 2007-2008, pp. 19-20). In both countries, there has been an increase in punitiveness for sex offenses in the last two decades. Despite these similarities, the legal contexts affecting men convicted of sex offenses differ markedly. In England & Wales, the punishment of men convicted of sex offenses is increasingly justified with reference to public protection and risk management; men convicted of sex offenses are discursively and institutionally treated as a distinct type of prisoner ("sex offender"); and the legal restrictions facing people upon release are many and comprehensive. In Norway, despite an increase in sentencing, the official aims of punishment are the same for men convicted of sex offenses as for other offenses; men convicted of sex offenses do not make up a distinct category either in legal regulations, policy documents, or in the placement of prisoners; and few restrictions are placed on them after release. The legal context, then, is based on a logic of exclusion in England & Wales. In Norway, the fact that men convicted of sex offenses are not treated differently from those convicted of other offenses suggests a legal context that is, at least formally, more inclusionary. We will detail these similarities and differences below.
The exclusionary nature of punishment in England & Wales is reflected in the fact that people convicted of sex offenses are treated as a separate category, one that has been the subject of increasingly harsh punishment. The Sexual Offences Act 2003 introduced several new offenses, in particular related to the Internet, and led to an increase of 31 percent in the number of people being sentenced for sex offenses between 2004 and 2011. These new offenses combined with a 13-month increase in average custodial sentence length over the same period to swell the population of people imprisoned for sex offenses (Ministry of Justice, 2013). In the mid-2010s, changes in public willingness to report sex offenses and in recording practices at police stations, as well as an increased public focus on sex offending, also led to significantly more sex offenses resulting in convictions (Ministry of Justice, 2017).
The Sexual Offences Acts 2003 covers a diverse range of acts including sexual acts with contact against adults and children, noncontact acts such as voyeurism, downloading or disseminating abusive images of children, and solicitation, but the penal system in England & Wales constructs everyone convicted of the offenses that it covers as "sex offenders." 4 To keep them safe from abuse and assault from other prisoners, most men convicted of sex offenses serve their sentences separately from other men, reinforcing the idea that "sex offenders" are a fundamentally different type of person and of prisoner and contributing to the widespread use of the term "sex offender" by both staff and prisoners (Ievins, 2020). The term is also used by policy makers: A recent white paper on prison reform, for instance, called for "a more coherent approach to how we manage specific cohorts of prisoners, including young adults and sex offenders, to make sure that they are in the right places to meet their safety and reform needs" (Ministry of Justice, 2016, p. 60).
Men convicted of sex offenses have a shared identity in legal consciousness, then, but their treatment in custody is less uniform, and the majority of people considered to be "sex offenders" do not receive rehabilitative interventions. 5 The treatment programs that are available-most of which are cognitive-behavioral and group based, and many of which are specific to people convicted of sex offenses-are assigned following risk assessment by the actuarial risk assessment tool RM2000, with programs offered to those who are deemed to pose a medium or high risk of sexual reoffending. People convicted of sex offenses also serve diverse types of sentences, both determinate and indeterminate, and there is no legislative provision for civil commitment. 4 Men convicted of offenses added in subsequent legislation-such as "upskirting"-are also classed as "sex offenders." Following their release from prison, all men are subjected to tight license restrictions, the breach of which can lead to their recall to custody; in practice, these restrictions are determined by prisoners' convictions and risk level, and frequently they involve limits on supervised or unsupervised contact with persons younger than 18 or on Internet access, or require people to disclose their convictions before starting new relationships. All men convicted of a sex offense are also required to attend a police station within three days of their release to sign the Sex Offender's Register (for life, if they receive a custodial sentence of more than 30 months). The Register is not publicly available (with some exceptions) but being on the Register places several notification requirements on people. In addition, some men convicted of sex offenses are given a Sexual Harm Prevention Order (previously a Sexual Offences Prevention Order, or a SOPO), a civil order that aims to inhibit sexual offending by preventing the individual from engaging in certain forms of behavior, such as accessing the Internet or having contact with persons younger than 18.
In Norway, a separate chapter in The Penal Code 2005 covers the criminalized actions defined as sex offenses. These include, among others, rape of adults and children, sexual acts without consent, sexual acts against children, insulting sexual behavior, possession/production/publication of sexualized content including children, incest, and the purchase of sex ( § § 291-320). For each of these articles, an upper limit of punishment is specified (e.g., "aggravated rape" is punishable with up to 21 years of imprisonment, which is the maximum sentence in Norway). For sexual offenses as a whole, there has been, according to Hansen (2013, p. 688), an "unprecedented" increase in sentence lengths in the last decades, and in the case of "sleep rapes," where the victim is unconscious, sentencing has increased 12-fold from 2000 to 2013 (Jacobsen, 2017, p. 127). This development has occurred because minimum sentences have been applied to some types of sex offenses-a controversial practice uncommon in Norwegian criminal law as it leaves less discretion to judges-and because the maximum sentences have been lifted for the most aggravated sexual crimes (ibid.). Whereas crime prevention is the main justification of punishment in Norway, the recent increase in punishments for sex offenses implies a more expressive element to sentencing (Skilbrei & Stefansen, 2018, pp. 103-104), and it challenges the jurisdiction's "exceptional" moderation in sentencing practices (Pratt & Eriksson, 2013).
Although The Penal Code 2005 includes a separate chapter for sex offenses, the term "sex offender" is not in use in the law and rarely in other legal and policy texts. Unlike in England & Wales, then, people convicted of sex offenses are not discursively constructed as a separate group. 6 This is further reflected in the "crime-blind" allocation of prisoners in Norwegian prisons (Friestad et al., submitted for publication), and it means that men convicted of sex offenses serve their sentences alongside "mainstream" prisoners. This formal "inclusion" of men convicted of sex offenses in the prisoner society has raised concerns about their safety as this group of prisoners is at risk of victimization as a result of their low moral status in the prisoner community (Ugelvik, 2014). A few sex offense-specific, cognitive-behavioral treatment programs are provided, but only for a minority of prisoners, although the Prison Service is currently introducing a new program (The Prison Service, 2020, pp. 40-41). 7 Unlike in England & Wales, structured risk assessment 6 The Norwegian equivalent to "sex offender" is seksualforbryter or tools are not used in the Norwegian Prison Service, but a needs assessment instrument (BRIK) is provided on a voluntary basis.
Postrelease restrictions for persons convicted of sex offenses can be issued by the Prison Service. They only apply, however, when prisoners are granted parole (or serving a conditional sentence), and they are lifted when the determinate sentence ends (The Execution of Sentences Act of 2001). This follows from an important principle of the Norwegian prison system, namely that the punishment is settled when the sentence is completed (White Paper no. 38, 2007. None of the restrictions target men convicted of sex offenses specifically. To be banned from having contact with specific individuals, which is a restriction frequently imposed on men convicted of sexual offenses in England & Wales, should in Norway, according to the Regulations to the Execution of Sentences Act 2002 (p. 4, our translation), "be applied with caution," and it "should preferably be restricted to situations where offenders are convicted of serious violent offences, threats as well as sexual offences." Norway does not have a Sex Offenders' Register.

FINDINGS
Our findings indicate that the Norwegian state intervened relatively little in the lives of men convicted of sex offenses during and after their imprisonment. Punishment operated to censure people for wrongdoing, and men convicted of sex offenses were therefore formally included and judicially rehabilitated; however, their punishment was not necessarily experienced as morally meaningful or substantively inclusionary. We therefore describe the Norwegian approach as representing "laissez-faire inclusion," using this ambiguous term in a way that reflects both negligence and a liberal respect for individual autonomy. In England & Wales, the state intervened much more in the lives of men convicted of sex offenses, and punishment primarily involved the management of risk. This led to people convicted of sex offenses being deprived of many rights and imposed with many burdens, and this legal dehabilitation resulted in an enduring exclusion from the citizenry. Capturing the coercive and invasive nature of these interventions, we describe the English and Welsh approach as "authoritarian exclusion." The overall findings that emerged from the comparative analysis are presented in condensed form in table 2. In the analysis that follows, we provide a detailed account of these differences.

England & Wales: Authoritarian exclusion
Our interview participants in England & Wales described their punishment as marked by a highly authoritarian form of exclusion, one that saw men convicted of sex offenses as risky objects who needed to be controlled to protect the public. They were understood as permanently and uniquely dangerous, and although this risk could be reduced, it could never be removed. Kyle, a lifesentenced prisoner, encapsulated this discourse well: Sex offenders are never ever cured. Right? All that happens is they are taught how to manage their risk. How to recognize when situations are becoming dangerous, like who to go to and see for help. (Ethnography study) Risk is here used as a noun: It is something people carry with them. This sense of enduring and extreme danger is reflected in the tight legal restrictions placed on prisoners both within and beyond the prison, which are much stricter than they are for people with other convictions (McAlinden, 2007).
Such was the density and tightness of these restrictions that it was common for people who had been released to say that they still felt like they were in prison. Joel, for example, said that he refused to let himself feel like his sentence was over or that he was free. Otherwise, he risked breaking a rule and being recalled into custody: Complacency sets in, and it's very easy then to do something which may then go against a SOPO or a licence condition, even a minor thing, that's still a recall [to custody], you're back inside. I don't let myself feel that I'm. . . I'm still on my sentence. Interview participants argued that these restrictions, which represent a form of judicial dehabilitation, showed the legal system operating on the presumption that they were not able to change. Carlton's words were representative: I don't think the system is geared up or designed to rehabilitate people, if I'm honest. The system has fixed notions of offenders that [they] never change. That much is evident in that fact that when you go into society, you're required to disclose your conviction, you're barred from doing all kinds of things. So if there was any sense, any genuine sense that people could be rehabilitated, then they would be allowed to forget their past and move on. But you're never able to forget, especially if you're a sex offender in this society, you're never allowed. (Ethnography study) This belief that prisoners represented a static danger was reflected in the prison's apparent lack of interest in encouraging change. When asked how the prison wanted them to behave, almost all interview participants emphasized institutional compliance rather than individual transformation. Mick said that the prison wanted them to be "submissive," and Reece insisted that the prison did not try to shape him, that "they look at it this way: If you don't give them no hassle they don't give you no hassle" (ethnography study). Arjun, like many other prisoners, felt that the prison's lack of interest in promoting personal and psychological rehabilitation betrayed its neglect: "Prison doesn't care what kind of person you want to be, as long as you stay out of their way. The less effort you give them, the less requirement for them to do any work is what they want" (ethnography study).
Over the course of their sentences, however, some of our research participants underwent structured treatment programs. These programs encouraged prisoners to think about themselves in ways that were tightly organized by cognitive-behavioral discourse, and some described this discourse as alienating: Emmett said that programs aimed to give things "a psychological title," leaving those who undertake them only able to use "program speak" (ethnography study). Many prisoners complained in interviews about the low availability of these courses and the fact that they were apparently unlikely to "work." 8 In most cases, prisoners said that they completed these programs for procedural reasons either because the conditions of their release depended on them lowering their risk level or because they considered programs to be the forum for change that was most likely to be recognized in their parole hearing. Joel, for example, said that he completed the Sex Offender Treatment Program even though he found it uncomfortable because "it was on my sentence plan, it was what I needed to do" (entry-exit study, prison interview). Others talked at length about struggling to get a place on programs that they thought would help them to be released, indicating some of the serious frustrations with being provided with a narrow official route to self-transformation. In such cases, they described themselves as "objects of procedures" rather than as participants in a process (Burke et al., 2019, p. 3).
That said, some prisoners described programs as providing a useful opportunity for personal development. Jesse, for instance, said that he had appreciated the opportunity to spend time 'rummaging around [my] head and checking everything is alright and ready for the next stage' (ethnography study). Some men described the tight structure of programs as providing a useful scaffold. Aaron had been introduced to the idea of cognitive schemas (patterns of thought) while undergoing treatment, and years later he kept a schema diary that he felt helped him to reflect positively on his behavior. Emmett saw past his frustration with psychological discourse and said that he valued the official "affirmation" (ethnography study) that he received in programs that his personal reflection on his offending was on the right track. Overall, then, the form of personal rehabilitation to which these men were exposed can be described as "authoritarian" (Rotman, 1994) but inconsistent-offering a path to change that is in theory traversable but is in practice narrow, deceptive, and laden with obstacles (see also Crewe & Ievins, 2020;Österman, 2018).
As people approached release, they received little practical assistance with finding employment or accommodation. Furthermore, the legal burdens placed on them actively blocked them from reintegrating back into society once they left the prison. Many men said that they had restrictions on supervised or unsupervised access to children, which made it difficult for them to spend time with their families. Carlos described how these restrictions affected his relationship with his girlfriend's children: 8 During the fieldwork period, a report was released stating that the Sex Offender Treatment Program, which had been widely used for the preceding three decades, increased people's risk of reoffending (Mews et al., 2017). The program was canceled as a result. I want to take my surrogate children out for lunch. Can you not take them out. . . ? I can take them with [girlfriend] or with my mum or whoever, but I can't on my own. And would that be the case. . . ? For the rest of my life, yeah. Like my niece, for instance, I can't even just take her down the shop. I turned up there the other day, so I knocked on the door, she opened but her nan wasn't in and her mum wasn't in so I had to go away and wait for them to come back. (Entry-exit study, postrelease interview) Such was the impact of these restrictions that Carlos said that he wanted to move abroad as soon as his license period was over, a common desire among men convicted of sex offenses: "I'm not trying to run away from it but I want to live a normal life." Several other interview participants said that being unable to access the Internet, a standard license condition for anyone convicted of an Internet-related offense, meant that it was difficult for them to apply for benefits. Elijah lost his job after his probation officer told him that it was a legal requirement that he disclose his conviction to his employer. This experience led him to state in his postprison interview that "the biggest punishment is what happens after [prison], like losing everything and trying to start again, but having certain things holding you back" (entry-exit study, postrelease interview). He, like others, did not describe the state failing to facilitate social rehabilitation; instead, he described state policy serving to actively prevent it.
The system's preoccupation with public protection made it harder for prisoners to engage in serious moral reflection on what they had done. Aaron's story serves as a good example. As a teenager, and after being seriously abused by his stepfather, he raped his younger sister. After going to prison, he was so ashamed that he broke off contact with his family. Several years later, and following the intervention of another family member, he reestablished contact with his mother, and they used phone calls and visiting periods to talk about what had happened and to make sense of their past experiences of trauma. When the Security department and the Public Protection Unit within the prison realized that his victim lived with his mother, however, they immediately blocked contact between Aaron and his mother. After a lot of effort, the decision was eventually reversed, but the fact that it was made is illustrative of the state focus on managing risk over enabling reconciliation. Aaron made a similar point when he described the pain of being significantly over tariff on his indeterminate sentence, as the Parole Board believed that he had not yet demonstrated that he had reduced his risk: It's like prolonged punishment. Whatever they're punishing me for now, it's just prolonging the punishment on my family and me not being able to help them or me not being around them. They've [my family has] forgiven me for what I've done. It's took them a while, it's took them years but they've forgiven me so why can't society? It's got nothing to do with anybody else. (Ethnography study) Aaron's case raises difficult questions, and it is possible that allowing closer contact between him and his mother would endanger his victim physically and emotionally. What his case illustrates, though, is the intrusive and paternalistic nature of risk management work, as well as the potential conflicts between managing risk and enabling moral rehabilitation.
Overall, then, men imprisoned for sex offenses in England & Wales experienced the state as both formally and substantially exclusionary: It did not treat them as citizens, and it blocked them from reentering society after they were released. Although some prisoners appreciated the programs they were offered, the psychological rehabilitation this represented was not supported by formal processes of inclusion, and the continuation of their punishment after their release from prison meant that they rarely felt like they had "done their time." In part, the continuation of punishment derived from the public stigma attached to these sorts of crimes. Manny, for example, said that "anything to do with any form of sex offense, they're branded, they're not allowed a second chance" (ethnography study). The continuation of this stigma was ossified by the legal restrictions imposed on people convicted of sex offenses, as Carlos acknowledged: "The added extras at the end I do feel are slightly excessive because they don't give you that second chance" (entry-exit study, prison interview). Luca, interviewed after his release as part of the entry-exit study, summarized the morally communicative effects of this authoritarian exclusion: "I'm sick of the state interfering in my life and telling me I'm a bad person, I'm a bad man, and I'm going to take all this away from you."

Norway: Laissez-faire inclusion
Our participants in Norway described their relationship to the state differently, implying that they continued to be seen as citizens during and after their incarceration. The discourse of risk was significantly less pronounced in these interviews perhaps as a result of the fact that neither men convicted of sex offenses nor "mainstream" prisoners are subjected to standardized risk assessments. They also face very few restrictions or demands while they are in prison and after release and those that are imposed are more often enacted on an individualized and discretionary basis. Julius, for example, who had served several years for having a sexual relationship with a girl younger than 16, benefited from this discretionary and individualized decision-making and had no restrictions imposed on him when he was granted early release: "Everyone who gets out has a duty to meet with the Prison and Probation Service, but because of my behavior in prison and because of the support system around me, they didn't think it necessary in my case" (ethnography study, postrelease interview). Unlike in England & Wales, then, where a preoccupation with risk means that people convicted of sex offenses are presumed to be dangerous, the Norwegian state is much more willing to place trust in its citizens or at least not to remove trust merely on the basis of membership of a category. Contrary to common descriptions of the Norwegian penal state as intrusive and interventionist, we here see a state that functions with a very light touch. Dagfinn offered a clear example (entryexit study). He was in his late fifties and had a 20-year-old girlfriend and was sentenced to eight months in prison (after having waited for four years in the prison queue; see Laursen et al., 2020) for having sex with her when she was under age. He steadfastly maintained innocence and was supported in his trial by both his girlfriend and his girlfriend's mother. If he had been imprisoned in England & Wales, he would have been unable to have contact with her during his sentence to protect her and would probably not have been able to live with her after release. The low-security Norwegian prison in which he was held, however, not only allowed her to visit but also granted him additional days of visits as she lived so far away. Furthermore, he faced no contact bans or other restrictions upon release. This is an example of laissez-faire inclusion: The state did not see Dagfinn as carrying so much risk that it had the responsibility to protect his victim from him, and instead it stayed in the background and prioritized him retaining family contact.
Interview participants described the primary purpose of their sentence as censure, that is, as expressing the wrongness of the offense. Bertram captured this well when he said that his sentence said to him and the community "that this is not acceptable behavior . . . that this is not okay, this is not accepted" (ethnography study). Those of our participants who were provided with behavioral interventions (generally those convicted of the most serious crimes) considered rehabilitation to be an important purpose too, although they seldom described feeling enticed or threatened into completing treatment because of the fear of not being released. In fact, they often welcomed the opportunity to address what they felt went wrong. Orri, for example, had previously served a shorter sentence and completed only parts of a sex offense treatment program. He reoffended not long after his release, leading to a significantly longer sentence, and even though he regretted what he had done, he knew that he needed treatment and that he would get it in his current prison: So it was-if not carefully planned then at least a planned act that meant that I knew that I'd come in here again. I knew with myself that I needed more treatment. I needed to get it-get things-if nothing else then at least to highlight things in a more thorough way. So I could learn more and see more and understand more. (Ethnography study) Orri's account may be unusual, as he was one of the few who suggested that his desire for treatment impacted on his (re)offending, but the genuinely positive way he described treatment programs-as an opportunity to understand and learn-was shared by the majority of research participants on the treatment wing. Because few Norwegian prisons offer treatment programs, however, most men convicted of sex offenses did not receive any formal interventions. Very few men who were not on the treatment wing had had any conversations with prison officers or other staff members about their offending. Although they may have reflected privately on their own journeys, some were frustrated not to receive more specific guidance. These complaints were particularly common among men serving short sentences and in open prisons, where few resources were available for "recurrence prevention" (Lars-Petter, ethnography study).
Petter's case illustrates some of the painful consequences of the gentle approach to discipline that dominated in many Norwegian prisons. He was convicted of sending explicit photographs to an underage girl and served two months in an open prison unit. He was provided with no formal opportunities for change and was not expected to undertake any. Although he did not believe that he was attracted to children or was likely to reoffend, he occasionally doubted himself, and he described the prison's failure to offer him interventions as negligent: But I feel like maybe, what I have found out now is that there is a support group you can go to [after release] if you are a convicted sex offender. Why did I only find that out afterwards? Why did no one mention it in prison? Why don't they tell you that it exists? Is it because they think that you don't need it? Or is there a different reason? I wonder about that. Why was there no one in prison who said, "Okay, you are convicted of this and that. In order to not come back to prison for the same things, you should do this." Is it because they think that I am not at risk of doing it again? Or because that option doesn't exist, or because they don't care enough to offer it to me? I mentioned it, we talked about that last time, that there are programs for people who drive drunk, anger management programs. . . Drug programs. Yes. Why isn't there a "I send naked photos to 15-year-olds" program?
Because you would do it? You just have to, right? You should. Even though I think that for me it was a one-time incident, that is still something that I did. And, in a way, I'm afraid of doing it again. (Entry-exit study, postrelease interview) Petter spent much of his sentence in reflection, but his lack of trust in himself meant that he wanted to experience a more authoritative form of psychological rehabilitation. In Burke et al.'s terms, he wanted to be subjected to a "mechanism" and not just to be allowed to undergo a "process" (2019, p. 62). Without this opportunity, he described his sentence as meaningless: "You take the individual out of society and move them to a different place in society for a prison sentence, store them there and then put them back, then I don't think that can be called 'rehabilitation.' . . . It is just moving people" (entry-exit study, prison interview).
Nevertheless, the majority of people were hopeful about their lives after release and about their ability to rejoin society. Julius reflected this optimism, insisting that "I have my whole life ahead of me" (entry-exit study, postrelease interview). Almost everyone reported receiving some form of help with accommodation and employment from the welfare services if they needed it, although several people complained about bureaucratic inefficiencies. Very few interview participants had restrictions on what they were able to do, and these were rarely described as burdensome. Igor, for instance, was released on parole under the condition that he continued to see his therapist once a month and his probation officer once a week, for the duration of a year ("but normally they shorten it a bit," as he said). He described these restrictions as unproblematic: "A few months and you are finished with everything" (entry-exit study, postrelease interview). That these men were not subjected to legal restrictions meant that those who had been able to maintain strong networks to family and friends were able to capitalize on them after their release. Petter, for instance, was accompanied by friends and family when he went out in public in case he was harassed; they also hosted a themed "release party" to literally and symbolically welcome him back into their community. These "rituals of reintegration" (Maruna, 2011) were the creation of civil society, but they were facilitated by a state that ensured that people regained their rights after release and that encouraged strong family contact during the period of imprisonment. Those who had not been able to maintain these networks, however, experienced or anticipated a much more fugitive and fragile inclusion. Although the Norwegian state was able to create the conditions that made social rehabilitation possible, it was not able to ensure it if prisoners were not welcomed by the community. Ole-Jørgen, for example, whose interview was conducted in English, described the specific pain of leaving the prison on temporary leave (similar to Release on Temporary Licence in England & Wales) and being provided with a phone but having no one to talk to: And it is funny when you got out for permission [leave], you have this phone but nobody is calling, you have nobody to call, and this is hard. You have bad days and you have. . . . I always had somebody to lean on, you can call a friend or a cousin or whatever. . . . [Now] I have nobody to lean on, you know, empty. . . . And the other way when the few times you feel a little bit happiness that is a long time ago, and you have nobody to tell, that is hard too. (Ethnography study) Lars-Petter said that he dreaded his release: "I don't really have anything to go to" (ethnography study). Bertram, a self-proclaimed "not very social" (ethnography study) man convicted of sexual offenses against his daughters, feared that he would face harassment and death threats when he was released, as he had done during his trial. He wanted to move to a rural area so that he could more easily move house and keep his job if he faced problems. We here see the limits of the state's inclusionary capacity as it is unable to counteract the bottom-up exclusionary processes of civil society.
Taken together, then, for many men convicted of sex offenses, imprisonment in Norway represented a sort of "benign storage:" 9 The prisons in which they were held were safe, decent, and harmless, but their sentences were often experienced as empty and meaningless. Dagfinn exemplified this discourse when he described his sentence as "nothing," "a blip," "just four months of my life," and "just a small event in my life" (entry-exit study, prison and postrelease interviews). The prisoners who described their experience as meaningful tended to be those who had experienced forms of intervention that allowed them to discuss their offending and process their guilt. Tesfay, for instance, had struggled with feelings of shame before he undertook a treatment program in a high-security prison. Bertram said that seeing a psychologist and then undertaking a treatment program gave him "peace" (ethnography study), which he deepened by writing and posting a letter of apology to his victim and her mother (something he would be unable to do in England & Wales; see Ievins, 2019). These stories were rare, however, and it was more common for prisoners to echo the claims of Lars-Petter, who served more than five years for sexually assaulting his young niece and explicitly denied that his sentence felt morally meaningful: Similarly, Petter maintained that his sentence had neither helped him process his feelings of guilt nor helped him to pay back. He suggested that this was partly because no one had discussed his offending with him-because his sentence had no concrete communicative content-but it was also a result of weaknesses with the metaphor of "paying your debt:" I don't think that the prison did that. It would be better to talk about it. To have some sort of group where you can sit and talk about those things. Or, yeah, talk things through. I don't know. I don't know how much shame you should feel, either. Or how much blame you should feel when you have made up for it in a certain way. When you have served your prison sentence and admitted what you did. Like, I am now legally done with that case. Legally done with it. Society has said that, or at least how I read society, society has said 'You've done something, you've served your sentence, now you're done with it. Until you eventually do something wrong again. But if you don't, then okay. We're done with it. It's a debt, you've paid your debt, and your debt is done.' But I don't feel like that is what you feel yourself. If you really felt sorry for what you did, I don't think you get rid of that through being in prison. (Entry-exit study, postrelease interview) Here Petter drew on the distinction between legal and moral rehabilitation. Regaining your rights certainly matters, but it is not the same as paying your dues or fixing the harm, and he described Norway's formally inclusionary system as insufficiently active to serve as a meaningful ritual. It is worth noting that not all prisoners were as frustrated as Petter was with this lack of engagement, but nevertheless his case speaks to the effects of prioritizing legal rehabilitation, a form of rehabilitation which is inherently collectivistic, over its other forms. According to the legal framework, Petter is rehabilitated when he regains his rights, and the state thus imagines him primarily as a citizen. Petter, however, is preoccupied by something altogether more personalthe harms he has caused and what these mean about who he is. For him to feel rehabilitated in a meaningful and moral way would require him to be seen and acknowledged as an individual.

DISCUSSION
The way we have described these two penal systems obscures significant variations in prisoners' experiences within each jurisdiction, but we nevertheless maintain that there were clear patterns and that the systems had very different aims and effects. The punishment of people convicted of sex offenses in England & Wales was marked by its authoritarian exclusion. The purpose of the sentence, as experienced by the men in our study, was to manage risky objects, and although disciplinary demands were placed on prisoners while they were in custody, they often lacked the opportunity to meet them. The state also tightly intervened in people's lives in prison and after release, imposing rigid burdens and restrictions that limited people's ability to rejoin their communities. In Norway, on the other hand, the punishment of men convicted of sex offenses reflected a form of laissez-faire inclusion. The sentence signaled the wrongness of the offense, and prisoners retained many of their rights (to education and work, for example) while in prison, and regained most of the rest after release. This liberal approach meant that the state was reluctant to intervene in social spaces after release, where the community was the main driver of inclusion and exclusion. Although rehabilitative opportunities were sometimes available, many prisoners described their sentence as an empty ritual. Our analysis adds important nuance to binary conceptions of punishment in neoliberal and social democratic countries (Cavadino & Dignan, 2006;Pratt, 2008aPratt, , 2008bPratt & Eriksson, 2013) and interrupts some of the evaluative assumptions built into these conceptions. Contrary to what might be expected, we find that the punishment of men convicted of sex offenses is more paternalistic and interventionist in England & Wales, as well as more liberal-in that it respects the autonomy of the punished person-in Norway. Whereas our description of paternalistic punitivism in England & Wales is perhaps not novel and reflects both Wacquant's (2009) description of the neoliberal "centaur state" and accounts of the punishment of people convicted of sex offenses in the United States (e.g., Tewksbury, 2005), the liberal laissez-faire approach we have identified in Norway has not been sufficiently accounted for in descriptions of Nordic penality. That is, although we might have expected people convicted of sex offenses to be the ideal target of intrusive interventions by a state eager "to change, to 'cure' such propensities in the individual concerned" (Pratt & Eriksson, 2013, p. 190), in reality the Norwegian penal system is laissez-faire when it punishes men convicted of sex offenses, in the ambiguous meaning of the term. On the one hand, it communicates to prisoners that the deprivation of liberty is the only punishment the state can and should impose on them, and that how they choose to engage with or respond to this punishment is their responsibility. The state offers opportunities for rehabilitation, mainly through work and education programs in prison, yet prisoners maintain autonomy to decide whether or how they make use of them. On the other hand, this "hands-off" approach to punishment is easily interpreted as careless, negligent, or uninterested by prisoners who speak of a desire for meaningful conversations about their offending.
The distinction between authoritarian exclusion and laissez-faire inclusion speaks directly to the varying configurations of rehabilitation in both countries and, in particular, to the weight given to legal rehabilitation. In England & Wales, the authoritarian exclusion of men convicted of sex offenses reflects the growing significance of a managerial conception of rehabilitation. As rehabilitative ideologies have evolved, the "ontological subject of rehabilitation" has changed from "sinner" to "patient" to "risk-bearing subject: one whose risks must be assessed and then managed and, where possible, reduced" (Robinson, 2008, p. 440, emphasis in original). This emphasis means that people are subjected to tight legal constraints through measures such as contact bans, strict license conditions, and the sex offenders' register. These constraints have actively exclusionary consequences, creating obstacles for people who wish to desist from crime, and obstructing social and moral rehabilitation.
The configuration is different in the Norwegian case, where the most prominent type of rehabilitation is legal. People, including those convicted of sex offenses, face fewer restrictions on their release from prison, and on the whole they enjoy the same rights and duties as the rest of the citizenry. On paper, then, completing a sentence for a sexual offense involves something that comes close to Beccaria's aim of the "full restoration of the formerly errant citizen" (McNeill, 2012, p. 5). Furthermore, although the "what works" wave of cognitive-behavioral programs had some influence in the Norwegian Prison Service (Ugelvik, 2014), its broad and penal-welfarist form of rehabilitation anchored in legal rights and continued access to work and education has never been replaced by a narrow and reductionist form of psychological rehabilitation. The laissez-faire tendencies of the Norwegian state, however, at least when it is operating on people convicted of sex offenses, combines with the lack of significance afforded to psychological rehabilitation to mean that imprisonment in Norway is not necessarily experienced as either meaningful or actively inclusionary.

CONCLUSION
We opened this article by briefly telling two interlinked stories about penal change-the first was of growing exclusion in neoliberal states, and the second was of the persistence of inclusionary penal practices in the "exceptional" Nordic states. The findings we have presented here nuance both stories. First, they highlight the level of regulatory effort that goes into enforcing the exclusion of people convicted of sex offenses in a neoliberal country. Second, they demonstrate that the Norwegian legal system punishes people convicted of sex offenses in a significantly less interventionist way than might be expected. Future research should consider whether the patterns we have identified hold for other groups of convicted people, and whether they hold in other neoliberal and social democratic punishment systems. Our findings also present a challenge to binary conceptualizations of exclusion and inclusion, as well as to the assumption that moving away from the former and toward the latter is straightforwardly desirable. As Cohen suggested, although inclusionary forms of social control can be deeply invasive and normalizing, they can also be symbolically thin (1985, p. 234; see also Braithwaite, 1989, pp. 155-157), and can result in "a benevolent licence for neglect" when enacted by a "conservative laissez faire state" (p. 268, emphases in original). As our participants' experiences highlight, there are multiple different forms of inclusion, and restoring legal rights is not the same as meeting people's expressed desires for psychological intervention, social integration, or morally meaningful rituals. Our participants' experiences also support Cohen's argument that the choice neither to construct nor to emphasize difference can lead to the specific needs of particular groups being overlooked. This was arguably the case in Norway, where the Prison Service has been reluctant to emphasize the difference of men convicted of sex offenses by separating them from "mainstream" prisoners. This may have made formal interventions or morally meaningful conversations with prison staff members harder to facilitate, as many men convicted of sex offenses self-isolate to reduce the risk of victimization. That said, the research we conducted in England & Wales should remind us that interventions are often damaging, and in particular that some forms of psychological rehabilitation-the only process we described in that jurisdiction that may have been inclusionary in its intent-impose more cognitive demands than they produce cognitive insights. Similarly, it should remind us that creating categorical differences can lead to new forms of control being imposed rather than to unique needs being met.
We are hesitant to conclude this article by calling for the more invasive forms of punishment that might facilitate substantive inclusion. Instead, we urge scholars to think seriously about the balance between formal and substantive processes of inclusion-how to recognize and address the pains that come with poor mental health, deviant sexualities, entrenched loneliness, and the desire for moral repair-while respecting the right for individuals to be protected from invasive or unreliable forms of state power. It seems possible to imagine how a state could balance the need to provide psychological treatment and social reintegration (through developing and funding community groups, for instance) without either imposing them or trampling over individual rights and liberties. It may be more challenging to develop a system of state punishment that takes a close enough position to its participants to enable moral rehabilitation, at least without the system undergoing radical, structural change (Sered, 2019). Criminologists may be more familiar with criticisms of the potential excesses of morally and emotionally expressive forms of punishment (Garland, 2001) than they are with critiques of its more expressionless, bland, and meaningless configurations. As Durkheim (1933), Braithwaite (1989), Duff (2001), and Maruna (2011) reminded us, however, our responses to crime are ritualized, and rituals must mean something if they are to work.