<p>This paper is based on a Japanese Society of Cultural Anthropology Prize lecture that I delivered at Kyoto University on 1 June 2008. Its major objective is to explore ways in which people constitute a community through their imagining and reflexive practices by drawing on the reflections of my anthropological career since the late 1960s. The 'community' is here regarded not only as existing social relations already constituted, but as being constituted and situated in terms of people's desires, imagination and thinking through their everyday practices. I begin with my earlier and rather crude assumption that novel communality, which has never come into being before or only potentially existed, could be materialized in this world through the struggles, resistance or negotiations of people against ongoing power relations in the midst of the contradictory development of the capitalist system. The argument subsequently focuses on multiple effects of power relations that lead to the construction of communities, by examining the practice theories put forward by P. Bourdieu since the 1970s, by S. Ortner in the 1980s and by J. Lave and E. Wenger in the 1990s. It is argued that whilst the theory of the 'community of practice' by Lave and Wenger sheds light on the processes of gradual participation that enable the participants to acquire knowledge and skills, together with their identity of being members of the community, it largely fails to reveal the power relations involved in such participatory processes. The effect of power refers not only to coercion, control or domination over individuals and groups from outside. It also signifies, as the post-structuralist tradition lucidly maintains, the effects extending to their mode of cognition and system of evaluation, as seen in ideology or discourse that eventually leads to the recognition of the existing social order. To describe instances of resistance and struggle against such power relations thus became a major ethnographic topic towards the end of the twentieth century, in which the actuality of agency, quite different from the modern notion of the subject, is a prominent feature. My ethnographic studies on a peasant leader, spirit mediumship and HIV/AIDS self-help groups in northern Thailand are also concerned with the processes in which villagers, patients or sufferers recreate, through their imagination and reflexive practices, their relationship between the self and other, and the power relations around them. As revealed in other cases, media or resources of their practices are derived from hermeneutic knowledge concerning their own life situated within the communities, rather than from modern rational knowledge. During the last decade of the twentieth century in northern Thailand, people infected with HIV were able to organize themselves into self-help groups to maximize their lives through the establishment of such practices as intra-group counselling, home visiting, holistic health care, as well as struggling against the social discrimination by campaigns and negotiations with medical institutions and government offices. We could thus detect among those self-help groups the emergence of what P. Rabinow calls 'bio-sociality', namely, sociality based on life itself, as opposed to the individualization and fragmentation that proceeded radically under the modern medical system. On the other hand, however, under the recent development of social management, we have to pay special attention to the communality itself that has emerged in such self-help groups, which tends to be transformed into a target of intervention by a variety of powers, such as international agencies, medical bureaucracy, industrial enterprises and NGOs. The recent introduction of an ARV (antiretroviral) scheme since the early 2000s is a case in point. Along with that new system of curing, a great number of HIV-infected</p><p>(View PDF for the rest of the abstract.)</p>