Towards automated identification and analysis of argumentation structures in the decision corpus of the German Federal Constitutional Court

Abstract

Argumentation is an essential task in every scientific discipline. The development of strong and convincing argumentation as well as the analysis of existing argumentation structures is important in the field of humanities, and especially in the field of jurisprudence. Judicial argumentation requires sophisticated intellectual effort and the knowledge of as much potentially relevant background information as possible. Considering that the fulfillment of this task is limited by the natural human information processing capacity, the field of digital humanities investigates how such information-intensive and time-consuming tasks can be supported by computers. Against the background of the ever-growing availability of different corpora of jurisdiction in Germany, a software prototype supporting automated identification, analysis and recommendation of argumentation structures in electronically available corpora of jurisdiction is currently developed in the project ARGUMENTUM. In this article, we present the basic concept for the preparation and processing of the decision corpus of the German Federal Constitutional Court which shall provide the basis for the future ARGUMENTUM prototype.