The Prevention of Extremism and the Role of Safety: Essay on a Well-Balanced Relationship Between Social Work and Law Enforcement

The prevention of extremism experienced two developments in recent years: the securitisation of the discourse and the diversification of the involved actors. Both trends caused a certain divergence because the different stakeholders often follow different logics, methods and goals and the influence of security in many cases does not match the needs and interests of civil actors offering prevention work. Especially social work has its own ethical and methodological understanding that requires an own conceptual approach. Above all, universal or primary prevention plays an important role. The aim of the paper is not to separate social work offers in the field of extremism prevention from security politics, but rather to develop an understanding of security that better matches social work profession. In order to solve the actual divergence, a novel conceptual approach—based on the term safety—is developed and applied on preventive measures driven by social or youth work in the field of extremism prevention. The paper develops the outlines of such a concept and various proper methods. In the same time, it reminds of concepts like social cohesion or resilience that also play a role in the discourse about extremism prevention. The paper describes the safety concept by distinguishing it from these related terms.

1 3 stakeholders and their different logics cause a certain divergence because of their different interpretive frameworks used to understand and address extremism.
Extremism usually is perceived as a threat to democracy and therefore as a security issue (Fischer 2021). With the jihadist terror attacks since the beginning of the 2000s, many EU countries elaborated national strategies against violent radicalisation and extremism, that integrated traditional law enforcement and preventive approaches. The strategies resulted in the founding of preventive measures by the different national ministries of the interior. This development caused a direct linkage of security politics and civil preventive offers and thus an upcoming systematic securitisation of extremism prevention (about the German development: Figlestahler and Schau 2021). Although security may have very different meanings, characteristics of securitisation of prevention work often follow a quite narrow understanding, that manifests in a certain distrust of the authorities towards civil actors, a lack of trust between addressees of social work and the professionals (Figlestahler and Schau 2021), and sometimes the perception of addressees as a potential threat of public security (Hamm 2021).
Since the end of the twentieth century a new idea of police work, namely community policing, was developed in the US and-more or less in the same period-in Europe. The crucial intention of this police work is to shift from a mode of reaction to prevention, to get into closer contact with the population, and to integrate public and private organisations into police work (Dehbi 2019: 16-21). Nevertheless, in applied community policing "there is a natural tendency to look exclusively to law enforcement for solutions to problems of crime and antisocial behaviour" (Sabet 2014: 245, in Loeffler 2018. In the context of extremism prevention "new community policing tactics [were] developed and deployed in the name of national security" (Nguyen 2019a: 323) that in many cases have created new informal hierarchies within immigrant communities. An example Nguyen gives are documented "immigrants, [who] sometimes mark themselves as 'deserving citizens' by demonizing undocumented immigrants" (Nguyen 2019b: 249). This is just one example from the US that may illustrate the outcomes of security policies, that tend to hastily stigmatize social groups.
In Europe however, more recently efforts at prevention have focused not only on averting danger, but also on social aspects related to the cohesion of the population. Indeed, several European research projects are currently addressing how conventional law enforcement and prevention work-whether in the form of (community) policing or social work-can be combined to combat extremism. 1 At the same time, this trend has encouraged the second phenomenon noted at the outset: the diversification of the actors involved in extremism prevention. Andreas

3
The Prevention of Extremism and the Role of Safety: Essay on… Armborst et al. have noted "the increasing blurring of the boundary between civil society and prevention of threats by the state" (Armborst et al. 2018: 1, my translation). Furthermore, various expert circles have been devoting attention to the diversification of security policy. For example, the international security conference Security, Democracy and Cities 2021 (organised by the European Forum for Urban Security, EFUS) dedicated a panel to precisely this topic. It concludes: "Besides the public institutions operating in the security field, other stakeholders contribute to security policies, such as the third sector, citizens, the commercial sector and private security companies" (EFUS 2021). The EFUS report additionally notes that divergent stakeholders' professional cultures, approaches and purposes engender questions related to the development of a common "security culture".
Social work has a special role within this diversification of actors. Various international actors involved in extremism prevention, including the European Radicalisation Awareness Network (RAN), have "identified social work methods as a major component in good practice models for PVE [prevention of violent extremism]" (Hutson 2021: 300). The present contribution is mainly inspired by the German and Austrian professional social work discourse. Anyway, in the context of research and transfer projects and of scientific conferences I could observe that social work practitioners share a very similar experience in many European countries. 2  The present paper does not aim to separate social preventive measures from the field of security, but rather to find an understanding of security that better fits to the logics and ethics of social or youth work. So, when it comes to the diversification of security and crime-prevention policies, it is helpful to define the concept of security itself. The German term Sicherheit has two equivalents in English: "security" and "safety". In this respect, English is more nuanced than German, yet the precise relationship between the two terms in English is often obscure. "Security" generally pertains to strategic or political domains, describing a general social condition for which the state is (at least partially) responsible. "Safety", on the other hand, tends to be more individualistic and related to the condition or perception of being protected from harm or threat.
In the paper, I argue that an emphasis on "safety" has inherent strengths compared to "security". Consequently, the two terms should be used as different concepts for different tasks. If the two concepts are invested with discrete meanings, responsibilities and interventions in the field of extremism prevention can be clearly separated.
In order to systematically discuss this topic, I first define some key terms in greater detail. At the outset, I outline the dynamics of radicalisation processes and examine the topic of prevention, especially primary prevention (Sect. 2). This discussion then informs Sect. 3, which is devoted to the concept of security in relation to extremism prevention. Next, different facets of "safety" are examined in more detail, and I explain how this concept is able to compensate for deficits attendant to the concept of security in the domain of extremism prevention (Sect. 4). Finally, safety as a term is outlined and differentiated from related terms such as social cohesion and resilience. In addition, methods of action resulting from the concept are presented. The advantage of such a discrete concept is that it allows social work and law enforcement interventions to be distinguished from each other and thus better coordinated at the same time (Sect. 5).

Extremism and Radicalisation
The formulation of preventative measures inherently necessitates an adequate understanding of the conditions or processes they intend to avert-namely, a movement toward violent, anti-democratic extremism, particularly among youths; in other words, "political extremism", which "refers to positions and movements that are directed against the existing (democratic) principles and institutions, aiming to establish a different political order" (EU 2017: 14). Yet in this regard it is crucial that a critical position-even if extremely critical-never should be considered extremist "as long as it does not leave constitutionalism, popular sovereignty, pluralism and fundamental human rights" (Jesse and Mannewitz 2018: 15, my translation). Extremism is a broad term that is not connected to a specific ideological context. It can be a matter of various phenomena that ultimately only "share the common feature of being incompatible with individual (or all) institutions of constitutional democracy" (Jesse and Mannewitz 2018: 15).
Violent radicalisation is understood here as a process that leads to anti-democratic violent extremism. Various factors foster this process. There is agreement that causal relationships cannot explain the dynamics of the radicalisation process. "Radicalisation is therefore not exclusively about the presence of certain factors and influences, but also-and especially-about their interaction, development and course" (Neumann 2013: 3, my translation). Prevention is directed towards such factors, which include psychological-biographical factors (personality traits, family socialisation), social-psychological conditions (relative deprivation, group-sociological, socioeconomic factors), political and ideological factors, as well as cultural and religious factors (Miliopoulos 2018). There is no consensus in academia about the connection 1 3 The Prevention of Extremism and the Role of Safety: Essay on… between extremist attitudes and extremist actions. Some experts assume that only people who cultivate an extremist ideology will also commit extremist acts. Others argue that there have been terrorist acts committed by people who were not ideologically motivated per se, but rather impelled by a sense of belonging to an extremist group (Neumann 2013).
There is also no evidence that extremists are psycho-pathologically abnormal in any way (Kaya 2020: 7). Rather, questions of collective and personal identity and collective grievance are relevant (Sageman 2017: 135-138). As young people are in a phase of identity formation, they are particularly susceptible to extremist attitudes. "From early adolescence (i.e. between the ages of 12 and 15) until their mid-twenties, young people are specifically vulnerable to the threat of radicalisation and violent extremism" (RAN Paper 2018: 2). Particularly young people who have some form of collective experience of grievance can come to an orientation that is hostile to prevailing social norms. But also privileged young people may radicalize if they get in contact with extremist groups or ideologies. The radicalisation process can develop in three stages: "social affiliation, progressive intensification of beliefs and faith" (Bove and Böhmelt 2015: 8, my translation) and formal acceptance of the need for violent acts or even terrorism. Therefore, social relations or affiliations are central variables in this development: "Throughout these steps, social bonds play the most important role, as they provide mutual emotional and social support, development of a common identity, and encouragement to adopt a new faith" (Sageman 2004: 135).

Prevention
Prevention is usually understood as measures that aim to prevent future undesirable conditions or actions. Prevention can begin from different starting points. For example, one can reach out to already radicalised persons, in the effort to prevent them from carrying out (further) extremist acts. This is known as tertiary prevention (Koehler 2017: 67). In the area of social interventions, prevention aims to reintegrate extremists into society (e.g. via "exit offers").
Another approach, known as secondary prevention, attempts to reach social groups who express extremist ideologies or appear to be particularly at risk-for example, due to socio-economic disadvantage or relative deprivation, or who express extremist ideologies. It is a form of targeted or selective prevention that draws on predictive forecasts concerning specific societal subgroups (Greuel 2020). While forecasts of susceptibility to radical behaviour may be justified by statistics, the literature finds no empirical evidence for a direct link between so-called cognitive and violent extremism (Neumann 2013). A second disadvantage of such an approach is that it can have a self-fulfilling effect. For example, when young people who seem particularly likely to take a criminal or extremist path become the subject of interventions different to less at-risk young people, this segregation can produce the life trajectory it aims to avert, as Royce Hutson notes: Past experience demonstrates that ‚at-risk' youth has often been a dog whistle for minority and/or disempowered youth. That path has met with justifiably serious social justice critiques, especially in this context [i.e. the prevention of violent extremism], and at its core a psychologizing deficits-based approach (Hutson 2021: 302/303).
While the secondary prevention is politically and socially accepted, it nevertheless contains hidden dangers. Nguyen (2019a) analyses community-police partnerships in the US and concludes that in many cases they focus on so-called at-risk groups with the aim to especially observe and address them. From this perspective, they may be considered secondary preventative measures that produce so-called "suspect communities" (Nguyen 2019a: 324) with a re-stigmatising effect.
Since individuals may detach themselves from majority groups due to collective offenses in their social environment-a phenomenon to which the young are particularly susceptible-another prevention approach appears promising, namely primary prevention (Koehler 2017: 67) or "generic prevention" (EU 2017: 20-25). This approach does not address specific individuals or groups, but the community as a whole, seeking to promote social affiliation in the community before individuals succumb to radicalisation. The prevention of extremism is thus sought through a higher degree of social inclusion (Reich 2012). However, this form of prevention can hardly be carried out by the police; rather, it is the purview of social and youth work. Additional goals include encouraging all participants to formulate, articulate and discuss their concerns. Such measures thus indirectly promote trust in democracy and social cohesion. Primary prevention is a sort of universal prevention, which "starts with target groups that do not show any conspicuous features or an increased risk before a certain problem occurs" (Greuel 2020: n.p., my translation). Kessl and Reutlinger (2009) vividly describe the potentially close proximity between stigmatising selective prevention and socio-spatial universal prevention. Focusing on the example of a group of adolescents from a youth residential group who have a food stand at a district festival, the authors show how the same measure may have a stigmatising or an inclusive effect, depending on how the target audience is defined. If the youths are the addressees of the preventive measure-which, in this case, was designed to promote integration into the neighbourhood-the initiative may be stigmatising. "The young people are thus placed under special observation simply because they live in the residential group" (Kessl and Reutlinger 2009: n.p. my translation). If the neighbourhood is the addressee, the intervention is imbued with an inclusive character. Therefore, even universal or primary prevention measures have to reflect on the circumstances under which action is taken, even if their actual goal is to address all participants in a social space.

Social Security and the Prevention of Extremism
Which aspects of the term security are relevant when it comes to the prevention of extremism? Security-and as a consequence: securitization-do not necessarily mean exclusively, that society has to be defended from threats from outside or from explicit forms of delinquency or extremism growing within the society. Thus, action concepts or methods either may vary, depending on the 1 3 The Prevention of Extremism and the Role of Safety: Essay on… understanding of security. A brief look at the history of the term security shows that it can take on many different meanings, and thus can be loaded with divergent political implications. I argue that social security is of special importance for the topic of extremism prevention. Social security refers to economic security, that was systematically elaborated for the first time in the so-called "Social Security Act" of the US-American president Franklin D. Roosevelt. In terms of social security he highlights "decent homes", access to "productive work", and "safeguard against misfortunes" (see Kaufmann 2003: 81). But also a life in dignity and free development of personality are a crucial part of social security, as the Human Rights Declaration states: Everyone, as a member of society, has the right to social security and is entitled to realization, through national effort and international co-operation and in accordance with the organization and resources of each State, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality. (UN 1948: §22) At first glance, social security-that often refers to economic security-might not seem urgent where it comes to the prevention of extremism. With regard to extremism, a type of security that can be established through police intervention seems to be more important. But when it comes to fostering the social conditions that help prevent the emergence of delinquency and, by extension, extremism, access to housing, employment and protection against misfortune all play a crucial role in early universal prevention.
With the rise of economic insecurity in the 1930s, the term security was used not only to refer to social security, but also to insecurity in relation to a loss of traditional values. This fact, which became important in the rapidly changing world of that time, can also be related to our present-day society, which is characterised by ever-increasing globalisation, digitalisation and pluralisation. Aside from its proximate effects, the deployment of security forces by the executive branch can be interpreted as an indirect effort to guarantee traditional values. This phenomenon has a negative effect when groups of people are addressed who consciously or unconsciously question the traditional values held by the majority. Security is then only established for one part of society, at the risk of excluding other social segments. In such a case, the deployment of security forces may have unintended consequences, insofar as the devaluation of social groups leads to increased insecurity. General security for all, on the other hand, can only be achieved if social exclusion is prevented. This approach is particularly important in the field of extremism prevention, because: Radicalisation processes are not independent of socio-economic and political contextual factors. 'Fraternal relative deprivation', i.e. the perceived disadvantage of the group to which one belongs or with which one identifies, is the best predictor of radicalisation so far (Eckert 2013: 6, my translation).
Trust in such a complex system is a crucial factor. In this context, Kaufmann uses the terms external and internal security. Here, external and internal security do not mean the distinction between security within the nation and protection against threats from outside, but rather the distinction between the security of a system and the individual's perception of security. In addition to external or system security, internal security plays a key role in shaping security policy. According to Kaufmann, individuals have the need and desire to shape their own environment. Two approaches can be harnessed to address this situation. "Low confidence in one's own ability to act" (Kaufmann 2003: 94, my translation) can be counteracted with the help of empowerment (Solomon 1976) and inclusion (Reich 2012). Especially under the conditions of a rapidly changing society due to economic change and internationalisation, it is important to aim for inclusion. The second approach supports "the re-embedding of systemic contexts in familiar social contexts characterised by spatial proximity or intimacy" (Kaufmann 2003: 95, my translation). Social-space oriented forms of social intervention are appropriate (cf. Deinet 2009), since they promote the participatory shaping of concrete living environments and the development of a common culture.
An important sphere of prevention is the internet and social media. The internet is a virtual social space that interacts with the urban social space (cf. Kreß 2010) and in which inclusive, empowering interventions can be made. But the rise of virtual space has engendered new issues: through the internet, perspectives from distant locales may enter the local context and shape peoples' perceptions. This also brings a more recent approach into focus-namely, "situated security" (Bonacker 2021, my translation). A given spatial area or territory may host different "situations"-that is, divergent mixtures of perceptions and experiences. In this way, influences with various origins-e.g. from other local areas or from virtual spaces-may appear within one community. The main idea behind this approach is that, on the one hand, a current understanding of security develops from the perspectives of those actors who are able to publicly articulate their concerns and interests. On the other hand, the horizons of experience and perspectives of the actors involved will mix, so that within a territory, such as a municipality, different understandings of security may arise. Security therefore takes on "a relational character" (Bonacker 2021: 15, my translation). Thorsten Bonacker's proposal is to make "strategies of silencing, i.e. exclusion of the possibility of articulating threat experiences" (Bonacker 2021: 17, emphasis in original, my translation), analytically visible, in order to comprehensively survey security perceptions. In relation to our present concerns, this analytical approach can be harnessed to inform possible preventive action measures. To be sure, the understanding of security that underlies the prevention of extremism must be developed from the perspective of all residents of a neighbourhood or community, in order to avoid paternalistic decisions about the needs and interests of residents and to avoid excluding or even stigmatising individuals, or producing "suspect communities" (Nguyen 2019a: 324). Communication about the different understandings of security must therefore be an incremental part of prevention measures.

3
The Prevention of Extremism and the Role of Safety: Essay on… 4 Safety

Towards Safety
In the last section, various facets of the concept of security were elucidated. Some aspects-such as protection from social and economic insecurity, inclusion, and empowerment-furnish a basis for activities in the area of universal extremism prevention. Other aspects, such as the prevention and prosecution of delinquency and extremist acts, match secondary or tertiary prevention of extremism. A central argument of this paper is that these two approaches are categorically different, and, as such, they should be used to elaborate different concepts of action, for this allows extremism to be combatted from alternating directions.
Such a separation already takes place within the domain of law enforcement by means of community policing. Although "public safety goals" (Zhao et al. 2001: 370) are addressed by such policing, law enforcement is generally not informed by an understanding of security that appropriately addresses socio-economic problems or promotes the inclusion of local milieus.
These facts are responsible for the diversification of stakeholders in the field of extremism prevention mentioned at the outset of this paper. However, it is problematic that conceptually, there is no clear distinction between areas of responsibility, and associated fields of activity are sometimes interpreted differently (Schreiber 2013: 11).

Safety: Fundamental Considerations
There is no common or universal definition of safety. Usually, it refers to the perception of individuals, as defined in relation to risk or threat. For example, Nicole De Wet et al. state: Perceived safety is a subjective measure which speaks to an individual's state of mind including their fears and is partially based on their own experiences and that of others (De Wet et al. 2018: 3).
Under this definition, safety depends on an individual's subjective assessment of a situation, which can be described with the help of the security concept: Depending on the danger or risk that emanates from a given environment, the individual adopts a certain attitude or renders an emotional judgement, which can then be described with recourse to the concept of safety. Abraham H. Maslow already in the 1940s had a similar idea of the relationship between safety and security: The peaceful, smoothly running, 'good' society ordinarily makes its members feel safe enough from wild animals, extremes of temperature, criminals, assault and murder, tyranny, etc (Maslow 1943, 374-375) Risks and threats in the environment do not hinge entirely on subjective assessment, however, for they relate to actual, empirical conditions. At the same time, beyond being a source of risk or danger, objective conditions normally offer opportunities for inclusion and participation. In this way, safety can be measured at least partially with metrics that do not depend on individual perceptions or assessments. Thus, while the security situation of a neighbourhood may be measured in terms of the number or types of crimes committed, one could measure safety in terms of opportunities for participation at the local level; in terms of the socio-economic situation of the residents; or in terms of residents' perceptions, expectations and plans. While the last component is once again subjective and individual, but it can only be meaningfully assessed in the context of external and objective conditions. This brings safety in close propinquity to the concept of social cohesion (De Wet et al. 2018: 2).
In the following, I seek to introduce safety as a concept for social interventions that are practised in parallel to law enforcement work.

Safety: Sources for a Concept
The literature differentiates between forms of safety according to areas of application. These are: (1) industry, which is primarily concerned with safety standards: (2) management, which aims to promote personnel and organisational development through "psychological safety"; and (3) social or community safety, which addresses vulnerable persons or groups, and risky individual behaviour.
Ad (1): In industry, safety is usually understood as operational safety (Endreß and Petersen 2012). Safety can be achieved with the help of accident prevention measures, such as the covering of machines to prevent the splintering of material during the processing of a workpiece. In this way, it differs from security, which aims to prevent outside intruders from compromising physical or virtual assets (e.g. through cyberattack). A central difference between safety and security is that safety measures are regulated by law through standards and industry norms, while security is left to the individual company (Geiger n.d.). It is clear-and relevant in this context-that safety refers to internal, operational security while security refers to external security.
Adapted to the broader sphere of society, this conceptual distinction can be used to describe the relationship between external attacks and social security. Examples include interference in democratic elections by international hackers, or the hijacking of discussions in social media by trolls, or bots that disseminate hate speech. In both of these examples, the underlying aim is to disrupt democracy, social cohesion and social peace. Yet another example of external influence on internal security is the recruitment of young people by foreign extremist or terrorist groups. Internal, social security or "operational security" in this context means social cohesion and trust in the democratic system. It is precisely here that safety can be adapted as a concept to the subject of extremism prevention. The means that can be used to prevent violent or hostile action are of special importance in this regard. In the understanding of security cultivated by law enforcement agencies, a focus is placed on the apprehension and prosecution of hostile actors. Within the domain of social security or safety, by contrast, the goal should be to strengthen social actors so they are empowered in order not to be influenced by anti-democratic attitudes or even to be recruited by extremist groups. In the case of hate speech, this could take the form of the targeted moderation of discussion boards, or the deployment of counter-or alternative-narratives (Schlegel 2021). In relation to the problem of youth recruitment, open youth work or school social work could represent possible forms of action.
Ad (2): In the field of management, one speaks of psychological safety. This initially referred to trust and good cooperation within a group. In contrast to security, safety is typically discussed in the domain of management from the perspective of the individual in his or her environment, and often refers to trust within companies. In this context, safety usually means that individuals feel free to express their thoughts within a team of colleagues (Newman et al. 2017: 523).
In the context of safe work environments, trust is a counter-concept to verbal injuries, especially in the relationship between supervisors and employees, as well as between colleagues (Luria 2010). Accordingly, trust is concerned with the relationship between individuals. In this way, psychological safety includes.
items that capture shared perceptions amongst team members as to whether they believe that others will not reject members for being themselves, team members care about each other as individuals, team members have positive intentions to one another, and team members respect the competence of others (Newman et al. 2017: 523).
These characteristics of safety remind of social interventions that work with the concept of safe space. The conception of safe space seeks in part to foster a certain tolerance of conflict. The goal is not to ensure a homogeneous group or opinion, but rather to create the conditions for the individual to express and discuss his or her own opinions without danger (Bergold and Thomas 2012;Berner et al. 2020: 125).
Of special importance for psychological safety are supportive superiors, organisational practices such as diversity measures, and relationship networks within the company (Newman et al., 2017). Social support and social capital in particular are "key determinants of psychological safety" (Newman et al. 2017: 525). At the team level, rather unexpected conditions also promote psychological safety. For example, in one study, the heterogeneity of a group was found to be a positive factor, because "the presence of strong fault-lines within teams (i.e., the existence of sub-groups with non-overlapping demographic characteristics) led to greater psychological safety among team members" (Newman et al. 2017: 526). Both characteristicssocial capital and the coexistence of diversity within groups-belong to an understanding of social security as it is discussed in relation to the prevention of extremism, because the isolation of individuals within a group can encourage them to seek other arenas for belonging (such as membership in extremist groups).
A key difference between extremism prevention and psychological safety in management relates to desired effects. Usually, a productive, creative and innovative workforce is the aim of psychological safety. In the context of extremism prevention, by contrast, the goal is to achieve the democratic, harmonious living arrangements with equal opportunities. The strength of the concept therefore lies less in its intended outcomes than in the characteristics of psychological safety, which represent values in themselves. These can be including trusting cooperation, a constructive culture of conflict, and social equitable conditions, including equal access to resources. Nevertheless, the effects of successful psychological safety are also relevant to the present subject matter. For example, a significant effect of psychological safety lies in the "voice behaviour" of employees, i.e. in their verbal participation in group meetings (Newman et al. 2017: 525). In this way, the concept also counters the phenomenon of "silencing" (Bonacker 2021: 17).
Ad (3): Unlike psychological safety, which refers to teams or organisations, social/community safety focuses on the relationships between institutions and citizens or between citizens within a society. But community safety itself is not uniformly defined. On the one hand, the concept can relate to the threat of crime in a neighbourhood (Loeffler 2018). On the other hand, it may concern the inclusion of vulnerable groups (Terras et al. 2019). The concept of community safety seems particularly suitable for adaptation to extremism prevention, as it is informed by notion of inclusion, which is essential in relation to social security. Fundamentally, this form of safety is based on the observation that "engaging citizens in meaningful participation in local practices and decisions that shaped their lives led to increased participation at a community level which protected and advanced citizens' interests within broader society" (Terras et al. 2019: 40). The central idea is that safety must apply to all residents of a community.
Safety therefore means making inclusion and participation attractive for all members of society, including the "marginalised and disadvantaged" (Terras et al. 2019: 40). This may require strengthening trust in institutions within these segments of the population. Safety in this regard also refers to personal perceptions and attitudes, such as belonging or a sense of community. Social interventions aimed at combating stigmatisation and discrimination thus become the focus of action. One challenge of such an approach, however, is that-with the greater diversification of the populace-new perspectives and associated requirements are emerging within communities. This "may change the dynamic and initial shared vision of the group" (Terras et al. 2019: 41). This often does not succeed without further social intervention, since an increase in opportunities for participation in a community does not necessarily lead to the participation of vulnerable or stigmatised groups.
In terms of social action, this means that people who have not participated much so far require active encouragement. Alternatively, the reasons for a lack of participation need to be identified. Causes may include low self-esteem or social isolation as highlighted by Terras et al. in reference to persons with mental disability or dementia. Other stigmatised groups of people will have other reasons for lack of participation, and the reasons may vary within a city, or from district to district. Therefore, an individually tailored package of social action measures may be necessary, based on individual needs and the specific characteristics of the concerned districts or neighbourhoods. Social work knowledge applied in the local context plays a special role in this regard-for example, in terms of "community resilience building, a common community-level practice, […] strengths focus, empowerment, trauma-informed, and ecological models of prevention/intervention (aka 'whole-of-society')" (Hutson 2021: 300). Terras et al. recommend "kindness" (cf. Ferguson 2017) as a general concept that addresses the needs of all inhabitants 1 3 The Prevention of Extremism and the Role of Safety: Essay on… of a community, since "the development of inclusive communities has a further advantage in providing a unifying framework for considering all members of the community" (Terras et al. 2019: 51). Finally, this approach means a reversal of the security logic, such that one moves away from a focus on threats and towards the provisioning of community resources.
In sum, I argue that the concept of safety should be viewed as closely linked to an understanding of social security that addresses the internal security of a community and, at the same time, the internal security of the persons concerned, i.e. their perceptions and needs for participation and belonging. Safety-discussed here along the lines of industry, management and community-refers to these two forms of inner security. Social work informed by this understanding of safety aims to foster harmonious social relations between milieus, and also seeks to integrate heterogenous needs and perspectives. Furthermore, this conceptual understanding strives to develop concrete measures that are able to support the goal of a democratic, participative, inclusive togetherness. In the following concluding section, these findings are distilled into a novel conceptual understanding of safety.

Safety: A Novel Conceptual Understanding
Securitisation of extremism prevention often results in secondary preventive approaches, that may have a negative impact on the relationships between social workers and their addressees but also foster a certain distrust between social work and law enforcement. Secondary prevention causes ethical problems to social work, because it fosters stigmatization and exclusion. While such approaches may be justified in reference to concrete, actual threats, there is an attendant risk of stigmatising or discriminating against individuals or groups. In this case, such measures may have the opposite effect of their original goal, because they can encourage the marginalised to turn their actions and individuality to groups. When such groups are extremist, this may lead to the growth of extremism within a community. It is precisely here that social work has a role to play. However, in an increasingly securitised environment, social work lacks a conceptual basis to position itself in the field of extremism prevention. In this foregoing, I have attempted to show that the term safety can furnish such a conceptual basis.
The concept of safety encompasses universal primary prevention at the level of the social space. It addresses the vulnerability of communities resulting from conditions that exclude, stigmatise or disadvantage individuals. It focuses on the wellbeing of citizens, mostly young people, and seeks to promote inclusion, social cohesion and community resilience. Questions arising from this approach pertain to community resources rather than deficits. They may be, for example: What is going well? What strengthens the community? Safety takes place in districts, in neighbourhoods and in small-scale settings-such as classrooms-and relates to the wellbeing of the individuals involved. Table 1 provides an overview of the main conceptual features of a security concept and a safety concept, as elaborated in this paper. Indicators for the success of safety are partly socio-economic and socio-spatial, and partly they concern the perceptions of residents within a community. Safety integrates some characteristics of security, especially those that can be described with social security. At the same time, it addresses all persons of a community and supports their participation and inclusion. Together with indicators that can be used to operationalise security in a narrower sense-such as crime rates-they are thus closely related to the concept of social cohesion (Güntner 2009;Jenson 2010). One could say that if safety and security are present in a community, this corresponds to a high degree of social cohesion.
Another term that is often used in relation to extremism prevention is resilience. A distinction is made between the resilience of individuals, including in particular young people (Benjamin et al. 2021), and the resilience of communities (Stephens et al. 2021). Resilience in the context of extremism prevention represents resilience to ideological capture by extremist groups. Safety refers less to the resilience of individuals, but in relation to resilient communities, there is overlap between the concepts: "Resilience, associated with a strength-rather than deficit-oriented perspective, becomes an attractive concept in seeking an alternative to explicitly security-driven approaches" (Benjamin et al. 2021: 54). Safety, however, differs from resilience in that it has a stronger emphasis on political participation and solidarity, whereas resilience is aimed more at health promotion and is therefore depoliticised in its origins (Sørensen et al. 2012: 8). Safety is in this respect closer to an understanding of empowerment that promotes political activism (Herriger 2014: 21-38).
Safety as presented here provides concrete practical methods. Some of these have already been mentioned: • Universal understanding of early prevention with reflexion of measures (Are they inclusive or stigmatizing?) • Social space-oriented approaches • Empowerment of stigmatised and/or vulnerable persons/groups; safe spaces • Anti-discrimination work and diversity measures • Support in building social networks and social capital • Establishing institutions that promote networking, exchange and constructive dispute and support in accessing such institutions • Promotion of local political participation and other forms of participation • Kindness-oriented approaches • Focus on the strengths and resources of a community Safety provides the opportunity to re-embed security in local contexts. Security on a larger social scale is then created by concentrating on smaller units such as neighbourhoods or city districts. Safety thus means at the same time (1) the safety of each individual participant in a community (protection from being excluded); (2) as a result, the safety of the community (spatial safety; "a safe district"); and (3) if applied to a wide range, social security as the sum of many instances of local safety.
Nevertheless, one must also consider different understandings of safety based on experiences and relationships that extend beyond a concrete local context, as well as import situational demands in a local context based on outside influences (Bonacker 2021). Safety is a concept that mediates these different understandings and provides a framework for community residents to negotiate their perspectives.
Safety is not understood here as a substitute for security and the work of law enforcement agencies. Rather, the concept intends to allow both actors to meet at the same level, to define their tasks clearly and separately from each other, and to examine which overlaps, symmetries, support possibilities, but also dividing lines exist. In concrete terms, this means that the actors should meet regularly with city and community leaders in order to coordinate their respective goals, measures, and successes-as well as their problems with one another. Existing programmes like community policing are a first step in this direction and allow a meeting of the actors with low threshold.