Becoming Peoples

While the census is sometimes understood to be an objectifying practice that constructs and makes up a population, in this paper I am concerned with how it is necessary to produce census subjects in order to construct population. By drawing on formulations by Latour, Deleuze and Law, I conceive of census taking as a practice performed by heterogeneous socio-technical arrangements of actors -- humans, paper forms, categories, concepts, definitions, topography, geography -- whose mediations, interactions and encounters produce census subjects. It is through the relays and interactions between varying and never fixed technological, natural and cultural actors that census taking is performed. I analyse these arrangements as constituting agencements, which focuses our attention on how agency and action are configured by and contingent upon the socio-technical arrangements that make them up. Agencements assume different socio-technical configurations and thus construct different social realities and populations that cannot be captured in a single account. The argument is advanced through an account of the taking of what was declared the first ‘scientific’ enumeration of ‘Indians’ and ‘Eskimos,’ the Aboriginal inhabitants of the Canadian Far North in 1911. I argue that the agencements were not able to bring forth the subjectivities necessary to construct population in the Far North. Not able to find subjects then, census taking could not produce nor construct a population in the Far North and the practice of census taking ended up creating a record of a census ‘other’ -- an indeterminate multitude that could not identify and could not be identified as part of the population.