Research programme 2011-2016: A Constructive and Computational Theory of Law

The Leibniz Center for Law is one of the leading research centers in the field of legal knowledge management in the world. The aim of its research programme A constructive and computational theory of law is to develop reusable components and structures, methods of implementation, evaluation metrics and diagnostics, and simulation theories, of law, for purposes of policy making and legislative drafting, effective public administration, and business compliance.

A research programme forms the framework in which a series of studies and experiments are conducted that, taken together, aim at answering an important scientific question. This research programme is aimed at developing a constructive and computational theory of law, integrating theories of law (jurisprudence), political science, economics, cognitive science, and artificial intelligence. The theory aims to support an engineering approach to policy making, legislative drafting, public administration, law enforcement, business process compliance, and legal decision support systems. It suggests reusable design components and structures, methods of implementation, appropriate evaluation metrics and diagnostics, and cognitively plausible noncompliance behaviours. Designs based on the theory are suitable for simulation and automated decision support. The programme should provide organizations deeply invested in the law, for instance public administration, with tools that help them to design effective policy, regulations, administrative procedures, and decision support systems. The research programme connects theory of law with constructive theories in the social sciences that are relevant to the policy making process and evaluation of the effectiveness of law in general.

The research programme is relevant to society: public administration, citizens, and businesses struggle with complex legal requirements, legal pluralism, and multi-reality in business reporting, and these issues undermine legitimacy and predictability of legal consequences, and hence create uncertainty and inequality before the law. Another important practical aspect are the costs of compliance, or administrative burdens, which also affect perceived legitimacy.

In order to make sure that we find a right balance between scientific and societal relevance we aim at conducting our research together with relevant organizations invested deeply in the law. This not only helps us to work on actual and relevant problems, but also increases the likelihood that our scientific results will have practical implications.

The nature of the research allows for contributions from legal theory as well as from artificial intelligence, economics, and (administrative) law.

In our work we distinguish the following four loosely connected knowledge domains in the field of legal knowledge engineering, and always, at least implicitly, distinguish them ontologically in our work:

  1. The organization of the sources of law;
  2. The organization of legal institutions;
  3. Implementation and production of law in social structures; and
  4. The application of law in individual cases.

The first two, and in particular the second, are characteristic of a legal positivist perspective on law. The first knowledge domain addresses the structural organization of the text, and the structural organization of the corpus of texts. It deals with reference, discourse context, reuse of terminology, the use of model sentences to express institutional design patterns, perhaps even the intentional use of legal principles like lex superior, lex specialis, and lex posterior in design of legislation, etc. Knowledge about this domain plays a role in legislative XML, metadata vocabularies for linked open government data, legal text retrieval, self-organizing concept maps, and text parsing approaches to knowledge representation. This is the core domain of law for the information sciences.

The second knowledge domains address the abstract components of the legal system, its institutional structures, and rules. These are posited in the sources of law. This domain is understood well, and it is the core subject of legal theory, and of artificial intelligence & law.

The third knowledge domain is characteristic of legal realism, with its focus on the sociological aspects of law. It covers the pragmatics of enforcement, legal service delivery, and judicial decision-making. It covers the political arena. It also, in our view, covers what we will tentatively call the organization of contextualization, for instance:

  1. Lex specialis as a resolution to the confluence of norms in some context of application (i.e. the discovery of exceptions);
  2. law interpreted as requirements and constraints in design processes (i.e. compliance); and
  3. theory construction about and measurement of the effectiveness of law in contexts of application.

This list covers the major sources of input to policy-making, and has, in our view, potential to explain a major part of the dynamics of the legal system.

The fourth knowledge domain is clearly the one most legal professionals act in most of the time: interpretation of the law in context, from a specific perspective, given certain knowledge, expectations, motives, a concrete problem, a plan.

The problem

A legally proficient player has working knowledge of these domains, and uses that knowledge effectively to attain its goals. The legally proficient player is moreover not just a norm subject, but also a stakeholder in the legal system.

The problem is: how do we create an agent working model of the legally proficient player.