Category: Data

New concepts for the telephonic era

In a recent study on the Supreme Court’s switch to hearing arguments by telephone, we* showed that the new format gave the Chief the power to alter the liberal-conservative balance of argument, as well as to perpetuate the ongoing problem of female justices being interrupted disproportionately.

In determining how best to measure judicial activity at oral argument in the new forum, we wanted to be able to assess how much opportunity each justice has to speak, not just how many words or for how long they spoke. This is important because different justices use their time at oral argument differently.

For instance, if one justice asks three quick questions and gets three quick answers from an advocate, and another justice makes one long speech and gets one short answer from an advocate, how do we compare these two interactions? It is possible that both interactions may take the same amount of time, in which case the justice making the long speech will measure as more active in traditional measures of judicial activity—having had more words and longer duration in their individual activity levels. But, arguably, that interaction is not as substantive as the justice who speaks fewer words but who has multiple interactions with the advocate.

To better measure judicial activity at oral argument, especially in an era of highly structured time—with the justices speaking in order of seniority—and where that the length of that time is at the whim of the Chief Justice, we came up with a new tool for understanding judicial activity at oral argument.

We coded oral argument transcripts using a precise definition of an “interchange” between a justice-advocate pair. By interchange, we mean all of the time taken or words spoken between a specific justice and a specific advocate, by either of them, until a new justice-advocate interchange begins.

By switching to looking at the justice-advocate interchanges, rather than activity measured just from the point of view of the justice, we are able to capture how much of an opportunity for dialogue—and thus how much opportunity to influence the oral argument—each justice has. The many uses of this concept are illustrated below.

ScotusOA will be changing format

Each week, once the audio is released by the Court, we will post some graphical analysis of individual cases and also some analysis of the term to date.

We can divide each justice’s interchange with each advocate into the amount of time that the justice spends talking versus the amount of time that the justice spends listening to the advocate answers. We have excluded the time the Chief Justice spends in “traffic management”—i.e., calling on a new justice, switching between justices. We have also excluded the advocates’ quiet zone time under the Two Minute Rule and any uninterrupted rebuttal time the advocates are accorded. 

We can illustrate our new approach with a case from last Term, McGirt v. Oklahoma, on the question of whether a state can prosecute an enrolled member of the Creek Tribe for crimes committed within the historical Creek boundaries:

There were four advocates in McGirt because the Muscogee (Creek) Nation, as amicus curiae supporting the petitioner, and the United States, as amicus curiae supporting the respondent, were represented separately from the petitioner and the respondent. Each justice is listed on the left in order of the total duration of their interchanges with each advocate. The time each justice spends speaking is shown in orange, and the time the justice allows the advocate to answer is shown in blue.

Take Justice Sotomayor as an example. She had the longest interchanges with both Riyaz A. Kanji, representing amicus curiae supporting petitioner Muscogee (Creek) Nation, as well as with Edwin S. Kneedler, representing  the United States as amicus curiae supporting respondent. But Justice Sotomayor used her time very differently for each: in the interchange with Kanji, with whom she ultimately sided, she mostly listened, whereas in the interchange with Kneedler, supporting the opposite position, she spent more time talking than listening. This is consistent with what we have found previously in regard to the disagreement gap: justices talk more during the time of the advocate they ultimately rule against.

The next figure provides similar information but in a form that is very helpful in predicting case outcomes. Using McGirt again, it shows the number of words each justice speaks to each side of the case. In McGirt, with two advocates on each side, each graph in the figure represents the total time  each justice spends talking to the advocates on each side.

The figure suggests that the outcome in McGirt should not have come as a surprise: Justice Gorsuch provided the fifth vote along with the liberal justices in favor of petitioner, as was predictable given that he spent the most time talking during the time of the two advocates supporting respondent, and spoke at a much lower rate during the time of the two advocates supporting petitioner.

The figure shows the conservative justices in electric green, and the liberal justices in light blue. The division of the justices into these blocs is important, particularly in light of our finding that, in the new telephonic forum, the Chief Justice is giving advantage to the conservative justices.

Looking at trends that emerge over the Term

We can also use the concept of interchanges to measure not only how each justice fares compared to every other justice in a given case, but also what happens over time. As the Term progresses, we will report the average amount of time each justice takes (or is allowed by the Chief) in all of their individual interchanges with the advocates. If a justice passes, their average will be unaffected. If a justice has two separate conversations with an advocate, then those time periods will be added together and will be reflected in their average.

That concept is illustrated in the next figure, which shows not only the mean duration of justice-advocate interchanges across both sides of the argument, but across cases. We use combined data from Trump v. Mazars and Trump v. Vance as an illustration. These are two related cases that the Court heard argument on by telephone, each of which addressed attempts to subpoena President Trump’s financial records—the former by Congress to subpoena a third party, and the latter by the New York state Attorney General. Once again, we divide by ideology.

Even looking at just these two cases, we can see how active Justice Alito is in the telephonic cases—as we showed in our recent study, Justice Alito received the most time of any justice in the telephonic cases, whereas he normally sits around the middle of the Court in terms of relative level of activity.

It is important to look at both individual cases and activity in aggregate, as variation in activity levels can also be telling. For instance, Justice Alito was permitted 16 minutes to engage in a dialogue with the advocate he disagreed with in Little Sisters, whereas Justice Sotomayor was interrupted by the Chief Justice after only about one quarter of the same time with the advocate—in that case, both Justice Alito and Chief Justice Roberts sided with the Trump administration that they could promulgate rules exempting employers with religious or moral objections from providing contraceptive coverage to their employees. Justice Sotomayor did not agree.

As such, looking at when justices are unusually active is also informative. In the rest of the telephonic cases, as in the in-person cases, Justice Kavanaugh was ordinarily around the middle in terms of his relative activity levels. But in the two Trump cases, which concerned the man who had recently nominated him to the Court and defended him in the face of a credible but unproven accusation of attempted rape, Kavanaugh was unusually active. In the end, Kavanaugh was in the majority in both of the Trump tax returns cases, siding with the President, and he wrote a separate concurring opinion in Trump v. Vance—we have shown elsewhere that justices are more active in cases in which they write ultimately opinions.

Keeping a constant eye on gender relations at the Court

As mentioned, our most recent analysis shows that the gendered differences in the treatment of female versus male justices at Supreme Court oral argument continues. At the end of the 2018 term, our analysis showed there had been some improvement in terms of interruptions—although the problem was still apparent.

But female justices are not the only ones subject to gendered treatment. Female advocates are still underrepresented at the Supreme Court, and when they are given a chance to stand before the justices, they get less speaking time.

With these disparities in mind, we will track the average number of words each female advocate speaks to the Supreme Court, as the 2020 Term progresses. The numbers for the 2019 Term are illustrated in the next figure—if a female advocate appears more than once, her average words per argument are shown (the number of appearances appear in parentheses after each person’s name).

To illustrate just how underrepresented women are at the Court, we compare how each female advocate is treated compared to the average for the three men with the most common names appearing in the term to date.

The figure shows that in 2019, men with just three names—Jeffrey, Paul, and Eric (in green)—appear more times in the term than the entire cadre of women appearing before the Supreme Court that year (in blue). In 2019, Jeffreys, Pauls, and Erics appeared 26 times; all of the women put together appear 20 times. We will update this graph for every case heard, to see if there is improvement.

To see how the combination of female underrepresentation and muting of female voices really adds up over the course of the term, the next figure shows the numbers in aggregate of how many words are spoken by male and female advocates appearing in the 2019 term.

We have provided citation suggestions for each figure, but readers should feel free to contact us if they want more information, such as specific numbers represented in any graph.

By Tonja Jacobi and Matthew Sag

* Oral Argument in the Time of COVID: The Chief Plays Calvinball is an article by Tonja Jacobi, Timothy Johnson, Eve Ringsmuth & Matthew Sag.