ICYMI: Goldfarb on A (Mostly Corpus-Based) Linguistic Reexamination of D.C. v. Heller and the Second Amendment @NealGoldfarb

ICYMI: Neal Goldfarb has published A (Mostly Corpus-Based) Linguistic Reexamination of D.C. v. Heller and the Second Amendment. Here is the abstract.

This is an in-depth linguistic analysis of the key language in the Second Amendment ("the right of the people to keep and bear Arms") that is based primarily on evidence of actual 18th-century usage. That evidence comes from two corpora that have been developed and made available by the BYU Law School as resources for researching the original meaning of the language used in the Constitution: COFEA (the Corpus of Founding Era American English) and COEME (the Corpus of Early Modern English). The corpus data provides powerful evidence that contrary to what the Supreme Court held in District of Columbia v. Heller, "bear arms" was used in the Second Amendment in its idiomatic military sense, and in fact that it was most likely understood to mean serve in the militia. Thus, the right to bear arms was most likely understood as being the right to serve in the militia. The analysis proceeds roughly as follows: "BEAR" and "ARMS": The Supreme Court’s interpretation of "bear" and "arms" in District of Columbia v. Heller was accurate as far as it went, but it is clear from evidence of historical usage that was unavailable at the time that the Court’s interpretation failed to reflect how "bear" and "arms" were actually used in the late 18th century. Although "bear" was sometimes used to mean ‘carry,’ the two words weren’t generally synonymous. The ways in which "bear" was used differed substantially from those for "carry." While "carry" was often used to denote the physical carrying of tangible objects (e.g., "carry baggage"), "bear" was seldom used that way. In fact, "carry" had by the end of the 1600s replaced "bear" as the verb generally used to convey the meaning ‘carry.’ In addition, although "arms" was often used to mean ‘weapons,’ it was also used roughly as often to convey a variety of figurative meanings relating to the military. "BEAR ARMS": The corpus data for "bear arms" was overwhelmingly dominated by uses of the phrase in its idiomatic military sense. (This is unsurprising given the conclusions, above, regarding "bear" and "arms.") The Supreme Court in Heller was therefore mistaken in declaring that the “natural meaning” of "bear arms" was essentially, ‘carry weapons in order to be prepared for confrontation.’ The phrase was ordinarily used to convey the meaning ‘serve in the military’ (specifically, ‘in the militia’) or ‘fight in a war.’ "THE RIGHT OF THE PEOPLE TO...BEAR ARMS": Consistently with how "bear arms" was ordinarily used, the right to bear arms was most likely understood as conveying its idiomatic military sense, and in particular as meaning ‘the right to serve in the militia.’ That conclusion is based to a large extent on the fact that there is reason to think that "bear arms" was understood to mean the same thing as to the right to bear arms as it meant with respect to the duty to bear arms — and the duty to bear arms was understood as a duty to serve in the militia. In addition, there is reason to believe, contrary to what the Court said in Heller, that as used in the Second Amendment, "the people" referred to those who were eligible for militia service. The interpretation described above is not ruled out by the fact that "bear arms" appears as part of the phrase "keep and bear arms." Although that interpretation requires that arms be understood as being simultaneously literal (as part of "keep arms") and figurative (as part of "bear arms") there is reason to believe that that was in fact how "keep and bear arms " was understood at the time of the Second Amendment’s framing and ratification.

Download the article from SSRN at the link.

Phillips and Blackman on Corpus Linguistics and Heller @JoshMBlackman

James Cleith Phillips, Chapman University School of Law, and Josh Blackman, South Texas College of Law, are publishing Corpus Linguistics and Heller in volume 56 of the Wake Forest Law Review (2021). Here is the abstract.

In District of Columbia v. Heller, the Supreme Court sharply divided over the meaning of the twenty-seven words in the Second Amendment. Justice Scalia wrote the majority opinion. He concluded that the Second Amendment “protects an individual right to possess a firearm unconnected with service in a militia, and to use that arm for traditionally lawful purposes, such as self-defense within the home.” In short, an “individual” right. Justice Stevens, in his dissent, contended that the Second Amendment “is most naturally read to secure to the people a right to use and possess arms in conjunction with service in a well-regulated militia.” That is, a “collective” right. Justice Scalia and Justice Stevens both made linguistic claims about four elements of the Second Amendment: “right of the people,” “keep and bear arms,” “keep arms,” and “bear arms.” Both the majority and the dissent used various textualist approaches to consider these four phrases, but their toolkit in 2008 was limited. They considered only a fairly narrow range of sources to interpret the text. Today, we can do better. In this Article, we will grade the four linguistic claims made in the Heller case using corpus linguistics. We rely on the Corpus of Founding Era American English (“COFEA”). In 2015, one of us conceptualized and oversaw the initial development of COFEA. We performed five queries with COFEA. First, we queried right of the people. Second, we queried keep and bear arms (and synonyms). Third, we queried the word right within six words of arms. Fourth, we queried the word keep, and variants of keep, within six words of arms. Fifth, we queried the word bear, and variants of bear, within six words of arms. We used multiple coders who independently coded their results using a type of double-blind methodology. Both the majority and the dissenting opinions erred with respect to some of their linguistic claims. Justices Scalia and Stevens should have expressed far more caution when reaching their textualist conclusions based on the narrow subset of founding-era sources they reviewed. Additionally, corpus linguistic theory reveals that there are inconsistencies in both Justice Scalia’s and Stevens’s descriptions of the Second Amendment’s original public meaning.

Download the article from SSRN at the link.

Smith and Peterson on Big Data Comes For Textualism: The Use and Abuse of Corpus Linguistics in Second Amendment Litigation

Mark W. Smith, Oxford University Department of Pharmacology; The King's College, and Dan M. Peterson, Independent, have published Big Data Comes for Textualism: The Use and Abuse of Corpus Linguistics in Second Amendment Litigation. Here is the abstract.

Some scholars, judges, and advocates have recently urged that legal corpus linguistics, a methodology that uses computerized searches of large volumes of texts known as “corpora,” can determine the original meaning of constitutional provisions. More particularly, certain of these advocates have argued that corpus linguistics searches of Founding era corpora prove that the Second Amendment right to keep and bear arms protects only a collective, militia right and not an individual, private right to arms, contrary to the Supreme Court’s interpretation of that amendment in District of Columbia v. Heller, 554 U.S 570 (2008). In this article, we argue that relying on corpus linguistics to determine the meaning of the Second Amendment suffers from severe conceptual and practical difficulties. One of the most fundamental flaws concerns the central methodological assumption of corpus linguistics—the “frequency hypothesis”—which posits that the most frequent meaning of a word or phrase returned by a corpus search should be the meaning adopted for purposes of constitutional interpretation. Even if the phrase “bear arms” most frequently appears in a military context, that does not mean that the constitutional language excludes an individual right to bear arms for self-defense and other private purposes. Military and militia references were more likely to appear in public discussions of the right to bear arms simply because they were more “newsworthy” than the mundane acts of ordinary people carrying a firearm for hunting or defense, which would rarely be recorded. Contemporary examples, including references by the Founders themselves, show that the right to “bear arms” included protection of an individual right as well as furthering a well-regulated militia. In addition, corpus linguistics suffers from serious problems concerning the composition of the corpora, which are biased in favor of elite language usage and are critically incomplete, missing some of the key texts that historians and legal scholars have long relied upon in discerning the Second Amendment’s meaning. Use of legal corpus linguistics also raises serious practical difficulties in actual constitutional litigation, including the absence of the usual safeguards applicable to expert or “scientific” evidence. In the end, the counting of words resulting from a corpus search cannot overcome the history and traditions at the time of the Founding that allowed free carry and use of firearms, and the core conception by the Founders that self-protection with arms is a pre-existing right that cannot be taken away from the individual by any act of civil society.

Download the article from SSRN at the link.

Tobia on Dueling Dictionaries and Clashing Corpora @kevin_tobia

Kevin Tobia, Georgetown University Law Center; Georgetown University Department of Philosophy, has published Dueling Dictionaries and Clashing Corpora. Here is the abstract.

Judges increasingly look to corpus linguistic tools in legal interpretation, as scholars advance corpus linguistics arguments about statutory and constitutional language. Corpus linguistics is sometimes offered as a preferred interpretive tool, avoiding the pitfalls of dueling canons or cherry-picked dictionary definitions. However, this short essay proposes, legal corpus linguistic tools are unlikely to resolve most difficult debates about the ordinary or public meaning of law. The essay articulates ten emerging “arguments” and “counterarguments” of legal corpus linguistics. Of course, the existence of “clashing corpora” does not imply legal corpus linguistics will be abandoned. It’s been decades since the observation of “dueling canons” and “dueling dictionaries.” Courts today regularly look to both tools.

Download the essay from SSRN at the link.

Bernstein on How Technologies of Language Meet Ideologies of Law @UBSchoolofLaw @anyabernstein

Anya Bernstein, University at Buffalo Law School, is publishing Technologies of Language Meet Ideologies of Law (Symposium: Law, Language, and Technology) in the 2020 volume of the Michigan State Law Review (forthcoming in 2021). Here is the abstract.

A new technology of interpretation is taking the legal world by storm. Legal corpus linguistics, an approach generally unknown in the field until a few years ago, has suddenly become a focus for articles, conferences, legal briefs, and even judicial opinions. Taking advantage of evolving computational approaches and data collection abilities, legal corpus linguistics searches big data sets of language use to help interpret legal texts. This Article puts legal corpus linguistics in the context of other meaning-making technologies and suggests an approach for analyzing any technology of language in the law. One of my aims is to caution against technological exceptionalism—a view that computerized, automated, or big-data approaches are somehow special, perhaps more trustworthy, less subjective, and most likely to succeed. Rather, I argue that we should ask the same questions, and make the same demands, of any method of interpretation. As science and technology studies (STS) and related scholarship has demonstrated, technology is not neutral or passive. It is a cause in its own right. That makes it particularly important to examine the underlying assumptions that help construct, and are perpetuated through, a given technology. To elucidate these points, I draw on theorists who have influenced our understandings of the production of knowledge and technological development, showing how he key contributions by Bruno Latour, Ian Hacking, and Michel Foucault should inform our evaluation of legal language technologies. I then introduce legal corpus linguistics, describing its origins in academic linguistics and the somewhat different way it has been practiced in legal interpretation. Having laid this groundwork, I ask how we should evaluate this emerging technology in legal interpretation. I argue that legal corpus linguistics fails to coherently relate its methods, questions, aims, and claims. Moreover, it inscribes a peculiar view of legal meaning: a narrow, asocial, and abstracted notion of things that are in fact broad, social, and practice-based. The illusion of simplicity that legal corpus linguistics propagates undermines our evolving understanding of the real complexities of law and leaves out participants and contexts that are crucial to the production of law as a social force. To probe its implications further, I then put legal corpus linguistics in the context of some other ways of giving laws meaning. I choose two that sit at the extremes of simplicity and complexity: dictionary definitions, on the one hand, and administrative rulemaking procedures, on the other. These may seem unrelated or incommensurable, but in fact they all are technologies of legal interpretation that should be considered in comparison to one another. Comparison also helps illuminate those aspects of legal corpus linguistics that fit it snugly into particular legal ideologies, but blind it to the realities of how law functions in society.

Download the article from SSRN at the link.