Automated classification of modes of moral reasoning in judicial decisions

Abstract

What modes of moral reasoning do judges employ? We attempt to automatically classify moral reasoning with a linear Support Vector Machine (SVM) trained on applied ethics articles. The model classifies paragraphs of text in holdout data with over 90 percent accuracy. We then apply the classifier to a corpus of circuit court opinions and find a significant increase in consequentialist reasoning over time. We report rankings of relative use of reasoning modes by legal topic, by judge, and by judge law school. Though statistical techniques inherently face significant limitations in this task, we show some of the promise of machine learning for understanding human moral reasoning.