Abstract Dissenting opinions are common in the US Supreme Court even though they take time and effort, risk infuriating colleagues, and have no precedential value. In spite of these drawbacks, dissents can potentially contribute to future legal development. We theorize that dissenting justices who use more memorable language are more successful in achieving such long-term impact. To test this theory, we amass an original dataset of citations to dissenting opinions extracted directly from majority opinion text. We further leverage these texts to build an algorithm that quantifies the distinctiveness of dissenting language within a dynamic context. Our results indicate that dissents using more negative emotion and more distinctive words are cited more in future majority opinions. These results contribute to our understanding of how language can influence long-term policy development.
Stored only when both papers are in the corpus — 33 citations exist upstream.