Accounting quantification and the limits of remote government: the case of accounting in the colonial prisons of Guyana (1859-1873).

Summary This thesis focuses on the possibilities of exercising power at a distance through accounting quantification, from a Foucauldian perspective. While the existing literature analyzes the conditions of effectiveness of action at a distance via accounting (Robson 1992), the objective of this work is, on the contrary, to identify factors that limit this effectiveness, by mobilizing the concept of territorialization (Mennicken and Miller 2012). This analysis was conducted using a historical approach, based on the case of accounting for forced labor in the penal colony of Guyana (1859-1873), placing our work within the current of New Accounting History. We have developed an interpretative methodological approach, ethno-accounting, which has allowed us to map the appearance of a certain number of choices and effects, which were confronted by the actors of the prisons involved in the process of accounting quantification imposed from the metropolis. We then show how the combination of a certain number of these choices and effects can generate a certain number of dynamics, potentially having unexpected consequences, which can ultimately prove to be contrary to the initial governmental program that the accounting was supposed to operationalize. This work, in highlighting the problems generated by accounting quantification in prison institutions, is part of broader historical and current questions. Thus, the methodology of ethno-accounting has allowed us to initiate new research perspectives on the current phenomenon of prison privatization linked to New Public Management. Moreover, our research object, prison accounting, has led us to revisit the controversies initiated by Foucault's Surveiller et punir (1975) among social historians of the prison, as well as the issues related to Foucauldian-inspired historical approaches in Organization Studies.