Legal corpus studies, proportionality, and the promise of LLMs

How do courts justify their decisions? What role do legal arguments play in court decisions? With corpus-based legal studies we want to bridge the gap between NLP, empirical methods and the research interests of legal studies. We will give a brief overview of our previous work on proportionality, in particular the data we have annotated, our empirical research, and our attempts to classify court decisions. In addition, we discuss our ongoing work on the use of LLMs to identify proportionality and our attempt to combine them with a domain-specific German public law knowledge graph. Our broader aim is a new paradigm of empirical research on courts, focusing on legal reasoning and argumentation techniques.