Restorative Justice, Crime Victims and Penal Welfarism. Mapping and Contextualising Restorative Justice Policy in Scotland

While the criminological literature is abundant in studies on the functioning and effectiveness of restorative justice, there is a lack of engagement with policy on this subject, in spite of the increasing incorporation of restorative justice into policy frameworks. This paper contributes towards addressing this gap by mapping and discussing Scottish policy on restorative justice. The focus is placed on how policy frames restorative justice around certain problems, subjects and objects, reconstructing their underlying assumptions. Additionally, the paper analyses the cultural and political context within which those representations have emerged. From this perspective, it enhances the legibility of the particularly slow and fragmentary development of restorative justice in Scotland, compared to the rest of the UK. More generally, the paper provides an original (and relevant beyond British borders) case study on the interplay between cross-national and local, cultural and political factors in influencing policy change.


Introduction
Over the last three decades, there has been a relentless proliferation of studies on restorative justice (RJ) (Daly, 2016). This subject has been investigated from multiple (and often contrasting) normative, theoretical and methodological perspectives. However, while the criminological literature is abundant in empirical investigations on 'how it works' (see Crawford and Newburn, 2003;McEvoy et al., 2002;Rossner, 2013;Shapland et al., 2006;Zehr, 2005;Zernova, 2007), there is a lack of engagement with policy on RJ, in spite of the increasing regulation of this subject (Aertsen et al., 2006;von Hirsch et al., 2003). When policy on RJ is considered, then, this rarely happens in an analytically nuanced, historically documented and critical fashion. Moreover, the geographical distribution of studies on RJ in general, and on RJ within policy in particular, is uneven (Aertsen et al., 2006), mainly due to the unequal development of RJ in practice. In Scotland, the extant research on this subject consists mainly of empirical evaluations of specific RJ services (Kearney et al., 2006;Kirkwood, 2009) and appears as scant and possibly outdated. The lack of interest around how policy frames RJ is problematic, since the Scottish Government has recently established a new policy framework on RJ aiming to support the wide development of RJ services across Scotland by 2023 (Scottish Government, 2017.
Addressing this gap, this paper scrutinises the shifting representations of RJ within criminal and penal policy in Scotland, investigating both their underlying assumptions (Bacchi, 2000;Foucault, 1970Foucault, , 1972 and cultural and political context (Maglione, 2017a(Maglione, , 2017b(Maglione, , 2018a. Conceptually, the paper traces original historical links between the 'return of the victim', penal welfarism and RJ. From this viewpoint, it argues that the recent emergence in Scotland of a victim-centred RJ is related to global cultural pressures mediated by local professional conflicts set against a specific political landscape and that this dynamic may continue to influence the development of RJ in this geographical context. Furthermore, it brings about new material for policy comparisons between Scotland and the rest of the UK and Europe, providing a significant case study on the dynamic interplay between cross-national and local, cultural and political factors in influencing policy change.

Background
Scottish criminal justice is characterised by 'its own legal framework, set of institutions, actors and processes and, some would say, set of values' (Morrison, 2017: 1). Scotland has its own police, prosecution service, criminal justice courts and penal system often described as underpinned by a commitment to the traditional public sector model which emphasises egalitarianism and cooperation with policy communities (Keating, 2005).
On closer inspection, the distinctiveness of such institutions with respect to the rest of the UK (particularly England and Wales) appears as partial and contested. In fact, the relationships between Scottish and Anglo-Welsh criminal justice have contemplated historically both divergences and convergences, particularly over the last 30 years (Croall, 2006;McAra, 2008). Significant convergences were apparent immediately after devolution, 1 when the Labour-Liberal Democrats coalition, leading the Scottish parliament between 1999 and 2007, launched a number of initiatives mirroring New Labour's 2 criminal justice reforms in England and Wales (Hutton and Tata, 2010: 274;Mooney et al., 2015: 213). This was especially the case of youth crime and antisocial behaviour policy or community justice initiatives inspired by 'third way' values (Croall, 2006: 588). The rise of the Scottish Nationalist Party (SNP), since 2008, has instead fuelled policy divergences animated by civic nationalism infused with social-democratic values. The creation of a national police force and the simplification of community justice local arrangements can be read as informed by such local political commitments (Buchan and Morrison, 2020). Nevertheless, managerialism and austerity have also played their role here, exactly like 'south of the Border' (Grant and McNeill, 2014: 149;McAra, 2008: 490;McVie, 2011: 108).
Within this context, the development of RJ occupies an interesting position. Similarly to England and Wales, in Scotland, the emergence of RJ was anticipated by mediation and reparation projects run by third sector organisations, in the late 1980s and throughout the 1990s (Marshall, 1999: 16;Miers, 2004: 27). However, while in the rest of the UK similar projects turned into established services backed up by ad hoc policy, in Scotland their development has been much slower and with very limited impact on the local criminal justice system (Miers, 2004: 28). Only in the early 2000s, in fact, specific yet sparse policy regulations on RJ emerged, particularly in the youth and community justice sectors. In the first case it consisted of declarations of principles, often sitting at odds with the purported Scottish youth justice's uniqueness, while in the latter, initiatives described as 'restorative' were characterised by a distinctive moral-authoritarian approach, which possibly contributed to their dismissal.
During the 2010s, policy anchorages for RJ were provided by the development of the 'Whole System Approach' (WSA) and to the creation of the Community Payback Order (CPO). The WSA, a set of diversionary instruments enacted locally by partner agencies (police, social work, etc.) to address young people offending, included the possibility of referrals to targeted interventions like RJ (Gillon, 2018). Similarly, the CPO offered an instrument to do RJ within the context of community penalties.
In the area of crime victims, until very recently, policy was completely silent on RJ (differently from England and Wales). However, it was by means of crime victims' policy that in 2017 the Scottish Government fully recognised the existence of RJ. The first national 'Guidance for the Delivery of Restorative Justice', in fact was meant to implement the EU Directive 29/12 on victim's rights, promoting the development of RJ services in a 'victim-focused manner ' (2017: 2). In 2019, the Guidance was integrated by the first 'Restorative justice: Action Plan' aiming to develop RJ widely across the country by 2023 (2019: 3).
In spite of the relatively specific historical development of Scottish RJ policy and of the recent thrust of policy recognition, there is a complete lack of knowledge regarding how RJ is framed within policy in this geographical context, i.e. how policy-makers have 'appropriated' RJ, which RJ's values, aims and practices have been foregrounded and which silenced and under which social and political circumstances.
Against this backdrop, this paper offers a contextualised 'policy map' of RJ in a bid to further the critical understanding of (a dimension of) the institutionalisation of RJ in Scotland. This work is more broadly significant due to the fact that the increasing incorporation of RJ into policy is a worldwide phenomenon, also fuelled by the RJ movement's conviction that policy is a necessary condition for doing 'more' RJ (Poama, 2015). However, such an incorporation is neither culturally nor politically neutral, rendering progressive scholars' critical engagement with policy on RJ particularly timely.

Methodological and Theoretical Considerations
This paper is premised on the idea that policy 3 creates its own 'objects', 'problems' and their 'subjects' rather than responding pragmatically to objective issues 'out there' (Bacchi, 2000(Bacchi, , 2009(Bacchi, , 2016Bacchi and Goodwin, 2016). Compared to more conventional technical analyses of the implementation and impact of policy (Weimer and Vining, 1999), this approach reconstructs how policy defines and addresses the social world by emphasising certain aspects while silencing others, questioning both the political rationales and the implicit cultural framings upon which these generative framings rest.
From this perspective, this paper firstly maps policy texts on RJ and then, 'working backward', draws inferences from such documents regarding the relations between context-specific actors (e.g. political parties, criminal justice professionals, victims' organisations, etc.) and their larger cultural and political structures, which have plausibly worked as the conditions of possibility for those texts (cf. Loader and Sparks, 2004: 12). The goal, then, is not improving policy, but providing granular insights into policy texts' internal economy and political drives, that is, critique (Foucault, 1970(Foucault, , 1972. The first step in applying this approach is to generate a policy corpus. This paper examines policy documents produced in Scotland up to date (2019) which comprehensively address RJ in domestic criminal justice settings. Documents were retrieved by using official Scottish online publications repositories 4 as well as the online repository held by main provider of RJ in Scotland (SACRO) 5 (10 documents). Linear snowballing from the retrieved policies to other relevant material was also used, in order to widen the dataset. This led to the inclusion of 4 further documents with no reference to 'restorative justice' and 3 international policy instruments both recalled by Scottish policy on RJ (seven documents). Finally, the paper drew on available scholarly literature on Scotland's criminal justice system (McNeill, 2010;Moody, 1999;Munro, 2015;Warner, 1992) to further locate relevant material (9 documents) (n ¼ 26). The next step unpacks the 'problems', 'subjects' and 'objects' of RJ (Bacchi, 2009) within such a policy corpus. The target is on how RJ policy conceptualises certain issues as matters unwelcome or harmful and needing to be dealt with and overcome; on how it foregrounds the ideal actors of such 'problems', excluding other possible stakeholders; and on how it shapes certain phenomena as 'objects' upon which those subjects act, within the broader context of those problems.
The final part of the paper explores the shifting cultural and political milieu within which such representations have developed. The aim, here, is to enrich our understanding of policy on RJ by placing it within specific cultural, political and professional transformations involving the Scottish criminal justice field over the last 30 years. In this way it is possible to capture some long-term links between policy and society which help appreciate critically the regulation of RJ, unearthing some of the rationales for both its slow and fragmentary development and for its recent popularity.
In terms of limitations, policy documents are written by different actors, for different audiences and with different goals. This constitutive diversity may hinder the identification of common problems, subjects and objects. However, since the paper's primary aim is to discern overarching policy representations, generalisation plays the role of a heuristic strategy consistent with that purpose (Garland, 2001: viii). Moreover, a gap exists between policy programmes and their implementation. This paper does not address such a gap, for it is not a research on 'how RJ works' but an investigation of how policy shapes RJ. A further limitation is that locating 'grey' policy documents published by state agencies is challenging, for they are often stored incoherently in online repositories. Therefore, it should be cautionary acknowledged that the corpus generated here is not exhaustive of the entire Scottish policy on RJ; yet, the rich and diverse range of official documents retrieved offers a preliminary, wide and reliable ground for an insightful analysis. Finally, regarding the policy contextualisation, this is constitutively partial since it is built on inferences based on policy texts and discussed by using the available literature. Hence, it should be read as a range of interpretive hypotheses aiming to critically appraise extant policy and generate further debate.

Mapping Restorative Justice Policy in Scotland
RJ occupies a relatively marginal space within Scottish criminal and penal policy. This section seeks to delimit such a space, supplying the reader with a 'map', preliminary to the historical investigation presented in the second part of this paper.

Youth Restorative Justice
Early youth policy on RJ consisted of a few declarations of principles on the use of RJ within the youth justice context, mainly aiming to encourage 'the expansion of young offender mediation and reparation schemes' (Scottish Executive, 6 2002a: 9) and to create the opportunity to 'engage in a [restorative justice] scheme, where appropriate' (Scottish Executive, 2002b: 7).
In 2005 the first ad hoc policy on the role of RJ in youth justice was adopted by the Scottish Executive (2005a). This defined RJ as 'any process in which relevant parties participate together actively in the resolution of matters arising from the offence, generally with the help of a facilitator' (Scottish Executive, 2005a: 13). A basic condition to activate such a process was that the 'person charged', aged between 8 and 17, accepted 'some or all responsibility for the offence as described by the Children's Reporter 7 ' (Scottish Executive, 2005a: 3). Overall, while this definition seemed to reflect a wellknown description of RJ in the UK, 8 it included a noteworthy detail: '[RJ is] primarily designed to address an individual offence or episode, rather than patterns of offending behaviour' (Scottish Executive, 2005a: 2). Although the theoretical assumptions underlying this addition will be addressed in the second part of this paper, it should be cursorily noted here that this clause introduced a significant restriction to the RJ's remit. In fact, as the Scottish Executive stated, RJ processes could meet only a limited a range of 'needs of children who offend: for example, the need (a) to have access to educative experiences that will enable them to reduce their offending and develop as mature and responsible citizens; and the need (b) to be given the opportunity to restore their moral status and reputation [ . . . ]' (Scottish Executive, 2005a: 1).
Such basic values and conditions were reiterated in the following years (Scottish Executive, 2008;Scottish Government, 2013) adding some emphasis on youth offenders' duty to 'take moral responsibility for their part in what happened ' (Scottish Executive, 2008: 1.15) within the wider aim of 'empowerment of all parties [and] prevention of further offending' (Scottish Government, 2013: 142).
This overall approach still informs the extant Scottish policy framework on youth RJ which requires referrals in compliance with the welfarist principles 9 expressed by The Children (Scotland) Act 1995 and, from the 2010s, according to the Scottish Government's 'Whole System Approach for Children and Young People who Offend' (WSA) (Robertson, 2017). This means that RJ for youths is mainly a preventative tool, organised locally by WSA partner agencies (local authorities, police, educational services, etc.) targeting minor and first time offending and antisocial behaviour (Robertson, 2017: 38-39).
Youth Restorative Justice and Policing. There is a limited but significant number of policy regulations on police and RJ for youth offenders in Scotland. Police-led RJ programmes vary in terms of victim involvement, i.e. could contemplate some form of communication between 'victim' and 'offender' within a carefully supervised context or could be run without involving the direct victim at all (Scottish Government, 2010a: 31). 'Restorative warnings' and 'restorative warnings conferences', are the two instruments to implement RJ at police level (Scottish Executive, Children's Reporter and ACPOS, 2004). A police restorative warning is: 'A process, facilitated by trained personnel that involves the warning of an offender [first time young offenders from the age of 8 to 15 inclusive] while addressing the impact on the victim and the community', mainly by giving the victim the opportunity to be informed of the warning's 'outcome' (Scottish Executive, Children's Reporter and ACPOS, 2004: 2). In terms of rationales, a restorative warning is considered as a 'quick' measure, 'carried out by trained officers, and impresses upon the young person the impact of the offence on the victim or the community and on his/her parents, encouraging him/her to take responsibility for his/her actions and to understand the implications of future offending' (Scottish Executive, Children's Reporter and ACPOS, 2004: 2).
A 'restorative warnings conference', instead, involves also the victim and 'community members'. Such measures are described 'part of a wider range of preventative, diversionary and more intensive youth justice measures which the Scottish Executive has asked local youth justice strategic teams to develop to prevent, address and reduce offending behaviour among young people' (Scottish Executive, Children's Reporter and ACPOS, 2004: 2-3).
Overall, these measures profile Scottish restorative policing similarly to English and Welsh counterparts (Liebmann, 2007): a 'street-level' instrument of youth offender responsibilisation, with limited procedural safeguards, essentially justified by speedy proceeding, reduced labelling and less stigmatising effects, and wherein the 'victim' is often an optional participant.

Crime Victims' Policy and Restorative Justice
In Scotland, over the last 20 years, crime victims' policy has exponentially grown (Munro, 2015). However, the formal recognition of links between this and RJ is a very recent phenomenon. While the 'Scottish Strategy for Victims' (Scottish Executive, 2001), i.e. the first national document on crime victims in criminal justice, was completely silent on RJ, 10 the following 'Victims in the Scottish Criminal Justice System' (Scottish Executive, 2002c) recalled the EU definition of 'mediation in criminal cases 11 ' (p. 2), however postponing its implementation to 2011 (p. 18). More recently, the Scottish Government has acknowledged that RJ 'support[s] the strategic intention of involving victims more in the criminal justice system' (Scottish Government, 2010b: 6) as an instrument able to 'give victims voice' (Scottish Government, 2010b: 16).
A more important step in the victim-policy reception of RJ was the enactment of The Victims and Witnesses (Scotland) Act in 2014 in order to implement the EU Directive 29/12 on victim's rights. Here, nevertheless, the ratification of the international policy was selective, for the Directive's section on RJ (12.1) was not transposed into domestic legislation. The only reference was in the section 5 of The Victims and Witnesses (Scotland) Act, which deferred to the Scottish Ministers the possibility of issuing guidance about RJ. A definition of RJ, based on the EU Directive 29/12 on victim's rights, was however provided: (3) 'restorative justice services' means any 'process in which the persons such as are mentioned in subsection (1)(a) participate with a view to resolving any matter arising from the offence or alleged offence with the assistance of a person who is unconnected with either person or the offence or alleged offence'. In line with this apparent indifference towards RJ, the recent Victim's Code for Scotland (Scottish Government, 2015) was also devoid of any reference to RJ. The Code formally recognised the right to information, participation (including victim impact statements for more serious crimes), protection, support and compensation, but nothing was said about RJ (differently from the English 'Code of Practice for Victims of Crime' (Ministry of Justice, 2015)).
Recent Developments. The Scottish Government implemented the section 5 of the Victims and Witnesses (Scotland) Act in 2017 by publishing the first, ad hoc national document on RJ, that is, the 'Guidance for the Delivery of Restorative Justice in Scotland' (Scottish Government, 2017). This instrument is an attempt to 'to ensure that, where restorative justice processes are available, these are delivered in a coherent, consistent, victimfocused manner across Scotland, and are in line with the EU Victims' Rights Directive' (p. 2). RJ is described as a process of direct or indirect communication between the victim (who could also be a corporate or legal person) and the 'person who has harmed' (p. 4), characterised by the fact that 'the needs and interests of the victim [ . . . ] are imperative' (p. 7). Operationally, RJ ' [ . . . ] gives victims the chance to meet, or communicate with, the relevant people who have harmed, to explain the impact the crime has had on their lives' (p. 6). Such a process intends to generate benefits for the victim 'by giving them a voice within a safe and supportive setting and giving them a sense of closure' (p. 6) as well as for the person who has harmed 'providing [them] with an opportunity to consider the impact of their crime and take responsibility for it, with the aim of reducing the likelihood of re-offending' (p. 6). Indirectly, also the 'Principles and Purposes of Sentencing Draft Sentencing Guideline' (Scottish Sentencing Council, 2017) recognise the use of victim-centred RJ: 'Sentencing may also aim to recognise and meet the needs of victims and/or communities by requiring the offender to repair at least some of the harms caused; this may be with the co-operation of those affected' (p. 3). Finally, this victim-centred vision is recently echoed by the Programme for Government for Scotland 2018-19, within which the Government committed to the large-scale expansion of RJ 'with the interests of victims at their heart' (Scottish Government, 2018: 105).
In addition to the Guidance, the Scottish Government in 2019 enacted a 'Restorative Justice: Action Plan'. The Action Plan commits the Scottish Government 'to have Restorative Justice services widely available across Scotland by 2023, with the interests of victims at their heart.' (Scottish Government, 2019: 3). In doing so, it encourages a 'partnership approach involving organisations across the Scottish justice sector' which should then lead to develop a 'nationally-available model for RJ' (Scottish Government, 2019: 9). It also gives consideration to some of the challenges which may hinder such an ambitious plan, namely lack of public awareness, issues with information sharing, lack of resources and of training opportunities (Scottish Government, 2019: 8).
The 2017 Guidance and the following Action Plan are significantly different from the 'Best Practice Guidance' on RJ produced by Sacro and then acknowledged by the Scottish Executive in 2008 (Scottish Executive and Sacro, 2008). The 2017 document has as principal subject the 'victim', while the 2008 Guidance intended to strike a balance between the 'person harmed' (this semantic shift from 'person harmed' to 'victim' will be discussed below in the section 'The "return of the victim" and restorative justice'), the 'person who has harmed' and the community. Additionally, the 2017 Guidance links RJ to direct or indirect communication while the 2008 document also to cases with no communication involved between parties. It will be shown in the remainder of this paper that these apparently minor divergences can be interpreted in light of the different political and cultural conjunctures underpinning those documents.

Community Justice Policy and Restorative Justice
Historically, the articulation between community justice and RJ was expressed by the use of 'reparation' as an instrument to respond to crime 'in the community'. Placing the focus on recent policy on this subject a few instruments should be considered.
The Labour-Liberal Democrats Scottish Executive in 2003 stated that 'visible reparation to the community', through Community Reparation Orders (CROs), is one of the ways to hold offenders to account for their actions (Scottish Executive, 2003: 18-19). Within this context, 'programmes with elements of restorative justice, including reparation' could be a condition of a supervision requirement (Scottish Executive, 2003: 28). CROs were introduced by the Antisocial Behaviour (Scotland) Act 2004 as communitybased disposals for low antisocial behaviours, for aged 12 or more. However, a report on their effectiveness (MacQueen et al., 2007) showed limited numbers compared to target, contributing to CROs' dismissal in 2010, under significantly different political circumstances (see below the section 'Restorative justice, politics and penal policy').
In these early post-devolutionary years, RJ in community justice policy was mainly conceptualised as a measure for tackling antisocial behaviours in order to responsibilise the offender (Justice Committee, 2003: Part 2) and to help 'revitalise communities' (Scottish Executive, 2005b: 19).
A few years later, with the new Government led by the SNP, the Scottish Prisons Commission explicitly acknowledged in the 'Scotland's Choice Report ' (2008), that RJ could facilitate the offender 'pay[ing] back in and to the community' (p. 33), recommending the legislator to embed RJ into community justice. The Criminal Justice and Licensing (Scotland) Act (2010), partly responding to this recommendation, created the current policy anchorage for RJ in (adult) community justice settings, as a possible postsentencing instrument. The section 14, in fact, introduced the Community Payback Order (CPO), an instrument designed to ensure that those involved in offending behaviour 'payback' to communities and society in two ways: by obliging the offender to make reparation (normally unpaid work) or to address and change their offending behaviours to improve the safety of local communities, providing opportunities for reintegration as law-abiding citizens. CPO's underlying philosophy is not only to contribute towards community safety and protection but also social inclusion to support desistance from crime (Scottish Government, 2010b: 47). Currently, RJ finds in CPO a possible context of application, as linked to one of the requirements which the offender has to comply with. RJ could be organised either following the social worker's recommendation, as part of a supervision requirement or as part of a programme requirement, playing the role of a process by which to 'payback' to the community and to 'restore the balance upset by the offence' (McNeill, 2010: 4).

Unpacking Restorative Justice Policy
Scottish RJ policy is underpinned by a range of unspoken assumptions which this section seeks to unearth and discuss. Namely policy tends to justify RJ as a response to takenfor-granted 'problems' involving discrete 'subjects' and 'objects'. These problems/subjects/objects are formulated within different policy initiatives in ways which express different, and often inconsistent, understandings of the practice, staged against a changing cultural and political context.

Problems
RJ in Scottish criminal and penal policy seems to tackle only discrete, micro-social 'problems', that is, petty crimes and antisocial behaviours. This is not obvious since RJ has been slowly but increasingly used in the rest of the UK and worldwide to address complex conflicts ranging from domestic violence to terrorism (e.g. Liebmann and Wootton, 2008;Varona, 2014). Within the Scottish youth justice domain, RJ is intended to address a supposed interpersonal harm caused by the offender to the victim. Conversely, there is a firm denial that RJ could address social-structural issues driving youth crime and deviance (e.g. Scottish Executive, 2005a: 2). There is a tacit assumption that the type of offending targeted by RJ is rooted in the child's lack of understanding of the consequences of their behaviour. Offending has as preeminent consequence the child's loss of reputation and moral standing which can be restored by repairing the practical and/or moral harm (cf. McNeill, 2010). This representation is quite divergent from the construction emerging from Scottish crime victims' policy over the last decade. Here, the issue at stake is fundamentally the victims' 'lack of voice' within criminal justice, as a consequence of their predicated historical marginalisation, a common theme within RJ theory and practice (Marshall, 1996;Zehr, 2005). Greater participation entails having a say in the handling of a case and therefore counting as an active part of the process, in this way producing positive psychological effects, such as a sense of 'closure' and 'empowerment' (e.g. the 2017 Guidance). From this angle, the 2017 Guidance links RJ practice to direct communication as the main form of intervention, entailing the victim's empowerment by speaking out and being heard. Community justice policy, similarly to youth justice, stresses the responsibilising aim of RJ for petty criminals whose antisocial behaviour damages the fabric of idealised communities (e.g. Justice Committee, 2003: Part 2). Reparation, within this domain, is sufficient to mend that relational fracture caused by criminal actions (cf. Maglione, 2017b). Finally, from the police perspective, informal 'street-level' RJ warnings are efficient instruments to tackle antisocial behaviours, preventing prosecution's stigmatising effects and lengthy proceedings. In this context, the victim's lack of voice is traditionally a secondary 'problem' while it is possible to read through the lines a role played by RJ in terms of improving community-police relationships. This idea of RJ resonates with both the language of early reparative measures (as informal responses to petty crimes) and with communitybased moral discipline and multi-agency interventions (see 5.2).
Other possible 'problems' are silenced within Scottish policy. For instance, RJ as an instrument to address long-term conflicts instead of one-shot offending or victim's marginalisation. Offending is seen as an individual moral pathology which ought to be neutralised. RJ aspires to 'cure' this pathology, healing the victim and/or involving the offender, while neglecting e.g. the emancipatory potential of transgressions of established legal frameworks or the power relationships which contribute towards the definition of behaviours as 'crimes' and peoples as 'victims' or 'offenders'.

Subjects
Youth offenders are the typical targets of RJ interventions. Within Scottish policy, they are described as fundamentally immature one-time-offenders, who need reintegrative forms of responsibilisation (e.g. Scottish Government, 2010a: 31; Scottish Government, 2013: 142). This 'offender' exemplifies a wrongdoer who harms a discrete victim often due to a lack of emotional understanding (Gray, 2005). Shining through policy is the image of a 'wayward child' (Wright, 1996), to be reprimanded and reformed. The community's ties with such an offender are weak, but still exerting some forms of control over them. This offender is a flesh-and-blood individual encouraged to 'pay back' the human costs of their actions (as the 'Scotland's Choice Report' explicitly states). Such a rendering, which echoes early (1980s-1990s) RJ literature and policy in England and US (Zernova, 2007), ignores the possible macro-constraints of the offender's choice as well as the fluid transactions between these two dimensions, emphasising a distinctive type of agency and conveying images of physical harm. The 2017 Guidance proposes a significant shift, with respect to this representation: the replacement of the 'offender', still recurrent in the early 2000s policy, with the 'person who has harmed', possibly expressing the intention of reducing 'semantic' stigmatisation. Additionally, the 2017 Guidance explicitly states that such a person can 'be an adult, a child, a young person or a representative of a corporate or other body' challenging the idealised image of an 'embodied' youth offender.
The second subject is the 'victim'. This was an elusive actor in (youth) reparative measures until the early 2000s. In youth RJ policy, in fact, the emphasis is on the child/ offender, while the victim plays a less central role, mainly as recipient of reparation and instrumental to the restoration of the child's moral standing in the community. The 'victim' entered the Scottish RJ policy scene thanks (mainly) to international legislation, appearing as a central subject in the 2010s documents (e.g. Scottish Government, 2013). It should be noted, here, that in the Anglo-Welsh RJ policy the victim has occupied a central role since the late 1990s (Liebmann, 2007;Marshall, 1996Marshall, , 1999. In the most recent 2017 Guidance, RJ is conceptualised as a victim-centred instrument, in compliance with the EU Directive 29/12 on victim's rights. Here, the victim is portrayed as a two-faced subject: on one hand, an agentic and resilient individual undertaking decisions and actions; on the other, a disempowered and embodied entity, in search of safety and recognition. RJ grants victims a proactive 'agency-based' solution to victimisation, driven by their needs (or at least by some of them, i.e. their 'lack of voice') which are explicitly recognised by policy (cf. Maglione, 2017a).
The 'community' appears as delineated in slightly different ways over time. In the early 2000s (particularly post-2003, i.e. during Labour-Liberal Democrats coalition) policy, the community was the recipient of 'visible reparation' activities undertaken within a RJ setting (e.g. Scottish Executive, 2003: 28). Offenders were expected to account in this way for their wrongdoing to the idealised community they 'belonged to'. Such a representation resonated with an authoritarian-moralising understanding of the community (an 'antisocial community' (Mooney et al., 2015: 215)), typical of New Labour approach to antisocial behaviour. This community, in fact, is construed as a lawabiding collective subject, a local complement to 'state' and 'society', harmed by crimes. Such an actor was slowly eclipsed in Scottish RJ policy from 2010s by the increasingly central role played by the 'victim'. The 2017 Guidance, for instance, dedicates minimal space to this subject. It conceptualises the community as a possible victim (1.4) and the other two references are to contrast RJ to restorative practices and mediation. This only partly reflects the content of the 2008 Guidance which dedicated more space to this subject. Here, there were specific provisions stressing the facilitator's role to 'balance the concerns of the person harmed, the person responsible, and the community' (p. 12). Again, this may be a consequence of the changing political landscape (see below the section 'Restorative justice, politics and penal policy') as well as related to the influence of the EU legislation.
In short, the subjects which populate RJ appear as idealised 'victims', 'offenders' and 'communities' with certain needs, wants and desires (cf. Maglione, 2017aMaglione, , 2017bMaglione, , 2018a. It is arguable that such subjects do not properly fit many types of individuals with a stake in a crime. Collective victims of economic crimes, powerful victims, victims with shared responsibility, vulnerable offenders and deviant communities are hardly consistent with the ideal subjects of RJ. Consequently, problems in terms of proposing and practising RJ for these latter categories may likely arise.

Objects
The two central (and interrelated) objects of RJ policy are 'crime', construed as the legal expression of non-functioning interpersonal relationships, and 'harm', that is, the human consequences of legal transgressions. In both cases Scottish policy does not provide distinctive constructions, since similar representations can be found within other geographical contexts (O'Mahony and Doak, 2017). However, a few salient elements should be highlighted. In RJ policy, a 'crime' is a wrongful behaviour since it harms the material victim and/or the community; accordingly, the expected outcome of a RJ process is the reparation of harm and reintegration into the community (especially in the early 2000s policy) as well as the maximisation of the offender's awareness of the crime's impact and the victims' empowerment (post-2014 policy). Both discursive objects are underpinned by the idea that legal transgressions are the expression of interpersonal conflicts which 'belong' also to their direct actors, who are entitled to have a say in managing the crime/harm's consequences. At the same time, crimes remain public wrongs, insofar as the harm has a public relevance, justifying the criminalisation. It is irrelevant under which circumstances behaviours are labelled as crimes. It is accepted that certain actions are 'criminal', implicitly endorsing both primary and secondary criminalisation. If one participant contests the nature of 'crime/harm' of their action that would automatically exclude that person from participating in the restorative encounter (e.g. Scottish Executive, 2005a: 3;Scottish Government, 2017: 10). In policy documents enacted over the last decade, RJ is particularly portrayed as a response to crimes/harms by focussing on whole persons, their relationships and the whole of the problem, emphasising material repair and offenders' responsibilisation (early 2000s policy) and providing closure for victims (2010s). A further object of RJ policy is 'conventional' criminal justice. Such an object is not an autonomous entity, with defined conceptual boundaries, but a support of the definition of RJ, a means of stabilising RJ from outside (Halperin, 1995: 43). 'Conventional' criminal justice identifies a strategic backdrop against which RJ is consistently described. In this context, RJ is represented as a response to needs triggered by the (dis)functioning of criminal justice. It is worth noting that policy documents do not construe 'conventional' criminal justice as a competitor of RJ, in this way silencing the radical understandings of RJ as an alternative to criminal justice (Johnstone and Ward, 2010).

Contextualising Restorative Justice Policy
Scottish RJ policy has developed under a fluid range of historical circumstances. This section focuses on a number of transactions between cross-national socio-cultural shifts and the ideologies and actions of local actors whose appreciation may enhance the legibility of Scottish RJ policy, its rationales and possible future applications.

Restorative Justice, Politics and Penal Policy
'Criminal justice policy making [is] one of the emerging fault lines around which questions of devolution [in] Scotland were being articulated' (Croall et al., 2015: 3). From this perspective, Scotland's delayed interest in RJ (particularly in victim-centred RJ) can be interpreted as a 'diverging' response to the expansion of RJ in England and Wales since the late 1990s. Until very recently, in fact, RJ was seen by Scottish policymakers, victims' organisations and social work professionals as rather incoherent with Scottish penal welfarist values (McAra, 2008) (see next section). This led to a generalised political indifference towards RJ, paralleled by investments on 'local' justice initiatives such as the Children's Hearings System 12 and criminal justice social work (Croall, 2006: 590), in this way demonstrating how Scottish penal institutions could perform independent statehood by retaining (supposedly) distinctive policies (cf. Buchan and Morrison, 2020;Hobbs and Hamerton, 2014: 146). Yet, in the early 2000s, some signs of convergence appeared through the Anti-Social Behaviour Act (2004) (Croall, 2006: 599;Muir, 2010: 175). Within this legal framework, CROs intended to implement a moralising version of RJ, which resonated with Anglo-Welsh youth RJ initiatives (Crawford and Newburn, 2003). This convergence was undermined once the political atmosphere changed with the SNP gaining power in 2007, and with the purported 'return' of the penal welfarist approach (cf. Mooney et al., 2015: 218).
Currently, the Scottish political landscape is characterised by a stable SNP majority ruling the country, and with different manifestations of independent statehood compared to the early 2000s. RJ in this context assumes a less controversial position, becoming actually a point of partial convergence with Anglo-Welsh policy, promoted by the full spectrum of Scottish political forces and backed up by the EU legislation. One of the possible reasons for this, is the reframing of RJ as a victim-centred justice in a context where ideal victims (i.e. the key subjects of RJ) are considered as representatives of public interest, and therefore their protection is exploitable for political purposes (Garland, 2001). Promising protection for crime victims is an effective vehicle for attracting political consensus, since victims are presented as respectable, ordinary, law-abiding (i.e. ideal) citizens whose experience becomes universal as the (perceived) dangers they face (Maglione, 2017a). Protecting crime victims is integral to both the SNP and to other Scottish political groups' agenda. SNP, Labour and Conservative MPs, in fact, are now jointly upholding victim-centred RJ as a way forward for Scottish criminal justice, as a recent ad hoc parliamentary debate has shown. 13 Here, it was said that 'Restorative Justice is about the victims' voice being heard and questions being answered. [ . . . ] The main benefits of Restorative Justice are to the victim.' However, and once again, the convergence with England and Wales is partial, since the same rapporteur added at the end of summary that 'Consideration will be given to developing a National Strategic Framework, and the Scottish Government are working with Community Justice Scotland to consider this'. This means that an expert-led organisation with no equal 'south of the Border' i.e. Community Justice Scotland, is entrusted with the task of leading the development of RJ in Scotland, differently from England and Wales where politically elected figures -Police and Crime Commissioners -oversee the implementation of RJ.

The 'Return of the Victim' and Restorative Justice
The history of the connections between Scottish victims' organisations and penal policy making can further help making sense of the RJ policy imaginary.
The starting point here is that 'consideration for victims needs is new in Scotland' (Munro, 2015: 151). The rise of interest around crime victims in fact dates back to the 1980s thanks to three concomitant events: the publication of the first British Crime Survey (Scotland) (1984), the execution of pioneering research on victims (Munro, 2015: 154), and the emergence and strengthening of crime victims agencies, with Victim Support Scotland (VVS), established in 1985 (Moody, 1999: 322). In this period, victims' major concerns were: 'the lack of information regarding case progress and outcome; witness support and protection; the impact of the crime and how that is communicated to the court; court accommodation; and compensation' (Moody, 1999: 326). Nonetheless, the rise of sensitivity towards victims' role within criminal justice proceedings did not facilitate the development of RJ in Scotland, differently from England and Wales, where the Home Office produced three Victim's Charters which included thorough references to RJ (Ministry of Justice, 2015).
The relationship between victims' organisations (particularly Victim Support) and RJ in Scotland has been historically marked by scepticism and reticence. A possible reason for this has been the link between RJ and offender-led diversionary schemes, which in fact characterised early Scottish mediation and reparation policy and practice (Munro, 2015: 162). The growth of diversionary schemes in Scotland was triggered by the findings of the Stewart Committee (1983) and the subsequent introduction of legislation in Section 56 of the Criminal Justice (Scotland) Act 1987. The Stewart Committee encouraged the adoption of diversionary schemes as less labelling, crime sensitive, offender specific and cheap instruments. Victims' organisations saw in these measures 'a means of reducing the costs and pressures on the courts while also providing a better means of dealing with minor offences' (Warner, 1992: 198). However, they also highlighted that there was need to place 'greater emphasis on the benefits to the victim, reflecting an increased concern that the victims of crime be considered more in criminal justice processes' (Warner, 1992: 200;cf. Davis, 1992). From the victims' advocates point of view 'The problems of linking reparation and mediation and diversion [would have been] alleviated by a clearer separation of the approach from its function [ . . . ] The service could be offered to victims independently of criminal justice sanctions if there is serious commitment to provide a service for the victims of crime' (Warner, 1992: 207;cf. Kearney et al., 2006). This statement crystallised a recurrent objection against RJ put forth by victims' organisations: (diversionary) RJ is not a victim service, and its claims to be victim-sensitive should be cautiously (if not suspiciously) appraised.
Such a diffidence has been (partly) overcome only very recently due to the UN and EU policy, and particularly the EU Directive 29/12 on victim's rights. As a consequence, a victim-centred RJ policy in Scotland has emerged, which could be read as an expression of the larger global institutionalisation of the 'return of the victim' in criminal and penal policy (Garland, 2001: 11).

Penal Welfarism, Social Work and Restorative Justice
As cursorily mentioned above, an interpretive key to untangle the policy images of RJ in Scotland is to connect them with a discourse -penal welfarism -supposedly informing, inter alia, criminal justice social workers' activity particularly within youth justice (McAra, 2005). The focus is on the role that a number of beliefs held by these professionals have played in influencing Scottish penal policy making (Croall et al., 2015: 7). 'Penal welfarism' is a constellation of values such as rehabilitation, promotion of behavioural change and reintegration of offenders into community (McAra, 2005: 297). This broad social justice commitment to responding to crime, mediated by an expert-led approach to doing policy and cast against a cultural backdrop of 'greater collectivism of civic culture' (Croall, 2006: 590), stemmed from the work of the Kilbrandon Committee in 1964, forming the basis of the Social Work (Scotland) Act 1968 and then of the Children's Hearing System. Today, this discourse, residualised and re-shaped by global factors, is purported to underpin criminal justice agencies' workings, in the areas of youth justice and criminal justice social work (McAra, 2005). The coherence of this framework, the degree to which it is actually implemented, and its interaction with other (e.g. actuarialist, managerialist and populist) discourses, is a complex theme (see e.g. Croall, 2006;Mooney et al., 2015) whose investigation falls outside the remit of this paper. Here, instead, the penal welfarist discourse could be used heuristically to make sense of a popular critique towards RJ levelled by Scottish social workers in the late 1990s and early 2000s. Such a critique has possibly both limited RJ's development and shaped some of its policy representations, by putting forth two fundamental claims (Whyte, 2002). Firstly, it was highlighted a tension between the emphasis on the victim's needs, at the heart of RJ, and the attention to the child/ offender needs, at the heart of Scottish youth justice. As the then-director of the Criminal Justice Social Work-Development Center for Scotland argued (2002: 4): ' [ . . . ] there are signs in some jurisdictions that a primary motivation for introducing restorative measures may be to address the needs of victims and not the needs of the young person and to reduce the resource demands of adversarial court systems'. Secondly, it was pointed out that RJ fails to address structural issues underpinning offending and therefore it lacks social justice potential. Whyte denounced 'the limited capacity to impact on other social factors which sustain and support offending behaviour' and that 'For those with multiple difficulties, however, this process is likely to require supplementary resources and positive assistance for the young person and their family if it is to have a long term impact on their offending behaviour ' (2002: 4). This resonates with the Children's Hearing System's understanding of RJ as a process 'primarily designed to address an individual offence or episode, rather than patterns of offending behaviour' (Scottish Executive, 2005a: 2). Clearly, this view limits RJ's reach and potential, relegating this instrument to a response for petty crimes committed by young people, chiefly justified by a nonstigmatising and managerial approach -an image long overcome in the US and Europe, while still underlying Scottish RJ policy.

Restorative Justice Within Scottish Policy: Global Drives and Local Barriers
The problems, subjects and objects analysed above as well as their cultural and political platform are characterised by connections, tensions and overlaps.
Scottish RJ policy appears to suffer from what could be called the 'Elmira's complex 14 '. This means that RJ is construed as a response to minor crimes committed by one-time offenders against innocent victims, and not a rejoinder to behavioural patterns rooted in socio-structural inequalities or complex social conflicts. This makes RJ a 'lower rank justice' ancillary to penal responses inspired by Scottish penal welfarism. Communities feature within RJ in a way which is inversely proportional to the victims' rise. The stabilisation of a new (SNP-led) political landscape, in fact, was one of the conditions for re-signifying and limiting their role in RJ. 'Victims', who replaced 'the person harmed' of the early 2000s policy documents as a consequence of the influence of European policy, are now at the centre of RJ. These representations silence alternative images of RJ e.g. as an instrument to tackle with complex power imbalances (e.g. environmental damages, political violence or racialised conflicts), as a progressive framework to contest criminalisation processes or as a measure seeking to overcome the victim-offender dichotomy (Pavlich, 2005;Maglione, 2019aMaglione, , 2019b. While these representations can be found also in other geographical and policy contexts, in Scotland their cultural and political background is characterised by some specific features. The apparent limited consistency between RJ and penal welfarist values may have prompted a sustained suspicion around RJ and the delayed development of policy (and practice) on this subject in Scotland. Historically, criminal justice social workers have advocated for RJ as a pre-prosecution diversionary mechanism subsidiary to youth justice institutions. This has been partly countered by Scottish victims' organisations who have criticised diversionary RJ as an offender-led measure. The lack of support from both offender-centred agencies and victims' organisations has doomed RJ to a decades-long limbo of fragmentary projects with marginal impact, infused with perceived ambiguity. This local bottom-up dynamic, in turn, has been only recently (and partly) mitigated by a global phenomenon, the 'return of the victim' institutionalised by the EU Directive 29/12 on victim's rights. This has facilitated top-down the current emergence of victim-centred RJ policy in Scotland. 'Mitigating scepticism' does not mean overcoming the sedimented doubts around RJ. It is possible that forms of ideological opposition against RJ may still be enacted 'from the ground' (Scott, 1990: 192), but clearly this hypothesis demands an empirical test.
These preliminary findings afford significant material for comparison with England and Wales. Although this is not the place where such a comparison could be conducted, it should be cursorily noted a number of emerging convergences and divergences which would warrant further investigation. There are similarities between the England and Wales and Scotland regarding the early resistance of victims' organisations towards a perceived offender-led diversionary RJ (Davis, 1992;cf. Marshall, 1996). However, in Scotland this resistance has been further strengthened by policy actors, criminal justice agencies and stakeholders embracing penal welfarist values and performing independent statehood, which played a less significant role in shaping RJ policy 'south of the Border' (Crawford and Newburn, 2003;Liebmann, 2007). Furthermore, a partial commonality is the role played by global factors (EU policy) in shaping RJ as a victim-centred instrument. Yet, it should be emphasised that a victim-centred RJ policy in England and Wales was already available in the late 1990s and that the recent EU legislation has merely strengthened such an approach (Maglione, 2020). In Scotland, instead, the implementation of the EU Directive on victim's rights has facilitated the emergence of an unprecedented victim-centred RJ policy framework. A further difference is that in England and Wales the New Labour's idea of moral-authoritarian communities is still the backdrop of RJ policy (particularly within youth justice) (Crawford and Newburn, 2003). Else ways, in Scotland, the SNP has advanced ideas of 'community' (i.e. local/national source of civic participation, deliberative processes and partly site of cost-saving political reorganisation) relatively different from the New Labour's one.
Overall, the 'return of the victim' appears as the most powerful drive of policy convergence on RJ 'across the Border', even though declined in different ways and according to different time-scales, due to dissimilar local, cultural and political circumstances. Conversely, the combination between Scottish penal welfarism rhetoric and nationalism seems one of the main roots of policy divergence.

Conclusions
This paper is informed by the idea that bringing to the fore policy's unspoken assumptions and their cultural/political context, would generate some awareness of the bonds and signatures which drive the institutionalisation of RJ in context. With this in mind, after an original recognition of the Scottish policy on this topic, within youth justice, victim-policy, policing and community justice, the paper teased out the problems, subjects and objects of RJ, highlighting their underlying assumptions. An interpretive analysis of the cultural and political milieu of these representations was also profiled. A significant finding emerging from this work, which would call for further research, is around the relationships between the institutionalisation of 'the return of the victim' (embedded in international policy instruments), local politics, penal welfarist rhetoric and RJ. Particularly, it should be considered whether and how the slow but increasing expansion of RJ (related to the political exploitation of idealised victims as representatives of public interest) may interact with local values ingrained in criminal justice practitioners' cultures, possibly modifying in turn the nature of RJ services which rely on those professionals' support. This subject merits further (empirical) research since it could lead to a deeper understanding of the role of RJ in contemporary criminal justice systems as well as of its political import, from a larger perspective of locally acted and globally influenced penal policy change.
There is a final aspect which is at stake in this work and from which a number of implications could be drawn. The critical scrutiny of how state agencies 'think of' RJ, according to specific yet tacit rationales, provide material to assess how the 'Leviathan' appropriates community-based and practice-led forms of justice. It is certainly possible to imagine RJ targeting complex social harms, supporting stakeholders in dealing with their power differentials, recasting conflicts and harms along the lines of gender, ethnicity and economic precarity, using a non-stigmatising language and disavowing violence (Maglione 2018b). The problem is whether such a radical RJ is compatible with penal policy as an expression of the state control over the social world. From this angle, a context-sensitive analysis of penal policy is both a condition for discerning some of the rationales and possible effects of the institutionalisation of RJ, and a step towards imagining and practicing a different (and hopefully better) RJ. 9. They are: The welfare of the child is the paramount consideration; no compulsory intervention should be made unless it would be better for the child than no compulsory intervention at all; and that children should be given an opportunity to express a view and, if they do so, consideration should be given to the child's views. 10. This is interesting since the Strategy was based on the UN Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power (1985) which explicitly considered the role of RJ for supporting crime victims (sections 43. 1.4; 43.5.17; 44.1.6; 44.13.36). 11. 'The search, prior to or during criminal proceedings, for a negotiated solution between the victim and the author of the offence, mediated by a competent person' (EU Framework Decision On The Standing Of Victims In Criminal Procedure 2001: 2). 12. Scottish care and youth justice institution which works as a panel with lay judges. 13. Members Debate on the 22nd of May 2018. The motion was brought by MSP Liam Kerr (Scottish Conservative and Unionist). 14. Elmira is the Canadian town where the first RJ intervention was registered, involving two drunk teens who went on a night rampage damaging the unaware residents' properties (Peachey, 1989).