Skip to content

Latest commit

 

History

History
116 lines (66 loc) · 16.3 KB

index.md

File metadata and controls

116 lines (66 loc) · 16.3 KB
layout
default

U.S. Supreme Court Arguments: Auditing and Oddities

Below are the results of an on-going audit that correlates three sources of U.S. Supreme Court data:

The goal of this initial audit was to identify issues and work out kinks in the auditing process, and ultimately generate comprehensive lists of relevant data, including:

  • Audio present in the NARA collection but missing from the Oyez collection (and vice versa)
  • Other (non-argument) U.S. Supreme Court audio present in the NARA collection
  • Names of advocates who argued in all cases, extracted from the Journals and cross-referenced with Oyez

Ultimately, all terms from 1889 (the earliest year for which Journals are available digitally) through the present need to be audited.

Oyez omissions discovered in the initial audit:

  • Nilva v. United States (No. 37) was argued on November 8 and 13, 1956. Oyez appears to have combined the audio recordings and inadvertently dropped the November 8 date in the process.

  • Stanton v. United States (No. 546) was argued on March 24, 1960, following a related case, Commissioner of Internal Revenue v. Duberstein (No. 376) that was argued on March 23. However, the argument for Stanton appears to be missing from Oyez; only the argument for Commissioner v. Duberstein is available. The cases were decided together and reported in 363 U.S. 278.

  • Presser v. United States (No. 278) was argued in the 1961 term, affirmed by an equally divided Court, subsequently restored to the Court's 1962 calendar (as No. 25), and reargued and decided in the 1962 term. However, Oyez appears to provide only the 1961 argument.

  • Creek Nation v. United States (No. 124) was argued in the 1961 term on April 24, 1962, but it does not appear in Oyez. It was affirmed by an equally divided Court on June 4, 1962, so the lack of an opinion may have contributed to it being overlooked.

  • Chicago and North Western Railway Company v. The Atchison, Topeka and Santa Fe Railway Company (No. 8) and United States v. The Atchison, Topeka & Santa Fe Railway Company (No. 23) were argued together in the 1966 term on April 19, 1967, but all that appears in Oyez is an incomplete page which links to the correct opinion. [NOTE: For those interested, Quantity of Books v. Kansas does exist in Oyez as well.]

Examples of anomalies uncovered in Oyez during the initial audit:

  • Oyez filed a large number of reargued cases in the terms in which they were first argued, but with docket numbers from the terms in which they were reargued. For example, International Association of Machinists v. Street was argued in the 1959 term as No. 258, and then redocketed and reargued in the 1960 term as No. 4, but Oyez filed it under the older term (1959) with the newer docket number (4).

  • That filing practice, in turn, resulted in terms containing multiple cases with the same docket number. For example, Kennedy v. Mendoza-Martinez was argued in the 1961 term, reargued in the 1962 term, and filed under 1961 with its 1962 docket number (No. 2). Unfortunately, in the 1961 term, Metlakatla Indians v. Egan is also No. 2. In that case, Oyez had to resolve the filing conflict with a special suffix in the URL ("2_0").

  • Oyez did not consistently follow that filing practice. For example, Bartkus v. Illinois was originally argued in the 1957 as No. 39, and then redocketed and reargued in the 1958 term as No. 1. In this case, Oyez filed both arguments under the newer term and newer docket number.

  • Garrison v. Louisiana is a redocketed and reargued case in which Oyez accidentally swapped the audio for the arguments. The case was originally docketed as No. 400 and redocketed as No. 4, but Oyez's original argument transcript begins with "Number 4, Garrison -- Jim Garrison, Appellant, versus Louisiana" and its reargument transcript begins with "Number 400, Jim Garrison, Appellant, versus Louisiana." This type of mistake isn't something my audit looks for (my initial focus is missing or misfiled cases); I just happened to notice while looking at those particular transcripts.

  • Hamm v. City of Rock Hill (No. 2) combines two cases: the Hamm case (No. 2) and Lupper v. Arkansas (No. 5). They were argued separately and decided together, which is not uncommon, but what is uncommon is that Oyez combined both arguments into a single audio recording. The better (and more typical) approach is the one taken in cases like Miranda v. Arizona (No. 759), where multiple cases with similar facts and issues were argued together, but each case can be listened to separately. And again, this isn't something I'm looking for, it simply caught my eye while tracking down the Lupper case.

  • Immigration and Naturalization Service v. Errico (No. 54) includes the audio for Scott v. Immigration and Naturalization Service (No. 91), which is not obvious, because the second case isn't listed and Chief Justice Warren's introduction of the second case is truncated (the tape recorder wasn't started in time). A larger problem with combined recordings is that it makes argument accounting more difficult. Here, Thurgood Marshall argued in each of the two cases, so he should be credited with two arguments, not one. Oyez has the larger problem that it doesn't credit Marshall with any of these arguments.

  • Oyez's advocate data (ie, the names of the attorneys who argued the cases) has quite a few omissions and errors. In many cases, names were assigned in the Oyez-generated transcripts, but for some reason, those names were not always extracted and presented on the Oyez case pages. The Journals are a much more reliable and consistent source of advocate information, so extracting that information should result in a more accurate and comprehensive set of advocate data, as these audits continue.

With regard to docket numbers, it's worth noting that not even the Supreme Court was perfectly consistent. For example, Reid v. Covert (No. 701) was assigned docket number 701 in the 1955 term, but when it was redocketed for reargument in the 1956 term, the same docket number was used. This required the Supreme Court to add a qualifier to the docket number every time the case was mentioned in the 1956 Journal (e.g., "No. 701, October Term, 1955"). Fortunately, that was a rare occurrence, at least among argued cases, and the Court had a method for dealing with it.

The way Oyez defines a "term" created other problems. In the modern era, a term is generally a 9 to 12-month period starting in October, but in the past, the Court sometimes had multiple "regular terms" in a single year, as well as the occasional "special term" (ie, a term sandwiched between regular terms, often in the summer). Both regular terms and special terms are always identified by month and year (e.g., "February Term 1809", "October Term 1955", "August Special Term 1958", etc), so it would have been far better if Oyez (and others) had used unambiguous term identifiers (e.g., "YYYY-MM") instead of merely "YYYY".

Take the case of Aaron v. Cooper (No. 1 Misc) and its immediate successor, Cooper v. Aaron (No. 1), both of which Oyez filed in the 1957 term. Those two cases were actually argued and decided between regular terms, in "August Special Term 1958", but given Oyez's filing limitations, it had no choice but to file them under either 1957 or 1958. Unfortunately, each of those terms already had a case docketed as No. 1, which added to the confusion. Oyez circumvented that confusion by creating more confusion: it "consolidated" both arguments under name of the latter (Cooper v. Aaron) but with the docket number of the former (No. 1 Misc). In hindsight, the 1958 term would have a been better choice, if only because the Supreme Court listed those cases in the 1958 Journal (the Court doesn't produce separate journals for special terms).

Another anomaly arises when the Court invites amicus parties to argue on behalf of a set of cases, rather than a specific case. For example, in the 1962 term, the following cases were argued over a period of three days:

On the third day, the Solicitor General and others then argued as amici, and while the Solicitor General argued with respect to all the cases, other attorneys (e.g., Joseph Kaufman) argued only with respect to specific cases (e.g., No. 26). Unfortunately, Oyez didn't have an easy way to attach amicus arguments to multiple cases, so it simply tacked all the amicus arguments onto Shuttlesworth v. City of Birmingham (No. 67) [see November 7, 1962]. Since there are no notations to explain this, anyone looking for the complete set of arguments in another case (e.g., Griffin v. Maryland) will not know where to find the amicus arguments, or even that they exist.

These exceptions are what consume 90% of the time required to audit a single term. 90% of all argued cases match up perfectly between Journals, NARA records, and Oyez records -- quickly and automatically. It's figuring out what's wrong with the remaining 10% that can take hours of sleuthing. As the process moves forward in time, the process should improve and the number of exceptions shrink.

Working backwards before 1955, we will lose the benefit of any Oyez or NARA audio records that we can cross-reference, so the auditing process needs to incorporate other sources of data as well, such as:

The Supreme Court Database Revisited

As luck would have it, the very first case in this initial audit, Texas v. New Mexico (No. 9 Orig.), was missing from SCDB, because even though it was argued on October 10, 1955, it was dismissed the following term, on February 25, 1957, "because of the absence of the United States as an indispensable party." This was reported on p.166 of the 1956 Journal and in the decision at 352 U.S. 991.

This is another example of frustrating inconsistencies with SCDB. It does not consistently include cases that were 1) dismissed as improvidently granted ("DIG'ed"), 2) dismissed for want of a substantial federal question, 3) dismissed ("affirmed") by an equally divided Court, 4) dismissed pursuant to Rule 60 due to some mistake or omission in the lower court, or 5) dismissed for any other reason, as in the case above.

Texas v. New Mexico is a particularly unfortunate omission, because as The National Archives noted:

NARA’s holdings of Supreme Court audio recordings date to the start of the 1955 court session. The earliest recording in the collection is from the October 10, 1955 argument in the case of Texas v. New Mexico, an interstate water dispute which persists to modern cases and arguments before the Court.

Once a case has been granted, and particularly once a case has been argued, it needs to be recorded, regardless of outcome, so that researchers can identify how often cases are dismissed after being granted, why they were dismissed, and whether (and when) they were argued before being dismissed.

Cases "affirmed by an equally divided Court" are perhaps the least interesting cases to record, even though SCDB makes some effort to record them, because the Court said absolutely nothing about them. In all other dismissals, the Court invariably provides a reason for its action, and yet SCDB does not always record them.

The 1952 term is illustrative. Looking through U.S. Reports Volume 344, you can easily spot three deadlocked cases from that term, all "affirmed by an equally divided Court":

  • No. 13 at 344 U.S. 860 (SCDB ID 1952-160)
  • No. 24 at 344 U.S. 861 (SCDB ID 1952-159)
  • No. 218 at 344 U.S. 916 (SCDB ID 1952-143)

all of which SCDB dutifully records. And yet a number of other cases that were decided by a brief per curiam opinion were not recorded by SCDB:

  • Nos. 197 and 198 at 344 U.S. 923
  • No. 390 at 344 U.S. 924

Those are significant omissions, because all three cases (Nos. 197, 198 and 390) were fully briefed and argued and decided, the lack of a detailed opinion notwithstanding.

As an aside, I dislike the semantics of the phrase "affirmed by an equally divided Court." Affirmance implies intention, but when the Court is deadlocked, there is no intention to affirm. Words like "affirm" or "reverse" only make sense when accompanied by some conclusion, which can take the form of a full opinion, a brief per curiam opinion, or an even briefer dismissal (as when improvidently granted). In all those instances, a majority of the Court reached some decision -- even deciding not to decide is a decision -- whereas a deadlock is the complete absence of a decision.

That said, I'm not arguing that deadlocked cases are somehow less important. Every case, once granted, should be treated with equal importance.

Hicks v. District of Columbia (No. 51) from the 1965 term is another example of an unfortunate omission in SCDB. The case was "dismissed as improvidently granted", and while (as usual) the Court said nothing more about it, several Justices wrote separately, shedding more light on the Court's reasoning.

First, there was a brief concurrence from Justice Harlan:

"Among the several reasons which support the action of the Court in dismissing the writ in this case as improvidently granted, I rest my decision to join in this disposition on the lack of a record, without which I do not believe the constitutional issues tendered can properly be decided."

His concurrence referred to "several reasons", only one of which he mentioned (lack of a record). But there was also a dissent from Justice Douglas that made those reasons clear: an incomplete record and a time bar that the Court refused to waive.

"DIG'ed" cases rarely provide that much detail, and they never provide a vote, so we can't know if such decisions were unanimous or whether there were any Justices with reservations who nevertheless didn't feel strongly enough to say so. And those decisions are typically buried in the back of U.S. Reports, along with all the other cases that were not even granted, much less briefed and argued.

However, Hicks v. District of Columbia (No. 51) was briefed and argued, and the decision -- including the concurrence and dissent -- is on full display in U.S. Reports, at 383 U.S. 252.

{% include arguments.html term="1952" termEnd="1975" %}