Donald H. Berman

Representing teleological structure in case-based legal reasoning: the missing link

We argue that robust case-based models of legal knowledge that represent the way in which practicing professionals use legal decisions must contain a deeper domain model that represents the purposes behind the rules articulated in the cases. We propose a model for representing the teleological components of legal decisions, and we suggest a method for utilizing this representation in a HYPO-like framework for case-based legal argument.

Understanding precedents in a temporal context of evolving legal doctrine

In evaluating the precedential strength of a prior case, skilled attorneys take account of how the holdings of the case have been treated in subsequent decisions. This paper describes the process by which a formerly strong precedent may be weakened by over time, identifying five reasoning patterns by which attorneys may predict that a most-onpoint case is likely to be explicitly or implicitly overruled. We consider the requirements for implementing this form of analysis in a case-based legal reasoning system, proposing an extension to earlier schemes for case representation, and outlining an evidential reasoning algorithm to compute the degree to which the holdings of a prior case have been weakened.

Incorporating procedural context into a model of case-based legal reasoning

In this paper we analyze the procedural considerations that affect the use of legal casesas precedents and propose a model of procedural knowledge that can be combined with substantive legal reasoning models to produce a more robust theory of case-based legal reasoning in common law jurisdictions. Our model addresses one component of procedural knowledge the distinction between questions of fact and questions of law. We categorize 32 different procedural scenarios into 10 basic types of legal results. We then propose rules for determining the precedential value of these result types. Finally we suggest a method for incorporating procedural distinctions into case-based reasoning systems.

Developer's choice in the legal domain: the Sisyphean journey with DBR or down hill with rules (a working paper for the case-rules panel at the third international conference of artificial intelligence and law)

The technology of case-based systems directly addresses problems found in ndebased systems: First, . .[i]t is easier to articulate, examine and evaluate case-sthan rules. Second... [a] case-based system can remember its own performance and modifi its behavior to avoid repeating prior mistakes. Third... ~]y reasoning from analogy with paat cases, a case-baaed system should be able to construct solutions to novel problems [Slade 1991].

Cutting legal loops

riverrun, past EVE and Adams, from swerve of shore to bend of bay, brings us by a commodious vicus of recirculation back to Howth Castle and Enviorns. ***** A way a lone a last a loved a long the Joyce, FINNEGANS WAKE 1, 628 (1955) Recursion is the act of defining an object or solving a problem in terms of itself. A careless recursion can lead to an infinite regress. We avoid the bottomless circularity inherent in this tactic by demanding that the recursion be stated in terms of some “simpler” object, and by providing the definition or solution of some trivial base case. Properly used, recursion is a powerful problem solving technique, both in artificial domains like mathematics and computer programming, and in real life. Friedman & Felleisen, THE LITTLE LISPER ix (1986)