• 1 Introduction

  This article analyses the way in which restorative justice got onto the government agenda and developed in Western criminal systems, by applying the multiple streams framework (MSF). To this end, it addresses the problematic aspects of traditional criminal justice, the role that different actors play in the implementation of restorative justice and the constraints that emerge in judicial systems. It further applies, in Section 5, the novel theoretical adaptation of the MSF (Fowler, 2019; Howlett, McConnell & Perl, 2015) to analyse why restorative justice implementation has been unequal across Western European countries and even within countries.
  Although it may seem ambitious to present conclusions regarding how restorative justice is getting onto Western European criminal justice systems based on an analysis of international organisations and individual countries, we believe that the extensive literature review that was conducted allows us to identify general processes and trends, even though there may be exceptions among the countries that belong to the Western European criminal justice system. These exceptions are beyond the scope of this study. However, they may be studied at a later stage as individual cases.
  As the MSF has not been used to analyse restorative justice before, in bringing together the MSF and restorative justice, this article is an important contribution to the MSF, restorative justice and policy change literature. It introduces a new perspective that helps understand why certain issues get onto the government agenda and receive more attention than others, while placing special emphasis on the actions of actors in different processes. As such, it adds important elements to the debate on how restorative justice practices are initiated and how they differ across countries.

• 2 Object of study: restorative justice

  Conflict between people is constantly present in our lives, and it is much more than the criminal act (Madrid, 2019). However, it has not been appropriately addressed, for the criminal justice system has been increasingly judicialised, witnessing a rise both in the number of criminal offences and in the length of sentences (Perulero, 2012). This has led to widespread discontent and a crisis of effectiveness of the traditional criminal system (Flores, 2015). There are two distinct layers to the commission of a crime. The first has to do with the conflict that is generated between the State and the offender who has broken the rules that we have given ourselves for a peaceful coexistence. The second, which we refer to as the intersubjective dimension of the crime, has to do with the breakdown that the crime entails for the direct victim, who suffers the direct consequences of the crime. These consequences must be addressed – an end for which restorative justice can be particularly appropriate – whether it is through moral or economic reparation or both (Milquet, 2019). This intersubjective dimension has not been sufficiently taken into account in the course of the criminal process, which only focuses on the first layer without paying adequate attention to the victims and their interests when they relate to the justice system (Kenney, 2004).
  Against this discontent, the need to introduce mechanisms that promote the responsibility of offenders and reparation for victims has been advocated for. The means to achieve such responsibility lies in restorative justice, whose objective is ‘aimed at preventing the commission of crimes and practicing prevention by combining rule of law requirements with those of the social state’ (Roxin, 2000: 31-36). At least in procedural systems of Napoleonic tradition, which feature what remains a somewhat limited space for what has been called negotiated criminal justice, restorative justice provides a new perspective that brings to light the conflict between the parties that results from the crime, while considering the protection and needs of the victim and the achievement of social peace through dialogue. It is a complementary system to that of procedural justice, which reinstates the victim as a subject whose needs transcend economic or vindictive ones, increasing the possibilities of economic reparation (Gal, 2020; Soleto, 2019).
  It is difficult to offer a single definition of restorative justice, for its content is heterogeneous, given the various trends it exhibits (Doolin, 2007). However, its primary objective is, invariably, the promotion of consensus in the resolution of criminal conflicts (Flores, 2015).
  According to the United Nations’ Basic Principles on the Use of Restorative Justice Programmes in Criminal Matters, the restorative process is

  any process in which the victim and the offender and, where appropriate, any other individual or community members affected by a crime participate together actively in the resolution of matters arising from the crime, generally with the help of a facilitator. (ECOSOC, 2002)

  The United Nations Handbook on Restorative Justice Programs (UNODC, 2020) includes the same definition.
  The 2012 EU Directive establishing minimum standards on the rights, support and protection of victims of crime of the European Parliament and of the Council adds a fundamental element to this definition, which concerns the free consent of both victims and offenders.¹x Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012, Art. 2, 1(d).
  Likewise, due to its recent nature, it is worth pointing out the recent Recommendation CM/Rec (2023) 2 of the Committee of Ministers to Member States on rights, services and assistance to victims of crime. This Recommendation, adopted in March 2023, intends to update and substitute the well-known 2006 Recommendation on assistance to crime victims, with an emphasis on the active role of member states in removing any obstacles to access to justice that victims of crime may encounter. The new Recommendation urges member states to ensure that restorative justice practitioners comply with the guidelines set out in the aforementioned Recommendation CM/Rec(2018)8. Although references to restorative justice are present throughout the entire body of the Recommendation, Article 18 focuses specifically on the promotion of restorative justice services, emphasising its availability as a broadly accessible service. The Article states that factors such as the nature and seriousness of the offence or its geographical location should not, on their own and in the absence of other circumstances, prevent the availability of restorative justice, reinforcing the notion of case-specific appropriateness rather than relying on the establishment of referral guidelines that could ultimately become exclusive. In this sense, and as far as the referral of the case to restorative services is concerned, one of the key lines of this Recommendation is precisely to guarantee that the voice of the victims is heard when decisions of particular interest to them are taken, decisions such as the referral of the case. Information and participation will be, therefore, two strategic lines in the future trajectory of restorative justice.
  The origins of restorative justice can be traced back to the first half of the twentieth century, with the development of philosophical currents based on empathy, human dignity and human rights. Subsequently, in the 1970s, there arose movements that focused on the study of the suffering that arises from crime and the need to promote the humanisation of the criminal system. These types of positions were reinforced with the crisis of the postulates of the reintegration model.²x The crisis of the reintegration model is reflected in crime rates and recidivism when prison fails as a mechanism for the reintegration of offenders (Flores, 2015).
  The traditional model had relegated victims to the background (Dünkel, 2017). As a result, victimology emerged as a field of criminology. Its objective is to prevent the victim’s role from being limited to serve as evidence intended to override the presumption of innocence and of constituting the basis for a conviction in criminal proceedings. Likewise, it emphasises the need to consider the consequences that the commission of a crime has for the community.
  As such, restorative justice led to a paradigm shift in the criminal justice system – however, the speed of this shift is actually slower than desired. Similarly, the pace of development and implementation of restorative justice within the European panorama is uneven, as will be explained. At the turn of the twentieth century, restorative justice emerged as a ‘product of social and legal trends in countries of northern Europe, Canada, and the United States’, resulting in higher quality justice, including the recognition of the victim’s needs (Soleto, 2019). It led to an evolution of the dual model in the criminal process, which featured the state and the offender as the sole two parties, towards a three-party model featuring the state, the offender and the victim. When a conflict arises and restorative justice mechanisms are used, the objectives that are sought include stakeholders’ participation, the reparation of the damage based on a subjective understanding, accountability, the integration of the offender into the community and even the avoidance of recidivism (Barona, 2017, 2019). These principles are achieved because restorative justice provides a flexible response to crime, in which each party is considered individually (Braithwaite, 1999). It motivates the offender to be aware of the consequences that derive from their crime, pays special attention to the underlying causes of the crime, addresses the damages and needs of the victim, and recognises an important role for the community as a key participant in crime prevention (Aertsen & Peters, 1998). This is carried out following a methodology that adapts itself to the principles of the criminal system prevalent in each community, as per the United Nations’ Principles and the Council of Europe 2018 Recommendation concerning restorative justice in criminal matters.
  For this evolution towards restorative postulates to take place, it was necessary to decide to pay attention to the hurdles existing in traditional justice, to define their characteristics and to propose solutions to these problems. This refers to the agenda building process, that is, to the manner in which the demands of some groups in society compete for the attention of political authorities (Cobb, Ross & Ross, 1976). The analysis of the agenda building process emphasises the importance of governmental and non-governmental actors related to a specific area of public action, for the beliefs and assessments of different actors play a fundamental role in defining problems. The definitions provided by those actors give rise to a certain understanding of reality, which can determine the public policies implemented (Elder & Cobb, 1984). Therefore, it is important to emphasise that the participation of various actors with a specific understanding of restorative justice has been very relevant in addressing the problems that have plagued the traditional criminal justice system. The existence and actions of these actors was decisive for restorative justice to get onto the government agenda.
  The government agenda is the set of issues explicitly accepted for serious and active consideration by decision-makers. It is concrete and specific, being formed by precise problems (Elder & Cobb, 1984). For example, and focusing on our object of study, discontent and dissatisfaction with traditional criminal justice is a concrete challenge that became part of the government agenda. This same reasoning applies to the shared interests of victims of crime, who have been historically reduced to an absolutely passive role (at best), not only as far as the management of the criminal conflict is concerned, but also with regard to the regulation of their rights of participation, reparation or assistance. The harmonisation efforts of the European institutions in the field of victimology, undertaken within the framework of the looked-for European area of freedom, security and justice, has led some authors to refer to this set of initiatives as part of a whole called European victimology policy (Hall, 2010; Joutsen, 1994).
  Additionally, for restorative justice to be placed on the government agenda, it was essential that the competent political institutions be faced with favourable circumstances to deal with the matter (e.g. that they possess the material and economic resources), that there exists systematic information on the matter and that there be interest in the adoption of a decision regarding that matter. However, even though restorative justice was adopted throughout Western European countries, its implementation has been uneven across countries and even within countries.

• 3 Theoretical approach

  This research was carried out using the MSF. The MSF was proposed by Kingdon in his book Agendas, Alternatives and Public Policies (1984), subsequently developed and complemented by other authors (Zahariadis, 2014), and applied in numerous research works. The MSF was employed for this research because it provides important elements that help understand why certain issues get onto the government agenda and receive more attention than others, while placing special emphasis on the actions of the actors in the different processes. According to the MSF, issues get onto government agendas when three different streams – problems, policies and politics – couple to open a ‘policy window’. This window enables interest groups and ‘policy entrepreneurs’ to champion favoured policies. In applying the MSF, we managed to understand the way in which restorative justice was developed by governments in a context of great ambiguity and to respond to the question of why restorative justice got onto the government agenda in Western judicial systems. The research also applies the novel theoretical adaptation of the MSF (Howlett, 2019; Howlett et al., 2015) to identify the factors that explain the heterogenous implementation of restorative justice across Western European countries and even within countries.

• 4 The multiple stream framework and restorative justice: agenda-setting

  Government agenda-setting is one of the most important processes in the policy cycle, as it can determine the outcomes of public action. Under the MSF, a problem is more likely to be placed on the government agenda when three independent streams are coupled: the problem stream, the policy stream and the political stream. In other terms, it occurs when, simultaneously, a problem is accepted, a solution is available and the political climate is favourable to change. The coupling can occur when a ‘policy window’ opens, allowing interested actors brief moments to present their problems or to push for their preferred solutions, facilitating policy change.

  4.1 Factors that facilitated the development of a restorative justice system

  The MSF outlines a series of conditions that can help turn an issue into a problem, thus capturing the attention of decision-makers. Proposals that generally prosper are technically feasible, compatible with the values of policymakers, financially acceptable and attractive to the public. Additionally, in the process of identifying problems, the following factors play a very important role: indicators that show the existence of the problem in a systematic way, studies carried out by experts in the field, focusing events (or crises) and feedback from past practice (Kingdon, 1984).
  In the case of restorative justice, five factors can be identified as having helped produce a change of perspective and, consequently, having facilitated the development of a restorative justice system (Soleto, 2012). Below, some milestones in the North American development of restorative justice will be outlined, noting that they can provide a useful framework for understanding the development of restorative justice at the European level, based on the common features and the mirroring effect on the European landscape of what was happening at the same time in Canada and the United States. These factors are as follows:

  1. Retributive trends: Retributive trends originated in the United States in the 1960s, at which time it had become clear that the traditional criminal system was not able to respond to the needs of the victims or the claims of society to participate in justice, in a system lacking such figures as private accusation or actio popularis. Subsequently, restitution would begin to replace retribution as a finality, thanks to the 1982 President’s Task Force Final Report, following which legislative reforms aimed at protecting the rights of victims were launched.

  2. Social empowerment trends: Social empowerment trends also emerged in the 1960s, with the primary objective of empowering society. These trends hold that the active participation of the parties to a conflict helps its resolution, while equally promoting the development of values such as tolerance and integration. The main idea is that in communities characterised by important cultural differences, dialogue can act as a mechanism that ensures the peaceful coexistence of all citizens.

  3. Inefficiency and the search of satisfaction in the administration of justice: As discussed in the National Conference on the Causes of Popular Dissatisfaction with Justice, which was held in Minnesota (United States) in 1976, the cause of dissatisfaction with the administration of justice was that it was necessary to find the most appropriate resolution in view of the characteristics of each conflict (Soleto, 2017). However, these characteristics were not reflected in the judicial system. In order to know which conflict resolution system is the most appropriate, Sander (1976) suggests the application of five criteria: nature of dispute, relationship between disputants, amount in dispute, cost and speed, that is, pendency time. Based on these criteria, the multi-door courthouse system is configured. It is a system in which each ‘door’ constitutes a method of resolution, which allows broad participation of the parties. Undoubtedly, it constitutes a novel idea that moves away from the typical vision that understands that the traditional judicial process is always the appropriate means. Since the 1970s, this multi-door system has been introduced in different countries, especially those with an Anglo-Saxon legal culture.

  4. Reintegration purposes: In different studies carried out in the United States, a certain causality was observed between the mediation processes carried out with minors and the reduction in their recidivism (Allen, 2004; Soleto, 2012). This is so because in restorative justice processes, offenders have the possibility of receiving support, understanding the effects of their infraction, discovering their emotions and those of their victim, showing their repentance and remorse, and correcting those attitudes that threaten peaceful coexistence in society. In short, they have the possibility of concluding a phase and achieving their reintegration into society.

  5. Importance of the victim: The traditional criminal system does not take the emotions and feelings of crime victims into account. Although in some systems their participation in different procedural moments is permitted, it is a very regulated intervention in which they lack the space to broaden their claims.³x Faced with the prejudice of some operators who reject the victim’s participation on the assumption that their claims are only economic, Bazemore (1998) concluded that the victims had different objectives. For example, in the case of victims of crimes committed by minors, there is a special interest in the re-education of the juvenile offender. However, gradually, Western systems have begun adopting legislation aimed at configuring a safety space for the victim (Lauwaert, 2015). In this regard, and focusing on European territory, the Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012, which establishes the minimum standards for the protection of the victim, and which replaces the regulations of the year 2001, stands out. This piece of legislation constitutes an important step in that it recognises the need to protect the victim without any kind of discrimination, taking into account their situation and needs, and providing the necessary information in any case, while avoiding secondary victimisation (Jacquelin, 2015). This legally recognised possibility of resorting to restorative justice services is fundamental because it allows the empowerment of the victim by favouring dialogue with the offender and the possibility of expressing their emotions, receiving forgiveness and restitution or reparation (Christopher, 2007). In short, as with the offender, it allows the victim to get closure and healing.

  However, and beyond the reference to the European Directive 2012/29 – and, in particular, Article 12 – previous efforts undertaken in Europe in the field of restorative justice and criminal mediation should not be neglected, particularly as they originated in the same period as those first initiatives in the United States and Canada. Authors such as Aertsen have discussed how restorative justice found a fertile soil on the basis of experiences in the 1960s and 1970s under the influence of the discussions and debates around the consequences of crime and the participation of the victim and the offender in the management of the aftermath of the offence (Aertsen, n.d.; Lauwaert & Aertsen, 2016). In Europe, the differentiating feature was that the most common format for these restorative initiatives was the victim-victimiser experiences or, put more simply, penal mediation. Among these first initiatives are the work carried out in Nordic countries – such as Norway in 1981 or Finland in 1983 – and Austria (1988) (Dünkel, Grzywa-Holten & Horsfield, 2015).
  Beyond these early initiatives, Europe’s experience, especially in the 1980s and 1990s, reveals an advance of restorative justice in various contexts or institutional models. This includes restorative perspectives focused on probation programmes, applied in the cases of Austria, England and Germany. Likewise, within the heterogeneity of the European experience, one can identify initiatives focused on victim support, as observed in France and Belgium, as well as a third model that opts for developing restorative initiatives in a neutral way and hand in hand with local governments, as evidenced in Norway and Finland. By the late 1990s, the numbers of restorative programmes had grown steadily, with various initiatives in Northern Ireland, Scotland, Ireland, Poland, the Czech Republic, Albania, Denmark and Sweden. After 2010, countries such as the Netherlands and in Eastern Europe expanded their restorative justice programmes, while some Southern European countries experimented with innovative local projects, sometimes hindered by a lack of legislative basis or national policies (Aertsen, n.d.; Dünkel, 2015; Lauwaert & Aertsen, 2016).
  Fortunately, this is a problem that is gradually being left behind. An example of this is the approval of the Spanish Law 4/2015 on victims of crime – this regulation was essentially adopted in order to transpose Directive 2012/29/EU – or the Italian Law nº. 134 of 27 September 2021 (the so-called Cartabia reform), which envisages the implementation of a comprehensive regulatory framework for restorative justice.

  4.2 The ‘policy primeval soup’ and ‘policy entrepreneurs’

  The five factors as mentioned above led to a change in perspective, also for Europe, with respect to the traditional criminal justice system, which was in crisis for not adequately satisfying the claims of society. This was the occasion for the defenders of restorative justice, which was in the ‘policy primeval soup’, to promote its incorporation into the government agenda.
  Kingdon (1984: 122-151) understands the ‘policy primeval soup’ as a system that resembles the natural selection process. Ideas are found in the different communities of actors that exist in society. Actors have their own understanding of certain aspects of reality and constantly try to generate alternatives and proposals according to such understanding. As in the natural selection process, some of these proposals manage to survive and be considered in the government agenda building process, while other proposals do not succeed.
  Kingdon refers to these actors as ‘policy entrepreneurs’. They are

  advocates who are willing to invest their resources – time, energy, reputation, money – to promote a position in return for anticipated future gain in the form of material, purposive, or solidary benefits. (1984: 188)

  Among these policy entrepreneurs, Kingdon (1984) distinguishes ‘inside of government actors’ from ‘outside of government actors’. The first group comprises government actors (including public employees and members of parliament), while the second group includes interest groups, researchers, academics, consultants, media, parties and other elections-related actors, and public opinion. Although this distinction between the outside and the inside is somewhat artificial, it serves to organise research and debate.
  It is especially the second group of actors that has played a key role in the development of restorative justice. Undoubtedly, the performance of these actors highlights the motivation to invest their resources in order to obtain policies in line with their ideas. This motivation may be due to different factors, among which are their extensive knowledge about restorative justice, their desire to promote certain values in the justice system, and their interest in participating and influencing the creation of a justice system in accordance with the needs of society.
  Among the European organisations whose activities have contributed to the development of restorative justice, the following stand out: the European Commission for the Efficiency of Justice (CEPEJ), and two non-profit organisations: the European Forum for Restorative Justice (EFRJ) and the European Association of Judges for Mediation (GEMME, according to its French initials). The performance of these actors is important in reflecting the situation of dissatisfaction with the traditional system. These actors make suggestions concerning the application of instruments such as mediation and develop a detailed analysis of the legislation on the protection of victims.
  The CEPEJ stands out for having analysed the degree of application of the Recommendations promoted by the Council of Europe and related to the adequate conflict management – in the fields of Public and Private Law – throughout various stages of development (Lhuillier, 2007). Both the above-mentioned non-profit organisations have favoured the sharing of know-how and support for new projects, and they have influenced national regulation and practice. GEMME is a group formed by judges and prosecutors, and is present in most European countries. Its aims include sharing experiences among judges, studying mediation systems, promoting research and disseminating initiatives. The EFRJ is specialised in the criminal field and has experts in restorative justice who focus on research and the development of doctrine. It promotes the dissemination of knowledge, and it combines the sharing of experiences and research with training and experimenting (Soleto, 2015). These actors invested their resources to obtain policies that reflect the objectives of ensuring the responsibility of the offender and the protection of the victim (Pemberton, Winkel & Groenhuijsen, 2007).

  4.3 The ‘softening up’ process

  According to Kingdon (1984: 134-137), once these policy entrepreneurs put forward their proposals and alternatives, they perform a ‘softening up’ process, so that these are considered. This is necessary as some issues may take time to become accepted within government, policy networks and the public, and to prevent proposals from being ignored. In this regard, the actors mentioned in the previous paragraphs play a key role. As the EFRJ states, it was necessary to promote the knowledge of restorative justice in society in general and among legal operators in particular, in order for it to be recognised as a feasible means within the judicial process. This softening up process is carried out by exposing the numerous advantages that restorative justice has in relation to the different currents in criminal law. Thus, for the most liberal tendencies, it is emphasised that restorative justice offers a limitation of traditional punishment. For the more conservative trends, it is explained that restorative justice promotes the responsibility of offenders. For the defenders of the victims, it is stressed that, in restorative processes, the victim plays an essential role and can also ask questions that are relevant to overcoming the crime without the constraints of an interrogation of a party at trial. Finally, it is emphasised that restorative justice is adapted to the neoliberal ideology of limiting the role of the state in order to achieve more participatory citizens (Lynch, 2010) – an argument that seems more appropriate in the case of relationship models based on the idea of alternativity.
  In this softening up process, European Union lawmakers also played an important role, since they decided to opt for a restorative justice model focused on the victim, through the regulation of victim protection (Directive 2012/29/EU). They established that restorative justice is a mechanism to improve reparation for the victim, resulting in an incorporation of restorative justice into justice systems that is less contentious from a socio-legal point of view than when restorative justice focuses on the recovery of the offender. Introducing restorative justice through the Victims Protection Directive was like introducing it as a ‘Trojan Horse’, since linking it to the victims’ welfare could not be questioned (Soleto, 2019).
  In this way, an adequate context is achieved for the consideration of restorative justice as a possible solution to the problems of traditional justice. However, it is not only necessary to carry out that softening up process. It is also essential that these ideas meet a series of requirements to survive. These survival criteria are technical feasibility, value acceptability, adaptability to existing values in the community, political acceptance and the acceptance of the target population (Kingdon, 1984: 138-151). In the case of restorative justice, compliance with these requirements can be observed. This is reflected in the development of three models depending on the type of relationship between the existing criminal system and restorative justice mechanisms (Soleto, 2012):

  1. Complementary system to the courts: There is an indisputable connection with the courts that is reflected in procedural consequences derived from the application of a restorative justice procedure (which need not belong to the administration of justice). This is the case, for example, with a possible reduction in the penalty or the application of penitentiary benefits. Nevertheless, it should be stressed that, in this model, no new criminal or procedural institution is created.

  2. Alternative systems to prosecution: Amongst such systems, programmes involving juvenile offenders and theft crimes stand out. These programmes are characterised by the fact that the referral to the restorative justice procedure may occur even in the absence of a formal criminal proceeding. Cases in which the subject is a repeat offender are generally excluded.

  3. Initiatives unrelated to the proceedings and enforcement: They are characterised by the emotional character that permeates these types of processes. They can be undertaken, for example, in intrafamily conflicts in which the parties do not want criminal proceedings initiated. Likewise, they may occur after conviction, and may or may not influence the procedural situation. Examples of such initiatives include activities with the victim’s family members.

  We can thus observe how restorative justice adapts itself to the specificities of each community, giving rise to different models, for one of its essential characteristics is flexibility (Aertsen, Mackay, Pelikan & Willemsens, 2004).

  4.4 Problem, policy and political streams, and the policy window

  According to Kingdon (1984: 95-121), problems are issues presented to policymakers, such as indicators of a policy, serious events that raise concern about an issue and information about the results of existing programmes or activities. As we have seen, the need for a change in traditional justice becomes a public problem. It is an issue that enjoys significant public attention given the widespread dissatisfaction with the existing criminal justice system; this dissatisfaction entails the demand for a response that meets the needs of the users of the criminal justice system; and citizens are aware that this is a reform that the government must take up. Kingdon (1984: 113-121) identifies three circumstances in which issues become defined as problems: when they affect important social values, in comparison with the situation of other countries, and when people become convinced that something should be done to change it. To understand the magnitude of the problem, different indicators can be used, such as the level of citizen satisfaction with the justice system for example. This desire to act to change reality is the best opportunity for such situations to be identified as a problem and, therefore, for the issue to be placed on the agenda.
  ‘Policy stream’ refers to the different alternatives for change that are generated by bureaucrats, policymakers, experts and academics. There exists no rational action by the government that results in a concrete solution (Kingdon, 1984: 209-210). Rather, there are different actors who, with their actions (investment of their time, energy, reputation and material resources), favour the study, exploration, organisation and quantification of the issue (Hogwood & Gunn, 1984). The presence of the aforementioned actors specialised in restorative justice was essential for the coupling of the three streams included in the MSF to take place.
  ‘Political stream’ refers to the factors that influence the political process daily, giving rise to a public space in which there is room for a particular issue to become an agenda item. The factors of change of perspective with respect to traditional justice, referred to at the beginning of this section, directly influenced the determination of the agenda.
  Political events occur according to their own rules and dynamics, generating a public space in which the claims of different actors allow restorative justice to get onto the government agenda. Political events are very varied; for example, the establishment of a new administration after the elections, or the pressure of interest groups. In any case, policy entrepreneurs must be attentive to these events, to proceed to act, given the short duration of the policy window (Kingdon, 1984: 173-174). In the present analysis, the actors who propitiated the evolution of traditional justice clearly identified the deficiencies of this system and proposed new methods in which the offender’s responsibility, the protection of the victim and the well-being of the community can have their place.
  This led us to conclude that we were facing a policy window at the European level. This policy window constituted a favourable context for the introduction of restorative justice mechanisms in the justice system. The favourable context resulted from the coupling of the problem stream, the policy stream and the political stream, determining the likelihood of the issue becoming a policy agenda item.

• 5 Moving the multiple streams framework forward to explain restorative justice implementation

  The MSF is a widely used theoretical approach for the analysis of the policymaking process, in particular agenda-setting, but it has not paid much attention to implementation (Howlett, 2019; Saetren, 2016). However, in recent years, some authors have claimed that ‘many of its precepts could be stretched or adapted to address implementation activities’ (Howlett, 2019: 414). In this regard, Howlett et al. (2015) and Ridde (2009) have suggested expanding the MSF to cover all stages of the policy process, including implementation. But considering that policy implementation is a distinct stage of the policy process, involving different goals, processes, mechanisms and relationships, where other important actors are active, some authors claim that the three streams proposed by Kingdon to explain agenda-setting are not readily applicable to implementation (Fowler, 2019; Matland, 1995). That is why they identify a fourth stream: ‘process stream’. This stream is formed by actors’ concern with knowing the best administrative practices in the implementation of programmes (Mukherjee & Howlett, 2015). It includes the necessary activities and events to bring about policy results. This stream includes new actors, new resources and new tactics that join the policymaking process after an issue gets onto the government agenda (Howlett, 2019).
  According to these authors, the dynamics of the interaction of these actors are in a fifth stream: the ‘programme stream’, which is key to implementation. This stream reflects the incorporation of new actors in the implementation process who have not participated in the policymaking process: above all bureaucrats and programme managers, and also people belonging to the affected public, and, sometimes, non-governmental organisations. In some cases, street-level bureaucrats and other actors involved in the implementation process have little room for manoeuvre in relation to what was previously established. In other cases, there is a more controversial implementation process, where disputes over procedures exist between actors, and a space opens for the discretion of street-level bureaucrats. Policy implementation often depends on civil servants and administrative officials, who set and manage the necessary actions. They use their knowledge, experience, expertise and values to implement policy decisions. They are therefore key actors in the programme stream (Howlett, Ramesh & Perl, 2009).

  Different bureaucratic agencies at various levels of government (national, regional, or local) are usually involved in implementing policy, each carrying particular interests, ambitions, and traditions that affect the implementation process and shape its outcomes, in a process of ‘multi-level’ government or governance. (Howlett, 2019: 420)

  Adding these two streams to the three streams developed by Kingdon, these authors present a more comprehensive framework of analysis, suitable for analysing the entire policy process and helping to understand the different modalities of the implementation process (Howlett et al., 2015; Mukherjee & Howlett, 2015). Once the policy window has opened, and the agenda-setting process has occurred, specific actors engage in deliberations and propose policy alternatives (Howlett & Craft, 2013). This exchange of ideas on policy alternatives creates the basis for a second critical point, when the politics stream connects with the process stream, creating the space for decision-making. The third critical point occurs when a specific action is chosen to implement. The policy stream and the programme stream couple, incorporating new actors and interests, who choose programme instruments to generate new outputs (Howlett, 2019).
  Additionally, Herweg, Huß & Zohlnhöfer (2015) identify a second window (apart from the policy window described by Kingdon), and consequently two coupling processes: one for agenda-setting; and one for decision-making, called ‘decision window’, with the related decision coupling. According to them, the results of the decision window opening is the selection of a programme (Herweg, Zahariadis & Zohlnhöfer, 2017).
  These new elements that have been added to the MSF emphasise the idea that actors involved in policymaking may be different from those who implement it (Aberbach & Christensen, 2014). Therefore, the analysis of the implementation process should primarily take into account those who implement the policy (Boswell & Rodrigues, 2016). Especially when policies are ambiguous, implementers may interpret them in a variety of ways (Fowler, 2019).
  At the same time, the application of the MSF to the implementation process underscores that policy implementation is often a costly and multi-year endeavour, which implies that continued funding of public actions is often not guaranteed. On the contrary, it requires ongoing negotiations and debates within and between the political and administrative spheres of the state. This creates opportunities for politicians, public managers and other actors to use the implementation process as another opportunity to continue discussions they have lost in the early stages of the policy process. These debates complicate implementation, as it is not simply a technical matter of implementing previously adopted decisions (Nicholson-Crotty, 2005; Ziller, 2005). Such a complex reality requires an analysis of policy implementation that highlights the different actors involved and their interests in order to better understand the nuances and dynamics of this stage.
  In the previous section, the incorporation of restorative justice into Western European countries’ government agendas was addressed. However, evidence suggests that the tempos and modalities of the development and implementation of restorative justice vary across countries and even within countries, for example countries with a federal state structure (also, e.g. the United States) (Dünkel, Grzywa-Holten & Horsfield, 2015; Galaway & Hudson, 1996; Sherman & Strang, 2007). While some countries have been implementing restorative justice experiences for several decades (as is the case of Canada or the United States since the 1970s⁴x With this reference to the United States, the authors do not intend to convey the idea that restorative justice developed fairly homogeneously in that particular country. On the contrary, the heterogeneity in the development, implementation and availability of different restorative practices is a noteworthy subject in the scientific literature (Sliva & Lambert, 2015).), others have only recently adopted national strategies or committed to restorative principles and practices (efforts in Belgium, Norway or the United Kingdom can be highlighted) (Dünkel, 2017; Gavrielides, 2016; Soleto, 2015). Two of the best-known examples of the latter are New Zealand, whose experience in restorative initiatives has been rich and diverse since the enactment of the Sentencing Act 2002, Parole Act 2002 and the Victims’ Rights Act 2002 (especially after the publication in August 2019 of the Restorative Justice: Best Practice Framework by its Ministry of Justice) (Wood & Suzuki, 2020), and Scotland, whose government launched in June 2019 its ‘Restorative Justice Action Plan’, according to which it was expected that by 2023 Scottish citizens will have widespread access to restorative justice services (Maglione, 2021).
  As restorative justice is a greatly diverse reality, factors such as methodological differences in data collection, lack of statistics in some cases, absence of homogeneity in the concepts applied, and the fact that restorative justice is still a growing discipline make it complex and difficult to contextualise differences in the implementation of restorative justice across countries (Lemonne, 2008). Nevertheless, after observing the practice, it is possible to glimpse some factors that may explain these differences:

  1. Different trends in procedural common and civil law legal cultures

  2. Conservative resistance against cultural change among legal practitioners

  3. Absence of general training guidelines and homogenised action guides

  4. Diversity of agents along with the restorative procedure and diverse intervention possibilities

  5. Lack of sufficient structures to support restorative services

  6. Difficulties for the professionalisation of restorative justice facilitators

  7. Lack of an evaluation system for restorative services as well as a facilitator’s statute for an ethical performance

  While these factors are intimately linked to each other, we will try to follow the proposed classification for the sake of clarity. This is without prejudice to the subsequent discussion of some of these causes and, consequently, to the exploration of other connected factors that are also useful to explain the differences in the implementation of restorative justice.

  1. A procedural perspective shows that the greater or lesser scope for negotiated criminal justice depends on the pre-eminence, within a legal culture, of either the principle of legality or the principle of opportunity, the latter of which plays a more important role in Anglo-Saxon models. Negotiated criminal justice has enjoyed greater development in countries with a large acceptance of the procedural principle of opportunity, as to promote wider spaces for agreement and private settlement.

  2. Somewhat connected to the above, national differences regarding cultural change among legal practitioners and the development of a restorative leadership able to boost the professionalisation of restorative services are explaining factors for the dissimilar implementation of restorative justice. In terms of cultural change, it is important not to lose sight of the difficulties involved in making room for restorative justice in structures with deeply rooted operating inertias. As such, it is necessary to work on structural transformations and changes within the organisational culture (Benier, 2017; Daly, 2011, 2017). One of the clearest consequences of this change of perception regarding the role of justice agents can be seen in restorative experiences with those who often address the conflict first: police officers (Hoyle & Batchelor, 2018: 178). Some remarkable experiences have gone along the line of having what has been called ‘proximity police’, whose agents have a role in the referral and adequate management of conflicts beyond its conservative security functions. Also, in relation to the need of a cultural change within the justice system, Bazelon and Green (2020) point out how in the United States a conservative resistance speech has effectively reduced the success of restorative justice experiences (even among the most progressive circles of prosecutors), as well as restricted its actual scope to rather juvenile offences or non-violent, low-level felonies committed by adults. For the authors, one of the reasons for this is that in many states, the use of restorative justice is not allowed without the consent of the prosecutor, who tends to have a more or less biased attitude towards restorative justice. This circumstance emphasises the key role that prosecutors play in the referral of cases and diversion mechanisms within the criminal justice system.

  3. The latest version of the United Nations Handbook on Restorative Justice Programs highlights the development of guidelines related to the training and supervision of facilitators, mediators and volunteers as a key measure for the promotion of restorative justice initiatives (UNODC, 2020). Thus, the absence of general training guidelines and homogenised action guides may be another factor explaining the difficulties in implementing restorative justice. This is something to be taken into account, especially in the field of European procedural law, which is an increasingly relevant area for international judicial cooperation policy. Successful implementation requires not only initial training, but also adequate ongoing training, which must be provided to all professionals involved, both within the restorative programme and in those instances playing a role in the referral of cases. This involves, for example, including training in restorative justice in the curriculum of future judges, prosecutors or lawyers, or setting restorative justice quality control for those who handle the case (Vedananda, 2020). A nationwide analysis on the training needs and a further development of a national strategy which takes into account both already existing provisions and the possibility of specialised training in particularly challenging situations (such as sexual violence or hate crimes) would strongly contribute thereto (Benier, 2017; Daly, 2017; Marsh & Wager, 2015; Pereira, Craen & Aertsen, 2022; Walters, 2014). Furthermore, this strategy should include accreditation plans and an ongoing training system so as to ensure that facilitators are up to date and able to provide a service in accordance with the highest quality standards (Neff, Patterson & Johnson, 2012). Thus far, these proposals are far from being implemented.

  4. Another factor explaining the uneven stages of implementation in restorative justice across countries, and even within the borders of the same country (Ghoshray, 2014), is the diversity of agents present along the procedure, as well as their irregular participation in the design of these policies and their scope of discretion. The Spanish case is a good example of how the diversity of agents involved, each with different levels of involvement within the process, has a significant impact on the effective implementation of restorative justice. While the analysis of the Spanish national scenario is not the aim of this article, it is still worthy of emphasising the much broader role that prosecutors in Spain play in the development of restorative procedures with juvenile offenders when compared to their function in criminal mediation with adults. As regards minors, prosecutors are allowed to desist from pursuing a case, after factors such as the seriousness and the circumstances of the crime (especially the lack of serious violence or intimidation in the crime commission), the minor’s own circumstances or the existence of conciliation between victim and offender (or alternatively a commitment given by the latter to repair the damage or to comply with the proposed educational activity) have been taken into account (Aguilera, 2012; Colomer, 2016). Employing the new concepts added to the MSF some decades after the publication of Kingdon’s seminal book (1984), we can affirm that the differences in the implementation of restorative justice between different countries, and even between regions within the same country, are related, among other factors, to the process stream, for the implementation of restorative justice involves actors, interests, resources and strategies that were not in all cases part of the policymaking process. The dynamics between these actors (programme stream) are also key to the restorative justice implementation process. Once restorative justice has gotten onto the government agenda, the different actors involved in implementation can determine the modality under which restorative justice is carried out, and the speed of its development.

  5. The extent to which restorative justice services are seen as a public, permanent and professional service also explains the differences across national settings. There also are in the Spanish case some examples of how there is still a lack of sufficient structures to support a homogeneous service. Article 29 of the Spanish Statute of the Victim of Crime states that the Victims’ Assistance Offices are responsible for providing ‘support to the services of restorative justice and other legally established out-of-court settlement procedures’. Lack of further provisions thereof permit to infer that Victims’ Assistance Offices are legally excluded from conducting restorative programmes, as they are only entrusted with supporting those who actually provide the service. However, the questions as to at which stages of the procedure that support should materialise and how it should be implemented remain unanswered. In fact, quite a few organisational models have been legally adopted to host the court-connected mediation services. Accordingly, in some Autonomous Communities, regulations assign the coordination and supervision of the restorative procedure to the Judicial Bureau. This encompasses the registration of the referrals, the management of the agenda, the supervision of the facilitator’s assignments, support preparing the file and minutes of the sessions, and responsibility for all the procedures that allow the mediation to be undertaken (communication with the parties, supervision of the mediation sessions and preparation of the referral protocols for the proper functioning of the mediation or any other restorative approach). However, sometimes the implementation of the service has been made dependent on the Consejerías, that is, the General Directorates of the Autonomous Government. In other Autonomous Communities, a zero-cost model with the collaboration of non-profit organisations has been opted for, which, as such, is not always available, nor is it available in all the criminal courts of the Autonomous Communities.

  6. The existence of a mix of volunteer and professional staff in restorative initiatives can account for the differences in the implementation of restorative justice. Certainly, including volunteer personnel provides considerable incentives, the most important of which is the symbolic representation of the communal aspect of restorative justice through voluntary community agents. However, relying only on volunteers may not contribute to the creation of trust in the restorative model, thus perpetuating the frictions between the traditional and restorative models of justice (Gresson, 2018). Besides, a restorative programme cannot function effectively unless it adequately supports its volunteers, and even community-based programmes that are largely dependent on volunteers need resourcing. As Sherman and Strang (2007) highlight, the zero-cost model does not work and will not work. Public investment in restorative justice should also be understood as part of a public policy that pursues the achievement of an efficient and quality criminal justice model committed to the victims’ needs and the reduction in secondary victimisation.

  7. Closely related to the stabilisation and professionalisation of the service, another factor that helps explain the differences in the implementation of restorative justice is the existence or not of a quality-of-service evaluation system and a facilitator’s statute with guidelines for an ethical performance. This framework contributes positively to both the confidence of legal practitioners in restorative justice programmes and the reputation of the programme in the eyes of the victims and offenders involved.

• 6 Conclusions

  As the MSF has not been used to analyse restorative justice policy before, in bringing together the MSF and restorative justice policy, this article makes an important contribution to understanding how restorative justice got onto the government agenda and how it was implemented across countries. In particular, the analysis of the ‘softening up process’, of the ‘process stream’, of the ‘programme stream’ and the identification of ‘survival criteria’ stands out.
  The generalised dissatisfaction of citizens with the traditional criminal justice system, their empowerment and the performance of actors specialised in the matter constituted a favourable context for restorative justice to get onto the government agenda. This access was produced thanks to the policy window that was generated by the coupling of the three streams: problem, policy and political. This led to the introduction of restorative justice mechanisms into the justice system.
  Five conditions have been in the ‘policy primeval soup’ since the 1960s, drawing the attention of decision-makers: the emergence of retributive currents, the empowerment of society, the inefficiency of and search for satisfaction with the administration of justice, the objective of achieving the reintegration of offenders and the importance acquired by the victim in the criminal process. Widespread discontent with the traditional criminal justice system becomes a public problem as it creates situations that are unsatisfactory to citizens.
  Regulations aimed at protecting the victim have been developed both by the Council of Europe and by the European Union, each one in their specific soft or hard law role. In fact, European guidelines – including the active work of both the Council of Europe and the European Union – have been pivotal in the development of restorative justice, directly influencing the member states, especially in the countries of Southern Europe, as recent examples such as the Italian case attest.
  In the restorative justice government agenda making process in European countries, certain specialised actors played an essential role in facilitating the incorporation of restorative justice into the government agenda, such as the CEPEJ, the EFRJ and the GEMME. They are what the MSF refers to as ‘policy entrepreneurs’. These actors decided to invest their resources and promote a change in traditional justice to ensure that restorative justice is configured as a space in which the needs of the victim are addressed, the responsibility of the offender is promoted and, ultimately, its application contributes to the achievement of social peace.
  These policy entrepreneurs performed the ‘softening up’ process. The main objective of this process was to prevent their proposals from not being considered because governments, policy networks and the public were not prepared to address these alternatives. With this objective in mind, the above-mentioned actors showed how the principles of restorative justice can be adapted to the movements of liberal, neoliberal and conservative ideology, and to the defenders of the victims.
  Restorative justice meets the survival requirements established by the MSF in order to survive. Its development is technically and economically viable, and it enjoys political acceptance and the acceptance of the target population – particularly, its ability to adapt to existing values in each community. Thus, we can distinguish three models resulting from the relationship between the existing criminal system and restorative justice mechanisms: (1) complementary systems to the courts, (2) alternative systems and (3) initiatives unrelated to the proceedings and enforcement.
  Restorative justice manages to successfully address the factors that made certain aspects of traditional justice problematic, including retributive currents, social empowerment, the inefficiency of and dissatisfaction with the administration of justice, and the promotion of the rights of the victim and the reintegration of the offender. According to the MSF, policy windows can open by the emergence of compelling problems or by happenings in the political stream. In our case study, the policy window opened in the problem stream.
  Factors such as the different trends in procedural common and civil law legal cultures, conservative resistance against cultural change among legal practitioners, difficulties for the professionalisation of restorative justice agents, lack of a sufficient structure to support restorative services, and lack of an evaluation system for restorative services as well as a facilitator’s statute for an ethical performance mark an uneven progress and rooting of restorative justice across Western European countries and even within countries.
  In summary, this analysis has confirmed, as in previous research,⁵x Kagan (2019), Liu, Yamaguchi and Yoshikawa (2017), Zahariadis (2005), among others. the value of both the original MSF and its novel theoretical adaptation in analysing policy change in a variety of environments. The MSF provides a useful lens with which to examine the complexity and dynamics of restorative justice policymaking and implementation processes.

References

• Aberbach, J. & Christensen, T. (2014). Why reforms so often disappoint. The American Review of Public Administration, 44(1), 3-16. doi: 10.1177/0275074013504128.

• Aertsen, I. (n.d.). The idea of restorative justice and how it developed in Europe. European Forum for Restorative Justice. Retrieved from https://www.euforumrj.org/sites/default/files/2020-01/the_idea_of_restorative_justice_and_how_it_developed_in_europe.pdf.

• Aertsen, I., Mackay, R., Pelikan, C. & Willemsens, J. (2004). Rebuilding community connections-mediation and restorative justice in Europe. Strasbourg: Council of Europe Publishing.

• Aertsen, I. & Peters, T. (1998). Mediation for reparation. European Journal of Crime, Criminal Law and Criminal Justice, 6(2), 106-124.

• Aguilera, M. (2012). Análisis crítico de la regulación normativa de la mediación en la Justicia penal de menores. In P. Garciandía, H. Soleto & S. Oubiña (eds.), Sobre la mediación penal (pp. 639-660). Pamplona: Thomson Reuters-Aranzadi.

• Allen, T. (2004). Ordering ADR: a guide for judges and those who apply to them. London: CEDR.

• Barona, S. (2017). Justicia Penal líquida. Teoría & Derecho, 1(22), 64-91.

• Barona, S. (2019). Mirada restaurativa de la justicia penal en España. In H. Soleto & A. Carrascosa (eds.), Justicia Restaurativa, una justicia para las víctimas (pp. 55-94). Valencia: Tirant lo Blanch.

• Bazelon, L. & Green, B. (2020). Restorative justice from prosecutors’ perspective. Fordham Law Review, 88, 1-32.

• Bazemore, G. (1998). Crime victims and restorative justice in Juvenile courts: judges as obstacle or leader?. Western Criminology Review, 1(1), 1-37.

• Benier, K. (2017). The harms of hate: comparing the neighbouring practices and interactions of hate crime victims, non-hate crime victims and non-victims. International Review of Victimology, 23(2), 179-201. doi: 10.1177/0269758017693087.

• Boswell, C. & Rodrigues, E. (2016). Policies, politics and organisational problems: multiple streams and the implementation of targets in UK government. Policy & Politics, 44(4), 507-524. doi: 10.1332/030557315X14477577990650.

• Braithwaite, J. (1999). Restorative justice: assessing optimistic and pessimistic accounts. Crime and Justice, 25, 1-127.

• Christopher, B. (2007). Satisfying the needs and interests of victims. In G. Johnstone & D. Van Ness (eds.), Handbook of restorative justice (pp. 247-264). Cullompton: Willan Publishing.

• Cobb, R., Ross, J. & Ross, M. (1976). Agenda building as a comparative political process. The American Political Science Review, 70(1), 126-138. doi: 10.2307/1960328.

• Colomer, I. (2016). Derechos de la víctima y mediación penal con menores infractores. In S. Barona (ed.), Mediación, arbitraje y jurisdicción en el actual paradigma de justicia (pp. 233-250). Pamplona: Thomson Aranzadi.

• Daly, K. (2011). Conventional and innovative justice responses to sexual violence. ACSSA Issues: Australian Centre for the Study of Sexual Assault, 12, 1-35.

• Daly, K. (2017). Sexual violence and victims’ justice interests. In E. Zinsstag & M. Keenan (eds.), Restorative responses to sexual violence (pp. 108-139). New York: Routledge.

• Doolin, K. (2007). But what does it mean? Seeking definitional clarity in restorative justice. The Journal of Criminal Law, 71(5), 427-440. doi: 10.1350/jcla.2007.71.5.427.

• Dünkel, F. (2017). Restorative justice in penal matters in Europe. In J. De la Cuesta & I. Subijana (eds.), Justicia Restaurativa y terapéutica (pp. 125-194). Valencia: Tirant lo Blanch.

• Dünkel, F., Grzywa-Holten, J. & Horsfield, P. (2015). Restorative justice and mediation in penal matters in Europe – comparative overview. In F. Dünkel, J. Grzywa-Holten & P. Horsfield (eds.), Restorative justice and mediation in penal matters. A stock-taking of legal issues, implementation strategies and outcomes in 36 European countries (pp. 1015-1096). Mönchengladbach: Forum Verlag Godesberg.

• ECOSOC (2002). Resolution 2002/12, basic principles on the use of restorative justice programmes in criminal matters (24th July 2002).

• Elder, C. & Cobb, R. (1984). Agenda-building and the politics of aging. Policy Studies Journal, 13(1), 115-129. doi: 10.1111/j.1541-0072.1984.tb01704.x.

• Flores, I. (2015). Algunas reflexiones sobre la justicia restaurativa en el sistema español de justicia penal. Revista Internacional de Estudios de Derecho Procesal y Arbitraje, 2, 1-45.

• Fowler, L. (2019). Problems, politics, and policy streams in policy implementation. Governance, 32, 403-420. doi: 10.1111/gove.12382.

• Gal, T. (2020). Restorative justice myopia. The International Journal of Restorative Justice, 3(3), 341-355.

• Galaway, B. & Hudson, J. (ed.) (1996). Restorative justice: international perspectives. New York: Criminal Justice Press, Monsey.

• Gavrielides, T. (2016). Repositioning restorative justice in Europe. Victims & Offenders, 11, 71-86. doi: 10.1080/15564886.2015.1105342.

• Ghoshray, S. (2014). An equilibrium-centric interpretation of restorative justice and examining its implementation difficulties in America. Campbell Law Review, 35(3), 287-332.

• Gresson, E. (2018). Restorative justice in criminal offending. Arizona Summit Law Review, 1, 1-47.

• Hall, M. (2010). Victims and policy making: a comparative perspective. New York: Willan Pub.

• Herweg, N., Huß, C. & Zohlnhöfer, R. (2015). Straightening the three streams: theorising extensions of the multiple streams framework. European Journal of Political Research, 54(3), 435-449. doi: 10.1111/1475-6765.12089.

• Herweg, N., Zahariadis, N. & Zohlnhöfer, R. (2017). The multiple streams framework. In C. Weible (ed.), Theories of the policy process (pp. 17-53). Boulder: Westview Press.

• Hogwood, B. & Gunn, L. (1984). Policy analysis for the real world. Oxford: Oxford University Press.

• Howlett, M. (2019). Moving policy implementation theory forward: a multiple streams/critical juncture approach. Public Policy and Administration, 34(4), 405-430. doi: 10.1177/0952076718775791.

• Howlett, M. & Craft, J. (2013). Policy advisory systems and evidence-based policy. In S. Young (ed.), Evidence-based policy-making in Canada (pp. 27-44). Ontario: Oxford University Press.

• Howlett, M., McConnell, A. & Perl, A. (2015). Streams and stages: reconciling Kingdon and policy process theory. European Journal of Political Research, 54(3), 419-434. doi: 10.1111/1475-6765.12064.

• Howlett, M., Ramesh, M. & Perl, A. (2009). Studying public policy. Toronto: Oxford University Press.

• Hoyle, C. & Batchelor, D. (2018). Making room for procedural justice in restorative justice theory. The International Journal of Restorative Justice, 12(1), 175-186. doi: 10.5553/IJRJ/258908912018001002001.

• Jacquelin, M. (2015). Victim’s participation in French criminal proceedings. In L. Lupária (ed.), Victims and criminal justice (pp. 83-99). Milano: Wolters Kluwer.

• Joutsen, M. (1994). Victimology and victim policy in Europe. CJ Europe, 4(5), 9-12.

• Kagan, J. (2019). Multiple streams in Hawaii: how the Aloha State adopted a 100% renewable portfolio standard. Review of Policy Research, 36(2), 217-241. doi: 10.1111/ropr.12323.

• Kenney, J. (2004). Human agency revisited. International Review of Victimology, 11, 225-257.

• Kingdon, J. (1984). Agendas, alternatives and public policies. Boston: Little Brown.

• Lauwaert, K. (2015). European criminal justice policies on victims and restorative justice. In I. Vanfraechem, D. Bolivar & I. Aertsen (eds), Victims and restorative justice (pp. 239-272). New York: Routledge.

• Lauwaert, K. & Aertsen, I. (2016). Restorative justice and criminal justice systems in Europe. Minorigiustizia, 24-32. doi: 10.3280/MG2016-001003.

• Lemonne, A. (2008). Comparing the implementation of restorative justice in various countries. British Journal of Community Justice, 6(2), 43-54.

• Lhuillier, J. (2007). Analysis on assessment of the impact of Council of Europe recommendations concerning mediation. Retrieved from https://rm.coe.int/1680747c1f.

• Liu, D., Yamaguchi, K. & Yoshikawa, H. (2017). Understanding the motivations behind the Myanmar-China energy pipeline. Energy Policy, 107, 403-412. doi: 10.1016/j.enpol.2017.05.005.

• Lynch, N. (2010). Restorative justice through a children’s rights lens. The International Journal of Children’s Rights, 18(2), 161-183.

• Madrid, S. (2019). Las víctimas en los procedimientos extrajudiciales del ámbito penal de menores infractores. In H. Soleto (ed.), Justicia Restaurativa, una justicia para las víctimas (pp. 641-666). Valencia: Tirant lo Blanch.

• Maglione, G. (2021). Restorative justice, crime victims and penal welfarism. Social & Legal Studies, 30(5), 745-767. doi: 10.1177/0964663920965669.

• Marsh, F. & Wager, N. (2015). Restorative justice in cases of sexual violence: exploring the views of the public and survivors. Probation Journal, 62(4), 336-356. doi: 10.1177/0264550515619571.

• Matland, R. (1995). Synthesizing the implementation literature. Journal of Public Administration Research and Theory, 5(2), 145-174.

• Milquet, J. (2019). Strengthening victims’ rights. Luxemburg: Publications Office of the European Union.

• Mukherjee, I. & Howlett, M. (2015). Who is a stream? Epistemic communities, instrument constituencies and advocacy coalitions in public policy-making. Politics and Governance, 3(2), 65-75. doi: 10.17645/pag.v3i2.290.

• Neff, J., Patterson, M. & Johnson, S. (2012). Meeting the training needs of those who meet the needs of victims: assessing service providers. Violence and Victims, 27(4), 609-632. doi: 10.1891/0886-6708.27.4.609.

• Nicholson-Crotty, S. (2005). Bureaucratic competition in the policy process. Policy Studies Journal, 33(3), 405-430.

• Pemberton, A., Winkel, F. & Groenhuijsen, M. (2007). Taking victims seriously in restorative justice. International Perspectives in Victimology, 3(1), 4-14.

• Pereira, A., Craen, B. & Aertsen, I. (2022). Restorative justice training for judges and public prosecutors in the European Union: what is on offer and where are the gaps?. The International Journal of Restorative Justice, 5 (Online First). doi: 10.5553/TIJR.000119.

• Perulero, D. (2012). Hacia un modelo de justicia restaurativa. In P. Garciandía & H. Soleto (eds.), Sobre la mediación penal (pp. 69-89). Pamplona: Thomson Reuters-Aranzadi.

• Ridde, V. (2009). Policy implementation in an African State: an extension of Kingdon’s multiple-streams approach. Public Administration, 87(4), 938-954. doi: 10.1111/j.1467-9299.2009.01792.x.

• Roxin, C. (2000). La evolución de la política criminal, el derecho penal y el proceso penal. Valencia: Tirant lo Blanch.

• Saetren, H. (2016). Lost in Translation: re-conceptualizing the Multiple-Streams Framework back to its source of inspiration. In R. Zohlnhöfer (ed.), Decision-making under ambiguity and time constraints (pp. 1-34). Essex: ECPR.

• Sander, F. (1976). Varieties of dispute processing, National Conference on the Causes of Popular Dissatisfaction with the Administration of Justice (Pound Conference). Minnesota: National Center for State Courts.

• Sherman, L. & Strang, H. (2007). Restorative justice: the evidence. London: Smith Institute.

• Sliva, S. & Lambert, C. (2015). Restorative justice legislation in the American States: a statutory analysis of emerging legal doctrine. Journal of Policy Practice, 14, 77-95. doi: 10.1080/15588742.2015.1017687.

• Soleto, H. (2012). La justicia restaurativa como elemento complementario a la justicia tradicional. In P. Garciandía (dir.), Sobre la mediación penal (pp. 41-69). Pamplona: Thomson Reuters-Aranzadi.

• Soleto, H. (2015). Development and resistance in South Europe justice systems to restorative justice. In H. Dalla (ed.), Contemporary tendencies in mediation (pp. 281-311). Madrid: Dykinson.

• Soleto, H. (2017). La conferencia Pound y la adecuación del método de resolución de conflictos. Revista de Mediación, 10(1), 1-6.

• Soleto, H. (2019). Justicia restaurativa para la mejor reparación a la víctima. In H. Soleto (dir.), Justicia restaurativa: una justicia para las víctimas (pp. 491-520). Valencia: Tirant lo Blanch.

• UNODC (2020). Handbook on restorative justice programmes. Vienna: Criminal Justice Handbook Series.

• Vedananda, N. (2020). Learning to heal: Integrating restorative justice into legal education. New York Law School Law Review, 64(1), 95-113.

• Walters, M. (2014). Hate Crimes and Restorative Justice: Exploring causes, repairing harms. Oxford: Oxford University Press.

• Wood, W. & Suzuki, M. (2020). Are conflicts property? Re-examining the ownership of conflict in restorative justice. Social & Legal Studies, 29(6), 903-924. doi: 10.1177/0964663920911166.

• Zahariadis, N. (2014). Ambiguity and multiple streams. In P. Sabatier (ed.), Theories of the policy process (pp. 25-58). Boulder: Westview Press.

• Ziller, J. (2005). Public law: a tool for modern management, not an impediment to reform. International Review of Administrative Sciences, 71(2), 267-277. doi: 10.1177/0020852305053885.

Noten

• 1 Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012, Art. 2, 1(d).

• 2 The crisis of the reintegration model is reflected in crime rates and recidivism when prison fails as a mechanism for the reintegration of offenders (Flores, 2015).

• 3 Faced with the prejudice of some operators who reject the victim’s participation on the assumption that their claims are only economic, Bazemore (1998) concluded that the victims had different objectives. For example, in the case of victims of crimes committed by minors, there is a special interest in the re-education of the juvenile offender.

• 4 With this reference to the United States, the authors do not intend to convey the idea that restorative justice developed fairly homogeneously in that particular country. On the contrary, the heterogeneity in the development, implementation and availability of different restorative practices is a noteworthy subject in the scientific literature (Sliva & Lambert, 2015).

• 5 Kagan (2019), Liu, Yamaguchi and Yoshikawa (2017), Zahariadis (2005), among others.