This paper presents an analysis of Community Intervention Teams (CIT), a risk-oriented multi-agency crime prevention program in Sweden targeting youth who are at risk of engaging in criminal gang activities. We have analysed policy documents and information material created by the Swedish government, the Swedish National Police Board and the National Board of Health and Welfare in conjunction with the national implementation of the CIT program. The results show the use of a risk assessment manual is perceived as the only legitimate tool for archiving success in work with CIT, both from an individual and societal perspective. This simplifies the criminological field by focusing solely on the individual as the starting point for crime prevention and ignores the impact of structural factors behind crime and crime prevention. The use of the concept of risk constructs the target group as both potentially dangerous criminals and as a group of vulnerable youth, which needs to be saved to a better life. This duality creates what we choose to call benevolent surveillance, namely controlling interventions that are legitimized by rehabilitative ideals.

This article is an empirical contribution that adopts Arlie R. Hochschild’s theoretical approach to emotions and emotional work to examine the emotional dimensions of working in secure units for youth in Sweden. The institutions in focus in the study are so-called special residential homes for young people. Characteristic of these institutions is a risk-oriented approach towards the targeted group, expressed in an increasing organisational focus on security, such as high fences, locked doors and windows, cameras, alarms and far-reaching restrictions for the youth within these institutions. The aim of the present study is to analyse how these high safety demands become important and are understood by the treatment staff working at the institutions, with a focus on how this understanding is expressed in emotional work. In the analysis, we show that the safety of special residential homes cannot be reduced only to spatial and material dimensions but that safety is also something that is largely achieved done by the treatment staff through the emotional work they perform. We describe two prominent strategies in this emotional work, which we call being on guard without showing it and backing each other up. With Hochschild, we can also capture the interaction that takes place between the institutional context and the emotional work of the treatment staff’s emotional work and how the treatment assistants’ work demands a comprehensive surface action. The empirical basis for the study includes two ethnographic research projects, which comprise seven departments for boys and young men aged 13–21, divided into five special residential homes. The material for the present article is based on participatory observations and interviews with the institutions’ treatment staff.