Arguing by Analogy in Law: A Case-Based Model

Abstract

In this chapter we focus on arguing by analogy in law, the way in which attorneys argue in favor of deciding a problem situation by analogizing it to precedent cases. We describe a 3-ply, turn-taking structure of analogical legal arguments in which analogous precedents are cited in points and responded to by distinguishing and citing counter-examples. After working through a brief example, we examine the traditional theoretical account of legal analogical reasoning and two criticisms of the traditional account, that it does not explain: (1) what similarities and differences are important, or (2) how competing analogies are resolved. We present a more complete account of arguing by analogy in law and show how the model is implemented in HYPO, a computer program that makes case-based, analogical arguments in the domain of trade secret law. We describe how HYPO uses “dimensions” and “claimlattice” mechanisms to perform indexing and dynamic relevancy assessment of precedent cases, compares and contrasts cases to come up with the best precedents pro and con a decision and makes a skeletal argument with points and responses that pose and distinguish analogous precedents. We show how the HYPO approach addresses the criticisms of the traditional model and compare it to the approaches of other AI research on analogical reasoning.