Proximity and Neighbourhood: Using Topic Modelling to Read the Development of Law in the High Court of Australia

There is significant interest in the computational analysis of legal text to enhance search capabilities, to navigate the ever-growing corpus of electronic legal texts, and to aid in the application of artificial intelligence approaches to legal work and analysis. However, the feasibility of utilising this technique to trace the emergence and evolution of particular legal doctrines or substantive areas of law over time has not been achieved so far, despite the centrality of this task to legal practice and the use of legal text. Here, we report on the application of computational methods, including topic modelling and visualisation techniques, to measure the emergence and development of particular doctrines or areas of substantive law. We do so using two case studies, firstly the Mabo litigation regarding native title, and the rise and fall of 'proximity' as a test of foreseeability and the duty of care in tort.