Governance after planning.
An ethnography of how a postcolonial state continues to govern when the epistemological foundations of planning come undone — and what new political forms emerge in their place.
For most of the twentieth century, the developmental authority of postcolonial states rested on a particular epistemic claim — that the state could know, through statistics, surveys, and resource assessments, the social and economic processes it sought to govern. By the early twenty-first century, that claim no longer holds steady. National statistical infrastructures have lost authority and coverage; new digital infrastructures have not yet consolidated reliability.
The dissertation pursues what it means to govern through this interregnum. Drawing on three years of fieldwork and archival research reaching back to the postcolonial planning institutions of the 1950s, it argues that contemporary modes of governing do not retreat from technical infrastructure when its epistemic ground gives way. Instead, they redeploy it for other purposes — to generate political rhythm, competitive community, and the spectacular performance of governance improvement.
The state is not the only actor in this story. The work also tracks the cast that has come to populate the conditions of national governance: transnational funding agencies, philanthrocapitalist foundations, multinational technology corporations, and financial experiments such as development impact bonds — whose tools, money, instruments, and expertise reshape what the state is able to do, and what it is asked to do, by those it governs.
Three threadsThe contours of the project
I. Regional inequality
The postcolonial state has long been concerned with the geography of disadvantage — the historical distinction between regions deemed “backward” and those deemed advanced. That concern is being reorganised through new vocabularies of aspiration, competition, and ranked performance. The work traces how the political and administrative labour once done by redistributive planning is now being done by other means.
II. Digital infrastructures
Dashboards, real-time data portals, ranking systems, machine learning algorithms, and automated decision systems have become central to how the contemporary state addresses developmental questions. These are not merely tools of measurement but a medium through which the state constitutes itself as an object of public attention. The work reads the technical apparatus ethnographically — attending to how it is built, maintained, and inhabited by frontline workers, bureaucrats, consultants, and the multinational technology corporations whose platforms increasingly underwrite governance — and to the gap between what such systems are supposed to know and what they actually do.
III. Federalism
The relations between centre, states, and districts are reshaped each time a new mode of governing development is introduced. The work tracks how programmatic reform recomposes what it means for sub-national units to “implement” or to “perform” — and how the centre, in turn, performs its own reach across an uneven terrain.
Field sitesWhere the research happened
Ethnography in bureaucratic settings; archival work in national and state collections; participant-observation alongside policy consultants and frontline workers; document analysis across cabinet resolutions, programme guidelines, and digital portals. The project moves between offices and districts, between papers tabled in parliament and forms filled at the block level.
Two strands precede the dissertation. A Master’s thesis at the University of Chicago (2014, advised by John D. Kelly) on the Criminal Tribes Act and police practice in British India, 1871–1927, argued against scholarship that read the Act as a uniform homogenising apparatus, drawing on police manuals to show classification at the level of practice as variable, multidimensional, and uneven. The thesis pre-empts contemporary debates on AI-driven facial recognition, carcerality, and predictive policing — tracing how nineteenth-century apparatuses already operated through racialised, hereditary classification of populations marked for surveillance.
A stint on the City of Chicago’s Smart Cities team (2014–15) contributed to a predictive model for food inspections used by the city’s Department of Public Health. The dissertation inherits both threads — the suspicion of homogenising classifications and a working knowledge of what predictive systems actually do inside a state apparatus.