Connections built and broken: The ontologies of relapse.

'Relapse prevention' has become a familiar concept and practice for those engaged with drug treatment services. The ways that 'relapse prevention' is currently practised and talked about departs primarily from research produced within the discipline of psychology, and especially by researchers and practitioners adopting cognitive behavioural (Marlatt & Donovan, 2005; Witkiewitz & Marlatt, 2009) and neurocognitive approaches (Tapert et al., 2004). The outcome has been the production of 'tools' and 'mechanisms', put in place to 'prevent' people from relapsing. This way of thinking about relapse has generated the assumption that once access to these 'tools' has been granted, relapse becomes a problem of the individual, a personal 'success' or 'failure', depending on how these tools are used, a measurement of how much one 'really' wants to recover. This system of thought reproduces longstanding discourses of blame against AOD users and fuels the discussion on the 'revolving doors' of recovery (White & Kelly, 2010), holding treatment services accountable for 'failing' to produce and maintain 'recovered' bodies. In this paper my aim is to challenge the production of relapse as a 'threat' and to rethink it as a desire to connect, a desire that can be either enhanced, or broken. Drawing on empirical data produced in two recovery services, one in Liverpool (UK) and one in Athens (Greece), analysed through a Deleuzo-Guattarian system of thought, I discuss relapse in two different ways: (a) as part of the temporality of recovery, a way to start building connections with services; as the expression of an emerging desire under exploration, and(b) as the consequence of broken and interrupted connections when policy fails to support the encounters emerging in the recovery space, disrupting thus the recovery process.