Social Science & Medicine 317 (2023) 115636 Available online 24 December 2022 0277-9536/© 2022 Elsevier Ltd. All rights reserved. The ends of an assemblage of health Cameron Duff Centre for People, Organisation and Work College of Business and Law RMIT University, Australia A R T I C L E INFO Handling editor: S Susan J Elliott Keywords: Assemblage Power Processes of selection Health geography ABSTRACT This paper identifes and responds to three key challenges that have emerged in discussions of the assemblage across health geography. These challenges concern the problem of identifying the borders or limits of an assemblage of health; the problem of clarifying how such assemblages change over time; and the more general problem of identifying the affective and material character of the assemblage such that one may distinguish ‘therapeutic’ from ‘oppressive’ arrangements. The paper argues that each challenge calls for a novel analytics of power grounded in assessments of the generative forces of stratifcation and selection expressed within an assemblage. Assemblages of health are composed in relations of power, affect and desire that stratify the assemblage in ongoing processes of selection, acting upon heterogeneous entities (material and immaterial, intensive and extensive, human and nonhuman), bringing them into contact, causing them to affect one another, transforming their activity. Analysis of these processes provides potent tools for rethinking how relations, events, spaces and encounters mediate experiences of health and illness, and novel grounds for intervening in the for- mation of an assemblage of health. 1. The ends of an assemblage of health The assemblage offers health geographers a remarkably versatile ontology. From urban life and the segmentations of individuals and groups to the relatively durable form of social structures, health prac- tices, norms and beliefs, all assemblages trace the “co-functioning” of heterogeneous entities as they encounter one another, acting on and affecting other assemblages (Deleuze and Parnet, 1987:51). This is the realism proper to the assemblage (DeLanda, 2016), and a clue perhaps to the enduring appeal of this idea across health geography (Andrews, 2019; Bell et al., 2018). All that is real, all that we may be inclined to assert exists, is the expression of a unique assemblage of matter, affects, bodies and signs, conditioned by the events, encounters and relations that give these elements their formal composition (Deleuze and Guat- tari, 1987). The generative theorising inaugurated within this ‘specu- lative realism’ (DeLanda and Harman, 2017) has yielded potent tools for rethinking how relations, events, spaces and encounters mediate expe- riences of health and illness (see Andrews, 2019; Andrews and Duff, 2019; Atkinson, 2013; Duff, 2014; Foley, 2011; Gorman, 2017). Yet this growth has sometimes come at the cost of the idea’s conceptual rigor, what Deleuze and Guattari call the assemblage’s ‘cutting edges’, or the capacity to identify “with precision the specifc characteristics and fea- tures of a concept” (Buchanan, 2015:383). This paper attends to the assemblage’s cutting edges by responding to three key challenges that have emerged in recent discussions of the assemblage across health geography. These challenges have emerged both in my own work (see Duff, 2014; Duff, 2016; Duff, 2018), and in my reading of cognate accounts of assemblages of health and illness. In addressing these challenges, I will refect on some of the ways “assem- blage thinking” (Anderson and McFarlane, 2011:124) has shaped research across the discipline, though I do not propose to systematically review the varied fndings that have emerged in this research. Rather, my hope is to contribute to the ongoing development of what Thomas Nail (2017:21) calls the “general logic” of the assemblage by identifying what I regard to be key challenges in empirical and conceptual accounts of the assemblage in recent health geographies, and the beginnings of what I propose to be effective responses to these challenges. Considered in this light, the paper contributes to ongoing efforts to determine “how ontology translates into methodology” (Fox and Alldred, 2022:625) by further specifying how the assemblage ought to be approached as a fgure of social science inquiry. Studies of the ways assemblages shape experiences of health and illness have grown rapidly in health geography (see Andrews, 2019; Andrews and Duff, 2019; Crooks et al., 2018; Power et al., 2019 for reviews), alongside cognate work in sociology (Coffey, 2020; Fox, 2011; Fox and Alldred, 2016; McLeod, 2017) and anthropology (Brown and Reavey, 2019; Zigon, 2015). Drawing from Deleuze and Guattari (1987) original treatment of ‘agencement’, and the subsequent refnement of E-mail address: cameron.duff@rmit.edu.au. Contents lists available at ScienceDirect Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed https://doi.org/10.1016/j.socscimed.2022.115636 Received 17 October 2022; Received in revised form 8 December 2022; Accepted 23 December 2022