Close insecurity: shifting conceptions of security in prison confinement The study of security within a prison environment implies the observation of a complex phenomenon: on the one hand, inmates are defined as agents of insecurity, insofar as they are the authors of criminal acts, which to the outside world represent everything that is perceived as a threat – in terms of the law, order and general well-being. On the other hand, the prison is often characterised as a space riddled with fear, uncertainty and in- security, manifest in the everyday life of prisons. In this article, based on a two-year fieldwork in three Portu- guese male prisons, I explore the meaning attributed to security from inmates’ perspectives and discourses. This analysis, which includes inmates with different ages, origins, types of crime and sentence length, as well as spec- ificities inherent to the chosen field sites, allows us to expand and deepen our understanding of the significance of security within a population that is often excluded from this discussion, albeit invariably related with it. Key words prison, security, insecurity, Portugal, confinement Introduction For some reason one stupidly thinks a criminal act has to be more thought out, more deliberate than an innocuous one. But there’s really no difference. Actions know an elasticity that ethical judgments are ignorant of. (Saviano 2004: 14) ‘Security’ does not mean the same thing to an adult man as it does to a woman, a child or a group of youths (Eriksen et al. 2010). It is not the same when uttered by a police officer, by a politician, a common citizen, or a given group or community. It assumes different features in a small village, a suburban district or the centre of a major city. An- thropology has only very recently begun to address security as a central theme (Maguire et al. 2014; Frois 2013; Holbraad and Pedersen 2013), as opposed to an ancil- lary or subjacent motif; one that, as Daniel Goldstein points out in his call for a critical anthropology of security, ‘reveals not only the ways in which global discourses are situated and manipulated in the face-to-face contexts of ethnography – it also transform the way security itself is conceptualised in a historical and contemporary global reality’ (2010: 499). Despite the abundance of anthropological works on violence, racism, xenophobia, human rights or poverty (e.g. Comaroff and Comaroff 2006; Parnell and Kane 2003; Kaldor and Stiglitz 2013; Scheper-Hughes and Bourgois 2004), security has rarely found a place as a theme in its own right in the discipline. 1 Even though prison is studied by several other disciplines, this security gap is also confirmed. For example, we can notice that several works from criminology develop 1 A different focus on security is related with the employment of anthropologists by private and official security institutions in the USA working in collaboration with governmental strategies to fight terrorism (Albro et al. 2012). Social Anthropology/Anthropologie Sociale (2016) 0,01–15. © 2016 European Association of Social Anthropologists. doi:10.1111/1469-8676.12299 CATARINA FROIS