Text Reuse in Judicial Decision-Making: Cutting Corners or Leveraging Efficiency Gains?

How can busy judges decide cases more quickly? One strategy that judges can use is to reuse text from previous decisions. Text reuse could reflect over-worked judges and clerks cutting corners, sacrificing the quality of their legal reasoning for efficiency, but it could also reflect experienced judges appropriately using boilerplate text to make the process of decision-writing more efficient. In this paper, we examine patterns of text reuse at the Court of Justice of the European Union (CJEU). Using a new text corpus that includes the universe of CJEU decisions (1952-2021), we estimate sentence embeddings and calculate the similarity between sentences to identify verbatim and non-verbatim text reuse. We use quantitative text analysis techniques to assess whether reused text tends to be boilerplate text or text containing legal reasoning. We also compare patterns of text reuse to the Court’s case law citation network to determine when Judge-Rapporteurs are paraphrasing legal reasoning when citing precedent. We examine whether there is variation across Judge-Rapporteurs in patterns of text reuse to assess the degree to which text reuse is a practice used by particular judges (and their référendaires) or by the Court as a whole.