The thesis investigates how social norms are enforced. It consists of two parts. The first part establishes the concept of "realistic constraints for moral theory" based on the "ought implies can principle". Different notions of feasibility lead to different degrees of moral realism. Game theory and computational modelling are the appropriate instruments to determine feasibility constraints for realistic moral theory. They allow for a dynamic perspective on norm enforcement, in contrast to more static approaches. The thesis discusses the use of computational models and game theory from a philosophy-of-science point of view. I conclude that computational models and game theory can inform moral theory if they are understood as sources of realistic constraints. The second part uses two agent-based models to explain the enforcement of social norms. In the first model, agents play one-shot, two-person prisoner's dilemmas. Before the game, agents have a better than random chance to predict which strategy the others are going to play. Cooperative agents do well if they are able to pool their information on the strategies of others and exclude defectors. The second model analyses repeated multi-person prisoner's dilemmas with anonymous contributions. The players are situated in a social space represented by a graph. Agents can influence with whom they are going to play in future rounds by severing ties. Cooperative agents do well because they are able to change the interaction network structure. I conclude by connecting the findings with debates in moral philosophy and evolutionary theory. The results obtained have implications not only for the emergence of cooperation and social norms literature, but also for theories of altruism, research on social network formation, and recent inquiries by behavioural economists into the effects of group identity.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:645732 |
| Date | January 2008 |
| Creators | Spiekermann, Kai |
| Publisher | London School of Economics and Political Science (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://etheses.lse.ac.uk/2319/ |
Page generated in 0.0021 seconds