Norms and rules A framework for the interpretation of regulations and social behavior

Norms and rules govern behavior. They evolve from customs when communities start to enforce a custom, rather than leaving individuals free to choose whether to follow it, or not. In this thesis, we present a method for extracting norms and rules from normative text by interpretation. The goal of the method is to create a shared opinion on the meaning of laws and regulations between lawyers and non-lawyers as a foundation for instructions for people and specifications for computers.
In part one, we give an overview of literature on norms and rules in the social sciences, philosophy, legal science and software engineering. We argue that norms and rules are synonyms: both require relations between agents, conditions for valid actions, and results. The main theoretical basis of this thesis is the work of Salmond, Hohfeld and Kocourek on fundamental legal relations, and the work of Searle on institutional facts.
In part two, we construct an action-based framework consisting of frames for acts, facts and duties, also known as FLINT. In this framework the act is of primary importance: it can be used as an instruction for an agent to gain a benefit, or as a power to punish non-compliant behavior.
In part three, we present two case studies in the domain of immigration law to show how acts can be extracted from laws and regulations.
Currently we are building a case-base of applications of the FLINT method. To enable large-scale application, we are developing an open source tool.