Infrastructure is a means to control and organize life through far-reaching technological systems.

For some time now, scholars in science and technology studies and geography have analyzed how infrastructures mediate exchange over distance, bringing different people, objects, and spaces into interaction and forming the base on which to operate modern economic and social systems (Graham & Marvin 1996, 2001; Lefebvre 1991). Graham & Marvin (1996) have written a series of influential works examining how new systems of telecommunications are reconfiguring urban space and how infrastructures bundle together water, energy, people, and streets into a series of networked infrastructures that define modern life (2001). The emphasis here is on the formation of what Hughes (1987, 1993) terms large-scale technical systems, the massive infrastructural networks that have come to organize everyday life. Hughes (1987, 1993) and those coming in his wake (Bijker 1997; Bijker et al. 1987; Bowker & Star 2000; Edwards 1997, 2003; Edwards et al. 2009; Hecht 2011; Jackson et al. 2007; Star & Ruhleder 1996; Yates 1993) argue that infrastructures typically begin as a series of small, independent technologies with widely varying technical standards. They become infrastructures when either one technological system comes to dominate over others or when independent systems converge into a network. For Hughes, as a systems thinker, the ontology of a technology is composed of elements internal to the object itself but is formed as part of a system. The invention of the Edison light bulb, for instance, simultaneously involved the invention of the Edison Jumbo Generator and the Edison Main and Feeder. It innovated financial instruments and management structures necessary to accommodate a rapidly expanding corporation. The light bulb in this conception does not develop from an unfolding of elements internal to the technology itself—an assemblage of glass, filament, and vacuum—as some theories of technics would have it. Rather, as an infrastructure it is an amalgam of technical, administrative, and financial techniques. Two major insights that emerge from this work are of central importance to anthropology. Placing the system at the center of analysis decenters a focus on technology and offers a more synthetic perspective, bringing into our conception of machines all sorts of nontechnological elements. For Hughes, a holding company or an accounting practice is as much a technical invention as is a dynamo or a telephone, and all are essential to producing an infrastructural system. Second, the focus is on system building. A technical system originates in one place, growing in response to particular ecological, legal, political, and industrial techniques native to that area. But as it grows into a networked infrastructure, it must move to other places with differing conditions, technological standards, and legal regulations, elaborating techniques of adaptation and translation. This conception places focus on practices of routinization and extension, requiring an account of translation (which can be technical, but also managerial and financial) as a process inherent to system building. It is no surprise that the scholars coming after Hughes (Bowker & Star 2000, Edwards et al. 2009) drew heavily on actor-network theory and its emphasis on tracing associations between heterogeneous networks (Latour 1996, 2007) and on the necessity of accounting for how translation occurs (Callon 1986, 1998; Latour 1993). This work has been extremely productive for recent anthropological research (Appel 2012a,b; Anand 2011, 2012; Carse 2012; Collier 2011; Collier & Lakoff 2008; Von Schnitzler 2008), as one can see in two excellent recent studies on water supply in Mumbai (Anand 2011, 2012) and Soweto (Von Schnitzler 2008). Anand’s ethnography draws together engineers, political fixers, slum dwellers, politicians, activists, and bureaucrats into a single system through the technical operation of water supply (see also Gandy 2008, Graham et al. 2013). In Mumbai, the density of the population causes water supply to be scarce and periodic. Municipal engineers account for this limitation in technical terms (locating slums on a hilltop, the pressure required to move water great distances), but for slum residents it is an issue of political mobilization. Slums without water engage dadas, powerful patrons who use their connections to pressure elected officials to provide infrastructural connection. In return, the dada rewards those representatives by delivering electoral support. In Anand’s analysis, essentially two infrastructural systems interact: water delivery, with its systems of pipes, engineers, and bureaucracy that make up the technical end of water provision (the aspect of infrastructural supply that constitutes the entire large-scale system for Hughes and others); and the social networks, forms of patron-clientship, and “phatic labor” (Elyachar 2010) that are as important to water delivery as are pumps and pipes. Infrastructure in this sense is a kind of mentality and way of living in the world (Hansen & Verkaaik 2009, Simmel 1972, Simone 2001), and Anand (2011) brings together these two differing conceptions of infrastructure not, in the final instance, to analyze water supply but to reveal the production of what he terms “hydraulic citizenship, a form of belonging to the city enabled by social and material claims made to the city’s water infrastructure” (p. 545). Von Schnitzler (2008), like Anand, sees water supply as revealing a larger logic of citizenship in the postapartheid state in South Africa. But her approach differs by focusing analysis on a specific technology, the water meter, which was introduced as a technical device to monitor water usage and end the problem of wastage in townships. Von Schnitzler traces the technical history of the water meter back to its invention in Britain but then examines how its introduction to South Africa was intended not just to regulate water use but to produce a new sort of citizen. This citizen was one who was responsible, self-monitored his or her own actions, and engaged in practices of calculation. Taking a more biopolitical turn, Von Schnitzler argues that meters do not just regulate water provision; they reveal a strategy of government and, ultimately, produce an ethics. “[C]ouched in a moral-pedagogical language” (Von Schnitzler 2008, p. 906) city officials aimed to reform the culture of township residents (who wasted water) by producing a “calculating subject,” one who switches the tap off when brushing teeth or reuses bath water when flushing the toilet (see also Fennell 2011). The meter, Von Schnitzler argues, delivered moral behavior as well as water. Both Von Schnitzler and Anand demonstrate how the operation of technologies, ostensibly a neutral practice, becomes the grounds around which forms of citizenship are contested (Barry 2001; Joyce 2003; Mitchell 2002, 2011; Otter 2008) and where technological systems are entangled with other religious and political domains. Collier (2011) adds a twist to this literature by organizing his analysis of infrastructures entirely around the question of the biopolitical (Foucault 2010, 2011). Infrastructures, for Collier, are a mixture of political rationality, administrative techniques, and material systems, and his interest is not in infrastructure per se but in what it tells us about practices of government. Soviet electricity provision, through this lens, is analyzed for how it reveals a system of total planning in a command economy rather than for what it tells us about the effects of electricity on users in Russia. Unlike in the West, electricity supply in Soviet Russia was not regulated by user demand but was an allocation to a particular area decided in advance by technocrats as part of a total system of electricity planning. The post-Soviet transition required a dismantling of this idea of government and a reconceptualizing of society as organized around an individual consumer and his or her demands rather than around a collective assigned a fixed amount of units.