More-than-rational modelling : the pragmatics and politics of social simulation in fisheries management

Social simulation methods such as agent-based modelling are making inroads into the tool sets of economists and environmental scientists. Responding to critiques of mainstream economic models, agent-based modellers promise increased behavioural realism by modelling the more-than-rational aspects of economic action in socio-ecological systems. This thesis examines agent-based modelling in fisheries management by ethnographically following the efforts of an international scientific consortium to develop an agent-based fisheries model on to the sites of its intended use at the West Coast of the United States. In a dialogue with pragmatic and performative accounts of modelling in science and technology studies, economic sociology, and geography, this project offers an empirically grounded contribution to debates about the politics of knowledge in environmental science and the performativity of economics. The thesis elaborates a relational and materialist account of social simulation and documents how the practices of agent-based modellers are entwined with the politics of fisheries management. The thesis finds that scientific aspects of the agent-based model like agent algorithms and coupling decisions were co-produced with its political context of model use; that efforts to reform the array of calculative practices enacting fisheries economies contend with infrastructural relations between science and policy; and that computational policy optimisation rests on assumptions about policy process that are at odds with its political dimension. As a result, the thesis argues that more-than-rational modelling unsettles rationalist notions of science, economics, and policy, disclosing instead the socio-materiality of scientific modelling, the normative dimension of economic calculation, and the irreducibly antagonistic aspects of politics at work in a scientific effort to reform economic modelling for fisheries management. The thesis concludes by suggesting that recovering the normativity of rational calculation and the politics of knowledge in the analytical framework of economic performativity contributes to understanding the production of economic difference in the context of geographical debates about marketised environmental governance.