The overall aim of the thesis is to explore the possibilities for an integrated research perspective on men's violence and to exemplify how such research can be conducted. The specific purpose is to increase knowledge about how violent men's childhood experiences, socialization, masculinity construction and emotions can be related to their violence against other men, against themselves and against women, and how therapeutic interventions against violence can be analyzed and developed in correspondence with this knowledge. With epistemological starting points drawn from critical realism and ecological methods, the study relates research from different schools of thought to each other; - psychological: on childhood experiences and socialization, social psychological: on emotions and interaction and sociological: on social class, gender power structures and hegemonic masculinity. This is done to gain access to knowledge about how different factors interact in men's violence.
Studies I and II investigated the possibilities of examining the social bonds between therapist/therapy and client in therapeutic treatments against violence.
In study I, indicators of the emotions pride and shame were operationalized and in study II these were tested on therapists in a CBT-oriented therapy. Study III examined men in different positions of masculinity, where the sample for one group was drawn from the population of men sentenced to therapy for violence and abuse and the other from the population of men who organized for equality and against violence against women. The study compared the two groups' attitudes to factors related to violence and violence against women in previous research. Study IV examined the careers of men convicted of violence up to their current position as violent criminals in order to increase knowledge of the interplay of factors that in different situations lead to their violence against other men, themselves and women. All empirical studies used qualitative methods for data collection and analysis. Study IV used individual interviews and biographical analysis, Studies II and III used group interviews and deductive content analysis. In Study I, the theoretical review article, sociological, social psychological and psychological theories were empirical.
The thesis shows that there are more advantages than disadvantages to a multi-level perspective. Level-integrating studies are hampered by the fact that they require a complex methodology to deal with the interaction between factors behind violence at different levels, but on the other hand provide a more holistic understanding of the phenomenon in question. The results show that integrative perspectives can reduce the risk of ecological fallacies and 5 increase the understanding of complex interactions between factors behind men's violence, which may contribute to the development of knowledge in the field of violence therapy. The theoretical review article (Study I) exemplified how theoretically and methodologically driven research on social bonds can be made pragmatically applicable by therapists in violence treatment. The applied study of a CBT therapy (Study II) provided examples of how operationalized indicators of pride and shame can be used practically to determine the quality of the social bond between therapist and client. As expected, the CBT therapy studied contained both shame- and pride-creating elements, which constitute valuable starting points for further research. The comparison between men in ideally opposite positions of masculinity (study III) showed that both the group of men who work against violence against women and the men sentenced to treatment for violence carry ambivalent attitudes towards violence and violence against women. The comparison further showed that the groups' constructions of masculinity and attitudes towards violence correspond to the groups' different access to economic, social and cultural resources. The biographically focused qualitative study of men in violence treatment (Study IV) explored exploratively how the career path to becoming a violent offender may look like and how childhood experiences, socialization, masculinity and emotions of individual violent men may have interacted with each other when violence takes place. The results showed that the men who testify to exposure to serious violence in childhood are more shame-prone and, when offended by others, tend to react unconsciously and without prior feelings of shame immediately with aggression and violence against both sexes. Other men were indeed shame-prone but described a more controlled violent reaction. Two men who had been brutally physically bullied in primary school reported more controlled violence. A preliminary hypothesis is that the men may have learnt to cognitively take control of the process of replacing shame with aggression in order to escape further bullying. The parents' personal problems, together with their lack of social control and care, were hypothesized to be associated with several of the men's school problems, their association with deviant youth, their later difficulties in earning a living by conventional means, and their violent careers.