Ex Uno Plura: Identifying Issues in CJEU Jurisprudence

Research of judges and courts traditionally study court decisions, treating each decision as a unit of observation. However, court decisions often address multiple distinct and more or less unrelated issues. Studying judicial behavior on a decision-level therefore loses potentially important details and, more problematically, risks drawing false conclusions from the data. This contribution presents and validates a method for automatically splitting decisions by issues using a supervised machine learning classifier.