Between the 2004 and 2019 Supreme Court Terms, Justice Thomas spoke in five oral arguments for a total of fewer than 200 words. In that window of time he chose not to speak in over 1,100 arguments.  In that same timeframe, Justice Breyer spoke over 320,000 words at oral argument.  During those same years, Justice Thomas authored a combined 417 majority and separate opinions compared with Justice Breyer’s 319. This comparison highlights an underdiscussed element of Supreme Court behavior – that of the relationship between oral argument speech and opinion engagement.

Several camps emerged in the 1990’s and early 2000’s looking to understand the justices’ goals in oral arguments. Those like Epstein and Knight on one hand and Segal and Spaeth on the other posited preference based arguments that focused on how justices are goal-oriented in both oral arguments and with their opinion writing, with policy preferences driving their decisions.

Tim Johnson wrote several pieces that focused on multiple dimensions of oral arguments ranging from how they can be used to predict case outcomes to the justices’ other goals. In the vein of information gain, Johnson’s work describes how the justices seek comprehensive information during oral arguments to help them make informed decisions that align with their goals while adhering to institutional norms. Briefs provide a substantial amount of information, yet they also reflect the argument positions of the litigants and so they are often biased in that regard, which underscores the importance of justices seeking additional perspectives to inform their decisions.

According to Johnson and others, oral arguments are a critical venue for justices to gather information that is independent of the potentially biased materials found in briefs. Justices can ask questions during these proceedings to clarify issues and obtain insights that might not be included in the written submissions. Johnson’s articles demonstrate that a portion of the issues in majority opinions are raised uniquely during oral arguments. This finding supports the idea that oral arguments introduce new issues that may not have been considered previously.

Justices may also use oral arguments to gain insight into the positions of their colleagues on the Court and may even attempt to convince them to join along with their preferred position on the merits.

The Focus

This article looks at three relationships between oral arguments and written decisions: first between the extent of a justice’s engagement in oral argument and the decision to author an opinion at all, second between the extent of oral argument engagement and the extent of written engagement in the same cases, and third between the language justices use at oral argument and what they write in their decisions.

The first two measures – between the extent of oral argument speech and the choice with whether to write at all and if so how much – are performed by looking at the correlations in these aspects of engagement. The third – the language similarity aspect uses cosine similarity

If justices who speak more tend to write more comprehensive opinions or write opinions at all, it may suggest a direct connection between the justices’ oral discourse during proceedings and their written analyses. Examining the similarity between the language used in oral arguments and written opinions can shed light on how justices frame their thoughts and arguments. If their language is consistent across both formats, it may indicate a deliberate effort to maintain coherence between what they say and do.

To construct this measure of similarity the algorithm looks at each justice’s words in each oral argument transcript and corresponding opinion and counts how many times each word appears after removing “stop words” which hold little to no information value.  Then it counts how many times each word appears in both documents. Cosine similarity looks at these two lists and checks how similar they are.

The data used are oral arguments transcripts and opinions from the 2004 through the 2022 Supreme Court terms. The oral argument transcripts were supplied by Jake Truscott’s compilation via his R based tool SCOTUSText.  The justices of focus are the nine on the current Court and the most recent retiree, Justice Breyer.  These data are provided on the justice level, thereby breaking these patterns down to focus on individual behavior along with that of the justices generally.

Understanding these relationships may reveal strategic behavior in how justices use oral arguments to influence their colleagues or shape public perception. A justice who uses more persuasive language during arguments may also adopt similar rhetoric in their opinions, suggesting a conscious strategy to convey specific innuendo.

Findings

The first analysis looks at the likelihood of a justice authoring an opinion at all, based the volume of their speech at oral argument. The plot is a combination of a bar and box plot with confidence intervals in black.

Justice Kavanaugh exhibited the highest correlation with .294 (with a p-value of .000112) indicating a statistically significant relationship. This means that as Kavanaugh speaks more, the likelihood of authoring a decision in the same case increases markedly.  Justices Alito and Kagan both showed significant correlations of .226. This indicates that their speaking practices are also closely tied to the eventual written decisions.  Conversely, Justice Roberts presented a negligible correlation of .00519 indicating no significant relationship between his speech and the presence of a written opinion.

The overall correlation across all justices between the volume of speech and the presence of written decisions is low (.0431). This finding underscores the notion that, on average, the amount of speech produced does not significantly predict the presence of written decisions among the justices collectively. This also implies that individual justices’ relationships on this front vary widely in terms of when opinions are ultimately rendered.

Moving forward along the same train of thought, the next figure looks at the relationship between the amount of oral argument speech and the length of opinions when authored.

Based on arguments where each justice engaged at all, Justice Jackson had the highest average number of words spoken, followed by Justices Kagan and Breyer. Justice Thomas had the lowest number of average words in oral argument with 308.6.  Looking at decision length, Justice Breyer had the highest average length with 5623.59 words.  Justice Barrett with 2331.58 words per decision had the least words on average.

The overall correlation between words spoken and words written across justices is .2316 indicating a weak positive correlation. This suggests that, in general, justices who tend to speak more during oral arguments also tend to write longer decisions, but the relationship is not very strong. 

Justices Jackson and Kagan have high average oral argument word counts and relatively high average decision lengths, indicating a robust engagement in both oral arguments and written opinions. Thomas who notably has lower engagement in oral arguments, produces a differentially high word count in opinions. While the correlation indicates a relationship between speaking amount and decision length, it also highlights that the relationship is not strong enough to conclude that higher speaking amounts directly lead to longer written opinions. Factors such as the nature of the cases, individual styles of writing, and the complexities involved in each decision likely contribute to the observed outcomes.

The last figure shows the relationship between the similarity in language used in oral argument and written decisions.

The plot shows the median similarity per justice with a red dot with other datapoints around them showing different levels of similarity. The bulk of the observations by justice are in the thicker hubs in a similar manner to a box plot, with the outliers towards the edges.

Justice Jackson had the highest average similarity .782, suggesting that her language is most similar between oral argument and decision text. Justices Alito and Sotomayor also have relatively high average similarities (.748 and .720, respectively). Justice Thomas has the lowest average similarity (.599), indicating a lower alignment between his word choice in oral argument and decision text compared to other justices. 

Concluding Thoughts

The general insights from these results highlight the variability in how justices engage with both oral arguments and written opinions, reflecting different approaches. First, there appears to be a notable distinction between justices who actively engage in oral arguments and those who are more reserved in that setting, with some justices demonstrating a clear connection between how much they speak and their likelihood to author opinions. This suggests that for some justices, speaking in oral arguments may serve as an important aspect of their decision-making process.

Second, the weak but present correlation between the amount spoken in oral arguments and the length of written opinions suggests that while there is some connection between oral and written engagement, it is not a strong or consistent one.

Lastly, the variation in language similarity between oral arguments and written opinions underscores differences in how justices approach communication in these two formats. Justices who exhibit higher language similarity may strive for consistency in how they present their ideas in both settings, possibly reflecting a strategic effort to ensure coherence. Conversely, justices with lower similarity might adapt their language for different contexts—perhaps viewing oral arguments as more dynamic and fluid, while treating written opinions as more formal.

This variability highlights that there is no one-size-fits-all pattern in how justices move from oral argument to written decision, with some showing strong connections between the two and others exhibiting more independence in their methods of engagement.

Knowledge of how justices engage in oral arguments compared to their written opinions can inform legal practitioners on how to better present cases. Understanding which justices favor oral discourse may guide attorneys in tailoring their arguments.

Researching these relationships can contribute to a more comprehensive understanding of judicial behavior by bridging gaps in existing literature on how oral arguments influence judicial outcomes and enhance the understanding of the Supreme Court’s decision-making as a complex interplay of verbal and written communication.

---

Find Adam on X/Twitter and LinkedIn. He’s also on Threads @dradamfeldman and on Bluesky Social @dradamfeldman.bksy.social