Corpus linguistics a statutory interpretation tool for lawyers
Corpus linguistics has been in the news lately, giving us the opportunity to discuss this interesting tool of law interpretation and, in doing so, revisit some Sixth Circuit views on it.
What is corpus linguistics? We’ll let Circuit Judge Amul Thapar explain:
[Corpus linguistics] builds on the common knowledge of the layman by showing us the common uses of words in our common language. How it works? Corpus linguistics allows lawyers to use a searchable database to find specific examples of how a word was used at a given time. . . These databases, available primarily online, contain millions of everyday word usage examples (from spoken words, works of fiction, magazines, newspapers, and scholarly works). . . Lawyers can search these databases for the ordinary meaning of statutory language. . . Matching search results will give a broader and more empirical understanding of the ordinary meaning of a word or phrase by giving us different situations in which the word or phrase has been used in a wide variety of common usages. . . In short, corpus linguistics is a powerful tool for discerning how the public would have understood the text of a law at the time it was enacted.
Wilson vs. Safelite Grp., Inc.930 F.3d 429, 440 (6th Cir. 2019) (Thapar, J., concurs) (internal citation omitted).
Corpus linguistics is a tool used to identify the original public meaning of words – no small feat when the outcome of a case often depends on the meaning of a single word in a statute or the Constitution. And if you fail to address corpus linguistics in your brief, you may receive a letter from the court asking you to submit an additional brief correcting the omission.
Consider what happened a few days ago in the highly anticipated Ninth Circuit case of Jones v. Bonta, which involved the Second Amendment. Judges Nelson, Lee, and Stein “directed the parties to file a supplementary briefing note addressing in part the applicability of corpus linguistics to [the] Case.” No. 20-56174, 2022 US App. LEXIS 12657, at *16 n.6 (9th Cir. May 11, 2022). Circuit Judges Thapar and Siler and Judge Hood of the Eastern District of Kentucky “directed the parties to file additional briefs on the original meaning of the case or controversy requirement of Article III, particularly whether the corpus of American English of the founding era helped illuminate this meaning. Wright v. Spaulding, 939 F.3d 695, 700 n.1 (6th Cir. 2019). In neither Jones v. Bonta neither Wright v. Spaulding did corpus linguistics control the outcome of the case. But they show judges’ appetite for bringing new tools to discern the original public meaning of the Constitution or other legislation.
Regardless of the prior briefing, judges at the district or circuit level can use corpus linguistics to rule for or against you. Two cases illustrate this point. See United States vs. Woodson960 F.3d 852 (6th Cir. 2020); Freedom Health Def. Fund, Inc. vs. Biden, no. 8:21-cv-1693-KKM-AEP (MD Fla. 18 April 2022). These examples underscore the usefulness of corpus linguistics in determining the meaning of laws.
In Woodson, the accused and his accomplices robbed more than a dozen diamond shops in several states. At sentencing, the district court determined that the defendant’s sentence should be increased because the defendant “moved or participated in the relocation of a fraudulent scheme to another jurisdiction to escape the forces law enforcement or regulatory officials”.
Justice Readler, writing for the majority, helpfully broke down the law into four elements that trigger improvement: “(1) relocation or participation in relocation, (2) a fraudulent scheme, (3) to another jurisdiction, (4) to evade law enforcement or regulatory officials. At issue were the district court’s reading of the first two items. Defendant argued that because the “basis of tether” or the “hub” of the scheme remained Toledo, Ohio, throughout the robberies, it never moved the scheme. The district court held otherwise, finding that the defendant “deliberately targeted stores” in multiple states “to impede communication between law enforcement, triggering improved relocation.”
The panel disagreed with Woodson’s interpretation that “a schematic” is something tangible, like a “hub”. Rather, Judge Readler, referring to numerous dictionaries, found that a scheme is something intangible like a plot or a plan. Thus, the district court was correct that the deliberate targeting of diamond shops in multiple states was sufficient to satisfy the first two elements. Not stopping there, Judge Readler conducted a corpus linguistic analysis of the term “schema”, and found the analysis to be consistent with the dictionaries cited. Corpus linguistics extinguished all doubt.
In Freedom Health Def. Funds, Inc., Florida Intermediate District Judge Kathryn Kimball Mizelle used corpus linguistics to fortify her opinion in one of the most high-profile cases so far this year. The Health Freedom Defense Fund challenged the CDC’s imposition of a mask mandate on all U.S. air travel under the Public Health Services Act of 1944 (“PHSA”).
The PHSA authorized the CDC to “make and enforce those regulations” necessary to “prevent the introduction, transmission, or spread of disease” through “fumigation, disinfection, sanitation” and other actions. The United States argued that the airlines’ mask mandate was a sanitation measure. Justice Mizelle found that at the time the PHSA was enacted, sanitation had one of two meanings: 1) “measures that clean something or remove dirt” or 2) “measures that keep something clean”. If sanitation was given the old meaning, the warrant failed because the masks aren’t actively cleaning anything. If the latter is given, the mask mandate must be respected because masks keep the air clean from COVID-19.
Justice Mizelle relied on “all the traditional tools of statutory interpretation” to decide the issue. First she looked at the context. “Sanitation” was accompanied by active (non-preventive) words such as fumigation and disinfection. It favored the first sense. The structure and history of the law was also consistent with the first definition.
Furthermore, using corpus linguistics, Justice Mizelle determined that “customary usages of the time agree[d]” with his discoveries. She searched for uses of ‘sanitization’ in the relevant corpus over the relevant period and found that the most frequent use of sanitization was in the context of ‘a positive act to make a thing or a clean place” while only 5% were “sanitation”. as a measure to maintain a state of cleanliness, or as a barrier to keep something clean. As such, Judge Mizelle concluded that “sanitation” carried the first meaning and that the mask mandate was not authorized.
Corpus linguistics is not a silver bullet. As Justice Thapar noted, “Corpus linguistics is a tool – new to lawyers and still developing – but not the whole toolkit. Its primary value may come from those difficult cases where laws divide and dictionaries diverge. Wilson vs. Safelite Grp., Inc.930 F.3d 429, 440 (6th Cir. 2019); see also id. at 445-448 (J Stranch, concurring with her own opinion in response to Justice Thapar’s “endorsement of ‘corpus linguistics'” and noting in particular: the practical problems of privileging newsworthy connotations of a term” versus the “ordinary meaning”, the difficulty of weeding out “irrelevant results” and concerns about courts performing such statistical analyses).
Still, corpus linguistics is probably something litigants should be familiar with when interpreting the meaning of a statute or constitutional provision. You may be asked to inform the judges about this. And it could end up being the tool that persuades the court to accept an interpretation of a law or the Constitution in favor of your client.
Alon Farahan also contributed to this article.
© Copyright 2022 Squire Patton Boggs (USA) LLPNational Law Review, Volume XII, Number 133