🎯 Core Question
Do Justices telegraph their preferences during oral arguments? This study shows that Supreme Court Justices implicitly reveal leanings in their voices during arguments — often before arguments and deliberations have concluded.
🎙️ Audio Evidence: Three Decades, 3,000+ Hours
- Over 3,000 hours of U.S. Supreme Court oral-argument audio spanning 30 years were analyzed.
⚙️ Measuring Arousal and Linking It to Votes
- Emotional content was extracted from the audio recordings.
- Emotional arousal was operationalized using vocal pitch for each Justice during arguments and then used to predict that Justice's eventual vote on the case.
📈 Key Findings
- Vocal-pitch–based arousal measures accurately predict many of the Justices' eventual votes.
- These predictions are both statistically and practically significant.
- The results are robust to inclusion of a range of controls.
- Signals appear early: vocal cues reveal leanings even before arguments and deliberations are finished.
🔎 Why It Matters
The findings suggest that subconscious vocal inflections carry information not captured by legal, political, or textual data alone. This has implications for understanding judicial behavior, the informational content of oral arguments, and how scholars and practitioners interpret in-court signals.
