Public or private violence? Understanding the overlap between intimate partner abuse and susceptibility to violent extremism

The link between intimate partner abuse (IPA) and violent extremism has become an area of media and policy interest – sparked both by newly emerging extremes (e.g., Involuntary Celibates; incels) and high-profile attacks where the perpetrator had previously perpetrated IPA. These developments blur the boundaries between forms of violence traditionally treated in silos of public and private (domestic) violence. However, while IPA has been observed as present in the backgrounds of a range of violent offenders, including extremists, it is important for the risk assessment and management of both to understand if and how IPA is relevant to violence risk. To unpack this relationship, we apply psychometric network analysis to survey responses from a sample of men from the UK general population. A series of network graphs visualise how IPA perpetration, attitudes towards violence against women, attitudes towards violence, exposure to violent extremism, and violent extremist attitudes and intentions relate to one another. Our findings suggest that IPA may be an observable (although admittedly crude) indicator of the types of attitudes which underpin both types of violence, and that exposure to violent extremism may, in part, explain the blurring of boundaries increasingly observed in practice. We introduce a number of protective factors into the models to visualise how they might mitigate risk, to inform the design and delivery of risk management in this space. • Intimate partner abuse and violent extremism overlap via gendered attitudes, specifically support for violence against women, which underpin both types of violence • How IPA relates to violent extremism is via exposure, where men seek out worldviews which align with their own, including contexts supportive of violent extremism, given the often explicit support for violence against women across extremists narratives • Empathy, resilience, internal locus of control, and social support, demonstrate protective affects that buffer against different features of the risk network – these have implications for creating bespoke risk management plans to prevent both types of violence • However, security agendas challenging patriarchy must ask two questions: not only, does this make PCVE more effective, but does this make the prevention of violence against women more effective? Violence against women is far more widespread than violent extremism, and this work should not be subordinated to a security agenda