About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

LOG IN
LOG IN


17 J. Legal Analysis 2 (2025)

handle is hein.journals/jlegan17 and id is 1 raw text is: Journal of Legal Analysis, 2025, 17, 2-13 https://doi.org/10.1093/j~la/laafOOl Advance access publication 27 March 2025 Article Lifting the American Supreme Court Veil: Identifying Authorship in Unsigned Opinions 0 Ronen Avraham1', Rami Nasser2 , Itamar Kohn', Tamar Kricheli-Katz' , and Roded Sharan' 'School of Law, The University of Texas at Austin, Austin, TX, United States 0 3 2School of Computer Science and AI, Tel Aviv University, Tel Aviv, Israel 'Buchman Faculty of Law, Tel Aviv University, Tel Aviv, Israel *Corresponding author. Ronen Avraham, Buchman Faculty of Law, Tel Aviv University, Israel. E-mail: Ronen3112@gmail.com 0 Abstract 3 The Supreme Court of the United States (SCOTUS) issues 10-15 % of its opinions unsigned, con- cealing authorship. Traditionally, unveiling authors required the posthumous release of Justices' per- sonal papers. We trained our AI algorithm to achieve real-time authorship probabilistic identification, 6 encompassing 17 Justices and 4,069 opinions from 1994 to 2024. Our algorithm identified the likely 2 authors of the March 2024 Trump v. Anderson case, which enabled Donald Trump to run for office. Moreover, our algorithm unveiled the likely authorship in significant unsigned COVID-19 era cases, estimated with high probability individual parts of the joint dissent in the Obamacare Case (2012), 5_ and discerned the likely authors of the landmark cases of Bush v. Gore (2000). Applications range from legal research to decoding SCOTUS internal dynamics. Compared to prior methods, our study demonstrates a substantially higher accuracy rate of 91 per cent over a much longer period of time, offering timely insights into the nuances of SCOTUS decision-making. To facilitate further research, 6 we provide a public web server at https://raminass.github.io/SCOTUS_AI/. 0 1. Introduction Approximately 10-15 per cent of the official opinions of the Supreme Court of the United States (SCOTUS) remain unsigned, leaving the authorship unknown. Traditionally, the main method for determining the authorship of these opinions involved waiting for the release of personal papers of Justices, typically several years posthumously. This delay is exemplified in the recent unveiling of the personal papers of Justice John Paul Stevens, who retired in 2010 and passed away in 2019. These papers shed light on the deliberations surrounding the 2000 decision in Bush v. Gore,' a significant case where the majority opinion granting George W. Bush the presidency over Al Gore, was unsigned. However, such documents face extended public exposure delays more than 22 years in the case of Justice John Paul 0 Stevens. Furthermore, they are considered private property, and Justices are not obligated to share or publicize them after retiring. In fact, only about 1 in 3 Justices have chosen to donate their papers to the Library of Congress (Gresko 2023). Few studies have explored SCOTUS opinion authorship through classical statistical and machine learning methods. Rosenthal and Yoon (2011a, 2011b) utilized function words with a linear or naive Bayes classifier to distinguish between only 2 justices. Chandler, Muenster, and Lichtblau (2023) focused Bush v. Gore, 531 U.S. 98 (2000). Ronen Avraham, Buchman Faculty of Law, Tel Aviv University, Israel. This work was sup- ported by a grant from the Tel Aviv University Center for AI and Data Science (TAD). © The Author(s) 2025. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site-for further information please contact journals.permissions@oup.com.